Kindergarten students at Guilford Elementary school in 2015. The school system in Loudoun County, one of the wealthiest counties in the U.S. by some measures, only provides full-day kindergarten at Title I schools like Guilford, which receive special federal funding because of high rates of poverty. (Evelyn Hockstein/For The Washington Post)

 Child poverty is in the political news. But is there really a political commitment to ending it?

“We have the highest rate of childhood poverty of any major country on Earth” is one of Senator Bernie Sanders’ regular stump speech lines. In January, Speaker Paul Ryan convened a forum in which six Republican presidential candidates debated anti-poverty policy. And in the nationwide “Fight for $15” campaign, advocates claim that a $15 minimum wage will lift children out of poverty.

But as I’ll show below, the cross-national data show that really reducing child poverty requires a government much larger than most politicians have acknowledged.

Yes, the U.S. has exceptionally high rates of children in poverty

For as long as we’ve had credible social science on the topic, scholars have known that the U.S. has had an unusually high rate of child poverty. The world’s leading source on child poverty, the Luxembourg Income Study or LIS, shows that the U.S. child poverty rate has been high for decades. At least since the mid-1980s, more than one in five American children have been in poverty.

The LIS uses the classic relative measure of poverty, which is the most well-established measure in the international social sciences of poverty. This measure defines someone as poor if they live in a household that gets by on less than 50 percent of the median income, adjusted for household size.

These high rates of child poverty are troubling because child poverty has clear negative consequences. Scholars have convincingly linked child poverty with greater mortality, poor health, undermined educational attainment and lower adult income. Children who grow up poor are more likely to be victims of crime, to get sick, to experience family instability, and to suffer from a variety of disadvantages.

Is Sanders’ correct that the U.S. has the highest rate of childhood poverty of any major country? Scholars often compare the U.S. to “high income” countries that have been democratic since 1980, and with populations over a million. Within this group of what Hal Wilensky called “rich democracies,” the U.S. has consistently had one of the highest child poverty rates. Another country may pass the U.S. occasionally. For instance, child poverty has skyrocketed in Spain from 16 percent in 2000 to over 23 percent in 2013. And in recent decades, Canada, Greece, Ireland, and the United Kingdom all had at least one year in which more than 15 percentage of their children were poor. But the U.S. consistently has among the highest child poverty rates in the rich democracies.

Both Democratic Party presidential candidates acknowledge that changing that fact — and bringing more children out of poverty — would require greater tax revenue. For instance, Sanders has called for higher taxes and identified Denmark, which has a much larger government, as a country to emulate.

Is there a relationship between child poverty rates and government revenues?

They’re right, as we can see by looking at the relationship between child poverty and government revenue in the rich democracies. In the LIS, we can find child poverty estimates for 19 rich democracies for an average of 6 points in time since 1990. (Technical note: the countries are observed across similar but not the same years; some countries have as many as eight time points although Japan has only one. The list of countries and years can be found here.) If we take those 106 “country years,” the average child poverty rate was 11.8, and ranged from 2.6 to 24.7.

For each of those country-years, we checked that country’s government revenue as a percent of gross domestic product (GDP) in the Comparative Welfare States Database. Across all country-years, government revenue averaged 44.2 percent of GDP.

Let’s examine the most recent estimates that the LIS provides for the 19 rich democracies.


Child poverty (y-axis) and government revenue (x-axis) in 19 rich democracies in most recent LIS data (r=-.66, N=19).

Above, we see that when government revenue is high, child poverty is low (r=-.66). For instance, in the lower right hand corner, we see the Scandinavian countries, including Denmark, have low child poverty and high government revenue. By contrast, in the upper left corner, we see the U.S., with high child poverty and very low government revenue. Germany and the United Kingdom are in the middle.

Correlation isn’t the same as causation, of course. But a great deal of social science research explains how these two are connected. Evelyne Huber and John Stephens’ classic Development and Crisis of the Welfare State convincingly demonstrates that higher government revenue is essential to pay for generous social policies and public services. And it is those social policies and public services that are most responsible for lower child poverty.

As Lee Rainwater and Timothy Smeeding explain in Poor Kids in a Rich Country, higher government revenue implies that affluent families keep less of their income after taxes while disadvantaged families’ incomes are raised by welfare transfers. Thus, higher government revenue substantially narrows the gap between the top and bottom half of the income distribution.

At the same time, higher government revenue pays for public services that allow low-income families to devote their income to better meet their needs. As a result, and as shown in my book Rich Democracies, Poor People, generous social policies are much more important to achieving low poverty than economic performance or demographic factors like single parenthood.

What’s more, in the figure above, I draw a horizontal line to identify countries that achieve “low child poverty.” I define “low child poverty” as those that were in the bottom third of the distribution of child poverty rates across the 106 country-years in the 19 rich democracies from 1990 to 2015. This means the country has below 8.2 percent child poverty. I also draw a vertical line to mark the average level of government revenue in those country-years (44.2%).

As you can see, there are no countries in the graph’s lower left quadrant. Zero countries have below average government revenue and low child poverty. Apparently, low child poverty doesn’t happen unless a government has above average government revenue.

What social benefits make the biggest difference?

The benefits that successfully lift children out of poverty include what scholars and policymakers call “universal child cash benefits” — for example, what Germans call “Kindergeld.” Every family with children, regardless of need, receives a cash payment as an investment in the country’s children.

Children are also far less likely to grow up poor in societies that heavily subsidize healthcare and childcare, which enables parents to work and prevents illness from bankrupting the family. Child poverty can also be combated by supplying “near cash” benefits like food stamps and housing vouchers, and offering tax credits to supplement low-income parents’ earnings.

One clear illustration that these policies work comes from the Blair and Brown Labor Party governments in the U.K, which pledged to end child poverty in a generation. According to the LIS, in 1995, fully 20 percent of the UK’s children lived in poverty; and in 1999, 17.7 percent did.

As Jane Waldfogel shows in her book Britain’s War on Poverty, the U.K. effectively reduced child poverty through a variety of welfare transfers, tax credits, and programs. By 2004, only 13 percent of UK children were poor. By 2010, fewer than ten percent — 9.2 percent, to be exact — were poor. Hundreds of thousands, perhaps even millions, of children were lifted out of poverty.

And there should be no doubt: These policies required substantial government revenue.

Let’s redo the figure above with a little more detail, including all 106 LIS estimates for our 19 rich democracies across several years between 1990 and 2015. The figure below labels the countries with two-letter country names (so, for instance, “AU” is Australia, “AT” is Austria, and “SZ” is Switzerland) and two numbers for the year (e.g. “10” is 2010, “95” is 1995). The figure below is more crowded than the one above. But we can see that the correlation between government revenue and child poverty is even stronger (r=-.74), with essentially the same distribution, with most countries in the upper left with high child poverty and low government revenue, or in the lower right, with low child poverty and high government revenue.

Only one of our 106 country-years, Switzerland in 2002, deviated from this, with low government revenue and low child poverty. And Switzerland couldn’t sustain it; in every other year we have on record, Switzerland’s child poverty topped 8.2 percent.


Child poverty (y-axis) and government revenue (x-axis) in 19 rich democracies 1990-2015 (r=-.74, N=106).

Bringing down child poverty requires bringing up government revenue. 

For the U.S. to reduce low child poverty, it will require increases in government revenue. Indeed, the experience of rich democracies in the past 25 years suggests that above average government revenue is a necessary condition for low child poverty. Some argue that child poverty could be lower with higher marriage rates, more employment, efficiently targeted benefits, or private charity. But no rich democracy has succeeded with those strategies alone.

It takes taxes, in other words, to raise a child.

David Brady is a professor in the University of California Riverside School of Public Policy and director of its Blum Initiative on Global and Regional Poverty. He is author of Rich Democracies, Poor People: How Politics Explain Poverty, co-editor of The Oxford Handbook of the Social Science of Poverty, and on twitter @DaveBrady72.