Relative treatment of aliens: firm-level evidence from developing countries

University of Oxford | December 2016

Relative treatment of aliens: firm-level evidence from developing countries

by Emma Aisbett and Lauge Poulsen

Introduction

Foreign firms routinely complain about mistreatment by host state governments in the developing world. Yet, opaque, unpredictable, and intrusive government treatment is not just a concern for foreign firms. Domestic firms are also exposed to these risks. So is one group of firms treated better than the other? Even when foreign firms are treated poorly, they may still be treated better by host governments than local firms – and vice versa.

There are two main reasons why developing country governments might discriminate in favour of domestic firms. In the presence of clientilist networks between domestic firms and political elites in host states, bureaucrats, judges, and politicians can make life harder for foreign firms resulting in a competitive advantage for local business interests. In addition, to the extent foreign investments involve high sunk costs, host governments can be incentivised to extract greater value from the foreign investment after it has taken place.

Yet, there could also be significant political advantages from a being foreign firm. Political elites in the host state may expect that foreign investment brings specific advantages compared with domestic investment. Equally, whereas some foreign firms may be subject to hold-up problems others can have more bargaining power than local firms due to the greater mobility of foreign firms. In addition, foreign firms can have unique sources of economic and political power over host states stemming from their connections with international financial institutions and home state governments. Finally, foreign firms often have considerable resources to negotiate favourable terms with host state officials. These potential advantages of foreign firms should be particularly pronounced in the poorest developing countries, where foreign capital is scarce, host governments poorly resourced, and domestic firms and industries small. In these countries, in particular, foreign firms may be treated considerably better than their domestic counterparts. Furthermore, in countries with weak legal systems local courts may also treat foreign firms better than comparable domestic firms due to the higher profile of cases involving aliens.

Understanding the relative treatment of foreign firms is not only relevant for literature on the political economy of foreign investment, but also has important policy implications for governments seeking to achieve competitive equality between firms. A political liability of foreignness could under some circumstances justify incentives and subsidies targeted at foreign firms so as to level the playing field. If there is an advantage of foreignness, on the other hand, such policies would be even more difficult to justify on economic terms. A case in point is the global web of investment treaties, which provide potent property right protections for foreign, but not domestic, investors. Investment treaties give foreign investors the right to avoid domestic courts in host states and file compensation claims against a wide range of government behaviour. One premise of this regime is that courts are biased against foreigners and, more generally, that political risks are particularly pervasive for foreign firms. Yet, this has not been subject to rigorous empirical testing.

Our study is the first to systematically assess the relative government treatment of foreign firms across a wide range of indicators – including by courts - and show how this treatment evolves in line with economic development. Our primary analysis is based on the World Bank’s Enterprise Surveys, where we find that foreign firms receive better treatment than domestic firms in both low and middle-income countries. Moreover, the political advantage of foreignness is particularly prevalent in the poorest countries of the world.

To cross-check our results, we show that subjective and objective measures in the data align well – as has been the case in previous studies using the World Bank Enterprise Surveys. In addition, we show that whereas foreign firms are treated relatively better than domestic firms as host income decreases, the absolute treatment of foreign firms improves as host income increases. This fits the well-established stylized fact that the investment climate is better in higher income countries, and shows that the determinants of relative treatment are not the same as those of absolute treatment. Moreover, our findings are backed up by qualitative feedback from private lawyers who act on behalf of foreign and domestic firms vis-à-vis host governments in developing countries.

In addition to these substantive findings, we offer a methodological contribution in our matching of foreign firms with comparable domestic firms. In particular, we address the concerns about specification bias in generalized linear models raised by Gary King and co- authors (Ho, Imai, King, and Stuart 2007; Iacus, King, and Porro 2011). Models like tobit and probit are widely used, including in previous studies using the Enterprise Services. But whereas the linear specification of these models can incorporate higher order terms and interactions between variables, such changes can have major impacts on the results obtained. The researcher is then left to choose which set of results to report, giving rise to the potential for substantial (intentional or otherwise) researcher bias. We avoid these model selection issues by employing a novel conditional likelihood estimator of relative risk (CLERR), introduced by Aisbett, Aisbett & Steinhauser (2016), that is useful not just for our study but whenever exact matching or coarsened exact matching can be used.

Read more (pdf)