Direct international aid flows directly to individuals and communities in the form of such essentials as food, water, medicine, and cash. As humanitarian crises erupt around the world, this form of assistance has become both politically contentious and more necessary than ever. Yehonatan Givati’s timely new paper, Income and Preferences for International Redistribution: Theory and Evidence, offers an enlightening framework to think about patterns of support for such aid programs. It builds on theoretical and empirical research that has established a relationship between preferences for domestic redistribution and income: poor citizens will likely support rich-to-poor redistributive policies within their own countries more than rich citizens. But what about international redistribution? How might an individual’s relative income level influence her support for internationally redistributive policies?
Givati begins by observing that the relationship between income and preferences for international redistribution is much less obvious than in the domestic context. Because lump-sum international redistribution must be financed through higher taxes, higher-income individuals have more to lose from international redistribution. This implies a negative relationship between income and support for international redistribution. At the same time, lower-income individuals may perceive that, given fixed funding levels, international redistribution will come at the expense of domestic redistribution. This implies a positive relationship between income and support for international redistribution.
To untangle these intuitions, Givati proposes a model in which a rational, utility-maximizing representative individual works and consumes. The individual chooses amounts of labor and leisure to maximize her utility, which is derived from consumption and leisure, plus two further elements: (1) she has altruistic preferences for a lump-sum “domestic grant” (like a universal basic income) that is paid to each individual within her country and (2) she also has altruistic preferences for a lump-sum “international grant” that is paid to each individual within a group of foreign countries. (P. 7.) In the model, the lump-sum domestic and international grants are financed by proportional (flat) income tax rates, so each of the taxes, by definition, is redistributive. (P. 6.)
The model allows Givati to assess the effect on an individual’s utility of an increase in the tax that finances international redistribution. Again, there are arrows pointing in opposite directions. On one hand, a tax increase (weakly) increases her utility because it provides more international redistribution and the individual has altruistic preferences. On the other hand, a tax increase distorts her labor-leisure choice. The increased tax causes her to work less even if the domestic grant tax rate does not change. This increases her leisure but decreases her income and leaves less income to be taxed to fund domestic redistribution. (P. 10.) This implies that an increase in taxes for international redistribution decreases utility. Balancing these effects on utility, the model gives us an expression for the optimal, utility-maximizing level of the tax that finances international redistribution.
From this expression, the paper can answer its central question of how the optimal level of tax that finances international redistribution varies with an individual’s income. Givati finds that an increase in income reduces the optimal level of the tax for international redistribution (P. 10): the cost of the program increases with income, but the benefit, including the indirect utility from the program, is fixed. This implies that there should be an overall negative effect on utility of an increase in taxes for international redistribution.
All of this makes theoretical sense. But is it true, in the real world? Does support for international redistribution really decrease as income rises? Or, as columnist Nicholas Kristof has put it, is it simply that “Americans are weary of international burdens and don’t feel it is their job to save every impoverished person?”
Somewhat amazingly (and Givati believes he is the first to use this data), four existing surveys share a question that asks about support, through taxes, for international redistribution. The 2000 and 2021 General Social Surveys (GSS), which are U.S.-focused, and the 1999 and 2019 International Social Survey Programs (ISSP), which are focused on other countries, include identical prompts, with which participants indicate their agreement on a 5-point scale: “People in wealthy countries should make an additional tax contribution to help people in poor countries” (emphasis mine). The prompts work well with Givati’s model because they contemplate direct aid to individuals and suggest the existence of two separate taxes, one for international redistribution and one for domestic redistribution.
These highly-relevant surveys allow Givati to assess whether poor people are more or less likely than rich people to support international redistribution within a given country. He first looks at the U.S. using the 2021 GSS data and, using both graphical evidence and ordinary least squares regression analysis with a range of controls, shows that there is a negative correlation between respondents’ support for international redistribution and their income. (P. 21.) Next, he investigates whether the correlation holds in other countries using the 2019 ISSP data, and finds that it does. (P. 23, Table 4.) He notes the possibility that a country’s status as rich or poor may affect whether the rich in that country support international redistribution more than the poor. He seeks to rule this out in two ways: first, by interacting the coefficient for the measure of respondents’ income with that for country-level GDP per capita (no effect found!) and, second, by breaking up the countries in the ISSP data into deciles by GDP per capita, and running regressions within each decile. On this latter test, for every one of the deciles, the coefficient on support for international redistribution is negative and statistically significant (P. 25, Table 5), meaning that no matter how rich or poor a country is, the negative relationship between income and support for international redistribution holds.
One might wonder: are the 2019 and 2021 data that Givati uses simply creatures of their times? Might preferences surrounding international redistribution have changed rapidly with the recent rise of populist politics and isolationist foreign policies? Only future surveys can tell. However, Givati shows that the correlation has historical stability: using the earlier datasets (GSS from 2000 and ISSP from 1999), he confirms a negative relationship between taxpayers’ income and their support for international redistribution. (P. 31-33, Table 8.)
Or, alternatively, maybe the poor are simply more generous than the rich? And the rich should aspire to be like the poor in this regard? These questions are beyond Givati’s scope. That said, I came away from the paper with a fervent-if-small-and-research-oriented hope: that the GSS and ISSP surveys will start asking, on an annual basis, about support for international redistribution. Regardless, Givati’s contribution offers needed insight into how support for international redistribution is itself distributed across societies.
