1 IntroductionAfter more than twenty years of empirical research in the relationship between taxes and foreign direct investment (FDI), there is consensus among most re- searchers that taxes do matter for the decision of multinational enterprises (MNE) on where and how much to invest. Correspondingly, policy-makers often rely on the eectiveness of corporate tax policy reforms in order to attract FDI, and the current debate suggests that they will continue to do so in the future. But, lower-ing the tax burden on investment necessarily implies a cut in public expenditure or a shift of the tax burden to other tax bases like e.g. labor or consumption.Therefore, it is necessary not only to know whether taxes do matter but also how much they do. In other words, the quantitative dimension of the tax impact on FDI is decisive for the design of sound tax policy which carefully weighs the bene-.ts of a corporate tax reduction to the economy as a whole against the cost.1 The question is: How much additional investment or production do we get for a given loss of tax revenue?The purpose of this paper is to measure the elasticity of FDI with respect to corporate tax reductions. We do so by analyzing the eect of the German tax reform in 2000, which came into force in January 2001. This reform implied substantial corporate tax rate cuts and broadened the corporate tax base. A frequently cited goal of the tax reform was to attract foreign direct investment in order to foster economic growth and mitigate the high unemployment rate. Now,.ve years after the reform, we ask whether the tax reform reached its goal. We analyze this question by using the very richMiDi data set fromthe Deutsche Bundesbank with .rm-speci.c balance sheet data of foreign-held companies. Our analysis contributes to a literature that tries to clarify the incentive eects of existing tax systems on corporate investment. As corporate investment is assumed to be crucial for the generation of new jobs and growth, we think that this question is at the heart of future debates on corporate tax reforms. 2 The German tax reform of 2000The main goals of the German Tax Reform 2000 were to improve the competit- iveness of .rms in Germany, to foster investment, to increase Germany.s attract- iveness to foreign investors and to adapt the corporate tax system to the rules of the EC common market. With regard to the corporate tax system, the formerly dierent tax rates on retained earnings (40 percent) and distributed pro.ts (30 percent) were replaced by a single and lower tax rate on all pro.ts (25 percent).In addition, the reform eliminated a long list of loopholes.4 including local trading taxes (Gewerbesteuer) the combined statutory tax rate of the old system was 54,3% while the new combined rate is on average 39,4%, see Spengel (2001). The corporate tax base was broadened substantially. Therules for thin capitalization of foreign companies and related party .nancing were tightened. Depreciation allowances were reduced in terms of expected value fortangible assets, like machines, and structures.The decline of the statutory tax rate ensures that both the marginal and the average eective tax rate for all assets decrease with the tax reform of 2000. The tax base broadening for investment in machines and structures, though, leads to a relative disadvantage for those assets compared to .nancial and inventory assets.4For a complete description of the reform please refer to Keen (2002), Homburg (2000) and Schreiber (2000).Next to the changes in the corporate tax system the reform lowered the top personal income tax rate and the imputation system for the taxation of dividends was replaced by a shareholder relief system, the so called half income method (Hal- beinknfteverfahren) which stipulates that 50% of the diovidend after corporate tax is subject to personal income taxation.The reform of the personal income tax system and the change in the integration technique of corporate and shareholder taxation are important and probably rel- evant for the investment decision of foreign investors. But, as we lack appropriate data on shareholders, we cannot use these reform features for our purpose. In the following we will restrict ourselves to the reform of the corporate tax system itself, but we will discuss in how far the other reform parts might play a role in shaping the investment process.3 The theoretical underpinningThis section develops the conceptual framework for our empirical analysis. We have to de.ne the nature of decisions that multinational companies make and that could potentially be distorted by taxation. As pointed out by Devereux and Gri th (2003), it is helpful to distinguish three dimensions of the investment decisions of multinational .rms: The .rst is to decide whether to export or to produce abroad (internationalization decision). The second decision is where toproduce (location decision). The third decision is how much to produce, or: how much to invest (investment decision).Correspondingly, there are two types of tax eects on FDI. First, taxes may reduce the average return of a project and thus in.uence the internationalization and location decisions by .rms. Second, taxes may change the user cost of capital and thus have an impact on the investment decision. Our dataset does not to analyse the internationalization or the location decision of MNEs. We just observe existing capital stocks and their variation over time. Therefore, our main focus is on the choice of the optimal capital stock, i.e. the investment decision. However,discrete jumps in the balance sheet capital stock suggest that we can observe quasi- location decisions where .rms decide to locate the production of new goods in one country or another. As explained further below, our analysis will therefore take into account both the marginal and the average tax burden on investment.53.1 The main hypothesisAssume that there is an MNE located in a country outside Germany, which has an a liate in Germany and - potentially - in other countries as well.5 Using a very general formulation, we can state that the MNE chooses the size of the capital stocks depending on a .nite vector x including global, country-speci.c, activity- speci.c and .rm-speci.c parameters:_ = _(K1; :::;Kn) with Kh = Kh (xh) (1)with h = 1; :::; n. The x is a m _ 1 vector of parameters which are candidates for in.uencing the investment decision. One of these parameters is supposed to be taxation. Total dierentiation yields:dK =@K@x1dx1 + ::: +@K@xtaxdxtax + ::: +@K@xndxm (2)xtax is some tax variable to be operationalized later on. Held everything else constant, i.e. dxg = 0 8 g 6= tax with g = 1; :::;m, the partial eect of the tax variable on the capital stock K is:dK dxtax= @K @xtax _ _tax (3)Our main hypothesis is that taxation has a negative impact on the foreign-held capital stock, i.e. _tax < 0. There are two channels through which this relation can be established. First, consider the case in which taxes increase the cost of capital, and assumethat the pre-tax cost of capital does not depend on country-speci.c characteristics but rather on some world capital market. In this case the size of the optimal capital stock falls because some marginal investment projects are not realized any more.Here, the capital stock of the foreign parent company and the one of the German a liate are not systematically linked: i.e. dKh dKh = 0. Given that the a liate is not liquidity-constrained, the investment is independent of any factor in.uencing the other a liates, the parent company or their locations characteristics.Second, if taxes increase the average tax burden of a given project, the probabil- ity rises that this project will be realized elsewhere, e.g. in the country of residence of the parent company: dKhdKh < 0. In this case, the a liate investment depends on factors in.uencing the other capital stocks as well, since they change the relative 5This assumption is justi.ed by the type of data that we use. We only consider foreigninvestors which already have a liates in Germany. Thus, the decision is just how much to invest and not if at all, i.e. not two-fold like in Devereux and Gri th (1998) or in Razin, Rubinsteinand Sadka (2004).attractiveness of the a liate location. These factors include, in particular, foreign taxes.Whether FDI are better described by the .rst or the second model has consid- erable policy implications. Sinn (1990) argues that a country can immunize itself against tax competition by lowering the eective marginal tax rate to zero and by only taxing intra-marginal pro.ts. This is true for the .rst model but not for the second. If .rms and/or projects are mobile and not only capital, i.e. if the second model is valid, there will be no immunization strategy and even the oppos-ite policy, a tax rate cut cum base broadening strategy, might become optimal, asis shown in Becker and Fuest (2005). It is one of the objectives of our analysis to .nd hints at which of the two distinct models performs better in describing the empirical investment data.The literature often dierentiates between cost-driven investment and market- entry investment, suggesting that the latter are not or only weakly tax-sensitive.The idea behind this statement is that every unit of a market-entry investment has some complementary units of production in another country. In this case, the tax eect is weaker because domestic taxes only have an impact on the domestic part of the whole investment. In contrast, cost-driven investment is part of a strategy of disentangling the production chain. If the goods are not produced in the country under consideration, they will be produced elsewhere. Proximity to consumers or market-entry reasoning do not play a role.In principle, these two types of investment could be used as control and treat- ment group in order to identify the tax impact on investment. The di culty is that the data does not directly reveal the type of investment. One often used approximation is the dierentiation between horizontal investment and vertical investment, see e.g. Buch et al. (2005). Horizontal (or intra-branch) investment issupposed to be realized because of market-entry reasons, vertical (or inter-branch) investment because of cost dierences. As we will outline later on we do not have any means to dierentiate between these two, either, due to data limitations..

Appendix: Data sourcesFDI dataThe Deutsche Bundesbank carries out annual full sample surveys on inbound and outbound direct investment stocks based on the provisions of the German Foreign Trade and Payments Regulation. Due to this legal regulation, foreign companies with investments in Germany have to report selected balance sheet information of their German subsidiaries. The balance sheet data are calculated using the German accounting regulations. Similarly, German multinational companies have to report the same information about their foreign a liates. For an extensivedescription of the database, see Lipponer (2003a) and Lipponer (2003b). The data is available for the years 1989 to 2003. Time series for individual companies are available for the years 1996 to 2003. In 2002, about 6,000 domestic investors .led reports on around 22,000 foreign subsidiaries abroad. With respect to inward FDI, in 2002 data are available for about 10,000 a liates in Germany, in which some 7,000 foreign investors had a participating interest.For the purpose of our paper we used only the data on inbound FDI. In order to create a balanced data set, we only use data from 1996 on and exclude all .rms which do not have full coverage of the whole period 1996-2003. The balanced panel dataset contains 2830 subsidiaries in Germany.The investors are based in the following countries: France, Belgium, Nether- lands, Luxembourg, Italy, UK, Ireland, Denmark, Spain, Norway, Sweden, Fin- land, Austria, Switzerland, USA, Canada, South Korea and Japan. If a subsidiary is owned by investors from dierent countried we assumed that the home country is the one where the investor with largest share in the subsidiary comes from. If two or more investors are equal in size, we choose the one from the country which comes .rst in alphabetic order.The most important branches of activity in our data set are Manufacturing, Wholesale, Holdings, Services to Companies and the Financial Services. We con- centrate on incorporated non-public companies that have either the legal form of a corporation (Aktiengesellschaft (AG)) or a limited liability company (Gesellschaft mit beschrnkter Haftung (GmbH)).

Tax related data and other dataFor the calculation of the economic depreciation rates in equation (10) we adopt the assumptions reported in Spengel (2001): Intangible assets depreciate over a period of 12.5 years, buildings over 53 years and machinery over 11 years. To calculate Ai in equation (9), we assume a nominal interest rate of 10 percent and an in.ation rate of 3.5 percent. These values are adopted from Devereux, Grif- 31 .th and Klemm (2002) who also provide the data on other countries.depreciation allowances and statutory tax rates. The _i are taken from branch-speci.c aggregate data of the Deutsche Bundes- bank15 assuming that the a liates held by foreign owners have the same tangible capital structure as the industry average.

CESifo Working Paper Series___________________________________________________________________________1657 Daniel Haile, Abdolkarim Sadrieh and Harrie A. A. Verbon, Cross-Racial Envy andUnderinvestment in South Africa, February 20061658 Frode Meland and Odd Rune Straume, Outsourcing in Contests, February 20061659 M. Hashem Pesaran and Ron Smith, Macroeconometric Modelling with a GlobalPerspective, February 20061660 Alexander F. Wagner and Friedrich Schneider, Satisfaction with Democracy and theEnvironment in Western Europe a Panel Analysis, February 20061661 Ben J. Heijdra and Jenny E. Ligthart, Fiscal Policy, Monopolistic Competition, andFinite Lives, February 20061662 Ludger Woessmann, Public-Private Partnership and Schooling Outcomes acrossCountries, February 20061663 Topi Miettinen and Panu Poutvaara, Political Parties and Network Formation, February20061664 Alessandro Cigno and Annalisa Luporini, Optimal Policy Towards Families withDifferent Amounts of Social Capital, in the Presence of Asymmetric Information andStochastic Fertility, February 20061665 Samuel Muehlemann and Stefan C. Wolter, Regional Effects on Employer ProvidedTraining: Evidence from Apprenticeship Training in Switzerland, February 20061666 Laszlo Goerke, Bureaucratic Corruption and Profit Tax Evasion, February 20061667 Ivo J. M. Arnold and Jan J. G. Lemmen, Inflation Expectations and InflationUncertainty in the Eurozone: Evidence from Survey Data, February 20061668 Hans Gersbach and Hans Haller, Voice and Bargaining Power, February 20061669 Franoise Forges and Frdric Koessler, Long Persuasion Games, February 20061670 Florian Englmaier and Markus Reisinger, Information, Coordination, and theIndustrialization of Countries, February 20061671 Hendrik Hakenes and Andreas Irmen, Something out of Nothing? Neoclassical Growthand the Trivial Steady State, February 20061672 Torsten Persson and Guido Tabellini, Democracy and Development: The Devil in theDetails, February 20061673 Michael Rauber and Heinrich W. Ursprung, Evaluation of Researchers: A Life CycleAnalysis of German Academic Economists, February 20061674 Ernesto Reuben and Frans van Winden, Reciprocity and Emotions when ReciprocatorsKnow each other, February 20061675 Assar Lindbeck and Mats Persson, A Model of Income Insurance and Social Norms,February 20061676 Horst Raff, Michael Ryan and Frank Staehler, Asset Ownership and Foreign-MarketEntry, February 20061677 Miguel Portela, Rob Alessie and Coen Teulings, Measurement Error in Education andGrowth Regressions, February 20061678 Andreas Haufler, Alexander Klemm and Guttorm Schjelderup, Globalisation and theMix of Wage and Profit Taxes, February 20061679 Kurt R. Brekke and Lars Srgard, Public versus Private Health Care in a NationalHealth Service, March 20061680 Dominik Grafenhofer, Christian Jaag, Christian Keuschnigg and Mirela Keuschnigg,Probabilistic Aging, March 20061681 Wladimir Raymond, Pierre Mohnen, Franz Palm and Sybrand Schim van der Loeff,Persistence of Innovation in Dutch Manufacturing: Is it Spurious?, March 20061682 Andrea Colciago, V. Anton Muscatelli, Tiziano Ropele and Patrizio Tirelli, The Role ofFiscal Policy in a Monetary Union: Are National Automatic Stabilizers Effective?,March 20061683 Mario Jametti and Thomas von Ungern-Sternberg, Risk Selection in Natural DisasterInsurance the Case of France, March 20061684 Ken Sennewald and Klaus Waelde, Its Lemma and the Bellman Equation forPoisson Processes: An Applied View, March 20061685 Ernesto Reuben and Frans van Winden, Negative Reciprocity and the Interaction ofEmotions and Fairness Norms, March 20061686 Franoise Forges, The Ex Ante Incentive Compatible Core in Exchange Economieswith and without Indivisibilities, March 20061687 Assar Lindbeck, Mrten Palme and Mats Persson, Job Security and Work Absence:Evidence from a Natural Experiment, March 20061688 Sebastian Buhai and Coen Teulings, Tenure Profiles and Efficient Separation in aStochastic Productivity Model, March 20061689 Gebhard Kirchgaessner and Silika Prohl, Sustainability of Swiss Fiscal Policy, March20061690 A. Lans Bovenberg and Peter Birch Srensen, Optimal Taxation and Social Insurance ina Lifetime Perspective, March 20061691 Moritz Schularick and Thomas M. Steger, Does Financial Integration Spur EconomicGrowth? New Evidence from the First Era of Financial Globalization, March 20061692 Burkhard Heer and Alfred Maussner, Business Cycle Dynamics of a New KeynesianOverlapping Generations Model with Progressive Income Taxation, March 20061693 Jarko Fidrmuc and Iikka Korhonen, Meta-Analysis of the Business Cycle Correlationbetween the Euro Area and the CEECs, March 20061694 Steffen Henzel and Timo Wollmershaeuser, The New Keynesian Phillips Curve and theRole of Expectations: Evidence from the Ifo World Economic Survey, March 20061695 Yin-Wong Cheung, An Empirical Model of Daily Highs and Lows, March 20061696 Scott Alan Carson, African-American and White Living Standards in the 19th CenturyAmerican South: A Biological Comparison, March 20061697 Helge Berger, Optimal Central Bank Design: Benchmarks for the ECB, March 20061698 Vjollca Sadiraj, Jan Tuinstra and Frans van Winden, On the Size of the Winning Set inthe Presence of Interest Groups, April 20061699 Martin Gassebner, Michael Lamla and Jan-Egbert Sturm, Economic, Demographic andPolitical Determinants of Pollution Reassessed: A Sensitivity Analysis, April 20061700 Louis N. Christofides and Amy Chen Peng, Major Provisions of Labour Contracts andtheir Theoretical Coherence, April 20061701 Christian Groth, Karl-Josef Koch and Thomas M. Steger, Rethinking the Concept ofLong-Run Economic Growth, April 20061702 Dirk Schindler and Guttorm Schjelderup, Company Tax Reform in Europe and itsEffect on Collusive Behavior, April 20061703 Franoise Forges and Enrico Minelli, Afriats Theorem for General Budget Sets, April20061704 M. Hashem Pesaran, Ron P. Smith, Takashi Yamagata and Liudmyla Hvozdyk,Pairwise Tests of Purchasing Power Parity Using Aggregate and Disaggregate PriceMeasures, April 20061705 Piero Gottardi and Felix Kubler, Social Security and Risk Sharing, April 20061706 Giacomo Corneo and Christina M. Fong, Whats the Monetary Value of DistributiveJustice?, April 20061707 Andreas Knabe, Ronnie Schoeb and Joachim Weimann, Marginal EmploymentSubsidization: A New Concept and a Reappraisal, April 20061708 Hans-Werner Sinn, The Pathological Export Boom and the Bazaar Effect - How toSolve the German Puzzle, April 20061709 Helge Berger and Stephan Danninger, The Employment Effects of Labor and ProductMarkets Deregulation and their Implications for Structural Reform, May 20061710 Michael Ehrmann and Marcel Fratzscher, Global Financial Transmission of MonetaryPolicy Shocks, May 20061711 Carsten Eckel and Hartmut Egger, Wage Bargaining and Multinational Firms in GeneralEquilibrium, May 20061712 Mathias Hoffmann, Proprietary Income, Entrepreneurial Risk, and the Predictability ofU.S. Stock Returns, May 20061713 Marc-Andreas Muendler and Sascha O. Becker, Margins of Multinational LaborSubstitution, May 20061714 Surajeet Chakravarty and W. Bentley MacLeod, Construction Contracts (or How toGet the Right Building at the Right Price?), May 20061715 David Encaoua and Yassine Lefouili, Choosing Intellectual Protection: Imitation, PatentStrength and Licensing, May 20061716 Chris van Klaveren, Bernard van Praag and Henriette Maassen van den Brink,Empirical Estimation Results of a Collective Household Time Allocation Model, May20061717 Paul De Grauwe and Agnieszka Markiewicz, Learning to Forecast the Exchange Rate:Two Competing Approaches, May 20061718 Sijbren Cnossen, Tobacco Taxation in the European Union, May 20061719 Marcel Grard and Fernando Ruiz, Interjurisdictional Competition for Higher Educationand Firms, May 20061720 Ronald McKinnon and Gunther Schnabl, Chinas Exchange Rate and InternationalAdjustment in Wages, Prices, and Interest Rates: Japan Dj Vu?, May 20061721 Paolo M. Panteghini, The Capital Structure of Multinational Companies under TaxCompetition, May 20061722 Johannes Becker, Clemens Fuest and Thomas Hemmelgarn, Corporate Tax Reform andForeign Direct Investment in Germany Evidence from Firm-Level Data, May 2006

FDI DataEuropean Stocks at the TopGermany ranks seventh in the world as a recipient of foreign direct investment (FDI), according to the United Nations Conference on Trade and Development (UNCTAD).

According to official Bundesbank (German Central Bank) statistics for 2012, fifty-eight percent (or EUR 346 billion) of all FDI stocks in Germany originate from within the EU-27, with a further eight percent stemming from the remaining European non-EU countries. Investments from outside the EU continue to grow. North America accounts for 24 percent of FDI stock, while Asia holds a nine percent share. Especially Asian countries increased their FDI stocks in Germany in recent years. Germany is the worlds largest recipient of new Chinese fdi projects.

First Choice in EuropeA recent study conducted by the American Chamber of Commerce highlights the positive regard in which the German business environment is held by US companies. Invited to indicate their main medium-term investment focus within the EU, 29 percent of participating American companies named Germany as their first choice followed by Eastern Europe (17 percent), and Western Europe (12 percent) respectively. The UNCTAD World Investment Prospects Survey 2013-2015 confirms Germany's reputation as the most attractive business locations in continental Europe. 161 Trans-National Companies ranked Germany first within Europe, and fifth internationally as top destination for 2013-2015. The Ernst & Young study on the attractiveness of the European economic area (2014) also shows that Germany is the number one business location in Europe and the number four worldwide.Competitiveness StudyGermanys Rank in EuropeGermanys Rank Worldwide

Ernst & Young European Attractiveness Survey (2014)14

A.T. Kearney FDI Confidence Index (2014)26

UNCTAD World Investment Prospects Survey 2013-2015 (2013)15

High Scores in Infrastructure and WorkforceIn the above mentioned Ernst & Young study, over 800 international decision-makers were surveyed on Europe and a further 201 companies were asked questions with a specific focus on Germany.Germany received many great marks for various business location factors. These include infrastructure (telecommunications and transport), workforce qualifications and the social climate. 55 percent of those surveyed predict that Germanys attractiveness will continue to grow in the coming years.The study further shows that Germanys growth drivers are in the transportation and automotive industries, environmental and energy technology, as well as the information and communication sector. Surveyed managers expect that in the future most of the investments will go into research and development as well as in manufacturing. It is no surprise that 64 percent of companies surveyed are convinced that Germany is pursuing a policy that attracts international investors.

Ongoing Interest in Greenfield InvestmentsBetween 2008 and 2013, fDi Markets recorded more than 4,700 investment projects in Germany by some 3,900 foreign companies. With 857 greenfield projects, 2012 proved to be the most successful year ever recorded, with Germany placing forth worldwide in terms of FDI projects attracted. In 2013, nearly 800 projects were recorded, making it the third best result ever achieved. FDI results are based on greenfield project announcements (including expansions and joint ventures) collected in the Financial Times Group's fDi Markets database. The most important countries as sources for new investment projects are the USA (24 percent of all investment projects), Switzerland (eight percent), and UK (eight percent), respectively. The ICT & software industry, and business & financial services are leading sectors in attracting new projects. The industrial machinery & equipment, textiles, consumer goods, and chemical industries are likewise attractive foreign investment sectors in Germany. Most new projects open sales and marketing & support offices. One in seven investment project is manufacturing-site located making this a very important business activity in Germany.

R&D FrameworkGerman firms are global leaders in the development of new technologies. For over a century, the Made in Germany label has stood for innovation and excellent quality. Exports of high-tech products in 2009 were worth a total of 102 billion euros. This makes Germany the top exporter in Europe and the second largest in the world.Strong Innovation PerformanceGermany enjoys an excellent reputation regarding its dynamic and innovative R&D environment. This is regularly confirmed by leading international comparisons on innovative capability. The latest example can be found in the currentInnovation Union Scoreboard (IUS), which is used as an instrument by the European Commission for the evaluation of the innovative performance of European nations. The IUS ranks Germany among the group of leading innovators. Within the EU,only Sweden and Denmark are placed ahead of the German R&D location. With its high innovation output, Germany is a first rate location for R&D projects.The 'Innovation Index' is made up from a total of 25 indicators for the measurement of innovation. Along with others, public and private R&D outlay, the education level, and the proportion of patent grants at the international patent offices per million inhabitants are considered. The Innovation Union Scoreboard (IUS) was developed in the scope of the Lisbon strategy and is an instrument of the European Commission. Source: European Innovation Scoreboard (2013) The unique innovative potential of Germany as a research location is also acknowledged by international executives. According to a survey by Ernst & Young, a quarter of the decision-makers asked, think Germany is the most attractive R&D location in the world, ahead of neighbouring Switzerland as well as the USA. A further expression of the 'world class performance' of German R&D departments is provided by the results of the surveys by the American Chamber of Commerce in Germany (AmCham). In these surveys 29 percent of the companies state that they are developing their R&D activities in Germany. For this reason, they are increasingly establishing their own research and development centres that are utilising R&D potentials with national as well as global responsibilities.

Rising R&D ExpenditureIn Germany, enormous sums of money are invested in the development of new technologies and innovations. No other country in Europe invests greater sums in research and development (R&D). Since 2000, Germany R&D expenditures have been continuously rising.In 2012, public and private spending on research projects in Germany amounted to approximately EUR 79.4 billion representing 2.98% of GDP. This share places fourth in Europe, behind the Scandinavian countries (Finland, Sweden, Denmark), but significantly ahead of France and the United Kingdom.This means that Germany is well on track to achieving the 3 percent goal specified by the European Union within the coming years. More than two thirds of the expenditure are accounted for by research intensive private business.

Industry PotentialInvestments in Research and DevelopmentGermany's position as a high-tech country is no mere accident. Companies invest significant sums in order to continually bring innovative products and services to the market. According to the joint initiative of German industry for promoting science and humanities (Stifterverband), economy-related research and development expenditure accounted for EUR 57.4 billion in 2009. In the following, we present a selection of the relevant fields of research in Germany for investors with ambitious R&D goals.Data refers to 2008 (as share of the manufacturing industry). Sectors are defined by industry classification codes WZ 2003. Total R&D expenditures include internal and external expenditures. Source: Stifterverband Wissenschaftsst

Economic ActivityEurope's Economic HubGermany is the largest market in Europe. It constitutes 21 percent of Europe's GDP (EU-28) and is home to 16 percent of the total European Union (EU) population. The German economy is both highly industrialized and diversified with equal focus placed on services and production. GDP(in EUR tn)SHARE OF TOTAL GDP(EU-28)POPULATION(in m)SHARE OF TOTAL POPULATION(EU-28)

Germany2.7421%8116%

France2.0616%6613%

UK1.9014%6413%

Spain1.028%479%

Netherlands0.635%173%

Poland0.393%398%

Czech Republic0.151%112%

Hungary0.981%102%

USA12.65316

Japan3.69128

EU-2813.07506

Eurozone9.60333

Key Driver: The Manufacturing IndustryAlmost 10 percent of Europe's manufacturing companies are German. They generate 30 percent of the EU's gross value added in manufacturing alone. In fact, they represent more than one fifth of all of Germany's value added one of the highest shares in Europe. Increasingly more foreign companies are placing their faith in Germany as an essential location for production sites and are benefiting from the country's excellent business framework and superior productivity rates.

Market ReachGermany's central location in the middle of Europe brings with it a closely tied infrastructure network. All European capital cities can be quickly reached by land, air or water from Germany.

Relations with fellow EU member nations are particularly strong. Customs duty exemptions within the EU allow free trade between the community's 28 member states. By the end of 2014, eighteen member countries had introduced the euro as its common currency which has eliminated exchange rate risks which would otherwise prove an obstruction to bilateral trade.

These close relationships are reflected in the significance of the EU as a foreign trade partner. Nearly two thirds of all German imports and exports come from or remain within the EU. Around three quarters of all foreign direct investments in Germany and more than half of all German direct investments overseas originate from or are made in EU countries.

IMPORTSOFGOODSEXPORTS OF GOODS

Country of OriginEUR bnCountry of Destination EUR bn

1 Netherlands89.11 France100.3

2 China73.62 USA88.4

3 France64.13 UK75.6

4 USA48.54 Netherlands70.9

5 Italy47.55 China67.0

6 UK42.36 Austria56.2

7 Russia40.47Italy53.3

8 Belgium38.98 Switzerland47.3

9 Switzerland38.29 Poland42.4

10 Austria36.810 Belgium42.3

ConclusionsIn this paper we evaluate the German tax reform of 2000 with respect to its eect on inward foreign direct investment. We deal with the identi.cation problem by re- ferring to the rather radical assumption that the aggregate eect of the tax reform is equal to zero. Nevertheless, we .nd signi.cant tax eects. The baseline regres- sion indicates that a reduction in the eective marginal tax rate of 10 percentage points increases net investment by 1 percentage point. Given an investment level of around 7,5% and a pre-reform EMTR of around 52%, the elasticity of invest- ment with respect to eective marginal taxation is approximately equal to 0; 7.In comparison to other empirical studies this estimate is rather at the bottom line, but it should be recalled again that our results are based on an assumed aggregate eect of zero. We employed two distinct approaches to measure the tax impact on inward FDI. The .rst de.nes the pre-reform period (1997-1999) as the control group and the post-reform period (2001-2003) as the treatment group. The second links periodical variations in the investment process to periodical variations in the tax incentive scheme. Both approaches yield virtually the same results.The data con.rm several predictions from standard tax theory. Pro.table .rms on average show a strong and signi.cant tax impact while non-pro.table .rms do not. Firms with a high debt level are less tax-sensitive than .rms with a high equity share. Tax considerations seem to be more relevant in branches in which investment is likely to be cost-driven, like manufacturing, and less so in wholesale trade activities and services, which are supposed to be complements to some foreign production units. Geographic proximity plays a major role. We can show that the tax eect is strong and signi.cant for European investors while American and Asian investments do not show any tax eect. Among Europeans, the investors from countries with a common border with Germany show the strongest tax-sensitivity.Besides those expected results, our data give some new and interesting insights, which deserve further testing and research. First, among the pro.table .rms, only the just-above-zero pro.table .rms show a signi.cant tax eect. Under certain assumptions, this could be interpreted as hint towards liquidity constraints. As a 26 matter of fact, the reduction of the eective tax burden always has an income eect and - under the assumption of imperfect capital market - this may induce .rms which are cash-.ow constrained to increase investment. Analyzing this questionthoroughly is beyond the scope of this paper, though, and would require better data. Second, .rms with a high fraction of non-.nancial assets do react strongly to the tax reform while others do less so. This is surprising because predictions derived by standard tax theory said that those .rms would lose relative to others from the tax reform. Third, our data fail to show signi.cant dierences between the marginal investment choice model and the discrete investment choice model.This may be due to the fact that our data sample is considerably reduced when we compare these two models and that .rms with a higher pro.tability (i.e. those which are candidates to con.rm the discrete investment choice model) do not show any tax impact. In addition, the two tax rate measures under consideration are highly correlated, so we would not expect great dierences in the coe cients or standard errors. The problem is that the tax rate cut is equal for all .rms, so welack an approach to identify the impact of this feature of the tax reform, given that we cleaned the time series from any aggregate impact.