Donors and domestic politics: Political influences on foreign aid effort

Abstract

The vast majority of scholarship on foreign aid looks at either the effectiveness of foreign aid or why particular countries receive aid from particular donors. This paper takes a different approach: what are the domestic sources of support for foreign aid? Specifically, how does the donor's domestic political and economic environment influence ‘aid effort’? This paper uses a time-series cross-sectional data set to analyze the influence of changes in political and economic variables. As governments become more conservative, their aid effort is likely to fall. Domestic political variables appear to influence aid effort, but only for aid to low income countries and multilaterals while aid effort to middle income countries in unaffected. This suggests that models solely emphasizing donor economic and international strategic interests as determinants of donor aid policy may be mis-specified. These results also suggest sources of aid volatility that might influence recipient growth prospects.

Introduction

Foreign aid is an important but variable source of income for developing countries. Part of that variation stems for fluctuating levels of donor aid effort (the percentage of its GDP a donor decides to allocate to foreign aid). What influences these decisions to change aid effort? While public opinion surveys show large differences in support for aid across a liberal–conservative political spectrum, evidence that the ideological position of political parties in government influences aid effort is mixed at best. These mixed results are surprising as they suggest a minor role for domestic politics in explaining aid policy. This paper provides new evidence that party liberal–conservative positions influence aid effort.

The literatures on aid effectiveness and allocation frequently highlight the role of politics in explaining aid allocation (Alesina and Dollar, 2000, Boone, 1996, Burnside and Dollar, 2000, Clemens et al., 2004, Maizels and Nissanke, 1984, McKinlay and Little, 1977). But the political variables employed in these literatures typically focus on relationships between the donor and recipient country. Domestic political variables in the donor country are absent from these analyses and the role of politics is cast at the international level. Other scholars stress the influence of domestic politics in donors on aid policy (Fleck and Kilby, 2001, Fleck and Kilby, 2006, Irwin, 2000, Lancaster, 2007, Milner and Tingley, 2010, Noel and Therien, 1995, O’Keefe and Nielson, 2006, O’Leary, 1967, Rieselbach, 1966, Ruttan, 1996, Therien and Noel, 2000). This literature suggests that political parties and domestic political institutions play an important part in shaping foreign aid policy.

This paper investigates the role of domestic political variables in determining aid effort. While the focus of the paper is establishing the role of domestic politics, the upshot of the analysis is a better understanding of factors that could influence both aid effectiveness and aid allocation. Discussions about aid effectiveness would benefit from a firmer understanding of donor domestic politics. For instance, changes in the power of donor political parties might lead to changes in foreign aid priority and hence aid volatility, which has been linked to negative growth effects (Arellano et al., 2009, Bulir and Hamann, 2003, Bulir and Lane, 2002, Eifert and Gelb, 2005, Lensink and Morrissey, 2000). Likewise, domestic political factors can also influence the motivation for giving aid and hence the characteristics of preferred recipients (Fleck & Kilby, 2006). Results in this paper suggest that support for aid to countries with various levels of development differs across the types of domestic political actors influential in government.

The analyses presented below focus on the influence of political party ideology on foreign aid effort. Here political ideology is the liberal–conservative orientation of political parties and the governments they compose. Do governments that become more liberal become more likely to increase their foreign aid effort? I conceptualize liberal/conservatism in terms of views on the role of government in the economy. I also examine other political variables, such as the influence of welfare state institutions (Noel and Therien, 1995, Therien and Noel, 2000) and economic variables that may influence aid flows, such as the trade position and economic health of the country. This analysis dovetails with a range of work in comparative political economy that stresses the role played by political parties and their ideological orientation in shaping foreign economic policies (Bearce, 2003, Boix, 1998, Garrett, 1998).

A second contribution is the analysis of within-country changes in foreign aid effort using time series cross-sectional data. Most previous cross-sectional studies on aid effort have only analyzed cross-sectional variation at a handful of ‘snapshots’ in time. This analysis examines more closely the within-country dynamics that change aid effort. A third contribution is that I break aid out by different categories (e.g., low-income versus high-income developing countries) and channel (bilateral versus multilateral). The influence of domestic political and economic factors may be more salient for one type of aid, and looking only at aggregate aid would obscure this relationship.

The results suggest that as governments become more conservative their foreign aid efforts are likely to fall. This relationship is statistically significant in many, though not all, models, and provides new evidence about the relationship between political ideology and foreign aid over time. This effect is most significant in aid to poorer developing countries; aid to countries with higher levels of income appears relatively unaffected by changes in the ideological orientation of parties in donor countries. Year to year changes in welfare state institutions, a commonly cited source of donor aid policy, have little effect on foreign aid policy within a country but remain an important explanation of broader trends.