979797979797979797979797979797979

79

E550

Business Conditions Analysis979797979797979797979797979797979

79

Expenditures

C + I + G

Gross Domestic Product

979797979797979797979797979797979

79

Dr. David A. DiltsDepartment of Economics

June 2006 first revisionMay 2005

Business Conditions Analysis, E550

© Dr. David A. Dilts

All rights reserved. No portion of this bookmay be reproduced, transmitted, or stored, byany process or technique, without the expresswritten consent of Dr. David A. Dilts

2006

Published by Indiana - Purdue University - Fort Waynefor use in classes offered by the Department of Economics, School of Business and Management Sciences at I.P.F.W. by permission of Dr. David A. Dilts

i

SYLLABUSE550, Business Conditions Analysis

Dr. David A. Dilts Office: Neff Hall 340DDepartment of Economics Office Phone: 481-6486School of Business and Management Sciences Email Office: Dilts@ipfw.eduIndiana University - Purdue University- Fort Wayne Home Phone: 486-8225

Email Home: Ddilts2704@aol.com

Course Policies

1. This course is an intersession course. It meets from 5:30 p.m. to 10:20 p.m. for ten straightweek days beginning August 7, 2006.

2. Grades will be determined through take-home quizzes. A package of quizzes will bedistributed to class, and you will be expected to turn one in at the beginning of class on August8, August, 10, August 14, August 16 August 17, and August 18, 2005. There are a total of sixquizzes, each worth twenty points. The best five quizzes will be used to calculate your finalgrade. The grade scale is:

89-100 A77-88 B65-76 C53-64 D52 and below F

3. Attendance will not be taken, however, it is strongly encouraged.

4. The quizzes may be turned-in early, but will not be accepted more than one class day late. Remember, as much as this is like taking a drink from a fire plug, it is worse for the instructorwho has to grade the work you turn in, and you people outnumber me. I’m not trying to beofficious, but I want to survive this too.

5. All other department, school, campus and university policies will be applicable to this courseand strictly observe.

Course Objectives:

This course has several objectives which are very much intertwined. These objectives include:

To solve problems innovatively, using the following tools, and concepts:

ii

1. A quick review of the essence of macroeconomics which is most important in understandingthe environment of business both domestically and globally, including:

A. National Income Accounts, Price Indices, and Employment DataB. Keynesian Model of Macroeconomic Activity

1. Business Cycles2. Fiscal Policy

C. Monetary Aggregates1. Monetary Policy2. Exchange rates3. Interest Rates

2. Economic data sources and their interrelationsA. Economic Indicators

1. Leading indicators2. Concurrent indicators3. Trailing indicators

B. Understanding sector performance using economic aggregates

3. Forecasting models and their usesA. Simple internal modelsB. Correlative methodsC. Limitations and value

4. Putting together the notion of business environment within a strategic view of a businessenterprise

With these issues mastered it is also the objective of this course to master these tools to beable to integrate and synthesize business conditions information and analysis for thepurposes of planning and decision making.

Reading AssignmentsEconomics:Monday, August 7, 2006Dilts, Chapters 1-2

Tuesday, August 8, 2006Dilts, Chapters 3

Wednesday, August 9, 2006Dilts, Chapters 4-5 Thursday, August 10, 2006Dilts, Chapters 6-7

iii

Friday, August 12, 2006Dilts, Chapters 8-9

Monday, August 14 2006Dilts, Chapter 10-11

Tuesday, August 15, 2006Dilts, Chapter 12

Data:

Wednesday, August 16, 2006Dilts, Chapter 13-14

Thursday, August 17, 2006Dilts, Chapter 15

Friday, August 18, 2006Dilts, Chapter 16 -17 and Catch-up.

iv

Chapter 1

Introduction to Economics

In general, the purpose of this chapter is to provide the basic definitions uponwhich the subsequent discussions of macroeconomics and business conditions will bebuilt. The specific purpose of this chapter is to define economics (and its majorcomponent fields of study), describe the relation between economic theory andempirical economics, and examine the role of objectivity in economic analysis, beforeexamining economic goals and their relations. For those of you who have had E201 orE202, Introduction to Microeconomics or Introduction to Marcoeconomics much of thematerial contained in this chapter will be similar to the introductory material contained inthose courses. However, you will find that there is considerable expansion of thediscussions offered in the typical principle courses, as a matter of foundation for themore in depth discussions of business conditions that will occur in E550.

Definitions

Economics has been studied since sixteenth century and is the oldest of thesocial studies. Most of the business disciplines arose in attempt to fill some of theinstitutional and analytical gaps in the areas with which economics was particularly wellsuited to examine. The subject matter examined in economics is the behavior ofconsumers, businesses, and other economic agents, including the government in theproduction and allocation processes. Therefore, any business discipline will have somedirect relation with the methods or at least the subject matter with which economistsdeal.

Economics is one of those words that seems to be constantly in the newspapersand on television news shows. Most people have some vague idea of what the wordeconomics means, but precise definitions generally require some academic exposure tothe subject. Economics is the study of the allocation of SCARCE resources tomeet UNLIMITED human wants. In other words, economics is the study of humanbehavior as it pertains to the material well-being of people (as either individuals orsocieties).

Robert Heilbroner describes economics as a "Worldly Philosophy." It is theorganized examination of how, why and for what purposes people conduct their day-to-day activities, particularly as relates to the production of goods and services, the

1

accumulation of wealth, earning incomes, spending their resources, and saving forfuture consumption. This worldly philosophy has been used to explain most rationalhuman behavior. (Irrational behavior being the domain of specialities in sociology,psychology, history, and anthropology.)

Underlying all of economics is the base assumption that people act in their ownbest interest (at least most of the time and in the aggregate). Without the assumptionof rational behavior, economics would be incapable of explaining the preponderance ofobserved economic activity. Consistent responses to stimuli are necessary for a modelof behavior to predict future behavior. If we assume people will always act in their besteconomic interests, then we can model their behavior so that the model will predict (withsome accuracy) future economic behavior. As limiting as this assumption may seem, itappears to be an accurate description of reality. Experimental economics, using rats inmazes, suggests that rats will act in their own best interest, therefore it appears to be areasonable assumption that humans are no less rational.

Most academic disciplines have evolved over the years to become collections ofclosely associated scholarly endeavors of a specialized nature. Economics is noexception. An examination of one of the scholarly journals published by the AmericanEconomics Association, The Journal of Economic Literature reveals a classificationscheme for the professional literature in economics. Several dozen specialities areidentified in that classification scheme, everything from national income accounting, tolabor economics, to international economics. In other words, the realm of economicshas expanded to such an extent over the centuries that it is nearly impossible foranyone to be an expert in all aspects of the discipline, so each economist generallyspecializes in some narrow portion of the discipline. The decline of the generalist is afunction of the explosion of knowledge in most disciplines, and is not limited toeconomists. Economics can be classified into two general categories, these are (1)microeconomics and (2) macroeconomics. Microeconomics is concerned withdecision-making by individual economic agents such as firms and consumers. Inother words, microeconomics is concerned with the behavior of individuals or groupsorganized into firms, industries, unions, and other identifiable agents. Microeconomicsis the subject matter of E201 and E321 Microeconomics.

Macroeconomics is concerned with the aggregate performance of theentire economic system. Unemployment, inflation, growth, balance of trade, andbusiness cycles are the topics that occupy most of the attention of students ofmacroeconomics. In E202 and E321 the focus is on macroeconomic topics. Thesematters are the topics to be examined this course E550, Business Conditions Analysis.

Macroeconomics is a course that interfaces with several other academicdisciplines. A significant amount of the material covered in this course involves public

2

policy and has a significant historical foundation. The result is that much of what iscurrently in the news will be things that are being studied in this course as they happen. In many respects, that makes this course of current interest, if not fun.

Methods in Economics

Economists seek to understand the behavior of people and economic systemsusing scientific methods. These scientific endeavors can be classified into twocategories, (1) economic theory and (2) empirical economics. Economic theory reliesupon principles to analyze behavior of economic agents. These theories aretypically rigorous mathematical models (abstract representations) of behavior. A goodtheory is one that accurately predicts future behavior and is consistent with the availableevidence.

Empirical economics relies upon facts to present a description ofeconomic activity. Empirical economics is used to test and refine theoreticaleconomics, based on tests of economic theory. The tests that are typically applied toeconomic theories are statistically based, and is generally called econometric methods. Much of the material involved in forecasting techniques draws heavily from thesemethods. The regression-based forecasting models are very similar to the econometricmodels used to analyze the macroeconomy.

In Business Conditions analysis we will rely heavily on a sound theoretical basisto understand the macroeconomic framework in which business activity arises. Withinthis theoretical framework a range of empirical models, and data will be examined whichwill allow us to both formally and informally forecast economic conditions and what sortsof phenomenon are associated with these conditions for specific markets.

Theory concerning human behavior is generally constructed using one of twoforms of logic. Sociology, psychology and anthropology typically rely on inductive logicto create theory. Inductive logic creates principles from observation. In otherwords, the scientist will observe evidence and attempt to create a principle or a theorybased on any consistencies that may be observed in the evidence. Economics reliesprimarily on deductive logic to create theory. Deductive logic involves formulatingand testing hypotheses. Often the theory that will be tested comes form inductivelogic or sometime informed guess-work. The development of rigorous modelsexpressed as equations typically lend themselves to rigorous statistical methods todetermine whether the models are consistent with evidence from the real world. Thetests of hypotheses can only serve to reject or fail to reject a hypothesis. Therefore,empirical methods are focused on rejecting hypotheses and those that fail to be rejectedover large numbers of tests generally attain the status of principle.

3

However, examples of both types of logic can be found in each of the socialsciences. In each of the social sciences it is common to find that the basic theory isdeveloped using inductive logic. With increasing regularity standard statistical methodsare being employed across all of the social sciences and business disciplines to test thevalidity of theories.

The usefulness of economics depends on how accurate economic theorypredicts behavior. Even so, economics provides an objective mode of analysis, withrigorous models that permit the discounting of the substantial bias that is usuallypresent with discussions of economic issues. The internal consistency brought toeconomic theory by mathematical models often fosters objectivity. However, no modelis any better than the assumptions that underpin that model. If the assumptions areeither unrealistic or formulated to introduce a specific bias, objective analysis ca still bethwarted (under the guise of scientific inquiry).

The purpose of economic theory is to describe behavior, but behavior isdescribed using models. Models are abstractions from reality - the best model is theone that best describes reality and is the simplest (the simplest requirement is calledOccam's Razor). Economic models of human behavior are built upon assumptions; orsimplifications that allow rigorous analysis of real world events, without irrelevantcomplications. Often (as will be pointed-out in this course) the assumptions underlyinga model are not accurate descriptions of reality. When the model's assumptions areinaccurate then the model will provide results that are consistently wrong (known asbias).

One assumption frequently used in economics is ceteris paribus which meansall other things equal (notice that economists, like lawyers and doctors will use Latin toexpress rather simple ideas). This assumption is used to eliminate all sources ofvariation in the model except for those sources under examination (not very realistic!).

Economic Goals, Policy, and Reality

Most people and organizations do, at least rudimentary planning, the purpose ofplanning is the establishment of an organized effort to accomplish some economicgoals. Planning to finish your education is an economic goal. Goals are, in a sense, anidea of what should be (what we would like to accomplish). However, goals must berealistic and within our means to accomplish, if they are to be effective guides to action. This brings another classification scheme to bear on economic thought. Economics canbe again classified into positive and normative economics.

Positive economics is concerned with what is; and normative economics isconcerned with what should be. Economic goals are examples of normative

4

economics. Evidence concerning economic performance or achievement of goals fallswithin the domain of positive economics.

Most nations have established broad social goals that involve economic issues. The types of goals a society adopts depends very much on the stage of economicdevelopment, system of government, and societal norms. Most societies will adopt oneor more of the following goals: (1) economic efficiency, (2) economic growth, (3) economic freedom, (4) economic security, (5) an equitable distribution of income, (6)full employment, (7) price level stability, and (8) a reasonable balance of trade.

Each goal (listed above) has obvious merit. However, goals are little more thanvalue statements in this broad context. For example, it is easy for the very wealthy tocite as their primary goal, economic freedom, but it is doubtful that anybody living inpoverty is going to get very excited about economic freedom; but equitable distributionsof income, full employment and economic security will probably find rather wide supportamong the poor. Notice, if you will, goals will also differ within a society, based onsocio-political views of the individuals that comprise that society.

Economics can hardly be separated from politics because the establishment ofnational goals occurs through the political arena. Government policies, regulations, law,and public opinion will all effect goals and how goals are interpreted and whether theyhave been achieved. A word of warning, eCONomics can be, and has often been used,to further particular political agendas. The assumptions underlying a model used toanalyze a particular set of circumstances will often reflect a political agenda of theeconomist doing the analysis. For example, Ronald Reagan argued that governmentdeficits were inexcusable, and that the way to reduce the deficit was to lower peoples'taxes -- thereby spurring economic growth, therefore more income that could be taxedat a lower rate and yet produce more revenue. Mr. Reagan is often accused, by hisdetractors, of having a specific political agenda that was well-hidden in this analysis. His alleged goal was to cut taxes for the very wealthy and the rest was just rhetoric tomake his tax cuts for the rich acceptable to most of the voters. (Who really knows?) Most political commentators, both left and right, have mastered the use of assumptionsand high sounding goals to advance a specific agenda. This adds to the lack ofobjectivity that seems to increasingly dominate discourse on economic problems.

On the other hand, goals can be publicly spirited and accomplish a substantialamount of good. President Lincoln was convinced that the working classes should haveaccess to higher education. The Morrell Act was passed 1861 and created Land Grantinstitutions for educating the working masses (Purdue, Michigan State, Iowa State, andKansas State (the first land grant school) are all examples of these types of schools). By educating the working class, it was believed that several economic goals could beachieved, including growth, a more equitable distribution of income, economic securityand freedom. In other words, economic goals that are complementary are consistentand can often be accomplished together. Therefore, conflict need not be thecenterpiece of establishing economic goals.

5

Because any society's resources are limited there must be decisions about whichgoals should be most actively pursued. The process by which such decisions are madeis called prioritizing. Prioritizing is the rank ordering of goals, from the most important tothe least important. Prioritizing of goals also involves value judgments, concerningwhich goals are the most important. In the public policy arena prioritizing of economicgoals is often the subject of politics.

Herein lies one of the greatest difficulties in macroeconomics. An individual caneasily prioritize goals. It is also a relatively easy task for a small organization or firm toprioritize goals. For the United States to establish national priorities is a far larger task. Adam Smith in the Wealth of Nations (1776) describes the basic characteristics ofcapitalism (this book marks the birth of capitalism). Smith suggests that there are threelegitimate functions of government in a free enterprise economy. These three functionsare (1) provide for the national defense, (2) provide for a system of justice, and (3)provides those goods and services that cannot be effectively provided by the privateeconomy because of the lack of a profit motive. There is little or no controversyconcerning the first two of these government functions. Where debate occurs is overthe third of these legitimate roles.

Often you hear that some non-profit organization or government agency shouldbe "run like a business." A business is operated to make a profit. If the capitalistmodel is correct, then the only reason for an entrepreneur to establish and operate abusiness is to make profits (otherwise the conduct of the business is irrational andcannot be explained as self-interested conduct). A church, charity, or school isestablished for purposes other than the making of a profit. For example, a church maybe established for the purposes of maximizing spiritual well-being of the congregation(the doing of good-works, giving testimony to one's religion, worship of God, and theother higher pursuits). The purpose of a college or a secondary/elementary schoolsystem, likewise is not to make profits, the purposes of educational institutions is toprovide access knowledge. A University is to increase the body of knowledge throughbasic and applied research, professional services, and (of primary importance to thestudents) to provide for the education of students. To argue that these public orcharitable organizations should be run like a business is to suggest that these matterscan be left to the private sector to operate for a profit. Inherent in this argument is theassumption (a fallacy) that the profit motive would suffice to assure that society receiveda quality product (spiritual or educational or both) and in the quantities necessary toaccomplish broad social objectives. Can you imagine what religion would become if itwas reduced to worldly profitability (some argue there's too much of that sort of thingnow), can you imagine what you would have to pay for your education if, instead of theState of Indiana subsidizing education, the student was asked to pay for the total cost ofa course plus some percentage of cost as a profit? Perhaps worse still, who would dothe basic research that has provided the scientific break-throughs that result inthousands of new products each year? Would we have ever had computers without thebasic research done in universities, what would be missing from our medicaltechnology?

6

Priorities at a national level are rarely set without significant debate,disagreements, and even conflict. It is through our free, democratic processes that weestablish national, state and local priorities. In other words, the establishment of oureconomic priorities are accomplished through the political arena, and therefore it isoften impossible to separate the politics from the economics at the macro level.

Objective Thinking

Most people bring many misconceptions and biases to economics. After all,economics deals with people's material well-being. Because of political beliefs andother value system components rational, objective thinking concerning variouseconomic issues fail. Rational and objective thought requires approaching a subjectwith an open-mind and a willingness to accept what ever answer the evidence suggestsis correct. In turn, such objectivity requires the shedding of the most basicpreconceptions and biases -- not an easy assignment.

What conclusions an individual draws from an objective analysis using economicprinciples, are not necessarily cast in stone. The appropriate decision based oneconomic principles may be inconsistent with other values. The respective evaluationof the economic and "other values" (i.e., ethics) may result in a conflict. If aninconsistency between economics and ethics is discovered in a particular application, arational person will normally select the option that is the least costly (i.e., the majorityview their integrity as priceless). An individual with a low value for ethics or morals mayfind that a criminal act, such as theft, as involving minimal costs. In other words,economics does not provide all of the answers, it provides only those answers capableof being analyzed within the framework of the rational behavior that forms the basis ofthe discipline.

Perhaps of greatest importance by modeling economic systems, it provide abasis of cold, calculation free from individual emotion and self-interest. Do not beconfused, anytime we deal with economic forecasts, there is always the problem of“hope” versus “what’s next.” By modeling economic activity, and empirically testing thatmodel, one may be able to produce forecasting models or even indicators that permitunbiased forecasts. To forecast is one thing, to hope is another.

There are several common pitfalls to objective thinking in economics. After all,few things excite more emotion than our material well-being. It should come as nosurprise that bias and less than objective reasoning is common when it comes toeconomic issues, particularly those involving public policy.. Among the most commonlogical pitfalls that affect economic thought are: (1) the fallacy of composition, and (2)post hoc, ergo prompter hoc. Each of these will be reviewed, in turn.

The fallacy of composition is the mistaken belief that what is true for the

7

individual must be true for the group. An individual or small group of individuals mayexhibit behavior that is not common to an entire population. In other words, this fallacyis simply assuming a small, unscientifically selected sample will predict the behavior,values, or characteristics of an entire population. For example, if one individual in thisclass is a I.U. fan then everyone in this class must be an I.U. fan is an obvious fallacy ofcomposition. Statistical inference can be drawn from a sample of individualobservations, but only within confidence intervals that provide information concerningthe likelihood of making an incorrect conclusion (E270, Introduction to Statistics,provides a more in depth discussion of confidence intervals and inference).

Post hoc, ergo prompter hoc means after this, hence because of this, andis a fallacy in reasoning. Simply because one event follows another does notnecessarily imply there is a causal relation. One event can follow another and becompletely unrelated. All of us have, at one time or another, experienced a simplecoincidence. One event can follow another, but there may be something other than adirect causal relation that accounts for the timing of the two events.

For example, during the thirteenth century people noticed that the black plagueoccurred in a location when the population of cats increased. Unfortunately, somepeople concluded that the plague was caused by cats so they killed the cats. In fact,the plague was carried by fleas on rats. When the rat population increased, cats wereattracted to the area because of the food supply (the rats). The people killed thepredatory cats, and therefore, rat populations increased, and so did the population offleas that carried the disease. This increase in the rat population also happened toattract cats, but cats did not cause the plague, if left alone they may have gotten rid ofthe real carriers (the rats, therefore the fleas). The idea that cats were observedincreasing in population gave rise to the conclusion that the cats brought the plague is apost hoc, ergo prompter hoc fallacy, but this example has an indirect relation betweencats in the real cause. Often, even this indirect relation is absent.

Many superstitions are classic examples of this type of fallacy. Broken mirrorscausing seven years bad luck, or walking under a ladder brining bad luck are nothingbut fallacies of the post hoc, ergo prompter hoc variety. There is no causal relationbetween breaking glass and bad luck or walking under ladder (unless something falls offthe ladder on the pedestrian). Deeper examination of the causal relations arenecessary for such events if the truth of the relations is to be discovered. However,more in depth analysis is often costly, and the cost has the potential of causingdecision-makers to skip the informed part and cut straight to the opinion.

Economic history has several examples of how uniformed opinion resulted invery significant difficulties for innocent third-parties, in addition, to those responsible forthe decisions. The following box presents a case where policy was implemented basedon the failure to recognize that there is a significant amount of interdependence in theU.S. economy.

8

This story shows fairly conclusively that private interests can damage society asa whole. While our economic freedom is one of the prime ingredients in making oureconomy the grandest in the world, such freedom requires that it be exercised in aresponsible fashion, lest the freedom we prize becomes a source of social harm. Likeanything else, economic freedom for one group may mean disaster for another, throughno fault of the victims. Government and the exercise of our democratic responsibilitiesis suppose to provide the checks on the negative results of the type portrayed in theabove box.

Statistical Methods in Economics

The use of statistical methods in empirical economics can result in errors ininference. Most of the statistical methods used in econometrics (statistical examinationof economic data) rely on correlation. Correlation is the statistical association oftwo or more variables. This statistical association means that the two variables movepredictably with or against each other. To infer that there is a causal relation betweentwo variables that are correlated is an error. For example, a graduate student oncefound that Pete Rose's batting average was highly correlated with movement in GNPduring several baseball seasons. This spurious correlation cannot reasonably beconsidered path-breaking economic research.

On the other hand we can test for causation (where one variable actually causesanother). Granger causality states that the thing that causes another must occurfirst, that the explainer must add to the correlation, and must be sensible. As withmost statistical methods Granger causality models permit testing for the purpose ofrejecting that a causal relation exists, it cannot be used to prove causality exists. Thesetypes of statistical methods are rather sophisticated and are generally examined inupper division or graduate courses in statistics.

As is true with economics, statistics are simply a tool for analyzing evidence. Statistical models are also based on assumptions, and too often, statistical methods areused for purposes for which they were not intended. Caution is required in acceptingstatistical evidence. One must be satisfied that the data is properly gathered, andappropriate methods were applied before accepting statistical evidence. Statistics donot lie, but sometimes statisticians do!

A whole chapter will be dedicated to forecasting methods in this course. It is notso much that forecasting methods are the cognitive content of this course, as much as itis presented to provide the student with a menu of the types of forecasting techniquesthat are available to help get a handle on what to expect. BUFW H509 is ResearchMethods in Business which focuses on data collection, statistical methods, andinference. Together with E550 these two courses should provide the student with a

9

“deep-dish” approach to both subjects.

Objectivity and Rationality

Objective thinking in economics also includes rational behavior. The underlyingassumptions with each of the concepts examined in this course assumes that peoplewill act in their perceived best interest. Acting in one's best interests is how rationality isdefined. The only way this can be done, logically and rigorously, is with the use ofmarginal analysis. This economic perspective involves weighing the costs against thebenefits of each additional action. In other words, if benefits of an additional action willbe greater than the costs, it is rational to do that thing, otherwise it is not.

Forecasting

Objectivity and rationality are critical in developing models and understandinghow business conditions arise. However, the forecaster is less interested in explaininghow the economy works, and more interested in what some series of numbers is goingto do over time, and in particular, in the near future. The end result is that forecastingand economic analyses are very often similar processes. If one is interested in theeconomic environment (business conditions), one must understand how the economyactually works. This is why much of the first half of this course is focused exclusively onthe macroeconomy.

Forecasting is like most other endeavors, you need to have a plan. The steps inforecasting provide an organized approach to permitting the best possible forecasts. The steps in forecasting involve the identification of the variable or variables you wish toforecast. Once these are identified, you need to determine the sophistication of theforecasting method, i.e., who is going to do the forecasting, and who is going to use theforecast will determine the sophistication. Once the sophistication level is determined,then the selection of the method, i.e., exponential smoothing, moving averages at thesimple end, at the more complex end, two-stage least squares, autoregressive models,etc. This, again, involves sophistication, but also the nature of the variable to beforecasted may have something to say about the method selected. Once themethodology is selected, predictor variables must be selected, tested, and incorporatedinto the model. Once the model is developed it will be subject to nearly constantevaluation, and re-specification. The efficacy, accuracy and usability of the results, andthe efficiency of the model in making the forecasts are the primary criteria used indetermining whether the model needs to be revised. Finally, forecasting reports willneed to be framed and circulated to the appropriate management officials for their use. These reports will also be subject to evaluation and re-specification.

10

In sum, the forecasting plan includes several, interrelated steps which are:

1. Selection of variables to be forecast2. Determining the level of forecast sophistication3. Specification of the model to be used to forecast4. Specification of the variables to be used in the model5. Re-evaluation of the model, using the results, their efficiency and efficacy6. Creating the reports, and evaluating the effectiveness of the reports

One should never lose track of the fact that a forecast will never be perfectlyaccurate, realistic goals for the forecasts must be set. If one can pick up the direction ofmoves in aggregate economic data, the models is probably pretty good. Arriving atexact magnitudes of the changes in economic aggregates is probably more a function ofluck than of good modeling in most cases. In other words, be realistic in evaluating theefficiency and efficacy of the forecasting models selected.

Conclusion

Business Conditions Analysis is focused on macroeconomic topics and howthese topics result in various aspects of the business environment in which the modernenterprise operates. It is not simply a macroeconomics course, nor is it simply a datacollection, modeling and inference course. E550 is the managerial aspects ofmacroeconomics, together with a quick and dirty introduction to analysis of themacroeconomy.

This course is far different than what the typical MBA course of the same titlewould have been twenty-five years ago. Twenty-five years ago this course would havefocused on a closed economy, with very little foreign sector influences, and an economythat was far more reliant on the goods producing sectors. Today, the U.S. economy isstill losing manufacturing, fewer people are employed in mining and agriculture, and it istruly an economy searching for what it wants to be. This is in some ways a moreexciting economy to study because it is at a cross-roads. The U.S. economy was thedominate economy when it was a manufacturing based system. At this cross-roads,there is the potential to re-double that economic dominance, a system full ofopportunities and challenges. On the other hand, there is always the path the Britishfollowed in the 1960s and since, towards less satisfactory results. In U.S. economyhistory we have had many mis-steps, but in general the long-term trend has beentoward ever-increasing prosperity. That is no guarantee, but it is certainly cause foroptimism, particularly if everyone masters the ideas that comprise this course.

11

KEY CONCEPTS

Economics

Microeconomics

Macroeconomics

Empirical economics v. Theoretical economics

Inductive logic v. Deductive logic

Model Building

Assumptions

Occam’s Razor

Normative economics v. Positive economics

Objective Thinking

Rationality

Fallacy of Composition

Cause and effect

Bias

Correlation v. causation

Cost-benefit analysis

Forecasting

Steps

Plan

Evaluation

12

STUDY GUIDE

Food for Thought:

Most people are biased in their thinking particularly concerning economic issues. Whydo you suppose this is?

Sample Questions:

Multiple Choice:

Which of the following is not an economic goal?

A. Full EmploymentB. Price StabilityC. Economic SecurityD. All of the above are economic goals

Which of the following methods can be applied to test for the existence of statisticalassociation between two variables?

A. CorrelationB. Granger causalityC. Theoretical modelingD. None of the above

True - false:

Non-economists are no less or no more biased about economics than physics orchemistry {FALSE}.

Assumptions are used to simplify the real world so that it may be rigorously analyzed{TRUE}.

13

Chapter 2

National Income Accounting

The aggregate performance of a large and complex economic system requiressome standards by which to measure that performance. Unfortunately our systems ofaccounting are imperfect and provide only rough guidelines, rather than crisp, clearmeasurements of the economic performance of large systems. As imperfect as thenational income accounting methods are, they are the best measures we have and theydo provide substantial useful information. The purpose of this chapter is to present themeasures we do have of aggregate economic performance.

Gross Domestic and Gross National Product

The most inclusive measures we have of aggregate economic activity are GrossDomestic Product and Gross National Product. These measures are used to describetotal output of the economy, by source. In the case of Gross Domestic Product we areconcerned with what is produced within our domestic economy. More precisely, GrossDomestic Product (GDP) is the total value of all goods and services producedwithin the borders of the United States (or country under analysis). On the otherhand, Gross National Product is concerned with American production (regardless ofwhether it was produced domestically). More precisely, Gross National Product(GNP) is the total value of all goods and services produced by Americansregardless of whether in the United States or overseas.

These measures (GDP and GNP) are the two most commonly discussed in thepopular press. The reason they garner such interest is that they measure all of theeconomy's output and are perhaps the least complicated of the national incomeaccounts. Often these data are presented as being overall measures of ourpopulation's economic well-being. There is some truth in the assertion that GDP andGNP are social welfare measures, however, there are significant limitations in suchinferences. To fully understand these limitations we must first understand how thesemeasures are constructed.

The national income accounts are constructed in such a manner as to avoid theproblem of double-counting. For example, if we count a finished automobile in thenational income accounts, what about the paint, steel, rubber, plastic, and othercomponents that go into making that car? To systematically eliminate double-counting,only value-added is counted for each firm in each industry. The value of the paint,

14

GROSS DOMESTIC PRODUCT by COMPONENT 1940-2000(billions of current U.S. dollars)

Personal Gross Domestic Government NetYEAR Consumption Investment Expenditures Exports GDP

1940 71.1 13.4 14.2 1.4 100.11950 192.1 55.1 32.6 0.7 286.71960 332.4 78.7 99.8 -1.7 513.41970 646.5 150.3 212.7 1.2 1010.71980 1748.1 467.6 507.1 -14.7 2708.01990 3742.6 802.6 1042.9 -74.4 5513.8 2000 6257.8 1772.9 1572.6 -399.1 9224.02005 8745.7 2105.0 2362.9 -726.5 12487.1

used in producing a car, is value-added by the paint manufacturing company, theapplication of that paint by an automobile worker is value-added by the car company(but the value of the paint itself is not). By focusing only on value-added at each step ofthe production process in each industry national income accountants are thus able toavoid the problems of double-counting.

The above box presents the GDP accounts in the major expenditurescomponents. GDP is the summation of personal consumption expenditures ©, grossdomestic private investment (Ig), government expenditures (G) and net exports (Xn),where net exports are total export minus total imports. Put in equation form:

GDP (Y) = C + Ig + G + Xn

GDP can also be calculated using the incomes approach. GDP can be found bysumming each of the income categories and deducting Net American Income EarnedAbroad. The following illustration shows how GNP and GDP are calculated using theincomes approach as follows:

15

__________________________________________________________________________________________________________________________________________

Depreciation +

Indirect Business Taxes +

Employee Compensation +

Rents +

Interest +

Proprietors' Income +

Corporate Income Taxes +

Dividends +

Undistributed Corporate Profits

= Gross National Product

- Net American Income Earned Abroad

= Gross Domestic Product

__________________________________________________________________________________________________________________________________________

In a practical sense it makes little difference which approach to calculating GDPis used, the same result will be obtained either way. What is of interest is theinformation that each approach provides. The sub-accounts under each approachprovide useful information for purposes of understanding the aggregate performance ofthe economy and potentially formulating economic policy. Under the expendituresapproach we have information concerning the amount of foreign trade, governmentexpenditures, personal consumption and investment.

The following accounts illustrates how GDP is broken down into another useful set of sub-accounts. Each of these additional sub-accounts provides information thathelps us gain a more complete understanding of the aggregate economic system. Thefollowing illustration demonstrates how the sub-accounts are calculated:

16

____________________________________________________________________________________________________________________________________________

Gross Domestic Product- Depreciation =

Net Domestic Product + Net American Income Earned Abroad

- Indirect Business Taxes =

National Income- Social Security Contributions- Corporate Income Taxes- Undistributed Corporate Profits+ Transfer Payments =

Personal Income- Personal Taxes =

Disposable Income_______________________________________________________________________________________________________________________________________________________________________________________________________________

The expenditures approach provides information concerning from what sectorproportions of GDP come. Personal consumption, government expenditures, foreignsector, and investment all are useful in determining what is responsible for oureconomic well-being. Likewise, the incomes approach provides greater detail to ourunderstanding of the aggregate economic output. Net National Product is the outputthat we still have after accounting for what is used-up in producing, in other words, thecapital we used-up getting GDP is netted-out to provide a measure of the output wehave left. National Income takes out of Net National Product all ad valorem taxes thatmust be paid during production and net American income originating from overseas. Appropriate adjustments are made to National Income to deduct those things that donot reach households (i.e., undistributed corporate profits) and adds in transferpayments to arrive at Personal Income. The amount of Personal Income thathouseholds are free to spend after paying their taxes is called Disposable Income.

So far the national income accounts appear to provide a great deal ofinformation. However, we do know that this information fails to accurately measure ouraggregate economic well-being. There are many aspects of economic activity that donot lend themselves well to standard accounting techniques and these problems mustbe examined to gain a full appreciation for what this information really means.

17

National Income Accounts as a Measure of Social Welfare

Accounting, whether it is financial, cost, corporate, nonprofit, public sector, oreven national income, provides images of transactions. The images that the accountingprocess provides has value judgments implicit within the practices and procedures ofthe accountants. National income accounting, as do other accounting practices, alsohas significant limitations in the availability of data and the cost of gathering data. Inturn, the costs of data gathering may also substantially influence the images that theaccounts portray.

GDP and GNP are nothing more than measures of total output (or income). However, the total output measured is limited to legitimate market activities. Further,national income accountants make no pretense to measure only positive contributionsto total output that occur through markets. Both economic goods and economic badsare included in the accounts, which significantly limits any inference that GDP or any ofits sub-accounts are accurate images of social welfare. More information is necessarybefore conclusions can be drawn concerning social welfare.

Nonmarket transactions such as household-provided services or barter are notincluded in GDP. In other words, the services of a cook if employed are counted, butthe services of a man or woman doing the cooking for their own household is not. Thismakes comparisons across time within the United States suspect. In the earliestdecades of national income accounting many of the more routine needs of thehousehold were served by the household members' own labor. As society becamefaster paced, and two wage earners began to become the rule for Americanhouseholds, more laundry, housecleaning, child-rearing, and maintenance worknecessary to maintain the household were accomplished by persons hired in themarketplace. In other words, the same level of service may have been provided, butmore of it is now a market activity, hence included in GNP. This is also the case incomparing U.S. households with households in less developed countries. Certainly lessmarket activity is in evidence in less developed countries that could be characterized ashousehold maintenance. Few people are hired outside of the family unit to performdomestic labor in less developed countries, and if they are they are typically paidpennies per hour. Less developed countries' populations rely predominately onsubsistence farming or fishing, and therefore even food and clothing may be rarelyobtained in the marketplace.

Leisure is an economic good but time away from work is not included in GNP. The only way leisure time could be included in GNP is to impute (estimate) a value forthe time and add it to GNP (the same method would be required for household servicesof family members). Because of the lack of consistency in the use of time for leisureactivities these imputation would be a very arbitrary, at best. However, commoditiesused in leisure activities are included in GNP. Such things as movie tickets, skis, and

18

other commodities are purchased in the market and may serve as a rough guide to thebenefits received by people having time away from work.

Product quality is not reflected in GNP. There is no pretense made by nationalincome accountants that GDP can account for product or service quality. There is alsolittle information available upon which to base a sound conclusions concerning whetherthe qualitative aspects of our total output has increased. It is clear that domesticautomobiles have increased in quality since 1980, and this same experience is likelytrue of most of U.S. industry.

No attempt is made in GDP data to account for the composition output. We mustlook at the contributions of each sector of the economy to determine composition. TheU.S. Department of Commerce publishes information concerning output and classifiesthat output by industry groups. These industry groupings are called Standard IndustrialCodes (S.I.C.) and permits relatively easy tracking of total output by industry group, andby components of industry groups.

Over time, there are new products introduced and older products disappear astechnology advances. Whale oil lamps and horse shoes gave way to electric lights andautomobiles between the Nineteenth and Twentieth Centuries. As we moved into thelatter part of this century vinyl records gave way to cassettes, which, in turn, have beenreplaced by compact disks. In almost every aspect of life the commodities that we usehave changed within our lifetimes. Therefore, comparisons of GNP in 1996 with GNP in1966 is really comparing apples and oranges because we did not have the sameproducts available in those two years.

As we move further back in time the commodities change even more. However,it is interesting to note the relative stability of the composition of output before theindustrial revolution. For centuries, after the fall of the Roman Empire, the compositionof total output was very similar. Attila the Hun would have recognized most of what wasavailable to Mohammed and he would have recognized most of what Da Vinci couldhave found in the market place. Therefore, the rapid change in available commoditiesis a function of the advancement of knowledge, hence the advancement in technology.

Another short-coming of national income accounting is that the accounts saynothing about how income is distributed. In the early centuries of this millennium, only aprivileged few had lifestyles that most of us would recognize as middle income orabove. With recent archaeological work at Imperial Roman sites, many scholars haveconcluded that over 95% of the population lived in poverty (the majority in life-threatening poverty), while a very few lived in extreme wealth. With the tremendousincreases in knowledge over the past two-hundred years, technology has increased ourproductivity so substantially that in the 28 industrialized nations of the world, themajority of people in those countries do not know poverty. However, the majority of theworld's population lives in less developed countries and the overwhelming majority ofthe people in those countries do know poverty, and a significant minority of these

19

people know life threatening poverty. In short, with the increase in output has come anincrease in the well-being of most people.

Per capita income is GDP divided by the population, and this is a very roughguide to individual income, which still fails to account for distribution. In the UnitedStates, the largest economy in the world, there are still over 40 million people (about14½ percent) that live in poverty, and only a very few these in life threatening poverty. Something just under four percent of the population (about 1 person in 26) are classifiedas wealthy. The other 81 percent experience a middle income lifestyle in the UnitedStates. The distribution of poverty is not equal across the population of this country. Poverty disproportionately falls to youngest and oldest segments of our population. Minority group persons also experience a higher proportion of poverty than do themajority.

Environmental problems are not addressed in the national income accounts. Thedamage done to the environment in the production or consumption of commodities isnot counted in GDP data. The image created by the accounts is that pollution,deforestation, chemical poisoning, and poor quality air and water that give rise tocancer, birth defects and other health problems are economic goods, not economicbads. The cost of the gasoline burned in a car that creates pollution is included in GNP,however, the poisoning of the air, especially in places like Los Angeles, Denver, andLouisville is not deducted as an economic bad. The only time these economic bads are accounted for in GNP is when market transactions occur to clean-up the damage, andthese transactions are added to GNP. The end result, is that GNP is overstated by theamount of environmental damage done as a result of pollution and environmentaldamage.

The largest understatement of GNP comes from something called theunderground economy. The underground economy is very substantial in mostless developed countries and in the United States. It includes all illegitimate(mostly illegal) economic activities, whether market activities or not. In lessdeveloped countries much of the underground economy is the "black market," but thereis also a significant amount of crime in many of these countries. Estimates aboundconcerning the actual size of the underground economy in the United States. Themiddle range of these estimates suggest the amount of underground economic activitiesmay be as much as one-third of total U.S. output.

The F.B.I. has, for years, tracked crime statistics in the United States andpublishes an annual report concerning crime. It is clear that drugs, organized theft,robberies, and other crimes against property are very substantial in the United States. But when these crimes result in income for the offenders, there is also the substantialproblem of income tax evasion from not reporting the income from the criminal activity. After all, Al Capone never went to jail for all of the overt criminal acts involved in hisvarious criminal enterprises, he went to jail for another crime, that is, because he didnot pay income taxes on his ill-gotten gains.

20

Drug trafficking in the United States is a very large business. The maximumestimates place this industry someplace in the order of a $500 billion per year businessin the U.S. Few legitimate industries are its equal. Worse yet, the news media reportsthat nearly half of those incarcerated in this nation's prisons are there on drug charges. The image that the national income accounts portrays is that the $100 billion, plus thatis spent on law enforcement and corrections because of drug trafficking is somehow aneconomic good, not a failure of our system. Drugs, however, are not the only problem. As almost any insurance company official can tell you, car theft is also another majorindustry. A couple of years ago CNN reported that a car theft ring operating in theSoutheast (and particularly Florida) was responsible for a large proportion of vehiclessold in certain Latin American countries. Further, that if this car theft ring were alegitimate business it would be the fourteenth largest in the United States (right aboveCoca-Cola in the Fortune 100).

In an economy with total output of $6 trillion, when nearly 10% of that is matchedby only one illegitimate industry - drugs - there is a serious undercounting problem. Ifestimates are anyplace close to correct, and $500 billion per year are the gross sales ofdrug dealers, and if the profits on this trade are only eighty percent (likely a lowestimate), and if the corporate income tax rate of forty-nine percent could be applied tothis sum, then instead of a $270 billion budget deficit, the Federal government would beexperiencing a surplus of something in the order of $130 billion, without any reduction inexpenditures for law enforcement and corrections (which could be re-allocated toeducation, health care or other good purposes). Maybe the best argument for thelegalization of drugs is its effect on the nation's finances (assuming, of course, drugswere only a national income accounting problem).

Price Indices

Changes in the price level poses some significant problems for national incomeaccountants. If we experience 10% inflation and a reduction of total output of 5% itwould appear that we had an increase in GNP. In fact, we had an increase in GNP, butonly in the current dollar value of that number. In real terms, we had a reduction inGNP. Comparisons between these two time periods means very little because the pricelevels were not the same. If we are to meaningfully compare output we must have amethod by which we can compare output with from one period to another by controllingfor changes in the price levels.

Nominal GDP is the value of total output, at the prices that exist at that time. Byadjusting aggregate economic data for variations in price levels then we have data thatcan be compared across time periods with different price levels. Real GDP is the valueof total output using constant prices (variations in price levels being removed).

Price indices are the way we attempt to measure inflation and adjust aggregate

21

economic data to account for price level variations. There are a wide array of priceindices. We measure the prices wholesalers must pay, that consumers must pay (eitherurban consumers (CPI(U) or that wage earners must pay (CPI(W)), we measure pricesfor all goods and services (GNP Deflator) and we also have indices that focus onparticular regions of the country, generally large urban areas, called StandardMetropolitan Statistical Areas -- S.M.S.A.).

Price indices are far from perfect measures of variations in prices. These indicesare based on surveys of prices of a specific market basket of goods, at a particular pointin time. The accuracy of any inference that may be drawn from these indices dependson how well the market basket of commodities used to construct the index match ourown expenditures (or the expenditures of the people upon whom the analysis focuses). Further complicating matters, is the fact that the market basket of goods changesperiodically as researchers believe consumption patterns change. Every five to tenyears (generally seven years) the Commerce Department (Current Population Surveys)changes the market basket of goods in an attempt to account for the currentexpenditure patterns of the group for which the index is constructed (total GNP,consumers, wholesalers, etc.).

For the consumer price indices, there is a standard set of assumptions used toguide the survey takers concerning what should be included in the market basket. Themarket basket for consumers assumes a family of four, with a male wage-earner, anadult female not employed outside of the home, and two children (one male, onefemale). There are also assumptions concerning home ownership, gift giving, diet, andmost aspects of the hypothetical family's standard of living.

The cost of living and the standard of living are mirror images of one another. Ifsomeone has a fixed income and there is a two percent inflation rate per year, then theirstandard of living will decrease two percent per year (assuming the index used is anaccurate description of their consumption patterns). In other words, a standard of living is eroded if there is inflation and no equal increase in wages (or other income, i.e.,pensions). Under the two percent annual inflation scenario, a household would need atwo percent increase in income each year simply to avoid a loss in purchasing power oftheir income (standard of living).

During most, if not all, of your lifetime this economy has experienced inflation. Prior to World War II, however, the majority of American economic history is marked bydeflation. That is, a general decrease in all prices. With a deflationary economy all onemust do to have a constant increase in their standard of living is to keep their incomeconstant while prices fall. However, deflation is a problem. Suppose you want to buy ahouse. Most of us have mortgages, we borrow to buy a house. If you purchase ahouse worth $50,000 and borrow eighty percent of the purchase price, $40,000 youmay have a problem. If we have five percent deflation per year, it only takes five yearsfor the market value of that house to reach $38689. In the sixth year, you owe more onyour thirty year mortgage than the market value of the house. Credit for consumer

22

purchases becomes an interesting problem in a deflationary economy.

On the other hand, if you owe a great deal of money, you have the opportunity topay back your loans with less valuable money the higher the rate of inflation. Therefore,debtors benefit from inflation if they have fixed rate loans that do not adjust the rates forthe effects of inflation.

The inflationary experience of the post-World War II period has resulted in ourexpecting prices to increase each year. Because we have come to anticipate inflation,our behaviors change. One of the most notable changes in our economic behavior hasbeen the wide adoption of escalator provisions in collective bargaining agreements,executory contracts, and in entitlement laws (social security, veterans' benefits, etc.). Escalator arrangements (sometimes called Cost of Living Adjustments, C.O.L.A.)typically tie earnings or other payments to the rate of inflation, but only proportionally. For example, the escalator contained in the General Motors and United Auto Workerscontract provides for employees receiving 1¢ per hour for each .2 the CPI increases. This protects approximately $5.00 of the employees earnings from the erosive effects ofinflation (.01/.2)100, assuming a base CPI of 100. (There is no escalator that providesa greater benefit to income receivers than the GM-UAW national agreement).

There are other price indices that focus on geographic differences. (Price datathat measures changes over time are called time series, and those that measuredifferences within a time period but across people or regions are called cross sections). The American Chamber of Commerce Research Association (ACCRA) in Indianapolisdoes a cross sectional survey, for mid-sized to large communities across the UnitedStates. On this ACCRA index Fort Wayne generally ranges between about 96.0 and102.0, where 100 is the national average and the error of estimate is between 2 and 4percent.

There are also producer and wholesale price indices and several component partof the consumer price indices that are designed for specific purposes that focus onregions of the country or industries. For example, the components of the CPI are alsobroken down so that we have detailed price information for health care costs, housingcosts, and energy costs among others.

Paashe v. Laspeyres Indices

There are two different methods of calculating price indices, these are theLaspeyres Index and the Paashe Index. The Laspeyres Index uses a “constant marketbasket”of goods and services to represent the quantity portion of the ration Q0 so thatonly price changes are in the index.

L = 3PnQ0 / 3P0Q0 X 100

This is the most widely used index in constructing price indices. It has the disadvantage

23

of not reflecting changes in the quantities of the commodities purchased over time. ThePaasche index overcomes this problem by permitting the quantities to vary over time,which requires more extensive data grubbing. The Passche index is:

P = 3PnQn / 3P0Qn X 100

Measuring the Price Level

The discussion here will focus on the Consumer Price Index (CPI) but isgenerally applicable. The CPI is based on a market basket of goods and is expressedas a percentage of the value of the market baskets' value in a base year (the year withwhich all prices in the index are compared). Each year's index is constructed bydividing the current year market basket's value by the base year market basket's valueand then multiplying the result by 100. Note that the index number for the base year willbe 100.00 (or 1 X 100).

By using this index we can convert nominal values into real values (real value areexpressed in base year dollars). We can either inflate or deflate current values to obtainreal values. Inflating is the adjustment of prices to a higher level, for years when theindex is less than 100. Deflating is the adjustment of prices to a lower level, for yearswhen the index is more than 100.

The process whereby we inflate and deflate is relatively simple andstraightforward. To change nominal into real values the following equation is used:

Nominal value/(price index/100)

For example, in 1989 the current base year the CPI is 100. By 1996 the CPI increasedto 110.0. If we want to know how much $1.00 of 1996 money is worth in 1989 we mustdeflate the 1996 dollar. We accomplish this by dividing 110 by 100 and obtaining 1.1;we then divided 1 by 1.1 and find .909 which is the value of a 1996 dollar in 1989. If wewant to know how much a 1989 dollar would buy in 1996 we must inflate. Weaccomplish this by dividing 100 by 110 and obtaining .909; we then divide 1 by .909 andfind 1.10 which is the value of a 1989 dollar in 1996.

Because the government changes base years it may be necessary to convertone or more indices with different base years to obtain a consistent time series if wewant to compare price levels between years decades apart. Changing base years is arelatively simple operation. If you wish to convert a 1982 base year index to beconsistent with a 1987 base year, then you use the index number for 1982 in the 1987series and divide all other observations for the 1982 series using the 1982 value in 1987index series. This results in a new index with 1987 as a base year. If inflation wasexperienced during the entire period then the index number for 1987 will be 100, for the

24

Cost of Living Adjustments

David A. Dilts and Clarence R. Deitsch, Labor Relations, New York: MacmillanPublishing Company, 1983, p. 167.

To the casual observer, COLA clauses may appear to be an excellentmethod of protecting the real earnings of employees. This is not totallyaccurate. COLA is not designed to protect the real wage of the employee butis simply to keep the employee's nominal wage, within certain limits, close toits original purchasing power. With a 1 cent adjustment per .4 increase in theCPI (if no ceiling is present) the base wage which is being protected from theerosive effect of inflation is $2.50 per hour, 1 cent per .3 increase in the CPIprotects $3.33 per hour, and 1 cent per .2 increase in the CPI protects $5.00per hour. This is quite easy to see; since the CPI is an index numbercomputed against some base year (CPI = 100 in 1967) and the adjustmentfactor normally required in escalator clauses is 1 cent per some increase x, inthe CPI, the real wage which is protected by the escalator is the inverse of theCPI requirement or 1/x.

years prior to 1987 the indices will be less than 100 and for the years after 1987 thenumbers will be larger than 100.

The price index method has problems. The assumptions concerning the marketbasket of goods to be surveyed causes specific results that are not descriptive of ageneral population. In the case of the consumer price index, families without children orwith more than two may find their cost of living differs from what the index suggests. Ifboth parents work, the indices may understand the cost of living. For families with tenyear old, fixed rate mortgages and high current mortgages rates, the CPI mayunderstate their current cost of living. Therefore, the CPI is only a rough measure, andits applicability differs from household to household.

The items which go into the market basket of goods to construct the price indicesand the proper inference which can be drawn from these data are published in severalsources. Perhaps the easiest accessible location of this information is at the Bureau ofLabor Statistics site www.bls.gov which also publishes a wealth of informationconcerning employment, output, and prices.

The following boxes provide some recent information of the type to be found atthe Bureau of Labor Statistics site. The first box provides annual data for consumerprices for both All Urban Consumers (CPI-U), and Urban Wage Earners and ClericalEmployees (CPI-W).

25

Notice that the trend of prices is upwards, and that the CPI (W) is always slightlylarger than the CPI (U). The difference between the two series is growing in 1996, thedifference was about 1.78% of CPI (U), whereby in 2005 the difference was 2.2 % ofCPI (U).

The market basket of goods and services that go into calculating the consumerprice index has been the subject of substantial controversy. For inference to beproperly drawn concerning the underlying prices in the CPI one must have some idea ofwhat goes into to the market basket and whether that mix of goods and service is of anyparticular significance for a consumer. The general categories of expenditures byconsumers which are represented by the CPI (U) are presented in the following table. The table presents the categories of expenditure, rather than each specific item that isincluded in the market basket. This category of expenditure is for the base period 1982-84 and is supposed to be representative of what the average urban consumerpurchased during this time frame.

Consider the consumer expenditure categories used in calculating the CPI (U)which are presented in the following table:

Consumer Prices 1996-2005, Annual Averages

Year CPI-U CPI-W Difference(U-W)1996 156.9 154.1 2.81997 160.5 157.6 2.91998 163.0 159.7 3.31999 166.6 163.2 3.42000 172.2 168.9 3.32001 177.1 173.5 3.62002 179.9 175.9 4.02003 184.0 179.8 4.22004 188.9 184.5 4.42005 195.3 191.0 4.3

Not seasonally adjusted1982-84 = 100.0Source: U.S. Department of Labor, Bureau of Labor Statistics

26

This is an exercise in positive economics. These are NOT the percentages ofincomes that consumers spend. These are the percentages of incomes assumed to bespend by consumers in constructing the CPI (U). It is clear that there are valuejudgments and biases included in these data, but is not so clear is whether the data isupwardly or downwardly biased. With the increases in health care costs and recentincreases in oil prices it is very likely that, in general, the CPI (U) is downwardly biased. However, there are those who argued that the CPI (U) is upwardly biased, and shouldbe adjusted downward. These economists, however, are those who also argue thatindexing government transfer payments and social security is a bad idea. Where thetruth actually lies is an empirical question.

CPI (U) Expenditure Categories 1982-84

Category Relative Importance Dec. 2005

Food 13.942Alcoholic Beverage 1.109

Shelter 32.260Fuels & Utilities 5.371Furnishings & Operations 4.749

Apparel 3.786

Transportation 17.415

Medical Care 6.220

Recreation 5.637

Education 2.967

Communication 3.080

Other goods & services 3.463___________________________

All Categories 99.999

Source: Bureau of Labor Statistics

27

Other Price Level Aggregates

The key price level measures are published data from the CommerceDepartment. Some simple relationship have been demonstrated in the economicliterature which are worth examining here. The consumer price index is divided intocore components and also include things which are more volatile, these items are foodand energy. Housing, clothing, entertainment, etc. are less volatile are often referred toas the core elements of the consumer price index.

Theoretically it is presumed that the Wholesale Price Index, an index of itemsthat wholesalers provide to retailers, and the Producer Price Index, an index ofintermediate goods are leading indicators of prices at the consumer level. In fact, thesetwo price indices, over the broad sweep of American economic history, have been prettyreasonable indicators of what to expect from consumer prices.

The Producer Price Index is a complex, but useful collection of numbers. ThePPI is divided into general categories, industry and commodity. The major groupingsunder each of these broad categories have a series of price data which has beencollected monthly for several years. The following two tables present simple examplesof the types of PPI data that are available. The first table presents industry data, thesecond box presents commodity data. There are also series on such specializedpricing information as import and export prices, and these are all available on thewww.bls.gov site.

Industry PPI data, 1996-2005

Petroleum Book Auto & Light General Year Refineries Publishers Vehicle Mfg Hospitals

1996 85.3 224.7 140.4 112.51997 83.1 232.1 137.8 113.61998 62.3 238.8 136.8 114.61999 73.6 247.6 137.6 116.62000 111.6 255.0 138.7 119.82001 103.1 263.3 137.6 123.42002 96.3 273.0 134.9 127.92003 121.2 283.0 135.1 135.32004 151.5 293.7 136.5 141.92005 205.3 305.4 135.1 147.3%Δ1996-05 140.7 35.9 -3.9 30.9

Source: Bureau of Labor Statistics

28

As can be easily seen from the above table, there is wide variation in PPI byindustry. Oil refineries’ costs of operation have gone up more than 140% over theperiod, while the costs for automobile and light vehicle manufacturers’ have actuallydeclined over the period by nearly 4 percent. In keeping with the high costs of healthcare, hospitals’ costs of operation have increased nearly as much as the costs to bookpublishers over the period. Remember, these are industry cost numbers. The followingtable present commodity cost statistics in the PPI series.

As can readily be seen in comparing the two series, there is a reason why oilrefineries costs of operations increased. The price of crude oil increase by just aboutwhat the costs to the refiners increased. However, book publishers need not look to theprice of paper to explain their cost increases. In sum, the costs of commodities went upover the nine year period by a rather small 23.3 percent or just over two and one-halfpercent.

Monetary aggregates have also been used as a rough gauge of what to expect inthe behavior price variables over time. Later in the course there will be substantialdiscussion of the Federal Reserve and monetary economics, but it is interesting to notehere that monetary aggregates have both a practical and theoretical relationship toaggregate price information in the economy. Monetary aggregates generally havesignificant influence over interest rates, which, in turn, will have both aggregate supply

Commodity PPI data, 1996-2005

All Petroleum Iron andYear Commodities Crude Paper Steel

1996 127.7 62.6 149.4 125.81997 127.6 57.5 143.9 126.51998 124.4 35.7 145.4 122.51999 125.5 50.3 141.8 114.02000 132.7 85.2 149.8 116.62001 134.2 69.2 150.6 109.72002 131.1 67.9 144.7 114.12003 138.1 83.0 146.1 121.52004 146.7 108.2 149.4 162.42005 157.4 150.1 159.6 171.1%Δ1996-05 23.3 139.8 6.8 36.0

Source: Bureau of Labor Statistics

29

and aggregate demand implications. A presentation and discussion of these economicstatistics is reserved for the monetary economics chapters in this course.

Employment and Output

Unemployment and employment data have implications for the business cycle,as will be discussed in detail later. However, it is important to note here thatemployment statistics are generally a lagging indicator, that is, they trail (or confirm)what is going on in the aggregate economy.

Forecasting can be as much a game of indications, more than precision. If wesee industrial output is on the rise, we can expect that unemployment rates will declineonce a trough in the business cycle has been reached. On the other hand, if industrialoutput is declining, we can expect future unemployment rates to be on the rise. Thispresumes that there has not been prolonged recession or structural changes in theeconomy. When there are structural changes or prolonged recession, then suchphenomenon as the discouraged worker hypothesis may operate to make theunemployment rates move in directions that would be otherwise inconsistent with whathad been recently observed in the aggregate economy.

In other words, even though the unemployment and employment statistics have agenerally predictable relationship with output, you can get fooled sometimes because ofunderlying phenomenon in the economy. These theoretical relationships are onlyindicative and can sometimes be overwhelmed by other economic pressures in theeconomy.

A discussion of the various economic statistics relevant to employment andoutput will be reserved for the following chapter where more discussion of the conceptswhich underpin these data are discussed.

KEY CONCEPTS

National Income Accounting

Social Welfare

Under-estimations

Over-estimations

Gross Domestic Product v. Gross National Product

30

Value added

Expenditures Approach

Income Approach

Criticisms

Net Domestic Product v. Net National Product

Depreciation

National Income

Personal Income

Disposable Income

Inflation v. Deflation

Cost-Push Inflation v. Demand-Pull Inflation

Pure Inflation

Monetarist School

Quantity Theory of Money

Price Indices

CPI

PPI

Other Indices

Inflating v. Deflating

Paasche v. Laspeyres

Price aggregates

CPI and WPI and PPI relations

Monetary aggregates

31

Employment and Unemployment

Relations with output

Indicators, not hard and fast rules

STUDY GUIDE

Food for Thought:

Critically evaluate the use of national income accounts as measures of social welfare.

Using the following data construct the price indices indicated:

Year Market basket $

1989 3501990 4001991 4401992 4651993 500

Calculate a price index using 1989 as the base year

Calculate a price index using 1991 as the base year

If the nominal price of new house in 1993 is $100,000, how much is the house in 1989dollars? In 1991 dollars?

32

If the price of a new house in 1989 is $90,000 what is the price in 1991 dollars?

Critically evaluate the use of price indices for comparison purpose across time.

Using the following data calculate GDP, NDP, NI, PI, and DI

Undistributed Corporate profits $40Personal consumption expenditures $1345Compensation of employees $841Interest $142Gross exports $55Indirect business taxes $90Government expenditures $560Rents $115Personal taxes $500Gross imports $75Proprietors income $460Depreciation $80Corporate income taxes $100Net Investment $120Dividends $222Net American income earned abroad $5Social security contributions $70

33

Sample Questions:

Multiple Choice:

If U.S. corporations paid out all of their undistributed corporate profits as dividends totheir stockholders then which of the following national income accountants would showan increase?

A. Gross Domestic ProductB. Net Domestic ProductC. Personal IncomeD. National Income

The following are costs of market baskets of goods and services for the years indicated:

1900 $1001910 $1021920 $1051930 $901940 $1001950 $1101960 $160

Using 1920 as a base year what is the price index for 1900 and for 1960?

A. 1900 is 105, 1960 is 160B. 1900 is 95.2, 1960 is 152.4C. 1900 is 111.1, 1960 is 177.8D. Cannot tell from the information given

True - False:

The GDP is overstated because of the relatively large amount of economic activity thatoccurs in the underground economy {TRUE}.

The difference between Personal Income and Disposable Income is personal taxes{TRUE}.

34

Chapter 3

Unemployment and Inflation

When one thinks of business conditions, the first thing to come to mind, typically,is the recurrent cycles that the economy repeats with near regularity. Business cyclesare a fact of economic life with a market based economy. Recessions are generallyfollowed by recoveries which give a sort of up and down vibration to the long-termsecular growth trend in the U.S. economy. The way we track this vibration is throughthe two major economic phenomenon that are used to track business cycles, these areemployment and price level stability. The purpose of this chapter is to examine two ofthe most important and recurrent economic problems that have characterized moderneconomic history throughout the industrialized world, including the United States. These two problems are unemployment (associated with recessions) and inflation(associated with the lose of purchasing power of our incomes). Most economic policyfocuses on mitigating these, most serious, of problems in the macroeconomy.

Mixed Economic System and Standard of Living

American capitalism is a mixed economic system. There are small elements of command and tradition, and some socialism. However, our economic system ispredominately capitalist. The economic freedom and ability to pursue our individualself-interest provides for the American people a standard of living, in the main, that isunprecedented in world history. Perhaps more important, our high standards of livingare widely shared throughout American society (with fewer than 17.5% of Americansliving in poverty).

The accomplishments of American society ought not be taken lightly, no otherepoch and no other nation, has seen a "golden" age as impressive as modern America. However, there are aspects of our freedom of enterprise that are not so positive. Freemarket systems have a troubling propensity to experience recessions (at the extremedepressions) periodically. As our freedom of inquiry develops new knowledge, newproducts and new technologies, our freedom of enterprise also results in theabandonment of old industries (generally in favor of new industries). At times we alsoseem to lose faith in accelerating rates of growth or economic progress. At other times,we have experienced little growth in incomes (at the extreme declines in consumerincomes). All of these problems have resulted in down-turns in economic activity. Atother times, consumer's income have increased at accelerating rates, people havebecome enthusiastic about our economic future, and the growth of new industries have

35

far outpaced the loss of old industries that have resulted in substantial expansions ofeconomic activity. Together these down-turns and expansions are referred to as thebusiness cycle.

Business Cycles

The business cycle is the recurrent ups and downs in economic activityobserved in market economies. Troughs, in the business cycle, are whereemployment and output bottom-out during a recession (downturn) or depression(serious recession). Peaks, in the business cycle, are where employment and outputtop-out during a recovery or expansionary period (upturn). These ups and downs(peaks and troughs) are generally short-run variations in economic activity. It isrelatively rare for a recession to last more than several months, two or three yearsmaximum. The Great Depression of the 1920s and 1930s was a rare exception. Infact, the 1981-85 recession was unusually long.

One of the most confusing aspects of the business cycle is the differencebetween a recession and depression. For the most part, recessionary trends aremarked by a downturn in output. This downturn in output is associated with increasedlevels of unemployment. Therefore, unemployment is what is typically keyed upon infollowing the course of a recession. In 1934, the U.S. economy experienced 24.9%unemployment, this is clearly a depression. The recession of 1958-61 reached only6.7% unemployment. This level of reduced economic activity is clearly only arecession. However, in the 1981 through 1986 downturn the unemployment ratereached a high of 12%, and in both 1982 and 1983 the annual average unemploymentrate was 10.7%. Probably the Reagan recession was close to, if not actually, a shortdepression, arguably a deep recession. This 1981-85 period was clearly the worstperforming economy since World War II, but it also was clearly nothing compared to theproblems in the decade before World War II.

The old story about the difference between a recession and depression probablyis as close to describing the difference between a recession and a depression asanything an economist can offer. That is, a recession is when your neighbor is out ofwork, a depression is when you are out of work!

In general, the peaks and troughs associated with the business cycle, are short-run variations around a long-term secular trend. Secular trends are generalmovements in a particular direction that are observed for decades (at least 25years in macroeconomic analyses).

Prior to World War II the secular trend started as relatively flat and limited growth

36

period and then it took a sharp downward direction until the beginnings of the War inEurope (a period of about twenty years). Since the end of World War II we haveexperienced a long period of rather impressive economic growth (a period of over fiftyyears).

The following diagram shows a long-term secular trend that is substantiallypositive (the straight, upward sloping line). About that secular trend is another curvedline, whose slope varies between positive and negative, this is much the same as thebusiness cycle variations about that long-term growth trend. If we map out economicactivity since World War II we would observe a positive long-term trend, with markedups and downs showing the effects of the business cycle.

There are other variations observed in macroeconomic data. These variations,called seasonal variations, last only weeks and are associated with the seasonsof the year. During the summer, unemployment generally increases due to studentsand teachers not having school in the summer and both groups seeking employmentduring the summer months. Throughout most of the Midwest agriculture andconstruction tend to be seasonal. Crops are harvested in the fall, then in the wintermonths farmers either focus on livestock production or wait for the next grain season. In the upper Midwest, north of Fort Wayne, outside work is very limited due to extremeweather conditions, making construction exhibit a seasonal trend. In the retail industry,from Thanksgiving through New Year's Day is when disproportionately large amounts ofbusiness are observed, with smaller amounts during the summer.

Most series of data published by the Commerce Department or the Bureau ofLabor Statistics concerning prices and employment have two different series. There willtypically be data which does not account for predictable seasonal variations (notseasonally adjusted) and those data which have had the seasonal variations removed(seasonally adjusted). From a practical standpoint, the raw numbers capture whatactually happens and may be useful for forecasts and analyses where you wish to have

37

the actual results. On the other hand, knowing what the seasonal variations are permitthe analyst to capture affects with their models that are from more fundamentallyeconomic or political processes. Therefore, both series of data are of significance to theeconomics profession.

Unemployment

Unemployment is defined to be an individual worker who is not gainfullyemployed, is willing and able to work, and is actively seeking employment. Aperson who is not gainfully employed, but is not seeking employment or who is unwillingor unable to work is not counted as unemployed, or as a member of the work force. During the Vietnam War unemployment dropped to 3.6% in 1968 and 3.5% in 1969. This period illustrates two ways in which unemployment can be reduced. Because ofthe economic expansion of the Vietnam era more jobs were available, but during thesame period many people dropped out of the work force who may otherwise have beenunemployment or, alternatively, vacated jobs that became available to others to avoidmilitary service. One way to keep from being drafted was to become a full time student,which induced many draft-age persons to go to college rather than risk military servicein Vietnam. Additionally, there was a substantial expansion in the manpower needs ofthe military with nearly 500,000 troops in Vietnam in 1969. Therefore, theunemployment rate was compressed between more job, and fewer labor forceparticipants.

Unemployment can decrease because more jobs become available. It can alsodecrease because the work force participation of individuals declines, in favor ofadditional schooling, military service, or leisure. Unemployment is more than idleresources, unemployment also means that some households are also experiencingreduced income. In other words, unemployment is associated with a current loss of output, and reductions in income into the foreseeable future.

Economists classify unemployment into three category by cause. These threecategories of unemployment are (1) frictional, (2) structural, and (3) cyclical. Frictionalunemployment consists of search and wait unemployment which is caused by peoplesearching for employment or waiting to take a job in the near future. Structuralunemployment is caused by a change in composition of output, changes in technology,or a change in the structure of demand (new industries replacing the old). Cyclicalunemployment is due to recessions, (the downturns in the business cycle). Structuralunemployment is associated with the permanent loss of jobs, however, cyclicalunemployment is generally associated with only temporary losses of employmentopportunities.

Full employment is not zero unemployment, the full employment rate of

38

unemployment is the same as the natural rate. The natural rate of unemployment isthought to be about 4% and is a portion of structural unemployment and frictionalunemployment. However, there is not complete professional agreement concerning thenatural rate, some economists argue that the natural rate, today is, about 5%. Thedisagreement centers more on observation of the secular trend, than any particulartechnical aspect of the economy (and there are those in the profession who woulddisagree with this latter statement).

The reason that frictional and structural unemployment will always be observed isthat our macroeconomy is dynamic. There are always people entering and leaving thelabor force, each year there are new high school and college graduates and secondarywage earners who enter and leave the market. There is also a certain proportion ofstructural unemployment that should be observed in a healthy economy. Innovationsresult in new products and better production processes that will result in displacement ofold products and production processes that results employees becoming unnecessaryto staff the displaced and less efficient technology. Therefore, the structural componentof the natural rate is only a fraction of the total structural rate in periods where there isdisplacement of older industries that may result from other than normal economicprogress. Perhaps the best example of this is the displacement of portions of thedomestic steel and automobile industries that resulted from predatory trade practices(i.e., some of the dumping practices of Japan and others).

The level of output associated with full utilization of our productiveresources, in an efficient manner is called potential output. Potential output is theoutput of the economy associated with full employment. It is the level of employmentand output associated with being someplace on the production possibilities curve (fromE201). This level of production will become important to us in judging the performanceof the economy.

Full employment is not zero unemployment and potential GNP is not totalcapacity utilization (full production), such levels of production are destructive to the laborforce and capital base because people fulfill other roles (i.e., consumer, parent, etc.)and capital must be maintained. The Nazi's slave labor camps (during World War II)were examples of the evils of full production, where people were actually worked todeath.

Unemployment rates

The unemployment rate is the percentage of the work force that isunemployed. The work force is about half of the total population. Retired persons,children, those who are either incapable of working or those who choose not toparticipate in the labor market are not counted in the labor force. Another way to look

39

at it, is that the labor force consists of those persons who are employed or unemployedwho are willing, able and searching for work.

People who are employed may be either full time or part time employees. Inaggregate the average number of hours worked by employees in the U.S. economygenerally is something just under forty-hours per week (generally between 38 and 39hours per week). This statistic reflects the fact that people have vacation time, areabsent from work, and may have short periods of less than forty hours available to themdue to strikes, inventories, or plant shut-downs.

Part time employees are included in the work force. You are not counted asunemployed unless you do not work and are actively pursuing work. Those who do nothave 40 hours of work (or the equivalent) available to them are classified as part timeemployees. At present there are about 6.5 million U.S. workers were are involuntarilypart-time workers, and about 12 million were voluntarily part-time employees, this is upabout 3 million from total part time employment in 1982.

The unemployment rate is calculated as:

UR = Unemployed persons ÷ labor force

There are problems with the interpretation of the unemployment rate. Discouraged workers are those persons who dropped out of labor force because theycould not find an acceptable job (generally after a prolonged search). To the extentthere are discouraged workers that have dropped out of the work force, theunemployment rate is understated. There are also those individuals who have recentlylost jobs who are not interested in working, but do not wish to lose their unemploymentbenefits. These individuals will typically go through the motions of seeking employmentto remain eligible for unemployment benefits but will not accept employment or makeany effort beyond the appearance of searching. This is called false search and servesto overstate the unemployment rate.

There has yet to be any conclusive research that demonstrates whether thediscouraged worker or false search problem has the greatest impact on theunemployment rate. However, what evidence exists suggests that in recent years thediscouraged worker problem is the larger of the two problem, suggesting that theunemployment rate may be slightly understated.

The following box provides a snapshot of available economic data concerningemployment aggregates:

40

Okun's Law

As mentioned earlier, unemployment is not just a single dimensional problem. Based on empirical observation an economist determined that there was a fairly stablerelation between unemployment and lost output in the macroeconomy. This relationhas a theoretical basis. As we move away from an economy in full employment, non-inflationary equilibrium, we find that we lose jobs in a fairly constant ratio to the loss ofoutput. Okun's Law states that for each one percent (1%) the unemployment rateexceeds the natural rate of unemployment (4%) there will be a gap of two and one-halfpercent (2.5%) between actual GDP and potential GDP. In other words, employment isa highly correlated with GDP. This is why it is not technically incorrect to look to theunemployment rate to determine whether a recession has begun or stopped. It is alsotrue that unemployment tends to trail behind total output of the economy. In otherwords, unemployment is a trailing indicator. As the economy recovers we could still seeworsening unemployment, but after a lag of a quarter or so, the improving economicconditions will result in less unemployment.

This relationship permits some rough guesses about what is happening to totaloutput in the economy, however, it is only a rough guide, because unemployment is atrailing indicator. In other words, as the economy goes into recession that last variableto reflect the loss of output is the unemployment rate. Okun’s Law, as it turns out, isalso very useful in formulating policy within the Keynesian framework which will be

Labor Market Data, 1996-2006 (seasonally adjusted January)

Civilian Labor Employment Unemployment UnemploymentYear Force (1000s) Level (1000s) Level (1000s) Rate (%)

1996 132616 125125 7491 5.61997 135456 128298 7158 5.31998 137095 130726 6368 4.61999 139003 133027 5976 4.32000 142267 136559 5708 4.02001 143800 137778 6023 4.22002 143883 136442 8184 5.72003 145937 137424 8513 5.82004 146817 138472 8345 5.72005 147956 140234 7723 5.22006 150114 143074 7040 4.7

Source: Bureau of Labor Statistics

41

developed later in the course.

Burden of Unemployment

The individual burden of unemployment is not uniformly spread across thevarious groups that comprise our society. There are several factors that have beenhistorically associated with who bears what proportion of the aggregate levels ofunemployment. Among the factors that determine the burden of unemployment areoccupation, age, race and gender.

The individual occupational choice will effect the likelihood of becomingunemployed. Those with low skill and educational levels will generally experienceunemployment more frequently than those with more skills or education. There are alsospecific occupations (even highly skilled or highly educated) that may experienceunemployment due to structural changes in the economy. For example, with the declinein certain heavy manufacturing many skilled-trades persons experienced bouts ofunemployment during the 1980s. As educational resources declined in the 1970s andagain recently, many persons with a Ph.D. level education and certified teachersexperienced unemployment. However, unemployment for skilled or highly educatedoccupations tends to be infrequent and of relatively short duration.

Age also plays a role. Younger people tend to experience more frictionalunemployment than their older, more experienced counterparts. As people enter thework force for the first time their initial entry puts them into the unemployed category. Younger persons also tend to experience a longer duration of unemployment. However, there is some evidence that age discrimination may present a problem forolder workers (the Age Discrimination Act covers those persons over 40, and it appearsthose over 50 experience the greatest burden of this discrimination).

Race and gender, unfortunately, are still important determinants of bothincidence and duration of unemployment. Most frequently the race and gender effectsare the result of unlawful discrimination in the labor market. There is also a body ofevidence that suggests there may be significant discrimination in the educationalopportunities available to minorities.

Edmund Phelps developed a theory called statistical theory of racism andsexism that sought to explain how discrimination could be eliminated as a determinateof the burden of unemployment. His theory was that if there was not a ideologicalcommitment to racism or sexism, that if employers were forced to sample fromminorities they would find that there was no difference in these employees' productivityand the productivity of the majority. This formed the basis of affirmative actionprograms. The lack of effectiveness of most of these programs suggests that theracism and sexism that exists is ideological and requires stronger action, than simple

42

reliance on economic rationality and sampling.

Inflation

The news media reports inflation, generally, as increases in the CPI. This is nottechnically accurate. Inflation is defined as a general increase in all prices (theprice level). The CPI does not purport to measure all prices, wholesale prices andproducer prices are not included in the consumer data. The closest we have to ameasure of inflation is the GNP deflator that measures prices for the broadest range ofgoods and services, but even this broader index is not a perfect measure, but its all wehave and some information (particularly when we know the short-comings) is better thanperfect ignorance.

One of the more interesting bits of trivia concerning inflation is something calledthe Rule of 70. The rule of 70 gives a short-hand method of determining how long ittakes for the price level to double at current inflation rates. It states that the number ofyears for the price level to double is equal to seventy divided by the annual rate ofincrease (i.e., 70/%annual rate of increase(expressed as a whole number)). Forexample, with ten percent inflation, the price level will double every seven years (70/10= 7).

There are three theories of inflation that arise from the real conduct of themarcoeconomy. These three theories are demand-pull, cost-push, and pure inflation. There is a fourth theory that suggests that inflation has little or nothing to do with thereal output of the economy, this is called the quantity theory of money. Each of thesetheories will be reviewed in the remaining sections of this chapter.

Demand - Pull Inflation

Using a naive aggregate supply/aggregate demand model, we can illustrate thetheory of demand-pull inflation. The following chapter will develop a more sophisticatedaggregate supply/aggregate demand model, but for present purposes the naive modelwill suffice. The naive model has a linear supply curve and a linear demand curve,much the same as the competitive industry model developed in the principles ofmicreconomics course (or in A524). However, the price variable here is not the price ofa commodity, it is the price level (the CPI for want of a better measure) and the quantityhere is the total output of the economy, not some number of widgets.

43

Using a naive aggregate demand\ aggregate supply model, as the aggregatedemand shifts to the right or increases, all prices increase. This increase in all prices iscalled inflation. However, this increase in aggregate-demand is also associated with anincrease in total output. Total output is associated with employment (remember Okun'sLaw?). In other words, even though this increase in aggregate demand causesinflation, it does not result in lost output, hence unemployment. Policy measuresdesigned to control demand-pull inflation, will shift the aggregate demand curve to theleft, (i.e., reduce aggregate demand) and this reduction in aggregate demand isassociated with loss of output, hence increased unemployment.

Demand-pull inflation is therefore only a problem with respect to price levelinstability. On the other hand, there is another model of inflation which createsproblems in both price level instability and unemployment, and that model is calledCost-Push Inflation.

Cost - Push Inflation

Again using a naive aggregate supply/aggregate demand approach, cost-pushinflation results from particular changes in the real activity in the macroeconomy. In thiscase, a decrease in the aggregate supply curve. A decrease in aggregate supply canoccur for several reasons. Government regulations which divert resources fromproductive uses, increases in costs of intermediate goods and factors of production, advalorem taxes, and sources of inefficiency (excess executive salaries etc.) can all resultin a leftward shift in the aggregate supply curve.

The following diagram shows a shift to the left or decrease in aggregate supplycurve.

44

The OPEC Oil embargo may present the best case for cost-push inflation. As oilprices doubled and then tripled, the costs of production that were comprised at least inpart from oil also dramatically increased. Therefore, the dramatic increase in the priceof oil shifted the aggregate supply curve to the left (a decrease) , resulting in cost-pushinflation.

The general case is, as the price of any productive input increases, theaggregate supply curve will shift to the left. This decrease in aggregate supply alsoresults in reduced output, hence unemployment. This is consistent with the economicexperience of the early 1980s during the Reagan Administration when the economyexperienced high rates of both inflation and unemployment.

It is also interesting to note that the experiences of the Nixon-Ford administrationand later the Carter administrations were not unlike that of Reagan’s era. In the Nixon-Ford administration the 1973 Mid-East war resulted in very high oil prices, and monetarypolicies that resulted in rapidly increasing interest rates which together created whateconomists labeled Stagflation – high rates of inflation together with high rates ofunemployment. Again, tracing the causes of this back to the origin results in aconclusion that the Cost-Push model provided an interesting, albeit, over-simplifiedexplanation of the Nixon-Ford experience.

Jimmy Carter fared no better. The double digit inflation and unemployment of theearly 1970s was repeated at the end of the decade. Even though Carter’s move toderegulate the airlines industry and the trucking industry may have had long-termpositive results for the economy, foreign policy was about to intervene and createessentially the same havoc as plagued the previous regime. In 1979 the Iranianrevolution and the Russian invasion of Afghanistan created a foreign policy disaster inthe U.S. At the time of the Iranian crises, Iran was one this country’s largest tradingpartner. With the loss of this source of foreign demand for U.S. goods, and the second

45

largest source of foreign oil, the shock resulted in returning to the same double digitunemployment and inflation that swept the Republicans out of office just three yearsearlier – and again, the experience was not unlike the results predicted by Cost-PushInflation models.

Pure Inflation

Pure inflation results from an increase in aggregate demand and a simultaneousdecrease in aggregate supply. For output to remain unaffected by these shift inaggregate demand and aggregate supply, then the increase in aggregate demand mustbe exactly offset by an equal decrease in aggregate supply.

The following aggregate supply/aggregate demand diagram illustrates the theoryof pure inflation:

Notice in this diagram, that aggregate supply shifted to the left, or decreased byexactly the same amount that the aggregate demand curve shifted to the right orincreased. The result is that output remains exactly the same, but the price levelincreased. Intuitively, this makes sense, with the loss of aggregate supply we wouldexpect an increase in prices, and with an increase in aggregate demand we would alsoexpect an increase in prices. As aggregate supply and aggregate demand move inopposite directions, it is not perfectly clear what happens to output. In this case, withequal changes in aggregate demand and supply output should remain exactly thesame.

Put the two models together, and it is a roll of the dice whether the economygrows or declines as the price level runs amuck. If aggregate supply declines more

46

severely than aggregate demand increases, you will get stagflation. On the other, handeconomic growth, oddly enough, is still possible if aggregate demand outpaces thedeclines in aggregate supply – an interesting quandary to say the least.

Quantity Theory of Money

The models of cost-push, demand-pull and pure inflation presented above, arerather naive simplifications. These models suggest that there are causes of inflationthat are generated by the “real” rather than the monetary economy. In other words, it ispossible to create inflation through real shocks to the economy (Real Business CycleTheory from Tom Sargent and the Minnesota School). In fact, there may be pressuresthat can influence both policy makers, and have implications for inflation if monetaryauthorities make mis-steps.

The Monetarist School of economic thought points to another possibleexplanation for inflation. These economists do not reject the idea that inflationarypressures can occur because of an oil embargo or increases in consumer demand. However, these economists argue that inflation cannot occur simply as a result of theseevents. They are quick to point out that a change in a single price in the price indexmarket basket can give the appearances of inflation, when all that happened was achange in the relative price of one commodity with respect to all others.

More of this theory will be discussed in the final weeks of the semester, however,one point is necessary here. Inflation, in the monetarist view, can only occur if themoney supply is increased which permits all prices to increase. If the money supply isnot increased there can be changes in relative prices, for example, oil prices can go up,but there has to be offsetting decreases in the prices of other commodities. An increasein all prices or in the price of a particular good, therefore, is a failure of the Fed toappropriately manage the money supply.

As oil prices rose from $15 a barrel to over $70 a barrel we would have expectedto experience very significant inflation if the histories of the Nixon-Ford, and Carteradministrations were instructive. However, so far at least, these histories have not beenrepeated. What is interesting to note is that the price level has remained remarkablystable in the face of the quadrupling of crude oil prices. There are two explanations forthis observation – together they illustrate the point of inflation and monetary economics.

The explanation is rather straightforward and simple. After 9/11/01 the FederalReserve drove the discount rate down to 1%. The result was that this extremely easymonetary policy prevented what should have been a serious recession. Theunemployment rate in August of 2001 was a mere 4.9% and immediately started asignificant increase in September. By December of 2002 the rate was over 6 percentseasonally adjusted. However, the Fed had brought interest rates to such a low point

47

that two dramatic changes occurred. We began a bull market in housing (interest ratesensitive) and the declines in automobile production reversed (again, interest ratesensitive commodities). These two portions of the economy kept the whole thing fromdisaster. As the United States became mired in military actions in Southwest Asia, thedollar began to lose value. The loss of the dollar’s value served to prevent even greaterpenetration of domestic markets by Japanese and European competitors, hence againkeeping unemployment from become a more serious problem.

However, this monetary stimulus resulted in the Fed having record low interestrates, at a time that the Federal budget deficits began run out of control. High Federaldebt levels being held, in large measure, by foreigners resulted in further downwardpressure on the value of the dollar. This downward pressure on the value of the dollarmeant that the price of foreign goods had an upward pressure. At the same time, Chinawas pegging the value of their currency on the value of the dollar and those exports tothe U.S. were increasing and driving rather dramatic economic growth in China. Inturn, the increasing economic prosperity in China resulted in increased demand foreverything in China, and particularly in commodities, like gold, copper, and oil. In turn,the prices of these commodities in the U.S. sky-rocketed because of world demandincreasing faster than supply, and the loss of purchasing power of the dollar. The tightmonetary policy which began some two years ago has generated a 4 1/4% increase inthe discount rate, and has essentially stopped inflation in its tracks. The quadrupling ofthe price of oil has had little effect in the overall price level simply because of thequadrupling of interest rates.

However, the second reason is not to be entirely ignored. Over the period that oilprices went up dramatically, the demand for oil in the U.S. has not kept pace. Increasedgas mileage, together with the loss of manufacturing has meant that oil demand hasincreased at lower rates than in recent decades in the U.S. As oil intensive industriesbecome less important, the inflationary effects of oil price increases that were sodevastating in the 1970s are less so in the first decade of this century.

Effects of Inflation

The effects of inflation impact different people in different ways. Creditors andthose living on a fixed income will generally suffer. However, debtors and those whoseincomes can be adjusted to reflect the higher prices will not, and perhaps this groupmay even benefit from higher rates of inflation.

If inflation is fully anticipated and people can adjust their nominal income or theirpurchasing behavior to account for inflation then there will likely be no adverse effects,however, if people cannot adjust their nominal income or consumption patterns peoplewill likely experience adverse effects. This is the same as if people experienceunanticipated inflation. Normally, if you cannot adjust income, are a creditor with a fixed

48

rate of interest or are living on a fixed income you will pay higher prices. The result isthat those individuals will see their standard of living eroded by inflation.

Debtors, whose loans specify a fixed rate of interest, typically benefit frominflation because they can pay loans-off in the future with money that is worth less. It isthis paying of loans with money that purchases less that harms creditors. It shouldcome as no surprise that the double digit inflation of fifteen years ago causedsubsequent loan contracts to often specify variable interest rates to protect creditorsfrom the erosive effects of unanticipated inflation.

Savers may also find themselves in the same position as creditors. If savingsare placed in long-term savings certificates that have a fixed rate of interest, inflationcan erode the earnings on those savings substantially. Savers that anticipate inflationwill seek assets that vary with the price level, rather than risk the loss associated withinflation.

Inflation will effect savings behavior in another way. If a person fully anticipatesinflation, rather than to save money now, consumers may acquire significant debt atfixed interest rates to take advantage of the potential inflationary leverage caused byfixed rates. Rather than to save now, consumers spend now. Therefore, inflationtypically creates expectations among people of increasing prices, and if people increasetheir purchases aggregate demand will increase. An increase in aggregate demand willcause demand-pull inflation. Therefore, inflationary expectations can create a spiralingof increased aggregate-demand and inflationary expectations that can feed off oneanother. At the other extreme, recessionary expectations may cause people to save,that results in reduced aggregate demand, and another spiral effect can result (butdownwards). This is what economists call rational expectations.

Rational expectations is the simple idea that people anticipate what is going tooccur in the economy. In other words, markets will have everything priced into thembecause both buyers and sellers know everything there is to know about the prices andavailability of the goods and services in the economy. While this view expectationsmakes it easier to manipulate economic models, it is not very realistic.

To account for expectations in a less harsh way in economic behavior, the theoryof adaptive expectations has been formulated. It is more likely that people will not takeextreme views of economic problems, including perfect knowledge. People willanticipate and react to relatively "sure things" and generally in the near term and wait tosee what happens. As economic conditions change, consumers and producers changetheir expectations to account for these changes; in another words, they adapt theirexpectations to current and near term information about future economic events.

The adaptive expectations model is supported by a substantial amount ofeconomic evidence. It appears that the overwhelming majority of the players in themacroeconomy are adaptive in their expectations.

49

Unemployment Differentials

David A. Dilts, Mike Rubison, Bob Paul, “Unemployment: which person'sburden - man or woman, black or white?" Ethnic and Racial Studies. Vol. 12,No. 1 (January 1989) pp. 100-114.

. . . The race and sex of the work force are significant determinants ofthe relative burdens of unemployment. Blacks are experiencing a decreasingburden of unemployment over the period examined while white females exhibita positive time trend (increasing unemployment over the period). Thedispersion of unemployment for blacks varies directly with the business cyclewhich suggests greater labor force participation sensitivity by this group. . . .The dispersions of white female unemployment vary countercyclically with thebusiness cycle which is consistent with the inherited wisdom concerningunemployment and macroeconomics.

. . . These results, together with the unemployment equation results,indicate that the unemployment rate for white males is not sensitive to thefluctuations in the business cycle nor do these data exhibit any significant timetrend. These are rather startling results. This evidence suggests that whilemales bear substantially the same relative unemployment rates over all rangesof the business cycle. . . .

KEY CONCEPTS

Business Cycle

Peaks

Troughs

Seasonal trends

Secular trends

Unemployment

frictional

structural

50

cyclical

Full employment

Natural rate of unemployment

Potential output

Labor Force

part-time employment

discouraged workers

false search

Okun’s Law

Burden of Unemployment

Inflation v. Deflation

CPI

Rule of 70

Demand-pull

Cost-push

Pure inflation

Monetarists

Stagflation

Historical experiences in the U.S. with Stagflation

Impact of Inflation

Expectations

51

STUDY GUIDE

Food for Thought:

Using the basic supply & demand model demonstrate cost-push, demand-pull, and pureinflation.

Critically evaluate the three categories of unemployment, be sure to discuss theproblems with conceptualizing and measuring each.

Critically evaluate the qualitative and quantitative costs of both inflation andunemployment. Which is worse? Why?

Sample Questions:

Multiple Choice:

Which of the following is most likely to benefit from a period of unanticipated inflation? (assuming fixed assets and liabilities)

A. Those whose liabilities are less than their assetsB. Those whose liabilities exceed their assets, and whose loans are variable rateC. Those whose liabilities exceed their assets, and whose loans are fixed rateloansD. None of the above will benefit

People who are unemployed due to a change in technology that results in a decline intheir industry fit which category of unemployment?

52

A. frictionalB. structuralC. cyclicalD. natural

True - False:

Seasonal variations in data are impossible to observe in annual data {TRUE}.

Cost-push inflation is often associated with increased unemployment {TRUE}

Chapter 4

53

Aggregate Supply & Aggregate Demand

The aggregate supply/ aggregate demand model of the macroeconomy will bedeveloped in this chapter. This model is one of two models (the other is the KeynesianCross) of the U.S. macroeconomic system that we will develop in this course. Theaggregate demand and aggregate supply model is the main mode of analysis thatcharacterized the classical school of economic thought and provides some usefulinsights. However, because the Keynesian Cross permits more direct analysis of themultiplier effects and other economic phenomenon it will be the model that is reliedupon for the majority of the course.

Aggregate Supply and Aggregate Demand

The aggregate supply/aggregate demand model is comparative static (slice oftime) model of the macroeconomy. Rather than to be able to observe changes duringeach second of a period of time (dynamic), we will compare the economy in one timeperiod with the model in a subsequent and distinct period. Its elegance arises from thefact that the model has foundations in microeconomics. In this view of themacroeconomy we simply aggregate everything on the supply side to obtain anaggregate supply curve and aggregate everything on the demand side to obtain anaggregate demand curve. Where aggregate supply intersects aggregate demand weobserve a macroeconomic equilibrium. However, because the two functions areaggregations, the horizontal axis does not measure the quantity of a particular good, itmeasures GNP. The vertical axis is not a price in the sense of a microeconomicmarket, but is the price level in the entire economy.

Aggregate Demand

The aggregate demand curve is a downward sloping function that shows theinverse relationship between the price level and gross domestic output (GDP). Thereasons that the aggregate demand curve slopes down and to the right differ from thereasons that individual market demand curves slope downward and to the right (i.e., thesubstitution & income effects - these do not work directly with macroeconomicaggregates, for among other reasons, we are dealing with all prices of all commodities,not a single price of a single commodity, as in a microeconomic sense).

The reasons for the downward sloping aggregate demand curve are:

(1) the wealth or real balance effect,

54

(2) the interest rate effect, and

(3) the foreign purchases effect.

As the price level increases, with fixed incomes, the given amount of savings andassets consumers have will purchase less. On the other hand, as the price leveldecreases the savings and wealth consumers have will purchase more. The negativerelation between the price level and the purchasing power of savings (wealth or realbalances) is called the real balances or wealth effect.

Assuming a fixed supply of money, an increase in the price level will increaseinterest rates. Increases in interest rates will reduce expenditures on goods andservices for which the demand is interest sensitive (e.g., consumer durables such ascars and investment). This negative relation between the interest rate and purchases iscalled the interest rate effect.

If the prices of domestic goods rise relative to foreign goods (an increase in theprice level), domestic consumers will purchase more foreign goods as substitutes,assuming stable exchange rates for currencies. Remember from Chapter 2 that ifexports remain constant and imports increase GNP declines (net exports). Therefore,as the price level increases, imports will also increase, ceteris paribus. The foreignpurchases effect is the propensity to increase purchases of imports, when the domesticprice level increases.

Each of these effects describes a negative relation between the price level andthe aggregate demand for the total output of the economy. Therefore, the aggregatedemand curve will slope downward and to the right, as shown in the diagram below:

The determinants of aggregate demand are the factors that shift the aggregatedemand curve. These determinants are:

55

(1) consumer and producer expectations concerning the price level, and real incomes,

(2) consumer indebtedness,

(3) personal taxes,

(4) interest rates,

(5) changes in technology,

(6) excess capacity in industry,

(7) government spending,

(8) net exports,

(9) national income earned abroad, and

(10) exchange rates.

The expectations of producers and consumers concerning real income orinflation (including profits from investments in business sector) will effect aggregatedemand. Inflationary expectations and anticipated increases in real income areconsistent with increased current expenditures. Should real income or inflation beexpected to fall, purchases may be postponed in favor of future consumption orinvestment.

As personal taxes, interest rates and indebtedness increase aggregate demandshould decrease due to a general reduction in effective demand. Should any of thesethree decrease there should be an increase in aggregate demand due to the increasedability to purchase output.

Changes in technology operate on aggregate demand in two distinct ways, thecreation of new products or new production processes, and the reduction of the costs ofproducing resulting in less expensive output. The amount of excess capacity inindustry will also impact the demand for capital goods. If there is substantial excesscapacity, output can be increased without obtaining more capital, on the other hand, ifthere is very little excess capacity an expansion of output will require purchasing morecapital.

Government expenditures account for a significant proportion of GNP. Ifgovernment expenditures decrease, so will aggregate demand; if they increase, so willaggregate demand. The same is true of national income earned abroad, as Americaneconomic activity moves abroad, the demand for domestic output will generally decline.

56

Exchange rates refer to how much of a foreign currency the U.S. dollar willpurchase. If the Japanese Yen loses value with respect to the U.S. dollar, thenJapanese goods will become cheaper. As the dollar buys more imports aggregatedemand will decrease simply because the U.S. dollar gains value relative to that foreigncurrency. As the dollar loses value relative to a foreign currency, such as the Yen, thenthe Japanese goods become more expensive and American consumers will substituteAmerican goods for Japanese goods and aggregate demand increases.

These determinants of aggregate demand act together to shift the aggregatedemand curve. Several of these determinants may change at the same time, andpossibly in different directions. The actual observed change in an aggregate demandcurve will result from the net effects of these changes. For example, if the dollar gainsone or two percent in value relative to the Yen this alone may cause a slight decrease inaggregate demand. However, if government purchases increase by two or three percentthis should offset any exchange rate tendency toward a reduction in aggregate demandand shift the aggregate demand curve to the right.

Aggregate Supply

The aggregate supply curve shows the amount of domestic output available ateach price level. The aggregate supply curve has three ranges, the Keynesian range(horizontal portion), the intermediate range (curved portion), and the classical range(vertical portion). These ranges of the aggregate supply curve are identified in thefollowing diagram.

The Keynesian range of the aggregate supply curve shows that any output isconsistent with the particular price level (the intersection of the aggregate supply curve

57

with the price level axis) up to where the intermediate range begins. In other words,wages and prices are assumed to be sticky (fixed) in this range of the aggregate supplycurve.

The classical range of the aggregate supply curve is vertical. Classicaleconomists believed that aggregate supply curve goes vertical at the full employmentlevel of output. In other words, any price level above the end of the intermediate rangeis consistent with the full employment level (potential) of GNP.

The intermediate range is the transition between the horizontal and verticalranges of the aggregate supply curve. As the macroeconomy approaches fullemployment levels of output the price level begins to increase, as output increases inthis range.

The determinants of aggregate supply cause the aggregate supply curve to shift. There are three general categories of the determinants of aggregate supply, these are:

(1) changes in input prices,

(2) changes in input productivity, and

(3) changes in the legal or institutional environment.

As input prices increase, aggregate supply decreases. For example, when theprice of oil increased dramatically in 1979-80, aggregate supply decreased because thecosts of production increased in general in the United States. In other words, for thesame level of production cost, we got less GNP. Should input prices decline, we shouldexpect to observe an increase in aggregate supply. In other words, for the same levelof production cost we got more GNP.

Changes in input productivity will also shift the aggregate supply curve. If laboror capital becomes more productive then producers will receive more output for thesame cost of production. This can occur because of better quality resources orbecause of the ability to use more efficient technology.

Changes in the legal or institutional environment will also increase aggregatesupply. One of the reasons that de-regulation is so popular in certain business circles isthat such changes in the legal environment will often result in lower costs of production. To the extent that there are no diseconomies, this increases aggregate supply. Moreefficient capital markets (i.e., the recent proposed S.E.C. changes concerned, amongothers, the New York Stock Exchange), better schools, better health care, and lesscrime all have the potential for increasing aggregate supply through the institutionalenvironment of business.

58

Macroeconomic Equilibrium

As mentioned earlier in this chapter, macroeconomic equilibrium in this model isthe intersection of aggregate supply and aggregate demand. The idea of equilibrium inthe macroeconomy is similar to that in microeconomic market. Where aggregate supplyand aggregate demand intersect, if there is no force applied to disturb the intersection(change in a determinant), then there is no propensity for the output and price levels tochange from those determined by the intersection. The following diagram portrays amacroeconomy in equilibrium. In this case the intersection of aggregate supply andaggregate demand is in the Keynesian range. Should aggregate demand increase upto where the intermediate range starts, only output will change. Through theintermediate range both output and the price level will increase as aggregate demandincrease. In the classical range if aggregate demand increases, output will remain thesame, but the price level will increase.

The following diagram shows a macroeconomy in equilibrium. The solidaggregate demand curve is the initial equilibrium. The two dotted lines show anincrease in aggregate demand (AD1) and a decrease in aggregate demand (AD2).

The changes in aggregate supply are analytically only marginally morecomplicated than aggregate demand. An increase in aggregate supply is simply a shiftof the entire curve to the right (downward). As aggregate supply shifts downward alongthe aggregate demand curve in the Keynesian and intermediate ranges, the price levelfalls, and output will increase. However, in the classical range a decrease in aggregatesupply changes neither the price level or total output. The following diagram portraysan increase in aggregate supply, the line labeled (AS1) and a decrease in aggregate

59

supply, the line labeled (AS2), the original aggregate supply curve is the solid line.

There is a more complicated view of changes in equilibrium in the aggregate supply andaggregate demand model called the Ratchet Effect. The Rachet Effect is where there isa decrease in aggregate demand, but producers are unwilling to accept lower prices(rigid prices and wages). Rather than to accept the lower price levels resulting from adecrease in aggregate demand, producers will decrease aggregate supply. Therefore,there is a ratcheting of the aggregate supply curve (decrease in the intermediate andKeynesian ranges) which will keep the price level the same, but with reduced output. Inother words, there can be increases in prices (forward) but no decreases in the pricelevel (but not backward) because producer will not accept decreases (price rigidity). The same is argued to exist for wages in the labor market, in other words, unions willresist decreases in wages associated with a decrease in aggregate demand, hencethey too, will place downward pressure on aggregate supply.

The following diagram illustrates the rachet effect:

60

An increase in aggregate demand from AD1 to AD2 moves the equilibrium frompoint a to point b with real output and the price level increasing. However, if prices areinflexible downward, then a decline in aggregate demand from AD2 to AD1 will notrestore the economy to its original equilibrium at point a. Instead, the new equilibriumwill be at point c with the price level remaining at the higher level and output falling tothe lowest point. The ratchet effect means that the aggregate supply curve has shiftedupward (a decrease) in both the Keynesian and intermediate ranges.

KEY CONCEPTS

Aggregate Demand

Real balance effect

Interest rate effect

Foreign purchases effect

Determinants of Aggregate Demand

Aggregate Supply

Keynesian Range

Classical Range

Intermediate Range

Determinants of Aggregate Supply

Equilibrium in Aggregate Supply and Aggregate Demand

Rachet Effect

STUDY GUIDE

Food for Thought:

Explain and demonstrate the operation of the determinants of aggregate supply and

61

aggregate demand.

Critically evaluate both the Keynesian and Classical ranges of the aggregate supplycurve.

Is the ratchet effect plausible? Explain.

Why does the aggregate demand curve slope downward? Why do we find ranges inthe aggregate supply curve? Explain.

Sample Questions:

Multiple Choice:

The aggregate demand curve is most likely to shift to the right (increase) when there isa decrease in:

A. the overall price levelB. the personal income tax ratesC. the average wage received by workersD. consumer and business confidence in the economy

A short-run increase in interest rates on consumer loans may cause a decrease in

62

aggregate demand. What would we expect to observe, if there is no ratchet effect?

A. In the classical range only a reduction in pricesB. In the Keynesian range only a reduction in outputC. Both A and B are correctD. Neither A or B are correct

True - False:

All economists are convinced that ratchet effect exists in today's economy. {FALSE}

One of the major reasons that the aggregate demand curve slopes downward is the realbalances effect. {TRUE}

Chapter 5

Classical and Keynesian Models

The purpose of this chapter is to extend the analysis presented in Chapter 4. Based on the foundations of the relatively simple aggregate supply and aggregatedemand model both the Classical and Keynesian theories of macroeconomics will bedeveloped and compared in this chapter.

Introduction

63

The Classical theory of employment (macroeconomics) traces its origins to thenineteenth century and to such economists as John Stuart Mill and David Ricardo. TheClassical theory dominated modern economic thought until the middle of the GreatDepression when its predictions simply were at odds with reality. However, the work ofthe classical school laid the foundations for current economic theory and a greatintellectual debt is owed to these economists.

During the beginnings of the 1930s economists in both Europe and the UnitedStates recognized that current theory was inadequate to explain how a depression ofsuch magnitude and duration could occur. After all, the miracle of free marketcapitalism was suppose to always result in a return to prosperity after short periods ofcorrection (recession). For a long term disequilibrium to be observed was bothdisconcerting and fascinating. It became very obvious by 1935 that the marketmechanisms were not going to self-adjust and bring the economy out of a very deepdepression.

John Maynard Keynes, an English mathematician and economist, is the father ofmodern macroeconomics. His book, The General Theory, (1936) was to change howeconomists would examine macroeconomic activity for the next six decades (untilpresent). Keynes' work laid aside the notion that a free enterprise market system canself-correct. He also provided the paradigm that explained how recessions can spiraldownwards into depression without active government intervention to correct theobserved deficiencies in aggregate demand. Some of Keynes' ideas were original,however, he borrowed heavily from the Swedish School, in particular, Gunar Myrdal, inhis explanations of the fact that there is no viable mechanism in our system of marketsthat provide for correction of recessionary spirals.

Many economists, on both sides of the Atlantic, were working on the problem ofwhy the classical theory had failed so miserably in explaining the prolonged, deepdownturn and in offering policy prescriptions to cure the Great Depression. In someways it resembled an intellectual scavenger hunt. As soon as Keynes had worked outthe theory, he literally rushed to print before someone beat him to the punch, so tospeak. The result is that The General Theory is not particularly well written and hasbeen subject to criticism for the rushed writing, however, its contributions tounderstanding the operation of the macroeconomy are unmistakable.

The Classical Theory

The classical theory of employment (macroeconomics) rests upon twofundamental principles, these are: (1) underspending is unlikely to occur, and (2) ifunderspending should occur, the wage-price flexibility of free markets will preventrecession by adjusting output upwards as wages and prices declined.

64

What is meant by underspending is that private expenditures will be insufficientto support the current level of output. The classicists believed that spending inamounts less than sufficient to purchase the full employment level of output is not likely. They also believed that even if underspending should occur, then price/wage flexibilitywill prevent output declines because prices and wages would adjust to keep theeconomy at the full employment level of output.

The classicists based their faith in the market system on a simple propositioncalled Say's Law. Say's Law in its crudest form states that "Supply creates its owndemand." In other words, every level of output creates enough income to purchaseexactly what was produced. However, as sensible as this proposition may seem thereis a serious problem. There are leakages from the system. The most glaring omissionin Say's Law, is that it does not account for savings. Savings give rise to gross privatedomestic investment and the interest rates are what links savings and investment. However, there is no assurance that savings and investment must always be inequilibrium. In fact, people save for far different reasons that investors' purchasecapital.

Further, the classicists believed that both wages and prices were flexible. Inother words, as the economy entered a recession both wages and prices would declineto bring output back up to pre-recession levels. However, there is empirical evidencethat demonstrates that producers will cut-back on production rather than to lower prices,and that factor prices rarely decline in the face of recession. The classicists believedthat a laissez faire economy would result in macroeconomic equilibria through theunfettered operation of the market system and that only the government could causedisequilibria in the macroeconomy.

One need only look to the automobile industry of the last ten years to understandthat wage - price flexibility does not exist. Automobile producers have not loweredprices in decades. When excess inventories accumulate, the car dealers will offerrebates or inexpensive financing, but they have yet to offer price reductions. There hasbeen some concession bargaining by the unions in this industry, but even where wageswere held down, it is rare that a union accepts a nominal wage cut.

Modern neo-classical macroeconomics takes far less rigid views. Even thoughthe neo-classicists have come to realize that the market system has its imperfections,they believe that government should be the economic stabilizer of last resort. Further,even though there is now recognition that the lauded wage-price flexibility is unlikely, byitself, to be able to correct major downturns in economic activity, the neo-classicistsstubbornly hold to the view that government must stay out of the economic stabilizationbusiness. The one exception they seem to allow, is the possibility of a major externalshock to the system, otherwise they claim there is sufficient flexibility to prevent majordepressions, with only very limited government responses (primarily through monetary,rather than fiscal policy). In other words, the differences between the Keynesians andthe neo-classicists are very subtle (magnitude of government involvement) and focus

65

primarily on the starting point of the analysis (Keynes with recession, neo-classicist withequilibrium).

Keynesian Model

Keynes recognized that full employment is not guaranteed, because interestmotivates both households and businesses differently - just because households savedoes not guarantee businesses will invest. In other words, there is no guarantee thatleakages will result in investment injections back into the system. Further, Keynes wasunwilling to assume that self-interest in a market system guaranteed that there would bewage-price flexibility. In fact, the empirical evidence suggested that wages and pricesexhibited a substantial amount of downward rigidity.

Under the Keynesian assumptions there is no magic in the market system. Themechanisms that the classicist thought would guarantee adjustments back toward fullemployment-equilibrium simply did not exist. Therefore, Keynes believed thatgovernment had to be pro-active in assuring that underspending did not spiral theeconomy into depression once a recession began.

Keynes' views were revolutionary for their time. However, it must beremembered that the Great Depression was an economic downturn unlike anythingexperienced during our lifetimes. Parents and grandparents, no doubt, have relatedsome of the traumatic experiences they may have endured during the Depression totheir children and grandchildren (perhaps you have heard some of these stories). However, such economic devastation is something that is nearly impossible to imagineunless one has lived through it. One-quarter of the labor force was unemployed atpoints during the Depression. The U.S. economy had almost no social safety net, nounemployment compensation, little in the way of welfare programs, no social security,no collective bargaining, and very small government. Our generations have gottenused to the idea that there is something between us and absolute poverty, there areprograms to provide income during times of unemployment, generally for a sufficientperiod to find alternative employment. Even though these programs may have apersonal significance, they were intended to prevent future demand-deficiencydepressions.

The array of entitlement programs, particularly unemployment, welfare, andsocial security provides an automatic method to keep spending from spiraling intodepression. Once recessionary pressures begin to build in the economy, the loss ofemployment does not eliminate a household's consumption as soon as savings aredepleted. Unemployment compensation and then welfare will keep the lights on (and I& M employees working), food on the table (to the relief of those working at Krogers andScotts), and clothes on the back of the those in need (keeping people working atWalmart through Hudsons). Further, the government has been proactive in

66

stabilization of economic downturns since the beginning of World War II. Therebyproviding us with the expectation that something will be done to get things back in orderas soon as possible once a recession starts.

The government's ability to tax and to engage in deficit spending provides theflexibility in the market system to deal with underspending that was only presumed toexist by deflating the economy. Without the automatic stabilizers and the governmentflexibility to deal with serious underspending the economy could theoretically producethe same results that it did in the 1930s.

Aggregate supply and aggregate demand can now be expanded to include thesavings and investment in the analysis to make for a more complete model. In sodoing, we can also create a more powerful analytical tool.

The Consumption Schedule

In beginning the development of the Keynesian Cross model, we return toaggregate supply and aggregate demand. Along the vertical axis we are going tomeasure expenditures (consumption, government, or investment). Along the horizontalaxis we are going to measure income (disposable income). At the intersection of thesetwo axes (the origin) we have a forty-five degree line that extends upward and to theright. What this forty-five degree line illustrates is every point where expenditures andincome are equal.

The consumption schedule intersects the 45 degree line at 400 in disposableincome, this is also where the savings function intersects zero (in the graph below theconsumption function). At this point, (400) all disposable income is consumed and

67

nothing is saved. To the left of the intersection of the consumption function with the 45degree line, the consumption function lies above the 45 degree line. The distancebetween the 45 degree line and the consumption schedule is dissavings, shown in thesavings schedule graph by the savings function falling below zero. Dissavings meanspeople are spending out their savings or are borrowing (negative savings). To the rightof the intersection of the consumption function with the 45 degree line, the consumptionschedule is below the 45 degree line. The distance that the consumption function isbelow the 45 degree line is called savings, shown in the bottom graph by the savingsfunction rising above zero.

This analysis shows how savings is a leakage from the system. Perhaps moreimportantly the analysis also shows that there is a predictable relation betweenconsumption and savings. What is not consumed is saved, and vice versa. However,there is more to this than savings plus consumption must equal income.

The Marginal Propensity to Consume (MPC) is the proportion of any increase indisposable income that is spent on consumption (if an entire increase in income is spentMPC is 1, if none is spent then MPC is zero). The Marginal Propensity to Save (MPS)is the proportion of any increase in disposable income that is saved. The relationbetween MPC and MPS is that MPS + MPC =1, in other words, any change in incomewill be either consumed or saved. This relation is demonstrated graphically in thefollowing diagram:

The slope (rise divided by the run) of the consumption function is the MPC andthe slope of the savings function is the MPS. The slope of the consumption function is.8 or (8/10) and the slope of the savings function .2 or (2/10). Total change in income

68

will be either spent or saved. If ten dollars of additional income is obtained then $2 ($10times .2) will be saved and $8 ($10 times .8) will be spent on consumption.

The marginal propensities to save and to consume deal with changes in one'sincome. However, average propensities to consume or save deal with what happens tototal income. The Average Propensity to Consume (APC) is total consumption dividedby total income, Average Propensity to Save (APS) is total savings divided by totalincome. Again, (just like the marginal propensities) if income can be either saved orconsumed (and nothing else) then the following relation holds, the average propensityto consume plus the average propensity to save must equal one (APC + APS = 1). Forexample, if total income is $1000 and the average propensity to consume is .95 and theaverage propensity to save is .05, then the economy will experience $950 isconsumption (.95 times $1000) and $50 in savings ($1000 times .05).

The non-income determinants of consumption and saving cause theconsumption and savings functions to shift. The non-income determinants ofconsumption and saving are: (1) wealth, (2) prices, (3) expectations concerning futureprices, incomes and availability of commodities, (4) consumer debts, and (5) taxes.

In general, it has been empirically observed that the greater the amount of wealthpossessed by a household the more of their current income will be spent onconsumption, ceteris paribus. All other things equal, the more wealth possessed by ahousehold the less their incentive to accumulate more. Conversely, the less wealthpossessed by a household the greater the incentive to save. An Italian economist,Franco Modigliani, observed that this general rule varied somewhat by the stage in life aperson was in. The young (twenties) tend to save for homes and children, in the latetwenties through the forties, savings were less evident as children were raised, and withthe empty nest, came savings for retirement. This is called the Life Cycle Hypothesis.

An increase in the price level has the effect of causing the consumption functionto shift downward. As prices increase, the real balances effect (from the previouschapter) becomes a binding constraint. As the value of wealth decreases, so too doesthe command over goods and services, so consumption must fall (and savingsincrease). If the price level decreases, then we would expect consumption to increase(and savings to fall).

The expectations of households concerning future prices, incomes andavailability of commodities will also impact consumption and savings. As householdsexpect price to increase, real incomes to decline or commodities to be less available,current consumption will increase (and current savings decline). If, on the other hand,households expect incomes to increase, prices to fall, or commodities to become moregenerally available, current consumption will decline (and savings will increase).

Consumer indebtedness will also effect consumption and savings. If consumersare heavily indebted, save a third of their income goes on debt maintenance then

69

current consumption will decline to pay off debts (dis-savings). However, ifindebtedness is relatively low, consumers will consume more of their current income,perhaps even engage in dis-savings (borrowing) to consume more currently.

Taxation operates on both the savings and consumption schedules in the sameway. Because taxes are generally paid partly from savings and partly from currentincome, an increase in taxes will cause both consumption and savings to decline. Onthe other hand, if taxes are decreased, then both the savings and consumptionfunctions will increase (shift upwards).

Investment

Investment demand is a function of the interest rate. The following diagramshows an investment demand curve. The investment demand curve is downwardsloping, which suggests that as the interest rate increases investment decreases. Thereason for this is relatively simple. If the expected net return on an investment is sixpercent, it is not profitable to invest when the interest is equal to or more than sixpercent. A firm must be able to borrow the money to purchase capital at an interest ratethat is less than the expected net rate of return for the investment project to beundertaken Therefore, there is an inverse relation between expected return and theinterest rate; and the interaction of the interest rate with the expected rate of returndetermine the amount of investment.

The determinants of investment demand are those things that will cause theinvestment demand curve to shift. The determinants of investment demand are: (1) acquisition, maintenance & operating costs of capital, (2) business taxes, (3) technology, (4) stock of capital on hand, and (5) expectations concerning profits infuture.

70

The investment demand depends on whether the expected net rate of return ishigher than the interest rate. Therefore, anything that increases the expected net returnwill shift the investment demand curve to the right, anything that cause the expectedreturn to fall will shift the investment demand curve to the left (decrease). As theacquisition, maintenance and operating costs of capital increase, the net expectedreturn will decrease, ceteris paribus, thereby shifting the demand curve to the left. If theacquisition, maintenance and operating costs decline, we would expected a higher rateof return on this investment and therefore the demand curve shifts to the right(increase).

Business taxes are a cost of operation. If business taxes increase, the expectednet (after tax) return will decline, this shifts the investment demand curve to the left. Ifbusiness taxes decrease, the expected net return on the investment will increase,thereby increasing the investment demand curve.

Changes in technology will also shift the investment demand curve. Moreefficient technology will generally increase expected net returns and shift the investmentdemand to the right (increase). By decreasing production costs or improving productquality through technological improvements competitive advantages may be reaped andthis is one of the most important determinants of investment since World War II.

The stock of capital goods on hand will also impact investment demand. To theextent that producers have a large stock of capital goods on hand, investment demandwill be dampened. On the other hand, if producers have little or no inventory of capitalgoods, then investment demand may increase to restore depleted stocks of capital.

Business investment decisions are heavily influenced by expectations. Expectations concerning the productive life of capital, its projected obsolescence,expectations concerning sales and profits in the future will also impact investmentdecisions. For example, expectations that technological break-throughs may makecurrent computer equipment less competitive may dampen current investment demand. Further, if competitors are tooling-up to enter your industry, you may be hesitant toinvest in more capital if the profits margins will be cut by the entrance of competitors.

Autonomous v. Induced Investment

Autonomous investment is that investment that occurs which is not related to thelevel Gross Domestic Product. Investment that is based on population growth,expected technological progress, changes in the tax structure or legal environment, oron fads is generally not a function of the level of total output of the economy and iscalled autonomous investment. In examining the following diagram, the autonomousinvestment function is a horizontal line that intercepts the investment axis at the level ofautonomous investment.

71

Induced investment is functionally related to the level of Gross Domestic Product. The following diagram has an investment function that slopes upward (increases asGDP increases). Induced investment is that investment that is "induced" because ofincreased business activity. In short, induced investment means that investment that iscaused by increased levels of GDP.

Throughout American economic history the level investment has been veryvolatile. In fact, much of the variation in the business cycle can be attributed to theinstability of investment in the United States. There are several reasons for thisinstability, including: (1) variations in the durability of capital, (2) irregularity ofinnovation, (3) variability of profits, and (4) the expectations of investors.

The durability of most capital goods means that they have an expectedproductive life of at least several years, if not decades. Because of the durability andexpense of capital goods, their purchase can be easily postponed. For example, a bankmay re-decorate and patch-up an old building, rather than build a new building,depending on their business expectations and current financial position.

Perhaps the most important contributors to the instability of investment in thepost- World War II period is the irregularity of innovations. With the increase in basicknowledge, comes the ability to develop new products and production processes. During World War II there was heavy public investment in basic research in medicineand the pure sciences. What was intended from these public expenditures was formilitary use, but many of these discoveries had important civilian implications for newproducts and better production methods. Again, in the late 1950s and early 1960s anexplosion of basic research occur that led to commercial advantages. The Russianslaunched Sputnik and gave the Western World a wake-up call that they were behind insome important technical areas, the government again spent money on education andbasic research.

72

For the private sector to invest there must be some expectation of profits flowingfrom that investment. Much of the decline in private investment during the GreatDepression was because private investors did not expect to be able to make a profit inthe economic environment of the time. In the late 1940s, automobile producers knewthat profits would be nearly guaranteed because no new private passenger cars werebuilt in the war years. There was very significant investment in plant and equipment inthe auto industry in those years (mostly to convert from war to peace-time production).

The expectations of business concerning profits, prices, technology, legalenvironment and most everything effecting their business are simply forecasts. Because the best informed forecasts are still guess-work, there is substantial variabilityin business conditions expectations. Because these expectations vary substantiallyacross businesses and over time, there should be significant variability in investmentdecisions.

KEY CONCEPTS

Classical Theory

Say’s Law

Keynesian Model

price-wage rigidity

No guarantee of full employment

Income v. Expenditures

Consumption Function

Marginal Propensity to Consume v. Marginal Propensity to Save

Determinants of consumption and savings

Investment Demand

Determinants of investment demand

73

Investment instability in U.S.

Induced investment v. autonomous investment

STUDY GUIDE

Food for Thought:

Compare and contrast the classical model of the economy with the Keynesian model. Are these models really that different? Explain.

Develop the consumption and savings schedules. Fully explain the assumptionsunderlying the models, its mechanics, and its implications.

What is the relation between savings and investment? Explain.

Critically evaluate Say's Law.

Sample Questions:

Multiple Choice:

If you receive $40 in additional income and you save $4, what is your marginalpropensity to consume?

A. .4

74

B. .9C. 1.0D. None of the above

Which of the following contribute to investment instability?

A. Irregularity in innovationB. Variable investor expectationsC. Purchases of capital durable goods are postponableD. All of the above

True - False:

MPC + MPS = 1 {TRUE}

The forty-five degree line in the consumption function model shows each point at whichdisposable income and consumption are equal. {TRUE}

Chapter 6

Equilibrium in the Keynesian Model

The purpose of this chapter is to extend the analysis of Chapter 4. The basicKeynesian Cross model will be developed and examine how equilibrium can beachieved in the macroeconomy. In Chapter 6, that follows we will complete the analysisof the Keynesian view of the macroeconomy.

Equilibrium and Disequilibrium

Equilibrium GDP is that output that will create total spending just sufficient to buythat output (where aggregate expenditure schedule intersects 45 degree line). Further,once achieved, an equilibrium in a macroeconomy has no propensity to change, unlessthere is a shock to the system, or some variable changes to cause a disequilibrium. Disequilibrium where spending is insufficient (recessionary gap) or too high for level ofoutput (inflationary gap).

75

The neo-classicists view the primary macroeconomic as one of maintainingequilibriums. Their analysis of the system begins with equilibrium, because unless thereis some external intervening problem, they believe this is the state of nature for themacroeconomy. Keynesians, on the other hand, begin their analysis with disequilibriumbecause this is the natural state for a macroeconomy. The constant changesassociated with policies, technological change, and autonomous influences will impactthe economy periodically and cause it to move out of equilibrium. In fact, this is themajor analytical difference between the neo-classical and Keynesian economists.

Keynes' Expenditures - Output Approach

From Chapter 2, remember that one of the ways that GDP can be calculated isusing the identity Y = C + I + G + X ; where Y = GDP, C = Consumption, I =Investment, G= Government expenditures, and X = Net exports (exports minusimports). This provides for us the formula by which we can complete the model webegan in the previous chapter. Consider the following diagram:

Remember that the 45 degree line is each point where spending is exactly equalto GDP. The above figure shows a simple economy with no public or foreign sectors. We begin the analysis by adding investment to consumption, and obtaining Y = C + I. The equilibrium level of GDP is indicated above where C + I is equal to the 45 degreeline. Investment in this model is autonomous and the amount of investment is thevertical distance between the C and the C + I lines.

Keynes' Leakages - Injections Approach

76

The same result obtained in the expenditures - output approach above can beobtained using another method. Remember that APC + APS = 1, and MPC + MPS = 1,this suggests that leakages from the system are also predictable. The leakages -injections approach relies on the equality of investment and savings at equilibrium in amacroeconomic system (I = S).

The reason that the leakages - injection approach works is that plannedinvestment must be equal savings. The amount of savings is what is available for grossprivate domestic investment. When investors use the available savings, the leakages(savings) from the system are injected back into the system through investment. However, this must be planned investment.

Unplanned investment is the cause of disequilibrium. Inventories can increasebeyond that planned, and inventories are investment (stock of unsold output). Wheninventories accumulate there is output that is not purchased, hence reductions inspending which is recessionary; or, on the other hand, if intended inventories aredepleted which this inflationary because of the excess spending in the system. Consider the following diagram, where savings is equal to I1, investment. If there isunplanned investment, the savings line is below the investment line, at the lowest levelof GDP, the vertical line label UPI (Unplanned Investment), if inventories are depletedbeyond the planned level, then the savings line is above I1, as illustrated with thehighest level of GDP, and that vertical line is labeled Dep. Inv. for Depleted Inventory.

The original equilibrium is where I1 is equal to S. The if the unplanned inventoryor unplanned depletion of inventory become planned investment the analysis changes. If we experience a decrease in planned investment we move down to I2, with areduction in GDP (Recession), just like an increase in unplanned investment, and if anincrease in investment is observed it will be observed at I3, which is expansionary, and

77

this is similar to unplanned depletion of inventories (which could also be inflationary).

Re-spending

The interdependence of most developed economies, results in an observed re-spending effect if there is an injection of spending in the economy. This re-spendingeffect is called the multiplier, and we will provide a more detailed analysis of themultiplier effects in the following chapter, however, the re-spending effect representedby the multiplier will be introduced here to provide a full understanding of the model.

If there is an increase in expenditures, there will be a re-spending effect. In otherwords, if $10 is injected into the system, then it is income to someone. That first personwill spend a portion of the income and save a portion. If MPC is .90 then the firstindividual will save $1 and spend $9.00. The second person receives $9.00 in incomeand will spend $8.10 and save $0.90. This process continues until there is no moneyleft to be spent. Instead of summing all of the income, expenditures, and/or savingsthere is a short-hand method of determining the total effect -- this is called theMultiplier, which is 1/1-MPC or 1/MPS. The significance any increase in expendituresis that it will increase GDP by a multiple of the original increase in spending.

The re-spending effect and the leakage - injection approach to GDP provides forcurious paradox. This paradox is called the paradox of thrift. To accumulate capital, itis often the policy of less developed countries to encourage savings, to reduce thecountry's dependence on international capital markets. What often happens is that asa society tries to save more it may actually save the same amount, this is calledthe paradox of thrift. The reason that savings may remain the same is that unlessinvestment moves up as a result of the increased savings, all that happens is that GDPdeclines. The higher rate of savings with a smaller GDP results in the same amount ofsavings if GDP declines proportionally with the increase in savings rates. If investmentis autonomous then there is no reason to believe that investment will increase simplybecause the savings rate increased. In fact, because of the re-spending effects of theleakages, generally savings will remain the same as before the rate went up.

Full Employment

Simply because C + I + G intersects the 45 degree line does not assure utopia. The level of GDP associated with the intersection of the C + I + G line with the 45degree line may be a disequilibrium level of GDP, and not the full employment level ofGDP. The full employment level of GDP may be to the right or to the left of theaggregate expenditures line. Where this occurs you have respectively, (1) arecessionary gap or (2) an inflationary gap. In either case, there is macroeconomicdisequilibrium, that will generally require appropriate corrective action (as will be

78

described in detail in the following chapter on fiscal policy). Both forms ofdisequilibrium can be illustrated using the expenditures - output approach. Consider thefollowing two diagrams:

Recessionary Gap

In the above diagram the dashed line labeled Full Employment GDP. Is the levelof GDP that is associated with potential GDP or full employment. The distance betweenthe C + I line and the 45 degree line along the dashed indicator line is the recessionarygap. The dotted line shows the current macroeconomic equilibrium. Okun's Law(Chapter 3) provides some insight into what this means, remember that every 2.5% oflost potential GDP is associated with 1% unemployment above the full employmentlevel. Therefore, this recession represents lost output and unemployment is fixedproportions of 1% to 2.5%.

In other words, if we know what the unemployment rate is, then theoretically weshould also know where we are with respect to potential GDP. For example, if we havean 8% unemployment rate and the current GDP is $1000 billion dollars, we cancalculate what the full employment, non-inflationary level of GDP is quite simply. UsingOkun’s law we know that for every 1% the unemployment rate exceeds the 4% naturalrate, we lose 2.5% of potential GDP. With 8% unemployment we know that we arecurrently 10% below potential GDP (2.5% times 4 = 10%). To calculate what potentialGDP is one need only a little 8th grade algebra, to wit:

.9GDPpotential = $1000 billion

dividing both sides of the equation by .9 yields:

GDPpotential = $1000 billion / .9 or

79

GDPpotential = $1,111.11 billion

Therefore,Okun’s law turns out tobe a handy road map towhere the economyshould be with respectto where it currently is. Most of what will bedone in this course withrespect to disequilibriumwill be concerned withrecessionary gaps. Inflation is best left tomonetary policy, therecord shows thatstagflation has beenobtained and therefore

a better system than just simple recessionary and inflationary gaps is needed. In thiscase, recessionary gaps have been appropriately dealt with using Keynesianeconomics, but inflation is a monetary phenomenon. However, in order to present abalanced view, a brief examination of the inflationary gap view of the rudimentaryKeynesian model will be presented.

Inflationary Gap

An inflationary gap can arise in the Keynesian view of the world by actions whichtake the economy past the potential GDP levels of output. Consider the followingdiagram below:

80

Again the full employment (non-inflationary) level of GDP is indicated by the dashed linelabeled full employment GDP. The distance between the C + I line and the 45 degreeline along the dashed indicator line is the inflationary gap. The dotted indicator lineshows the current macroeconomic equilibrium. In this case, there is too much spendingin the economy or some other (similar) problem that has resulted in an inflated pricelevel. A reduction in GDP is necessary to restore price level stability, and to eliminateexcess output.

The various C + I and 45 degree line intersections, if multiplied by the appropriateprice level will yield one point on the aggregate demand curve. Shifts in aggregatedemand can be shown with holding the price level constant and showing increases ordecreases in C + I in the Keynesian Cross model. Both models can be used to analyzeessentially the same macroeconomic events. However, from this point on willconcentrate on our efforts on mastering the Keynesian Cross.

With these basic tools in hand, we can now turn our attention to economicstabilization and the use of fiscal policy for such activities. In the following chapter, theKeynesian model will be applied to problems of economic stabilization using taxes, government expenditures, and the combination of the two to eliminate recessionary andinflationary gaps.

KEY CONCEPTS

Macroeconomic Equilibrium v. Macroeconomic Disequilibrium

Expenditures - Output Approach

Leakages - Injections Approach

Planned v. Unplanned Investment

Re-spending

Multiplier Effect

81

Paradox of Thrift

Economic Development constraints

Full Employment

Potential GDP

Natural Rate of Unemployment

Recessionary Gap

Inflationary Gap

STUDY GUIDE

Food for Thought:

Critically evaluate the paradox of thrift.

What specifically is meant by a recessionary gap? Explain, and how does this differfrom an inflationary gap? Explain.

What insight does the multiplier provide for the Bush tax cuts of 2003? Explain.

Sample Questions:

Multiple Choice:

With a marginal propensity to consume of .8 the simple multiplier is:

A. 0.25B. 2.00

82

C. 4.00D. None of the above

An inflationary gap is:

A. The amount by which C+I+G exceeds the 45 degree line at or above the fullemployment level of outputB. The amount by which C+I+G is below the 45 degree line at the full employment levelof outputC. The amount by which the full employment level of output exceed the current level ofoutputD. None of the above

True - False

A recessionary gap is how much the aggregate expenditure line must increase to attainfull employment without inflation. {TRUE}

The equilibrium level of output is that output whose production will create total spendingjust sufficient to purchase that output. {TRUE}

Chapter 7

John Maynard Keynes and Fiscal Policy

The Great Depression demonstrated that the economy is not self-correcting asalleged by many of the classical economists of the time. The revolutionary Keynesianview that government must take a proactive role in stabilization of the business cyclefocused in large measure on the powers of the federal government to tax and to makeexpenditures. That power and its uses is what comprises the majority of the discussionin this chapter. Together the government's activities involving taxing and spending arecalled fiscal policy. The purpose of this chapter is to examine the fiscal policy tools andtheir effectiveness.

83

The Keynesian Revolution

What is not well-understood even today is what a revolution John MaynardKeynes brought to economic reasoning. The classical model was, unfortunately, notcapable of explaining much of what was being observed in the first third of the twentiethcentury in the Western world. Keynes and other, primarily European economists, wereobserving and writing concerning the inability of the classical model to explain thefinancial shocks which rocked both the U.S. economy and the many of the Europeaneconomies both pre and post-World War I. What is perhaps somewhat disturbing is thatsome seventy years after the appearance of the General Theory there are those whostill cling to some of the more peculiar aspect of the old classical reasoning. Marketsare imperfect and they have tendencies which result in less than satisfactory results leftunrestrained. Yet in this system populated by imperfect people, we have not yetdevised ways of allocating scare resources which provide entirely satisfactory results foreveryone concerned. In our second-best world we struggle to get-by, and it is therecognition of the fact that we do live in a second-best world that is important to ourpractical progress. In fact, it is that very recognition that is probably the greatestachievement of Keynes’ work within the economic profession.

Chapter 1 of the General Theory of Employment, Interest and Money by JohnMaynard Keynes sums up the economy quandary of the day, and is replicated here,since it is less than a full page. It is this chapter that basically sums up what had beentranspiring, and its results for the economic system and for the profession. It also nicelyportrays what the real meaning of the idea classical versus neo-classical or as they arenow called, Keynesian economists are. The following box provides interesting insightinto the controversy:

84

From this beginning it can be seen that the immodest title “General Theory” hada specific meaning within the context of the inherited economic theory of the day, andnot something more conceited.

Discretionary Fiscal Policy

The Employment Act of 1946 formalized the federal government's responsibilityfor promoting economic stability. The economic history of the first half of the twentiethcentury was a relatively stormy series of financial panics prior to World War I, arelatively stagnant decade after World War I, and the Depression of the 1930s. AfterWorld War II, a new problem arose called inflation. It should therefore come as nosurprise that the Congress wished to assure that there was a pro-active role for thegovernment to smooth-out these swings in the business cycle.

The government has several roles to fulfill in society. Its fiscal powers arenecessary to providing essential public goods. Without the federal government nationaldefense, the judiciary, and several other critical functions could not be provided forsociety. There is also the ebb and flow of politics. The Great Society of Lyndon B.Johnson represents the public opinion of the 1960s, today's political agenda seems to

Chapter 1, The General Theory

I have called this book the General Theory of Employment, Interest and Money,placing the emphasis on the prefix general. The object of such a title is to contrastthe character of my arguments and conclusions with those of the classical* theory ofthe subject, upon which I was brought up and which dominates the economicthought, both practical and theoretical, of the governing and academic classes ofthis generation, as it has for a hundred years past. I shall argue that the postulatesof the classical theory are applicable to a special case only and not to the generalcase, the situation which it assumes being a limiting point of the possible positions ofequilibrium. Moreover, the characteristics of the special case assumed by theclassical theory happen not to be those of the economic society in which we actuallylive, with the result that its teaching is misleading and disastrous if we attempt toapply it to the facts of experience.

* “The classical economists” was a name invented by Marx to cover Ricardo andJames Mill and their predecessors, that is to say for the founders of the theory whichculminated in the Ricardian economics. I have become accustomed, perhapsperpetrating a solecism, to include in “the classical school” the followers of Ricardo,those, that is to say, who adopted and perfected the theory of Ricardian economicsincluding (for example) J. S. Mill, Marshall, Edgeworth and Prof. Pigou.

85

be substantially different. The result is that as political opinion changes so will thegovernment's pattern of taxation and expenditures.

The government's taxing and spending authority to stabilize the economyis called discretionary fiscal policy. Taxation and spending by the federalgovernment has been used, with some frequency, to smooth out the business cycle. Intimes of underspending, the short-fall is made up by government spending or reductionsin taxes. In times of inflation, cuts in spending or increases in taxes have been used tocool-off the economy. However, in recent year the discretionary fiscal policies of thefederal government have become extremely controversial. The first step inunderstanding this controversy, is to understand the role of fiscal policy in economicstabilization.

Milton Friedman and others, have argued that there is no role for discretionaryfiscal policy. Friedman’s position is that much of the business cycle is the result ofgovernmental interference and that the long lags in fiscal policy becomingoperationalized makes it only a potential force for mischief, and not hope for stability. Inother words, Friedman believes that the classical economists, while over simplifying theargument, were basically correct about keeping government out of the economy.

Simplifying Assumptions

As was discussed in E201, Introduction to Microeconomics, assumptions areabstractions from reality. The utility of these abstractions is to eliminate many of thecomplications that have the potential to confuse the analyses and to simplify thepresentation of the concepts in which we are interested. It must be remembered thatthe assumptions underlying any model determine how good an approximation of realitythat model is. In other words, a good model is one that is a close approximation of thereal world.

To analyze the macroeconomy using the Keynesian Cross some simplifyingassumptions are necessary. We will assume that all investment and net exports aredetermined by factors outside of GDP (exogenously determined), it is also assumedthat government expenditures do initially impact private decisions and all taxes arelump-sum personal taxes (with some exogenous taxes collected, i.e., customs duties,etc.). It is also assumed that there are no monetary affects associated with fiscal policy,that the initial price level is fixed, and that any fiscal policy actions impact only thedemand side of economy, in other words, fiscal policy is not offset by, nor does impactthe monetary side of the economy. It is also assumed that there are not foreign affectsof domestic fiscal policy.

86

The Goals of Government Expenditures

The primary and general goal of discretionary fiscal policy is to stabilize theeconomy. Often other goals, involving public goods and services as well as political areincluded in the discretionary aspects of fiscal policy. However, for present purposes weare concerned only with government expenditures used to stabilize the economy (taxeswill be examined elsewhere in this chapter).

To remedy a recession the government must spend an amount that will exactlymake-up the short-fall in spending associated with that recession. The government canalso mitigate inflation by reducing government expenditures. However, the amount ofincreased expenditures necessary to bring the economy back to full employment issubject to a multiplier effect (the same is true of decreases in governmentexpenditures). Therefore, the government must have substantial information about theeconomy to make fiscal policy work effectively. To determine the proper value of themultiplier the fiscal policy makers must know either the Marginal Propensity to Save orto Consume. Further, the policy makers will need to know the current level of outputand what potential GDP is, (potential GDP is that output associated with fullemployment). In a practical sense, in the near term the government may havereasonably accurate information upon which to base forecasts to conduct fiscal policy. For present purposes, we will assume the government has all the necessary informationat hand to conduct fiscal policy.

The Simple Multiplier

When the government increases expenditures, the effect on the economy ismore than the initial increase in government expenditures. In fact, this is also true ofinvestment and consumption expenditures, not just government expenditures. Whenthere is an increase in expenditures, through an increase in government expenditures,those expenditures become income for someone. They will save a portion of theexpenditures and spend the rest which then become income to someone else. Theresult of this chair-reaction re-spending is that there is a direct relation between the totalamount of re-spending and the marginal propensity to save. This is the re-spendingeffect discussed in Chapter 6.

The multiplier is the reciprocal of the marginal propensity to save:

The multiplier = 1/MPS

Because MPS + MPC = 1, there is an equivalent expression: 1/MPS = 1/1-MPC. Themultiplier is the short-cut method of determining the total impact of an increase or

87

decrease in total spending in the economy. For example, if the government spends $10more and the marginal propensity to consume is .5, then the multiplier is 2 and the totalincrease in spending resulting from the increase of $10 in government expenditures is$20.

This is called the simple multiplier because the only leakage in the re-spendingsystem is savings. In reality there are more leakages. People will use a portion of theirincome to buy foreign goods (imports), and will have to pay income taxes. These arealso leakages and would be added to the denominator in the multiplier equation. TheCouncil of Economic Advisors tracks the multiplier that contains each of these otherleakages, called the complex multiplier. The complex multiplier has remained relativelystable over the past couple of decades is estimated to be about 2.0.

The following diagram presents a case of recession, that will be eliminated byincreasing government expenditures by just enough to eliminate the recession, but notto create inflation. Assuming that our current level of GDP is $350 billion and we knowthat full employment GDP is $430 we which to eliminate this recession using increasesin government expenditures. We also know that MPC is .75 and therefore MPS is .25.

With an MPC of .75 we know the multiplier is 4 (1/.25). We also know that wemust obtain another $80 billion in GDP to bring the economy to full employment. Thedistance between the forty-five degree line and the C+I=G1 line at $430 is $20 billionwhich is our recessionary gap in expenditures. Therefore to close this gap we mustspend $20 billion and the multiplier effect turns this $20 billion increase in governmentexpenditures into $80 billion more in GDP. Another way to calculate this is, that we are$80 billion short of full employment GDP, we know the multiplier is 4, so we divide $80by the multiplier 4 and find the government must spend an additional $20 billion.

The leakages approach yields the same results.

88

An increase in $20 of government expenditures moves the investment plusgovernment expenditures line for I+G1 to I+G2 by a total of $20 billion dollars on theexpenditures axis, but because of the re-spending effects, this increase results in $80billion more in GDP, from $350 billion to $430 billion.

Taxation in Fiscal Policy

The government can close a recessionary gap by cutting taxes, just as effectivelyas it can by increasing government expenditures (bearing in mind, that there will bedifferences in the lag between the actual enactment of law and the impact of the policyon the economy, in general taxes show up later in affect, than do expenditures). Assuming that it is a lump-sum tax the government will use (lump-sum meaning it is thesame amount of tax regardless of the level of GDP). The lump sum tax must bemultiplied by the MPC to obtain the change in consumption, however, such taxes arealso paid proportionately from savings. So the effect is not the same as is observedwith government expenditures. The tax is multiplied by the MPC and the remainderpaid from savings. However, there is also no immediate increase in expenditures. Inother words, the impact of a change in taxes is always less than the impact ofexpenditures. In other words, the taxation multiplier is always the simple multiplierminus one or:

taxation multiplier = (1/MPS) - 1

Consider the following diagrammatic presentation of the impact of decrease intaxes on an economy.

89

If full employment GDP is $600 billion and we are presently at $500 billion withan MPC of .8, then if we are going to increase GDP by $100 billion we must cut taxesby $25 billion. The simple multiplier with an MPC of .8 is 1/.2 = 5; but the taxationmultiplier is the simple multiplier minus one, hence 4. The decrease in taxationnecessary to increase GDP by $100 billion is the $100 billion divided by 4 or $25 billion.

The major difference between increasing expenditures or decreasing taxes isthat the multiplier effect for taxation is less. In other words, to get the same effect onGDP you must decrease taxes by more than you would have had to increaseexpenditures.

Balanced Budget Multiplier

An alternative policy is to increase taxes by exactly the amount you increaseexpenditures. This balanced budget approach can be used to expand the economy. Remember that the simple multiplier results in 1/MPS times the increase inexpenditures, but the taxation multiplier is one less than the simple multiplier. In otherwords, if you increases taxes by the same amount as expenditures, GNP will increaseby the amount of the initial increase in government expenditures. That is because onlythe initial expenditure increases GDP and the remaining multiplier effect is offset bytaxation. Therefore, the balanced budget multiplier is always one.

For example, if full employment GDP is $700 billion, and we are presently at$650 billion, with an MPC of .75, then the simple multiplier is 4, (1/.25) and the taxationmultiplier is 3, [(1/.25)-1], therefore the government must spend $50 billion and increasetaxes by $50 to increase GDP by $50.

90

Tax Structure

Taxation is always a controversial issue. Perhaps the most controversial of alltax issues concerns the structure of taxation. Tax structure refers to the burden of thetax. Progressive taxation is where the effective tax rate increases with ability to pay. Regressive taxation is where the effective tax rate increases as ability to pay decreases. A proportional tax structure is where a fixed proportion of ability to pay is taken in taxes.

In general consumption is more greatly effected by taxation when the tax isprogressive, and savings are impacted more when the tax is regressive. Therefore, ifthe distribution of the tax across income groups will have a variable impact on theconsumption and savings.

At present, the federal income tax structure is nearly proportional, most advalorem taxes, tobacco etc., are regressive. State income tax structures also varysubstantially. States like Kansas, California and New Jersey have mildly progressiveincome taxes. States like Indiana and South Carolina use gross income tax schemesthat tend to be very regressive. Therefore, the current tax structures are not neutralwith respect to their re-distributional effects across income groups. In total, the taxescollected across the federal, state and local level are at best proportional and areprobably slightly regressive. Probably the most regressive of these taxes is thegasoline tax.

Automatic Stabilizers

During the New Deal period several social welfare programs were enacted. Thepurpose of these programs was to provide the economy with a system of automaticstabilizers to help smooth business cycles without further legislative action. Amongthese programs were:

(1) progressive income taxes,

(2) unemployment compensation, and

(3) government entitlement programs.Since the end of the Carter administration these automatic stabilizers have

become controversial. President Clinton moved in his first term in office to eliminatesome of these programs, but it was really the Republican congress and Governors likeTommy Thompson in Wisconsin who eliminated much of then welfare programs. Since2002 much of this “welfare reform” has also come under fire for not having produced theresults that were advertised. In 2002 the movie “Bowling for Columbine” by MichaelMoore, brought many of these issues to an international audience.

91

The idea behind a progressive income tax was basic fairness. Those with thegreatest wealth and income have the greatest ability to pay and generally receive morefrom government. As recessions occur, it is that segment of the population with thegreatest wealth that will have resources upon which to draw to pay taxes, the poor willgenerally be impacted the most by high level of unemployment and recessionsgenerally leave them with very little ability to pay.

Unemployment compensation is paid for in every state of the union by a payrolltax. The payroll tax is generally less than one percent of the first $10,800 of payrolls. Most states also impose an experience rating premium, that is those companies thathave laid people off in the last year will pay a higher rate based on their experience. The preponderance of this tax is paid during periods of expansion and is placed in atrust fund. Unemployment benefits are then paid from this trust fund as the economyenters a recession. The effect is that money is taken out of the system duringexpansion and injected during recession which dampens the top of the cycle (peaks)and eases the bottom of the cycle (trough).

Most social welfare programs have essentially the same effect, except that theexpansions tend not to be dampened as much because the funding comes from generalbudget authority rather than a payroll tax. The significant increases in the proportion ofpoor people during recessions, however, do not add as much to the downward spiral ofunderspending, that would have otherwise been observed in the absence of theseentitlement programs.

Problems with Fiscal Policy

There are several serious problems with fiscal policy as a method of stabilizingthe macroeconomy. Among these problems are (1) fiscal lags, (2) politics, (3) thecrowding-out effect, and (4) the foreign sector. Each of these will be addressed, in turn,in the following paragraphs.

Fiscal Lag

There are numerous lags involved with the implementation of fiscal policy. It isnot uncommon for fiscal policy to take 2 or 3 years to have a noticeable effect, afterCongress begins to enact corrective fiscal measures. These fiscal lags fall into threebasic categories.

There is a recognition lag. The recognition lag is the amount of time for policymakers to realize there is an economic problem and begin to react. Administrative lags

92

are how long it takes to have legislation enacted and implemented. Operational lagsare how long it takes for the fiscal actions to effect economic activities.

Because of the typical two to three years for fiscal policy to have its intendedeffects, they may cause as many problems as they cure. For example, it is notuncommon for the Congress to cut taxes because of a perceived recession thatsubsequently ends within months of the enactment of the legislation. When the effectsof the fiscal policies actually effect the economy, it may be in a rapid expansion and thetax cut or increase in government expenditures add to inflationary problems.

Politics, Crowding-out, the Foreign Sector and Fiscal Policy

Often politics overwhelms sound economic reasoning in formulating fiscalpolicies. Public choice economists claim that politicians maximize their own utility bylegislative action, and are little concerned with the utility of their constituents. Perhapsworse, is the fact that most bills involve log-rolling and negotiations. Special interestoften receive benefits simply to because they pay many of the election costs, and theinterests of these lobbyists may be inconsistent with the best interests of the nation as awhole. The end result is that politics confounds the formulation of policy designed todeal with technical ills in the economy.

The neo-classicist have long argued that government deficits (often associatedwith fiscal policy) results in increased interest rates that crowds-out private investment. there is little empirical evidence that demonstrates the exact magnitude of thiscrowding-out effect, but there is almost certainly some small element of this.

The neo-classicist also argue that there is another problem with governmentborrowing to fund deficit financing of fiscal policies. This problem is called RicardianEquivalence. David Ricardo hypothesized that the deficit financing of government debthad the same effect on GDP as increased taxes. To the extent that capital markets arenot open (foreign investors) the argument is plausible, however, in open economiesthere is little empirical evidence to support this view.

There are also problems that result from having an open economy. The mosttechnical of these problems is the net export effect. An increase in the interest ratedomestically (associated with a recession, or with an attempt to control inflation) willattract foreign capital, but this increases the demand for dollars which increases theirvalue with respect to foreign currencies. As the value of the dollar increases it makesU.S. goods more expensive overseas and foreign goods less expensive domestically. This results in a reduction of net exports, hence a reduction in GDP.

There have also been shocks to the U.S. economy that have their origins outsideof the United States and are difficult if not impossible to address with fiscal policy. The

93

Arab Oil Embargo is a case in point. The United States and Holland supported theIsraeli in their war with the Arabs in the early 1970s. The problem was we also hadtreaty obligations to some of the Arab states. Because of our support, the Arabsembargo oil shipments to the United States and Holland which had the result ofincreasing domestic oil prices and decreased aggregate supply, hence driving up theprice level. This all occurred because of American Foreign Policy, but little could bedone with fiscal policy to offset the problems for aggregate supply caused by theembargo.

KEY CONCEPTS

Social Welfare Programs

1946 Employment Act

Politics and change

Expenditures

Expansionary v. Contractionary Fiscal Policy

Political Goals

Public Goods and Services

Taxation

Expansionary v. Contractionary Fiscal Policy

Multipliers

Simple

Taxation

Balanced Budget

Tax Structure

Proportional

Regressive

Progressive

94

Automatic Stabilizers

Progressive Income Taxes

Unemployment Compensation

Entitlement Programs

Fiscal Lags

Recognition

Administrative

Operational

Politics and Fiscal Policy

Log-rolling

Public Choice Economics

Government Deficits

Crowding-out

Ricardian Equivalence

Open Economy

STUDY GUIDE

Food for Thought:

Develop the expenditures - output model and show an increase (decrease) in taxes toclose an inflationary (recessionary) gap. Now, do the same thing using increases anddecreases in government expenditures. Do the exercise one more time using thebalanced budget approach. [do this exercise using various MPCs].

Critically evaluate fiscal policy as an economic stabilization policy.

95

Critically evaluate the Ricardian Equivalence Theorem, and be careful to explain itsimplications for the national debt.

Which is most reliable as a fiscal policy tool, taxes or expenditures? Defend youranswer.

Sample Questions:

Multiple Choice:

With an inflationary gap of $100 million and a large budget deficit which has become aserious political issue and an economy with an MPC of .95, what would you do to closethe gap?

A. Increase taxes $5 millionB. Decrease expenditures $5 millionC. Increase taxes $100 and decrease expenditures $100D. None of the above

With a recessionary gap of $70 million and an MPS of .1 which of the following policywould close the gap?

A. Increase taxes and expenditures by $70 millionB. Increase expenditures by $7 millionC. Decrease taxes by $7.8 millionD. All of the above will work

True - False:

96

The crowding - out effect is the theory that a government deficit raises interest rates andabsorbs resources that could have been used for private investment. {TRUE}

Automatic stabilizers, such as unemployment compensation, provide counter cyclicalrelief from economic instability without additional government action. {TRUE}

97

Chapter 8

Money and Banking

The “real” economy has been the subject of most of the analysis to this point. There is, however, another part of this story. That other part of this story is money. Notice to this point we have assumed away any difficulties that may arise because ofthe impact of policy on monetary aggregates. This chapter is the first of three which willdevelop monetary economics.

In primitive, tribal societies the development and use of money occurs only afterthat society reaches a size and complexity where barter is no longer a viable method oftransacting business. Barter, the trading of one good or service for another, requires acoincidence of wants. When one individual has something another wants, and viceverse, trade can be arranged be there is a coincidence of wants.

Larger, more complex social orders generally require the division of labor andspecialization, which in turn, increases the number of per capita market transactions. When individuals live in a higher interdependent society, most necessities of life areobtained through market transactions. In a modern industrialized country it is notuncommon for an individual to make more than a dozen transactions per day. Thisvolume of business makes barter nearly impossible. The result is that societies willdevelop money to facilitate the division of labor and specialization that provide for higherstandards of living.

The purpose of this chapter is to introduce the reader to money and to thebanking system. This chapter will provide the basic definitions essential tounderstanding our complex monetary and banking systems. The following chapters willextend the analyses to demonstrate how money is created and how the monetarysystem is used to stabilize the fluctuations in the business cycle.

Functions of Money

There are three functions of money, these are:

(1) a medium of exchange,

(2) a measure of value, and

(3) a store of value.

98

Each of these functions will be examined, in turn, in the following paragraphs.

Money serves as a medium of exchange. It is generally accepted as "legaltender" or something of general and specified value, such that, people have faith thatthey can accept it for payment, because they can use it in exchange without loss ofvalue. Because barter requires a coincidence of wants, trade occurs only when twopeople have different commodities that the other is willing to accept in trade for theirown wares. A barter economy makes exchange difficult, it may take several trades inthe market before you could obtain the bread you want for the apples you have.

Money solves the barter problem. If you have apples and want bread, you simplysell the apples for money and exchange the money for bread. If barter persists it maytake a dozen or more transactions to turn apples into bread. In other words, money isthe grease that lubricates modern, sophisticated economic systems.

Money is also a measure of value. Without money as a standard by which togauge worth, value would be set by actual trades. The value of a horse in eighteenthcentury Afghanistan could be stated in monetary units in the more modern areas of thecountry. However, the nomads that wandered the northern plains of that country couldtell you in terms of goats, carpets, skins, and weapons what the range of values werefor horses. However, there were as many prices of horses as there were combinationsof goods and services that could be accepted in exchange for the animal. Moneypermits the value of each commodity to be stated in simple terms of a single anduniversally understood unit of value.

Money is also a store of value. Money can be saved with little risk, with virtuallyno chance of spoilage, and little or no cost. Money is far easier to store than areperishable foodstuffs, or bulky commodities such as coal, wool or flour. To store money(save) and later exchange it for commodities is far more convenient than having to storecommodities for future use, or to have to continually go through barter exchanges.

Together, these functions vest in money a critical role in any complex moderneconomy. Our prices are stated in terms of money, our transactions are facilitated bymoney, and we can store the things we which to consume in the future by simply savingmoney. In fact, money may not make the world go 'round, but it certainly permits theeconomic world to go 'round much more smoothly.

The Supply of Money

The supply of money has a very interesting history in both U.S. economic historyand world economic history. Several historians note that one of the contributing factorsto the fall of the Roman Empire was that there was significant deflation in the third andfourth centuries. The reason for this was that money, in those days, was primarily

99

coinage minted of gold, silver, and copper. As gold and silver was traded forcommodities from the orient there was a flow of coinage out of the west. In addition,"barbarians" were constantly raiding Roman territory and it was the gold and silver thatthey carried back with them (for trade). Further, as the population grew at a faster ratethan the availability of precious metals, the money supply fell relative to the need for it tomake the economy function efficiently. This rapid deflation, added to a extremelymaldistributed income, and loss of productive resources resulting in a rapidly decliningeconomy after the second century A.D. The Roman economy collapsed, with thecollapse of the economy the military and government were also doomed. Given thetimes, once the Roman government and military were ruined, it was short-work toeliminate the empire and social structure.

In 1792, the U.S. Congress enacted the first coinage act. The Congressauthorized the striking of gold and silver coins. The Congress set the ratio of the valueof gold to silver in the coinage at 15 units of silver equaled one unit of gold. Theproblem with this was that gold was worth more than silver as bullion than as coinage. This arbitrary setting of the coinage value of these metals resulted in the gold coinsdisappearing from circulation (being melted down as bullion) and only silver coinscirculating. At the same time, most coinage that circulated in the eighteenth centurywestern hemisphere (including the United States) was of Spanish origin. In fact, thestandard unit adopted for U.S. coinage was the dollar, however, the Spanish mintedcoins that were also called dollars (from a Dutch word, tolar). The Spanish one dollarcoins contained more silver, than did the U.S. dollar and the Spanish coins were beingmelted down and sold, at a profit, at the U.S. mint. Herein is the problem with usinggold or silver as minted coins or as backing for currency. Money has value in exchangethat is unrelated to the value of precious metals. Their relative values fluctuate andresult in money disappearing if the value of the metal is more than the unit of currencyin which the metal is contained.

One of the classical economists noted this volatility in the monetary history ofmost countries. As the value of the gold or silver made the currency worth more asbullion, that currency disappeared and was quickly replaced by monetary devices oflessor value. Gresham's Law is that money of lesser value will chase money of greatervalue out of the market.

A modern example of this is available from U.S. coinage. In 1964, the value ofsilver became nearly 4 times (and later that decade nearly 22 times) its worth incoinage. Therefore, the last general circulation coins that were minted in the UnitedStates that contained any silver was 1964. Today, the mint issues one dollar coins thatcontain one ounce of silver (rather than .67 ounces), but that silver is worth $5.30 perounce as bullion. Therefore, you do not see actual silver dollars in circulation, and willnot until the value of silver drops below $1.00 per ounce. This observation is Gresham'slaw in operation.

Since the American Civil War there have been various forms of money used in

100

this country other than coinage. The government issued paper money, particularly afterthe Greenback Act of 1861 (there were numerous examples of U.S. paper moneybefore that year, including bank notes, state notes, and colonial notes). The GreenbackAct provided for the denominations of bills with which you are familiar, but it alsoprovided for 5¢ 10¢, 25¢, and 50¢ bills (fractional currency).

Today, there are numerous definitions of money. The most commonly usedmonetary items are included in the M1 through M3 definitions of money; thesedefinitions are: (1) M1 is currency + checkable deposits, (2) M2 is M1 + noncheckablesavings account, time deposits of less $100,000, Money Market Deposit Accounts, andMoney Market Mutual Funds, and (3) M3 is M2 + large time deposit (larger than$100,000). The largest component of the M1 money supply is checkable deposits(checking accounts, credit union share drafts, etc.), currency is only the second largestcomponent of M1.

The presented above are from the historical tables at the Federal Reserve’s sitewww.Federalreserve.gov. The Fed routinely publishes money supply data on a weekly,monthly, and annual average basis, both seasonally adjusted, and not seasonallyadjusted. As the above box demonstrates, there is a fairly stable relationship betweeneach of the three major definitions of the money supply. The growth in each of thecategories over the calendar year 2005 (on a seasonally adjusted basis) is alsorelatively slow.

Monetary Aggregates, Monthly 2005 (Seasonally adjusted)

Month M-1 M-2 M-3January 1357.2 6439.3 9487.2February 1363.0 6446.9 9531.6March 1368.0 6464.9 9565.3April 1360.4 6471.9 9620.9May 1366.8 6482.8 9665.0June 1359.8 6502.6 9725.3July 1355.2 6515.4 9762.4August 1359.1 6548.3 9864.6September 1355.1 6580.2 9950.8October 1363.8 6612.9 10032.0 November 1363.1 6641.0 10078.5December 1363.2 6663.9 10154.0

Source: Federal Reserve System

101

Near Money

Near money are the items that fulfill portions of the requirements of the functionsof money. Near money can be simply a store of value, such a stocks and bonds thatcan be easily converted to cash. Credit cards are often accepted in transactions, andthe line of credit they represent can serve as a medium of exchange. Gold, silver, andprecious stones have historically served as close substitutes for money because thesecommodities have inherent value and can generally be converted to cash in almost anyarea of the world. Much of the wealth smuggled out of Europe at the end of World WarII by escaping Nazis was smuggled out in the form of gem stones, gold, and silverbullion.

The widespread use of near money is relatively rare in the world's economichistory. However, in modern times, there is a potential for problems. Indiana's historypresents an interesting example. The State of Indiana issued currency in the 1850s, inpart to help finance the canals that were being built across the State. In 1855 and 1856,the railroads put the canals out of business, even before they were completed, virtuallybankrupting the State, hence the 1857 Constitution, rather the year Indiana entered theUnion. State currency or even private bank currency is not controlled by a central bankand is worth only what faith people have in it or the intrinsic value of the monetary unit. If the full faith and credit of a bankrupt State or bank is all that backs the currency, it isworthless.

What Gives Money Value?

The value of the U.S. dollar (or any other currency) can be expressed as thesimple reciprocal of the price level:

D = 1/P

Where D is the value of the dollar and P is the price level. In other words, the value ofthe dollar is no more or less than what it will buy in the various markets. This is true ofany currency (money in general). However, there are reasons why money has value. The value of money is determined by three factors, these factors are: (1) its general acceptability for payment, (2) because the government claims it is legal tender (hencemust be accepted for payment), and (3) its relative scarcity (as a commodity).

Money can be used to buy goods and services because people have faith in itsgeneral acceptability. It is not that the coin or paper has intrinsic value that makes money of value in exchange, it is simply because people know that they can accept it inpayment and immediately exchange it for like value in other commodities, because

102

virtually everyone trusts its value in exchange.

Money also has value in international currency markets, not just domestic ones. The international currency markets provide a very good example of why trust providesmoney with value. The world's currencies are generally divided into two categories,hard currency and soft currency. A hard currency is one that can be exchanged forcommodities in any nation in the world. A soft currency is one whose value is generallylimited to nation that issued it, and often to some limited extent in the world currencymarkets (often with significant limitations or even discounts). The U.S. dollar, FrenchFranc, German Deutsche Mark, Canadian Dollar, Japanese Yen, British Pound Sterling,and Italian Lire were recognized as hard currencies (generally the Swiss Franc is alsoincluded in the hard currencies), at least prior to the Euro. These seven nations are theG-7 nations and are the world's creditor nations (even though each has a national debtof their own, the private sector extends credit to less developed countries). The reasonthat these countries are the creditor nations is that they are the largest free marketeconomies, have democratic and stable governments, and long histories of ratherstable financial markets. These nations' currencies are termed "hard" currenciesbecause they are relatively stable in value and can be readily exchanged for the goodsand services of these largest most advanced economies. In other words, the economicsystems and governments that generate these currencies are the markets from whicheveryone else imports a wide array of necessary commodities.

On the other hand, the Mexican Peso, Kenyan Dollar, and Greek Drachma(among 165 others) are currencies of less developed nations that have very little ofvalue, relative to any of the G-7 nations in the world's markets, do not have histories ofstable financial markets, governments, and they are typically in debt to the G-7 nations. Because of their limited value in exchange, and rather volatile value these currenciesare called "soft" currencies. The difference between a hard and a soft currency is trustin its present and future value in exchange for commodities. Hard currencies aregenerally trusted, hence accepted, soft currencies are not.

There is also an element of legality in the value of money. For example, theUnited States is a large, economically powerful country. Its government is also a large,powerful government that has always paid its bills. People have faith and trust in theU.S. government making good on its financial obligations, therefore people have takennotice when the United States government says that Federal Reserve Notes are legaltender.

Also contributing to the value of money is its scarcity. Because money is ascarce and useful commodity it also has value the same as any other commodity. It isinteresting to note that the U.S. $100 bill is the most widely circulated currency outsideof the United States, and more of these bills circulate outside of the U.S. than within theU.S. This suggests something of the commodity value of U.S. dollars, as well as thegeneral international trust in the U.S. dollar.

103

By spring 1971, the exchange rate problems had become acute. This wastrue for Japan and especially for West Germany, who had large trade surpluses withthe United States and held more dollars than they wanted. Under these conditions,the U.S. dollar was rapidly losing value against the German Deutsche mark. FromJanuary to April of 1971, in keeping with the Bretton Woods agreements, the GermanBundesbank had to acquire more than $5 billion of international reserves in order todefend the value of the dollar. To protect the value of the dollar and its exchangerate with the German mark, however, the German central bank was losing controlover its domestic monetary policy. By May 5, 1971, the Bundesbank abandoned itsefforts to protect the dollar and permitted the Deutsche mark to seek its own value inthe world's currency markets. The scenario was the same for all countries that hadtrade surpluses with the United States. The excess supply of dollars was causingthose countries to lose control over their money supplies. Given that the UnitedStates was rapidly losing its gold reserves, on August 155, 1971, by Richard Nixon'sorder, it was announced that the United States officially abandoned the BrettonWoods system and refused to exchange gold for U.S. dollars held by foreigners. Forthe first time in modern monetary history, the U.S. dollar was permitted to seek itsvalue in open markets. The move to a flexible exchange rate system where theexchange rates are determined by the basic market forces was the official demise ofthe Bretton Woods system.

Mashaalah Rahnama-Moghadam, Hedayeh Samavati, and David A. Dilts, DoingBusiness in Less Developed Countries: Financial Opportunities and Risks. Westport,Conn: Quorum Books, 1995, p. 74.

The Demand for Money

The demand for money consists of two components of total money demand,these are: (1) transactions demand, and (2) asset demand. Transactions demand formoney is the demand that consumers and business have for cash (or checks) toconduct business. Transaction demand is related to a preference to have wealth orresource in a form that can be used for purchases (liquidity). There is also an assetdemand for money. In times of volatility in the stock and bond markets, investors mayprefer to have their assets in cash so as not risk losses in other assets. Together, theasset and transaction demand for money comprise the total demand for money.

Money is much the same as any other commodity, it has a demand curve and aprice. The price of money is interest, primarily because money is also a claim on capitalin the financial markets. The demand curve for money is a downward sloping function,

104

that is a schedule of interest rates to the quantity of money.

The Money Market

The money market is a particularly interest market. In reality there are severalmarkets in which money is exchanged as a commodity. In examining only M1, thecurrencies of various countries are exchanged for one another for the purpose of doingbusiness across national boundaries. The price of one nation's currency is generallyexpressed in terms of another countries currency. For example, 105 Yen is the value ofa dollar, in the case of a Deutsche Mark, 1.26 DM is worth one dollar.

There is also the credit market. The credit market is where consumers andbusinesses go to borrow money. A consumer purchasing a house will typically need amortgage and will borrow to buy a house. Businesses will need to borrow to purchasecapital equipment (investing) to produce commodities to sell in product markets. Bothtypes of borrowing influence the credit markets, because money is a relatively scarcecommodity. One of the largest borrowers in the U.S. economy is the U.S. Treasury. Typically the government borrows by selling Treasury Bonds.

The following diagram is for a general money market (credit market):

The money supply curve is vertical because the supply of money is exogenouslydetermined by the Federal Reserve. The Federal Reserve System regulates the moneysupply through monetary policy and can increase or decrease the money supply by thevarious actions it has available to it in regulating the banking system and in selling orbuying Treasury Bonds. The money demand curve slopes downward and to the right. The intersection of the money demand and money supply curves represents equilibrium

105

in the money market and determines the interest rate (price of money).

Bonds are financial obligations. Both private companies and governments issuebonds and receive cash. The bonds typically state that the owner of the bond willreceive a specific payment in dollars periodically for holding the bond, and at the end ofthe bond's life it will be redeemed for its face value. This is the primary market, wherebonds are sold directly by the government or company. In the case of the Treasury thebonds are generally sold at auction.

This method of paying interest creates a "secondary" market for bonds. Bondsmay be resold for either a discount or a premium. As the market value of the bondincreases, it drives down the rate of return on the bond, conversely, if the market valueof the bond decreases the rate of return increases. For example, consider a $1000bond that the government agrees to pay $60 per year in interest over its life. If the bondremains at the $1000 face value the interest rate is 6%. However, if bonds are viewedas a safer investment than other possible investments, or there is excess demand forbonds, the market price may increase. If the market price of this $1000 bond increasesto $1200 then the rate of return falls to only 5% (60/1200 = .05). On the other hand, ifthe bond is viewed as more risky, or there is an excess supply of bonds the marketprice may fall, say to $800, then the rate of return increases to 7.5% (60/800 = .075).

Notice how bonds become a good investment. Bonds are good investmentswhen the interest rate is falling. As the interest rate falls, the market value of the bondincreases, (remember that the payment made by the original borrower is a fixedpayment each period). In other words, falling interest rates mean larger market valuesfor the bonds, and greater profits for investors in bonds.

An Overview of the U.S. Financial System

The U.S. financial system is a complex collection of banks, thrifts, savings &loans, credit unions, bond and stock markets, and numerous markets for other financialinstruments such as mutual funds, options, and commodities. The complete analysis ofthese markets is not a single course, its an entire curriculum called Finance. What isimportant for understanding the basics of the effects of money on the macroeconomy isthe banking system and the closely associated regulatory agencies, such as the FederalReserve System and Federal Deposit Insurance Corporation (F.D.I.C.).

Federal Reserve System (FED) is comprised of member banks. These memberbanks are generally large, nationally chartered banks that do significant amounts ofcommercial banking. The FED is owned by these member banks. However, under theFederal Reserve Act, the Board of Governors and Chairman are nominated by thePresident of United States and confirmed by the Senate. The structure of the system is:(1) Board of Governors - that governs the FED and is responsible for the operations of

106

the Fed, (2) Open Market Committee - buys and sells bonds (called open marketoperations, (3) Federal Advisory Council - provides advise concerning appropriatebanks regulations and monetary policies, and (4) The FED has 12 regional banks thatserve as check clearing houses, conduct research, and supervises banks within theregion. Fort Wayne is in the Chicago region, however, Evansville is in the St. Louisregion. The Federal Reserve Banks are for each of the 12 regions are located in thefollowing cities:

1. Atlanta, Georgia2. Boston, Massachusetts3. Chicago, Illinois4. Cleveland, Ohio5. Dallas, Texas6. Kansas City, Missouri7. Minneapolis, Minnesota8. New York, New York9. Philadelphia, Pennsylvania

10. Richmond, Virginia 11. San Francisco, California 12. St. Louis, Missouri

The functions of FED are basically associated with bank regulation and theconduct on monetary policy. The functions of the FED include:

1. Set reserves requirements,

2. Check clearing services,

3. Fiscal agents for U.S. government,

4. Supervision of banks, and

5. Control money supply through Open Market Operations (buying andselling of bonds).

The supervision of banks and check clearing services are routine FED functionsthat are focused on making the banking system safer and more efficient. Because theFED is the agency through which Treasury obligations are bought and sold the FED isthe fiscal agent for the federal government. The setting of reserve requirements for thebanking system and open market operations are the tools of monetary policy and arethe subjects of the following chapter.

The Federal Deposit Insurance Corporation is a quasi-governmental corporationwhose purpose is to insure the deposits of member banks. Credit unions, thrifts, andsavings & loans had separate independent agencies designed to provide the same

107

insurance. However, after the savings & loans crisis, these other agencies wereconsolidated under the control of F.D.I.C. The reason for these programs was theexperience of the banking industry during the Great Depression when many depositorslost their life savings when the banks failed. Many banks failed, simply because of"runs." Runs are where depositors demand their funds, simply because they have lostfaith in the financial ability of the banks to meet their obligations (sometimes the loss offaith was warranted, often it was nothing more than panic). Therefore, to fosterdepositor faith in the banking system, F.D.I.C. was created that provided a guaranteethat depositors would not loss their savings even if the bank did fail.

There is a problem with such insurance arrangements. This problem is calledmoral hazard. Moral hazard is the effect that having insurance reduces the insured'sincentive to avoid the hazard against which they are insured. The savings and loancrisis was at least in part the result of moral hazard. The managers of the failed savingand loans often would extend loans or make investments that were high risk, but wereless concerned because if it resulted in failure, the government would pay-back thedepositors. This is a classic example of moral hazards, but there were also otherproblems involving fraud, bad loans in Mexico, and shaky business practices, againstwhich it was not intended that F.D.I.C. would risk substantial exposure.

KEY CONCEPTS

Functions of Money

Medium of Exchange

Avoidance of Barter

Measure of Value

Store of Value

Supply of Money

M1, M2, and M3

Near Money

Value of Money

Demand for Money

Transactions Demand

108

Asset Demand

Total Demand

Money Market

Interest rates

Federal Reserve System

Board of Governors

F.M.O.C.

Federal Advisory Council

12 regions

Functions of the Fed

Sets Reserve Requirements

Check clearing

Fiscal agent for the U.S. government

Supervision of Banks

Control of Money Supply

Moral Hazard

STUDY GUIDE

Food for Thought:

Develop and explain each:

Functions of money,

Demand for money,

Money supply, and

109

Value of money.

Critically evaluate the Federal Reserve System as a regulatory agency.

What is near money? Explain and critically evaluate the use of near money.

Sample Questions:

Multiple Choice:

The U.S. money supply is backed by:

A. Gold B. Gold and SilverC. Gold, silver & government bondsD. People's willingness to accept it

If a U.S. government 30-year bond sold originally for $1000 with a specified interest rateof five percent, each year the bond holder receives $50 from the government. If thesales price of the bond falls to $900 what is the interest rate?

A. 5.56%B. 5.00%C. 4.50%D. None of the above

110

True - False:

If the Fed wishes to decrease the money supply they can buy bonds {FALSE}.

The FDIC provides insurance for deposits in member banks {TRUE}.

111

Chapter 9

Multiple Expansion of Money

The purpose of this chapter is to analyze how banks create money and how theFederal Reserve System plays an essential role in regulating the banking system. Aftera brief history of banking reserves is examined, we will proceed to analyze the multipleexpansion of money.

Paper money is produced at the United States Bureau of Printing and Engravingin Washington, D.C. Each day, the Bureau of Printing and Engraving produces about$22 million in new currency. Once the Bureau of Printing and Engraving produces themoney it is then shipped to the twelve regional Federal Reserve banks for distribution tothe member banks. The stock of currency is created and maintained by a simpleprinting and distribution process. However, the stock of paper money is only a smallpart of the M1 money supply. The majority of the M1 money supply is checkabledeposits.

Counterfeiting is of U.S. currency is a significant problem. The U.S. Bureau ofPrinting and Engraving has developed a new fifty dollar bill that will make counterfeitingmore difficult. Periodically for the next several years the new design will be extended toall denominations of U.S. currency. A few years ago, a strip was added to U.S.currency that when held to the light shows the denomination of the bill, so thatcounterfeiters cannot use paper from one dollar bills to create higher denomination bills. The government agency responsible for enforcing the counterfeiting laws in the UnitedStates is the U.S. Secret Service, the same agency that provides security for thePresident.

Assets and Liabilities of the Banking System

Assets are items of worth held by the banking system, liabilities are claims ofnon-owners of the bank against the banks' assets. Net worth is the claims of theowners of the bank against the banks' assets. Over the centuries a system of doubleentry accounting has evolved that presents images of businesses. The double entrysystem accounts for assets, liabilities, and net worth.

Accounts have developed a balance sheet approach to present the double entryresults of the accounting process. On the left hand side are entered all of the bank'sassets. On the right hand side of the ledger are entered all of the claims against thoseassets (claims by owners are net worth and claims by non-owners are liabilities). The

112

assets side of the ledger, must equal the net worth and liabilities side. This rathersimple method, is an elegant way to assure that claims and assets balance.

The balance sheet method will permit us a method to track how banks createmoney through the multiple expansion process.

Rational for Fractional Reserve Requirements

The fractional reserve approach to monetary stability dates from the middle-agesin Europe. Goldsmiths received gold to make jewelry, religious objects, and to hold forfuture use. In return the goldsmiths would issue receipts for the gold they received. Inessence, these goldsmith receipts were the first European paper money issued andthey were backed by stocks of gold. The stocks of gold acted as a reserve to assurepayment if the paper claims were presented for payment. In other words, there was a100% reserve of gold that assured the bearer of the receipt that the paper receipt wouldbe honored. The reserves of gold held by the goldsmiths created faith in the receipts asmediums of exchange, even though there was no governmental involvement in theissuing of this money.

However, the goldsmiths in Europe were not the first to issue paper money. Genghis Khan first issued paper money in the thirteenth century. Genghis did not holdreserves to back his money, it was backed by nothing except the Khan's authority(which was absolute). Therefore in the case of the Great Khan, it was the ability topunish the untrusting individuals that gave money its value. In Europe, two-hundredyears later, it was trust in reserves that gave money its value.

The U.S. did not have a central banking system, as we know it, from the 1840sthrough 1914. There were two early "national" banks whose purpose was to serve asthe fiscal agent of the U.S. government and to provide limited regulation for the U.S.monetary system. Both failed and were eliminated. In the early part of this centuryseveral financial panics pointed to the need for a central banking system and for strongfinancial regulations.

During the first half of this country's history both states and private companiesissued paper money. Mostly this paper money was similar to the gold receipts issuedby the European goldsmiths, except the money was not backed by gold, typically themoney was a claim against the assets of the state or company, in other words, themoney issued represented debt. It is little wonder that most of this currency becameworthless, except as collectors' items. Prior to 1792, Spanish silver coins were widelycirculated in the U.S. because they were all that was available for use as money (formore details see the previous chapter).

The first widespread issuance of U.S. paper money was during the Civil War

113

(The Greenback Act), which included fractional currency (paper dimes & nickels!). Earlier attempts to issue U.S. notes were less than successful, simply because peopletrusted coinage because of the silver and gold therein contained, and paper money wasa novelty (but not a very valuable one).

Today, the Federal Reserve requires banks to keep a portion of its deposits asreserves, to help assure the solvency of the bank in case of a financial panic, like thoseexperienced in the first decade of this century and again in the 1930s. The fact thatthese reserves are kept also helps to assure the public of the continuing viability of theirbanking system, hence the safety of their deposits. In turn, this public faith shouldprevent future runs on the banking system that have historically caused so mucheconomic grief due to bank failures.

The Required Reserve Ratio

The Required Reserve Ratio (RRR) is set by the FED's Board of Governorswithin limits set by statute. The minimum legally allowable RRR is three percent, wherethe current RRR has been set by the Board of Governors. The RRR determines by howmuch the banking system can expand the money supply. The RRR is the amount ofreserves that a bank must keep, as a percentage of their total liabilities (deposits).

Banks are permitted some freedom to determine how their reserves are kept. Abank can keep reserves as vault cash or deposits with the regional Federal Reservebank. Should a bank be short of the amount required to meet the reserves necessary,then a bank can borrow their reserves for short periods from either the FED or othermember banks. The FED regulates the borrowing of reserves, and sets an interest ratefor these short term loans if they are borrowed from the FED. The rate charged onborrowed reserves from the FED is called the discount rate. The rate of interestcharged on reserves borrowed from other member banks is called the Federal FundsRate (currently about 5.5%).

The banking system has three forms of reserves, these are actual, required, andexcess reserves. Actual reserves are the amounts the banks have received in depositsthat are currently held by the bank. The required reserves are the amounts the Board ofGovernors requires the banks to keep (as vault cash, deposits with the FED, orborrowed). The excess reserve is amount of actual reserves that exceeds the requiredreserves. It is the excess reserves of the banking system that may be used by themember banks to expand the checkable deposits component of the U.S. money supply.

114

Multiple Expansion of Checkable Deposits

The largest component of the M1 money supply is cheackable deposits. Ratherthan printing Federal Reserve Notes, the majority of the money supply is createdthrough a system of deposits, loans, and redeposits. Money is created by a bankreceiving a deposit, and then loaning that non-required reserve portion of the deposit(excess reserve), which, in turn, is deposited in another checking account, and loansare subsequently made against those deposits, after the required reserve is deductedand placed in the bank’s vault or deposited with the FED.

For example, if the RRR is .10, then a bank must retain 10% of each deposit asits required reserve and it can loan the 90% (excess reserves) of the deposit. Themultiple expansion of money, assuming a required reserve ratio of .10, can, therefore,be illustrated with the use of a simple balance sheet (T-account):

Deposit Loans________________________

|$10.00 | 9.00 9.00 | 8.10

8.10 | 7.29 . | . . | . _______ | _______ $ 100.00 | $90.00

| 10.00 (required reserves)| $100.00

The total new money is the initial deposit of $10 and an additional $90 of multipleexpansion for a total of $100.00 in new money. The T-account used to illustrate thismultiple expansion of money is really a crude balance sheet for the banking system withthe liabilities on the left side and the assets on the right side of the ledger. Notice,however, that the T-account balances, and that there is $100.00 on each side of theledger.

There is a far easier way to determine how much the money supply can beexpanded through the multiple deposit - loan, re-deposit - loan mechanism. This short-cut method is called the money multiplier (Mm). The money multiplier is the reciprocalof the required reserve ratio:

115

Mm= 1/RRR

The money multiplier is the short-hand method of calculating the total entries inthe banking systems' T-accounts and shows how much an initial injection of money intothe system can generate in total money supply through checkable deposits. One of thetools of the FED to expand or contract the money supply is to make or withdrawdeposits from its member banks. This element of monetary policy will be discussed ingreater detail in the following chapter. The following chapter examines monetary policyand how the FED operates to maintain control over the money supply.

The potential creation of money is therefore inversely related to the requiredreserve ratio. For example, with a required reserve ratio of .05 the money multiplier is20. This means that a $1.00 increase in deposits can potentially create $20 in newcheckable deposits as it is loan and re-deposited through the system. On the otherhand, with a required reserve ratio of .20 the money multiplier is 5 and only $5 of newmoney can be created from an initial deposit of $1.00.

With the current required reserve ratio of .03, the money multiplier is 33.33. Currently, an initial deposit of $1.00 can potentially create $33.33 in new money throughincreased checking deposits. The reason that the word potential is used to describethis process, is that there is no guarantee that the banking system will be able to loan allof its available excess reserves. The amount that will be loaned still depends on thedemand for money and investment in plant and equipment.

The monetary history of several nations illustrates how well the multipleexpansion of money has been understood. The United States has had recurrent boutsof inflation during the post-World War II period. As the Fed struggled withunderstanding how the system worked in this country, the Swiss understood since theturn of the century. Switzerland is a relatively small economy, and the money supply inthis small nation was very competently managed. The result is that the Swiss haveexperienced usually stable price levels ever since 1945.

The recurrent bouts of inflation in the post-World War II economic history of theUnited States, are not consistent with the economic history of this country prior to WorldWar II. Most of American history experienced significant deflation. Deflation is rarelydiscussed today, but is, in many ways, a far more destructive problem than inflation. The reason this problem persisted for so many decades in this country was that therewas no stable central banking system to manage the money supply and that the valueof the U.S. dollar was tied to an increasingly rare commodity - gold. This is what wascalled the gold standard. The deflation that was a symptom of the Great Depressionmoved then President Herbert Hoover to order that gold coinage be no longer minted. A couple of years later President Roosevelt ordered that the gold coinage be taken outof circulation, and later in his administration gold was not permitted to be held in

116

coinage by private citizens. Finally, the destabilizing affects of the lack of allowing theU.S. dollar to find its own value on a free market motivated President Nixon to eliminatethe gold standard altogether in 1971.

Gold prices have been allowed to float in the market place. In 2001 the annualaverage price of gold was just under $272, today the price of gold is nearing the levelsreached right after the demise of the gold standard in 1971. $650 an ounce for gold inMay of 2006 is merely a couple of hundred dollars below the lofty levels set in 1972.

Need for Regulation and Inflation

One of the serious implications of the money multiplier is that the banking systemhas the potential for significant harm to economic stability. In Chapter 3 we examinedthe various theories of inflation. One of those theories is called the quantity theory ofmoney, where MV = PQ. Because V and Q grow very slowly, they are generallyregarded as nearly constant. The implication is that what happens to the money supplyshould be nearly directly reflected in the price level.

During expansions in the business cycle, investment demand is generally highand banks can often loan all of their excess reserves. If we reduced the requiredreserve ratio to .01 then banks could expand the money supply $100 for each $1.00 inadditional deposits made in the system. If the required reserve ratio was only .001, then$1000 of new money can be created for every dollar of new deposits. Without anyrequired reserve ratio, the money supply could be theoretically be expanded infinitely foreach dollar of new deposits. These high multiple potential expansions could createserious inflationary problems for the economy, and therefore the required reserve ratioof the central bank is an essential portion of any nation's economic stabilization policies.

During this period of a political swing to the right of public opinion, the idea of de-regulation has gained some new support. However, there is no serious, and informedmove toward de-regulating the money supply. Without monetary controls imposed by acentral bank, there will most certainly be serious economic problems associated withthe loss of some modicum of sensible control of the money supply.

This need for regulation has been long recognized by economists. Since at leastthe 1890s classical economists described the role of money in the economy, and theneed to control the money supply if price stability was to achieved. Most prominentamong the classical economists writing at the beginning of the twentieth centuryconcerning monetary economics was Irving Fisher, who developed the modern quantitytheory of money presented in Chapter 3.

117

We come back to the conclusion that the velocity of circulation either ofmoney or deposits is independent of the quantity of money or of deposits. Noreason has been, or, so far as is apparent, can be assigned, to show why thevelocity of circulation of money, or deposits, should be different, when the quantityof money or deposits, is great, from what it is when the quantity is small.

There still remains one seeming way of escape from the conclusion that thesole effect of an increase in the quantity of money in circulation will be to increaseprices. In may be claimed -- in fact it has been claimed -- that such an increaseresults in an increased volume of trade. We now proceed to show that (exceptduring transition periods) the volume of trade, like the velocity of the circulation ofmoney, is independent of the quantity of money. An inflation of the currency cannotincrease the product of farms and factories, nor the speed of freight trains or ships. The stream of business depends on natural resources and technical conditions, noton the quantity of money. The whole machinery of production, transportation, andsale is a matter of physical capacities and technique, none of which depend on thequantity of money . . . . We conclude, therefore, that a change in the quantity ofmoney will not appreciably affect the quantities of goods sold for money.

Since, then, a doubling in the quantity of money: (1) will normally doubledeposits subject to check in the same ratio, and (2) will not appreciably affect eitherthe velocity of circulation of money or of deposits or the volume of trade, it followsnecessarily and mathematically that the level of prices must double . . . .

Irving Fisher, The Purchasing Power of Money. New York, Macmillan, 1911, pp.154-57.

KEY CONCEPTS

Fractional Reserve Requirements

Balance Sheet

Required Reserve Ratio

Money Multiplier

Monetary History

Genghis Kahn

Gold Receipts

118

U.S. had no central bank prior to 1914

Greenback Act

Reserves

Required

Excess

Actual

Money Creation

STUDY GUIDE

Food for Thought:

Trace the deposit of $100 through a banking system with a .25 RRR. Does an injectionof $100 really result in an increase in the money supply of $400? Explain.

Explain the sources of reserves for member banks. How does the Fed control themoney supply when reserves can be borrowed? Explain and critically evaluate thispractice.

Critically evaluate the Fed as maker and implementer of monetary policy in the UnitedStates.

119

Sample Questions:

Multiple Choice:

With a required reserve ratio of .05 by how much will the money supply be increased inthe Federal Reserve deposits a check of $1000 with National City Bank?

A. $1000B. $5000C. $20,000D. None of the above

Which of the following is the interest rate that member banks charge one another forborrowing excess reserves?

A. Federal funds rateB. Money Market rate C. Discount rateD. Prime rate

True - False:

Excess reserves are the funds in the Federal Reserve Banks that the Fed does notneed to assure the solvency of the banking system. {FALSE}

Money multiplier with a RRR of .25 is 4. {TRUE}

120

Chapter 10

Federal Reserve and Monetary Policy

The neo-classicists continue to reject the idea that the government has a pro-active role to play in economic stabilization. However, where the neo-classicists seeany role for government in stabilizing the macroeconomy it is in monetary policy arena. There is some merit to this point of view. Fiscal policy has significant lags that oftenresults in counter-productive governmental actions, by the time the fiscal policy actuallyimpacts the economy. Monetary policy, on the other hand, has very short lags betweenthe recognition of a problem and the impact on the macroeconomy. At a minimum,monetary policy is quicker to change the economy's direction, is involved in far lesspolitical game-playing, and is more efficient than is fiscal policy.

The purpose of this chapter is the examine the role of the Fed in the formulationand implementation of monetary policy for the purpose economic stabilization. Each ofthe major tools of monetary policy will be examined, and their potential effectspresented.

Monetary Policy

The monetary policies of a government focus on the control of the money supply. These policies directly control inflation or deflation, but also can influence real economicactivity. The focal point of control over real economic activity through the managementof monetary aggregates is the interest rate. The demand for investment is dependentupon the relation between expected rates of return from that investment, and theinterest rate that must be paid to borrow the money to buy capital.

Monetary policy in the United States is conducted by the Federal Reserve, eitherthrough the Board of Governors or the Federal Open Market Committee. The monetarypolicies of the United States have focused primarily on the control of various monetaryaggregates, i.e., the money supply, which, in turn, influences interest rates and henceaggregate economic activity. The fundamental objective of monetary policy is to assistthe economy in attaining a full employment, non-inflationary macroeconomicequilibrium.

121

Tools of Monetary Policy

The Federal Reserve has an arsenal of weapons to use to control the moneysupply. The Fed can buy and sell U.S. Treasury bonds, called open market operations,it can control the required reserve ratio, within statutory limits, and it can manipulate thediscount rate (the interest rate charged by the Fed for member banks to borrowreserves). This array of tools provides the Fed with several options in taking correctiveactions to help stabilize the economy. Each of these tools will be examined in thefollowing paragraphs.

Open Market Operations

Open Market Operations (OMO) involves the selling and buying of U.S. Treasuryobligations in the open market. OMO directly influences the money supply throughexchanging money for bonds held by the public (generally commercial banks andmutual funds). Expansionary monetary policy involves the buying of bonds. When theFed buys bonds, it replaces bonds held by the public with money. When someone sellsa bond, they receive money in exchange, which increases the money supply.Contractionary monetary policy involves the Fed selling bonds to the general public.When the Fed sells bonds it removes money from the hands of the public and replacesthat money with U.S. Treasury bonds.

The Fed sells and buys both long-term (thirty year) Treasury bonds, and short-term (primarily two, five and ten year) Treasury bonds. In theory the Fed can focus its influence on either long-term interest rates or short term-interest rates. However, whatmost security analysts argue is that the Fed's bond buying and selling behavior is not aleadership position. In recent years, the Fed buys and sell bonds after the moneymarket establishes a direction for interest rates. In times, of high inflation or steeprecession, however, the evidence suggests that the Fed's OMO leads the market, ratherthan follows. In reality, this is what we should expect to observe in the responsibleexercise of monetary policies.

The Fed also buys and sells lower denomination and shorter term obligationscalled Treasury Bills. While bonds are sold only periodically in Treasury auctions,Treasury Bills are available at any time through the Federal Reserve Banks. TheTreasury Bill is also the buying and selling of U.S. government debt, but is not theprimary tool of open market operations, even though it has the same impact ofincreasing or decreasing the money supply. The market for Treasury Bills is smallrelative to the Bond market.

122

The Required Reserve Ratio

As discussed in the previous chapter, it is the required reserve ratio thatdetermines the size of the money multiplier. The money multiplier determines howmuch money can be created by the member banks through the deposit - loan process. Therefore, the Fed can directly control how much the banking system can expand themoney supply through the manipulation of the required reserve ratio.

The Fed can raise or lower the required reserve ratio, within statutory limits.Increasing the required reserve ratio, reduces the money multiplier, hence reduces theamount by which multiple expansions of the money supply can occur. By decreasingthe required reserve ratio, the Fed increases the money multiplier, and permits moremultiple expansion of the money supply through the deposit - loan process described inthe previous chapter.

Most of the new deposits that result in the multiple expansion of money occurbecause the Fed bought bonds from the public. When the Fed buys bonds, this isinjecting new money into the system that is, in turn, deposited -- which set the moneymultiplier (multiple expansion) into motion.

The Discount Rate

The Discount Rate is the rate at which the Fed will loan reserves to memberbanks for short periods of time. To tighten monetary policy, the Fed will raise thediscount rate. The raising of the discount rate will discourage the borrowing of requiredreserves by member banks, hence encourages using their reserves as requiredreserves, rather than excess reserves (which they loan and start the multiple expansionof money). By lowering the discount rate, the Fed encourages the borrowing of requiredreserves, which may result in more excess reserves hence potentially more loans, anda greater expansion of the money supply through the loan - deposit process describedin the previous chapter (Chapter 9).

It is the Discount Rate over which the Fed may exert direct control. However,member banks may borrow reserves from one another, as well as the Fed. The rate atwhich member banks borrow from one another is called the Fed Funds Rate. The FedFunds Rate is heavily influenced by what the Fed does with respect to the DiscountRate, and the Fed’s supervisory impact on member banks. Therefore, the Fed FundsRate is often targeted by the Fed, and is not entirely immune to Fed influence, eventhough not under the direct control of the Fed.

It is also worth mentioning that the Fed sets the margin rate minimums thatbrokerages may charge customers to buy stocks with borrowed money. This function of

123

the Fed is not an important current monetary policy tool. In fact, it is pretty much asymptom of the Federal Reserve Act being passed as a result of a series of financialpanics immediately prior to the out-break of World War I and of the devastation broughtlater by the Great Depression.

Fed Targets

The Fed must have benchmarks to determine the need for and effectiveness ofmonetary policies. The quantity theory of money suggests that the money supply itselfis the appropriate policy target for the Fed. As noted by Irving Fisher, the velocity ofmoney is how often the money supply turns-over, and is unrelated to economic activity. Further, Fisher argued that money does not directly influence the real output of theeconomy. Therefore, in any policies aimed at controlling inflation or deflation the Fed'smonetary targets should simply be the money supply.

However, the economy is not as simple as the quantity theory of money wouldsuggest. Many consumer purchases and most investment is interest rate sensitive. Therefore, to the extend that the Fed's policies do impact interest rates, the Fed canalso correct downturns in the business cycle. If investment is too low to maintain full-employment level of GDP, the Fed can reduce the required reserve ratio or buy bondsthereby increasing the supply of money and lowering the interest rate. The lowerinterest rate may encourage consumption expenditures and investment, therebymitigating recession.

There are dilemmas for the Fed in selecting targets for their monetary policies.Interest rates and the current business cycle may present a dilemma. Expansionarymonetary policy may result in higher interest rates, by increasing the rate of inflation,which will be reflected in the interest rates. As the interest rate increases and people'sinflationary expectations develop, these may serve to dampen the expansionary effectsof the Fed's monetary policies. At the same time, there is no necessary coordination offiscal and monetary policies. At the time an expansion monetary may be necessary toreverse a recession, contractionary fiscal policies may begin to affect the economy. Therefore, the Fed must keep an eye on the Congress and account for any fiscalpolicies that may be contradictory to the appropriate monetary polices. The presents adelicate balancing act for the Fed. Not only must the Fed correct problems in theeconomy, it may well have to correct Congressional fiscal mistakes, or at least, accountfor this errors when implementing monetary policies.

Tight and Easy Money

Discretionary monetary policy, therefore, fits into one of two categories, (1) easy

124

money, and (2) tight money policies. Easy money policies involves the lowering ofinterest rates, and the expansion of the money supply. The purpose of easy moneypolicies are typically to mitigate recession and stimulate economic growth. Tightmonetary policies involve the increasing interest rates, and contracting the moneysupply. The purpose of tight money policies is generally to mitigate inflation and slowthe rates of economic growth (typically associated with inflation). These monetarypolicies may also have a significant impact on the value of the U.S. Dollar in foreignexchange markets. In general, tight monetary policies are associated with increasingthe value of the U.S. Dollar, whereas easy monetary policies will generally have theopposite affect.

Consider the following diagram showing both tight and easy money policies’impact on the money market.

Assuming that the money supply remains constant, we can analyze the changesin the money supply imposed by the Fed. As the Fed engages in tight money policiesthe supply curve is shifted to the left (dashed line labeled tight), this increases theinterest rate and lowers the amount of money available in the economy. On the otherhand, easy money policy is a shift to the right of the money supply curve (dashed linelabeled easy). With easy money policies, the quantity of money increases and theinterest rate falls. The effectiveness of such policies in influencing GDP result fromchanges in autonomous investment. In the case of an easy money policy, as theinterest falls, investment will increase which results in an increase in the C+I+G line asillustrated below:

125

The decrease in the interest rate is associated with an increase in investment(the vertical distance between C+I+G1 and C+I+G2) which results in an increased GDPand levels of spending.

The acceptance of discretionary monetary policies are often associated withKeynesian views of a pro-active role for government in economic stabilization. Eventhough most neo-classicists argue that monetary policy is necessary to the properfunctioning of a market economy. However, there is another view. The most extremeof the neo-classicists argue that there is simply no role for either discretionary fiscal ormonetary policies, except in dealing with extreme variations in economic activity, i.e.,like the Great Depression of the 1930s.

Friedman's Monetary Rules Argument

The leading economist of the neo-classical school (Chicago School of Thought orMonetarist) is the Nobel Prize winning economist, Milton Friedman. Professor Friedmanwon his Nobel Prize for, among other contributions, his work on the monetary history ofthe United States. Friedman argues, with some persuasion, that Irving Fisher's workestablishes the appropriate standard for monetary policy. Based on the presumptionthat discretionary fiscal policy can be abolished, Friedman would have all monetarypolicies based on a simple rule that follows directly from Fisher’s formulation of thequantity theory of money.

Assuming that discretionary fiscal policy has been eliminated, and that theeconomy is operating at a full-employment, non-inflationary equilibrium, monetary policyshould be nothing more or less than estimating the growth rate of economy (change inQ) and matching the growth rate of the money supply to the growth rate of theeconomy. Such monetary policy leaves nothing to the discretion of policy makers. The

126

Fed's sole role is to make sure that the money supply simply facilitates economic growthby expanding at the same pace as the real economic activity. If the Fed underestimatesgrowth, there could be small deflations, that could be eliminated but subsequentadjustments to the money supply, and overestimations of the growth rate causinginflation could be dealt with in the same type of subsequent adjustments.

If nothing else, Friedman's suggestion would eliminate any government inducedvariations in economic activity. Real Business Cycle Theory is another paradigm thathas arisen out of the ashes of the classical school, and these economists would not findmuch to argue with Friedman about, as far as the analysis goes. These economists,primarily Thomas Sargent from the University of Minnesota, argue that recessions andinflations result from either major structural changes in the economy or external shocks,such as the Arab Oil Embargo. When these types of events occur, the Real BusinessCycle Theorists would have the government play a pro-active role, but focusedspecifically on the shock or structural problem. In this sense, there is a role fordiscretionary fiscal and monetary policies, but very narrowly focused on very specificevents.

KEY CONCEPTS

Monetary Policy

Expansionary

Contractionary

Tools of Monetary Policy

Bonds

Required Reserves Ratio

Discount Rate

Velocity of Money

Quantity Theory of Money

Target Dilemma in Monetary Policy

Discretionary Monetary Policy vs. Monetary Rules

127

STUDY GUIDE

Food for Thought:

Critically evaluate the tools and the independence of the Fed to use them in monetarypolicy.

If the MV = PQ equation is correct then inflation can be dealt with as a monetaryproblem. If we are experiencing 10 percent inflation with $4 billion real economy (inother words nominal GDP is $4.4 billion) and the velocity of money 2.8 how do wecontrol inflation? Would we want to? Explain.

What must the Fed do with each of its tools to create (1) a tight money policy, and (2)an easy money policy? Critically evaluate each.

Sample Questions:

Multiple Choice:

If the Fed wanted to create a tight money policy which of the following is inconsistentwith this goal?

A. Increasing the Required Reserve RatioB. Increasing the Discount RateC. Buying bondsD. All of the above are consistent with a tight money policy

128

If the Fed wishes to stimulate the economy during a recession what might we expect toobserve?

A. Lowering the Required Reserve RatioB. Lowering the Discount RateC. Fed Buying BondsD. None of the above

True/False:

The main goal of the Federal Reserve System is to assist the economy in achieving andmaintaining a full-employment, non-inflationary, stability. {TRUE}

A tight money policy assists in bringing the economy out of recession, but at the risk ofpossibly causing inflation. {FALSE}

129

Chapter 11

Interest Rates and Output: Hick’s IS/LM Model

The final issue with which to build a complete picture of the macroeconomy is thedevelopment of a model to explain the relationship of interest rates to the overall outputof the economy. Sir John Hicks developed a model which directly equates the interestrate with the output of the economy. This model is called the IS/LM model and providesvaluable insights into the operation of the U.S. economy – this chapter offers a quickand simple view of these relations.

IS Curve

The IS curve shows the level of real GDP for each level of real interest rate. Thederivation of the IS curve is a rather straightforward matter, observe the following diagram.

INCOME/EXPENDITURE Expenditure or Aggregate Demand

GDP AUTONOMOUS SPENDING

Interest Rate Interest Rate

IS

Autonomous Spending GDP

130

The Income-Expenditure diagram determines for each level of aggregatedemand the associated level of GDP. There is a portion of this GDP which isdetermined within the system (autonomous spending) and for each level of autonomousspending there is a specific interest rate. The indicator line from the Income-Expenditure diagram provides levels of GDP, which are associated with specific interestrates, which in turn are associated with specific level of autonomous spending. Theresulting IS curve is a schedule of levels of GDP associated with each interest rate.

The intercept of the IS Curve is the level of GDP that would obtain at a zero realinterest rate. The slope of IS Curve is the multiplier (1/1 - MPC) times marginalpropensity to Invest (resulting from a change in real interest rates) and the marginalpropensity to export (resulting from a change in real interest rates).

The IS curve can be shifted by fiscal policy. An increase in governmentpurchases pushes the IS curve to the right, and a decrease will shift it left. Conversely,an increase in taxes pushes the IS curve left, while a decrease in taxes will push thecurve to the right. However, almost anything that changes the non-interest-dependentcomponents of autonomous spending will move the IS curve. For example, foreignersspeculating on the value of the dollar may result in a shift in the IS curve to the extent itimpacts purchases of imports. Changes in Co or Io will also move the IS curve.

Anytime the economy moves away from the IS curve there are forces within thesystem which push the economy back onto the IS curve. Observe the followingdiagram:

Real Interest Rate

¹ ! A

B ! ¸ IS Curve Real GDP

131

If the economy is at point A there is a relatively high level of real interest rates whichresults in planned expenditure being less than production, hence inventories areaccumulating, and production will fall, hence pushing the economy toward the IS curve. On the other hand, at point B the relatively low real interest rate results in plannedexpenditure being greater than production, hence inventories are being sold into themarket place and product will rise to bring us back to the IS curve.

LM Curve

The LM Curve is derived in a fashion similar to that of the IS curve. Consider thefollowing diagram:

Interest Rate Money S Interest Rate

LM Curve

D 1

D 2

D 3

GDP (Y)

Real Qty of Money High Y Moderate Y Low Y

As can be readily observed from this diagram the LM curve is the schedule ofinterest rates associated with levels of income (GDP). The interest rate, in this case,being determined in the money market.

With a fixed money supply each level of demand for money creates a differentinterest rate. If the money market is to remain in equilibrium, then as incomes rise sotoo, then must the interest rate, if the supply of money is fixed. The shifting of the LMcurve is obtained through inflation or monetary policy. If the price level is fixed, then adecrease in the money supply will shift the LM curve to the left, and an increase in themoney supply will push the LM curve to the right. Conversely, if the money supplyremains fixed and there is inflation, then the real money supply declines, and the LM

132

curve shifts left, and vice versa.

Equilibrium in IS-LM

Just as in the Keynesian model, the intersection of the IS and LM curve results inthere being an equilibrium in the macroeconomy. The following diagram illustrates aneconomy in equilibrium at where LM is equal to ISe:

Interest Rate LM

r

IS1 ISe IS2

GDP Y

Where the IS and LM curves intersect is where there is an equilibrium in thiseconomy. With this tool in hand the affects of the interest rate on GDP can be directlyobserved. As the IS curve shifts to the right along the LM curve notice that there is anincrease in GDP, but with a higher interest rate (IS1) and just the opposite occurs as theIS curve shifts back towards the origin (IS2). From above it is clear the sorts of thingsthat shift the IS curve, fiscal policy or changes in Co or Io.

Fiscal policy is associated with changes in the position of the IS curve. If thegovernment decreases taxes, and there is no change in monetary policy associatedwith this fiscal change we would expect to observe a shift to the right of the IS curve,hence an increase in GDP, but with an increase in interest rates. For interest rates toremain the same with this increase in GDP, the Fed would have to engage in easymonetary policies and shift the LM curve to the right which would also have the effect ofincreasing GDP. The same analysis would result in the case of an increase ingovernment expenditures.

In the case of an increase in taxes, the GDP would fall, but so too would interestrates. If the Fed were to increase the interest rates back to previous levels it wouldneed to engage in tight monetary policies to shift the LM curve back to the left. In otherwords, the interest rate is the driving mechanism behind the equilibrium in themacroeconomy as suggested by the monetarists and Keynesians alike. It is simply thatneither the Keynesian cross nor the quantity theory of money made it easy to examine

133

these relations in any significant detail. It is therefore clear that the results of fiscalpolicy are dependent on what happens with the money supply.

Expansionary and Contractionary Monetary Policies

The LM curve, too, can be moved about by changes in policy. If the moneysupply is increased or decreased that will have obvious implications for the LM curveand hence the interest rate and equilibrium level of GDP. Consider the followingdiagram:

Interest Rate LM2 LMe

LM1

r

IS

GDP Y

As the Fed engages in easy monetary policies the LM Curve shifts to the right,and the interest rate falls, as GDP increases. This is the prediction of quantity theory ofmoney and is consistent with what is presumed to be the case if there is not monetaryneutrality in the Keynesian model. If the Fed engages in tight monetary policy then wewould expect to observe a shift to LM2. In the case of LM2 the decrease in the moneysupply results in a higher interest rate, and a lower GDP. Again, these are resultsconsistent with the predictions of the quantity theory of money and consistent with whatwe would expect without monetary neutrality in the case of the Keynesian model.

Foreign Shocks

The U.S. economy is not a closed economy, and the IS-LM Model permits us toexamine foreign shocks to the U.S. economy. There are three of these foreign shocksworthy of examination here; these are (1) increase in demand for our exports, (2)increases in foreign interest rates, and (3) currency speculators expectations of anincrease in the exchange rate of our currency with respect to some foreign currency. Each of these foreign shocks results in an outward expansion of the IS Curve asportrayed in the following diagram:

134

Interest Rate

LM

r2 r1

IS1

ISoGDP

Y1 Y2

As the demand for domestic exports increases overseas the IS curve shifts fromIS o to IS1. The interest rate increases domestically which will have the tendency toreduce the exchange rate, which will have two countervailing affects, which are to putdownward pressure on further exports, and to increase imports. The result is initially weexport more, and that increase is dampened somewhat by the countervailing effects inthe second round.

An increase in the foreign real interest rates has the effect of increasing ourdomestic imports because foreign exchange rate has become more favorable toforeigners, and allows their currency to go further in our economy. The end result isthat the same shift in the IS curve will be observed with same results. Speculators incurrency markets may also generate the same result as an increase in foreign interestrates, if they believe that a weak dollar will result in their currency gaining value. Hence,much of the foreign exchange and balance of payments phenomena can also be readilyanalyzed using the IS-LM Model.

KEY CONCEPTS

IS Curve

Slope and Intercept

LM Curve

Money Market

Equilibrium in the IS-LM Model

135

Fiscal Policy

Monetary Policy

Real Interest Rates

Foreign Sector

Demand for exports

Foreign interest rates

Currency speculation

STUDY GUIDE

Food for thought:

Describe the role that interest rates play in determining equilibrium in themacroeconomy. Is this role clear in the quantity theory of money or the Keynesianmodels? Explain.

What if currency speculators believed that the dollar was going to become stronger, orthat the demand for exports decreased abroad? What would happen to the analysisoffered above?

Critically evaluate monetary and fiscal policy within the context of the IS-LM model. ISthis different than you believed coming into this course? Explain.

136

Sample Questions:

Multiple Choice:

If we are off of the IS curve, and above it, what will happen to bring us back to the IScurve?

A. Monetary and / or fiscal policy will be requiredB. Inventories will accumulate and bring us backC. This cannot occur in theory or realityD. None of the above

To shift the LM curve to the right which of the following must occur?

A. Increase taxesB. Increase government expendituresC. Increase the money supplyD. Any of the above

True / False:

The main determinants of the LM Curve are in the money market. {True}

Anything that affects the non-interest-dependent components of autonomous spendingshifts the position of the IS curve. {True}

137

Chapter 12

Economic Stability and Policy

The focus of this course is managerial, however, public policy has a significantimpact on business conditions. It is therefore of importance that some discussion occurwith respect to these issues. Perhaps the most important macroeconomic problems ofthe last forty years are unemployment and price level instability. During the post-WorldWar II period price level instability has primarily been inflation. There are several othermatters that have also been the focus of economic policy, including poverty (incomedistribution), and the federal debt and budget deficit. The purpose of this chapter is toexamine these policy issues.

The Misery Index

During the Reagan administration both unemployment and inflation topped tenpercent. This record gave rise to an economic statistic called the misery index. Themisery index is the summation of the civilian unemployment rate and the consumerprice index. For example, with 10 percent unemployment and twelve percent inflation,the misery index would be 22.

The consumer price index is based on a market basket of goods that householdtypically purchase. Therefore, the CPI measures the impact of increasing prices on thestandard of living of consumers. The unemployment rate also focuses on the welfare offamilies, when the household wage earning is out of work it has serious implications forthat household's income. The misery index can be interpreted as a measure of the lossof well-being of households.

Since 1973 American households have not fared well. The U.S. Department ofCommerce, Current Population Survey, tracks economic and demographic dataconcerning households. Between 1973 and 1993 the median real income of Americanhouseholds has not changed. The slight increases enjoyed in the 1970s were all givenback during the 1980s. However, between 1973 and 1993 there has been a dramaticincrease in the number of two wage earner households, and households where a full-time worker also has a part-time job.

Increasingly, economists are warning that the distribution of misery in the U.S.economy is becoming more extreme and that the social ills associated with this miseryare increasing. Crime, drug abuse, social and economic alienation, and stress related

138

If the real GNP is up and real wages are down for two thirds of the workforce, as an algebraic necessity wages must be up substantially for the remainingone third. That one third is composed of Americans who still have an edge in skillson workers in the rest of the world -- basically those with college educations. In the1980s educational attainment and increases or decreases in earnings were highlycorrelated. American society is now divided into a skilled group with rising realwages and an unskilled group with falling real wages. The less education, thebigger the income reduction; the more education, the bigger the income gains.

These wage trends have produced a sharp rise in inequality. In the decadeof the 1980s, the real income of the most affluent five percent rose from $120,253to $148,438, while the income of the bottom 20 percent dropped from $9,990 to$9,431. While the top 20 percent was gaining, each of the bottom four quintiles lostincome share; the lower quintile, the bigger the decline. At the end of the decade,the top 20 percent of the American population had the largest share of totalincome, and the bottom 60 percent, the lowest share of total income ever recorded.

Lester Thurow, Head to Head: The Coming Economic Battle Among Japan,Europe, and America. New York: William Morrow and Company, 1992, p. 164.

illnesses have become epidemic. The fraudulent arguments that what was needed was a return to traditional family values, is used as a substitute for what is reallytranspiring, something must be done to reduce economic disparity, particularly as itnegatively effects the family. As Lester Thurow has noted:

It is beyond dispute that since 1981 there has been a fundamental change inAmerican society. As Lester Thurow notes, America is becoming two separate andunequal societies, one group with an increasing misery index, and another one withincreasing affluence.

The Bush Administration’s 2003 Tax Cut has also been heavily criticized ashaving provided significant tax relief for the wealthy in the form of reduced rates ondividends. At the same time, the deficits necessary to fund those dividend tax cuts willincrease interest rates over the next few quarters, which impact mortgage rates andconsumer loan interest rates. This is an empirical question and will be answered bycold, objective evidence within the year.

139

The Phillips Curve

Since the Kennedy administration much of American economic policy has beenbased on the idea that there is a trade-off between unemployment and inflation. It wasnot until the recession of 1981-85 that we experienced very large amounts ofunemployment together with high rates of inflation. It was thought that there wasalways a cruel choice in any macroeconomic policy decision, you can haveunemployment and low inflation, or you can have low rates of unemployment, but at thecost of high rates of inflation. This policy dilemma, results from acceptance of astatistical relation observed between unemployment and inflation named for A. W.Phillips who examined the relation in the United Kingdom and published his results in1958. (Actually Irving Fisher had done earlier work on the subject in 1926 focused onthe United States).

The Short-Run, Trade-off Phillips Curve

The following diagram presents the short-run trade-off view of the Phillips curve. Actually, A. W. Phillips' original research envisioned a linear, downward sloping curvethat related nominal wages to unemployment. Over two years after A. W. Phillips'paper was published in Economica, Richard Lipsey replicated Phillips' study specifyinga non-linear form of the equation and using the price level, rather than nominal wages inhis model. It is Lipsey's form that is commonly accepted, in the literature, as the short-run, trade-off view of the Phillips curve.

The short-run, trade-off view of the Phillips curve is often used to support anactivist role for government. However, the short-run, trade-off view of the Phillips Curveshows that as unemployment declines, inflation increases, and vice versa. It is thisnegative relation between unemployment and inflation, portrayed in the above diagram,

140

that gives rise to the idea of cruel policy choices between unemployment and inflation. However, there are alternative views of the Phillips Curve relation.

Natural Rate Hypothesis

Classicists argue that there can only be a short-run, trade-off type Phillips Curveif inflation is not anticipated in the economy. Further, these economists argue that theonly stable relation between unemployment and inflation that can exist is in the long-run. In the long run the Phillips Curve is alleged to be vertical at the natural rate ofunemployment, as shown in the following diagram:

In this view of the Phillips Curve any rate of inflation is consistent with the naturalrate of unemployment, hence the Natural Rate Hypothesis. It is based on the idea thatpeople constantly adapt to current economic conditions and that their expectations aresubject to "adaptive" revisions almost constantly. If this is the case, then businessesand consumers cannot be fooled into thinking that there is a reason for unemploymentto cure inflation or vice versa, as is necessary for the short-run, trade-off of the PhillipsCurve to exist.

If taken to the extreme, the adaptive expectations view can actually result in apositively sloped Phillips Curve relation. The possibility of a positive sloping PhillipsCurve was first hypothesized by Milton Friedman. Friedman was of the opinion thatthere may be a transitional Phillips Curve, caused by people adapting both theirexpectations and institutions to new economic realities.

In fact, the experience of 1981-85 may well be a transitional period, just like thatenvisioned by Friedman. The beginning of the period was marked by OPEC driving up

141

the price of exported oil, and several profound changes in both the American economyand social institutions. The Reagan appointments to the N.L.R.B., Justice Department(particularly the Anti-Trust Division), the Supreme Court (and Circuit Courts of Appeals,and District Courts) and significant changes in the tax code changed much of the legalenvironment significantly. Further, the very significant erosion of the traditional baseindustries in the United States (automobiles, transportation, steel, and other heavymanufacturing) together with massive increases in government spending on defensearguably created a transitory economy, consistent with the increases in both inflationand unemployment. The positively sloped Phillips Curve is show in the followingpicture:

The positively sloped transitional Phillips Curve is consistent with theobservations of the early 1980s when both high rates of unemployment existed togetherwith high rates of inflation (the positive slope) -- a condition called stagflation (economicstagnation accompanied by inflation).

Cruel choices may only exist in the case of the short-run, trade-off view of thePhillips Curve. However, there maybe a "Lady and Tiger Dilemma" for policy makersrelying on the Phillips Curve to make policy decisions. If fiscal policy is relied upon, onlythe timing of the impact of those fiscal policies will result in any positive influence on theeconomy. Therefore, to act, through taxes or expenditures, may result in having acounter-productive effect by the time the policy impacts the economy (the tiger). On theother hand, if accurate two and three year into the future forecasts can be acted upon intime, recession or inflation could be mitigated by current action -- a real long-shot! (thelady).

However, if the rational expectations theories are correct, then the long-shot isexactly what would be predicted. Rational expectations is a theory that businesses andconsumers will be able to accurately forecast prices (and other relevant economic

142

variables). If the accuracy of consumers' and businesses' expectations permit them tobehave as though they know what will happen, then it is argued that only a verticalPhillips Curve is possible, as long as political and economic institutions remain stable.

Market Policies

Market policies are focused on measures that will correct specific observedeconomic woes. These market policies have been focused on mitigating poverty,mitigating unemployment, and cooling-off inflation. For the most part, these marketpolicies have met with only limited success when implemented.

One class of market policies have been focused on reducing poverty in theUnited States. These equity policies are designed to assure "a social safety net" at theminimum, and at the liberal extreme, to redistribute income. In part, the distribution ofincome is measured by the Lorenz Curve, and more completely by the Gini coefficient. The following diagram presents the Lorenz Curve:

The Lorenz curve maps the distribution of income across the population. The 45degree line shows what the distribution of income would be if income was uniformlydistributed across the population. However, the Lorenz curve drops down below the 45degree line showing that poorer people receive less than rich people. The further theLorenz Curve is bowed toward the percentage of income axis the lower the income inthe poorer percentiles of the population.

The Gini coefficient is the percentage of the triangle (mapped out by the 45degree line, the indicator line from the top of the 45 degree line to the percentage ofincome axis, and the percentage of income axis) that is accounted for by the areabetween the Lorenz curve and the 45 degree line. If the Gini coefficient is near zero,

143

income is close to uniformly distributed (and the 45 degree line); if is near 1 then incomeis distributed in favor of the richest percentiles of the population (and the Lorenz curve isclose to the horizontal axis). If that distribution is consistent with the productivity ormeritorious performance of the population, there may be an efficiency argument thatcan be used to justify the distribution. However, if the high income skewness of thedistribution is not related to productivity, then the skewed distribution is inefficient andunfair, hence mal-distributed. In the United States the Gini coefficient exceeds .5, andwhile incomes in the lower three-quarters of the upper quintile are highly correlated witheducation (and presumably productivity) the overwhelming amount of the highest part ofthe distribution does not have any prima facie evidence to prove the justice or efficiencyof such high incomes. Recently, (December 8, 1995) CNN reported that ChiefExecutive Officer salaries in the entertainment industries appeared to be out of line withsimilar officials in more productive industries, and that stock holders in several of thesecompanies were beginning to revolt over these high levels of compensation. Inparticular, Viacom and Disney were experiencing stock holder queries concerningexecutive salaries. In general, most economists familiar with the income distribution inthe United States, would probably agree that income in this country is mal-distributed,because it does not reflect the market contributions of those at either the highest end, orin the lower end of the distributions.

Productivity has also the subject of specific policies. The Investment Tax Credit,WIN program, and various state and federal training programs have been focused onincreasing productivity. For the most part, there has been very little evidenceconcerning most of these programs that give reason for optimism. The one exceptionwas the work of Mike Seeborg and others, that found substantial evidence that JobCorps provided skills that helped low income, minority teenagers find and keepreasonably well-paying jobs.

Many of the recent treaties concerning international trade have aspects that canbe classified as market policies. To the extent that trade barriers to American exportshave been reduced through NAFTA and GATT it was hoped that there may have beenpositive effects from these treaties. In fact, little if any, positive effects have beenobserved from these initiatives.

There have been recurrent attempts have to directly control inflation throughprice controls. These controls worked well during World War II, mainly because ofappeals to patriotism during a war in which the United States was attacked by a foreignpower. Further, during World War II the idea of sacrifice was reinforced by manyfamilies having relatives serving in the military, which made the idea of sacrifice moreacceptable to most people. However, absent the popular support for these policiescreated by World War II for rationing and wage and price controls, these policies havebeen uniformly failures. President Carter tried voluntary guidelines that failed, andRichard Nixon had earlier tried short-lived wage and price controls that simply were apolicy disaster.

144

Debt and Deficit

The national debt is often argued to have an adverse effect on interest rates,which, in turn, crowds out private investment. The empirical evidence concerning this“crowding out” hypothesis suggests that increased public debt often results in higherinterests rates, assuming that the debt was not acquired to mitigate recession.

The supply side economics of the Reagan Administration were based on thetheory that stimulating the economy would prevent deficits as government spending forthe military was substantially increased. This failed theory was based on somethingcalled the Laffer Curve.

The Laffer Curve (named for Arthur Laffer) is a relation between tax rates and taxreceipts. Laffer's idea was rather simple and straightforward, he posited that there wasoptimal tax rate, above which and below which tax receipts fell. If the government wasbelow the optimal tax rate, an increase in the rate would increase receipts and in therate was above the optimal rate, receipts could be increased by lowering tax rates. TheLaffer Curve is shown below:

The Laffer Curve shows that the same tax receipts will be collected at the rateslabelled both "too high" and "too low." What the supply-siders thought was that taxrates were too high and that a reduction in tax rates would permit them to slide downand to the right on the Laffer Curve and collect more tax revenue. In other words, theythought the tax rate was above the optimal. Therefore Reagan proposed and obtainedfrom Congress a big tax rate reduction and found, unfortunately, that we were below theoptimal and tax revenues fell. While tax revenue fell significantly, the Reagan

145

Administration increased the defense budget by tens of billions of dollars per year. Thereduction in revenues, combined with substantial increases in government spendingmade for record-breaking federal budget deficits and substantial additions to ournational debt.

There were other tenets of the supply-side view of the world. These economiststhought there was too much government regulation. They would have de-regulatedmost aspects of economic life in the United States. However, after Jimmy Carter de-regulated the trucking and airlines industries, there was considerable rhetoric and littleaction concerning the de-regulation other aspects of American economic life.

Politics and Economic Policy

Unfortunately, the realities of American economic policy is that politics is oftenmain motivation for policy. Whatever anyone may think of Reagan's Presidency there issimply no doubt that he was probably the most astute observer of the political arena ofany of his competitors. Reagan argued that taxes were too high and needed to be cut. This is probably the single most popular political theme any candidate can adopt. Remember McGovern? He said if he were elected President he would raise taxes, hedid not have to worry about how, because he won only two states. The surest way tolose a bid for public office is to promise to raise taxes.

As it turned-out McGovern was right, there should have been a tax increase, atthe time Reagan cut them. Being right, does not have anything to do with beingpopular. What we now face is a direct result of the unprecedented deficits run duringthe Reagan years. In fact, during those eight years this country acquired nearly $1.7trillion of its national debt. No other President in U.S. history has generated this amountof debt in nominal dollars. However, before Reagan is judged too harshly, it mustremembered that we were in the midst of a major recession during his first term inoffice, and the cold war was still at its zenith. Again, in Mr. Reagan's defense the firstthree years of his administration also witnessed exceedingly high rates of inflation, thatare reflected in the nominal value of the deficits.

The following table shows the national debt for the period 1980 through 1988,notice, if you will, how the national debt accelerated during the Reagan administration,remember that the first three years were deepening recession.

146

__________________________________________________________________________________________________________________________________________

Table I: National Debt 1980 - 2005_____________________________________________________________________

NATIONAL DEBT YEAR (billions of U.S. dollars)

1980 908.31981 994.31982 1136.8 1983 1371.21984 1564.1 1985 1817.01986 2120.11987 2345.61988 2600.61989 2868.01990 3206.61991 3598.51992 4002.11993 4351.41994 4643.71995 4921.01996 5181.91997 5369.71998 5478.71999 5606.12000 5629.02001 5803.12002_____ 6228.22003 6783.22004 7379.12005 7932.7

_____________________________________________________________________

Nearly $1692.3 of additional debt was accumulated during the years thatPresident Reagan was in office. The first nine months of this period was under Carter'sbudget, but the first nine months of 1989 was under Reagan's and this data makesReagan's record look better than it really was. At a six percent interest rate, the currentyield on the 30 year Treasury Bond (as of December 8, 1995) the debt accumulatedduring these eight years adds $101.5 billion to the federal budget in interest payments.

147

There is a lesson here, that has nothing to do with political propensities, tax cutsthat do not generate continuing economic growth, will add to the national debt, and theinterest payments on that debt will add to even further budget deficits. It cannot bedenied that Reagan was one of the most popular Presidents of this century. The verythings that made him popular, tax cuts, and military build ups are responsible for muchof our current controversy concerning the balanced budget.

Is the Debt Really a Problem?

The most valid criticism of the debt is its potential for disrupting credit markets, byartificially raising interest rates and crowding out private investment. Naturally, a debtas large as ours will have an upward influence on interest rates, but the evidence doesnot suggest that it is substantial.

Today, we find ourselves in roughly the same position this country was in 1805. President Jefferson borrowed about the same amount as our GDP from England andHolland to buy Louisiana from the French (we owe an amount almost the same asGDP). History has taught us that even when we added the debt acquired in the War of1812, it did not bankrupt the government or its citizens. By the end of World War II, weagain had a national debt of $271 billion and a GDP of $211.6 billion. The predictionsthat my generation would be paying 60 and 70 percent effective tax rates to pay off thedebt never materialized. Frankly, I wish my debt was only equal to my annual income,and I suspect most people in their thirties and forties have the same wish. Surely thesize of debt and current deficits are not a source of any grief. If it is then little is learnedfrom history and little is known about economics.

The serious question is why did we acquire the debt and what must we give up ifwe are to pay it off. When Reagan took office, we were in the midst of a cold war. Theincreases in government expenditures for the military, caused our chief rivals to spendlarger proportions of their GNP on the military and eventually caused the economic andpolitical collapse of the Soviet bloc. Was the end of the Cold War worth the debt weacquired? Are the benefits of education worth a few million dollars in current debt? Isproviding for the poor, the elderly, and children worth a few billion in debt? Areveterans' pensions for those who stood between us and our enemies worth a few billionin debt? What about the Interstate Highway System, subsidies for the company whereyou work, research for medical reasons, the pure sciences, the social sciences, and thetens of thousands of things on which the government spends our tax dollars? Theseare the priorities that any society must set. Probably the only realistic answer towhether the debt is a problem is what we do about it and what priorities we set. Historywill judge this society, whether we are judged as compassionate, or barbarians, oreconomically astute or fools is for future generations of historians. Let us all hope thatwe choose correctly.

148

Hedayeh Samavati, David A. Dilts, and Clarence R. Deitsch, "The PhillipsCurve: Evidence of a 'Lady or Tiger Dilemma,'" The Quarterly Review ofEconomics and Finance. Vol. 34, No. 4 (Winter 1994) pp. 333-345.

During the period examined, January 1974 to April 1990, the evidencereported (here) suggests that there is a unidirectional causal relation from theinflation rate to the rate of unemployment. If, the Phillips Curve were verticalover this sixteen year period, one should have observed no causality betweenthe unemployment rate and the rate of inflation (the natural rate hypothesis,i.e., any rate of inflation can be associated with the natural rate ofunemployment). The empirical findings reported here, however, suggest anon-vertical Phillips Curve.

Finally, the results reported here suggest the proper specification of theempirical models used to test the Phillips Curve relation. Friedman arguedthat the proper specification of the regression equations used to estimate thePhillips Curve relation is of significance (both theoretically and empirically). "The truth of 1926 and the error of 1958" (as Friedman argued) is supportedby the evidence presented in this paper. The statistical evidence presentedhere supports Friedman's claim that inflation is properly specified as theindependent variable in "Phillips Curve" analyses. That is, rather that thespecification proposed by A. W. Phillips (1958), the proper specification is thatproposed by Irving Fisher (1926) as asserted by Friedman. This finding is ofsignificance to those researchers using ordinary least squares to examinerelations between inflation and unemployment. Irving Fisher's specification isconsistent with Friedman's well known theoretical arguments concerningposited relations between inflation and unemployment, hence, his taste forinflation as the independent variable.

Maybe this box does not represent the final word in the Phillips curvecontroversy, but this research suggests that there is a short-run view of the Phillipscurve and that the idea of rational expectations may not be as good as it looks at firstblush. This study employed the Granger causality methods to inflation andunemployment data in the United States for the period identified to see if there wascausality -- the evidence suggests that inflation causes unemployment.

149

KEY CONCEPTS

Misery Index

Inflation

Unemployment

Phillips Curve

Short-run Trade-off

Long-term, natural rate hypothesis

Positively sloped

Rational Expectations

Market Policies

Lorenz Curve

Gini Coefficient

Investment Tax Credits

NAFTA & GATT

Wage-Price Policies

Laffer Curve

Supply Side Economics

Budget Deficit

National Debt

150

STUDY GUIDE

Food for Thought:

Compare and contrast the various views of the Phillips Curve. Map out each anddemonstrate how there may be a cruel choice in economic policy.

Develop the Laffer Curve. What does it tell us? How then can we have witnessed thelarge increases in the deficit during the years this model held center stage? Criticallyevaluate.

What are market policies? Explain the major market policies observed in the U.S.economy today. Compare and contrast these.

Sample Questions:

Multiple Choice:

In the short-run view of the Phillips curve policy-makers are left with a cruel choice. What is this cruel choice?

A. Inflation will exist regardless of the level of unemploymentB. Unemployment will exist regardless of the level of inflationC. Policies that improve unemployment will create inflation and vice versaD. Inflation appears not to have a causal relation with unemployment, hence adownward sloping Phillips curve is not plausible.

Stagflation is:

A. general decreases in the price levelB. excess employment which causes inflationC. both high rates of unemployment and inflationD. both very low rates of unemployment and inflation

151

True - False:

The Laffer curve suggests that the higher the tax rate the lower will be tax revenues.{FALSE}

Supply side economics was the view of the Reagan administration which believed Say'sLaw. {FALSE}

152

Chapter 13

Data: Categories, Sources, Problems and Costs

Business conditions analysis requires that the analyst have a solid understandingof the theoretical underpinnings of macroeconomics. However, theory is simply notenough. One must have data to monitor or analyze what is occurring in the economy. Data imply empirical economics, and a mastery of a tool-kit of empirical methods toanalyze the economy and perhaps make forecasts from those empirical methods. Thepurpose of this chapter is present the types of data, their sources, and the practicalproblems in the gathering, storing, and use of data. A following chapter, Chapter 15 willintroduce you to some preliminary types of data analysis to be used to become familiarwith and understand the data set with which you are working.

Data Types

Data may be categorized in several different ways, but all classification tend tobe by the nature of the underlying processes that generate that data. Economic dataare generally of two basic categories. There is time series data, and cross sectionaldata. Time series data are metrics for a specific variable observed over time. Cross-sectional data, on the other hand, are data for a specific time period, for a specificvariable, but across subjects. For example, unemployment rates are measured overtime. The U.S. annual average unemployment rate is calculated for months, quarters,and years – such data is time series data. On the other hand, for each month, quarterand year, the Bureau of Labor Statistics publishes unemployment rates for each State inthe U.S. Unemployment rate data for a specific month, quarter or year is called cross-sectional data.

Economists sometimes find that cross-sectional data over time provides forinsights not available by looking at only time-series or just cross-sectional data. Thecombination of cross-sectional data over time is called panel-data. The empiricalexamination of each type of data has its own peculiarities and statistical pitfalls. Thesewill be examined in greater detail, in a latter chapter.

There is discrete data, and continuous data. Discrete data are metrics thathave specific and finite number of potential values. Continuous data, on the other hand,are metrics which can take on any value either from plus infinity to negative infinity, orwithin specific limits. Discrete data generally require more sophisticated statistical

153

models than just simple ordinary least squares. Continuous variables are things suchas Federal budgets, GDP, and work force. Discrete variable include things like Likertscales in survey research.

Data can also be classified as to stocks and flows. A stock variable is generallya variable in which there is a specific value at some point. A flow variable is generallysomething that has some beginning, ending or changing value over time. Inventorydata is a useful example to illustrate the difference between stocks and flows. At anypoint in time a retail establishment will have a certain dollar value of goods on hand andfor sale. Such a metric is a stock variable. On the other hand, the amount of turn-overof that inventory each week or month is a flow variable.

Data is sometimes generated by surveys, rather than the measurement ofsome state of nature. Survey data is generally reported by an individual who may ormay not have some motivation to accurately report the data requested in the survey. Therefore, when surveys are used to gather data, two specific issues arise with respectto the data. These two issues are validity and reliability. Validity has two broaddimensions. The response rate from the sample of a population will define howrepresentative that sample is of the population. The sample must be large enough tohave confidence that inference can be drawn from the sample about the population. The second issue involves whether the items on the survey actually capture what theresearcher intends. Validity in this respect means that care must be exercised toassure that the items on the survey are constructed so as to eliminate ambiguities. Field testing and editing of the instruments make survey research often time consumingand difficult.1

This second issue with validity also has three distinct dimensions these are: (1)content validity, (2) criterion-related validity and (3) construct validity. Content validityconcerns whether the measures used actually are capturing the concepts that are thetarget of the research. Face validity is often what is relied on concerning content, that ison the face of it, it appears that the measures are adequate. Otherwise, one must havea theoretical basis for the measures. Criterion-related validity involves whether themeasures differentiate in a manner that predicts values in the dependent or criterionvariable. Finally, construct validity involves whether or not the empirical measures areconsistent with the hypothesis or theory being tested. Each of these validity issues aredescribed in greater detail in Sekaran’s book.2

Reliability has to do with assuring the accuracy of responses. Opinion surveys

1 Uma Sekaran, Research Methods for Business: A Skill Building Approach,second edition. New York: John Wiley and Sons, 1992 presents a detailed discussion ofthe nuts and bolts of survey and interview methods for business analyses.

2 Ibid.

154

and self-reporting is often very risky with respect to inference. A well-constructedsurvey will ask the same thing is more than one way so that correlations can becalculated to at least assure consistency in a respondent’s answers. Reliability ofsurvey results is often very problematic, but does not render these methods necessarilyunreliable. It just means thought must be given in ways to check the accuracy andconsistency of the data gathered.

States of nature data, involve things like market prices, number of persons, tonsof coal and other such metrics that can be physically observed or measured. Oftenthese data are estimated from surveys, i.e., Consumer Price Index, so that there aresituations where there are combinations of methods in gathering data.

Interviews and observational methods are also often utilized to gather what issometimes referred to as primary data. These data also have particular difficultiesconcerning the metrics used and the validity and reliability of the data gathered usingthese methods.

There are seasonally adjusted and not seasonally adjusted data. Seasonaladjustment methods vary by the type of data examined. The Consumer Price Indexcomes both seasonally adjusted, and without seasonal adjustment. There are almostalways published technical reports that discuss the methods used to adjust for seasonalvariations in the data.

Conceptually, there are a number of economic time series data which havespecific, predictable biases created by the time of year. Unemployment predictablyincreases every summer because students and teachers enter the work force during thesummer months, and then exit again at the end of summer or beginning of fall whenschool resumes. The Christmas season always results in a jump up in retail sales, thatwill decline when the return rush is over in January. In these cases seasonaladjustments permit long term trends to be identified that may be hidden in the short-term variations.

Data Sources

Data sources are also diverse. There are public data sources and private datasources. Public data sources are governments, and international organizations. Privatedata sources include both proprietary and non-proprietary sources. Each of these willbe briefly examined in the following paragraphs.

Public data sources for the United States includes both Federal and Stateagencies. Economic data is gathered, compiled and published by virtually everyFederal Agency. State and local agencies also gather data, much of which is reportedto a Federal agency which compiles and publishes the data. For example, crimestatistics are available annually from the Federal Bureau of Investigation for both

155

Federal and State crimes. The Justice Department directly gathers statistics on Federalcrimes, but has to rely on State reporting to compile statistics concerning crimes instates, such as car-theft, murder, and battery. The same relationship holds for severalother types of data. Unemployment is compiled in two distinct ways. There is anestablishment survey conducted by the Commerce Department’s, Current PopulationSurveys, which is then compiled and given over to the Bureau of Labor Statistics. There is also another series, called “covered” unemployment which is reported by theStates through their Employment Security Divisions, and those numbers are alsocompiled and reported by the Commerce and Labor Departments.

The Commerce Department gathers industry data, foreign trade data, and awealth of statistics concerning the National Income Accounts. Most developedcountries follow the lead of the United States in closely monitoring data concerningvarious demographic, economic, infrastructure and health statistics. However, there arealso several international organizations which are involved in these sorts of activities. Perhaps one of the easiest to use sources of international statistics comes from theU.S. Central Intelligence Agency. The CIA routinely publishes something called theWorld Fact Book which is available online at www.cia.gov and contains a wealth ofeconomic, political, demographic, and health care statistics for the majority of theworld’s countries.

The Board of Governors of the Federal Reserve System and several of theFederal Reserve Banks publish statistical series on various monetary aggregates. Mostof these data are available in historical time series.

The World Bank and International Monetary Fund monitor various nation’s trade,inflation, and indebtedness data. The World Bank routinely publishes the World DebtTables which reports various aspects of the debt markets for LDCs’ debts, both privateand sovereign. The United Nations, through the auspices of their dozens ofinternational agencies also gathers and publishes data concerning various aspects ofthe demographics and economics of the world’s nations.

These public sources of information publish current statistics, and most haveavailable historical time series data. Most of the data gathered by the U.S. Governmentis available online at sources such as www.bls.gov and www.federalreserve.gov. Thesedata also generally have accompanying handbooks or articles which describe how thedata are gathered, de-seasonalized, or otherwise technically handled.

There are also private sources of information, both proprietary and non-proprietary. Stock, bond, and commodity prices are generally proprietary informationavailable for a fee from a brokerage or investment bank. However, there are sources ofmuch of this information which have little or no cost. The S&P 500 stock price data isavailable on www.Standard&Poors.com however, if you need historical data that comesfor a price.

156

There are non-profit organizations that publish data. The National Bureau ofEconomic Research and other such organizations keep tabs on various aspects of thebusiness cycle and National Income Accounts and have data available. There are alsoseveral professional organizations and industry organizations that have data sources. Perhaps one of the more interesting of these is the ACCRA Price Index which is across-sectional price index published quarterly, to wish a person may subscribe for asmall fee. The market basket assumes a mid-management household and surveysmany smaller cities, Fort Wayne is included in the survey and has been pretty stable atabout 88% of the national average cost of living www.aacra.org.

Problems and Costs

How we know something is a matter of individual perception. That perception isshaped by the data we have and the observations we have made. These perceptionswill also be colored by a persons values and their training. As a result people untrainedin the scientific method will often attach significance to evidence they believe supportstheir position, it is a form of bias, and is often associated with anecdotal evidence. As itturns out, anecdotal evidence is sometimes very misleading. The natural tendency is torely on a single isolated observation and use inductive logic to form conclusions. What,however, is required scientifically is the systematic gathering of a large number ofobservations, and the use of deductive logic to test theories concerning why that data isthe way that it is. This however, is expensive in time and resources. Information isalmost never free.

It is the cost of information that often becomes a limiting factor in makingbusiness decisions, based on such things as business conditions analysis. Over thepast several years there have been numerous studies concerning the problems with theuse of information for making managerial decisions, even including strategicmanagement. Kathleen M. Sutcliff and Klaus Weber, published an article in the HarvardBusiness Review in May of 2003 which takes a very practical and managerial approachto the problems associated with “High Cost of Accurate Knowledge.”

Sutcliff and Weber argue that the accuracy of data is an expensive propositionand that too often managers have neither the expertise nor the time to fully analyzewhat imperfect information they can gather. What Sutcliff and Weber contend is thatintuition is perhaps a better modus operandi, what they call a humble optimism butbased on what knowledge can be gleaned without becoming a victim of the old cliche“paralysis by analysis.

The effects of accuracy are summarized in the following charts:

157

Performance

Perceptual accuracy

Sutcliff and Weber studies suggest that organizational change has an inverted Ushaped relation to perceptual accuracy. In other words, over the initial ranges there areincreases in the return to perceptual accuracy for positive organizational change up tosome optimal point. Beyond that optimal, additional accuracy actually results in negativereturns. On the other hand, organization performance is negatively correlated withperceptual accuracy. The more resources (time, etc.) are devoted to perceptualaccuracy of analysis or the underlying data the more likely performance will decline.

These authors account for this relation perceptual accuracy to both organizational

158

change and performance with several different factors. Chief among these factors arethe individual characteristics of the executives, growth trends in the industry, complexity,controllability of the business’ fortunes, positiveness in the culture, and the companies’contingent strategies. The more capable a manager, the more familiar with thebusiness, and more experienced with the specific issues which arise the more likely thatmanager is to make the right call. The more optimistic the organizational culture, themore likely it is that there will be a “get it done” ethic which guides the organization. Theability to control the business’ success regardless of environment makes data and itsanalysis less critical to the business, whereas the more complex the business’ relation toits environment the more critical data and its analysis becomes to the success of thecompany.

There is also an old cliche which is operable here. “A higher tide, floats all boats,”is true in this case. If the business is in a market in which there is rapid growth, thendata becomes less important to the extent that everyone is prospering. The dot comfirms in the 1990s, mini-steel mills in the last two decades, and numerous otherexamples exist that suggest some validity to this view.

There will be more concerning these issues in the final chapter of this book. Suffice it to say that there are significant internal costs of the management of data and itsuses.

There are other writers who point to other problems with data and its analysis. Inmany enterprises there are external requirements that force a firm to invest heavily in thegathering and analysis of data. Among the causes of this behavior is the fact that thereare substantial government regulations involving health and safety that require datagathering and analysis. The automobile industry is required to engage in quality controlmeasures, and document those measures with data for safety critical parts. Pharmaceutical companies have very stringent regulations from FDA concerning notother the safety of their products, but there must also be evidence of their efficacy beforethose products can go to market. While these data issues are not necessarily businessconditions in a macroeconomic sense, they are part of the regulatory environment whichface businesses and are of critical importance to the success of the enterprise. Further,just like market analysis, the development of this information and its analysis must bepart of the firm’s strategic planning functions. Again, issues to be addressed later in thiscourse.

Professor James Brian Quinn has observed that the move into “high-tech”business (science based) has forced many firms to examine their strategies and howthey deal with information. In fact, Dr. Quinn believes that these science basedindustries have been forced into a situation where knowledge sharing is of criticalimportance to the success of industries and firms within those industries. The followingquotation states his point:

159

Conclusions

Clearly the gathering and analysis of data has implications for both the success,and the cost structure of the firm. But it is also clear that much of the problemassociated with data, its acquisition and its use is also a function of the culture andbehavior within the organization. It is sort of amazing how “humility” seems to be arecurrent theme in this literature. In many organizations the control of information is asource of power within that business, and that is almost universally destructive to theenterprise.

Finally, economists have long recognized that there is cost associated withinformation. In the analysis of the macroeconomy it is worthy to note, that information,through both rational and adaptive expectations played very important roles in the resultsobtained from the economic models presented. The matter of expectations also playedsignificant roles in policy decisions – fiscal lags, inflationary expectations, and aggregatedemand are shaped heavily by what information people have and what they do with it.

James Brian Quinn “Strategy, Science and Management” Sloane ManagementReview, Summer 2002

No enterprise can out-innovate all potential competitors, suppliers and externalknowledge sources. Knowledge frontiers are moving too fast. In almost every majordiscipline, up to 90% of relevant knowledge has appeared in the last 15 years. . . .

To exploit such interactions, leading companies develop flexible core-knowledgeplatforms and, equally important, the entrepreneurial skills to seize opportunitywaves. They systematically disseminate and trade knowledge and, if necessary,share proprietary information, recognizing that larger, unexpected innovations mayarise to increase their own innovations’ value by orders of magnitude. They know it’simpossible to foresee all combinations and profit opportunities.

As was drive surfers, external forces often drive companies’ success.Managers must read the waves expertly. But that means adopting a humilityunfamiliar to traditional titans of industry. Accomplishments depend on ensembles ofothers’ work and unfold in way managers can’t anticipate. Because leaders can’tpredict which combinations will succeed, they can’t drive their organizations towardpredetermined positions.

160

KEY CONCEPTS

Data Categories

Time Series

Cross-section

Panel Data

Stocks and Flows

Discrete and Continuous

Surveys and states of nature

Validity and reliability

Seasonally adjusted versus not seasonally adjusted

Sources of Data

Public

Government

Relations between levels of government

International Organizations

Private

Non-Proprietary and proprietary

Costs of data acquisition and analysis

Determinants of performance

Organizational cultural

Performance-Organizational Change

Humility versus Power and the role of information

161

STUDY GUIDE

Food for Thought:

Why has information traditionally been power? Why has this concept being replacedwith the idea that humility should reign with respect to information and its uses? Explain.

Critically evaluate the idea that there is some optimal level of information for anorganization or a manager to utilize.

Compare and contrast the various types of data, their uses, and their short-comings formanagement of an enterprise.

Have advances in science resulted in the need for information in business enterprises? Explain.

What would you use a seasonally adjusted series of data for, over data that has not beenseasonally adjusted? Explain.

162

Sample Questions

Multiple Choice:

Data that is concerned with the levels of a particular variable at a specific time is:

A. FlowB. StockC. DiscreteD. Continuous

Perceptual accuracy is related to performance of a firm, according Sutcliff and Weber:

A. Negatively B. PositivelyC. Positively to a point then negativelyD. Not at all

True / False:

Panel data is a mixture of discrete and continuous date. {False}

Survey data has problems not associated with data which are measurements of states ofnature, and require greater attention be paid to reliability and validity. {True}

163

Chapter 14

Economic Indicators

Perhaps the easiest method of analysis of macroeconomic trends and cycles iswith the use of rather simple devices called “economic indicators.” Leading EconomicIndicators are often focused on by the press, particularly the financial press (i.e., CNBCand the Wall Street Journal). By examining specific data, without the use of complicatedstatistical methods one can often obtain a sense of where the macroeconomy is headed,is currently, and has been. The purpose of this chapter is to present the economicindicators, and describe their usefulness and limitations. Why these indicators are beinggiven such attention is not just their popularity in the press, they are also theoreticallyconnected to the business cycle and are useful, quick and dirty methods of forecastingwhere we are in the business cycle.

Cyclical Indicators: Background

The Arthur F. Burns and Wesley C. Mitchell at the Bureau of Economic Researchbegan to examine economic time series data in the 1930s to determine if there weredata which were useful in predicting business cycles.3 What this research producedwas the classification of approximately 300 statistical series by their response to thebusiness cycle. These data are classified as leading indicators, concurrent indicators,and trailing indicators.

The selection of variables to be included in the indices are done on the basis ofcriteria, discussed below. The index for leading indicators is designed to provide aforecasting tool which permits calls to be made in advance of turning points in thebusiness cycle. The index of concurrent indicators is a confirmatory tool, and the trailingindicator index suggests which each phase of the business cycle has been completed. There is nothing complicated about the application of these indices, as opposed to bigmultiple regression models. However, the simplicity is at the expense of model theunderlying processes which produce the time series data used to construct theseindicator indices.

These indicators are described in Bureau of Economic Analysis’ Handbook of

3 W. C. Mitchell and A. F. Burns, Statistical Indicators of Cyclical Revivals, NewYork: NBER Bulletin 69, 1938.

164

Cyclical Indicators.4 These indicators were selected using six criteria. These criteria aredescribed in John J. McAuley’s Economic Forecasting for Business: Concepts andApplications, to wit:

In other words, not only is there a basis in economic theory for these indicatorvariables, these indicator variables appear to be statistically well-behaved over time. Asa result the indicators are considered, in general, to be useful, simple guides to thevariations in the business cycle in the U.S.

Leading Indicators

There are twelve variables that are included in the index of leading economicindicators. Leading indicators are those variables which precede changes in thebusiness cycle. In a statistical sense, these variables are those which rather consistentlyand accurately lead turning points in the business cycle, as measured by other series –i.e., lagging indicators and coincident indicators which will be presented below, as well asthe national income accounting data, and employment data which are used to define thebusiness cycle.

Table 1 presents these variables and the weights of the series within the index ofleading indicators.

John McAuley,Economic Forecasting for Business: Concepts and Applications

Cyclical indicators are classified on the basis of six criteria: (1) the economicsignificance of the indicator; (2) the statistical adequacy of the data; (3) the timing ofthe series – whether it leads, coincides with, or lags turning points in overalleconomic activity; (4) conformity of the series to overall business cycles; a seriesconforms positively if it rises in expansions and declines in downturns; it conformsinversely if it declines in expansions and rises in downturns: a high conformity scoreindicates the series consistently followed the same pattern; (5) the smoothness of theseries’ movement over time; and (6) timeliness of the frequency of release – monthlyor quarterly – and the lag in release following the activity measured.

4 Handbook of Cyclcial Indicators. Washington, D.C.: Government Printing Office,1977.

165

______________________________________________________________________

Table 1: Leading Economic Indicator Index

Variable Weight in Index

1. Average workweek of production workers, manufacturing 1.014 2. Average weekly initial claims, State Unemployment Insurance

(Inverted) 1.041 3. Vendor performance (% change, first difference) 1.081 4. Change in credit outstanding .959 5. Percent change in sensitive prices, smoothed .892 6. Contracts and orders, plant and equipment (in dollars) .946 7. Index of net business formation .973 8. Index of stock price (S&P 500) 1.149 9. M-2 Money Supply .93210. New Orders, consumer goods and materials (in dollars) .97311. Building permits, private housing 1.05412. Change in inventories on hand and on order (in dollars, smoothed) .985

Source: Business Conditions Digest, 1983______________________________________________________________________

From the discussions of macroeconomic activity in the first eleven chapters of thistext, there should be hints as to why this selection of twelve variables form the leadingindicator index. Rather straightforward reasoning concerning the first two variables leadsdirectly to household income in base industries which drive the remainder of theeconomy. As the workweek increases so too does consumer income, and the inverse ofinitial unemployment claims also suggests something about the prospects for consumerincome. The base industries, manufacturing and other goods producing activities are thepro-cyclical industries, which support the service sector. As these industries becomemore capable of sustaining household incomes, this bodes will for industriesdownstream. The same reasoning extends to credit outstanding. As consumersanticipate their incomes declining they typically extend themselves less with respect totaking on new debt obligations, and attempt to repay the debts they currently have. Prices of sensitive items include those things which are cyclical sensitive. Consumerdurables and other large purchases which are interest rate sensitive are typicallypostponed when recession is anticipated, exerting downward pressure on the prices ofthose commodities.

To the extent that producers’ expectations account for these perceived changes inhousehold income, then we would expect producers to react prior to upturns anddownturns in economic activity. These producer reactions will be reflected in the VendorPerformance, plant and equipment, business formation, building permits, and inventory

166

series. If producers react, we would also expect retailers and wholesalers to react,hence orders for consumer goods and materials.

This leaves two components of the leading index which have not yet beendiscussed, these are stock prices, and the M-2 money supply. The M-2 money supply isa function of the monetary policies of the Fed. As the Fed wishes to induce economicrecovery with monetary policy, it is typically observed that the Fed moves to an easymonetary policy from a neutrality or if the Fed erred, a tight monetary policy. Hence theM-2 money supply increases prior to an upturn in economic activity. The opposite istypically observed in times of recession. As the Fed wishes to eliminate the risk ofinflation from an overheated economy, the Fed will retreat from an easy money policytowards neutrality or even a tight monetary policy which typically precedes a recession. In other words, the money supply changes observed conform with the coming businesscycle and precedes the inflection point.

Stock prices are perhaps the most interesting of the leading indicators. Goingback to the finance literature, there is an interesting hypothesis. Eugene Fama andothers working on rational expectations in securities markets proposed that the financialmarket price-in all available information, and therefore everything one needs to knowabout prices are already in price of stocks and bonds. This is called the efficient markethypothesis. In other words, a bear market signals that there is a recession about tooccur, and a bull market signals that economic expansion is on the way. In fact, theempirical evidence suggest that the financial markets are very good predictors of comingbusiness cycles and that the S&P 500 index is the best of those predictors.

Why, one may ask, is the stock market such a good predictor of future fortunes inthe U.S. economy. It is a simple matter of the fundamentals underpinning investments. Investors put their money where they believe the best returns are. When earnings beginto increase, inventories start to decline, and revenues start to grow, this is the time tobuy the stock. If you wait until the analysts start recommending the stock, and the pricesgo through the roof, you are late to the party and have pretty much nothing but downsiderisk, and little upside potential. Hence, it is clear the stock market should be an excellentpredictor of future economic fortunes.

Coincident Indicators

These indicators are the ones which happen as we are in a particular phase of thebusiness cycle, in other words, they happen concurrently with the business cycle. Statistically their variations follow the leading indicators and occur prior to the laggingindicators. There are four variables which comprise the Index of Coincident Indicatorsand they are presented in Table 2 below:

167

______________________________________________________________________

Table 2: Index of Coincident Indicators

Variable Weight

1. Employees on nonagricultural payrolls 1.0642. Index of industrial production, total 1.0283. Personal Income, less transfer payments 1.0034. Manufacturing and trade sales .905

Source: Business Conditions Digest, 1983.

______________________________________________________________________

The coincident indicator variables are those which happen at the particular pointsin the business cycle. That is, this variables tend to confirm what is going on presently. At the top of recovery cycle we should observe that each of these variables havereached their highpoint during the cycle, and their low point is generally observed duringthe trough of a recession. As these variables are increasing, then we should be inrecovery, and as they are declining we should be moving toward recession.

Employees on nonagricultural payrolls and the Index of Industrial Production arestock variables. Whereas Personal income and Manufacturing and trade sales are flowvariables. The stocks of output should be growing during recovery, as should personalincome (all industries) and sales. The exact opposite is true during recessions. Therefore, these variables are simply confirming what phase of the business cycle theeconomy is experiencing currently.

Lagging Indicators

Lagging indicators are those which trail the phase in the business cycle that theeconomy just experienced. Again, confirming what the leading indicators predicted, andwhat the coincident indicators also confirmed as it occurred. There are six variables thatcomprise the Index of Lagging Indicators. The movements of these variables trail whatactually transpired in the phases of the business cycle.

These six variables are presented in Table 3:

168

______________________________________________________________________

Table 3: Lagging Indicators

Variables Weights

1. Average duration of unemployment 1.0982. Labor cost per unit of output, manufacturing .8683. Ratio constant dollar inventories to sales, and manufacturing and sales .8944. Commercial and industrial loans outstanding 1.0095. Average prime rate charged by banks 1.1236. Ratio, consumer installment debt to personal income 1.009

Source: Business Conditions Digest, 1983.

______________________________________________________________________

The Index of Lagging Indicators follows what has already transpired in thebusiness cycle. The average duration of unemployment trails the recession in aneconomy. For the stock variable, duration of unemployment, to increase means thatthere are a lot of people out of work, who have been out of work for a considerableperiod of time. Hence, this variables tends to continue to increase once the recovery isin its early stages.

Labor costs in manufacturing also tend to increase after the recession has ended,because producers are hesitant to hire additional labor, and are willing to work overtimeat premium pay, to avoid hiring which may yet turn out to be temporary. The averageprime rate reflects the banks risk premium not yet being mitigated as a result ofexperience with a recovery. Hence the prime rate is sticky downward, much the sameas employment is sticky upward in the beginnings of a recovery.

Commercial and industrial loans tend to increase during periods in which liquidityis a problem for enterprise. These loans tend to reach their peak shortly after arecession has ended and recovery has begun. The consumer installment debt to incomeratio, also tends to reflect a recent history of a lack of income which has, in part, beensupplemented by credit at the worst of a recession for some households.

The time necessary to work down an inventory excess generally extends past theend of a recession for goods producing entities. Manufacturing companies will oftenaccumulate inventories which cannot be brought down during a recession, and must becarried forward until the recovery becomes strong enough that customers are willing andable to buy down those inventory levels. Hence, inventories for the base industries tend

169

to persist into the recovery phase of the business cycle.

Other Indicators

There are other indicators which are constructed and published concerningvarious aspects of the business cycle. The University of Michigan publishes a“Consumer Confidence” index. The University of Michigan has developed a surveyinstrument and surveys consumers concerning what they expect in their economicfuture. The Conference Board also does something similar for producers.

Real Estate firms have a lobby and research group that does surveys of newhome construction, and sales of existing housing which plays on essentially the samesort of economic activities captured by “Building permits, private housing” variable foundin the Index of Leading Indicators.

The Federal Reserve is not to be outdone in this forecasting business. ThePhiladelphia Federal Reserve Bank does an extensive survey of business conditions andplans in the Philadelphia Federal Reserve District and publishes their results monthly. This report is commonly referred to as “The Beige Book” and it eagerly anticipated asproviding insights into what to expect in the near future in the eastern part of the UnitedStates.

There are also numerous proprietary forecasts and indices constructed for everypurpose imaginable. Stock price data have been subjected to nearly every tortureimaginable in an attempt to gain some sort of trading advantage. Bollinger Bands, DowTheory, and even Sun Spots have been used in an attempt to forecast what will happenin the immediate short run with financial markets. As one might guess virtually all ofthese schemes are proprietary. The normal warning applies, if it sounds too good to betrue, it generally is little more than fun and games with little substance. Care should beexercised in taking advantage of these sorts of things.

Frankly with financial markets there is no substitute for one’s own due diligence,however, there are a couple of good sources of information. If one is interested in thecumulative wisdom of stock analysts, then Zacks (www.Zacks.com) is a good reportingservice which provides not only average analyst ratings, but their own recommendationsbased on sound fundamentals. Standard and Poors provides a rating system based onup to five stars for the best performing stocks, again based on sound fundamentals. Formutual funds, Morningstar provides both proprietary (a premium service) and non-proprietary ratings of mutual funds and considerable information concerning themanagement charges and loads on respective funds www.morningstar.com.

170

Conclusions

It is interesting to note that these indices have been constructed and are roughguides to what is occurring in the business cycle. However, it is also not encouraging toknow that as the economy changes, becomes more global, and more service oriented,that the efficacy of these indices becomes less clear. These have never been perfectlyaccurate, they are at best only rough guides to the landscape of business cycles in theU.S. economy, However, it also must be remembered that a subscription to these indicesis far cheaper than having a battalion of econometricians on staff gathering andanalyzing economic data.

Further, these numbers are also reasonable indicators of what is occurring andcan give considerable empirical pop for the buck. To the extent that all economicforecasting is, at best, little more than sophisticated number grubbing these sorts ofapproaches are worth careful consideration.

It is also interesting to note that OECD (Organization for Economic Cooperationand Development) gathers and publishes indices of economic indicators for severalEuropean nations at their website www.OECD.org. These numbers provide essentiallythe same sorts of leading, coincident, and lagging indicators as are described here.

The U.S. publication of the Economic Indicators is not longer shouldered entirelyby the Bureau of Economic Analysis. The Conference Board now publishes these seriesfor the U.S. These indicators are available through the Conference Board at theirwebsite and requires a subscription www.ConferenceBoard.org. However, theconstruction of the indices are still accomplished using data provided by the U.S.Department of Commerce, and many organizations have constructed their own indicesmirroring what was done by the Bureau of Economic Analysis.

There are specialized data systems mostly for stocks, and mutual funds. Theseare almost all proprietary for the premium services, but most brokerages provide one ormore these services, and some are available without a fee and without the premiumportions of the service. Many of these are quite good.

KEY CONCEPTS

Economic Indicators

Based in economic theory

Six criteria are used for variable selections

171

Different weighting used for each variable

Index of Leading Indicators

Prediction variables

Efficient Market Hypothesis

12 variables make up the index

Index of Coincident Indicators

Occurs with the cycle

4 variables make up the index

Index of Lagging Indicators

Confirms where we’ve been

6 variables make up the index

Variables are selected for their efficacy in the index

Other Indicators and Sources of Information

Proprietary services

Philadelphia Fed’s Beige Book

Investment information

Zacks

S&P

Morningstar

172

STUDY GUIDE

Food for Thought:

Critically evaluate the idea that an index can be created to predict future economicactivity, measure current economic activity, and confirm where we’ve been in thebusiness cycle.

Why do you suppose there are different weights for different variables in each of theseindices? Explain.

“Keep it Simple!!!” There is no doubt these indices are simple, but what do we give upfor the simplicity? Explain.

What other sorts of information is available based loosely on this index of indicatorsidea? Explain.

173

Sample Questions:

Multiple Choice:

Which of the following variables are included in the Index of Leading Indicators?

A. Vendor PerformanceB. Building Permits, Private HousingC. Index of Stock PricesD. All of the above

The duration of unemployment is included in the lagging indicators index. This variableindicates the average length someone has been off work. What is this variable?

A. A flow variableB. A stock variableC. A discrete variableD. None of the above

True / False

There are indicator indices constructed for and available for countries other than theUnited States. {True}

High conformity scores for variables in the respective economic indicator indices suggestthe series consistently followed the same pattern. {True}

174

Chapter 15

Guide to Analytical and Forecasting Techniques

Since time immemorial knowledge of the future has been a quest of all mankind. Frankly, we are all probably better-off not knowing what’s in store for us. However, whatis undisputable is that some insight into what business conditions are going to be, allowsus more an opportunity to prepare to take advantage of the opportunities which await us,and avoid the threats that may also be present. It is this risk that motivates one to haveas much information as is optimally available so as to effectively deal with thecontingencies we face. It is this motivation that brings most business organizations tothe need for forecasting.

The purpose of this chapter is to present a menu of forecasting techniques andprovide some guidance as to their utility and to what ends they are best suited. It ispresumed that students will have had a basic grounding in statistical methods so thatmost of the discussion presented here will be readily understood. However, it is not thepurpose of this chapter to go into the “nuts and bolts” of the methods presented. Asecond course, M509, Research Methods, will provide the computational expertisenecessary to deploy these methods.

Methods to be Presented

This chapter is divided into three basic sub-sections. The first sub-section willpresent a class of models referred to collectively as Time Series Methods. The nextsection deals with correlative methods, including ordinary least squares and variants ofthat model. The final section will present a sampling of other statistical methods whichmay be useful in dealing with special cases.

Time Series Methods

This class of methods is the most simple in dealing with time series data. Timeseries methods includes. There are several different approaches varying from the mostsimple, to somewhat more complex. The methods to be briefly presented here include:(1) trend line, (2) moving averages, (3) exponential smoothing, (4) decomposition, and(5) Autoregressive Integrative Moving Average (ARIMA).

Each of these methods is readily accessible, and easily mastered (with the

175

possible exception of certain variants of ARIMA). The first four of these methodsinvolves little more than junior high school arithmetic and can be employed for a varietyof useful things.

1. Trend Line

This method is also commonly called a naive method, it that is simply fitting arepresentative line to the data observed. One method, presented later, to determinewhat the representative line is, is called ordinary least squares, or regression analysis. In fitting a naive trend line, however, we rely on something more basic, that isextrapolation. In other words, if we have a pool of data and the plot of that data revealsa predictable pattern we simple fit the line over what that observed pattern is.

X ! ! ! ! ! !

! ! ! ! !

Y

The Nobel Prize winning economist Paul Samuelson (a Gary, Indiana native)described this method as the black string method. That is, if you have a collection ofdata point, simply pull a string out of your black socks and position in such a manner as itbest fits your data. In other words, best judgement suggests that the dotted line is mostrepresentative of this collective of data points. The line either passes through or touchesfive of eleven data points, with three points being above and three points being below thedotted line. However, there is nothing any more rigorous that visual observation andletting the eye tell you what looks like it best represents the data.

2. Moving Averages

Moving averages are the next most rigorous method of best guess. If we havethree observations over time of a particular variable, and we wish to obtain an estimateof the next observation we might choose to use something that gives use a particular

176

number, rather than a slope and intercept from which to extrapolate as in the trend linemethod above.

If we have three data points 6, 7, and 8. The moving average is calculated as:

0 = 6+7+8 ÷ 3 = 7

The moving average of this series is 7. However, it is clear that if the seriescontinues that the next number will be 9. Therefore, a weighted moving average mayresult in a more satisfactory result. If the we use a method that weights the lastobservation in the series more than the first observation we will get closer to the nextnumber in the series, which we presume is 9. Therefore, the following weighted averageis used:

0 = (.5) 6+7+ (1.5)8 ÷ 3 = 7.333

This weighting scheme still undershoots the forecast. Therefore we try usingsomething in which the last number is weighted more heavily.

0 = (.5) 6+7+(2)8 +1 ÷ 3 = 9

This weighting scheme with the addition of a constant term one provides a moresatisfactory result. Unfortunately, using a weighted average scheme results in guessingfor the best weighting scheme. Hence a better method is need.

This sort of method is often useful for forecasting very simple, “shop floor” sorts ofthings. If you are dealing with a single data series, and need a very rough estimate ofwhat is going to happen next with this data, the moving average, or exponentialsmoothing technique is often the most efficient.

3. Exponential Smoothing

One of the problems with moving averages is determining how much to weightobservations to obtain reasonably accurate forecasts. The weighting process is simplyguessing, albeit hopefully informed guessing, about what sort of weights will produce theleast error in future forecasts. Exponential smoothing is an effort to take some of theguess work out, or at least formalize the guess work. The equation for the forecastmodel is:

F = (%) X2 + (1 - %) S1

177

Where alpha is the smoothing statistic, X is the last observation in the data set, and S isthe first predicted value from the data set. For example, if we use the following series:

25, 23, 22, 21, 24

To obtain the initial value of S or S1 we add 25 and 23 and divided the sum by 2 to obtain

24. The X2 in this case is our last observation or 25. All that is left is the selection of %.Normally, an analyst will select .3 with a series which does not a linear trend upwards or

downwards. This gives a systematic method of determining the appropriate value of %.

Make predictions using % beginning at .1 and work your way to .9 and determine which

% gives the forecast with the least error term over a range of the latest available data,

and use that % until such time as error begin to increase. To complete the example, letsplug and chug to a get a forecast.

F = (.3) 25 + (1 - .3) 24 =

7.5 + 16.8 = 24.3

The mechanics are simple, far simpler than name implies, and the method toadjust for the best alpha is easy to implement.

3. Decomposition

There are several ways in which to decompose a time series data set. Outliers,trend, and seasonality are among the most common methods to strip away variations inthe data so as to be able to look at fundamentals in the data. Each of these methodsrelies upon an indexing routine in order to provide a basis to “normalize” the data. Perhaps the most complicated of these is seasonality, and therefore we will examine thatexample.

The process begins with calculating a moving average for the data series. Forquarterly data the moving average is:

MA = (Xt-2 + Xt-1 + Xt + Xt+1) / 4

178

We will apply this method to the following data:

Year 1 XFirst Quarter 10Second Quarter 12Third Quarter 13Fourth Quarter 11

Year 2 XFirst Quarter 11Second Quarter 13Third Quarter 14Fourth Quarter 11

Year 3 X First Quarter 10

Second Quarter 13Third Quarter 13Fourth Quarter 11

Applying our moving average method:

MA3 = (10+12+13+11)/4 = 11.50MA4 = (12+13+11+11)/4 = 11.75MA5 = (13+11+11+13)/4 = 12.00MA6 = (11+13+14+11)/4 = 12.25

The data are now presented for year as a moving average of the quarterscentered on the quarter represented by the number following the MA. For the year, thedata for this series centered on the third quarter is 11.50, centered on the fourth quarterit is 11.75 and for the first quarter of the previous year it is 12.00.

Now we can calculate an index for seasonal adjustment, in a rather simple andstraightforward fashion. The index is calculated using the following equation:

SI t = Y t / MA t

where SI t is the seasonal index, Y t is the actual observation for that quarter, and MA t isthe moving average centered on that quarter, from the method shown above. Aseasonally adjusted index can then be created which allows the seasonality of the data

179

to be removed.

Using the three years worth of data above we can construct a seasonality indexwhich allows us to remove the seasonal effects from the data. The average for eachquarter is calculated:

Y1 = (10+11+10)/3 = 10.33Y2 = (12+13+13)/3 = 12.67Y3 = (13+13+13)/3 = 13.00Y4 = (11+11+11)/3 = 11.00

Therefore plugging in the values for Y t and MA t the equation above allows us toconstruct an index for seasonality from this data:

SI 1 = 10.33/11.50 = .8983SI 2 = 12.67/11.75 = 1.0783SI 3 = 13.00/12.00 = 1.0833SI 4 = 11.00/12.25 = .8988

To seasonally adjust the raw data is then a simple matter of divided the raw databy its appropriate seasonal index.

Year 1 X Seasonally AdjustedFirst Quarter 10/0.8983 = 11.1321Second Quarter 12/1.0783 = 11.1286Third Quarter 13/1.0833 = 12.0004Fourth Quarter 11/0.8988 = 12.2385

Year 2First Quarter 11/0.8983 = 12.2453Second Quarter 13/1.0783 = 12.0560Third Quarter 14/1.0833 = 12.9235Fourth Quarter 11/0.8988 = 12.2385

Year 3 First Quarter 10/0.8983 = 11.1321

Second Quarter 13/1.0783 = 12.0560Third Quarter 13/1.0833 = 12.0004Fourth Quarter 11/0.8988 = 12.2385

Decomposition of a data series can be accomplished for monthly, weekly, even

180

daily data. It can also be used to eliminate variations that occur for reasons other thanjust seasonality. Any cyclical or trend influences in data can be controlled using theseclassical decomposition methods.

ARIMA

ARIMA is an acronym that stands for Autoregressive Integrated Moving Average. This method uses past values of a time series to estimate future values of the same timeseries. According to John McAuley, The Nobel Prize winning economist Clive Granger isalleged to have noted: “it has been said to be difficult that it should never be tried for thefirst time.” However, it is useful tool in many applications, and therefore at leastsomething more than a passing mention is required here.

There are basically three parts of an ARIMA model, as are described in McAuley’stext these three components are:5

“(1) The series can be described by an autoregressive (AR) component,such that the most recent value (X1) can be estimated by a weightedaverage of past values in the series going back p periods:

X1 = Θ (X t-1) + . . . + Θ p (X t-p) + δ + ε

where Θ p = the weights of the lagged values δ = a constant term related to the series mean, and

ε = the stochastic term

(2) The series can have a moving average (MA) component based on a qperiod moving average of the stochastic errors:

X1 = 0 + (εt - Θ1) εt-1 - . . . - Θq + εt-u

where 0 = the mean of the seriesΘq = the weights of the lagged error terms and

ε = the stochastic errors.

5 John J. McAuly, Economic Forecasting for Business: Concepts andApplications, Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1985.

181

(3) Most economic series are not stationary (can be transformed throughdifferencing using a stationary process), but instead display a trend orcyclical movements over time. Most series, however, can be madestationary by differencing. For instance, first differencing is the period-to-period change in the series:

ΔX t = X t - X t-1 “

As can readily been seen the application of an ARIMA model requires a certainmodicum of statistical sophistication that is beyond the scope this course. For thoseinterested, consider taking M509.

However, a few concluding remarks are necessary here. To apply an ARIMAmodel to data there are three procedures necessary. First, the analyst must determinewhat autoregressive process is most appropriate to the series. Second, the model mustbe then fitted to the data, and thirdly, diagnostic tests for check for the models adequacyand that the data are stationary are minimally necessary before attempting to drawinference from the results.

Clive Granger mentioned above won his Nobel Prize for extending ARIMAmodels to where they were useful in testing any casual relations that may exist betweentwo time series variables. The method is called, “Granger Causality.” In manysophisticated forecasting activities it is necessary to use models such as ARIMA, andunfortunately such models generally require a professional statistician and are generallybeyond the reach of most managerial personnel. However, that does not mean that youcan’t learn enough to be able to draw inference from the results, and understanding thelimitations and benefits of such modeling.

Correlative Methods

Correlative methods rely on the concept of Ordinary Least Squares, or O.L.S.,which is also sometimes referred to as regression analysis. O.L.S. refers to the fact thatthe representative function of the data is determined by minimizing the sum of thesquared errors of each of the data points from that line.

Consider the follow diagram:

182

X ! ! ! ! ! ! ! ! !

Y

The lines from each data point to the representative line are the error terms. Squaring the error terms eliminates the signs, and the regression analysis minimizesthese squared errors in order to determine what the most representative line is for theredata points.

The general form of the representative equation for a bi-variate regression is:

Y = α + βX + ε

where Y is the dependent variable, and X is the independent variable; α is the interceptterm, β is the slope coefficient and ε random error term.

Correlation

To be able to determine what a representative line is, requires some furtheranalysis. The model is said to be a good fit if the R squared is relatively high. There Rsquared can vary between 0 (no correlation) to 1 (perfect correlation). Typically an Fstatistic is generated by the regression program for the R squared which can bechecked against a table of critical values to determine if the R squared is significant

183

different than zero. If the F statistic is significant, then the R squared is significantlymore than zero.

The simple r is calculated as:

_ _ _ _ r = Σ (X - X) (Y - Y) / ([ Σ (X - X)2 ][ Σ (Y - Y)2 ])-1

and the simple r is squared to obtain the R squared.

Significance of Coefficients

There are also statistical tests to determine the significance of the interceptcoefficient and the slope coefficient, these are simple t-statistics. The test ofsignificance is whether the coefficient is zero. If the t-statistic is significant for thenumber of degrees of freedom, then the coefficient is other than zero.

In general, the analyst is looking for coefficients which are significant and of asign that makes theoretical sense. If the slope coefficient for a particular variable iszero, it suggests that there is no statistical association between that independentvariable and the dependent variable.

There is also a problem in which additional data being added to the series oranother variable will sometimes cause the sign of a coefficient already in the model toswap signs. This is a symptom of problem with the data called heteroskedasticity. Thisproblem is a result data not conforming to the underlying assumptions of ordinary leastsquares that is that the errors or residuals do not have a common variance. Youbasically have two ways in which to deal with this problem, select a different analyticaltechnique or try transforming the data. If you transform the data, log transformations ordeflating the data into some normalized series will sometime eliminate the problem andgive unbiased, efficient estimates.

With time series data there also arises a problem called serial correlation. Thatis when the error terms are not randomized as assumed. In other words, the errorterms are correlated with one another. There is test for this problem, called the Durbin-Watson (d) statistic. When dealing with time series data one should, as a matter ofroutine, have the program calculate a DW (d) and check with a table of critical values soas to eliminate any problems in inference associated with correlated error terms. If thisproblem arises, often using the first difference of observations in the data will eliminatethe problem. However, often a low DW (d) statistic is an indication that the regressionequation may be mis-specified which results in further difficulties, both statistically andconceptually.

184

Other Issues

Correlative methods are often used because of there accessibility. You can doregressions on Excel Spreadsheets. However, there are a couple of other issuesworthy of some mention here before moving on. If there is something that hashappened that is suspected of fundamentally changing the data series at someparticular point in time. Statisticians use a method called a dummy variable. If youbelieve that when we went off the gold standard in 1971 that fundamentally changed theprices of oil, then you would add a dummy variable to test that hypothesis. For eachyear that the country was off the gold standard the value of that variable would be 1 andfor all other years (i.e., when we were on the gold standard) the value would be zero. You may use multiple dummy variables, but there must be variation in the observations,otherwise you create for yourself something called a dummy variable trap - whichresults in a zero coefficient and other unpleasantries.

Instrumental variables are also sometimes used. These variables areconstructed for the purpose of substituting for something we could not directly measureor quantify. For example, in years in which we had heavy military commitments, wecould also have been at war. Clearly during the cold war there we heavily militarycommitments, but only intermittent combat. Rather than use a dummy for years at war,an instrumental variable such troops abroad, or Department of Defense budgets maycapture more of what the research was looking for.

OTHER METHODS

This chapter has simply scratched the surface of methods useful in statisticalanalysis of data and in forecasting. There are an array of other statistical methodswhich are useful for both purposes.

Included in these methods are nonparametric methods. Nonparametric methodsare useful with data in which quantification of the data is problematic. For example, theuse of survey methods generate results that are, in large measure, arbitrary. A Likertscale asking someone to strongly agree, agree, disagree or strongly disagree isimprecise. The ordering is all we know for sure, and that the magnitude is set in twooptions, rather than the continuous data used in parametric methods. Just becausedata is discrete does not mean there are not good ways to analyze that data.

There are also other methods of analyzing continuous data. There are uniquecircumstances in continuous data that do not necessarily lend themselves to the O.L.S.methods described above. We will also briefly describe some of these methods in thissection.

185

Continuous Data

One of the more simple and practical methods of examining data is the analysisof variance and analysis of covariance. Analysis of variance is typically used to isolateand evaluate the sources of variation within independent treatment variable todetermine how these variable interact. An analysis of variance is generally part of theoutput of any regression package and provides additional information concerning thebehavior of the data subjected to the regression.

There are variations on the analysis of variance theme. There is a multivariateanalysis of variance which provides information on the behavior of the variables in thedata set controlling for the effects of the other variables in the analysis., and theanalysis of covariance which goes a set further, as the name suggests, and providesan analysis of how the multiple variable interact statistically.

Factor analysis and cluster analysis are related processes which rely oncorrelative techniques to determine what variables are statistically related to oneanother. A factor may have several different dimensions. For example, men andwomen. In general, men tend to be heavier, taller, and live shorter lives. Thesevariables would form a factor of characteristics of human beings. Cluster analysis isanalogous in that rather than a particular theoretical requirement that results in factors,there may be simple tendencies, such as behavioral variables.

Canonical correlation analysis is a method where factors are identified and theimportance of each variable in a factor is also identified. These elements are calculatedfor a collection of variables which are analogous to a dependent variable in a regressionanalysis and a collection of variables which are analogous to independent variables in aregression. This gives the research the ability to deal with issues which may be multi-dimensional and do not lend themselves well to a single dependent variable. Forexample, strike activity in the United States has several different dimensions, and datameasuring those dimensions. The competing theories explaining what strikes happenand why they continue can then be simultaneously tested against the array of strikemeasurements, lending to more profound insights than if just one variable was used.

There are methods of analysis which involving the mixing of discrete andcontinuous data. Probit analysis uses categorical dependent variables (discrete) andexplains the variation in that variable with a mixture of continuous and discretevariables. The output of these computer programs yields output which permitsinference much the same as is done with regression analysis. If there is a stochasticdependent logit analysis is an alternative to probit.

186

Discrete Data

There are several nonparametric methods devised to deal specifically withdiscrete data. These methods are typically useful with survey data, such as marketingresearch or in some types of social science research in psychology or sociology. Wewill briefly examine a couple of the generally applicable tests for comparing datapopulations, and two of the most commonly used tests dealing with ordinal dataobtained from surveys, or other such methods.

The Mann-Whitney U test provides a method whereby two populationsdistributions can be compared. The Kruskal-Wallis test is an extension of the Mann-Whitney U test, in that this method provides an opportunity to compare severalpopulations. These tests can be used with the distributions of the data do not fit theunderlying assumptions necessary to apply an F test (random samples drawn fromnormally distributed populations with equal variances).

There are also methods of dealing with ordinally ranked data. For example, theWilcoxon Signed-Rank test can be utilized to analyze paired-differences inexperimental data. This is particularly useful in educational studies or in some types ofbehavioral studies where the F statistic assumptions are not fulfilled or the two testabove do not apply (not looking at population differences).

Finally, there is a test that is somewhat analogous to regression analysis to dealwith ranked data. Spearman Rank Correlations are useful in determining how closelyassociated individual observations may be to specific samples or populations. Forexample, item analysis in multiple choice exams can be analyzed using this procedure. Students who perform well on the entire exam can have their overall score assessedwith respect to specific questions. In this example, if the top students all performed wellon question 1, their rank correlation would be near 1, however, if they all performedwell, but on question 2 they did relatively poorly, that correlation would be closer to zero,say .35. If, on the other hand, none of the students who performed well got question 2correct, then we would observe a zero correlation for that question. This method isuseful in marketing research and in behavioral, as well as college professors with toomuch time on their hands.

There are several other nonparametric methods available. This section simplypresented the highlights so that you are aware that discrete data can be analyzed withdeveloped and generally accepted statistical techniques which are well supported with“canned” computer programs which are generally available commercially. Just becauseyou have discrete data does not mean that you have no reasonable methods to analyzethat data.

187

Conclusions

This chapter presented a brief overview of some of the commonly usedforecasting and statistical methods commonly used in business for analysis of specificissues concerning the business environment. This was done more as a menu of what’savailable rather than a seven course meal in statistical fun and games.

There is a well developed literature concerning these methods and numerousexcellent sources on the detailed nuts and bolts in applying these and other relatedstatistical methods. For the multivariate techniques6 there is an excellent text by Hair,Anerson et al, which is often the basic textbook for the M509 course. Also for a basictreatment of the nonparametric methods discussed here see Mendenhall, Reinmuth andBeaver’s text.7 This text also presents an excellent discussion of regression analysisand some of the related topics. In this discussion very little was offered concerningsurvey research and the particular problems associated with this specialized topic. Forthose desiring more concerning survey methods Sekaran’s text is an excellent primarysource.8

For those whose tastes run along the lines of the basic models used forforecasting described in the first section of this chapter, you may wish to consult a textby Wilson and Keating for further details.9 Of course, no course would be completeconcerning forecasting without some mention of the most advanced text books in thefield. For more in-depth and advanced discussions of forecasting see Makridakis,Wheelright and McGee.10 There are also several classic textbooks concerningeconometric methods. Still the most complete and comprehensive of these books is theone by Henri Theil.11 Finally, there is an article that is pretty much the authoritativepiece on manager’s guide to statistical and forecasting methods. This article appears in

6 J. F. Hair, R. E. Anderson, R. L. Tatham, and W. C. Black, Multivariate DataAnalysis with Readings, third edition. New York: Macmillian Publishing Company, 1992.

7 W. Mendenhal, J. E. Reimth, and R. Beaver, Statistics for Management andEconomics, sixth edition. Boston: PWS-Kent Publishing Company, 1989.

8 U. Sekaran, Research Methods for Business, second edition. New York: JohnWiley & Sons, 1992.

9 J. H. Wilson and B. Keating, Business Forecasting. Homewood,IL: Richard D.Irwin, Co., 1990.

10 S. Makridakis, S. C. Wheelright, and V. E. McGree, Forecasting: Methods andApplications, second edition. New York: John Wiley & Sons, 1983.

11 H. Theil, Principles of Econometrics. New York: John Wiley & Sons, 1976.

188

the Harvard Business Review in 1986.12 The reason for all of these citations, when weseemed to live quite nicely without them to this point in the course, is that knowledgecomes in two varieties, either you know something yourself, or you know where to gofind the information you need. These citations fall into the later category.

KEY CONCEPTS

Time Series Methods

Trend Line

Moving Averages

Exponential Smoothing

Decomposition

Autoregressive Integrative Moving Average

Correlative Methods

Regression

Correlation

Statistical Significance

Heterosckdasticity

Serial Correlation

Dummy Variables

Other Continuous Data Methods

Analysis of Variance

Multivariate Analysis of Variance

Analysis of Covariance

12 D. M. Georgoff and R. G. Murdick, “Manager’s guide to forecasting,” HarvardBusiness Review, January-February 1986.

189

Factor Analysis

Cluster Analysis

Canonical Correlation

Probit Analysis

Nonparametric Methods

Mann-Whitney U

Kruskal-Wallis

Wilcoxon Signed-Rank

Spearman Rank Correlations

STUDY GUIDE

Food for Thought:

What are the problems encountered in the data that may face you in applyingregression analysis to time series data? Explain.

Keeping things simple has obvious advantages with respect to cost and expertise, butwhat are the drawbacks? Be specific.

Compare and contrast the occasions in which the various methods presented here maybe appropriate or inappropriate to the analysis of business conditions.

190

What is meant by statistical significance? Why is this of importance? Explain.

Critically evaluate de-seasonalizing time series data.

Sample Questions:

Multiple Choice:

Which of the following models are methods which rely in some respect on correlation?

A. Exponential SmoothingB. Wilcoxon Signed-RankC. Trend LineD. None of the above

Serial correlation is:

A. The value of the simple rB. Correlation between the slope coefficients in a bi-variate regressionC. The residuals being correlated with one another in a multi-variate regressionD. None of the above

True / False

A significant t-statistic for a regression coefficient means we do not know that coefficientdiffers from zero. {False}

Correlation is the strength of statistical association between two variables, and Rsquared is the amount of variation explained in the dependent variable by theindependent variables. {True}

191

Chapter 16

Data: In the Beginning

This chapter is the follow-on of Chapter 13, but based, in part, on the discussionsfound in Chapter 15. We again, return to the subject of data, and what to do with it soas to accomplish the analyses or forecasts we wish to complete.

The purpose of this chapter to present brief discussion of the preliminaryexamination of data for the purposes of various forecasting and analytical techniques. This chapter presents some basics idea on how to become familiar with and handleproblems which commonly arise in empirical economics. It is through these methodsthat an analyst can best decide what methods are appropriate to her needs, and how tobest torture the data to obtain the desired results.

Data Inspection

Just like in the goods producing segments of the economy there are certainquality control procedures that must be implemented with data before you begin using itfor testing theories or forecasting. Familiarity with the data, its characteristics and itslimitations will help determine, not only the methods that will be applied to it, but theproblems and benefits would may expect to encounter with its use.

The following sections provide a basic set of tools and considerations for the“quality control” of data. The use of these methods can save a great deal of grief andaggravation and are highly recommended.

Becoming Familiar with the Data

The simple single variable forecasting techniques are not very sensitive to dataproblems which may create considerable difficulties with correlative methods. However,the time series data gathered may very well determine what single variable forecastingtechniques are best suited for forecasting purposes. Trend line, quite naturally is a bestguess method and is not subject to many of these problems, it is simply "eye-balling"the data. However, exponential smoothing is not particularly robust with data with largevariations, and little trend to it. Whereas, decomposition may be better suited to datawhich can have some of the variation removed through elimination of seasonality,

192

outliers, or other identifiable issues with the data.

Calculating means, medians, and modes will give you some idea what thedistribution of the data looks like. Mean is the arithmetic average, median is themid-point in the data series, and mode is the most commonly appearing observation inthe data. In a normal distribution, the mean, median, and mode are all equal to oneanother.

The first thing one should do when examining a data set, is to calculate each ofthese descriptive statistics to determine how its distribution looks. Further, somemeasure of variation should also be calculated, variance σ2 or standard deviation σwhich will provide some insight into how the data may be distributed. Plots of the datawill also help the analyst to come to some better understandings of how the data aredistributed as well.

Consider the following equations for the mean and for the variance for data thatis roughly normally distributed:

Mean:

0 = / nfmii n

k

i

Variance:

σ2 = 2 - [( )2 / n] / n - 1fmii n

k

i

fmi

i n

k

i

Standard Deviation:

σ = [ 2 - [( )2 / n] / n - 1] -1fmii n

k

i

fmi

i n

k

i

where: mi = midpoint of data series I fi = frequency of observation in series

193

The mean is used in the calculation of variance, however, it is a descriptivestatistic which give the analyst some indication of the magnitude of the observations. When compared with the median and the mode, the mean also provides someindication of how that data may be distributed.

Variance is also a useful idea. If data are distributed very tightly around themean, the variance will tend to be small. If, on the other hand, there are large numbersof outliers, or the data is tightly distributed the variance will tend to be large. Therefore,variance or the square root of variance which is standard deviation is useful indetermining how tightly distributed the data set is. Variance is commonly used, but asone can easily see from the equation is the square of the standard deviation. Standarddeviation is more easily interpreted because it is in the same numeraire as the dataitself.

The data series may be systematically distributed in some fashion that will havesome impact on modeling or technique decisions. Perhaps the most common of theseproblems is the data being skewed. The following graph shows skewed data:

This graph is presented with three variables, X, Y, and Z. The colored-in area, isthe plain created in this three-dimensional space, rather than the two dimensional lineyou are used to seeing. This data set is skewed to the left. If the tail observed is tothe other side of the distribution this is referred to as the data being skewed to the right. Notice that the mean, median and mode are all different numbers in this graph.

194

In examining a data set with the tail to the other side, and this time in twodimensions (just variables X and Y) we can observe a data set that is skewed to theright.

There is a statistical principle called the law of large numbers. For purposes ofmost statistical analyses, if the data are random sample from a population with thesame variance and continuous, then if you have more than 30 observations it isassumed that those number of observations or more will approximate a normaldistribution.

If the mode appears precisely in the middle of the plot, and mean, mode andmedian are all the same number, with a sombrero-shaped distribution, this is would bewhat is called a normal distribution. The following diagram presents a normaldistribution.

195

Correlated Independent Variables

In the discussion of the regression analysis in the previous chapter, there was

mention of correlated residuals sometimes referred to as serial correlation, ormulticolinearity. This most often occurs because there are two or more of theindependent variables – explanatory variables, which are correlated with one another. That is that they have a statistically significant correlation. That is a correlation which is.3 or higher (ignoring the sign).

The best way to handle this issue with respect to forecasting models is to makesure beforehand that you are not using correlated independent variables. Unfortunately, you may not have a choice in the selection of variables with a structuralmodel, or if you do, the instrumental variable you choose instead of the correlatedvariable may still exhibit a significant correlation with the remaining independentvariables. The simplest and quickest method for dealing with multicolinear variables isto run a correlation matrix of the variables being considered for use in the forecastingmodel. A correlation matrix is simply the correlation between each of the variablespresented.

Consider the following correlation matrix where we have correlatedunemployment rate, with Manufacturing inventories (in millions of dollars), Coincidentindex of copper prices, and the Standard and Poor’s 500 index of common stock pricesin the U.S.

196

_____________________________________________________________________ Unemployment Inventories Coincident S & P Rate Mfg. Copper Prices 500

_____________________________________________________________________

Unemployment rate 1.000

Inventories Mfg. -0.881 1.000

Copper Prices (coincident) -0.793 0.535 1.000

S&P 500 -0.679 0.790 0.243 1.000

_____________________________________________________________________

From inspection of this matrix it is clear that all of these variables are highlycorrelated with one another, except for Coincident Copper Prices and the S&P 500stock index. The unemployment rate is negatively associated with each of the othervariables, and manufacturing inventories are positively associated with both the price ofcopper and the S&P 500 index – although the statistical association between inventoriesand copper prices is not as high as the remaining variables. Copper prices and theS&P 500 may be useable in the same regression equation without causing significantstatistical problems.

In examining these variables it should come as no surprise that these variablesare highly correlated. Unemployment rate is highly correlated with persons unemployed(.98) which is also statistically associated with the lagging economic indicator “averageduration of unemployment.” The S&P 500 is very highly correlated with most of themajor categories within the National Income Accounts and is leading economicindicator. With some understanding of the economic theory which underpins theseindicators an analysis could very easily pick out the types of variables which would beexpected to highly correlated with one another.

Factor analysis will provide essentially the same sorts of information, and manyscholars will utilize factor analysis because it also give some basis for making otherdecisions. If there are highly correlated variables which are important to the analysis,and you have these sorts of difficulties on both sides of the equation, independent andpotential dependent variables which are highly correlated, then this is precisely whatcanonical correlation techniques are for. Therefore, these sorts of preliminary analysesmay be very useful in determining what methods are most appropriate to the task athand.

197

Forecasting versus Structural Models

It is wise to remember that business conditions analysis may involve two differentsorts of activities. There is the need to confirm theory or academic econometrics andthe need to get some fix on future developments – forecasting. There are differences inboth the rigor and the activities required when conducting research for the purposes ofunderstanding and explaining economic phenomenon or testing theories, and justforecasting. With empirical examination of a theory, there is the need to fully specifyyour model. Missing variables will bias results, and such under-specified models willrarely get past a referee. However, the formalities of academic research are lessrigorous when it comes to forecasting.

The large econometric forecasting models, such as the Wharton Model and othersuch simultaneous equations models are formal models of the U.S. economy andattempt to model each aspect of what is measured in the National Income Accounts. This sort of forecasting borders on academic research and is not the nuts and bolts sortof thing we expect from simple forecasts of a single variable or focused on somethingfar less complex. Forecasting is more pragmatic. Forecasters will look to find thosevariables that give consistent guidance to what might be expected in the selecteddependent variable, and if a variable can be found which predicts turn-around points,the forecaster is typically elated, if some connection can be theoretically found asconformation for the result.

One should always remember why you came to the swamp. Draining theswamp doesn’t necessarily mean you have to wrestle every alligator that wanders intocamp. Forecasting is a down and dirty application of statistical and economicknowledge which is clearly more focused on pragmatics than what is expected ofempirical tests of economic theory.

Conclusions

The more information you have concerning the nature and statistical properties ofthe data you have with which to work, the better choices you will be able to makeconcerning the selection of variables and statistical techniques. You are generally welladvised to make choices for variable to explain variations in your dependent variablewhich have both sound theoretical reasons for their inclusion, but which cause the leastamount of statistical headaches.

Prior planning does prevent poor performance. In this case, get to know the datayou are going to torture, it will help in the long - run.

198

KEY CONCEPTS

Know the Data

Independent variables

Dependent variables

Descriptive statistics

Mean

Mode

Median

Variance

Skewed Data

Correlation matrix

Choice of techniques related to data

Forecasting versus Structural equations

STUDY GUIDE

Food for Thought:

Why sorts of things would be useful to know about the variables you are going to use forindependent variables? Explain.

199

What are the differences between simple forecasting, and academic researchconcerning testing theories? Explain.

In what ways will getting to know your data help in analyzing it? Explain.

Critically evaluate the use of descriptive statistics, correlations matrices, and data plotsto select variables for inclusion in a forecasting model.

Multiple Choice:

When the mode is greater than the mean and median what is this called?

A. Normal distributionB. Skewed to the left dataC. Skewed to the right dataD. None of the above

Which of the following problems is associated with independent variables which arehighly correlated?

A. Residuals are correlatedB. The estimated coefficients may be biasedC. The estimated coefficient may be unstableD. All of the above

200

True / False

Standard deviation is one measure of the variability of a data set. {True}

A normal distribution has a mean, mode and median which are all equal, among othercharacteristics. {True}

201

Chapter 17

Epilog:

Management and Business Conditions Analysis

The purpose of this chapter is to examine the managerial aspects of businessconditions analysis. Data gathering and data sources are the starting point of anyattempt to analyze business conditions. But this must be done on a firm foundation ofsound economic analysis. The uses of the analyses and by whom will, in largemeasure, determine the rigor of analyses, and how it is communicated. Finally, in thischapter, the discussion of the managerial aspects will end with an examination of therole of this subject in strategic planning.

Information and Managerial Empowerment

To this point in the course, the discussion has focused almost entirely on thedevelopment of macroeconomic models used to examine and explain the businessenvironment in which an enterprise must function. In the broadest sense, it is themacroeconomic environment in which an enterprise must function. However, thiscourse, in a more practical sense, is a management course. All of the economics anddata analysis in the world doesn’t mean very much unless it can be translated into someproductive activities.

Information should, first and foremost, be one means by which people in anorganization are empowered to contribute to success of the organization. If theinformation technology gurus and the economists are the only ones with access toinformation, and the results of analyses then a barrier is being erected between thepeople who are responsible for the organization’s performance and the informationnecessary to discharge those responsibilities. Empowerment in a managerial senserequires that those closest to the events should be making the calls, within broad policyconstraints. To do this effectively, it is also clear that these people must have access tothe analyses and the data necessary to effective decision making. How this is done isof critical importance.

First and foremost, it must be remembered that information (its acquisition anddissemination) is a support function. Accounting, information technology and economicanalysis are staff function which are useful to line functions in determining the beststrategies for the organization to follow. These support functions are subsidiary

202

functions. There is a serious danger here, too many times these support functions takeon decision-making authority simply because they control the information necessary toeffective decision-making. THIS SHOULD NEVER HAPPEN. Economists andaccountants are not physicians, engineers or Indian chiefs and not the people with theline authority or expertise to properly execute that line authority.

Policy parameters must be set for the gathering, analysis and use of data. Openaccess to these data and the resultant analyses, regardless of where it is performed, isessential if it is to be useful within the context of overall management of theorganization. It is an extremely delicate balancing act, but discretion within the policyparameters is the essence of managerial empowerment. If you delegate authority, youmust permit discretion to exercise that authority, and information is one tool to makesure that discretion is exercised wisely.

Policy and Strategy

The purpose of a strategic plan is to provide not only guidance for managerialactions, hence policy, but also to provide a set of goals by which the directed actions ofthe enterprise can be assessed. An effective strategic plan is not just simply aroadmap. An effective strategic plan accounts for contingencies, and provides forguidance so as to reduce ambiguities with respect to what and how a firm should beperforming. Naturally, business conditions analysis plays a very important role informulating business strategies. Both opportunities and risks arise from the economicenvironment in which an enterprise operates. There for anticipation of those conditions,and contingencies for situation in which the anticipation was inaccurate are importantresults of business conditions analysis.

Decision-making is a fact of life in any organization. The culture that isdeveloped in an organization concerning how decision-making occurs is often one of themost important determinants of the organization’s success or failure. Consistently,studies of why enterprises fail suggest that poor planning is often at the heart of failure,and that effective planning is predictive of success. How this planning getsimplemented is part and parcel of decision-making.

Effective organizations empower those with line authority throughout theorganization. However, to effectively empower line managers, requires that their effortsbe focused and coordinated. Unfortunately, to some that means micro-managing, andthe organization falls victim to all the ills associated with this bankrupt managerialmodel. Effective organizations delegate authority, and have a mission statement thatprovides general directions, and a series of well understood and reasonable policies toguide those individuals to whom that authority has been delegated.

The role of information and its analysis in formulating and evaluating the goals

203

these policies are developed to achieve is critical. Not only are external businessconditions opportunities and threats, but how the organization responds to these issues,given the information, is also of critical importance and should be a crucial part of theplanning process.

Qualitative Issues

What has been done in this course is primarily quantitative analysis, however,this is not to minimize the qualitative aspects of a business environment. It is difficult tomeasure ethics, or business regulations, however both of these have significantimplications for the environment in which businesses operate.

There are other issues which tend to be more internal that are also shaped bythe business environment. The culture and design of organizations are often heavilyinfluence by what the business environment is anticipated to be. Enron’s culture was inlarge measure the result of rise of derivatives and fast moving financial markets. Enron’s management was successful in playing a sort of shell game moving losses andassets to give the appearance of a thriving company. It is doubtful this culture wouldhave developed at all, except for the financial market environment in which theyoperated.

As intrusive as Sarbanes - Oxley is for businesses, the legislative result waspredictable from the behavior of Ken Lay, Jeff Skilling, and Bernie Ebbers. When thesesorts of scandals arise which cost pension funds, and mutual funds billions of dollarsthere will be a public outcry for action to prevent such conduct in the future. Suchqualitative aspects of the business environment does not lend itself to quantification andstatistical analysis. Yet these matters are important aspects of the conditions in whichbusiness operates.

Forecasting and Planning Interrelations

Forecasting and planning cannot be separated, as a practical matter. Strategiesrequire a vision of the future. However, as pointed out earlier in the course, there arecosts to the gathering and analysis of information. Therefore, there is some optimalamount of analysis and forecasting that is consistent with an effective business plan.

There are some general rules which help in finding the optimal amount ofanalysis for the purposes of planning. Perhaps the foundation of these rules is arequisite flexibility to react to forecast "vision" errors. The religious adherence to plansis a formula for disaster. Plans are subject to forces which alter the results expected. Itis absolutely essential that the feed-forward, feed-back processes that generate the

204

original plans remain open during implementation so contingencies can be madeoperable and goals revised as conditions require. This culture of flexibility requires acertain humility and willingness to permit participation across the organization. Forthose with work experience, it is clear that this requirement is far easier said, then done.

Over time and across organizations we have become far better at understandingthe planning process than we were even twenty years ago. There are clearrelationships between goals specified in the plan and what the authors of those plansanticipated. A good strategic plan is going to specify specific objectives as guides towhether goals are being met. Those specific objectives will be measureable eitherqualitative or quantitatively. There will be provision for revisions of goals that theobjectives suggest are not being obtained or are being exceeded. Perhaps mostimportantly the process will be one of continuous improvement and ongoing.

Forecasting and business conditions analysis is just one critical cog in this overallwheel of strategic management. Goal are in an important sense nothing more thanforecasts, and the enabling objectives are simply road-signs along the way to allow usto gauge how well we are doing.

Risk and Uncertainty

Risk is simply the down-side of opportunity. We could call this chapter clichescentered on quantitative methods, but sometimes those cliches are accuratedescriptions of the real world. Just because they are cliche doesn’t mean you shouldn’tpay attention. Risk is the reason entrepreneurs are rewarded. Risk is also the reasonenterprises fail. The trick is to take those risks you understand and can appropriatelymanage. That sounds very simple, but its not or we would all be rich. To understandthe risk one faces requires insight and hard work – hard work to do the data gatheringand crunching necessary to have a handle on what you are up against. Most peoplesimply don’t gain the insights because they are either unwilling or unable to do thehomework necessary to make some enterprise work.

Stochastics are what breeds uncertainty, which, in turn, breeds ambiguity. Whenwe don’t know what to expect, that is sometimes the most anxiety producing situation. Remember when you were a kid, and you transgressed on some rules your parentsmade. The anxiety waiting for the verdict as to the appropriate corrective action wasalmost always worse than the punishment. That’s what uncertainty does for you. Tothe extent that the knowledge acquired in this course is useful to you as an individual itcan make your economic well-being less stressful. However, the lessons here properlyapplied can help you to understand your world a little better, but if extended to yourorganization can make the culture of the place you work less stressful and moreinformed. Understanding the macroeconomy and having at your disposal a fewforecasting techniques can help minimizes uncertainty and if extended through your

205

organization can help co-workers and the enterprise deal with ambiguities – at leastthose that arise from the business environment.

Conclusions

The issues examined in this course have not only a practical implication for thegathering and analysis of information. It also is concerned with where that analysis andinformation gathering fits in with the successful management of an organization. Information and its uses is a major part of what a manager does in the day to dayoperations of her part of an organization. However, it is unlikely that anyone in anM.B.A. program will be put in charge of a major forecasting model for a big brokerage orFederal Reserve Bank. Just as unlikely is that the typical M.B.A. will progress throughtheir careers without dealing with the issues presented here on a routine basis. Therefore it is of critical importance that you at least know what the macroeconomicenvironment is, and how analysts deal with the business cycle.

Go forth and prosper!!!

206

E550

Lecture

Notes

1. Introduction to Economics

Lecture Notes

1. Economics Defined - Economics is the study of the allocation of SCARCE resourcesto meet unlimited human wants.

a. Microeconomics - is concerned with decision-making by individual economicagents such as firms and consumers.

b. Macroeconomics - is concerned with the aggregate performance of the entireeconomic system.

c. Empirical economics - relies upon facts to present a description of economicactivity.

d. Economic theory - relies upon principles to analyze behavior of economicagents.

e. Inductive logic - creates principles from observation.

f. Deductive logic - hypothesis is formulated and tested.

2. Usefulness of economics - economics provides an objective mode of analysis, withrigorous models that are predictive of human behavior.

3. Assumptions in Economics - economic models of human behavior are built uponassumptions; or simplifications that permit rigorous analysis of real world events,without irrelevant complications.

a. model building - models are abstractions from reality - the best model is theone that best describes reality and is the simplest.

b. simplifications:

1. ceteris paribus - means all other things equal.

2. problems with abstractions, based on assumptions. Too often themodels built are inconsistent with observed reality - therefore they arefaulty and require modification. When a model is so complex that it cannot

1

be easily communicated or its implications understood - it is less useful.

4. Goals and their Relations - Positive economics is concerned with what is; Normativeeconomics is concerned with what should be. Economic goals are value statements.

a. Most societies have one or more of the following goals:

1. Economic efficiency,

2. Economic growth,

3. Economic freedom,

4. Economic security,

5. Equitable distribution of income,

6. Full employment,

7. Price level stability, and

8. Reasonable balance of trade.

5. Goals are subject to:

a. interpretation - precise meanings and measurements will often become thesubject of different points of view, often caused by politics.

b. complementary - goals that are complementary are consistent and can oftenbe accomplished together.

c. conflicting - where one goal precludes or is inconsistent with another.

d. priorities - rank ordering from most important to least important; againinvolving value judgments.

6. The Formulation of Public and Private Policy - Policy is the creation of guidelines,regulations or law designed to affect the accomplishment of specific economic goals.

a. Steps in formulating policy:

2

1. stating goals - must be measurable with specific stated objective to beaccomplished.

2. options - identify the various actions that will accomplish the statedgoals & select one, and

3. evaluation - gather and analyze evidence to determine whether policywas effective in accomplishing goal, if not re-examine options and selectoption most likely to be effective.

7. Objective Thinking:

a. bias - most people bring many misconceptions and biases to economics. Because of political beliefs and other value system components rational,objective thinking concerning various issues requires the shedding of thesepreconceptions and biases.

b. fallacy of composition - is simply the mistaken belief that what is true for theindividual, must be true for the group.

c. cause and effect - post hoc, ergo propter hoc - after this, because of thisfallacy.

1. correlation - statistical association of two or more variables.

2. causation - where one variable actually causes another. Grangercausality states that the thing that causes another must occur first, that theexplainer must add to the correlation, and must be sensible.

d. Forecasting - Steps1. Selection of variables2. Sophistication3. Model4. Specification5. Re-evaluation of model6. Reports etc.

3

2. National Income Accounting

Lecture Notes

1. Gross Domestic Product - (GDP) the total value of all goods and services producedwithin the borders of the United States (or country under analysis).

2. Gross National Product - (GNP) the total value of all goods and services producedby Americans regardless of whether in the United States or overseas.

3. National Income Accounts are the aggregate data used to measure the well-being ofan economy.

a. The mechanics of these various accounts are:

Gross Domestic Product

- Depreciation =

Net Domestic Product

+ Net American Income Earned Abroad- Indirect Business Taxes =

National Income

- Social Security Contributions- Corporate Income Taxes- Undistributed Corporate Profits+ Transfer Payments =

Personal Income

- Personal Taxes =

Disposable Income

4

4. Expenditures Approach vs. Incomes Approach

a. Factor payments + Nonincome charges - GNP/GDP adjustments = GDP is the incomes approach

b. Y = C + Ig + G + Xn is the expenditures approach (where Y = GDP)

5. Social Welfare & GDP - GDP and GNP are nothing more than measures of totaloutput (or income). More information is necessary before conclusions can be drawnconcerning social welfare. There are problems with both measures, among these are:

a. Nonmarket transactions such as household-provided services or barter arenot included in GDP.

b. Leisure is an economic good but time away from work is not counted,however, movie tickets, skis, and other commodities used in leisure time are.

c. Product quality - no pretense is made in GDP to account for product or servicequality.

d. Composition & Distribution of Output - no attempt is made in GDP data toaccount for the composition or distribution of income or output. We must look atsectors to determine composition and other information for distribution.

e. Per capita income - is GDP divided by population, very rough guide toindividual income, but still mostly fails to account for distribution.

f. Environmental problems - damage done to the environment in production orconsumption is not counted in GDP data unless market transactions occur toclean-up the damage.

g. Underground economy - estimates place the amount of undergroundeconomic activities may be as much a one-third of total U.S. output. Criminalactivities, tax evasion, and other such activities are the underground economy.

6. Price Indices - are the way we attempt to measure inflation. Price indices are farfrom perfect measures and are based on surveys of prices of a specific market basketof goods.

a. Market-basket surveys - The market basket of goods and services are

5

selected periodically in an attempt to approximate what the average family of fourpurchases at that time.

b. CPI (U) is for urban consumers & CPI (W) is for urban wage earners. GDPDeflator is based on a broader market basket and may be more useful inmeasuring inflation.

1. Standard of living - is eroded if there is inflation and no equal increasein wages.

2. COLA - are escalator clauses that tie earnings or other payments to therate of inflation, but only proportionally.

3. Other indices - American Chamber of Commerce ResearchAssociation in Indianapolis does a cross sectional survey, there arewholesale price indices and several others designed for specific purposes.

c. Inflation/Deflation - throughout most of U.S. economic history we haveexperienced deflation - which is a general decline in all prices. Inflation isprimarily a post-World War II event and is defined to be a general increase in allprices.

d. Nominal versus Real measures - economists use the term nominal todescribe money value or prices (not adjusted for inflation); real is used todescribe data which are adjusted for inflation.

7. Paashe and Laspeyres Indices

L = 3PnQ0 / 3P0Q0 X 100

Laspeyres index is the most widely used index in constructing price indices. It has thedisadvantage of not reflecting changes in the quantities of the commodities purchasedover time. The Paasche index overcomes this problem by permitting the quantities tovary over time, which requires more extensive data grubbing. The Passche index is:

P = 3PnQn / 3P0Qn X 100

8 Measuring the price level

a. CPI = (current year market basket/ base year market basket) X 100the index number for the base year will be 100.00 (or 1 X 100)

6

b. Inflating is the adjustment of prices to a higher level, for years when the indexis less than 100.

c. Deflating is the adjustment of prices to a lower level, for years when the indexis more than 100.

1. to change nominal into real the following equation is used:

Nominal value/(price index/100)

d. Changing base years - a price index base year can be changed to create aconsistent series (remembering market baskets also change, hence the processhas a fault). The process is a simple one. If you wish to convert a 1982 baseyear index to be consistent with a 1987 base year, then you use the indexnumber for 1982 in the 1987 series and divide all other observations for the 1982series using the 1982 value in 1987 index series.

7

3. Unemployment and Inflation

Lecture Notes

1. Business Cycle - is the recurrent ups and downs in economic activity observed inmarket economies.

a. troughs are where employment and output bottom-out during a recession(downturn)

b. peaks are where employment and output top-out during a recovery (upturn)

c. seasonal trends are variations in data that are associated with a particularseason in the year.

d. secular trends are long-run trend (generally 25 or more years inmacroeconomic data.

2. Unemployment - there are various causes of unemployment, including:

a. frictional - consists of search and wait unemployment which is caused bypeople searching for employment or waiting to take a job in the near future.

b. structural - is caused by a change in composition of output, change in

8

technology, or a change in the structure of demand.

c. cyclical - due to recessions, (business cycle).

3. Full employment - is not zero unemployment, full employment unemployment rate isthe same as the natural rate.

a. natural rate - is thought to be about 4% and is structural + frictional unemployment.

1. potential output - is the output of the economy at full employment.

4. Unemployment rate - is the percentage of the workforce that is unemployed.

a. labor force - those employed or unemployed who are willing, able andsearching for work; the labor force is about 50% of the total population.

b. part-time employment - those who do not have 40 hours of work (orequivalent) available to them, at 6 million U.S. workers were involuntarily part-time, and about 10 million were voluntarily part-time employees in 1992.

c. discouraged workers - those persons who dropped out of labor force becausethey could not find an acceptable job.

d. false search - those individuals who claim to be searching for employment,but really were not, some because of unemployment compensation benefits.

5. Okun's law

a. Okun's Law states that for each 1% unemployment exceeds the natural ratethere will be a gap of 2.5% between actual GDP and potential GDP.

6. Burden of unemployment differs by several factors, these are:

a. Occupation - mostly due to structural changes.

b. Age young people tend to experience more frictional unemployment.

c. Race and gender reflect discrimination in the labor market and sometimes in

9

educational opportunities.

7. Inflation - general increase in all prices.

a. CPI - is the measure used to monitor inflation.

b. Rule of 70 -- the number of years for the price level to double = 70/%annualrate of increase.

8. Demand - pull inflation

Using a naive aggregate demand - aggregate supply model (similar to the supplyand demand diagrams for a market, except the supply is total output in allmarkets and demand is total demand in all markets, as the aggregate demandshifts outwards prices increase, but so does output.

9. Cost - push inflation - again using a naive aggregate supply - aggregate demandapproach cost-push inflation results from a decrease in aggregate supply:

10

a. pure inflation results from an increase in aggregate demand that is equal to adecrease in aggregate supply:

10. Effects of inflation impact different people in different ways. If inflation is fullyanticipated and people can adjust their nominal income to account for inflation thenthere will be no adverse effects, however, if people cannot adjust their nominal incomeor the inflation is unanticipated those individual will see their standard of living eroded.

a. Debitors typically benefit from inflation because they can pay loans-off in thefuture with money that is worth less, thereby creditors are harmed by inflation.

11

b. Inflation typically creates expectations among people of increasing prices,which may contribute to future inflation.

c. Savers generally lose money because of inflation if the rate of return on theirsavings is not sufficient to cover the inflation rate.

12

4. Aggregate Supply & Aggregate Demand

Lecture Notes

1. Aggregate demand is a downward sloping function that shows the inverse relationbetween the price level and domestic output. The reasons that the aggregate demandcurve slopes down and to the right differs from the reasoning offered for individualmarket demand curves (substitution & income effects - these do not work withaggregates). The reasons for the downward sloping aggregate demand curve are:

a. wealth or real balance effect - higher price level reduces the real purchasingpower of consumers' accumulated financial assets.

b. interest-rate effect - assuming a fixed supply of money, an increase in theprice level increases interest rates, which in turn, reduces interest sensitiveexpenditures on goods and services (e.g., consumer durables - cars etc.

c. foreign purchases effect - if prices of domestic goods rise relative to foreigngoods, domestic consumers will purchase more foreign goods as substitutes.

13

2. Determinants of aggregate demand - factors that shift the aggregate demand curve.

a. expectations concerning real income or inflation (including profits from investments in business sector),

b. Consumer indebtedness,

c. Personal taxes,

d. Interest rates,

e. Changes in technology,

f. Amount of current excess capacity in industry,

g. Government spending,

h. Net exports,

i. National income abroad, and

j. Exchange rates.

3. Aggregate Supply shows amount of domestic output available at each price level. The aggregate supply curve has three ranges, the Keynesian range (horizontal), theintermediate range (curved), and the classical range (vertical).

a. Keynesian range - is the result of downward rigidity in prices and wages.

b. Classical range - classical economists believed that the aggregate supplycurve goes vertical at the full employment level of output.

14

c. Intermediate range - is the range in aggregate supply where rising output isaccompanied by rising price levels.

4. Determinants of Aggregate Supply cause the aggregate supply curve to shift.

a. Changes in input prices,

b. Changes in input productivity, and

c. Changes in the legal/institutional environment.

5. Macroeconomic Equilibrium - intersection of aggregate supply and aggregatedemand:

6. Ratchet Effect - is where there is a decrease in aggregate demand, that producersare unwilling to accept lower prices (rigid prices and wages) therefore there is aratcheting of the aggregate supply curve (decrease in the intermediate and Keynesianranges) which will keep the price level the same, but with reduced output. In otherwords, there can be increases in prices (forward) but no decreases (but not backward).

15

An increase in aggregate demand from AD1 to AD2 moves the equilibrium from point ato point b with real output and the price level increasing. However, if prices areinflexible downward, then a decline in aggregate demand from AD2 to AD1 will notrestore the economy to its original equilibrium at point a. Instead, the new equilibriumwill be at point c with the price level remaining at the higher level and output falling tothe lowest point. The ratchet effect means that the aggregate supply curve has shiftedupward (a decrease) in both the Keynesian and intermediate ranges.

16

5. Classical and Keynesian Models

Lecture Notes

1. Classical theory of employment (macroeconomics) rests upon two foundingprinciples, these are:

a. underspending unlikely - spending in amounts less than sufficient to purchasethe full employment level of output is not likely.

b. even if underspending should occur, then price/wage flexibility will preventoutput declines because prices and wages would adjust to keep the economy atthe full employment level of output.

2. Say's Law "Supply creates its own demand" (well not exactly)

a. in other words, every level of output creates enough income to purchaseexactly what was produced.

b. among others, there is one glaring omission in Say's Law -- what aboutsavings?

3. Savings

a. output produces incomes, but savings is a leakage

b. savings give rise to investment and the interest rates are what links savingsand investment.

4. Wage-Price flexibility

a. the classicists believed that a laissez faire economy would result inmacroeconomic equilibria and that only the government could cause disequilibria.

5. Keynesian Model - beginning in the 1930s the classical models failed to explain whatwas going on, hence a new model was developed -- the Keynesian Model.

a. full employment is not guaranteed, because interest motivates bothconsumers & businesses differently - just because households save does notguarantee businesses will invest.

b. price-wage rigidity, rather than flexibility was assumed by Keynes

17

6. The Consumption schedule - income & consumption

a. consumption schedule - the 45 degree line is every point where disposableincome is totally consumed.

b. saving schedule - shows the amount of savings associated with theconsumption function.

The consumption schedule intersects the 45 degree line at 400 in disposableincome, this is also where the savings function intersects zero (in the graph below theconsumption function). To the left of the intersection of the consumption function andthe 45 degree line, the consumption function lies above the 45 degree line. Thedistance between the 45 degree line and the consumption schedule is dissavings,shown in the savings schedule graph by the savings function falling below zero. To theright of the intersection of the consumption function with the 45 degree line theconsumption schedule is below the 45 degree line. The distance that the consumptionfunction is below the 45 degree line is called savings, shown in the bottom graph by thesavings function rising above zero.

b. Marginal Propensity to Consume (MPC) is the proportion of any increase indisposable income spent on consumption (if all is spent MPC is 1, if none isspent MPC is zero). The Marginal Propensity to Save (MPS) is the proportion ofany increase in disposable income saved. The relation between MPC and MPSis:

1. MPC + MPS = 1

18

c. The slope (rise divided by the run) of the consumption function is the MPCand the slope of the savings function is the MPS. Add the slope of theconsumption function (8/10) to the slope of the savings function (2/10) and theyequal one (10/10).

d. The Average Propensity to Consume (APC) is total consumption divided bytotal income, Average Propensity to Save (APS) is total savings divided by totalincome. Again, if income can be either saved or consumed (and nothing else)then the following relation holds:

1. APC + APS = 1

7. The nonincome determinants of consumption and saving are (these cause shifts inthe consumption and saving schedules):

a. Wealth,

b. Prices,

c. Expectations concerning future prices, incomes and availability ofcommodities,

d. Consumer debts, and

e. Taxes.

8. Investment

a. investment demand curve is downward sloping:

19

b. determinants of investment demand are:

1. acquisition, maintenance & operating costs,

2. business taxes,

3. technology,

4. stock of capital on hand, and

5. expectations concerning profits in future.

c. Autonomous (determined outside of system) v. induced investment (functionof GDP):

1. Instability in investment has marked U.S. economic history.

20

2. Causes of this instability are:

a. Variations in the durability of capital,

b. Irregularity of innovation,

c. Variability of profits, and

d. Expectations of investors.

21

6. Equilibrium in the Keynesian Model

Lecture Notes

1. Equilibrium GDP - is that output that will create total spending just sufficient to buythat output (where aggregate expenditure schedule intersects 45 degree line).

a. Disequilibrium - where spending is insufficient (recessionary gap) or too highfor level of output (inflationary gap).

2. Expenditures - Output Approach

a. Y = C + I + G + X is the identity for incomewhere Y = GDP, C = Consumption, I = Investment, G= Governmentexpenditures, and X = Net exports (exports minus imports)

The equilibrium level of GDP is indicated above where C + I is equal to the 45degree line. Investment in this model is autonomous and the amount of investment isthe vertical distance between the C and the C + I lines. This model assumes nogovernment and that net exports are zero.

3. Leakages - Injections Approach relies on the equality of investment and savings atequilibrium.

a. I = S is equilibrium in the leakages - injections approach.

22

b. planned v. actual investment, the reason that the leakages - injectionapproach works is that planned investment must equal savings. Inventories canincrease beyond that planned, hence output that is not purchased which isrecessionary; or intended inventories can be depleted which is inflationary.

The original equilibrium is where I1 is equal to S and that level of GDP is shownwith the solid indicator line. If we experience a decrease in investment we move downto I2 and if an increase in investment is observed it will be observed at I3.

4. If there is an increase in expenditures, there will be a respending effect. In otherwords, if $10 is injected into the system, then it is income to someone. That first personwill spend a portion of the income and save a portion. If MPC is .90 then the firstindividual will save $1 and spend $9.00. The second person receives $9.00 in incomeand will spend $8.10 and save $0.90. This process continues until there is no moneyleft to be spent. Instead of summing all of the income, expenditures, and/or savingsthere is a short-hand method of determining the total effect -- this is called theMultiplier, which is:

a. Multiplier M = 1/1-MPC or 1/MPS

b. significance - any increase in expenditures will have a multiple effect on theGDP.

5. Paradox of thrift - the curious observation that if society tries to save more it mayactually save the same amount - unless investment moves up as a result of the savings,all that happens is that GDP declines and if investment is autonomous then savingsremain the same.

23

6. Full Employment level of GDP may not be where the aggregate expenditures lineintersects the 45 degree line. There are two possibilities, (1) a recessionary gap or (2)an inflationary gap, both are illustrated below.

a. Recessionary gap

The distance between the C + I line and the 45 degree line along the dashedindicator line is the recessionary gap. The dotted line shows the currentmacroeconomic equilibrium.

.9 GDP potential = $1000 billion divide each side by .9

get $1,111.11

b. Inflationary gap

The distance between the C + I line and the 45 degree line along the dashed

24

indicator line is the inflationary gap. The dotted indicator line shows the currentmacroeconomic equilibrium.

7. Reconciling AD/AS with Keynesian Cross the various C + I and 45 degree lineintersections, if multiplied by the appropriate price level will yield one point on theaggregate demand curve. Shifts in aggregate demand can be shown with holding theprice level constant and showing increases or decreases in C + I in the KeynesianCross model. Both models can be used to analyze essentially the samemacroeconomic events.

25

7. John Maynard Keynes and Fiscal Policy

Lecture Notes

1. Discretionary Fiscal Policy - involves government expenditures and/or taxes tostabilize the economy.

a. Employment Act of 1946 - formalized the government's responsibility inpromoting economic stability.

b. simplifying assumptions for the analyses presented here:

1. exogenous I & X,

2. G does initially impact private decisions,

3. all taxes are personal taxes,

4. some exogenous taxes collected,

5. no monetary effects, fixed initial price level, and

6. fiscal policy impacts only demand side.

2. Changes in Government Expenditures - can be made for several reasons:

a. Stabilization of the economy,

1. To close a recessionary gap the government must spend an amountthat time the multiplier will equal the total gap.

2. To close an inflationary gap the government must cut expenditures byan amount that times the multiplier will equal the inflationary gap.

b. Political goals, and

c. Provision of necessary goods & services.

26

An increased government expenditure of $20 billion results in an increase in GDPof $80 billion with an MPC of .75, hence a multiplier of 4.

3. Taxation effects both consumption and savings.

a. If the government uses a lump sum tax increase to reduce an inflationary gapthe reduction in GDP occurs thusly:

1. The lump sum tax must be multiplied by the MPC to obtain thereduction in consumption;

2. The reduction in consumption is then multiplied by the multiplier.

b. A decrease in taxes works the same way, the total impact is the lump sumreduction times the MPC to obtain the increase in consumption, which is, in turn,multiplied by the multiplier to obtain the full impact on GDP.

c. A short-cut method with taxes is to calculate the multiplier, as you would withan increase in government expenditures and deduct one from it.

4. The balanced budget multiplier is always one.

a. Occurs when the amount of government expenditures goes up by the sameamount that a lump sum tax is increased.

b. That is because only the initial expenditure increases GDP and the remainingmultiplier effect is offset by taxation.

27

5. Tax structure refers to the burden of the tax:

a. progressive is where the effective tax rate increases with ability to pay,

b. regressive is where the effective tax rate increases as ability to paydecreases,

c. proportional is where a fixed proportion of ability to pay is taken in taxes.

6. Automatic stabilizers help to smooth business cycles without further legislativeaction:

a. Progressive income taxes,

b. Unemployment compensation,

c. Government entitlement programs

7. Fiscal Lag - there are numerous lags involved with the implementation of fiscalpolicy. It is not uncommon for fiscal policy to take 2 or 3 years to have a noticeableeffect, after Congress begins to enact fiscal measures.

a. Recognition lag - how long to start to react.

b. Administrative lag - how long to have legislation enacted & implemented.

c. Operational lag - how long it takes to have effects in economy.

8. Politics and Fiscal Policy.

a. Public choice economists claim that politicians maximize their own utility bylegislative action.

b. Log-rolling and negotiations results in many bills that impose costs.

9. Government deficits and crowding - out. It is alleged that private spending isdisplaced when the government borrows to finance spending:

a. Ricardian Equivalence - deficit financing same effect on GDP as increasedtax.

28

10. Open economy problems. Because there is a foreign sector that impacts GDPthere are potential problems for fiscal policy arising from foreign sources.

a. increased interest - net export effect

1. An increase in the interest rate domestically will attract foreign capital,but this increases the demand for dollars which increases their value andthereby reduces exports, hence GDP.

b. foreign shocks - in addition to currency exchange rates.

1. Oil crises increased costs of production in the U.S.

29

8. Money and Banking

Lecture Notes

1. Functions of Money - there are three functions of money:

a. Medium of exchange - accepted as "legal tender" or something of general andspecified value.

1. Use avoids reliance on barter.

2. Barter requires a coincidence of wants and severely complicates amarket economy.

b. Measure of value - permits value to be stated in terms of a standard anduniversally understood standard.

c. Store of value - can be saved with little risk, chance of spoilage and virtuallyno cost and later exchanged for commodities without these positive storagecharacteristics.

2. Supply of money

a. There are numerous definitions of money M1 through M3 most commonlyused.

1. M1 is currency + checkable deposits

2. M2 is M1 + noncheckable savings account, time deposits of less$100,000, Money Market Deposit Accounts, and Money Market MutualFunds.

3. M3 is M2 + large time deposit (larger than $100,000).

3. Near Money - are items that fulfill portions of the requirements of the functions ofmoney.

30

a. Credit cards - fulfill exchange function, but are not a measure of value and ifthere is a credit line, can be used to store value.

b. Other forms of near money:

1. Precious metals - store of value, but not easily exchanged

2. Stocks and Bonds - earnings instruments, but can be used as store ofvalue.

c. Implications for near money - stability, spending habits & policy

4. What gives money value

a. No more gold standard

1. Nixon eliminated gold standard

b. The Value of money depends upon:

1. acceptability for payment,

2. because the government claims it is legal tender, and

3. its relative scarcity.

c. Exchange rates

5. Value of dollar = D = 1/P

6. Demand for Money - three components of money demand:

a. Transactions demand

b. Asset demand

c. Total demand

31

The money supply curve is vertical because the supply of money is exogenouslydetermined by the Federal Reserve. The money demand curve slopes downward andto the right. The intersection of the money demand and money supply curvesrepresents equilibrium in the money market and determines the interest rate (price ofmoney).

7. Money market

a. With bonds that pay a specified interest payment per quarter then:

1. interest rate and value of bond inversely related

8. U.S. Financial System

a. FDIC - Federal Deposit Insurance Corporation - guarantees bank deposits.

b. Federal Reserve System - is comprised of member banks. The Board ofGovernors and Chairman are nominated by the President of United States. Thestructure of the system is:

1. Board of Governors

2. Open Market Committee

3. Federal Advisory Council

4. 12 regions, Atlanta, Boston, Chicago, Cleveland, Dallas, Kansas City,Minneapolis, New York, Philadelphia, Richmond, San Francisco, St. Louis

32

c. Functions

1. Set reserves requirements,

2. Check clearing services,

3. Fiscal agents for U.S. government,

4. Supervision of banks,

5. Control money supply through FOMC,

9. Moral hazard - insuring reduces insured's incentive to assure risk does not happen

33

9. Multiple Expansion of Money

Lecture Notes

1. Balance sheet (T accounts -- assets = liabilities + net worth)

a. is nothing more than a convenient reporting tool.

2. Fractional Reserve Requirements

a. Goldsmiths used to issue paper money receipts, backed by stocks of gold. The stocks of gold acted as a reserve to assure payment if the paper claims werepresented for payment.

1. Genghis Khan first issued paper money in the thirteenth century - itwas backed by nothing except the Khan's authority.

b. The U.S. did not have a central banking system from the 1820 through 1914. In the early part of this century several financial panics pointed to the need for acentral banking and financial regulation.

1. States and private companies used to issue paper money.

2. In the early days of U.S. history Spanish silver coins were widelycirculated in the U.S.

3. The first U.S. paper money was issued during the Civil War (The Greenback Act), which included fractional currency (paper dimes &nickels!).

c. Today, the Federal Reserve requires banks to keep a portion of its depositsas reserves.

1. purposes to keep banks solvent & prevent financial panics

3. RRR (Required Reserve Ratio) and multiple expansion of money supply through Taccounts

a. How reserves are kept

34

1. Loans from Fed - discount rate at which Fed loans reserves tomembers

2. Vault cash

3. Deposits with Fed

4. Fed funds rate - the rate at which members loan each other reserves

b. RRR = Required reserve/demand deposit liabilities

c. actual, required, and excess reserves

4. Money created through deposit/loan redepositing

a. Money is created by a bank receiving a deposit, and then loaning that non-reserve portion of the deposit, which is deposited and loans made against thosedeposits.

1. If the required reserve ratio is .10, then a bank must retain 10% of eachdeposit as a reserve and can loan 90% of the deposit; the multipleexpansion of money, assuming a required reserve ratio of .10, istherefore:

Deposit Loan

$10.00 9.00 9.00 8.10 8.10 7.29 . . . . _______ _______$ 100.00 $90.00

Total new money is the initial deposit of $10 + $90 of multiple expansion for atotal of $100.00 in new money.

5. Money multiplier Mm = 1/RRR

a. Is the short-hand method of calculating the entries in banks' T accounts andshows how much an initial injection of money into the system generates in totalmoney supply.

35

b. With a required reserve ratio of .05 the money multiplier is 20 & with arequired reserve ratio of .20 the money multiplier is 5.

c. the Federal Reserve needs to inject only that fraction of money that time themultiplier will increase the money supply to the desired levels.

36

10. Federal Reserve and Monetary Policy

Lecture Notes

1. Monetary policy, defined and objectives

a. Monetary policy is carried out by the Federal Reserve System and is focusedon controlling the money supply.

b. The fundamental objective of monetary policy is to assist the economy inattaining a full employment, non-inflationary equilibrium.

2. Tools of Monetary Policy

a. Open Market Operations is the selling and buying of U.S. treasury obligationsin the open market.

b. Expansionary monetary policy involves the buying of bonds.

1. The Fed buying bonds, it puts money into the hands of those who hadheld bonds.

c. Contractionary monetary policy involves the selling of bonds.

1. The Fed sells bonds it removes money from the system and replaces itwith bonds.

3. Required Reserve Ratio - the Fed can raise or lower the required reserve ratio.

a. Increasing the required reserve ratio, reduces the money multiplier, hencereduces the amount by which multiple expansions of the money supply canoccur.

1. decreasing the required reserve ratio increases the money multiplier,and permits more multiple expansion of the money supply.

4. The Discount Rate is the rate at which the Fed will loan reserves to member banksfor short periods of time.

37

5. Velocity of Money - is how often the money supply turns-over.

a. The quantity theory of money is: MV = PQ

1. This equation has velocity (V) which is nearly constant and output (Q)which grows slowly, so what happens to the money supply (M) should bedirectly reflect in the price level (P).

6. Target Dilemma in Monetary Policy

a. Interest rates and the business cycle may present a dilemma. Expansionarymonetary policy may result in higher interest rates which dampen expansionarypolicies.

b. Fiscal and monetary policies may also be contradictory.

7. Easy Money - lowering interest rates, expanding money supply.

a. mitigate recession and stimulate growth.

8. Tight Money - increasing interest rates, contracting money supply.

a. mitigate inflation and slow growth.

9. Monetary rules - Milton Friedman

a. Discretionary monetary policy often misses targets in U.S. monetary history

b. Suspicion of Fed and FOMC – perhaps overblown

38

11. Interest Rates and Output: Hick’s IS/LM Model

Lecture Notes

1. IS Curve

The IS curve shows the level of real GDP for each level of real interest rate. Thederivation of the IS curve is a rather straightforward matter, observe the following diagram.

INCOME/EXPENDITURE Expenditure or Aggregate Demand

GDP

AUTONOMOUS SPENDINGInterest Rate Interest Rate

IS

Autonomous Spending GDP

39

The Income-Expenditure diagram determines for each level of aggregatedemand the associated level of GDP.

The intercept of the IS Curve is the level of GDP that would obtain at a zero realinterest rate. The slope of IS Curve is the multiplier (1/1 - MPC) times marginalpropensity to Invest (resulting from a change in real interest rates) and the marginalpropensity to export (resulting from a change in real interest rates).

The IS curve can be shifted by fiscal policy

Changes in Co or Io will also move the IS curve.

Anytime the economy moves away from the IS curve there are forces within thesystem which push the economy back onto the IS curve. Observe the followingdiagram:

Real Interest Rate

¹ ! A

B ! ¸ IS Curve Real GDP

If the economy is at point A there is a relatively high level of real interest rates whichresults in planned expenditure being less than production, hence inventories areaccumulating, and production will fall, hence pushing the economy toward the IS curve. On the other hand, at point B the relatively low real interest rate results in plannedexpenditure being greater than production, hence inventories are being sold into themarket place and product will rise to bring us back to the IS curve.

40

2. LM Curve

The LM Curve is derived in a fashion similar to that of the IS curve. Consider thefollowing diagram:

Interest Rate Money S Interest Rate

LM Curve D 1

D 2

D 3 GDP (Y)

Real Qty of Money High Y Moderate Y Low Y

As can be readily observed from this diagram the LM curve is the schedule ofinterest rates associated with levels of income (GDP). The interest rate, in this case,being determined in the money market.

With a fixed money supply each level of demand for money creates a differentinterest rate.

If the money market is to remain in equilibrium, then as incomes rise so too, thenmust the interest rate, if the supply of money is fixed.

The shifting of the LM curve is obtained through inflation or monetary policy.

41

3. Equilibrium in IS-LM

The Intersection of the IS and LM curve results in there being an equilibrium inthe macroeconomy.

Interest Rate

LM

r

IS1 ISe IS2

GDP Y

Where the IS and LM curves intersect is where there is an equilibrium in this

economy. With this tool in hand the affects of the interest rate on GDP can be directlyobserved. As the IS curve shifts to the right along the LM curve notice that there is anincrease in GDP, but with a higher interest rate (IS1) and just the opposite occurs as theIS curve shifts back towards the origin (IS2). From above it is clear the sorts of thingsthat shift the IS curve, fiscal policy or changes in Co or Io.

4. Expansionary and Contractionary Monetary Policies

The LM curve, too, can be moved about by changes in policy. If the moneysupply is increased or decreased that will have obvious implications for the LM curveand hence the interest rate and equilibrium level of GDP. Consider the followingdiagram:

42

Interest Rate LM2 LMe

LM1

r

IS

GDP Y

As the Fed engages in easy monetary policies the LM Curve shifts to the right, and theinterest rate falls, as GDP increases.

5. Foreign Shocks

The U.S. economy is not a closed economy, and the IS-LM Model permits us toexamine foreign shocks to the U.S. economy. There are three of these foreign shocksworthy of examination here; these are (1) increase in demand for our exports, (2)increases in foreign interest rates, and (3) currency speculators expectations of anincrease in the exchange rate of our currency with respect to some foreign currency. Each of these foreign shocks results in an outward expansion of the IS Curve asportrayed in the following diagram:

43

Interest Rate

LM

r2 r1

IS1

ISo

GDP Y1 Y2

44

12. Economic Stability and Policy

Lecture Notes

1. Inflation, Unemployment and Economic Policy

a. The misery index is the inflation rate plus the unemployment rate.

2. The Phillips Curve is a statistical relation between unemployment and inflationnamed for A. W. Phillips who examined the relation in the United Kingdom andpublished his results in 1958. (Actually Irving Fisher had done earlier work on thesubject in 1926).

a. Short run trade-off

Often used to support activitist role for government, however, the short-run trade-off view of the Phillips curve demonstrates that there is a cruel choice betweenincreased inflation or increased unemployment, but low inflation and unemploymenttogether are not possible.

b. Long run Phillips Curve is alleged to be vertical at the natural rate of unemployment.

45

This long-run view of the Phillips Curve is also called the Natural RateHypothesis. It is based on the idea that people constantly adapt to current economicconditions and that their expectations are subject to "adaptive" revisions almostconstantly. If this is the case, then business and consumers cannot be fooled intothinking that there is a reason for unemployment to cure inflation or vice versa.

c. Possible positive sloping has hypothesized by Milton Friedman. Friedmanwas of the opinion that the may be a transitional Phillips curve while people adaptboth their expectations and institutions to new economic realities. The positivelysloped Phillips curve is show in the following picture:

The positively sloped transitional Phillips Curve is consistent with theobservations of the early 1980s when both high rates of unemployment existed togetherwith high rates of inflation -- a condition called stagflation.

46

d. Cruel choices only exist in the case of the short-run trade-off view of thePhillips Curve. However, there maybe a "Lady and Tiger Dilemma" for policymakers relying on the Phillips Curve to make policy decisions.

3. Rational expectations is a theory that businesses and consumers will be able toaccurately forecast prices (and other relevant economic variables). If the accuracy ofconsumers' and business expectations permit them to behave as though they knowwhat will happen, then it is argued that only a vertical Phillips Curve is possible, as longas political and economic institutions remain stable.

4. Market policies are concerned with correcting specific observed economic woes.

a. Equity policies are designed to assure "a social safety net" at the minimumand at the liberal extreme to redistribute income.

1. The Lorenz Curve and Gini Coefficients are used to measure incomedistribution in economies.

The Lorenz curve maps the distribution of income among across the population. The 45 degree line shows what the distribution of income would be if income wasuniformly distributed across the population. However, the Lorenz curve drops downbelow the 45 degree line showing that poorer people receive less than rich people.

The Gini coefficient is the percentage of the triangle mapped out by the 45degree line, the indicator line from the top of the 45 degree line to the percentage ofincome axis, and the percentage of income axis that is accounted for by the areabetween the Lorenz curve and the 45 degree line. If the Gini coefficient is near zero,

47

income is close to uniformly distributed; if is near 1 then income is mal-distributed.

b. Productivity is also the subject of specific policies. The Investment TaxCredit, WIN program, and various state and federal training programs arefocused increasing productivity.

c. Trade barriers have been reduced through NAFTA and GATT in hopes offostering more U.S. exports.

5. Wage-Price Policies

a. Attempts have been made to directly control inflation through price controls,this seemed to work reasonably well during World War II. Carter tried voluntaryguidelines that failed, and Nixon tried controls that simply were a disaster.

6. Supply Side Economics of the Reagan Administration were based on the theory thatstimulating the economy would prevent deficits as government spending for the militarywas increased. This failed theory was based on something called the Laffer Curve.

a. Laffer Curve (named for Arthur Laffer) is a relation between tax rates and taxreceipts. Laffer's idea was rather simple, he posited that there was optimal taxrate, above which receipt went down and below which receipts went down. TheLaffer curve is shown below:

The Laffer Curve shows that the same tax receipts will be collected at the rateslabelled both "too high" and "too low." What the supply-siders thought was that tax

48

rates were too high and that a reduction in tax rates would permit them to slide downand to the right on the Laffer Curve and collect more revenue. In other words, theythought the tax rate was above the optimal. We got a big tax rate reduction and found,unfortunately, that we were below the optimal and tax revenues fell, while wedramatically increased the budget. In other words, record-breaking deficits and debt.

b. There were other tenets of the supply-side view of the world. Theseeconomists thought there was too much government regulation. After JimmyCarter de-regulated trucking and airlines, there was much rhetoric and littleaction to de-regulate other aspects of American economic life.

7. Unfortunately, the realities of American economic policy is that politics is often mainmotivation for policy.

a. Tax cuts are popular, tax increases are not.

b. Deficits are a natural propensity for politicians unwilling to cut pork from theirown districts and unwilling to increase taxes.

8. Politics - economics provides a scientific approach to understanding, politics is the“art of the possible” what is good economically maybe horrible politically and vice versa

a. Public choice literature

1. Politicians act in self-interest just like the rest of us

49

13. Data: Categories, Sources, Problems and Costs

Lecture Notes

1. Data Types

A. Time series data are metrics for a specific variable observed over time.

B. Cross-sectional data, on the other hand, are data for a specific time period,for a specific variable, but across subjects.

C. The combination of cross-sectional data over time is called panel-data.

D. Discrete data are metrics that have specific and finite number of potentialvalues.

E. Continuous data, on the other hand, are metrics which can take on any valueeither from plus infinity to negative infinity, or within specific limits. .

F. A stock variable is generally a variable in which there is a specific value atsome point.

G. A flow variable is generally something that has some beginning, ending orchanging value over time.

2. Survey data is generally reported by an individual who may or may not have somemotivation to accurately report the data requested in the survey.

A. Validity has two broad dimensions.

1. The response rate from the sample of a population will define howrepresentative that sample is of the population.

2. The second issue involves whether the items on the survey actuallycapture what the researcher intends.

a. This second issue with validity also has three distinctdimensions these are: (1) content validity, (2) criterion-related

50

validity and (3) construct validity.

B. Reliability has to do with assuring the accuracy of responses.

1. Internal checks

2. States of nature data, involve things like market prices, number ofpersons, tons of coal and other such metrics that can be physicallyobserved or measured.

3. Interviews and observational methods are also often utilized to gatherwhat is sometimes referred to as primary data.

3. Seasonal Adjustments – Conceptually, there are a number of economic time seriesdata which have specific, predictable biases created by the time of year.

4. Data Sources

A. Public, Federal, State, International Organizations, and Foreign Countries1. Commerce Department2. Bureau of Labor Statistics3. CIA4. Fed5. OECD6. World Bank, IMF, and UN

B. Private data sources include both proprietary and non-proprietary sources.

1. Market data - S&P Dow Jones etc.2. ACCRA

5. Information is not free, there are costs associated with gathering and analyzing data.

A. Sutcliff and Webber

1. Perceptual Accuracy

2. Dangers in Anecdotal evidence

3. Training, perception, experience, and humility

Sutcliff and Weber argue that the accuracy of data is an expensive proposition

51

and that too often managers have neither the expertise nor the time to fullyanalyze what imperfect information they can gather. What Sutcliff and Webercontend is that intuition is perhaps a better modus operandi, what they call ahumble optimism but based on what knowledge can be gleaned withoutbecoming a victim of the old cliche “paralysis by analysis.

Performance

Perceptual accuracy

52

Sutcliff and Weber studies suggest that organizational change has an inverted Ushaped relation to perceptual accuracy. In other words, over the initial rangesthere are increases in the return to perceptual accuracy for positive organizationalchange up to some optimal point. Beyond that optimal, additional accuracyactually results in negative returns. On the other hand, organization performanceis negatively correlated with perceptual accuracy. The more resources (time, etc.)are devoted to perceptual accuracy of analysis or the underlying data the morelikely performance will decline.

6. Professor James Brian Quinn has observed that the move into “high-tech” business(science based) has forced many firms to examine their strategies and how they dealwith information. In fact, Dr. Quinn believes that these science based industries havebeen forced into a situation where knowledge sharing is of critical importance to thesuccess of industries and firms within those industries. The following quotation stateshis point:

53

14. Economic Indicators

Lecture Notes

1. Cyclical Indicators: Background

A. The Arthur F. Burns and Wesley C. Mitchell at the Bureau of EconomicResearch began to examine economic time series data in the 1930s to determineif there were data which were useful in predicting business cycles.

B. These indicators are described in Bureau of Economic Analysis’ Handbook ofCyclical Indicators. These indicators were selected using six criteria. Thesecriteria are described in John J. McAuley’s Economic Forecasting for Business:Concepts and Applications, to wit:

C. Indicators are considered, in general, to be useful, simple guides to thevariations in the business cycle in the U.S.

2. Leading Indicators.

A. Leading indicators are those variables which precede changes in the businesscycle.

John McAuley,Economic Forecasting for Business: Concepts and Applications

Cyclical indicators are classified on the basis of six criteria: (1) the economicsignificance of the indicator; (2) the statistical adequacy of the data; (3) the timing ofthe series – whether it leads, coincides with, or lags turning points in overall economicactivity; (4) conformity of the series to overall business cycles; a series conformspositively if it rises in expansions and declines in downturns; it conforms inversely if itdeclines in expansions and rises in downturns: a high conformity score indicates theseries consistently followed the same pattern; (5) the smoothness of the series’movement over time; and (6) timeliness of the frequency of release – monthly orquarterly – and the lag in release following the activity measured.

54

______________________________________________________________________

Table 1: Leading Economic Indicator Index

Variable Weight in Index

1. Average workweek of production workers, manufacturing 1.014 2. Average weekly initial claims, State Unemployment Insurance

(Inverted) 1.041 3. Vendor performance (% change, first difference) 1.081 4. Change in credit outstanding .959 5. Percent change in sensitive prices, smoothed .892 6. Contracts and orders, plant and equipment (in dollars) .946 7. Index of net business formation .973 8. Index of stock price (S&P 500) 1.149 9. M-2 Money Supply .93210. New Orders, consumer goods and materials (in dollars) .97311. Building permits, private housing 1.05412. Change in inventories on hand and on order (in dollars, smoothed) .985

Source: Business Conditions Digest, 1983______________________________________________________________________

B. Efficient Market Hypothesis – Eugene Fama

3. Coincident Indicators

A. These indicators are the ones which happen as we are in a particular phase ofthe business cycle, in other words, they happen concurrently with the business cycle. Statistically their variations follow the leading indicators and occur prior to the laggingindicators. ______________________________________________________________________Table 2: Index of Coincident Indicators

Variable Weight

1. Employees on nonagricultural payrolls 1.0642. Index of industrial production, total 1.0283. Personal Income, less transfer payments 1.0034. Manufacturing and trade sales .905

Source: Business Conditions Digest, 1983.

______________________________________________________________________

55

3. Lagging Indicators

A. Lagging indicators are those which trail the phase in the business cycle thatthe economy just experienced.

______________________________________________________________________

Table 3: Lagging Indicators

Variables Weights

1. Average duration of unemployment 1.0982. Labor cost per unit of output, manufacturing .8683. Ratio constant dollar inventories to sales, and manufacturing and sales .8944. Commercial and industrial loans outstanding 1.0095. Average prime rate charged by banks 1.1236. Ratio, consumer installment debt to personal income 1.009

Source: Business Conditions Digest, 1983.______________________________________________________________________

4. Other Indicators

A.. The University of Michigan publishes a “Consumer Confidence” index and theConference Board also does something similar for producers.

B. Real Estate firms have a lobby and research group that does surveys of newhome construction, and sales of existing housing which plays on essentially thesame sort of economic activities captured by “Building permits, private housing”variable found in the Index of Leading Indicators.

C. The Federal Reserve is not to be outdone in this forecasting business. ThePhiladelphia Federal Reserve Bank does an extensive survey of businessconditions and plans in the Philadelphia Federal Reserve District and publishestheir results monthly. This report is commonly referred to as “The Beige Book”and it eagerly anticipated as providing insights into what to expect in the nearfuture in the eastern part of the United States.

56

D. There are also numerous proprietary forecasts and indices constructed forevery purpose imaginable. Stock price data have been subjected to nearly everytorture imaginable in an attempt to gain some sort of trading advantage. BollingerBands, Dow Theory, and even Sun Spots have been used in an attempt toforecast what will happen in the immediate short run with financial markets.

E. Investment Services

1. Zacks2. Standard and Poors3. Morningstar4. Others

57

15. Guide to Analytical and Forecasting Techniques

Lecture Notes

1. Time Series Methods

These are: (1) trend line, (2) moving averages, (3) exponential smoothing, (4)decomposition, and (5) Autoregressive Integrative Moving Average (ARIMA).

Each of these methods is readily accessible, and easily mastered (with thepossible exception of certain variants of ARIMA). The first four of these methodsinvolves little more than junior high school arithmetic and can be employed for a varietyof useful things.

A. Trend Line

1. This method is also commonly called a naive method, it that is simplyfitting a representative line to the data observed. Black thread method.

X ! ! ! ! ! !

! ! ! ! !

Y

58

B. Moving Averages

1. If we have three data points 6, 7, and 8. The moving average iscalculated as:

0 = 6+7+8 ÷ 3 = 7

2. The moving average of this series is 7. However, it is clear that if theseries continues that the next number will be 9. Therefore, a weightedmoving average may result in a more satisfactory result. If the we use amethod that weights the last observation in the series more than the firstobservation we will get closer to the next number in the series, which wepresume is 9. Therefore, the following weighted average is used:

0 = (.5) 6+7+ (1.5)8 ÷ 3 = 7.333

3. This weighting scheme still undershoots the forecast. Therefore we tryusing something in which the last number is weighted more heavily.

0 = (.5) 6+7+(2)8 +1 ÷ 3 = 9

C. Exponential Smoothing

1. Exponential smoothing is an effort to take some of the guess work out,or at least formalize the guess work. The equation for the forecast modelis:

F = (%) X2 + (1 - %) S1

2. Where alpha is the smoothing statistic, X is the last observation in thedata set, and S is the first predicted value from the data set. For example,if we use the following series:

25, 23, 22, 21, 24

3. To obtain the initial value of S or S1 we add 25 and 23 and divided thesum by 2 to obtain 24. The X2 in this case is our last observation or 25.

All that is left is the selection of %. Normally, an analyst will select .3 with aseries which does not a linear trend upwards or downwards. This gives a

59

systematic method of determining the appropriate value of %. Make

predictions using % beginning at .1 and work your way to .9 and determine

which % gives the forecast with the least error term over a range of the

latest available data, and use that % until such time as error begin toincrease. To complete the example, lets plug and chug to a get a forecast.

F = (.3) 25 + (1 - .3) 24 =

7.5 + 16.8 = 24.3

D. Decomposition

1. The process begins with calculating a moving average for the dataseries. For quarterly data the moving average is:

MA = (Xt-2 + Xt-1 + Xt + Xt+1) / 4

We will apply this method to the following data:

Year 1 XFirst Quarter 10Second Quarter 12Third Quarter 13Fourth Quarter 11

Year 2 XFirst Quarter 11Second Quarter 13Third Quarter 14Fourth Quarter 11

Year 3 X First Quarter 10

Second Quarter 13Third Quarter 13Fourth Quarter 11

Applying our moving average method:

60

MA3 = (10+12+13+11)/4 = 11.50MA4 = (12+13+11+11)/4 = 11.75MA5 = (13+11+11+13)/4 = 12.00MA6 = (11+13+14+11)/4 = 12.25

The data are now presented for year as a moving average of the quarterscentered on the quarter represented by the number following the MA. For theyear, the data for this series centered on the third quarter is 11.50, centered onthe fourth quarter it is 11.75 and for the first quarter of the previous year it is12.00.

Now we can calculate an index for seasonal adjustment, in a rather simple andstraightforward fashion. The index is calculated using the following equation:

SI t = Y t / MA t

where SI t is the seasonal index, Y t is the actual observation for that quarter, and MA t is the moving average centered on that quarter, from the method shownabove. A seasonally adjusted index can then be created which allows theseasonality of the data to be removed.

Using the three years worth of data above we can construct a seasonality indexwhich allows us to remove the seasonal effects from the data. The average foreach quarter is calculated:

Y1 = (10+11+10)/3 = 10.33Y2 = (12+13+13)/3 = 12.67Y3 = (13+13+13)/3 = 13.00Y4 = (11+11+11)/3 = 11.00

Therefore plugging in the values for Y t and MA t the equation above allows us toconstruct an index for seasonality from this data:

SI 1 = 10.33/11.50 = .8983SI 2 = 12.67/11.75 = 1.0783SI 3 = 13.00/12.00 = 1.0833SI 4 = 11.00/12.25 = .8988

To seasonally adjust the raw data is then a simple matter of divided the raw databy its appropriate seasonal index.

61

Year 1 X Seasonally AdjustedFirst Quarter 10/0.8983 = 11.1321Second Quarter 12/1.0783 = 11.1286Third Quarter 13/1.0833 = 12.0004Fourth Quarter 11/0.8988 = 12.2385

Year 2First Quarter 11/0.8983 = 12.2453Second Quarter 13/1.0783 = 12.0560Third Quarter 14/1.0833 = 12.9235Fourth Quarter 11/0.8988 = 12.2385

Year 3 First Quarter 10/0.8983 = 11.1321

Second Quarter 13/1.0783 = 12.0560Third Quarter 13/1.0833 = 12.0004Fourth Quarter 11/0.8988 = 12.2385

E. ARIMA – Autoregressive Integrated Moving Average.

1. This method uses past values of a time series to estimate future valuesof the same time series. According to John McAuley, The Nobel Prizewinning economist Clive Granger is alleged to have noted: “it has beensaid to be difficult that it should never be tried for the first time.” However,it is useful tool in many applications, and therefore at least something morethan a passing mention is required here.

There are basically three parts of an ARIMA model, as are described in McAuley’stext these three components are:1

“(1) The series can be described by an autoregressive (AR) component,such that the most recent value (X1) can be estimated by a weightedaverage of past values in the series going back p periods:

X1 = Θ (X t-1) + . . . + Θ p (X t-p) + δ + ε

where Θ p = the weights of the lagged values δ = a constant term related to the series mean, and

1 John J. McAuly, Economic Forecasting for Business: Concepts andApplications, Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1985.

62

ε = the stochastic term

(2) The series can have a moving average (MA) component based on a qperiod moving average of the stochastic errors:

X1 = 0 + (εt - Θ1) εt-1 - . . . - Θq + εt-u

where 0 = the mean of the seriesΘq = the weights of the lagged error terms and

ε = the stochastic errors.

(3) Most economic series are not stationary (can be transformed throughdifferencing using a stationary process), but instead display a trend orcyclical movements over time. Most series, however, can be madestationary by differencing. For instance, first differencing is the period-to-period change in the series:

ΔX t = X t - X t-1 “

2. Correlative Methods

A. Correlative methods rely on the concept of Ordinary Least Squares, or O.L.S.,which is also sometimes referred to as regression analysis. O.L.S. refers to the fact thatthe representative function of the data is determined by minimizing the sum of thesquared errors of each of the data points from that line.

B. Graphical depiction of Ordinary Least Squares:

63

X ! ! ! ! ! ! ! ! !

Y

1. The lines from each data point to the representative line are the errorterms. Squaring the error terms eliminates the signs, and the regressionanalysis minimizes these squared errors in order to determine what themost representative line is for there data points.

C. The general form of the representative equation for a bi-variate regression is:

Y = α + βX + ε

where Y is the dependent variable, and X is the independent variable; α is the interceptterm, β is the slope coefficient and ε random error term.

D. Correlation

1. To be able to determine what a representative line is, requires somefurther analysis. The model is said to be a good fit if the R squared isrelatively high. There R squared can vary between 0 (no correlation) to 1(perfect correlation). Typically an F statistic is generated by theregression program for the R squared which can be checked against atable of critical values to determine if the R squared is significant differentthan zero. If the F statistic is significant, then the R squared is significantlymore than zero.

64

The simple r is calculated as:

_ _ _ _ r = Σ (X - X) (Y - Y) / ([ Σ (X - X)2 ][ Σ (Y - Y)2 ])-1

and the simple r is squared to obtain the R squared.

E. Significance of Coefficients

1. There are also statistical tests to determine the significance of theintercept coefficient and the slope coefficient, these are simple t-statistics. The test of significance is whether the coefficient is zero. If the t-statisticis significant for the number of degrees of freedom, then the coefficient isother than zero.

In general, the analyst is looking for coefficients which are significant and of asign that makes theoretical sense. If the slope coefficient for a particular variable iszero, it suggests that there is no statistical association between that independentvariable and the dependent variable.

2. There is also a problem in which additional data being added to theseries or another variable will sometimes cause the sign of a coefficientalready in the model to swap signs. This is a symptom of problem with thedata called heteroskedasticity. This problem is a result data notconforming to the underlying assumptions of ordinary least squares that isthat the errors or residuals do not have a common variance. Youbasically have two ways in which to deal with this problem, select adifferent analytical technique or try transforming the data. If you transformthe data, log transformations or deflating the data into some normalizedseries will sometime eliminate the problem and give unbiased, efficientestimates.

3. With time series data there also arises a problem called serialcorrelation. That is when the error terms are not randomized as assumed. In other words, the error terms are correlated with one another. There istest for this problem, called the Durbin-Watson (d) statistic. When dealingwith time series data one should, as a matter of routine, have the programcalculate a DW (d) and check with a table of critical values so as toeliminate any problems in inference associated with correlated errorterms. If this problem arises, often using the first difference ofobservations in the data will eliminate the problem. However, often a lowDW (d) statistic is an indication that the regression equation may be mis-

65

specified which results in further difficulties, both statistically andconceptually.

F. Other Issues

1. accessibility; you can do regressions on Excel Spreadsheets.

2. dummy variable.

a. dummy variable trap - which results in a zero coefficient andother unpleasantries.

3. Instrumental variables are also sometimes used. These variables areconstructed for the purpose of substituting for something we could notdirectly measure or quantify.

3. Other Methods

A. Continuous Data

1. Analysis of variance is typically used to isolate and evaluate thesources of variation within independent treatment variable to determinehow these variable interact. An analysis of variance is generally part ofthe output of any regression package and provides additional informationconcerning the behavior of the data subjected to the regression.

2. Multivariate analysis of variance which provides information on thebehavior of the variables in the data set controlling for the effects of theother variables in the analysis.

3. Analysis of covariance which goes a set further, as the namesuggests, and provides an analysis of how the multiple variable interactstatistically.

4. Factor analysis relies on correlative techniques to determine whatvariables are statistically related to one another. A factor may haveseveral different dimensions. For example, men and women. In general,men tend to be heavier, taller, and live shorter lives. These variableswould form a factor of characteristics of human beings.

5. Cluster analysis is analogous in that rather than a particular theoreticalrequirement that results in factors, there may be simple tendencies, such

66

as behavioral variables.

6. Canonical correlation analysis is a method where factors areidentified and the importance of each variable in a factor is also identified. These elements are calculated for a collection of variables which areanalogous to a dependent variable in a regression analysis and acollection of variables which are analogous to independent variables in aregression. This gives the research the ability to deal with issues whichmay be multi-dimensional and do not lend themselves well to a singledependent variable. For example, strike activity in the United States hasseveral different dimensions, and data measuring those dimensions. Thecompeting theories explaining what strikes happen and why they continuecan then be simultaneously tested against the array of strikemeasurements, lending to more profound insights than if just one variablewas used.

7. Probit analysis uses categorical dependent variables (discrete) andexplains the variation in that variable with a mixture of continuous anddiscrete variables. The output of these computer programs yields outputwhich permits inference much the same as is done with regressionanalysis.

a. If there is a stochastic dependent logit analysis is an alternativeto probit.

B. Discrete Data – There are several nonparametric methods devised to dealspecifically with discrete data.

1. The Mann-Whitney U test provides a method whereby twopopulations distributions can be compared.

2. The Kruskal-Wallis test is an extension of the Mann-Whitney U test, inthat this method provides an opportunity to compare several populations.

3. These tests can be used with the distributions of the data do not fit theunderlying assumptions necessary to apply an F test (random samplesdrawn from normally distributed populations with equal variances).

4. There are also methods of dealing with ordinally ranked data. Forexample, the Wilcoxon Signed-Rank test can be utilized to analyzepaired-differences in experimental data. This is particularly useful ineducational studies or in some types of behavioral studies where the F

67

statistic assumptions are not fulfilled or the two test above do not apply(not looking at population differences).

5. Spearman Rank Correlations are useful in determining how closelyassociated individual observations may be to specific samples orpopulations. For example, item analysis in multiple choice exams can beanalyzed using this procedure.

a. Students who perform well on the entire exam can have theiroverall score assessed with respect to specific questions. In thisexample, if the top students all performed well on question 1, theirrank correlation would be near 1, however, if they all performedwell, but on question 2 they did relatively poorly, that correlationwould be closer to zero, say .35. If, on the other hand, none of thestudents who performed well got question 2 correct, then we wouldobserve a zero correlation for that question. This method is usefulin marketing research and in behavioral, as well as collegeprofessors with too much time on their hands.

68

16. Data: In the Beginning

Lecture Notes

1. Becoming Familiar with the Data

A. Calculating descriptive statistics to determine what the data “look” like.

1. Mode most frequent observation

2. Median - midpoint in the

3. Mean:

0 = / nfmii n

k

i

4. Variance:

σ2 = 2 - [( )2 / n] / n - 1fmii n

k

i

fmi

i n

k

i

5. Standard Deviation:

σ = [ 2 - [( )2 / n] / n - 1] -1fmii n

k

i

fmi

i n

k

i

where: mi = midpoint of data series I fi = frequency of observation in series

A. Skewed data

69

1. 3 dimensional, skewed left

2. Skewed right, 2 dimensional

4. Normal

70

Distribution

3.There is a statistical principle called the law of large numbers. Forpurposes of most statistical analyses, if the data are random sample froma population with the same variance and continuous, then if you havemore than 30 observations it is assumed that those number ofobservations or more will approximate a normal distribution.

.

2. Correlated Independent Variables

A. Multicolinearity, serial correlation, correlated error or residuals

B. Calculation of correlation matrix

71

____________________________________________________________________ Unemployment Inventories Coincident S & P Rate Mfg. Copper Prices 500

_____________________________________________________________________

Unemployment rate 1.000

Inventories Mfg. -0.881 1.000

Copper Prices (coincident) -0.793 0.535 1.000

S&P 500 -0.679 0.790 0.243 1.000

_____________________________________________________________________

C. Forecasting versus Structural Models

1. Structural equations versus forecasting models

2. Rigor and theory

72

17. Epilog: Management and Business Conditions Analysis

Lecture Notes

1. Information and Managerial Empowerment

A. Information should, first and foremost, be one means by which peoplein an organization are empowered to contribute to success of theorganization.

1. Empowerment in a managerial sense requires that those closestto the events should be making the calls, within broad policyconstraints.

2. Information (its acquisition and dissemination) is a supportfunction.

3. Policy parameters must be set for the gathering, analysis and useof data.

a. Open access

b. Discretion to exercise delegated authority, and informationis one tool to make sure that discretion is exercised wisely.

2. Policy and Strategy

A. The purpose of a strategic plan is to provide not only guidance formanagerial actions, hence policy, but also to provide a set of goals bywhich the directed actions of the enterprise can be assessed.

B. The culture that is developed in an organization concerning howdecision-making occurs is often one of the most important determinants ofthe organization’s success or failure.

C. Effective organizations empower those with line authority throughoutthe organization.

73

3. Qualitative Issues

A. Organizational Culture

B. Regulatory Environment

C. Other issues

4. Forecasting and Planning Interrelations – Forecasting and planning cannot beseparated, as a practical matter. Strategies require a vision of the future.

A. Flexibility to react to vision errors

B. Contingency planning

C. Measureable outcomes

D. Scientific approach to creation of plan and goals.

5. Risk and Uncertainty

A. Risk is simply the down-side of opportunity. Risk is the reason entrepreneursare rewarded. Risk is also the reason enterprises fail. The trick is to take thoserisks you understand and can appropriately manage.

B. Stochastics are what breeds uncertainty, which, in turn, breeds ambiguity. When we don’t know what to expect, that is sometimes the most anxietyproducing situation.

6. Go forth and prosper!!!

74

E550

Quizzes for Intersession2006

August 7 - August 20, 2006

E550, Business Conditions Analysis Name_____________________________________First Quiz, August 8 Chapters 1,2,3Dr. David A. Dilts

True/False (1 point each)

___1. Macroeconomics is concerned with the aggregate performance of the entire economicsystem.

___2. Inductive logic involve formulating and testing hypotheses.

___3. Positive economics is concerned with what is, and normative economics is concernedwith what should be.

___4. Correlation is the causal relation between two variables.

___5. Gross Domestic Product is the total value of all goods and services produced within theborders of the United States.

___6. The difference between Gross Domestic Product and Net Domestic Product is IndirectBusiness Taxes.

___7. The GDP is overstated because of the relatively large amount of economic activity thatoccurs in the underground economy.

___8. Non-economists are no less or no more biased about economics than they are aboutphysics or chemistry.

___9. Assumptions are used to simplify the real world so that it may be rigorously analyzed.

___10. Ceteris Paribus means “consumer beware.”

(Multiple Choice, each worth 2 points)

__1. People who are unemployed due to a change in technology that results in a decline in theirindustry fit into which category of unemployment?

A. FrictionalB. StructuralC. CyclicalD. Natural

Quiz 1, p. 2

___2. An increase in the price of natural gas will result in:

A. Demand-pull inflationB. Cost-Push inflationC. Pure inflationD. None of the above

___3. The aggregate demand curve is most likely to shift to the right (increase) when there is adecrease in:

A. The overall price levelB. The personal income tax ratesC. The average wage received by workersD. Consumer and business confidence in the economy

___4. An increase in the real wage will:

A. Increase aggregate demandB. Decrease aggregate supplyC. Cause cost-push inflationD. All of the above

___5. Which of the following are not determinants of aggregate supply?

A. Changes in input pricesB. Changes in input productivityC. Changes in the institutional environmentD. All of the above are determinants of aggregate supply

E550, Business Conditions Analysis NAME______________________________

Second Quiz, August 10, Chapters 4,5,6,7Dr. David A. DiltsTrue/False (1 point each)

___1. The reasons the aggregate demand curve slopes downward is (1)real balance effect, (2)interest rate effect, and (3) foreign purchase effect.

___2. The rachet effect results from the aggregate supply curve shifting upwards – which iswhat is often called price rigidity.

___3. Say’s Law is an accounting identity that is associated with aggregate demand being equalto aggregate supply in macroeconomic equilibrium.

___4. Marginal propensity to consume is the amount of consumption expenditures a personmakes on average.

___5. MPC+MPS = 0

___6. The Simple multiplier is 1/MPS.

Multiple Choice (2 points each) ___1. Which of the following is not a determinate of aggregate demand?

A. Personal taxesB. Government spendingC. Exchange ratesD. All of the above are

___2. The non-income determinates of consumption and savings are:

A. WealthB. Consumer debtsC. PricesD. All of the above are

___3. A recessionary gap is:

A. The amount by which C+I+G exceeds the 45 degree line at or above the full employmentlevel of outputB. The amount by which C+I+G is below the 45 degree line at or above the full employmentlevel of outputC. The amount by which the full employment level of output exceeds the current level of outputD. None of the above

Quiz 2, p. 2

Short Answer (points as indicated)

With a marginal propensity to consume of .4 the simple multiplier effect is ____ (2 points)

Okun’s Law states that if unemployment is 8% we have lost what percent of potential GDP____(3 points)

Briefly explain the paradox of thrift (3 points)_________________________________________

______________________________________________________________________________

E550, Business Conditions Analysis NAME______________________________Third Quiz, August 15, Chapters 8,9,10, 11,12

Dr. David A. Dilts

Short Answer (points as indicated)

1. The relationship between unemployment and inflation is shown by what model____________

_____________________________________________________________________(2 points)

2. An optimal tax rate and the revenues available at each rate are shownby:___________________________________________________________________(2 points)

3. (2 points) The way economist measure income distribution is with a:___________ curve and a______coefficient .

4. To close a recessionary gap of $100, with a full employment level of GDP of $1000 and amarginal propensity to consume of .8 requires how much in:

addition government expenditures?__________ (2points); tax cuts?______________ (2 points)

5. (2 points) What is Gresham’s Law?______________________________________________

True / False (1 point each)

___1. For barter to work requires a coincidence of wants.

___2. The U.S. dollar is backed by gold on deposit at Fort Knox, Kentucky.

___3. If the Fed wishes to decrease the money supply it can buy bonds.

___4. With a fractional required reserve ratio of .05, the money multiplier is 20.

___5. The Federal Funds Rate is the overnight rate of interest charged by the Fed to loanmember banks reserves.

___6. Tight monetary policy is associated with the Fed wishing to control inflation.

___7. Easy monetary policy is associated with the Fed wishing to bring the economy out ofrecession.

___8. The largest proportion of the money supply is currency.

E550, Business Conditions Analysis NAME______________________________Fourth Quiz, August 16, Chapters 13, 14

Dr. David A. Dilts

Short Answers (points as indicated)

1. Identify the criteria on which cyclical indicators are classified_____________________________________________________________________________________________(3 points)

2. (2 points) What is the largest proportion of GDP? _________________________

3. (2 points) What are NET exports ______________________ and are they positive or negativein the national income accounts? _________

4. (3 points) List the components of the index of coincident indicators_____________________

______________________________________________________________________________

True/ False (1 point each)

___1. The Laspeyres index uses a constant “market basket” whereas the Paasche index useschanging weights to always have a current “market basket.”

___2. Seasonally adjusted unemployment differs from not seasonally adjusted in that theentrants into the market in the summer and at Christmas time are washed out of the data.

___3. Establishment data for unemployment counts workers twice who are employed in twoplace, whereas Household data does not.

___4. Consumption expenditures are divided up among, consumer durables, consumernondurables, and services.

___5. Index of stock prices is a leading economic indicator.

___6. Index of industrial production is a coincident economic indicator.

___7. Average duration of unemployment is a lagging economic indicator.

___8. Two-thirds of consumer spending is accounted for by “other durable goods” and “housingservices.

___9. Producer’s durable equipment are reported in three categories: Motor vehicle, aircraft, andother producer durables.

___10. The simplest forecasting method is the use of leading indicators.

E550, Business Conditions Analysis NAME______________________________Fifth Quiz, August 17, Chapter 15

Dr. David A. Dilts

1. (3 points) What three ways do economic forecasts vary: a. _________________,b.________________, c.________________

2. What comprises the M3 money supply? (4 points) a.____________ b.____________ c. _____________ d. _________________

3. What are the two needs of an economic forecaster?__________________________(2 points)

4. What are the five phases of the business cycle? (2 points)_____________________________

______________________________________________________________________________

5. Briefly outline what a regression model is (3 points)_________________________________

______________________________________________________________________________

______________________________________________________________________________

True/False (1 point each)

___1. The domestic sector is C + I + G.

___2. Consumption is dependent on disposable income, and wealth is built in as a stream ofpotential earnings.

___3. Investment is sensitive to interest rates.

___4. Public savings is the amount the government spends minus the aggregate budgetsurpluses.

___5. A rise in interest rates abroad will reduce the value of the domestic currency at eachdomestic interest rate.

___6. A decrease in the domestic interest rate will cause an increase in the flow of savings in theloanable funds market domestically.

E550, Business Conditions Analysis NAME______________________________Sixth Quiz, August 18, Chapters 16, 17

Dr. David Dilts

True / False (1 point each)

___1. Quinn claims that successful strategic management with be to recognize patterns ofimpending change, anomalies, or promising interactions, then monitor, reinforce, and exploitthem.

___2. Perceptual accuracy seems to be negatively correlated with performance as measured byprofits according to Sutcliffe and Weber.

___3. Top managers should focus on managing ambiguity for subordinates, rather than worryingabout the accuracy of data or forecasts according to Sutcliffe and Weber.

___4. Forecasting is just one tool in managing ambiguity and uncertainty.

Short Answers (as indicated)

1. Describe briefly nine forecasting or analytical methods presented in this course. Tell whattype of data it uses, what the output will be, and whether it is accessible. (9 points)

a. f.

b. g.

c. h.

d. i.

e.

2. Briefly explain each of the following forecasting methods: (4 points)

Judgments methods

Counting methods

Time Series methods

Association or causal methods

3. Managers can improve their projections in the following three ways: (3 points)a. b. c.