President Joe Biden’s expansive American Families Plan, announced last month and now being debated in Congress, proposes almost $2 trillion in spending on a wide range of social services, from paid family and medical leave, subsidized child care, and universal prekindergarten to free community college and health insurance subsidies. Amidst these ambitious initiatives, it is easy to miss adjustments to the federal tax code. But one quieter proposal—extending the already increased child tax credit that came into law in March through the American Rescue Plan Act—has dramatic implications for reducing child poverty and supporting working families, and more broadly for the way we think about the provision of social welfare in the United States.

For almost a quarter century, from 1997 until the American Rescue Plan this March, the child tax credit systematically and categorically excluded the neediest families from support.

The March stimulus bill made three major changes to the standard child tax credit, which has been in effect since 1997. First, it raised the value of the credit for low- and middle-income families by more than 50 percent, from $2,000 for each child under age 17 to $3,000 for each child aged 6 to 17 and to $3,600 for each child under age 6. Second, it made the credit “fully refundable,” meaning that families who owe less in taxes than the credit is worth—a common situation for low- or zero-income earners—will get a refund for the full difference, instead of a reduced maximum that had been in effect since 2001. And third, it adopted a mechanism to pay out half of a family’s estimated credit in advance monthly installments throughout the second half of 2021, providing assistance throughout the year rather than all at once when families file their tax returns next spring. All three of these changes are set to expire at the end of the year, but the American Families Plan would extend them through 2025.

The changes may seem small, but they are likely to have huge consequences. Along with other COVID-19 relief provisions such as stimulus payments and liberalized food aid and unemployment insurance, the credit will make an enormous and immediate difference in the lives of thousands of working families struggling through the deeply uneven economic crisis brought on by the pandemic. According to Columbia University’s Center on Poverty and Social Policy, the combined effect of these policies could reduce the absolute level of child poverty as well as the racial poverty gap between white and non-white children by about 50 percent. The legislation will have the strongest impact among families of color, reducing the poverty rate for Hispanic children from 22.3 percent to 9.2 percent and for Black children from 24.8 percent to 9.7 percent, while for white children the poverty rate will fall from 9 percent to 3.1 percent.

Beyond this expansion of material support, the credit is also likely to breathe new life into discussions of the U.S. welfare state. Though the revamped credit is hardly a revolutionary step toward universal basic income, as some have claimed, its monthly payments do mark a significant innovation in the history of the U.S. social policy: they establish, for the first time, the functional equivalent of family allowances—direct state payments to families with children—that have been a core component of postwar states in Western Europe and elsewhere. Until this year, few would have considered such a measure to be within the realm of political possibility. Now that it is on the books, it may help to expose fundamental problems with what scholars have called the aggressive “fiscalization” of U.S. social support through the income tax system.

One of those problems is a basic matter of awareness of state support. Many parents are not even aware they qualify for subsidies such as the child tax credit and the earned income tax credit, and as a result millions of dollars of credits go unclaimed each year. Even those families who do file an annual tax return may not realize how much state support they receive—or that they are getting a break from the state at all. Burying social provisions in the tax code thus obfuscates not just the availability but the sizable extent of federal social support, leaving social spending vulnerable to attacks from the right and center and distorting citizens’ relationship to the state.

Routing social policy through tax breaks disproportionately benefits whiter, wealthier families who have a tax burden in the first place.

Another problem with this system, perhaps even more significant, is its uneven distributional profile. Routing social policy through tax breaks disproportionately benefits whiter, wealthier families who have a tax burden in the first place. Making the child tax credit fully refundable helps to correct this upward bias, removing the cruel cap on the refundable part of the credit that has kept millions of dollars of support from the neediest families for a generation. But we also need a deeper reckoning with the legacy and limitations of trying to pursue social policy through the back door of taxation, instead of through the front door of direct services.

A large and growing body of scholarship helps to illuminate these peculiar features of U.S. social governance. Contrary to the conventional portrait of a weak, underdeveloped, and underfunded U.S. welfare state, studies such as Christopher Howard’s The Hidden Welfare State (1999) and Suzanne Mettler’s The Submerged State (2011) have shown that U.S. spending on social services is not unusually low among wealthy nations once we account for its hidden elements in the U.S. tax system. Whereas classical welfare states levy high individual and corporate income taxes and then redistribute those funds through direct—and sometimes universal—public programs, the United States often forgoes collecting that revenue altogether, leaving it to be spent toward similar ends on the private market.

This system works via the U.S. tax code’s dizzying array of deductions and credits, from mortgage interest deductions and retirement savings credits to tax-deferred savings plans, the child tax credit, and dozens of others. Even the terminology can be difficult to master. Credits, only some of which are “refundable,” reduce your tax burden directly, dollar for dollar: if you owe $5,000 in taxes but qualify for a $3,000 credit, for example, you reduce your tax bill to $2,000. Deductions, by contrast, indirectly reduce your tax burden by lowering your taxable income, whether through itemization or the standard deduction. (A related category of tax breaks, personal exemptions, was eliminated through 2025 by the 2017 Tax Cuts and Jobs Act under Donald Trump.)

Many of these tax breaks are designed to facilitate specific social outcomes. The mortgage interest deduction, for example, incentivizes home ownership. Other tax breaks target social goals more typically associated with the welfare state. In lieu of a tax-financed system of universal health care, for instance, Americans can deduct expenses related to employer-provided health insurance. And though Americans lack universal government-funded child care (a situation the American Families Plan may change), working parents can deduct much of the expense of child care services purchased on the private market. In 2017 there were 80 such tax breaks, worth over half a trillion dollars each year. Seen in this light, social spending in the United States is not anemic but comparatively high, amounting to 29.6 percent of U.S. gross domestic product in 2017. That outstrips countries we tend to think of as exemplars of the welfare state, including Sweden and New Zealand, and it is far beyond the OECD average of 20.7 percent. Indeed, by this measure, the United States has the second highest rate of social spending among all OECD countries, trailing only that of France.

The tax code obscures not only the reach of the state in daily life but also the fundamental failure of capitalism to sustain social reproduction without state intervention.

The present system is the result of countless alterations in the convoluted U.S. tax code that have taken place over decades, typically in congressional revenue committees insulated from public scrutiny. These routine budgetary decisions rarely attract much in the way of press coverage, and their fairly anonymous nature means they are often heavily influenced by lobbyist pressure. The practice has become so prevalent in budgetary policymaking because it is often amenable to both sides of the aisle: Democrats achieve social welfare goals, and Republicans are satisfied by tax cuts and a greater role for the private sector to fulfill public needs. The historical development of the child tax credit illustrates how policymakers have used the tax deferrals for social policy ends, as well as the gross inequities that are often built into this system by design.

The credit’s origins lie in the 1940s, when the horrors of war and depression sparked a transnational effort to rethink the role of the state in the economy. A central pillar of this postwar period of social planning was the idea of family allowances: a monthly governmental payment to families with children.

Such policies had originated in France and Belgium in the late nineteenth century, originally as a voluntary practice among industrial employers. In 1932 the French government responded to union pressure to make these allowances mandatory for all wage earners, but before World War II most industrialized democracies provided more targeted benefits, whether “mother’s pensions” to widowed mothers or means-tested payments to very poor families. In the United States, policy in this vein first emerged on the federal level in 1935 with the Aid to Families with Dependent Children (AFDC) provision of the Social Security Act, which paid a direct subsidy to families with children when the father was deceased, absent, or incapacitated.

In the late 1940s many planners sought to make family allowances universal for all children regardless of parental profession or income. This vision got its most influential elaboration in the work of British economist William Beveridge, whose 1942 report Social Insurance and Allied Services laid the foundation for the postwar British welfare state. Citing a mixture of social egalitarianism, pronatalist concerns for future of the “British race,” and the macroeconomic imperative of protecting against income disruptions, Beveridge proposed a flat payment of eight shillings per child per week (enough to provide each child with a minimum of food, clothing, fuel), financed out of general taxation. During the war the United Kingdom and numerous commonwealth countries, including Canada and Australia, put Beveridge’s proposals into law with family allowance legislation. Immediately after the war, many countries in Europe followed suit, and by 1949 fifteen industrialized countries had a system of family allowances in place.

Last year virtually all children in households from the top half of the income distribution qualified for the full child tax credit, while the vast majority from the poorest 10 percent qualified for no credit whatsoever.

In the United States, it was a different story. Despite a wave of American interest in the Beveridge plan and a spate of legislative proposals for enhanced social security throughout the 1940s, family allowances would prove a political dead letter. After touring the country in 1943, Beveridge himself noted with exasperation that family allowances “had hardly yet risen above the horizon of discussion.” Regional cost-of-living disparities, bipartisan faith in postwar affluence, and pervasive racism all worked to keep family allowances from passing in a Congress dominated by Republicans and openly white supremacist southern Democrats.

Progressive social reformers of the era were well aware that race was the rock that would sink the prospect of family allowances. Economist and social security expert Eveline Burns wrote in 1945 that income guarantees faced “a real technical obstacle due to the very great variations in levels of wages and standards of living in different parts of the country, between different occupational groups and, it must not be forgotten, between different racial groups.” It was questions of race above all, Burns argued, that “obstruct adoption of any simple and single flat rate system such as Beveridge proposed.” Family allowances would “confer an especial benefit on Negroes,” which would encounter “emotionally based opposition,” especially from the white Southern public and Southern legislators. “Here, as in so many other fields of social policy,” Burns wrote, “national progress may be held back by the unsolved social and economic problems of the South,” which were of a “fundamental and long-run character.”

At the same time, however, many lawmakers and their economic advisors had become convinced that Keynesian demand management was the key to sustained postwar economic growth. But how to stimulate demand without income guarantees? They found a solution in the tax code: rather than legislate direct family allowances, families would be permitted to reduce their taxable income for each child.

The idea of exemptions was not new; personal income tax exemptions for dependents had been introduced in 1917 as way of excluding most people from tax liability. Indeed, prior to World War II, only very wealthy Americans, representing less than five percent of the population, owed federal income tax. During the war, however, defense spending and the threat of inflation led Congress to expand the tax base dramatically, subjecting working- and middle-class Americans to federal income tax liability for the first time. A mere 3.9 million Americans owed income tax in 1939. Four years later, that number had ballooned by an order of magnitude to over 40 million.

In this environment, child-based tax exemptions might stimulate consumer spending—so as Congress made more people subject to income tax, it also liberalized exemptions for dependents. “Increased exemptions,” one tax authority wrote in 1945, were designed so that “individual surpluses may be increased and spending enhanced” to sustain postwar consumption. Three years later Congress expanded dependent exemptions yet again, and they became a key feature of the postwar federal tax code. The policy, originally intended to exclude all but the rich from owing income tax, thus took on a new function in a new age of mass taxation and mass consumption: it augmented disposable income and purchasing power while bypassing the political hurdles of Beveridge-style family allowances.

Along with other COVID-19 relief provisions, the recent changes to the child tax credit could reduce child poverty by half.

By propping up the ideal of a white nuclear family headed by a unionized male breadwinner, these exemptions formed an indispensable, albeit largely invisible, bedrock of the Keynesian-Fordist industrial economy of the postwar decades. Like family allowances in much of the rest of the noncommunist Global North, dependent exemptions in the United States boosted the working-class purchasing power that fueled the consumption-driven economic order. But largely shut out of the postwar boom were mostly nonwhite, female-headed households that relied on stigmatized, means-tested social assistance such as AFDC.

This system began to erode in the 1970s. Deindustrialization, deunionization, and rising divorce rates worked in tandem to undermine the Fordist family model. Meanwhile, rampant inflation reduced working-class earning power and the value of child benefits throughout the industrialized Global North; in the United States, in particular, the dependent exemption lost a staggering 61 percent of its purchasing power between 1948 and 1984. At the same time, an increasingly ubiquitous logic of neoliberal austerity intensified policymakers’ hostility to direct social transfers. This policy orientation took hold throughout the developed world, but it was especially salient in the United States, where “welfare” was portrayed as sustaining a nonproductive racialized urban underclass.

The result was a widespread trend throughout OECD countries toward what Tommy Ferrarini and other scholars have called the “fiscalization” of social policy. By the 1990s and early 2000s, most wealthy countries cut direct social spending, replacing them with new or expanded tax breaks to fulfill social policy objectives. Some countries, such as France, Belgium, and the Netherlands, adopted a mixed model combining traditional social benefits with tax-based fiscal mechanisms. Other countries, such as Canada, entirely replaced universal systems of family allowances with income tax credits, exemptions, or deductions.

As Joshua T. McCabe has recently shown, in the United states this program of fiscalization appealed to neoliberal policymakers in both parties eager to rein in spending, combat “welfare dependency,” and find market-based solutions to social problems. In 1991 the both the Urban Institute and the National Commission on Children published influential reports that proposed supplementing the existing dependent exemption with a child tax credit. The proposals caught the attention of antitax lawmakers such as Representative Patricia Schroeder, a Democrat from Colorado who chaired the House Select Committee on Children, Youth, and Families. The same year the reports were published, Schroeder called hearings to “evaluate alternative tax policies that can make the tax code more ‘family-friendly,’” and suggested that an increased child tax break could serve as a way of “reclaiming the tax code for American families.”

One important element of this debate was the question of refundability. The proposals for child tax credits that first gained traction in the late 1980s were fully refundable, and many countries that replaced family allowances with tax credits, including Canada, adopted such a scheme. In the United States, however, social and fiscal conservatives balked at a direct payment to low-income families. In the 1991 hearings, conservative lobbying groups such as the Family Research Council and the American Enterprise Institute opposed refundable credits on the ground that they were not tax breaks for the middle class but rather “welfare-style cash transfers” in disguise. Most Republicans followed suit, along with many Democrats. Schroeder herself conceded that the credits should “target relief to middle-income parents” and not “increase welfare under the rubric of pro-child tax relief.”

By the time Bill Clinton took office in 1993, the direction of political change was clear. As sociologist Melinda Cooper argues in her recent book Family Values (2019), proposals for child tax credits found fertile ground in an era in which neoconservatives and neoliberals converged on the need for market-based solutions to prop up the traditional family. Such policies fit the “triangulation” politics of the Clinton administration and a 1990s-era political zeitgeist marked by bipartisan commitments to austerity and to dismantling systems of direct welfare benefits, and three years after coming into office Clinton fulfilled his campaign pledge to “end welfare as we know it” by killing AFDC with the 1996 Personal Responsibility and Work Opportunity Reconciliation Act.

The child tax credit is a prime example of how policymakers in the United States legislated uneven tax breaks where other wealthy countries have traditionally used direct social spending.

As with the contemporaneous debate over welfare, race was the implicit but obvious subtext of these debates over refundability. Child tax credits were a great tool for promoting austerity and downsizing government when they went to the white middle class. But a refundable credit, which might be of special benefit to low-income people of color, too closely resembled the form of welfare that both parties agreed had to be destroyed. After defeating welfare in 1996, the Clinton administration was not going to waste political capital by pushing for a refundable credit that critics would only smear as AFDC under a new name. Clinton thus signed a $500 per child tax credit into law with the 1997 Taxpayer Relief Act, but made it nonrefundable: families would only get a benefit if they made enough to owe federal income tax.

In 2001 the credit was made partially refundable for those who owed less tax than the credit was worth, and numerous Congressional acts throughout the 2000s and early 2010s reduced the threshold for refundability. The Trump-backed Tax Cuts and Jobs Act of 2017 further expanded access to the credit to some low-income families, but it also required recipient children to have a valid Social Security number. As a result, an estimated one million undocumented children became ineligible for the credit. Regardless of citizenship, children from families that earned too little to owe federal income tax remained excluded.

The result was that for almost a quarter century, from 1997 until the American Rescue Plan this March, the child tax credit systematically and categorically excluded the neediest families from support. An October 2020 study by Jacob Goldin and Katherine Michelmore for the National Bureau of Economic Research found that, before the March updates, virtually all children living in households from the top half of the income distribution were eligible for the full credit, while the majority of households in the bottom 30 percent were eligible for only a partial credit—and the vast majority of children from the poorest 10 percent of households were not eligible for any credit whatsoever. Children of color have borne the greatest brunt of this policy. Goldin and Michelmore’s study found that only half of Black and Hispanic children were eligible for the full credit, compared with over three quarters of white and Asian children. Black children were particularly unlikely to qualify. One in four children ineligible for the child tax credit were Black, despite comprising only 14 percent of all children.

As this long history shows, the child tax credit is a prime example of how policymakers in the United States employed targeted and uneven tax breaks where other wealthy countries have traditionally used direct social spending. The “hidden” welfare state thesis contends that this forgone tax revenue forgone is functionally equivalent to direct welfare spending, achieving the same social ends by different financial means. But the argument is not without its critics. Some scholars have argued that tax reductions are more reflective of an attenuated public sector rather than state power operating through indirect and subtle channels. Sociologist Monica Prasad, for one, has questioned whether “not taking equals giving.” Where some see a welfare state in disguise, skeptics such as Prasad see the “reprivatization of risks that had previously been met collectively and the waning of state capacity.” Others focus on the implications for political knowledge and engagement. In The Submerged State, for example, Mettler, argues that if Americans were only more aware of the support they receive from the federal government through tax deferrals, voters would be less hostile to social spending.

By occluding the state mechanisms on which capitalism depends, the hidden welfare state makes markets appear more naturally stable than they really are.

This is certainly right, but it arguably doesn’t go far enough. The tax code obscures not only the reach of the state in daily life but also the fundamental failure of capitalism to sustain social reproduction without state intervention. Theorists such as Nancy Fraser, drawing on Karl Polanyi’s analysis of the “double movement” of public protections to mitigate social dislocations caused by market forces, have shown how social policies such as family allowances are a means of resolving one of capital’s central contradictions: its tendency to undermine the family and the ability of workers to reproduce themselves and their labor power. By occluding the state mechanisms on which capitalism depends, the hidden welfare state makes markets appear more naturally stable than they really are.

Less emphasized, but by far the most insidious feature of the hidden welfare state, is its heavy tilt in favor of white, wealthier Americans. Hiding social provisions in the tax code disproportionately benefits Americans with significant tax liability. There are isolated exceptions—most notably the earned income tax credit, a fully refundable credit for families with very low incomes. But in general, because higher-income households typically pay more federal income tax in both absolute terms and as a proportion of income, they also benefit more from the tax deductions and credits they receive. Indeed, the value of these subsidies for the wealthy can be many times that of lower-income workers. As political scientist Chris Faricy has pointed out, an employee in the top tax bracket who makes a $10,000 contribution to a private pension plan might receive $3,900 in tax subsidies, while a worker in the bottom tax bracket would only get $1,000.

When tax subsidies are structured this way, our system of progressive taxation often perversely works to the disadvantage of low-income earners: the less money you make, the less tax you pay, and thus the less you stand to benefit from tax breaks. This becomes most egregious when income dips below the minimum threshold for federal income taxation. As of 2018, as many as 44 percent of American adults—often those with low incomes or outside of the formal labor market—had no federal income liability at all, and this percentage is unquestionably higher due to the economic disruptions of the COVID-19 pandemic. Many such families are shut of out the advantages of the hidden welfare state altogether.

Where do we go from here? It is too soon to say whether Biden’s agenda marks a decisive break with the pervasive logic of market fundamentalism that has informed policy for four decades. But if such a transformation does come to pass, it will depend on not just legislation and policy from the top, but also on a fundamental reorientation in the political culture of the electorate regarding the reach and limits of state power. And such a reorientation will only be possible through highly visible state solutions that have a tangible and measurable effect on people’s lives.

We need to reckon with the legacy and limitations of trying to pursue social policy through the back door of taxation, instead of through the front door of direct services.

The changes to the child tax credit under the American Rescue Plan are a step in the right direction—toward a welfare state that dispenses benefits as a universal right of citizenship rather than mystifies them through an inequitable tax code. Extending the changes to 2025 via the American Families Plan would help, but removing the expiration date—making it harder for Republicans to overturn them—would be even better. In March, six congressional Democrats, including Senators Sherrod Brown and Cory Booker, released a statement calling the expansion of the child tax credit “the most significant policy to come out of Washington in generations,” urging Biden to “provide a lifeline to the middle class and to cut child poverty in half on a permanent basis.” After announcing the American Families Plan in April, Biden indicated support for making refundability permanent, but not the other changes. To ramp up the pressure, Richard Neal, chair of the House Ways and Means Committee, has introduced a competing bill that would go all the way.

Allowing these important changes to expire, whether at the end of the year or in 2025, would squander a rare opportunity to contribute to a lasting and substantial overhaul of the social welfare system. To add up to a true paradigm shift, they need to outlive the current crisis.

Independent and nonprofit, Boston Review relies on reader funding. To support work like this, please donate here.