Last week, The Washington Post reported that the Trump administration might freeze $3 billion in foreign aid that Congress has already approved. Foreign aid regularly tops the list of the most unpopular U.S. government programs and is one of the few federal programs that a near-majority of Americans support cutting. Thus, many Americans (but not all) might support the move by the administration, as well as the hefty reductions in its last two budget proposals.

With inequality rising in the West, perhaps it’s no surprise that people cheer politicians who complain that their governments spend too much of their tax money abroad. But even the developed world’s poor and middle classes are, by global standards, extraordinarily rich. After adjusting for cost-of-living differences, a typical American still earns an income that is 10 times the income received by the typical person in the world.

Do Americans understand this fact? In short, no. Does their misperception of their comparative affluence help to explain deep-seated opposition to foreign aid and other forms of international redistribution? In short, yes. Let me explain.

Here’s how I did the research

To explore these questions, I recruited a nationally representative sample of about 1,500 U.S. residents through the GfK KnowledgePanel. I randomly assigned respondents to three groups. I asked members of the first group for their attitudes toward foreign aid, U.S. trade protections for domestic agriculture and charitable giving.

In the second group, before asking for these attitudes, I asked respondents to estimate the percentage of the world’s population receiving a lower annual income than they do, and to estimate the income received by the typical person in the world — the global median income. I then gave them accurate cost-of-living-adjusted information for both the percentage below them and the global median income, based on data collected by Branko Milanovic. Then I asked them the same questions as answered by the first group.

A third group of respondents was asked to estimate the median income and their percentile in the income distribution but not given any information. This makes it possible to quantify the effect of the increased mental salience of relative affluence on people’s attitudes even absent any information; the effect is about half of what the rest of the post discusses among people also given the correct information.

Americans profoundly underestimate how rich they are compared to the rest of the world.

So what did they say? The average U.S. resident estimated that the global median individual income is about $20,000 a year. In fact, the real answer is about a tenth of that figure: roughly $2,100 per year. Similarly, Americans typically place themselves in the top 37 percent of the world’s income distribution. However, the vast majority of U.S. residents rank comfortably in the top 10 percent.

What explains these misperceptions? Human beings draw heavily on their own local, lived experience to make judgments about the wider world. As individuals’ own incomes rise, and therefore the incomes of those around them, so too do their overestimates of the global median income. Upon being recruited into the sample, people shared their incomes. A $10,000 increase in a person’s annual household income is associated with about a $670 increase in that person’s estimate of the global median income.

Learning the facts did change respondents’ attitudes.

Of the control group that answered the survey without any new information, about 12 percent supported higher U.S. spending on foreign economic assistance. Of those whose misperceptions had been corrected, that rose to 22 percent — an increase of about 10 percentage points. For comparison, that’s similar to the gap in attitudes between an average Republican and average Democrat. The information also prompted a smaller increase in support for trade policies that would give farmers from developing nations more freedom to sell their products in the United States.

They put their money where their survey responses were.

Respondents put their new attitudes into action. One in 10 study participants were given a $20 bonus payment and told that they could keep the entire amount or donate a percentage to a domestic or international charity. The remainder were told that 10 percent of study participants would receive a $20 bonus and would be asked how they would distribute it. On average, respondents from the first, uninformed group kept 62 percent of the bonus, donated 28 percent to a domestic charity and donated 10 percent to an international charity. That switched among the respondents who had been educated with the facts: They kept 54 percent, donated 30 percent to a domestic charity — and gave 16 percent to the international charity. Total giving to international charities rose by 56 percent.

Interestingly, those whose estimates were accurate to begin with didn’t change their beliefs about foreign aid — suggesting that actually learning about one’s relative global affluence is what alters support for international aid generally.

Of course, this was an experiment. We don’t know how stable these new attitudes are, once people are influenced by other information in the “real world.” What’s more, the study was conducted before a presidential election that became a referendum on the United States’ economic engagement with the rest of the world; the results might be different today.

Reminding Americans of their amazingly prosperous position in the world won’t by itself transform the terms of that larger debate. But a little information would make a difference in people’s attitudes.

Gautam Nair is a PhD candidate in political science at Yale University.