Ideas to Action:

Independent research for global prosperity


Views from the Center


The world’s poorest people have been getting richer recently. But they remain incredibly poor. The 10 percent of the world’s population still consuming $1.90 or less a day are subsisting on a small fraction of the resources available to people at the US poverty line. So you’d hope that the governments of the countries where they live would be trying to raise their consumption levels. But the reality is more complex.

In the rich world, even in the United States, government tax and transfer systems like welfare payments reduce the gap between rich and poor. But a new effort to produce comparable data on the equity of fiscal spending around the world reveals that in the world’s poorest countries, taxes are less progressive, financial transfers are much smaller, and—with the bulk of social spending soaked up by broken health and education systems—the net effect is often to leave people poorer than they started.

Tax and transfer systems reduce inequality in the rich world, but often exacerbate poverty in the poorest countries

In the OECD, taxes and transfers reduce overall income inequality by about 25 percent on average. That’s the result of a study by Isabelle Joumard, Mauro Pisu, and Debbie Bloch, who find that about three quarters of the reduction in inequality between market and disposable income is due to transfers like social security and family, housing, disability, and unemployment benefits. One-quarter of the reduction is due to the rich paying more taxes—particularly income taxes. The US is an outlier in seeing considerably less redistribution, but even so, the impact of taxes and transfers is to reduce inequality. 

But what has only recently been documented systematically is how limited is the impact of government redistribution in many developing countries. That’s thanks to a series of studies released by Nora Lustig and colleagues at the Commitment to Equity Institute (CEQ) at Tulane University.

The Gini coefficient measures the level of inequality in a group where zero is perfect equality (everyone gets the same income) and 100 is perfect inequality (one person gets all of the income). Before government taxes and transfers, the Gini for the 29 developing countries in Lustig’s study is 47, compared to 45 in the United States. Fiscal redistribution reduces the Gini by 2.6 points on average in developing countries. That compares to 7.2 points in the United States—and again, the US performance is weak by OECD standards.

In the rich world, the poorest citizens tend to be net financial recipients from the government—they get more in transfers than they pay in taxes. But that’s not true in some developing countries. Tax regimes in those countries aren’t very progressive—taxing the rich a similar percentage of their income as the poor. That’s because the revenue authorities tend to rely on indirect taxes like VAT—which fall on all consumers—rather than direct taxes on high personal or corporate incomes. The mechanics of a VAT are easier to implement for a weak state, and so organizations like the IMF have pushed countries down this path, prioritizing the need for revenue over a concern for equity. At the same time, welfare programs that transfer more cash to the poorest citizens tend to be a far smaller part of government spending.

And because governments tax much more than they transfer overall, on net, poor people lose a similar proportion of their income to the government as do rich people. That means the impact of government can be to increase poverty rates. In four of the five sub-Saharan African countries where CEQ data is available, the net effect of taxes and transfers is to increase the number of people living below the World Bank’s extreme poverty line. In Tanzania, poverty is nearly 20 percent higher due to taxes and transfers.

The picture is different in Latin America where (richer) governments have focused on reducing inequality through more significant transfer programs. CEQ data shows that in Chile and Argentina, for example, taxes and transfers reduce extreme poverty by about two-thirds (although this from a far lower level of poverty than Tanzania in the first place).In Brazil, they reduce the number of extreme poor by about one-third. These fiscal efforts are still modest by OECD standards, but have contributed to a broad decline in inequality in the region since 2000.

To caricature a bit, rich countries give poor people money, while poor countries give them schools with no teachers

Poor people do get services from government spending—roads, security, health, and education. Take the two largest in-kind benefits that developing country governments provide: free or subsidized education and health. The 29 low-income and middle-income governments Lustig and colleagues study spend 4.3 percent of GDP on education and 3.0 percent on health. While that compares to an OECD average of 5.3 percent on education and 6.2 percent on health, it is still a considerable outlay.

But it isn’t clear who benefits most from that expenditure. In education, for example, the average government teacher in Tanzania is paid almost four times GDP per capita to teach. In Kenya, they are paid about five times GDP per capita. That compares to average teacher pay in the United States worth about the same as GDP per capita. Clearly, educators are comparatively well compensated for their work in East Africa, and teacher salaries account for the considerable majority of education spending. But despite that, they don’t turn up: public school teachers in Tanzania and Kenya are absent from the class 47 percent of the time they are meant to be there, according to World Bank data. And many fail simple ability tests: the average teacher score on test questions involving the addition of two decimals was only 64 percent in Tanzania, for example. Poor quality teaching is one obvious reason why student learning outcomes are so grim: only 39 percent of Tanzanian school children can pass a test involving simple addition, subtraction, and multiplication after three years of schooling in Tanzania, for example. That drops to 27 percent amongst poor children.

The point here is not to scapegoat teachers or doctors and nurses. The failure of health and education systems in poor countries reflects a complex web of factors. The point is that poor people in poor countries pay almost as large a proportion of their incomes as richer people do to the government. In return, they get anemic transfers and (often) low-quality government services.

Looking at the graph above that ranks countries from richest to poorest, there is a fairly clear pattern by which the poorest countries spend the lowest percent of GDP on social services, and of these services, the lowest share on cash (e.g., direct transfers or pensions) as opposed to in-kind services like health and education. There is some irony here. Where people are neediest, social spending is lowest, and where public services are the least reliable, the proportion of social spending devoted to cash payments is lowest.

Ultimately, eradicating global poverty will require decades of sustained economic growth, and a state capable of delivering high-quality education and health services to all. But in the short term, just making the current system of taxes and transfers slightly more progressive would be—on a technical level—a pretty easy fix for poverty and inequality in many countries. As this new data shows, it’s not that simple redistribution doesn’t work, it’s just that in the poorest parts of the world, it hasn’t been tried.

Thanks to Divyanshi Wadhwa for excellent research assistance, and to Nora Lustig for helpful clarifications.


CGD blog posts reflect the views of the authors drawing on prior research and experience in their areas of expertise. CGD does not take institutional positions.