A Service of

Leibniz-Informationszentrum econstor Wirtschaft Leibniz Information Centre Make Your Publications Visible. zbw for Economics

Desai, Mihir A.

Article Capital flows, taxation, and institutional variation

NBER Reporter Online

Provided in Cooperation with: National Bureau of Economic Research (NBER), Cambridge, Mass.

Suggested Citation: Desai, Mihir A. (2008) : Capital flows, taxation, and institutional variation, NBER Reporter Online, National Bureau of Economic Research (NBER), Cambridge, MA, Iss. 3, pp. 6-10

This Version is available at: http://hdl.handle.net/10419/61915

Standard-Nutzungsbedingungen: Terms of use:

Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Documents in EconStor may be saved and copied for your Zwecken und zum Privatgebrauch gespeichert und kopiert werden. personal and scholarly purposes.

Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle You are not to copy documents for public or commercial Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich purposes, to exhibit the documents publicly, to make them machen, vertreiben oder anderweitig nutzen. publicly available on the internet, or to distribute or otherwise use the documents in public. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, If the documents have been made available under an Open gelten abweichend von diesen Nutzungsbedingungen die in der dort Content Licence (especially Creative Commons Licences), you genannten Lizenz gewährten Nutzungsrechte. may exercise further usage rights as specified in the indicated licence. www.econstor.eu all, even thin people get heart attacks, just eases. Furthermore, becoming obese often NBER Working Paper No. 11529, at lower rates than obese people. In eco- imposes financial costs, such as a reduc- August 2005, and Advances In Health nomic jargon, there is a positive exter- tion in wages in some cases. My research Economics And Health Services nality from becoming obese, because the suggests that the vast preponderance of Research, 17 (2007), pp.279–318. benefit from the induced research on these costs is private and paid for by obese 2 J. Bhattacharya and M.K. Bundorf, heart attacks and diabetes will not accrue individuals themselves. Moreover, some “The Incidence of the Healthcare Costs solely to obese individuals. We estimate of the costs that are traditionally identi- of Obesity,” NBER Working Paper No. that this positive externality, which ben- fied as public costs may actually be ben- 11303, May 2005. efits thin people, is at least as large as the efits. Rising obesity rates, for example, 3 I. Rashad and S. Markowitz, transfer induced by Medicare from thin may bring forth greater innovative activ- “Incentives in Obesity and Health to obese people. ity and ultimately treatments that benefit Insurance,” NBER Working Paper No. The conclusions of this research thin people as well. In this setting, poli- 13113, May 2007. agenda cast doubt on the conventional cies such as on junk food may lower 4 J. Bhattacharya and M. Packalen, wisdom about the costs of obesity, social welfare rather than raising it while “Is Medicine an Ivory Tower?” NBER which fails to distinguish between pri- placing substantial burdens on obese indi- Working Paper No. 1382, March 2008. vate and public costs. There is no doubt viduals who already pay a substantial cost 5 J. Bhattacharya J and M. Packalen, that becoming obese imposes substan- for their girth. “The Other Ex-Ante Moral Hazard tial medical costs on the obese individ- in Health” NBER Working Paper No. ual, who is more likely to develop Type 1383, March 2008. II diabetes, heart disease, strokes, and a 1 J. Bhattacharya and N. Sood, “Health number of other unpleasant chronic dis- Insurance and the Obesity Externality,”

Capital Flows, Taxation, and Institutional Variation

Mihir A. Desai*

Tariff reductions, falling transport held by foreigners. Foreign holdings of folio decisions and firms’ operational and costs, and reduced barriers to interna- American stocks increased from $400 bil- financing decisions. tional capital flows have created extensive lion to $2.3 trillion over the last decade, Recent research has advanced our opportunities for multinational firms and while American holdings of foreign stocks understanding of the role of taxation and investors in increasingly integrated global increased from $600 billion to $3 trillion. investor protections on capital flows and markets. For example, the outbound for- In the midst of this rapid integration, patterns of FPI. We also have considered eign direct investment (FDI) position of investors and firms still face systems the causes and consequences of tax avoid- American firms grew at an average annual and investor protections that differ across ance activity; we have established how for- rate of 11 percent to $2.4 trillion from countries, and these differences have the eign and domestic activity interact in order 1982 to 2006 while inbound FDI to the potential to affect major investment and to inform new welfare measures; and we grew to $1.8 trillion. Foreign financing decisions. Governments anxious have elaborated on how investment and portfolio investment (FPI) has grown sim- to attract FDI often consider the use of tax financing decisions by multinational firms ilarly. By 2005, 16 percent of all U.S. long- incentives to lure multinational firms, and reflect the effects of taxes and varying insti- term securities (equity and debt) were governments of FDI source countries — in- tutional regimes. cluding the United States — often won- * Desai is a Research Associate in the der whether their tax treatment of foreign Portfolio Flows NBER’s Programs on income is appropriate. Similarly, investor and Corporate Finance and a Professor protections and the broader institutional Empirical efforts to isolate how taxa- of Business Administration at Harvard environment remain distinctive around the tion influences portfolio choice have pro- Business School. His profile appears later world and may influence investors’ port- duced mixed results. Investigating the in this issue.  NBER Reporter • 2008 Number 3 relationship between cross-sectional dif- ferential changes in the preferences of increase in a foreign country’s investor ferences in marginal tax rates and asset American investors, or investment oppor- protections is associated with a 24 per- holdings is complicated by the incom- tunities, or time trends in investment, or cent increase in U.S. investors’ equity FPI plete nature of most household portfolios any changed behavior towards tax eva- holdings. These results are robust to vari- and the fact that income levels can influ- sion. These results show how FPI, and ous controls, are not evident for debt ence both risk preferences and marginal portfolio decisions more generally, are capital flows, and are confirmed using an tax rates. Efforts to examine how port- influenced by taxation. instrumental variables analysis. The use of folios change in response to tax reforms Corporate taxes and investor protec- FPI to bypass home country taxation of must overcome the possibility that the tions also have the potential to influence multinational firms is also apparent using observed changes reflect endogenous sup- FPI by changing the relative attractive- only portfolio investment responses to ply responses or other general equilibrium ness of FPI and FDI as means of achieving within-country corporate changes effects that may confound the influence of international diversification. Dharmapala in a panel from 1994 to 2005. Investors taxation on portfolio choices. and I analyze whether the composition of appear to alter their portfolio choices sig- Dhammika Dharmapala and I U.S. outbound capital flows reflects efforts nificantly to circumvent home and host attempt to overcome these empirical diffi- to bypass home country tax regimes and country institutional regimes. culties by analyzing a that dif- weak host country investor protections.2 ferentially changed the tax treatment of The potential effects of taxation on FPI Causes and Consequences otherwise similar instruments in a man- result from the interaction between home of ner that is unlikely to have produced any and host country taxes. In particular, the endogenous supply response.1 Specifically, United States taxes multinational firms Changing patterns of multinational we investigate how taxes influence port- legally domiciled here on their worldwide firm activity have drawn attention to the folio choices by exploring the response income. As a consequence of this policy, role of tax havens and their effects on to the distinctive treatment of foreign U.S. investors should prefer FPI as a means neighboring countries. More generally, dividends in the Jobs and Growth Tax of accessing foreign diversification oppor- accounts of rising tax avoidance by firms Relief Reconciliation Act (JGTRRA). tunities, particularly in low-tax countries have generated interest in the effects of JGTRRA lowered the rate where the residual tax imposed by the tax avoidance on economies, tax authori- to 15 percent for American equities and United States will be most burdensome. ties, and investors. International variation extended this tax relief only to foreign In effect, FPI allows investors to avoid in tax systems and the activities of mul- corporations from a subset of countries. any residual tax on investment income tinational firms together have allowed The division of countries into two sepa- earned abroad arising from the worldwide insight into the causes and consequences rate groups was driven by regulatory con- tax regime. Conversely, the absence of the of tax avoidance. cerns and was unrelated to future changes residual U.S. tax should make U.S. equity Typical accounts of in investment opportunities or other reg- FPI sensitive to variations in foreign cor- avoidance characterize such activities as ulatory efforts to change investment in porate tax rates, even after controlling for transfers from the state to investors. This these countries differentially. Given the any effects of corporate taxes on levels of view has been questioned by a line of relatively small share of their stocks held U.S. FDI. As such, the worldwide system inquiry that emphasizes the nature of the by American investors, it is unlikely that of taxing income may vitiate the diversi- agency problem in firms. Such a perspec- supply responses by foreign firms would fication benefits of multinational firms. tive is recommended by the fact that the be large. It is similarly unlikely that the Additionally, concerns about the rights state can be characterized as the largest effects of the reform on U.S. investors’ available to minority investors might tilt minority shareholder in most firms given portfolios would have been offset by cli- them towards accessing those opportu- their claim on pretax cash flows via the entele effects in asset markets. JGTRRA nities via investments in U.S. multina- corporate tax system. Alexander Dyck, applied only to U.S. investor returns, leav- tional firms that globally undertake FDI, Luigi Zingales, and I analyze the inter- ing non-U.S. investor tax rates and asset to ensure that investor interests are better action between corporate taxes and cor- demands unaffected. protected. porate governance.3 We show that the This paper uses a difference-in-differ- Our cross-country analysis indicates characteristics of a taxation system affect ence analysis that compares U.S. equity that a 10 percent decrease in a foreign the extraction of private benefits by com- holdings in affected and unaffected country’s corporate tax rate increases U.S. pany insiders. A higher tax rate increases countries. The international investment investors’ equity FPI holdings by 21 per- the amount of income that insiders divert responses to JGTRRA were substantial, cent, controlling for effects on FDI. This and thus worsens governance outcomes. implying an elasticity of asset holdings suggests that the residual tax on foreign In contrast, stronger tax enforcement with respect to taxes of -1.6. This effect multinational firm earnings biases capi- reduces diversion and, in so doing, can cannot be explained by several poten- tal flows to low corporate tax countries raise the stock market value of a company tial alternative hypotheses, including dif- toward FPI. A single-standard-deviation in spite of the increase in the tax burden,

NBER Reporter • 2008 Number 3 7 as evidenced by patterns from the Russian tax gap in a related paper.6 the role of nearby tax havens in diverting stock market. We also show that the cor- Aside from these interactions with economic activity, these results indicate porate governance system affects the level the agency problem, tax havens are also that the opposite may well be the case, of tax revenues and the sensitivity of tax of interest because they may have effects as the ability to reduce tax obligations revenues to tax changes. When the corpo- on tax revenues and real activity. C. Fritz through judicious use of opera- rate governance system is ineffective (that Foley, James R. Hines, and I examine tions may stimulate greater investment in is, when it is easy to divert income), an what types of firms establish tax haven their high-tax neighbors. increase in the tax rate can reduce tax rev- operations and what purposes these oper- enues. We test this prediction in a panel ations serve.7 Analysis of affiliate-level Domestic and Foreign of countries. Consistent with the model, data for American firms indicates that Investment Interactions we find that corporate tax rate increases larger, more international firms, and those have smaller (in fact, negative) effects on with extensive intra-firm and high There is considerable debate over revenues when corporate governance is R and D intensities, are the most likely the likely domestic effects of the rapidly weaker. to use tax havens. Tax haven operations increasing foreign activity by U.S. multi- Dharmapala and I examine whether facilitate tax avoidance both by permit- national firms. In particular, FDI flows these interactions are relevant for ting firms to allocate away to rapidly growing foreign markets gen- American firms, particularly those that from high-tax jurisdictions and by reduc- erate fears that such investment displaces undertake activity in tax havens.4 We test ing the burden of home country taxation domestic employment, capital invest- alternative theories of corporate tax avoid- of foreign income. The evidence suggests ment, and . An alternative ance that yield distinct predictions on the that the primary use of affiliates in larger perspective suggests that growing foreign valuation of corporate tax avoidance. We tax haven countries is to reallocate taxable investment may instead increase levels of then use unexplained differences between income, whereas the primary use of affili- domestic activity by improving the prof- income reported to capital markets and to ates in smaller tax haven countries is to itability and competitiveness of domes- tax authorities to proxy for tax avoidance facilitate deferral of U.S. taxation of for- tic operations as firms expand globally. activity. These “book-tax” gaps are larger eign income. Very little empirical evidence is currently when firms are alleged to be involved in U.S. multinational firms are also more available with which to distinguish these tax shelters. OLS estimates indicate that likely to establish new tax haven opera- views. The absence of evidence in this the average effect of tax avoidance on firm tions if their non-haven investments are domain is particularly troubling because value is not significantly different from growing rapidly, which generally confirms a central motivation for of for- zero, but is positive for well-governed the notion that greater foreign investment eign income has been “capital neu- firms as predicted by an agency perspec- increases the potential return to using tax trality,” a notion in part predicated on the tive on corporate tax avoidance. havens. The analysis shows that 1 per- idea that outbound FDI represents lost We use an exogenous change in tax cent greater sales and investment growth investment. regulations that affected the ability of in nearby non-haven countries is associ- Foley, Hines, and I report time-series some firms to avoid taxes abroad — the ated with a 1.5 to 2 percent greater like- evidence that aggregate foreign and do– onset of so-called “check the box” reg- lihood of establishing a tax haven opera- mestic investment are positively corre- ulations — to construct instruments for tion. This evidence also suggests that tax lated for the United States.8 Such aggre- tax avoidance activity. The IV estimates havens may serve to increase economic gate evidence is open to many alternative yield larger overall effects and reinforce activity in nearby high-tax countries. Tax explanations. In one paper9 we expand on the basic result that higher quality firm havens serve this function by indirectly this line of inquiry by using firm-level evi- governance leads to a larger effect of tax reducing tax burdens on income earned dence and an instrumental variables strat- avoidance on firm value. The results are in high-tax countries, and by attracting egy to overcome identification difficulties robust to a wide variety of tests for alter- investment that may enhance the profit- in this setting. native explanations. Taken together, the ability of operations in those countries. Firms whose foreign operations grow results suggest that the simple view of Proximity allows firms to split up produc- rapidly exhibit coincident rapid growth corporate tax avoidance as a transfer of tion processes and increases the extent of domestic operations, but this pattern resources from the state to sharehold- to which firms can avoid taxes through alone is similarly problematic, as foreign ers is incomplete given the agency prob- . Evidence that firms with and domestic business activities are jointly lems characterizing shareholder-manager extensive nearby investments find it prof- determined. We use foreign GDP growth relations. This paper builds on previous itable to establish tax haven operations rates, interacted with lagged firm-specific work5 that develops this measure of cor- likewise implies that the availability of tax geographic distributions of foreign invest- porate tax avoidance and examines why haven opportunities increases the attrac- ment, to predict changes in foreign invest- managers undertake tax shelters. We dis- tiveness of investments in high-tax loca- ment by a large panel of U.S. manufactur- cuss the broader importance of the book- tions. While it is common to worry about ing firms. Estimates produced using this

8 NBER Reporter • 2008 Number 3 instrument for changes in foreign activity investor protections influence the cross- fact, affiliates experience 6.9 percent faster indicate that 10 percent greater foreign border operational, financing, and invest- annual growth of property, plant, and capital investment is associated with 2.2 ment decisions of firms.10 We develop a equipment investment subsequent to the percent greater domestic investment, and model in which product developers have a liberalization of controls, indicating that that 10 percent greater foreign employee comparative advantage in monitoring the capital controls impose significant bur- compensation is associated with 4 per- deployment of their technology abroad. dens on foreign investors. cent greater domestic employee compen- We demonstrate that when firms want to International variations in political sation. Changes in foreign and domestic exploit technologies abroad, multinational risk also can influence financing decisions sales, assets, and numbers of employees firm activity and FDI flows arise endoge- of multinational firms. Foley, Hines, and I are likewise positively associated. Foreign nously when monitoring is not verifiable demonstrate that American multinational investment also has positive estimated and financial frictions exist. The mech- firms respond to politically risky environ- effects on domestic and R and D anism generating MNC activity is not ments by adjusting their capital structures spending, suggesting that growth-driven the risk of technological expropriation by abroad and at home.12 Foreign subsid- foreign expansions stimulate demand for local partners but rather the demands of iaries located in politically risky coun- tangible and intangible domestic output. external funders who require MNC par- tries have significantly more debt than do These results do not support the popu- ticipation to ensure value maximization other foreign affiliates of the same parent lar notion that greater foreign activity by local entrepreneurs. The model dem- companies. American firms further limit crowds out domestic activity by the same onstrates that weak investor protections their equity exposures in politically risky firms, instead suggesting that the reverse is limit the scale of multinational firm activ- countries by sharing ownership with local true of foreign activity. ity, increase the reliance on FDI flows, partners and by serving foreign markets These findings further lend support and alter the decision to deploy technol- with exports rather than local production. to an alternative welfare metric for assess- ogy through FDI as opposed to arm’s The residual political risk borne by parent ing the appropriate tax policy for foreign length licensing. Using firm-level data, we companies leads them to use less domes- profits. The traditional welfare metric of test and confirm several distinctive pre- tic leverage, resulting in lower firm-wide capital export neutrality is predicated on dictions for the impact of weak investor leverage. Multinational firms with above- the substitutability of foreign and domes- protection on MNC activity and FDI average exposures to politically risky coun- tic activity and recommends the taxation flows. tries have 8.4 percent less domestic lever- of worldwide income with credits for for- Foley, Hines, and I examine the role age than do other firms. These findings eign taxes paid. An alternative welfare of capital controls in influencing local illustrate the broader impact of risk expo- benchmark, capital ownership neutrality, borrowing rates and patterns of invest- sures on capital structure. has been developed recently, emphasizing ment, financing, and transfer pricing.11 Foley, Kristin J. Forbes, and I study distortions to ownership decisions and Borrowing rates are 5.25 percentage points the effects of financial constraints on lost productivity in a setting where substi- higher in countries imposing capital con- firm growth by investigating whether tutability may not be complete, as sug- trols than they are elsewhere for affili- large depreciations differentially affect gested by these findings. The capital own- ates of the same multinational parents. multinational affiliates and local firms in ership neutrality benchmark recommends Multinational firms distort their reported emerging markets.13 U.S. multinational exemption of foreign income for national profitability and their dividend repatria- affiliates increase sales, assets, and invest- and global welfare maximization. tions in order to mitigate the impact of ment significantly more than local firms capital controls. Affiliates have 5.2 per- both during and after currency crises. The Nature of Multinational cent lower reported profit rates than com- The enhanced relative performance of Firm Activity parable affiliates in countries without cap- multinationals is traced to their ability ital controls, reflecting, in part, trade and to use internal capital markets to capi- Analysis of microdata on American financing practices that reallocate income talize on the competitiveness benefits of multinational firms collected by the within a firm. The distortions to reported large depreciations. Investment specifi- Bureau of Economic Analysis has allowed profitability are comparable to those that cations indicate that increases in lever- for new insights into how patterns of FDI stem from a 27 percent difference in cor- age resulting from sharp depreciations are shaped by variations in investor pro- porate tax rates. Dividend repatriations constrain local firms from capitalizing tections, political risk, capital controls, are also regularized to facilitate the extrac- on these investment opportunities, but and currency crises. These studies also tion of profits from countries imposing do not constrain multinational affiliates. shed light on how these varying institu- capital controls. If capital controls impose Multinational parents also infuse new tions influence local firms and how taxes costs through higher interest rates and capital in their affiliates after currency shape multinational firm decisionmaking. the distortions associated with avoid- crises. These results indicate another role Pol Antràs, Foley, and I examine how ance, then liberalizations of capital con- for foreign direct investment in emerging costly financial contracting and weak trols should have significant effects. In markets as multinational affiliates expand

NBER Reporter • 2008 Number 3 9 economic activity during currency crises 5 M. Desai and D. Dharmapala, 14 M. Desai, C. Foley, and J. Hines, “A when local firms are most constrained. “Corporate Tax Avoidance and High Multinational Perspective on Capital The role of taxes in shaping financial Powered Incentives,” NBER Working Structure Choice and Internal Capital and operating decisions has also been a Paper No. 10471, May 2004, published in Markets,” NBER Working Paper No. 9715, prominent feature of these studies. The Journal of Financial Economics. May 2003, published in Journal of Finance behavior of U.S. multinational firms con- 6 M. Desai, “The Corporate Profit Base, 59, no.  (December 2004), pp. 2451–88. sistently demonstrates that taxes play crit- Tax Sheltering Activity, and the Changing 15 M. Desai, C. Foley, and J. Hines, ical roles in shaping the volume and loca- Nature of Employee Compensation,” NBER “Dividend Policy Inside the Firm,” NBER tion of foreign investment, the financing Working Paper No. 88, March 2002, Working Paper No. 898, January 2002, of foreign investment, and the organiza- published as “The Divergence Between Book published in Financial Management 3, tional structures of multinational firms. Income and Tax Income,” in Tax Policy and no. 1 (2007), and “Repatriation Taxes and The papers also capitalize on the inter- the Economy 17 (2003), pp. 19–20. Dividend Distortions,” NBER Working national variation faced by multinational 7 M. Desai, C. Foley, and J. Hines, Paper No. 8507, October 2001, pub- firms to provide estimates of how taxes “Economic Effects of Regional Tax Havens,” lished in National Tax Journal 54, no. 4 influence financial and investment deci- NBER Working Paper No. 1080, October (December 2001). sions more generally. 2004, published as “The Demand for Tax 16 M. Desai, C. Foley, and J. Hines, For example, Foley, Hines, and I show Haven Operations,” in Journal of Public “Foreign Direct Investment in a World of that capital structure and internal capital Economics 90, no. 3 (March 200), pp. Multiple Taxes,” NBER Working Paper No. allocations decisions respond significantly 513–31, and as “Do Tax Havens Divert 8440, August 2001, published in Journal of to tax differentials.14 While other studies Economic Activity?” in Economic Letters Public Economics, 88, no. 12 (December have not found significant effects, the set- 90, no. 2 (February 200), pp. 219–24. 2004), pp. 2727–44. ting of a multinational firm facing multi- 8 M. Desai, C. Foley, and J. Hines, 17 M. Desai, C. Foley, and J. Hines, ple tax regimes provides a cleaner setting “Foreign Direct Investment and the “Chains of Ownership, Regional Tax for considering this question. Similarly, Domestic Capital Stock,” NBER Working Competition, and Foreign Direct we have explored the role of tax and non- Paper No. 11075, January 2005, published Investment,” NBER Working Paper No. tax factors in dividend policy by looking in American Economic Review 95, no. 2 9224, September 2002, published in at multinational firm repatriations.15 We (May 2005), pp. 33–8. Heinz Herrmann and Robert Lipsey (ed.), have also studied the sensitivity of invest- 9 M. Desai, C. Foley, and J. Hines, Foreign Direct Investment in the Real and ment to income and differen- “Foreign Direct Investment and Domestic Financial Sector of Industrial Countries tials.16 We have examined ownership and Economic Activity,” NBER Working Paper (Heidelberg: Springer-Verlag, 2003), organizational form decisions.17 Finally, No. 11717, October 2005; forthcoming in pp. 1–98, and “The Costs of Shared the incidence of export subsidies was the American Economic Journal: Economic Ownership: Evidence from International motivation behind our investigation of Policy. Joint Ventures,” NBER Working Paper No. the effects of WTO complaints against 10 P. Antras, M. Desai, and C. Foley, 9115, August 2002, published in Journal incentives for exports.18 “Multinational Firms, FDI Flows and of Financial Economics 73, no. 2 (August Imperfect Capital Markets,” NBER 2004), pp. 323–74; M. Desai and J. Hines, 1 M. Desai and D. Dharmapala, “Taxes, Working Paper No. 12855, January 2007. “Expectations and Expatriations: Tracing Dividends, and International Portfolio 11 M. Desai, C. Foley, and J. Hines, the Causes and Consequences of Corporate Choice,” NBER Working Paper No. 13281, “Capital Controls, Liberalizations, and Inversions,” NBER Working Paper No. July 2007. Foreign Direct Investment,” NBER 9057, July 2002, published in National Tax 2 M. Desai and D. Dharmapala, “Taxes, Working Paper No. 10337, March 2004, Journal 55, no. 3 (September 2002), pp. Institutions, and Foreign Diversification published in Review of Financial Studies 409–40, and “ ‘Basket’ Cases: International Opportunities,” NBER Working Paper No. 19, no. 4 (200), pp. 1399–1431. Joint Ventures After the Tax Reform Act of 13132, May 2007. 12 M. Desai, C. Foley, and J. Hines, 198,” published as, “ ‘Basket’ Cases: Tax 3 M. Desai, A Dyck, and L. Zingales, “Capital Structure with Risky Foreign Incentives and International Joint Venture “Theft and Taxes,” NBER Working Paper Investment,” NBER Working Paper No. Participation by American Multinational No. 10978, December 2004, published in 1227, June 2005; forthcoming in Journal Firms,” in Journal of Public Economics 71 Journal of Financial Economics 84, no.3, of Financial Economics. (March 1999), pp. 379–402. (June 2007), pp.591–23. 13 M. Desai, C. Foley, and K. Forbes, 18 M. Desai and J. Hines, “Market 4 M. Desai and D. Dharmapala, “Financial Constraints and Growth: Reactions to Export Subsidies,” NBER “Corporate Tax Avoidance and Firm Multinational and Local Firm Responses Working Paper No. 10233, January 2004, Value,” NBER Working Paper No. 11241, to Currency Crisis,” NBER Working Paper published in Journal of International April 2005; forthcoming in Review of No. 10545, June 2004; forthcoming in Economics 74, no. 2 (March 2008). Economics and Statistics. Journal of Financial Economics.

10 NBER Reporter • 2008 Number 3