Aiding Human Rights? The Effect of U.S., European, and Chinese Development Assistance on Rights Practices in Recipient Countries, 2001 to 2017

Abstract

Official development assistance (ODA) refers to aid intended to promote economic development and wellbeing in developing countries. The effect of ODA from Western donors continues to be debated, but the impact of aid from non-Western countries such as China is a relatively new field of inquiry and analysis. Using data on Western ODA and a new dataset of “ODA-like” disbursements from China, this article analyzes the relationship between bilateral aid receipts from three sources—the United States, major European donors, and China—and two sets of human rights practices: physical integrity and “empowerment” (i.e., civil and political) rights. Analyses are conducted using panel fixed-effects regression models with and without instrumental variables. U.S. ODA, in particular, improves human rights in recipient countries. Estimated effects of bilateral ODA from European donors and China are far less robust. These results suggest that U.S. aid is not as ineffective nor Chinese aid as pernicious as is commonly assumed.

Keywords:

• Official development assistance
• physical integrity
• civil and political rights
• instrumental variables
• China

Acknowledgement

I thank Evan Schofer for helpful comments on an earlier version of this paper.

Notes

1 The lone exception appears to be acceptance of the “One China” policy. Few countries that entertain diplomatic relations with Taiwan receive Chinese aid or investment (Fuchs and Rudyak 2019; Lee 2017; Regilme and Hodzi 2021).

2 Nevertheless, the Chinese government relishes any opportunity to expose human rights abuses in the United States. See, e.g., State Council Information Office (2021).

3 I thank an anonymous reviewer for raising this point.

4 These countries are Bahamas, Brunei, Cyprus, Israel, Kuwait, Libya, Qatar, Singapore, and United Arab Emirates (throughout the period); Slovenia (from 2003); Bahrain (from 2005); Saudi Arabia (from 2008); Barbados, Croatia, Oman, and Trinidad and Tobago (from 2011); and Saint Kitts and Nevis (from 2014).

5 I thank Evan Schofer and Wes Longhofer for generously sharing these data.

6 Nevertheless, many studies fail to show any consistent link between human rights conditions and receipt of Western aid (Carey 2007; Neumayer 2003; Zanger 2000).

7 I conducted these analyses in Stata 16.1 using the xtivreg2 command (Schaffer 2010).

8 Rotated factor loadings for these variables are .982, .988, and –.955, respectively, with an Eigenvalue of 2.852 (proportion of explained variance = .951).

9 This relationship transitions from negative to positive at roughly 16.4 deaths per 1,000 live births, just above the 17th percentile for former European colonies in the sample. ODA received from major European donors did not vary with infant mortality rates for all sampled developing countries, and the strictly linear (as opposed to curvilinear) association among former colonies was also insignificant.

10 Tests of overidentifying restrictions do not—and cannot—evaluate instrument validity directly, because the error terms are unobservable. Instead, they test whether the excluded instruments, considered individually, produce similar estimates for the effect of an endogenous variable. As a test of the exclusion restriction, one must therefore assume that at least one of the excluded instruments is valid.

11 I also analyzed bilateral ODA from France, Germany, and the United Kingdom individually and found only one statistically significant effect: German ODA improves respect for civil and political rights. These results are available upon request.

12 Hung (2015:143) makes a similar point about the global economic order: “China,” he says, “needs the perpetuation of the global neoliberal order to advance its economy.” He adds that China “would be shooting its own feet if it were to subvert this global order, the institutions related to this order, or the U.S. power underlying this order.”