International Trade and Domestic Legal Systems: Examining the Impact of Islamic Law

Abstract

What factors determine countries' international trade relations? Recent theories point to the potential importance of domestic legal traditions. Countries' legal systems influence the enforcement of contracts. This has been shown to affect trade flows in common law and civil law countries. However, these two legal systems do not constitute the universe of legal traditions. Islamic law is an important and fundamentally distinct legal system that has been largely overlooked. In this article, we offer the first direct test of the effect of Islamic law on countries' trade relations. We find that, on average, levels of bilateral trade are lowest among Islamic law states, holding all else constant. This finding suggests that, contrary to conventional wisdom, shared institutions alone are insufficient to enhance trade flows. Instead, levels of bilateral trade depend critically on the quality of shared institutions. The importance of countries' legal systems for trade declines over time, possibly due to the increased role of international arbitration bodies and/or the standardization of international sales contracts.

Keywords:

• contract enforcement
• Islamic law
• international trade
• legal systems
• UNIDROIT

We are grateful to Sara McLaughlin Mitchell, Charles Wesley Powell, Dale L. Smith, and Greg Vonnahme for their comments on this paper.

Notes

¹Throughout this paper, quality refers to contract enforcement.

²In fact, most contractual disputes that come before courts concern an issue arising from incomplete contracts (Dawson, Harvey, and Henderson 1982).

³Pakistan recently imposed Islamic law in a vast region of the north west called Malakand (Shah 2009).

⁴The Koran is the sacred book of the Muslims, and it literally means “the Reading.” The Sunna literally means “the path taken or trodden” by the Prophet Muhammad, and it contains his explanations, deeds, sayings, and conduct (Glenn 2007). Judicial consensus is established by “a common religious conviction” of major traditional legal scholars (Glenn 2007) and it regards specific points of Islamic law. Analogical reasoning, the fourth source of Islamic law, is used in circumstances not addressed by the Koran or the other two sources (Vago 2000).

⁵It is important to note, however, that at times these two mechanisms point in the same direction like, for example, in dyads with two common law countries. We thank an observant reviewer for pointing this out.

⁶This categorization is constructed using the CIA Fact Book, as well as other subsidiary legal sources, such as Glendon et al. (1994), Opolot (1980), website created at the Law Faculty of the University of Ottawa (http://www.droitcivil.uottawa.ca/world-legal-systems/eng-generale.html;), and Kritzer (2002). The Appendix contains description of the legal systems of all countries (also see Mitchell and Powell 2011).

⁷Mixed legal system, which is present in some states, represents a legal design where two or more systems apply interactively or cumulatively (systems where a clear form of any of the three legal systems cannot be observed). Although countries belonging to the mixed category constitute a rather small portion of the entire data set, for the purpose of correct estimates, inclusion of this category is critical. Examples include: Botswana, Brunei, Cameroon, China, Israel, or Japan. We do not have any theoretical expectations relating to the mixed law states, since this legal tradition constitutes an amalgamation of other legal systems.

⁸The unit of analysis is non directional dyad-year.

⁹We follow suit by including a variable, Common Language, that is equal to one if the primary language spoken in each state in a dyad is the same, and zero otherwise. Data come from Rose (2004a).

¹⁰Microfoundations for the gravity model have been provided by Bergstrand (1985, 1989), and Helpman and Krugman (1985).

¹¹One of the most important advantages of Rose's (2004a) data is that he deflates trade by a base Consumer Price Index. These data may be obtained at http://faculty.haas.berkley.edu/arose/RecRes.htm#Trade

¹²Legal structures need several years to become stable after a state becomes independent.

¹³Results of this model are supportive of our conclusions based on the OLS model.

¹⁴The validity of the Polity IV democracy scale is supported by its strong correlation (.85 to .92) with numerous conceptually and operationally different indicators of democracy developed by various researchers (such as Freedom House, and others) (Jaggers and Gurr 1995).

¹⁵Although GATT membership should also boost trade openness, Rose (2004a, 2004b) finds no evidence of this.

¹⁶See www.wto.org/english/tratop_region_e/region_e.htm. Some of the regional organizations include ASEAN, EEC/EC/EU, US-Israel FTA, NAFTA, CARICOM, SPARTECA, Mercosur, etc.

¹⁷The variable was assembled using the Expected Utility Generation and Data Management Program (EUGene) (Bennett and Stam 2000; Jones, Bremer, and Singer 1996).

¹⁸Since there are three possible dyadic combinations of states in a dyad (both states are communist (Communist Dyad), neither state is communist (Not Communist Dyad), or only one state in a dyad is communist (mixed Communist Dyad), in order to avoid “the dummy trap,” we excluded the Mixed Communist Dyad from the model.

¹⁹It is no surprise that CommonCommon dyads enjoy the highest levels of bilateral trade. These dyads are doubly blessed; they share similar institutions and the institutions they share have a strong record of contract enforcement.

²⁰This strategy is recommended by Rigobon and Rodrik (2005) who argue that failure to do so many introduce systematic bias.

²¹It is also consistent with research that finds market-protecting institutions matter more than democratic institutions for attracting foreign direct investment (Li and Resnick 2003).

²²Recall that Islamic law is the benchmark category. Therefore, the positive and significant coefficients on Common law and Civil law indicate that trade levels are higher on average in secular law countries as compared to Islamic law countries, all else equal.

²³This is true despite the negative and significant coefficient on the interaction term Common*Year. The positive and significant coefficient on Common law in this model is correctly interpreted as the marginal effect of Common law when Year equals zero. The reported coefficient indicates that trade is 26 percentage points higher in common law countries as compared to Islamic law countries in 1955 (i.e.that is, when Year equals zero). However, given the magnitude of the coefficients on Common law and Common*Year, we can conclude that trade is significantly higher in common law countries throughout the entire sample.

²⁴International law firms also assist in supervising litigation abroad (Schaffer, Earle and Agusti 2005).

²⁵In the Indonesian example discussed earlier, the international sales contract contained a clause requiring the parties to resolve all disputes by arbitration in Indonesia, applying Indonesian law, utilizing the Indonesian National Arbitration Body Rules (BANI).

²⁶Recall that our sample covers the period from 1955 to 1998.

²⁷ Figures 1, 2 and 3 were produced using the STATA do program created by Brambor, Clark, and Golder (2005)

²⁸However, the coefficients for civil law are estimated with a greater amount of error.