Bachelorarbeit Philipp Hänggi

7/31/2019 Bachelorarbeit Philipp Hnggi

1/26

University of Bern

Faculty of Social and Economic Sciences

Economic Department

Prof. Dr. Haris Dellas

Coached by Luzia Halter

An Estimation of Factors of Competitiveness

Philipp Hnggi

04-112-512

Grafenfelsweg 14

4500 Solothurn

November 2010


2/26

Table of Contents

1. Introduction ....................................................................................................................... 12. Theoretical Background .................................................................................................... 33.

The Model ......................................................................................................................... 5

3.1 Explanatory Variables.................................................................................................. 73.2 Dependent Variables ................................................................................................... 93.3 Countries in the estimation ........................................................................................ 10

4. Testing the model............................................................................................................ 104.1 Exports ...................................................................................................................... 114.2 Inward Investments ...................................................................................................134.3 Outward Investments ................................................................................................. 14

5. Conclusions.....................................................................................................................176. Appendix ......................................................................................................................... 19

6.1 Stata-Output ..............................................................................................................197. Sources ........................................................................................................................... 22

7.1 Data ........................................................................................................................... 227.2 Literature ...................................................................................................................22

8. Selbstndigkeitserklrung ............................................................................................... 24

Charts

Chart 1 World Exports as Percentage of World GDP (Source: Worldbank 2010a) .................2Chart 2 Sum of Exports and Imports of all OECD Countries (Source: OECD 2010) ...............2Chart 3 Average Export and Import Volume per Country (Source: OECD 2010) ....................2Chart 4 World Inward FDI (Source: Worldbank 2010b) .........................................................16Chart 5 World Outward FDI (Source: Worldbank 2010c).......................................................16

TablesTable 1 Estimation Results: Exports (Source: Compiled by the author) ................................11Table 2 Estimation Results: Inward Investments (Source: Compiled by the author) .............13Table 3 Estimation Results: Outward Investments (Source: Compiled by the author) ..........14


3/26

1

1.Introduction

The content of this paper discusses the determinants of international competitiveness

of industrialized countries based on a regression, testing several indicators on

competitiveness. With regards to the order of the topic, first of all there will be a

theoretical discussion about the topic, followed by an estimation of the model. The

paper finishes with certain conclusions from the regression. The objective is to

evaluate the influence of certain determinants on international competitiveness, such

as technology, education and economic performance indicators. The estimation

should show if some presumed main factors of competitiveness are really important

or just overrated and if there are some underestimated factors, which should be paid

more attention to.

The aim of the paper is to evaluate a base for approaches to avoid insufficient

competitiveness in an international embedded, industrialized economy in the future.

The empirical part is supposed to bring the basis for a further discussion on the topic.

The paper focuses merely on industrialized countries, because I was always

interested in finding out what could be the next step of development in those

countries. The beginning of the development in all industrialized countries was a

strong industrial sector with mainly mass production and farming. Until now, the

economies of those countries include also a mature service sector as well as mostly

a high technology sector. Farming and industrial mass production has become less

important. Some industries such as the textile industry in Switzerland become extinct.

This dynamic system implies enduring change of factor endowments and a countrys

infrastructure to be seen as new challenges. Therefore, the individual country must

supervise its infrastructure and endowment (in a qualitative and quantitative way) to

manage these continuous challenges.

Besides these developments, there is also a change in the export intensity. The chart

below shows world exports and services as a percentage of the worlds GDP from

1952 to 2007. It is evident that export intensity of GDP increased over the years. The

more countries and companies export, the more increases the need to be able to

manage competitive challenges. The chart also shows that the dependency of acountrys wealth on exports gets stronger.


4/26

2

Chart 1 World Exports as Percentage of World GDP (Source: Worldbank 2010a)

The second chart represents the sum of exports and imports of all OECD countries

from 1955 to 2009 in billions of 2005 US dollars. The upward slope demonstrates the

trend of the above chart in absolute numbers.

Chart 2 Sum of Exports and Imports of all OECD Countries (Source: OECD 2010)

The next chart shows the global average export and import volume for all countries

from 2005 to 2009. Except for the years of the global financial crisis, there is a strictly

upward sloping curve. The long-term trend is also increasing; meaning trade intensity

in general is growing.

Chart 3 Average Export and Import Volume per Country (Source: OECD 2010)

Another fact, that needs to be considered with regards to the changing economical

environment, is that money has become extremely versatile between countries. This

fact increases competition between countries on capital investment as well as it


5/26

3

increases competition in general, according to the factor equalization theorem. It

says that the remuneration of factors will equalize when trade between countries

happens (Caves et al. 2007). Thus, this paper evaluates both export market shares

and foreign direct investment (further FDI) together. The focus is on technology in

both of its kinds, embodied and dis-embodied as well as on economic performance

indicators.

In this paper, absolute advantages are taken as a proxy for national competitiveness.

The central theme in this paper follows the papers of Jan Fagerberg (International

Competitiveness, 1988) and Rajneesh Narula and Katharine Wakelin (Technological

Competitiveness, Trade and Foreign Investment, 1998). It is not to be seen as a

replica of those papers, but more as resumption of the topic, as some of the results

might be different after almost 20 years (the data Narula and Wakelin used were from

1975 to 1991). Therefore, it is of interest to observe the development of international

trade in the years since.

2.TheoreticalBackground

In another paper in 1997, Narula and Wakelin found out that multinational enterprises

(further MNEs) account for approximately 20% of the GDP in the major industrialized

countries. This explains why and where MNEs locate their activities, has become

crucial to understand economic growth and competitiveness. Even if the dominance

of MNEs on production and trade is growing, very few studies have tried to evaluate

the determinants of trade and FDI within a unified framework. Exceptions are

Cantwell, 1989 and Narula, 1995. There are remarkable similarities between trade

approaches and FDI theory, which underlines the differences in technology. This

section highlights three fundamental aspects, which render the basis for the empirical

part in this paper: the concept of competitiveness, the role of technology and the

nature of these two concepts on a country specific basis (Narula and Wakelin 1998).

Fundamental [] is the central role given to technology. Technology is

characterized as both firm specific (ownership advantages in the FDI literature), and

inherent in sector and country specific market structures. (Narula and Wakelin

1998). The treatment of technology can be summarized as follows (Narula and

Wakelin 1998):

1.) Privileged access to technology, and differences in technology, are seen as the


6/26

4

main motivation for trade and FDI, this is regarded as conferring ownership

advantages to firms when internalized.

2.) The cumulative nature of innovation and skills is used as an explanation for the

continuing existence of technology gaps over time.

3.) Location advantages prevail and are potentially available to all firms operating in

that environment, where such technology is not firm specific, but inherent in industry

and country specific structure of markets.

4.) The importance of learning-by-doing and learning-by-using effects in

technological change is underlined by both perspectives, along with the potential to

affect some of the benefits from innovation.

This characterization leads to the localization of the profits of innovation and the

obstacles of its unresisted diffusion. A common institutional framework and the

relations between producers and users result in a singular technological profile for

each country (Lundvall 1992). These national technological characteristics define an

important determinant of export performance as well as FDI. Due to this cumulative

nature of innovation, each country can create its own competitive advantages

(Wakelin 1997).

There are similarities in the concept of competitiveness. A companys or a locations

competitive advantage is an absolute advantage over other companies or locations.

Meanwhile, comparative advantage applies to the country level and influences the

pattern of a countrys trade specialization, hence the sectoral structure. A companys

competitive advantage influences the comparative which advantage can be

supportive or contradictive. Dosi et al. (1990) concluded that technological

differences are more influential than endowment-based comparative advantage with

regards to trade patterns. Even if comparative cost considerations may be relevant,

they suggest absolute differences in technology to be more important. It is to

underline that competitiveness is a company specific as well as a country specific

phenomenon. Even if country competitiveness represents the overall capability of

companies of a given nationality in a given sector to compete with firms of another

country in the same sector, this is not just the sum of the competitiveness of its

companies (Lall 1990). The competitiveness of companies is influenced also by the

general economic structure of the country in which they are located including

relations among sectors, the national system of innovation and trade specializationpatterns (Narula and Wakelin 1998).


7/26

5

Both, the level of development of a country and the type of international economic

activity vary the importance of the different components of structural competitiveness

in determining company and national competitiveness (Narula 1995). Companies in

industrialized countries usually have had a certain period of time to develop their own

characteristics, and so, in general, may be less dependent on the characteristics of

their home country (Narula and Wakelin 1998). Due to the fact that this paper

focuses on industrialized countries, it will be more important for the competitiveness

on how those companies and countries are engaged in international trade,

determined mainly by the extent of multinationality of the companies and the nature

of their international economical activity. This has got several consequences. First,

ownership advantages of purely domesticly owned companies producing for the local

as well as the international market are likely to have a stronger linkage with the

structural competitiveness of its home country, than a foreign owned multinational

company based in this country. Second, countries which are host to MNEs that do

considerable international business, e.g. the industrialized countries, are more likely

to be influenced by the country specific characteristics of the other countries in which

these MNEs operate, than countries with relatively less internationalized firms. Third,

different production sectors have different economies of scale or the demand may be

heterogeneous across countries, thus having different propensities for

internationalization (Narula and Wakelin 1998).

3.TheModel

The empirical model evaluated in this paper aims to estimate the main determinants

of international competitiveness. As dependent Variables I chose relative export

market share as well as relative market share of inward and outward investment,

respectively. The dependent variable for export market share is a countrys exports

relative to the exports of the entire sample, normalized by the ratio of that countrys

population, to the population of the whole sample. The same applies for the FDI

dependent variables. These dependent variables are normalized in order to take

account for the influence of different country sizes on them. The selected dependent

variables represent a countrys absolute advantage on international markets in terms

of exports and FDI (Narula and Wakelin 1998).


8/26

6

A traditional measure of a countrys international competitiveness is the relative unit

labor costs (Fagerberg 1988). In my opinion, relative unit labor costs are, if they have

an influence, just part of the model, namely as an explanatory variable. It cannot

stand as an indicator (dependent variable) on its own. It may be noted that this

approach is incompatible with the neoclassical equilibrium theory. In a perfectly

competitive market, prices and quantities will always adjust, resources will be

employed by the full extent and balance of payments equilibrium will be ensured

(Fagerberg 1988).

As an example - one might say that unit labor costs contain the level of technology,

but technology is not equal to productivity, which is evidently important for

competitiveness. Technological innovations can bring total new businesses and of

course at the beginning of a new product era, productivity is low. But the innovation

itself brings at least a temporary competitive advantage. Another reason for using

unit labor cost as an explanatory variable is its questionable effect on

competitiveness. Lets take Switzerland for instance. With a very high level of wages,

it is still one of the most competitive countries in the world. The same applies for

Scandinavian countries. This leads me to the assumption that the mix of all factors

really matters. Lets have a quick look at innovations in the industrial sector. Without

a competitive workforce (mainly low skilled workers) you cannot produce a new

computer generation, which was invented by a highly educated workforce and R&D

institutions. Therefore it is obvious that relative unit labor costs are neither useful as a

proxy for technology nor for the skill level of a work force. It is just what it is, the cost

of one unit of labor. Because costs are surely an important factor for decision makers

(also for FDI), I integrated relative unit labor costs in the model in order to find out

what influence they have on FDI and export market shares.

The estimation in this model will be based on certain explanatory variables as proxies

for the main factors of competitiveness. As mentioned, the model is based on the

papers of Narula and Wakelin (1998) and Fagerberg (1988). All descriptive variables

for a country are given on a per capita basis relative to the country with the highest

per capita value in the sample. Export market shares as well as inward and outward

investment will be tested with the following explanatory variables.


9/26

7

3.1ExplanatoryVariables

TLi = Technology level of country i relative to the most advanced country of the

sample

In this model, a nations technology level will be measured with patents granted in the

country. Technology is taken as one of the most important pillars of international

competitiveness. This variable represents the embodied sector of technology (i.e.

innovation). I expect it to be positively related to export market shares as well as

inward FDI. The effect on outward investments is supposed to be negative.

RULCi = relative unit labor costs in common currency for country i

The costs of labor are certainly to expect to have an influence on competitiveness.But the strength of this influence is not clear. As mentioned above, there are

countries with high unit labor costs such as Switzerland or Scandinavian countries,

which ultimately rank under the most competitive countries in global annual

competitiveness rankings (World Economic Forum 2010). Another value is that in this

model, there are only industrialized countries with similar levels of unit labor costs, so

they cannot make the difference. For this, the role of unit labor costs will be

interesting to observe in the estimation. RULC is measured by an index with level

100 in 2005. I expect RULC to be not that important (maybe insignificant) on every

variable tested.

RHCi = tertiary graduates

The amount of tertiary graduates compared to a nations population per year is the

models proxy for the quality of the education system. It also stands for the dis-

embodied technology sector, namely human capital. It is assumedly positive related

to competitiveness in general (Fagerberg 1988). It is expected to have a positive

coefficient in the export share model as well as in the inward FDI model. The effect

on outward FDI is questionable, but if there is one, it is supposed to be negative.

RGNPi = GNP per capita

GNP per capita is an indicator for the level of development of a country. As Narula

and Wakelin found out, it is the measurement with the highest explanatory power for

a countrys development status. It is assumed that a higher level of development is

associated with higher export as well as higher inward investment shares (Narula


10/26

8

and Wakelin 1998). In my opinion, the effect on outward investment is not clear to

predict in advance, due to a superior developed country has more money to invest,

and in contrary, it might have a lot of interesting investment opportunities within its

own territory.

RDEMDi = aggregate demand

RDEMD is a countrys aggregate private consumption. It is an indicator for the size of

a countrys market. A larger market is assumed to be more attractive for foreign

inward investment since economies of large-scale production is more likely to be

captured than in a smaller market. On the other side smaller markets are expected to

have higher export market shares as a result of the smaller domestic market. Smaller

countries are also assumed to have higher outward investments than larger ones

(Narula and Wakelin 1998).

RIWPOPi = stock of inward investment per capita

This variable is only tested in the export market share and outward investment

models. In both models RIWPOP is expected to be positive related. Its aim is to

consider the connection between trade and FDI. For the exports and outwards

investment models, this means including the inward investment variable as an

explanatory one. The relation is in both cases expected to be positive. A look on

exports shows that companies frequently make inward investments in a country to

export from that country afterwards. Particularly this might happen in countries that

are part of a costums union (i.e. EU) or a free trade agreement (i.e. NAFTA) and has

preferential access to markets. When a countrys domestic market is large enough, in

ward FDI are not supposed to have an impact. In the case of outward FDI, a higher

level is likely to attract high inward FDI, with companies seeking the sophisticatedassets of the country (Narula and Wakelin 1998).

RTIi = Exports plus imports

The main influence on the development of a companys ownership advantages is the

extent of its exposure to international competition. For the model testing the inward

investment share, trade intensity (RTI) is measured by the sum of exports and

imports. This captures the degree to which the country participates in international


11/26

9

markets (Narula and Wakelin 1998). It is expected to be positive related with inward

FDI.

Narula and Wakelin (1998) also included a proxy for the availability of natural

resources. They argued that a high share of primary exports of all exports is an

indicator for a less developed country. Therefore, in this paper, this variable is not

included due to testing industrialized countries only. Narula and Wakelins

argumentation on the result of this variable and the fact that there are high

competitive countries without natural resources (i.e. Switzerland) confirm the decision

to exclude this proxy. It is to mention that a lot of countries do have advantages from

a high abundance of natural resources, but to be abundant of natural resources is

just a sufficient and not a necessary condition to reach a high level of wealth. It is

important to me to only include proxies, which occur in all the countries tested.

Fagerberg (1988) also included terms of trade in his estimation, but with the RTI

variable in my model, there is already a proxy for trade intensity. To avoid any

unnecessary collinearity problems I decided to resign terms of trade in my

estimations.

3.2DependentVariables

The variable for the export share is defined as the ratio of the export (X) share to

population (pop) share. The market is given as the sum over i, where i represents all

countries in the sample. The relationship with outward and inward investment is

defined simultaneously (OW, IW). The models are estimated using ordinary least

squares (OLS) and logarithms are taken of all variables. Following are the identities

of the three models (Narula and Wakelin 1998).

X/ Xi

pop / pop

i

= f(rgnp,rhc, tl,rulc,rdemd,riwpop)

OW / OWi

pop / pop

i

= f(rgnp,rhc, tl,rulc,rdemd,riwpop)

IW / IWi

pop / pop

i

= f(rgnp,rhc,tl,rulc,rti,rdemd)


12/26

10

3.3Countriesintheestimation

The considered countries were:

Japan, Australia, New Zealand, United States, United Kingdom, France, Netherlands,

Denmark, Norway, Finland, Sweden, Austria, Italy and Switzerland.

4.Testingthemodel

Before discussing the results, a few things need to be mentioned: I will compare the

findings with those of Narula and Wakelin where my results were not equal

(concerning significance and signs of the coefficients). A complete comparison is still

not possible because Narula and Wakelin used pooled data, whereas I used panel

data. In addition, the time frame differs. Narula et al used data from 1975 to 1991,

therefore their data describes a longer time frame in another era. As mentioned in

the introduction, our global economic system is very dynamic, and therefore it is to

expect that my results might differ. This is ultimately the essence of this paper. The

three regressions were tested for multi-collinearity with the VIF test in Stata and a

partial correlation matrix, but the results are negative, so there is no multi-collinearity

problem. It is also important to highlight that only industrialized countries were tested,therefore the results will be different to other models testing also developing

countries. Almost complete satisfied markets like in Europe and northern America

also behave different from developing markets in certain ways. Last but not least,

whenever I am speaking about a variable being significant, it is significant at a 10%

level of significance or lower. The exact level of significance of each variable is

marked in the tables by footnotes.


13/26

11

4.1Exports

Variable Coefficient t-Statistics

.4831573 0.75

RGNP .5013181 3.251

RHC .0132483 0.62

TL -.0895908 -0.56

RULC -.0238627 -1.63

RDEMD -.4759862 -9.301

RIWPOP .1067626 7.231

Adjusted R2 0.7504

Observations 125

1Significant at 1%

Table 1 Estimation Results: Exports (Source: Compiled by the author)

The estimation of the model of export market shares showed a highly significant

result for the influence of GNP per capita on exports. This result was expected as a

higher level of production leads to more supply left to serve foreign markets. It is to

remark that the United States of America is an exception. There is a high level of

production per capita but the size of the market is large enough, that there is simply

no incentive to export goods and services to achieve higher economies of scale.

Another consent of my expectations is the significant positive relation between

exports and relative inward FDI. A country with a high level of inward investments

seems to have more capacities for production (ergo more exports), so this result

makes absolutely sense and shows a complementary relationship between trade and

FDI (see also Narula and Wakelin 1998).The results become more interesting when it comes to the technology variables. The

coefficient of granted patents is insignificant and surprisingly negative. The negative

sign of the coefficient is an enigma. There is no explanation that makes sense.

Therefore it is to be accepted, stressing the fact that it is anyway insignificant. The

number of tertiary graduates has a positive coefficient and is also insignificant.

Compared to the results of Narula and Wakelin where the patent variable is positive

and significant and the tertiary graduates variable is negative (at least in theregression without the inward investment variable) and insignificant, there is a need


14/26

12

for explanation. My interpretation for the insignificance of the patents variable in the

regression is that in this model, the variable only includes the patents granted, but it

does not say anything about how many of these patents became successful products

and how much return they created. It is also possible that those patents were sold to

companies in countries where production is cheaper. Another point of view would be

to say that today, industrial countries are more dependent on process-related

advancements than on product innovations. It is almost impossible to create an

indicator, which accounts for such learning-by-doing effects. Therefore, to find a

proxy for this would be very difficult.

A similar problem might be influential in the tertiary graduate variable case. Not every

field of tertiary education has the same influence on a modern economy. So it

depends on how many graduates every country has in every field that matters for

competitiveness. Fields like engineering sciences or natural sciences might be more

effective for the development of new products as social sciences or humanities. This

thesis relies on the fact that several industrialized countries have a lack of graduates

from engineering sciences (FAZ 2010), while the numbers of tertiary students tends

to increase (OECD 2010). To evaluate this topic, another regression would be

needed. Another Thesis would be that in industrialized countries the mobility of the

highly educated workforce brings in a bias. A lot of graduates from universities tend

to go abroad to gain work experience and to learn a foreign language. Some of them

might stay forever. In other words, every industrialized country educates people on a

global scale to a certain degree.

As expected, relative unit labor costs is scarcely not significant and also the

coefficient is very low. As mentioned before, we can find countries (i.e. Switzerland,

Sweden or Japan) with relative high wages and/or social charges, still being able to

compete with other countries with lower unit labor cost countries. The key is the

quality of labor supply of a country, which is almost impossible to measure on a

countrywide base. As mentioned above, all of the countries mentioned in the sample

have a similar level of unit labor costs (except for the Scandinavian countries).

As expected, the demand variable has a negative sign and is highly significant. This

confirms the expectations on this variable: A smaller domestic market is an incentive

to produce for foreign markets to achieve economies of scale, while larger countries

can attain the same scales in their own markets.


15/26

13

4.2InwardInvestments


-2.015034 -0.67

RGNP -.0565173 -0.07

RHC .2398521 2.421

TL .4130493 0.54

RULC .0840839 1.20

RTI 1.934637 7.011

RDEMD 1.204314 4.691

Adjusted R2 0.3615

Observations 1251Significant at 1%

Table 2 Estimation Results: Inward Investments (Source: Compiled by the author)

The inward FDI model surprisingly showed that GNP per capita has no significant

influence on inward FDI. With a closer look on this topic, it is visible that a higher

GDP per capita, which stands for a higher economic development of a country

(Narula and Wakelin 1998). This could implement a satisfied or highly competitive

market with low margins. Low margins, as commonly known, are not very attractiveto foreign investors. On the other hand, a high GNP per capita is not a proof for or

against investment opportunities. It also depends which sectors a country does

compete in and which of those are already highly competitive and which are new,

emerging markets with higher expected returns on investment.

The number of tertiary graduates per citizen has a positive coefficient and is

significant. According to the results, a higher amount of citizens with higher education

attracts foreign investments. On the other hand, more patents do not have an

influence on inward investments. The interpretation on this is the same as in the first

regression. The number of patents granted does not tell anything about how many of

them end up as products on markets and how high their returns are as well as if

those products will be produced in the same country. If it is known that patents in a

certain country were often sold to other countries, investors are harder to find.

Relative unit labor costs are not significant, as expected. Narula and Wakelin (1998)

received the exact the opposite results, so the roles of education level and the

amount of patents granted have changed over time.


16/26

14

Still a strong significant variable is RTI, which measures trade intensity of a country.

It is intuitive that a strong participation in trade attracts investments from abroad. I

also estimated the model with the RTI variable replaced by export share. The result

is very close to the model with RTI. The coefficient is approximately the same and it

is significant on the same level (1%). In conclusion, export market share is a good

proxy for trade participation as well.

The result of the last variable (relative demand) in the model is consistent with the

expectation. Larger markets attract more inward investments. But there is an

important point to take into consideration: in this regression, there are only

industrialized countries, but a lower demand could be the case in a less developed

country. In such case, it could still be interesting to invest in such a country, because

of the higher growth potential in developing and emerging countries than in

industrialized economies. In other words, this variable would change its sign in a

model for less developed countries.

4.3OutwardInvestments


-.2814231 -0.13

RGNP 1.532793 2.941

RHC -.1315598 -1.712

TL -.6541518 -1.20

RULC -.0370865 -0.73

RDEMD -.4223916 -2.401

RIWPOP .4654958 9.321

Adjusted R2 0.5205

Observations 1211Significant at 1%;

2Significant at 10%

Table 3 Estimation Results: Outward Investments (Source: Compiled by the author)

The outward FDI model shows a positive and significant influence of GNP per capita

on outward FDI. This result confirms the intuition that more developed countries have

more capital to invest. Therefore, they will spread their investments globally. The

significance at 1% underlines the importance of economic development for the ability

to invest globally and profit from emerging markets.


17/26

15

The result of the RHC variable is additionally not surprising. The negative sign of its

coefficient shows that the higher the level of education hence the quality of the work

force in a country is, the lower are the investments in other markets versus domestic.

The coefficient of the patents granted variable has the same sign, but it is not

significant. The reason might be in the generality of the used data. As mentioned

above, the number of patents granted includes all patents, but does not say anything

about the returns these patents have when they result in mature products as well

how many of those actually become products. Narula and Wakelin (1998) received

the opposite concerning significance and a positive sign for the patents variable.

After observing this in both FDI estimations, it seems that the level of education

becomes more important for the last ten years.

As expected, relative unit labor costs are not significant here as well as in the

regressions above. But the negative sign is a surprise. After three estimations with

the same result, namely insignificance, it is to scrutinize if relative unit labor costs are

a useful proxy in these models. As cited from Fagerberg in the theoretical section of

this paper, relative unit labor costs are not really qualified to measure

competitiveness. It would be helpful to have more specific and long-term data on this

topic. A productivity index could solve the problem, but how can one measure the

productivity of a whole economy? As an example, lets take natural scientists in R&D.

A lot of their work contains thinking about a problem and analyze the situation. How

can somebody measure the effort and quality of a scientists way of thinking about a

problem? It is almost impossible to find an indicator for sorts of productivity and bring

them down to an index.

The demand variable has the same influence on outward FDI as on export market

shares. As mentioned, a higher domestic demand implements a larger market. The

market size of a country is able to influence the number of investment opportunities.

If we take the United States of America for instance, we can see there is a large

enough market to profit from economies of scale. Therefore it is not necessary to

invest in foreign companies if the domestic market offers comparable attractive

alternatives. Narula and Wakelin received a positive sign in their estimation so it

seems that the role of demand, when it comes to outward FDI, has changed over the

years. Additionally it is more comprehensible that investors from a larger market are

more interested in domestic investments as investors from smaller markets.


18/26

16

The large proportion of FDI activity among industrialized countries explains at least

partly the positive relation of inward and outward investments (Narula and Wakelin

1998). It would be interesting to test if the same relation exists between import

volumes and export volumes. Anyways, the following charts show how symmetrically

inward and outward FDI behave.

Chart 4 World Inward FDI (Source: Worldbank 2010b)

Chart 5 World Outward FDI (Source: Worldbank 2010c)

I think that an explanation for the high significance of inward FDI on outward FDI

could be that in a country, which attracts a lot of foreign investments, the possibilities

to invest in this market decreases for domestic investors. In other words, the supply

of investment opportunities in a country stays the same while the demand for those

increases. Therefore, domestic investors are forced to invest in foreign countries.


19/26

17

5.Conclusions

After estimating three indicators for international competitiveness among

industrialized countries and comparing the results with those from a regression made

ten years ago (the data used is even older), some remarkable changes, or

differences, were observed. Even though the proxies used in the models are not

exactly the same, the comparable explanatory power of both estimations allows a

comparison of what are or were important indicators for measuring competitiveness.

First of all, I will discuss my own findings and then I will bring them in relation to the

situation ten years ago.

I start with one of the main findings of the estimation. It is a real disappointment that

the technological variables were not as significant as I expected. As previously

mentioned, the numbers I used might have not been specific enough to describe the

situation. A good proxy for measuring the efficiency of a work force is almost

impossible to create as well as one for a countrys quality of technology. It would be

necessary to find an index, which represents the performance of a work force in all

sectors compared to their education level discounted with its quality compared with

other countries quality of education. This index should include explicitly the efficiency

with which the GDP is reached. The level of unit labor cost does not say anything

about efficiency, neither does the share of tertiary students in population. The same

counts for technology. The index should ideally include the economic potential of

innovations and inventions made in a country and especially to which extent it is

exploited. The largest problem would be to find all the necessary data, because it is

not to expect that enough companies would be willing or able to supply the essential

economic numbers.

After the process of writing this paper, I realized that there is a high potential to

increase the explanatory power of economic indicators. For some purposes, there is

no sufficient data available. The technology variables used in this model are good

examples. Of course, the economic performance of an invention is private

information; therefore it would be difficult to argue that companies must make them

public. On the other hand, we live in a system, in which we require a lot of

information about the status quo in order to make responsible decisions for the

future. Maybe an acceptable trade-off for both sides might help. It is not my intensionto talk politics here, but for sustainable scientific work, specific data is indispensable.


20/26

18

Therefore I hope in the future, scientists will have access to more useful data to

conduct more valuable economic investigation.

More satisfying findings were the results of the economic indicator variables. Gross

National Product and domestic demand as well as inward FDI and trade intensity

play a significant role in the model. The most influential variable in the estimations

seems to be domestic demand. Since it stands for the size of a market, this implies

that large domestic markets face another form of competition than smaller markets:

in large markets (more import intensive), domestic companies compete with a lot of

foreign companies in their own country, while in smaller markets, domestic

companies compete more in foreign markets. In other words, companies from

smaller markets are more forced to get along with different legislatures than

companies from larger markets. As a consequence, companies from smaller markets

might be more able to answer challenges more effectively on new competitive

situations.

The grade of economic development, measured with GNP per capita, is crucial as

well. The more developed an economy is, the more it can compete on international

markets.

The estimations leave one question concerning inward FDI. Does a country increase

its exports if inward FDI increases, or vice versa? In my opinion, an economy, which

is strong in exports, attracts foreign investors (because of economies of scale etc.),

which enforces competitiveness even more on international markets.

The trade intensity variable has an implicit conclusion. If a country is strongly

engaged in trade, it attracts a lot of foreign investors. This implies that investors trust

in trade. This again leads back to a very often-mentioned phrase in international

trade books: With trade, everyone is better off!

After all findings on the regressions, there is one point left to mention. Chart number

three in section one shows the influence of the financial crisis on international trade.

In 2007, imports and exports dropped as a direct consequence of the crisis. Chart

number four and five state the same for FDI. One of the drivers in the crisis was the

lack of trust in companies (foremost banks) and institutions, as well as governments

like a German study showed (Musiol et al. 2009). Trust is fundamental when it comes

to economic interactions. Therefore, no matter how well endowed and competitive a

countrys economy is, when there is no trust, there is no trade.


21/26

19

6.Appendix

6.1Stata-Output


22/26

20


23/26

21


24/26

22

7.Sources

7.1Data

The data on outward and inward FDI, relative demand and relative unit labor cost was taken

from the OECD:

http://stats.oecd.org/index.aspx

The data on export market share, gross national product and trade intensity was taken from

the UN data explorer:

http://data.un.org/Explorer.aspx?d=KI

The data on patents granted was taken from the World Intellectual Property Organisation:

http://www.wipo.int/ipstats/en/statistics/patents/

The data on tertiary graduates was taken from the UNESCO data center:

http://stats.uis.unesco.org/unesco/TableViewer/document.aspx?ReportId=143&IF_Language

=eng

7.2Literature

Cantwell, J. (1989), Technological Innovation and Multinational Corporations, Oxford.

Caves, R. E. et al. (2007), World Trade and Payments, An Introduction, 10th Edition, Boston.

Dosi, G. et al. (1990), The Economics of Technical Change and International Trade,

Brighton.

Fagerberg, J. (1988), International Competitiveness, in: Economic Journal 98, pp. 355-374.

Frankfurter Allgemeine Zeitung (FAZ) (2010),Der Ingenieurmangel kommt mit Wucht,

Online on the Internet: URL:

http://www.faz.net/s/RubC43EEA6BF57E4A09925C1D802785495A/Doc~E0CBA323

7CCD241A8B2EDB606C18FFAAA~ATpl~Ecommon~Scontent.html[11/26/2010].

Lall, S. (1990), Building Industrial Competitiveness in Developing Countries, Paris.


25/26

23

Lundvall, B. A. (ed) (1992), National Systems of Innovation: Towards a Theory of Innovation

and Interactive Learning, London.

Musiol, K. G. et al. (2009), Vertrauen in der Krise II, Eine aktuelle Bestandsaufnahme und

Implikationen fr Markenverantwortliche, Online on the Internet: URL: http://www.mc-

brandnews.de [11/04/2010].

Narula, R. (1995), Multinational Investment and Economic Structure, London.

Narula, R. and K. Wakelin (1997), The Pattern and Determinants of US Foreign Direct

Investment in Industrialized Countries, TSER-TEIS Working Paper Series No. 8.

Narula, R. and K. Wakelin (1998), Technological Competitiveness, Trade and Foreign

Investment, in: Structural Change and Economic Dynamics 9, pp. 373-387.

OECD (2010),Stat Extracts, Online on the Internet: URL:http://stats.oecd.org/index.aspx

[7/18/2010].

Wakelin, K. (1997), Trade and Innovation: Theory and Evidence, Aldershot.

Worldbank (2010a), Exports of Goods and Services (% of GDP), Online on the Internet:

URL: http://data.worldbank.org/indicator/NE.EXP.GNFS.ZS/countries?display=graph

[11/04/2010].

Worldbank (2010b), Foreign Direct Investment, Net Inflows (% of GDP), Online on the I

Internet: URL:

http://data.worldbank.org/indicator/BX.KLT.DINV.WD.GD.ZS?display=graph

[11/04/2010].

Worldbank (2010c), Foreign Direct Investment, Net Outflows (% of GDP), Online on the

Internet: URL:

http://data.worldbank.org/indicator/BM.KLT.DINV.GD.ZS?display=graph [11/04/2010].

World Economic Forum (2010), The Global Competitiveness Report, Online on the

Internet: URL:

http://www.weforum.org/en/initiatives/gcp/Global%20Competitiveness%20Report/inde

x.htm[11/26/2010].


26/26

8.Selbstndigkeitserklrung

Ich erklre hiermit, dass ich diese Arbeit selbststndig verfasst und keine anderen als die

angegebenen Quellen benutzt habe. Alle Stellen, die wrtlich oder sinngemss aus Quellen

entnommen wurden, habe ich als solche gekennzeichnet. Mir ist bekannt, dass andernfalls

der Senat gemss Artikel 36 Absatz 1 Buchstabe o des Gesetzes vom 5. September 1996

ber die Universitt zum Entzug des aufgrund dieser Arbeit verliehenen Titels berechtigt ist.

Philipp Hnggi

Bachelorarbeit Philipp Hänggi

Documents

Transcript of Bachelorarbeit Philipp Hänggi