Abstract
The recent boom of investor-state disputes filed under international investment agreements has fueled a controversial academic and policy debate. We study the impact of these compensation claims on foreign direct investment (FDI) flows to the responding host country. Our econometric analysis focuses on differences in the FDI response from BIT-partner and non-partner countries of developing host countries. This approach allows us not only to distinguish competing hypotheses about BIT function, but also to address endogeneity concerns in earlier studies. We find that BITs stimulate bilateral FDI flows from partner countries—but only so long as the developing host country has not had a claim brought against it to arbitration. Our results provide an additional explanation for the policy-changes observed in many states subsequent to their first experience of an investor-state dispute.
Similar content being viewed by others
Notes
Note that we use the terms ‘disputes’ and ‘claims’ interchangeably in the following to indicate instances in which an investor brings a claim for compensation for alleged violation of a treaty’s provisions to formal ISDS.
Traditionally, relatively few claims were raised against respondent states with high per-capita income. UNCTAD’s database on ISDS lists just 78 claims against high income states up to 2010, accounting for less than 20% of all claims until recently. However, this share increased to about 30% in subsequent years. The most prominent recent example among high income respondent states is Spain which faced 33 claims since 2011.
It remains open to question, however, whether and through which mechanisms FDI in advanced host countries will be affected by the recent increase in the number of disputes with such hosts. This issue is taken up again by performing complementary estimations in Sect. 4.
We were unable to reproduce this aspect of Wellhausen’s findings. While we can only surmise the source of this discrepancy, it is possible that Wellhausen’s results were driven by her choice of transformation of the dependent variable.
For details on this case see: https://icsid.worldbank.org/ICSID/FrontServlet?requestType=CasesRH&actionVal=showDoc&docId=DC2551_En&caseId=C661; bilateral FDI stock data are from UNCTAD.
CAFTA stands for Central American Free Trade Agreement.
This includes the ability for the investor to extract higher compensation from the host through settlement or negotiation due to the threat of bringing a case to ISDS.
As we explain in more detail below, we also consider high income host countries by performing complementary regressions in Sect. 4.
Indeed, there is also substantial debate about whether BITs attract FDI at all. In addition to the mixed findings in the empirical literature linking BIT participation to increased FDI flows, surveys have shown that many investors were not aware of BITs or their potential (Yackee 2010; Skovgaard Poulsen 2015). Thus the theories about investor response presented here apply to that portion of investors (or their legal counsel) who were informed about BITs.
Proponents of the “BITs as signals” hypothesis argue that studies which examine the impact of BIT participation on the basis of bilateral FDI inflows underestimate the effectiveness of BITs by ignoring the host country’s “willingness to protect all foreign investment” (Neumayer and Spess 2005: 1572). However, assessing the “BITs as signals” hypothesis on the basis of aggregate FDI inflows is likely to overstate the effectiveness of BITs unless the endogeneity of BIT formation is taken into account. As we explain in more detail in Sect. 3 below, the use of bilateral FDI inflows renders it easier to mitigate endogeneity concerns.
If this causal effect is significant relative to other sources of variation in the data, it should be possible to use bilateral FDI data to identify the effect of BIT coming into force even once host-year effects are controlled for. The belief that the case for a causal impact of BITs on FDI flows rests on the identification of greater response from the partner country is one of the reasons that bilateral (c.f. host-level) FDI data dominates the economics literature on the impacts of BITs.
Jandhyala and Weiner (2014) are an important exception. They outline similar mechanisms to those explained here, but are not interested in determining which has the most impact on investor behavior.
The effectiveness of various post-establishment obligations (e.g., lawful expropriation, minimum standard of treatment, transfer of funds) depends to a great extent on strict and binding ISDS provisions.
A review of arbitration decisions by UNCTAD (2008: XXV) revealed that “less than half of the awards rendered favored the claimant, and that damages awarded were considerably smaller than the total claims made by investors.”
Once a developing country improves its investment climate, it will also be more likely to participate in BITs. Consequently, it is hard for the econometrician to distinguish to what extent any resulting increase in FDI flows is due to the improved investment conditions, or due to BIT participation. The lack of adequate and valid instruments renders it still more difficult to establish causal effects of BITs on (aggregate) FDI inflows. See Berger et al. (2013) for a short account of previous attempts to address endogeneity concerns in the literature on BITs and FDI. Kerner (2009) uses BITs signed by the source partner with neighboring hosts as an instrument for BIT formation. This instrument is, however, problematic due to the strong evidence of spatial correlation in FDI flows. Neumayer and Spess (2005) and Allee and Peinhardt (2011) take an even more minimalist approach to the problem by lagging explanatory variables one period.
Our most conservative specification includes dyad, host-year and source-year fixed effects. Thus any country-pairs that never report non-zero flows, any host-years and any source-years in which there are no reported flows are dropped from the regression. This decreases the proportion of zero or missing flows which are relevant for identification of the variables of interest.
This variable is set equal to 1 if there is a BIT in force and a claim against the host, and the date of first claim against the host is equal to or later than the date of entry of the BIT into force.
This variable is set equal to 1 if there is a BIT in force and a claim against the host, and the date of first claim against the host is earlier than the date of entry of the BIT into force.
Recall that both H2a and H3a predict similarly strong reactions of FDI from partner and non-partner countries to BIT signing.
In contrast, the effects of BIT-related claims on FDI flows are no longer statistically significant when considering high income host countries (see Sect. 4 for details).
The FDI data for financial offshore centers are highly likely to be biased. We exclude all countries that are on the list of offshore financial centers as reported by Eurostat (2005).
More precisely, we excluded host countries which were classified as “high-income” according to the World Bank’s World Development Indicators for more than half our sample period.
We are grateful to an anonymous reviewer for having alerted us to sample selection problems. On the other hand, the problem of “inappropriate pooling”, stressed by Blonigen and Wang (2005) with regard to widely varying economic growth effects of FDI in developing versus developed host countries, may be relevant in the present context, too.
Recall the fundamental endogeneity problem is that as a host improves its investment climate we would expect to see increased BIT participation and increased FDI inflows, without the BIT participation necessarily being the cause of the FDI growth.
Specifically, neither the sum of the coefficients for BIT in force and Claim with BIT in force, nor the sum of the coefficients for BIT in force and BIT after claim is positive. A Wald test of the constraint BIT-in force ijt + Claim with BIT ijt = 0 has a p value of 0.22 and the corresponding value for the constraint BIT in force ijt + BIT after claim ijt = 0 is 0.36. Thus, if anything, the sums suggest a slight (though statistically insignificant) negative net impact of BITs for hosts which have faced claims.
The magnitude of each of the variables is, however, somewhat diminished. This is consistent with the idea that controlling for host-year effects reduces endogeneity bias. For example Aisbett (2009) has argued that developing host countries which have increasing FDI inflows are more likely to form BITs. Similarly, in the current paper we have argued that claims against developing host countries are likely to be associated with general decreases in FDI flows due to a worsening investment climate.
The coefficient on BIT after claim even proves to be significantly positive in column (3), though just marginally at the 10% level. While not the focus of the current study, the combination of the above results for the high income host sample and the slight negative net impact for non-high income countries of BITs which are signed after there has been a claim may actually point to something systematic. Specifically, one of the functions of BITs is to improve the ease of repatriating assets and profits to the home country. It could be that this effect is showing up in the FDI data.
The coefficients of BIT in force and BIT after claim change together in such a way that the net impact of a BIT coming into force is slightly negative, as it was for the base sample which includes post-socialist countries.
The difference is that instead of showing the impact on FDI flows from protected investors of the first claim against the host, the new interaction term shows the average impact per claim of the cumulative number of claims against the host.
References
Aisbett, E. (2009). Bilateral investment treaties and foreign direct investment: Correlation versus causation. In K. P. Sauvant & L. E. Sachs (Eds.), The effect of treaties on foreign direct investment: Bilateral investment treaties, double taxation treaties, and investment flows (pp. 395–435). Oxford: Oxford University Press.
Aisbett, E., Karp, L., & McAusland, C. (2010). Compensation for indirect expropriation in international investment agreements: Implications of national treatment and rights to invest. Journal of Globalization and Development, 1(2), Article 6.
Allee, T., & Peinhardt, C. (2011). Contingent credibility: The impact of investment treaty violations on foreign direct investment. International Organization, 65(3), 401–432.
Berger, A., Busse, M., Nunnenkamp, P., & Roy, M. (2011). More stringent BITs, less ambiguous effects on FDI? Not a bit! Economics Letters, 112(3), 270–272.
Berger, A., Busse, M., Nunnenkamp, P., & Roy, M. (2013). Do trade and investment agreements lead to more FDI? Accounting for key provisions inside the black box. International Economics and Economic Policy, 10(2), 247–275.
Bertrand, M., & Mullainathan, S. (2003). Enjoying the quiet life? Corporate governance and managerial preferences. Journal of Political Economy, 111, 1043–1075.
Blonigen, B. A., & Wang, M. G. (2005). Inappropriate pooling of wealthy and poor countries in empirical FDI studies. In T. H. Moran, E. M. Graham, & M. Blomström (Eds.), Does foreign direct investment promote development? (pp. 221–244). Washington, DC: Institute for International Economics.
Blume, L., Rubinfeld, D. L., & Shapiro, P. (1984). The taking of land: When should compensation be paid? Quarterly Journal of Economics, 99, 71–92.
Buntin, M. B., & Zaslavsky, A. M. (2004). Too much ado about two-part models and transformation? Comparing methods of modeling medicare expenditures. Journal of Health Economics, 23(3), 525–542.
Burbidge, J. B., Magee, L., & Robb, A. L. (1988). Alternative transformations to handle extreme values of the dependent variable. Journal of the American Statistical Association, 83(401), 123–127.
Busse, M., Königer, J., & Nunnenkamp, P. (2010). FDI promotion through bilateral investment treaties: More than a bit? Review of World Economics, 146(1), 147–177.
Büthe, T., & Milner, H. V. (2014). Foreign direct investment and institutional diversity in trade agreements: Credibility, commitment, and economic flows in the developing world, 1971–2007. World Politics, 66(1), 88–122.
Egger, P., & Merlo, V. (2012). BITs bite: An anatomy of the impact of bilateral investment treaties on multinational firms. Scandinavian Journal of Economics, 114(4), 1240–1266.
Eurostat. (2005). European Union foreign direct investment yearbook 2005. Luxembourg: European Communities.
Gaukrodger, D., & Gordon, K. (2012). Investor-state dispute settlement: A scoping paper for the investment policy community (OECD Working Papers on International Investment, 2012/03). OECD Publishing. doi:10.1787/5k46b1r85j6f-en. Accessed January, 2016.
Hallward-Driemeier, M. (2003). Do bilateral investment treaties attract foreign direct investment? Only a bit…and they could bite (World Bank Policy Research Working Paper 3121). Washington, DC: World Bank.
Halvorsen, R., & Palmquist, R. (1980). The interpretation of dummy variables in semilogartihmic equations. American Economic Review, 70(3), 474–475.
Hicks, J. R. (1935). Annual survey of economic theory: The theory of monopoly. Econometrica, 3, 1–20.
Jandhyala, S., & Weiner, R. J. (2014). Institutions sans frontieres: International agreements and foreign investment. Journal of International Business Studies, 45, 649–669.
Johnson, N. L. (1949). Systems of frequency curves generated by methods of translation. Biometrika, 36(1–2), 149–176.
Kerner, A. (2009). Why should I believe you? The costs and consequences of bilateral investment treaties. International Studies Quarterly, 53(1), 73–102.
Markusen, J. R. (2001). Commitment to rules on investment: The developing countries’ stake. Review of International Economics, 9(2), 287–302.
Miceli, T. J. (1991). Compensation for the taking of land under eminent domain. Journal of Institutional and Theoretical Economics, 147(2), 354–363.
Mullahy, J. (1998). Much ado about two: Reconsidering retransformation and the two-part model in health econometrics. Journal of Health Economics, 17(3), 247–281.
Neumayer, E., & Spess, L. (2005). Do bilateral investment treaties increase foreign direct investment to developing countries? World Development, 33(10), 1567–1585.
Pence, K. M. (2006). The role of wealth transformations: An application to estimating the effect of tax incentives on saving. Contributions to Economic Analysis & Policy, 5(1), Article 20.
Santos Silva, J. M. C., & Tenreyro, S. (2006). The log of gravity. Review of Economics and Statistics, 88(4), 641–658.
Sauvant, K. P., & Sachs, L. E. (2009). The effect of treaties on foreign direct investment: Bilateral investment treaties, double taxation treaties, and investment flows. Oxford: Oxford University Press.
Simmons, B. A. (2014). Bargaining over BITs, arbitrating awards: The regime for protection and promotion of international investment. World Politics, 66(1), 12–46.
Skovgaard Poulsen, L. N. (2015). Bounded rationality and economic diplomacy: The politics of investment treaties in developing countries. Cambridge: Cambridge University Press.
Skovgaard Poulsen, L. N., & Aisbett, E. (2013). When the claim hits: Bilateral investment treaties and bounded rational learning. World Politics, 65(2), 273–313.
Tobin, J. L., & Rose-Ackerman, S. (2011). When BITs have some bite: The political-economic environment for bilateral investment treaties. Review of International Organizations, 6(1), 1–32.
UNCTAD. (2008). World investment report 2008. New York and Geneva: United Nations.
UNCTAD. (2014a). Foreign direct investment statistics (online database). http://unctad.org/en/Pages/DIAE/FDI%20Statistics/FDI-Statistics-Bilateral.aspx. Accessed October, 2014.
UNCTAD. (2014b). Bilateral investment treaties (online database). http://investmentpolicyhub.unctad.org/IIA. Accessed October, 2014.
UNCTAD. (2014c). World investment report 2014. New York and Geneva: United Nations.
UNCTAD. (2015a). World investment report 2015. New York and Geneva: United Nations.
UNCTAD. (2015b). Double taxation treaties (online database). http://unctad.org/en/Pages/DIAE/International%20Investment%20Agreements%20(IIA)/Country-specific-Lists-of-DTTs.aspx. Accessed March, 2015.
Van Harten, G., & Scott, D. N. (2015). Investment treaties and the internal vetting of regulatory proposals: A case study from Canada (Research Paper 71). Toronto: Osgoode Hall Law School.
Waibel, M., Kaushal, A., Kyo-Hwa Chung, L., & Balchin, C. (2010). The backlash against investment arbitration: Perceptions and reality. In M. Waibel, A. Kaushal, et al. (Eds.), The backlash against investment arbitration. New York: Kluwer Law International.
Wälde, T. (2005). The “umbrella” clause in investment arbitration: A comment on original intentions and recent cases. Journal of World Investment and Trade, 6(2), 183–236.
Wellhausen, R. L. (2015). Investor-state disputes: When can governments break contracts? Journal of Conflict Resolution, 59(2), 239–261.
World Bank. (2014). World development indicators (online database). http://data.worldbank.org/data-catalog/world-development-indicators. Accessed October, 2014.
World Bank. (2015). International Center for Settlement of Investment Disputes (online database). https://icsid.worldbank.org/apps/ICSIDWEB/cases/Pages/AdvancedSearch.aspx. Accessed August, 2015.
Yackee, J. W. (2010). Do bilateral investment treaties promote foreign direct investment—Some hints from alternative evidence. Virginia Journal of International Law, 51(2), 397–442.
Acknowledgements
Thanks to Maximillian Mantei for excellent research assistance and to participants at the DIE conference on The Political Economy of International Investment Agreements and anonymous reviewers for helpful comments and feedback. All remaining errors are the authors.
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix 1: Calculation of relative effects
The calculation of marginal effects and marginal elasticities from the coefficients of models linear in the inverse hyperbolic sine (IHS) transformation has been well described in Pence (2006). Since our variables of interest are (0,1) dummies, we are not interested in marginal effects, but rather the relative change in FDI flows associated with a change in a dummy variable(s) from zero to one (Halvorsen and Palmquist, 1980). Denote the FDI flow when the dummy variable(s) take(s) the value zero as y 0 , and the corresponding flow when the dummies are one as y 1 . We wish to calculate the estimated proportional change (also called relative effect), (y 1 − y 0)/y 0.
Consider a general estimating equation linear in the IHS transformation (with the scaling variable ϴ set equal to one as in our case), with dummy variable (D) whose relative effect we wish to calculate.
It is easy to see that
To ease exposition, define w ≡ y + (y 2 + 1)1/2, then, from the definition of the IHS function, IHS(y) = ln(w).
Substituting into the estimating equation and exponentiating we have:
Substituting for w and rearranging gives
Thus the relative effect of the dummy variable can be calculated from
where c is as defined in Eq. 1 and \({{\upalpha }}\) is the regression coefficient on the dummy variable of interest. Note that c approaches 1 quickly as \(y_{0}\) grows in absolute value. Already with \({y_{0} } = 100\), c equals 1 to 5 decimal places for any \({{\upalpha }}\) between minus one and one. Hence the relative effect can confidently be approximated by \(\exp \left( {{\upalpha }} \right) - 1\) in most economically relevant cases. By symmetry of the IHS function relative effect can be shown to approximate closely exp(-α)-1 for y0 < −100.
The exact solution to Eq. 2 can be found fairly easily by iteration. For illustrative purposes, Tables 7, 8, 9 and 10 show example correspondence values in the range relevant to our study. Note the symmetry of the IHS transformation means that the values in Table 7 are identical to those in Table 10; and those in Table 8 to those in Table 9.
Appendix 2: Definition of variables and data sources
Variable | Definition | Source |
---|---|---|
FDI | Bilateral FDI flows from source to host country in current (1000) US$ | UNCTAD (2014a) |
IHS of FDI | The inverse hyperbolic sine transformation of FDI is given by \({ \log }\left( {FDI + \left( {FDI^{2} + 1} \right)^{1/2} } \right)\) | Own calculation |
BIT in force | Dummy variable, set equal to one in the case of a bilateral investment treaty in force between source and host country | UNCTAD (2014b) |
Claim against host | Dummy variable, set equal to one in the case a claim has been brought against the host | World Bank (2015) |
BIT after claim | Dummy variable, set equal to one if there is a BIT in force and a claim against the host, and the date of first claim against the host is earlier than the date of entry of the BIT into force | Own calculation |
Claim with BIT in force | Dummy variable, set equal to one if there is a BIT in force and a claim against the host, and the date of first claim against the host is equal to or later than the date of entry of the BIT into force | Own calculation |
PTA | Dummy variable, set equal to one if there is any preferential trade agreement between host and source country | World Trade Organization (2015) |
DTT | Dummy variable, set equal to one if there is any double taxation treaty ratified between host and source country | UNCTAD (2015b) |
Host total BITs | Total number of bilateral investment treaties the host participates in | Own calculation |
Host Log of GDP | Log of host country GDP in current US$ | World Bank (2014), data for ARG and TWN from Penn World Table 8.0 |
Host Exch. Rate | Host country official exchange rate (LCU per US$, period average) | World Bank (2014), data for TWN, TKM and UZB from Penn World Table 8.0 |
Host Openness | Host country merchandise trade (% of GDP) | World Bank (2014) |
Host Inflation | Host country GDP deflator (base year varies by country) | World Bank (2014) |
Source Log of GDP | Log of source country GDP in current US$ | World Bank (2014), data for ARG from Penn World Table 8.0 |
Annual claims against host | Number of claims against host country in given year | Own calculation |
Cumulative claims against host | Total number of claims that have been brought against host country up to and including the given year | Own calculation |
Source annual claims against host | Number of claims against host country in given year brought by investors from the source country | Own calculation |
Source cumulative claims against host | Total number of claims that has been brought against host country up to and including the given year by investors from the source country | Own calculation |
Appendix 3: Descriptive statistics
Variable | Obs. | Mean | SD | Min | Max |
---|---|---|---|---|---|
IHS of FDI | 60,572 | 1.39 | 4.87 | −16.88 | 17.57 |
BIT in force | 60,572 | 0.25 | 0.43 | 0 | 1 |
Claim against host | 60,572 | 0.20 | 0.40 | 0 | 1 |
BIT after claim | 60,572 | 0.02 | 0.14 | 0 | 1 |
Claim with BIT in force | 60,572 | 0.10 | 0.30 | 0 | 1 |
PTA | 60,572 | 0.15 | 0.35 | 0 | 1 |
DTT | 60,572 | 0.26 | 0.44 | 0 | 1 |
Host total BITs | 60,572 | 17.6 | 18.1 | 0 | 100 |
Host Log of GDP | 60,028 | 23.74 | 1.76 | 18.99 | 29.74 |
Host Exch. Rate | 59,834 | 520.83 | 1939.59 | 9.33e−12 | 20,828 |
Host Openness | 58,894 | 60.28 | 31.09 | 5.00 | 203.04 |
Host Inflation | 59,129 | 209.53 | 416.82 | 1.75e−12 | 4004.39 |
Source Log of GDP | 60,495 | 26.64 | 1.45 | 22.44 | 30.42 |
Annual claims against host | 60,572 | 0.11 | 0.67 | 0 | 21 |
Cumulative claims against host | 60,572 | 0.76 | 3.39 | 0 | 48 |
Source annual claims against host | 60,572 | 0.0035 | 0.08 | 0 | 6 |
Source cumulative claims against host | 60,572 | 0.03 | 0.34 | 0 | 17 |
Appendix 4: Source country sample
Argentina (1992–2010), Australia (1992–2010), Austria (1985–2010), Belgium (2002–2010), Brazil (1992–2010), Canada (1980–2010), Chile (1992–2010), China (2003–2010), Colombia (1992–2010), Denmark (1985–2010), Finland (1985–2010), France (1986–2010), Germany (1980–2010), Greece (2003–2010), Iceland (1988–2010), Ireland (2001–2010), Israel (2001–2010), Italy (2001–2010), Japan (1980–2010), Republic of Korea (1990–2010), Luxembourg (2002–2010), Malaysia (1980–2010), Mexico (1990–2010), Netherlands (1982–2010), New Zealand (1980–2010), Norway (1986–2010), Poland (1996–2010), Portugal (1990–2010), Russia (2007–2010), South Africa (2001–2010), Spain (1992–2010), Sweden (1982–2010), Switzerland (1993–2010), Chinese Taipei (1980–2010), Thailand (1980–2010), Turkey (2000–2010), United Kingdom (1985–2010), United States (1982–2010), Venezuela (1990–2010) |
Appendix 5: Host country sample
Albania, Algeria, Angola, Argentina, Armenia, Azerbaijan, Bangladesh, Bolivia, Botswana, Brazil, Bulgaria, Burkina Faso, Burundi, Cameroon, China, Colombia, Democratic Republic of Congo, Republic of Congo, Costa Rica, Côte d’Ivoire, Dominican Republic, Ecuador, Egypt, El Salvador, Ethiopia, Gambia, Georgia, Ghana, Guatemala, Guinea, Guyana, Haiti, Honduras, Hungary, India, Indonesia, Jordan, Kazakhstan, Kenya, Kyrgyzstan, Macedonia, Madagascar, Malaysia, Mali, Mauritius, Mexico, Moldova, Mongolia, Morocco, Mozambique, Namibia, Nicaragua, Niger, Nigeria, Pakistan, Papua New Guinea, Paraguay, Peru, Philippines, Romania, Rwanda, Senegal, Seychelles, South Africa, Sri Lanka, Sudan, Swaziland, Syria, Taiwan, Tanzania, Thailand, Togo, Tunisia, Turkey, Turkmenistan, Uganda, Ukraine, Uruguay, Uzbekistan, Venezuela, Vietnam, Zambia, Zimbabwe |
About this article
Cite this article
Aisbett, E., Busse, M. & Nunnenkamp, P. Bilateral investment treaties as deterrents of host-country discretion: the impact of investor-state disputes on foreign direct investment in developing countries. Rev World Econ 154, 119–155 (2018). https://doi.org/10.1007/s10290-017-0285-1
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10290-017-0285-1
Keywords
- Bilateral investment treaties
- Investor-state dispute settlement
- Compensation claims
- Foreign direct investment
- Developing host countries