Is the dispute settlement system, “jewel in the WTO’s crown”, beyond reach of developing countries?

A Correction to this article was published on 18 October 2019

This article has been updated


Since the inception of the World Trade Organisation (WTO) in 1995, member countries have been heavily relying on the organisation’s Dispute Settlement System (DSS). Exploiting a new database on WTO litigations between 1995 and 2014, this paper describes disputes initiated over this period and identifies potential sources of bias concerning the participation of developing countries. The analysis builds on three different models to determine country i’s probability of initiating a dispute against country j. Either it depends on the two countries’ structure of trade (the rules-based model), or it is also affected by country i’s or country j’s specific characteristics (the unilateral power-based model), or it is also affected by bilateral economic and trade relations between countries i and j (the bilateral power-based model). We find that country i’s structure of trade with j plays an important role in explaining the probability that i initiates a dispute against j under the DSS. We also find clear evidence with regard to the importance of two independent variables: first the legal capacity of i (a variable related to the unilateral power-based model) and second the trade retaliatory capacity of i against j (a variable related to the bilateral power-based). Almost all these results hold when testing for both the likelihood of initiating a dispute and for the number of disputes initiated before the WTO.

This is a preview of subscription content, access via your institution.

Fig. 1

Source: World Trade Organisation (2015)

Fig. 2

Source: Authors

Change history

  • 18 October 2019

    In the original publication of the article, the third affiliation has been missed to be added.


  1. 1.

    “The products of the territory of any contracting party imported into the territory of any other contracting party shall be accorded treatment no less favourable than that accorded to like products of national origin in respect of all laws, regulations and requirements affecting their internal sale, offering for sale, purchase, transportation, distribution or use.”

  2. 2.

    Let us remind that the WTO gives developing countries special rights, under an umbrella called Special and Differential Treatment (SDT). This includes: (i) longer time periods and augmented flexibility for the implementation of agreements and commitments; (ii) specific provision targeting the expansion of their trade opportunities and the safeguard of their interests; (iii) technical support (handling of disputes, implementation of technical standards...), and (iv) specific provisions for least-developed countries. It may be useful to differentiate between SDT under the “substantive” WTO Agreements that affects the rights and obligations of developing and least developed countries and SDT available to developing and least developed countries under the DSU. For the most part, these countries get little preferential treatment in formal disputes.

  3. 3.

    Intra-EU trade is excluded.

  4. 4.

    On the other hand, it may be that, whatever the value of trade flows at stake is, a country may complain in order to create a reputation effect. This is an interesting direction that we will not follow in this paper.

  5. 5.

    Another potential explanation was suggested by an anonymous reviewer: issues that proceed to formal disputes are also cases where, for political or other reasons, such as the domestic influence of the stakeholder benefiting from the WTO-inconsistent measure, it is very difficult for the respondent to change its measure without the pressure of a WTO panel ruling.

  6. 6.

    Horn et al. (1999) used a binomial model and assumed that the expected number of bilateral complaints is proportional to the number of products exported to the respondent.

  7. 7.

    The consultations phase is different from the panel phase and it may be considered separately. In particular, it is not subject to multilateral supervision and frequently leads to a settlement. This remark could lead to split this initial phase into two phases: consultations phase and panel phase.

  8. 8.

    To the authors’ knowledge, there has never been a litigation case in the history of the WTO in which the DS body has rejected the final panel report. For this to happen, all member states involved must reach a consensus to reject the panel’s recommendations, which seems unlikely because both the respondent and the complainant are involved.

  9. 9.

    In 93 percent of cases that have been ruled between 1995 and 2014, the panel of expert has favored the complainant.

  10. 10.

    Let us also mention that results from this study question the impartiality of an international organisation such as WTO and therefore, should be used with caution.

  11. 11.

    This phenomenon is mostly due to the complexity and the duration of disputes.

  12. 12.

    Based on the reliance of poor countries on bilateral aid and preferential trade arrangements.

  13. 13.

    We suppressed a total of 155 cases, among which 128 were not specific to an HS6 line (these include cases related to services, tax measures, copyrights, and so on), 23 were renewals, and 4 cases were related to export restrictions.

  14. 14.

    Settled disputes are cases classified under the following categories from the WTO’s website: (i) reports adopted, no further action required; (ii) implementation notified by respondents; (iii) compliance proceedings completed without findings of non-compliance; (iv) authorisation to retaliate granted; (v) withdrawn (measure withdrawn, other reasons); (vi) mutually terminated; (vii) mutually agreed solution.

  15. 15.

    For the design of this table we used the 2014 income classification.

  16. 16.

    This includes the European Community before 2010.

  17. 17.

    Services (HS2 codes 98 and 99) are not included in our analysis.

  18. 18.

    We calculate the average duration of disputes based on the 218 settled cases.

  19. 19.

    Individual (exporter) fixed effects are not included in the model because the dependent variable, \(y_{ij}\), always equals zero for a set of exporters. Exporter fixed effects associated with countries that have never filed a complaint before the DSS perfectly predict failure (i.e. dependent variable equal to zero). As a result, these observations are dropped from the regression. When adding exporter fixed effects, the model uses only 19,655 observations, compared to 67,161 without exporter fixed effects. Moreover, results obtained from the model that includes exporter fixed effects are similar to results obtained from the model that does not include exporter fixed effects. The only difference resides in the share of agricultural products in a country’s total exports to the partner country, \(AGRI_{ij}\), which is no longer significant.

  20. 20.

    This model may be called the unilateral power-based model or the unilateral model. Explanatory variables are systematically related to the characteristics of a single country, either the complainant or the respondent. These variables may be the expression of an economic power, like the legal capacity of the complainant, or a structural feature like the degree of democracy.

  21. 21.

    Throughout the analysis section, i represents the complainant and j the respondent.

  22. 22.

    The first time period lasts from 1995 to 1999, the second from 2000 to 2004, the third from 2005 to 2010, and the fourth from 2010 to 2014.

  23. 23.

    Seven countries were not included in the analysis due to a lack of data: Botswana, Lesotho, Liechtenstein, Luxembourg, Montenegro, Namibia, and Swaziland. Furthermore, Afghanistan, Kazakhstan, Liberia, and Seychelles were not included since they became members of the WTO after 2014.

  24. 24.

    A one-unit increase in the number of products exported from i to j at time period t increases the odds that the exporter will become a complainant by a factor of 1.0004.

  25. 25.

    We exclude all products classified under the following HS2 codes 25–27.

  26. 26.

    Recall from Sect. 4 that about 55 percent of disputes included in our dataset concerns trade remedies implemented by respondents.

  27. 27.

    We have tested other indicators of trade remedies provided by Bown (2011, 2014), such as the stock of the respondent’s imports subject to TTBs weighted by the HS6 product-level value of import, the flow of the respondent’s imports affected by TTBs by count, and the flow of the respondent’s imported subject to TTBs weighted by the HS6 product-level value of import. Results obtained from these models confirm our main conclusions. We do not report these results, but they may be requested from the authors.

  28. 28.

    In the unilateral power-based model, \(POLITY_{jt}\) is no longer significant, as is the case in the robustness analysis. In the bilateral power-based model, \(PTA_{jit}\) is no longer significant whereas \(RETAL_{jit}\) becomes significant as is also the case in various robustness checks. Otherwise, coefficients associated with all other variables remain significant and of comparable values.

  29. 29.

    The logged value of the exporter’s GDP per capita has been used in previous literature as a proxy for the complainant’s legal capacity. See, for example, Sattler and Bernauer (2008), Guzman and Simmons (2005), Horn et al. (1999). We think this is not a right proxy for legal capacity: for illustration countries with high per capita GDP may have a religious legal system, which may not be adapted to the WTO DSS.

  30. 30.

    We thank the ACWL staff for rapidly providing data on membership.

  31. 31.

    A Poisson distribution assumes that the mean and the variance of the count variable are the same. In this case, detailed descriptive statistics of the dependent variable, \(y_{ijt}\), demonstrate that the variance (0.009) is more than two times larger than the mean (0.004). The distribution of \(y_{ijt}\) thus displays signs of overdispersion. Furthermore, we run a negative binomial regression that provides a test of the overdispersion parameter. In this case, the overdispersion parameter is significantly different from zero, which reinforces that the Poisson distribution is not the most appropriate.

  32. 32.

    We also use the Poisson model to estimate the three models. A Vuong test was conducted to identify the best model, and systematically the ZIP model is better, which looks consistent, given the overdispersion in the data. We ran the same tests based on the negative binomial model. Results are similar to those obtained with the ZIP model. We present in this paper only the results from the ZIP model. Results from the Poisson and Vuong tests and from the negative binomial model may be requested from the authors.

  33. 33.

    Throughout this section, we report the average marginal effects rather than the odds ratio since we are estimating the determinants of the number of disputes initiated by i against j, rather than the probability that i files a complaint against j.

  34. 34.

    According to the 2017 US president’s trade policy agenda report, “it is important to recall also that Congress had made clear that Americans are not directly subject to WTO decisions” (Office of the USTR 2017, 5)


  1. Azevêdo, R. (2014). Azevêdo says success of WTO dispute settlement brings urgent challenges. World Trade Organization, September 26.

  2. Bagwell, K., Bown, C. P., & Staiger, R. W. (2016). Is the WTO passé? Journal of Economic Literature, 54(4), 1125–1231.

    Article  Google Scholar 

  3. Besson, F., & Mehdi, R. (2004). Is WTO dispute settlement system biased against developing countries? An empirical analysis. Paper presented at international conference on economic modeling, Paris, June 30–July 2.

  4. Bohanes, J., & Garza, F. (2012). Developing countries as plaintiffs and defendants in GATT/WTO trade disputes. Trade, Law and Development, 4(1), 45–124.

    Google Scholar 

  5. Bown, C. P. (2004a). Developing countries as plaintiffs and defendants in GATT/WTO trade disputes. World Economy, 27(1), 59–80.

    Article  Google Scholar 

  6. Bown, C. P. (2004b). On the economic success of GATT/WTO dispute settlement. Review of Economics and Statistics, 86(3), 811–823.

    Article  Google Scholar 

  7. Bown, C. P. (2005). Participation in WTO dispute settlement: Complainants, interested parties, and free riders. World Bank Economic Review, 19(2), 287–310.

    Article  Google Scholar 

  8. Bown, C. P. (2010). Self-enforcing trade: Developing countries and WTO dispute settlement. Washington, D.C.: Brookings Institution Press.

    Google Scholar 

  9. Bown, C. P. (2011). Taking stock of antidumping, safeguards and countervailing duties, 1990–2009. World Economy, 34(12), 1955–1998.

    Article  Google Scholar 

  10. Bown, C. P. (2014). Temporary trade barriers database. The World Bank. Accessed May 2019.

  11. Bown, C., & Hoekman, B. M. (2005). WTO dispute settlement and the missing developing country cases: Engaging the private sector. Journal of International Economic Law, 8(4), 861–890.

    Article  Google Scholar 

  12. Bown, C. P., & Reynolds, K. M. (2015). Trade agreements and enforcement evidence from WTO dispute settlement (Policy Research Working Paper WPS7242). Washington, DC: World Bank.

  13. Busch, M. L., & Reinhardt, E. (2000). Bargaining in the shadow of the law: Early settlement in GATT/WTO disputes. Fordham International Law Journal, 24(1), 158–172.

    Google Scholar 

  14. Busch, M. L., & Reinhardt, E. (2002). Testing international trade law: Empirical studies of GATT/WTO dispute settlement. In D. L. M. Kennedy & J. D. Southwick (Eds.), The political economy of international trade law: Essays in honor of Robert E. Hudec. New York: Cambridge.

    Google Scholar 

  15. Busch, M. L., & Reinhardt, E. (2003). Developing countries and general agreement on tariffs and trade/World Trade Organization dispute settlement. Journal of World Trade, 37(4), 719–735.

    Google Scholar 

  16. Bütler, M., & Hauser, H. (2000). The WTO Dispute settlement system: A first assessment from an economic perspective. Journal of Law, Economics, and Organization, 16(2), 503–533.

    Article  Google Scholar 

  17. CEPII (Centre d’Etudes Prospectives et d’Informations Internationales). (2017). Base pour l’Analyse du Commerce International (BACI) database. Accessed March 2016.

  18. CEPII (Centre d’Etudes Prospectives et d’Informations Internationales). (2018). Gravity database. Accessed December 2018.

  19. Conybeare, J. A. (1987). Trade wars: The theory and practice of international commercial rivalry. New York: Columbia University Press.

    Google Scholar 

  20. Davis, C. L. (2008). The effectiveness of WTO dispute settlement: An evaluation of negotiation versus adjudication strategies (Technical Report). Boston: American Political Science Association.

  21. Davis, C. L., & Bermeo, S. B. (2009). Who files? Developing country participation in GATT/WTO adjudication. The Journal of Politics, 71(3), 1033–1049.

    Article  Google Scholar 

  22. Freedom House. (2016). Accessed January 2016.

  23. Guzman, A., & Simmons, B. (2002). To settle or empanel? An empirical analysis of litigation and settlement at the World Trade Organization. Journal of Legal Studies, 31(S1), S205–S235.

    Article  Google Scholar 

  24. Guzman, A., & Simmons, B. (2005). Power plays and capacity constraints: The selection of defendants in WTO disputes. Journal of Legal Studies, 34(557), 557–598.

    Article  Google Scholar 

  25. Hoekman, B. M., & Mavroidis, P. C. (2000). WTO dispute settlement, transparency and surveillance. World Economy, 23(4), 524–542.

    Article  Google Scholar 

  26. Holmes, P., & Rollo, J. (2003). Emerging trends in WTO dispute settlement: back to the GATT? (Policy Research Working Paper 3133). Washington, DC: World Bank.

  27. Horn, H., Mavroidis, P. C., & Nordström, H. (1999). Is the use of the WTO dispute settlement system biased? In P. C. Mavroidis & A. O. Sykes (Eds.), The WTO and international trade law dispute settlement (pp. 134–164). Cheltenham, UK: Edward Elgar.

    Google Scholar 

  28. Johnson, H. G. (1953). Optimum tariffs and retaliation. Review of Economic Studies, 21(2), 142–153.

    Article  Google Scholar 

  29. Kennan, J., & Riezman, R. (2013). Do big countries win tariff wars? In R. Riezman (Ed.), International trade agreements and political economy (pp. 45–51). Singapore: World Scientific Publishing.

    Google Scholar 

  30. Michalopoulos, C. (1999). The developing countries in the WTO. World Economy, 22(1), 117–143.

    Article  Google Scholar 

  31. Office of the USTR (United States Trade Representative). (2017). The president’s 2017 trade policy agenda. Washington, DC. Accessed 15 Sept 2017.

  32. Petersmann, E. U. (1997). The GATT/WTO dispute settlement system: International law, international organizations and dispute settlement. Boston: Kluwer Law International.

    Google Scholar 

  33. Polity IV. (2016). Accessed January 2016.

  34. Reinhardt, E. (2000). Aggressive multilateralism: The determinants of GATT/WTO dispute initiation, 1948–1998. Unpublished. Atlanta, GA, US: Emory University Department of Political Science.

  35. Sattler, T., & Bernauer, T. (2008). Litigation in the World Trade Organization: Power of gravity or power of discrimination? (on file with authors).

  36. Sevilla, C. R. (1998). Explaining patterns of GATT/WTO trade complaints (Working Paper 98-01). Cambridge, MA, US: Weatherhead Center for International Affairs, Harvard University.

  37. United Nations. (2015). Comtrade database. Accessed September 2015.

  38. UNESCO. 2016. UNESCO Institute for Statistics. Accessed January 2016.

  39. World Bank. (2016). World development indicators. Accessed February 2016.

  40. WTO (World Trade Organization). (2015). Dispute settlement. Accessed July 2015.

Download references

Author information



Corresponding author

Correspondence to Antoine Bouët.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Funding support for this study was provided by the CGIAR Research Program on Policies, Institutions, and Markets (PIM). The views and opinions presented do not necessarily reflect those of PIM, International Food Policy Research Institute or CGIAR. The title of this article was inspired by Bohanes and Garza (2012). The Dispute Settlement System is often described as the jewel in the crown of the WTO. The authors are solely responsible for the content of the article. We thank participants in a Research Group in Theoretical and Applied Economics (GREThA) seminar (June 24, 2016), a Governance and economic Integration through Free Trade Agreements (GIFTA) workshop (July 7 and 8, 2016) and an IFPRI work-in-progress seminar (February 7, 2017), in particular Tanguy Bernard, Xin Geng, Joseph Glauber, Christophe Gouel, Sébastien Jean, David Laborde, Will Martin, Kathryn Pace, Flor Paz, and Eric Rougier, for their helpful comments and suggestions. We also thank two anonymous referees for their valuable comments. Of course all errors are ours.

About this article

Verify currency and authenticity via CrossMark

Cite this article

Bouët, A., Métivier, J. Is the dispute settlement system, “jewel in the WTO’s crown”, beyond reach of developing countries?. Rev World Econ 156, 1–38 (2020).

Download citation


  • WTO
  • Dispute Settlement System
  • Legal capacity
  • Trade retaliation

JEL Classification

  • F02
  • F13