What Should We Talk About When Talking About Tax Incentives for FDI?
Associate Professor at the Allard School of Law, University of British Columbia (Canada). Member of the Permanent Scientific Committee of the International Fiscal Association. firstname.lastname@example.org
Abstract. 1. Introduction. 2. Do Developing Countries Try to Become Tax Havens?3. Do Developing Countries Offer Tax Incentives Due to Tax Competition? 4. Conclusion
The policy of adopting tax incentives to attract foreign direct investment (FDI) 1 has been a favorite topic for both scholars interested in tax and development and tax policy advisors at international organizations. The following statement by two leading authors captures the standard conception of the topic as well as of the potential contribution of the tax specialist: “[although] standard international tax policy advice cautions against the use of tax incentives for investment, many developing and transition countries…continue to operate or introduce them. Accordingly, [we] briefly [outline] the reasons why such incentives are often found to be unsuccessful and what the more important issues may be for encouraging investment in developing and transition countries. [We] then [consider] in more detail the design, drafting, and international taxation issues that such incentives present.” 2 The approach rests on the assumption that developing countries display much greater enthusiasm for tax incentives than such policy deserves, and delineates the role of tax policy advisors and tax scholars as one of explaining why such enthusiasm might be misplaced, and guiding the estray back to the normal path of tax policy design. 3
Fortunately, in this forum examining “International Tax Principles in BRIC and OECD Countries”, there seems little need to engage in such a didactic exercise. The conference presentations suggest that Russia, Brazil, India and China already do not pursue national policies of using tax preferences to attract FDI generally. And more broadly, the popularity of FDI-targeted incentives may have been in decline not only in middle-income countries but also in a wider range of developing countries. 4 This is not to say that various tax incentives having a bearing on FDI and that may be objectionable from the perspective of ideal policy design have disappeared from these countries (just as they have not in developed countries), or have diminished in practical importance. However, such incentives may not be well documented or easily accessible to outsiders, may take varied or complex forms that elude understanding that is based merely on cursory reviews, and their political origins may be even more opaque than older forms of tax incentives. In the face of this growing specificity of developing countries’ experiences, a question may be raised as to what should be the appropriate focus for future international dialogues about this subject.
In this short commentary, I would like to suggest that for “tax incentives in developing countries” to continue as a productive area of intellectual and policy discourse, one must take the relevant tax policies of developing countries more in their own terms. That is, the types of tax incentives actually adopted in practice must be more carefully classified and documented. And the political processes that lead to the adoption of tax incentives may need to become a subject of study itself. I will elaborate on these suggestions by reference to some recent work by both economists and legal scholars on tax incentives in developing countries. This recent work, I argue, may not have depicted developing country practices accurately, leading to conclusions that are both theoretically and empirically unsatisfying.
Do Developing Countries Try to Become Tax Havens?
The relationship between tax rates and the quantity of investment is, both theoretically and in commonsense, a ceteris paribus one: all other things equal, lower tax rates, by increasing the after-tax return, should increase the quantity of investment. When all other things are not equal, then one cannot predict investment decisions based on tax rates alone. This, I believe, is a truism. What lies beyond commonsense—and also, it seems, investigations by economists to date, if we assume that there is a truth of the matter here amenable to empirical investigation—is what exactly the determinants are of investment decisions and what precise weights they have. When the OECD makes the statements that “a low host country tax burden cannot compensate for a generally weak or unattractive FDI environment… Tax is but one element and cannot compensate for weak non-tax conditions,” 5 or when World Bank economists claim that “tax incentives are a poor instrument for compensating for negative factors in a country’s investment climate,” 6 we do not obtain much information beyond the above truism. How “weak”, “unattractive”, or “negative” must a country’s investment climate be so that no amount of tax incentives would serve as compensation is not stated. And because such statements and claims are quite weak, it is also questionable what information we gain from empirical evidence (especially weak empirical evidence) marshalled to support them.
To illustrate the same point slightly differently, consider how controversial the claim is that for multinational companies, the tax burden on investment is not the only, or even the most important, factor for deciding whether to operate in a given country. In most circumstances, this claim should not be controversial. Thus if business surveys of multinationals are conducted and most responses state that tax incentives are not the most critical factor for investment, 7 most people who have thought about the matter probably will not be surprised. On the other hand, such surveys have shown that exporters, unlike many other types of businesses, may be highly responsive to tax incentives. 8 This may be the more interesting and informative finding, and one can imagine policymakers in developing countries being keen to learn more about it: how sensitive are exporters, and what other businesses may also be responsive to tax incentives? If commonsense suggests that the null hypothesis should be that tax incentives are not the most critical factor, then the tax policy specialist who merely reminds us of this null hypothesis contributes very little insight.
The foregoing is not meant just to state personal theoretical preferences. It is not clear that policymakers from developing countries, in their eagerness to spur economic growth, have, at the present or in recent decades, ignored what (I believe is) commonsense and adopted the assumption that tax burden is generally the most critical factor for investment decisions. I am not sure that evidence has often been presented, aside from the proliferation of tax incentives in the developing world, that policymakers there have adopted “simplistic” theories of the determinants of FDI or economic growth 9 and displayed blind faith in the power of tax incentives. On the contrary, it is quite plausible that they think tax incentives must be well designed in order to work, and developing such designs is difficult (hence the decision sometimes to seek advice from international experts). 10 Nonetheless, critiques of tax incentives in developing countries tend to be heavy on textbook generalities (such as that tax incentives are justified only in the presence of market failures), and light on designs in specific contexts.
An important case in point is the wide use of industrial policy in connection with tax incentives. Many developing countries, including India and China, have had longer histories exploring various versions of industrial policy than experimenting with tax incentives, and when tax incentives were introduced, they tended to be coordinated with industrial policy. Only some forms of FDI are viewed as capable of generating sufficient spillovers to warrant special tax treatment. In China, for example, before FDI tax preference were repealed in 2007, not only were they limited explicitly in the tax law to specific sectors, 11 but FDI in general was subject to an investment control regime that implemented industrial policy. All tax preferences for FDI were thus implicitly targeted because most permitted FDI had already been screened by other regulatory agencies. 12 Such explicit or implicit targeting could have allowed tax incentives to be limited to those sectors more sensitive to tax rates. They could also have allowed complementary institutions (for example, special intellectual property, maritime, or other tribunals or arbitral forums) to improve the “investment climate” and enhance the return to the host country of offering tax incentives.
It is possible that many such attempts to properly target tax incentives have not been successful. On the other hand, economists from developed countries may have traditionally disfavored industrial policy, and thus are not interested in any type of tax policy design tailored to industrial policy. They are skeptical and lack expertise at the same time. But the important point is that industrial policy is anything but simplistic: 13 it just falls outside the ambit of orthodox public finance. The dismissal of developing countries’ naiveté may sometimes reflect an indifference to those countries’ complex ambitions.
The lack of interest in particular attempts by developing countries to fashion effective tax incentives can sometimes lead to glaring mistakes. Consider a recent study by Van Parys and James that purports to be of relevance for developing countries, examining the effect of investment climate on the relation between tax rates and FDI. 14 They start with the very sensible observation that a satisfactory “investment climate” is important to ensuring adequate return to capital, and that therefore the degree of responsiveness of capital to tax depends on the “investment climate”. They further theorize that because higher rates may generate greater revenue, and because greater tax revenue would allow governments to improve governance quality, a higher tax rate may lead to greater FDI, not lower, FDI, when the investment climate is very poor. This is a controversial claim—because developing countries are known to have high distortionary taxes on narrow tax bases, 15 and because there is no strong reason to believe that governance quality features such as the rule of law are functions of tax revenue—but the authors claim to find empirical support for it. The measure of tax rates they use, however, is national marginal effective tax rates (METR), 16 which may or may not be the tax rates applied by the countries studied to FDI. For example, the high METR associated with China in this measure precisely did not apply to FDI, and the constant METR measure for China during 2005-2008 17 completely fails to reflect the decisive policy change in China in 2007 to abandon FDI tax incentives. Depending how pervasive this problem is of associating the wrong measure of tax rates with FDI, 18 Van Parys and James’ empirical finding that in poor investment climate countries, high tax rates are be positively correlated with FDI, may be spurious.
To sum up the arguments above, it may be observed that to attribute to developing countries an eagerness to attract foreign investment simply by lowering tax rates without pursuing other related developmental policy is similar to saying that developing countries try to become tax havens. Interestingly, another recent study shows that the tax haven countries of the world generally possess good governance characteristics, 19 and the authors of the study view this as complementary to the fact that “poorly-governed countries do not generally attempt to become tax havens.” 20 But if it is a fact that developing countries with poor governance do not try to become tax havens, efforts to show that such countries should not try to become tax havens are in one obvious sense misplaced. Nonetheless, commentaries on tax incentives in developing countries often seem content to attack such a straw man.
Do Developing Countries Offer Tax Incentives Due to Tax Competition?
Another piece of received wisdom about tax incentives in developing countries is that they are offered in response to tax competition. 21 This is sometimes intended as expressing a more sympathetic view than implied by other advice of the predicament faced by policymakers in developing countries: they adopt tax incentives not because of naïve beliefs about the relationship between tax burden and investment, but because they are locked into a tragic race to the bottom for mobile capital.
The question can be raised, however, whether this is mere lore or whether it can be backed up by real evidence. Empirical support for international tax competition is actually scant. In a very sophisticated study of OECD countries, 22 Devereux, Lockwood, and Redoano find that these countries competed over statutory tax rates to attract mobile corporate profit, but that evidence is much weaker for competition over effective marginal tax rates to attract capital. Moreover, even evidence for tax competition over statutory rates is found only among countries that do not impose capital controls. These authors further estimate that had there been no relaxation of capital controls among OECD countries, there would have been no fall in average tax rates in them. These carefully calibrated findings do not bode well for casual hypotheses that developing countries are locked into tax competition. Such hypotheses usually conceive developing countries as competing for capital, not profit. And they ignore the fact that developing countries are much more likely to maintain capital control (including long-term capital control, as in India and China) than developed countries.
More recently, Klemm and Van Parys have attempted to study tax competition among developing countries, by examining corporate income tax rates, tax holidays and investment allowances and credits in Latin American, African, and Caribbean countries. 23 They were not able to establish strategic interactions over these variables among the countries, due to the fact that the variables displayed simultaneous decline in most countries studied during the relevant period. Moreover, their empirical specification did not allow discrimination among tax competition and mere tax mimicking (which does not require capital flow among different jurisdictions). 24 They do find, however, that lower corporate income tax rates and longer tax holidays were effective in attracting FDI.
These cross-country studies buttress skepticism at least for me that talks of international tax competition among developing countries are at least somewhat fanciful and removed from political realities in these countries. If one is truly interested in why developing countries are keen to adopt tax incentives (if they are), plenty of other explanations should be considered before international tax competition. For example, for large developing economies like the BRIC countries, tax competition among domestic, subnational jurisdictions may offer sufficient explanations. Even if there are no good empirical studies of subnational tax competition in these countries, 25 merely as a theoretical matter, internal factor mobility within these countries ought to matter more for tax policy than international factor mobility.
Let me elaborate on this point further in connection with China. Even though China abandoned the national policy of offering tax incentives to FDI after 2007, local tax incentives are still routinely offered to foreign investors. 26 These incentives usually take the form of various subnational (usually municipal- or county-level) governments rebating their shares of income and other tax revenue to taxpayers. Because many such incentives are arguably in violation of national tax law, 27 local governments prefer to avoid controversy by not publicizing them—but offering them through negotiated agreements with businesses that come. The pattern of local tax incentives is thus hard to establish empirically, even though negotiating and holding local governments to the terms of these incentives is an important part of tax management for foreign multinationals.
What are the incentives of Chinese local government officials in offering rebates of tax paid to foreign investors? There is now a sizeable political economy literature that Chinese politicians are engaged in “yardstick competition” established within the Communist Party: officials compete for promotional opportunities by generating economic growth within their short tenures in governing positions. 28 Because economic growth is often equated within this political competition with crude measures such as GDP or fixed capital investment and not with broader measures such as employment and income, it is generally believed that Chinese local officials are prone to pursue economic growth at all costs. This political incentive structure has been used explain government behavior in a wide range of areas, including the management of local land sales, the implementation of environmental policy, etc. Thus it comes as no surprise that despite efforts by the national government to limit local governments’ capacity to engage in tax competition, unauthorized tax incentives thrive.
Are these local tax incentives effective at attracting FDI? Well-designed empirical studies of this question are still few in number. One such study, 29 based on firm-level data on tax burdens for the years 2002-4 (and thus incorporating neither national nor local tax policy information), found that average firm tax burdens (including relative burdens of foreign firms to domestic firms) in a locality had no significant impact on the volume of FDI in that locality. However, the same study finds that having a young governing politician who is in the relevant office for a relatively short tenure contributes positively and significantly to FDI, which may be interpreted as consistent with the political yardstick competition hypothesis.
The BRIC countries highlight the heterogeneity of developing countries: they are at the same time so different from one another and from other developing countries. International tax policy advisors are inevitably challenged in making generalizations about them, as are academics. Perhaps this is why, in commentaries on tax incentives deployed by developing countries, we so often read passages that seem lifted from undergraduate textbooks. But in this commentary I have tried to suggest that we not confuse this discourse with the real issues faced by policymakers in these countries. Let us avoid the “ignorance hypothesis” 30 that developing country policymakers adopt simplistic theories about tax incentives, or the assumption that we already know what drive their judgment to adopt such incentives. Their real ambitions—as well as their shrewdness—may be more fitting subjects for our study.
- Paper written for the conference on International Tax Principles in BRICS and OECD Countries: Divergences and Convergences. ↩
- David Holland and Richard J. Vann, “Chapter 23: income tax incentives for investment,” in Tax Law Design and Drafting (volume 2; International Monetary Fund: 1998; Victor Thuronyi, ed.) ↩
- A similar framing of the topic is found in Howell Zee, Janet Stotsky, and Eduardo Ley, “Tax incentives for business investment: a primer for policy makers in developing countries,” World Development, Vol. 30(9), pp. 1497–1516 (2002). ↩
- See, e.g. Alexander Klemm and Stefan Van Parys, “Empirical evidence on the effects of tax incentives,” Int Tax Public Finance (2012) 19:393–423 (finding, based on PricewaterhouseCoopers summaries of world corporate tax regimes, that the average intensity of tax incentives has declined between 1985 and 2005 in a panel of countries from Latin America, Africa, and the Caribbean). ↩
- OECD, “Tax effects on foreign direct investment, recent evidence and policy analysis”. OECD Tax Policy Studies, No.17 (2007), quoted in Stefan Van Parys and Sebastian James, “Why tax incentives may be an ineffective tool to encouraging investment? – the role of investment climate,” working paper available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1568296. ↩
- Jacques Morisset, “Public policy for the private sector: tax incentive,” World Bank Note, No. 253 (2003), quoted in Van Parys and James, supra note 3. ↩
- Sebastian James, “Tax and non-tax incentives and investments: evidence and policy implications” (2009, available at https://www.wbginvestmentclimate.org/uploads/IncentivesandInvestments.pdf) ↩
- Id. ↩
- For a recent assertion that this is the case, see Yariv Brauner, “The future of tax incentives for developing countries,” Chapter 2 in Y. Brauner and M. Stewart (Eds.), Tax Law and Development (Edward Elgar, 2013), pp 159-181. ↩
- And even most international tax policy experts would not deny that tax incentives can ever be designed to work effectively. See Alexander Klemm, “Causes, benefits and risks of tax incentives,” IMF Working Paper 09/21 (Washington: International Monetary Fund, 2009). ↩
- See, generally, Jinyan Li, “The rise and fall of Chinese tax incentives and implications for international tax debates,” 8 FLA. TAX REV. 669 (2007) ↩
- Wei Cui, “’Establishment’: an analysis of a core concept in Chinese inbound income taxation,” 1 Columbia Journal of Tax Law 46 (2010), at 81. Thus regulatory law other than tax law must be taken into account to determine whether tax incentives are targeted, raising an important methodological challenge for empirical investigations of the effectiveness of tax incentives for FDI. ↩
- For recent theory and evidence that industrial policy may play an important role in middle-income countries, see e.g. P. Aghion, M. Dewatripont, L. Du, A. Harrison & P. Legros, “Industrial policy and competition,” NBER Working Paper 18048 (May 2012); L. Du, A. Harrison and G. Jefferson, “Testing for horizontal and vertical foreign investment spillovers in China, 1998–2007,” Journal of Asian Economics, 23(3), pp 234–243 (2012). ↩
- Van Parys and James, supra note 3. ↩
- Roger Gordon and Wei Li, “Tax structures in developing countries: many puzzles and a possible explanation”, Journal of Public Economics, 93(7-8), pp 855-866 (2009). It is thus either unlikely correct, or not meaningful, to say that these countries are on “the rising side of the Laffer curve”. ↩
- As computed in Duanjie Chen and Jack Mintz, “Taxing business investments: a new ranking of effective tax rates on capital,” FIAS World Bank, Washington, DC (2008). ↩
- Id. Table 1. ↩
- Notably, the BRIC countries are shown in the Chen and Mintz study as among the top seven countries in terms of high METR. Some of them also rank low in some of the regulatory quality indices used by Van Parys and James, such as the World Bank Doing Business indicators. ↩
- Dhammika Dharmapala and James R. Hines Jr., “Which countries become tax havens?” Journal of Public Economics 93 (2009) 1058–1068. ↩
- Id. 1058. ↩
- See, e.g. Zee et al, supra note 2; Brauner, supra note 8. ↩
- Michael Devereux, Ben Lockwood, and Michela Redoano, “Do countries compete over corporate tax rates?” Journal of Public Economics, Vol. 92, pp. 1210–35 (2009) ↩
- Klemm and Van Parys, supra note 3. ↩
- The Devereux, Lockwood, and Redoano study both controls for time trend and tests different hypotheses about the nature of strategic interactions among countries. ↩
- Based on my own research, such studies are hard to find for China. ↩
- See Wei Cui, “Fiscal Federalism in Chinese Taxation,” World Tax Journal 2011(3), pp 455-80, section 5. Putting a stop to rampant local tax incentives was named a top policy item in the public finance agenda announced in the recent Third Plenum of Eighteenth Congress of the Chinese Communist Party. ↩
- Id. ↩
- See Hongbin Li and Li-An Zhou, “Political Turnover and Economic Performance : the Incentive Role of Personnel Control in China”, Journal of Public Economics , 89 (9-10), 1743-62 (2005). ↩
- Xiaozu Wang, Lixin Xu and Tian Zhu, “Foreign direct investment under weak rule of law: theory and evidence from China,” Economics of Transition 20 (3), 401–424 (2012). ↩
- James A. Robinson and Daron Acemoğlu, Why Nations Fail (Harvard University Press, 2012), Chapter 2. ↩