America the Undertaxed

The most important debates in U.S. politics today center on the cost and the role of government. Cutting taxes, limiting expenditures, and reducing debt have become the chief concerns of Republicans, whereas Democrats generally seek to preserve or even expand government spending and are willing to raise taxes to do so. The looming expiration of the George W. Bush tax cuts at the end of 2012 and the economy’s weak recovery give these debates special urgency, as decisions made in the next few months are likely to shape the nation’s economic, social, and political trajectory for years to come.

Behind each party’s position lies not only a particular collection of interest groups but also a story about what the government’s role in the U.S. economy is and what it should be. Democrats think Washington can and should play a more active part, using taxation, regulation, and spending to keep the economy growing while protecting vulnerable citizens from the ravages of volatile markets. Republicans, in contrast, think Washington already does too much; they want to scale government back to liberate markets and spur economic dynamism.

When mulling these stories, it can be useful to put U.S. fiscal policy in perspective. Compared with other developed countries, the United States has very low taxes, little redistribution of income, and an extraordinarily complex tax code. These three aspects of American exceptionalism deserve more attention than they typically receive.


The first striking feature of the fiscal state of the United States, when compared with those of other developed countries, is its small size. As of 2009, among the 34 members of the Organization for Economic Cooperation and Development (OECD), a collection of the world’s most economically advanced democracies, the United States had the third-lowest ratio of taxes to GDP (see chart). But it is important to look at pre-recession data, which better reflect long-term trends. In 2006, before the financial crisis struck, OECD tax statistics showed that total taxes in the United States — at all levels of government: federal, state, and local — were 27.9 percent of GDP, three-quarters the percentages in Germany and the United Kingdom and about half of those in Denmark and Sweden. Among the rich democracies in 2006, only South Korea had lower taxes.

The reason for this discrepancy is not that the United States has lower personal income tax revenues than its OECD counterparts. In fact, in 2006, personal income taxes at the federal, state, and local levels in the United States came to 10.1 percent of GDP, just above the OECD average of 9.2 percent. Instead, the disparity results from the low effective rates — or nonexistence — of other forms of taxation. To take one example, in 2006, the U.S. corporate income tax at all levels of government collected 3.4 percent of GDP, compared with an average of 3.8 percent across the OECD. During that same year, according to the OECD, U.S. social insurance taxes brought in 6.6 percent of GDP, compared with an average of 9.2 percent among the OECD nations. Yet the biggest difference between the United States and other OECD countries is in consumption tax revenue. Most U.S. states have sales taxes, and the federal government maintains excise taxes (taxes on such goods as alcohol, cigarettes, and fuel) and customs duties (taxes on imported goods). Yet none of those taxes currently collects the same amount of revenue as a value-added tax (VAT) would (a VAT is a consumption tax that collects revenue from the value added by each business at each stage in the chain of production of a given product). OECD statistics show that VATs bring in an average of 6.7 percent of GDP among the OECD nations, accounting for the majority of the difference in total tax revenues between the United States, which does not have a VAT, and the rest of the OECD.

U.S. tax revenue is not only low but also consistently low, having equaled roughly the same share of the economy for 60 years. Since the tremendous growth of the federal government during World War II, federal tax revenues have hovered around 18 percent of GDP. This stability has also proved to be true of state and local tax levels, which have fluctuated between eight and ten percent of GDP over the same period. Over that time, taxes in the other OECD countries have grown more than in the United States. In 1965, total tax revenues stood at about 25 percent of GDP in the United States and across the rest of the OECD. But by 2000, tax revenue represented 30 percent of GDP in the United States and 37 percent in the rest of the OECD. The enacting of VATs throughout the OECD during the 1960s and 1970s accounts for much of the difference. It also accounts for the steadiness of European tax revenues through the global financial crisis. By 2009, total tax revenues had dropped to 24 percent of GDP in the United States, but they had fallen just two points, to an average of 35 percent of GDP, in the other OECD countries.

Although tax receipts have composed approximately the same share of GDP for decades in the United States, their composition has changed. In particular, the corporate tax has plunged as a source of federal revenues, from 30 percent in the 1950s to ten percent today. As Republicans are quick to point out, the United States does have one of the highest statutory corporate tax rates in the developed world. Combining the federal and state levels, the top rate of these taxes is 39 percent, compared with an average of 36 percent across the G-7 and 31 percent across the OECD. Yet as with the individual income tax, the United States applies these statutory rates to a narrower base of taxpayers than other advanced countries do, due to various corporate tax credits and breaks, such as the accelerated depreciation of machinery and equipment and the deferral of taxes on income earned abroad. As a result, according to a report issued by the U.S. Treasury Department, between 2000 and 2005, on average, U.S. businesses paid an effective tax rate of only 13 percent, nearly three percent below the OECD average and the lowest rate among the G-7 countries.

Whereas corporate tax revenues have fallen, revenues from payroll taxes for programs such as Social Security and Medicare have grown. The Urban-Brookings Tax Policy Center found that these taxes rose from 23 percent of federal revenue in 1970 to 40 percent in 2010. In fact, the majority of Americans pay more in payroll taxes than in federal income taxes. This is the case in part because the United States imposes payroll taxes on all wages without the exemptions and deductions so common to individual and corporate income taxes and in part because the Earned Income Tax Credit, which helps offset the federal income and payroll taxes of low-wage workers, reduces or eliminates income taxes for many with low earnings.

Even as payroll tax revenues have risen, the individual income tax, which in 2010 accounted for 42 percent of national revenue, has remained the main source of federal income. According to the Urban-Brookings Tax Policy Center, for decades prior to the Bush tax cuts of 2001-3, despite many alterations to tax bases and rates, the individual income tax provided a steady and large percentage of federal revenue. That is because the government tended to compensate for changes in rates by expanding or shrinking the tax base when necessary. During the 1970s, the tax code featured 25 income brackets and a top rate of 70 percent. Legislation passed during Ronald Reagan’s presidency reduced the number of brackets to just two, dropped the top rate to 28 percent, ended a number of tax breaks, and pegged the brackets to inflation, ending so-called bracket creep, in which inflation forced taxpayers into higher tax brackets even though their real incomes stayed flat. President George H. W. Bush brought the top rate back up to 35 percent, and President Bill Clinton further raised it to 39.6 percent, but each administration added a number of new tax breaks, from an expansion of the Earned Income Tax Credit to a credit for a child’s tuition. The Bush tax cuts reduced taxes on capital gains and dividends and on estates and cut the top tax rate yet again, to 35 percent.

The largest tax reductions from these changes went to high-income households. In fact, the United States currently taxes top earners at some of the lowest effective rates in the country’s history. Data from the Internal Revenue Service (IRS) show that the top one percent of taxpayers paid an average federal income tax rate of 23 percent in 2008, about one-third less than they paid in 1980, despite the fact that their incomes are now much higher in both real and relative terms. Although the rich enjoyed by far the largest tax cuts, the middle class is also paying lower taxes. In 2011, the effective federal income tax rate for a family of four with a median income was just 5.6 percent, compared with 12 percent in 1980. And because of the Earned Income Tax Credit, about 40 percent of low-income U.S. households do not pay any federal income tax.

Altogether, the adoption and continuation of the Bush tax cuts has slashed federal revenues by about three percent of GDP, to levels not seen since shortly after World War II. As a result, the individual income tax now constitutes a smaller share of the economy than it did 30 years ago, falling from 10.4 percent of GDP in 1981 to 8.8 percent in 2005. By permitting extensive loopholes, failing to create effective consumption taxes, and cutting individual income taxes, the United States has created a tax system that collects far less revenue relative to GDP than many of its OECD counterparts.


On the surface, U.S. tax policies seem quite progressive. The individual income tax, for example, is scaled to match relative earnings, so that those who take in less income pay less in taxes and those who earn more income pay more in taxes. And as mentioned above, the United States has not implemented a VAT, which is considered regressive because lower-income households tend to spend everything they earn, meaning that the VAT takes a greater share of their earnings than it does for high-income households, which spend only part of their incomes and save the rest. But appearances can be deceiving unless fiscal policy and government spending are considered together. In Europe, regressive taxes are matched with highly redistributive states. In the United States, mildly progressive taxes are matched with a not very redistributive state. As a result, the United States experiences greater inequality than most other advanced nations, with the tax-and-transfer system doing little to alleviate it.

The Occupy Wall Street movement has cast into sharp relief the vast and growing income inequality in the United States. Analyzing IRS data, the economists Thomas Piketty and Emmanuel Saez have found that the share of total income going to the top one percent of earners — those with annual incomes of $400,000 or more — increased from nine percent in 1970 to 23.5 percent in 2007. The 2007 amount represented the highest level in the United States since 1928 and exceeded those for Europe and Japan that year, where the share going to the top one percent of earners was 11 percent in Germany, 9 percent in Japan, 8.7 percent in France, and 5 percent in the Netherlands. Although the 2008 financial crisis reduced the incomes of the top one percent in the United States by a fifth, by 2010 their earnings had largely recovered. And wealth is even more concentrated than income. According to the economist Edward Wolff, in 2007, the top one percent in the country earned just over 20 percent of all income but held more than 30 percent of all wealth.

As the top has risen, the bottom and the middle have faltered. Congressional Budget Office data show that between 1979 and 2007, before-tax incomes increased by 240 percent for the top one percent but by just 20 percent for the middle fifth of earners and by ten percent for the bottom fifth. Although the bottom 90 percent lost less income than the top one percent as a result of the financial crisis, their earnings have not recovered as much as those of the top earners. In fact, according to Saez, average income in the bottom 90 percent remains at the lowest level since 1983. This dearth of earnings exists in large part thanks to the prevalence of low-wage work in the United States compared with other advanced countries. The Center on Budget and Policy Priorities, a research and policy institute on fiscal matters, has found that almost one-third of Americans have “low incomes,” meaning ones below 200 percent of the poverty line. By analyzing OECD data, the economist Timothy Smeeding discovered that the proportion of full-time workers earning less than 65 percent of the median wage in the United States — around 25 percent — is twice as high as in France and Germany and five times as high as in Finland and Sweden.

To be sure, other advanced democracies also suffer from high market-generated inequality. The economists Smeeding and Katherin Ross Phillips have shown that rates of market-income poverty — the proportion of people living in households with incomes that are below 40 percent of the median disposable income — are quite high across the advanced democracies. In fact, according to Smeeding and Ross Phillips, the United States’ market-income poverty rate of 17.2 percent for working-age adults is only slightly higher than Germany’s (14.9 percent) and Sweden’s (15.8 percent) and even lower than Canada’s (18.4 percent) and the United Kingdom’s (25 percent). Yet unlike the United States, these countries reduce such market-generated poverty through redistribution — less through their tax systems than through their social welfare programs. Although each of these countries imposes generally progressive personal income taxes, they earn much of their revenue from regressive VATs. But these nations steer income from their VATs to wide-ranging social safety nets, thereby redistributing income more evenly across their societies. Smeeding and Ross Phillips found that after the implementation of universal transfer and social assistance programs, poverty among those aged 25 to 64 fell: to 6.9 percent in Canada, 5.9 percent in the United Kingdom, 3.5 percent in Germany, and 1.8 percent in Sweden. Meanwhile, in the United States, it remains at 10.9 percent. Working-age populations in most advanced countries pay higher taxes but receive extensive benefits from social welfare systems, meaning reduced poverty and inequality.

The United States, on the other hand, appears at first glance to have a more progressive tax system than many other OECD members — one that, by shifting tax burdens from the poor to the rich, would theoretically alleviate poverty along the same lines as redistributive programs. Yet the overall U.S. tax system is only mildly progressive. In part, that is because the United States carries out much of its redistributive effort not through social programs but through tax expenditures, that is, forgone revenue in the form of tax credits, deductions, and exemptions. By introducing such measures, the government aims to subsidize certain economic activities. For example, the interest deduction for individuals with home mortgages supports homeownership, and the tax credit for research and experimentation subsidizes corporate innovation. Such expenditures now amount to $1.1 trillion in forgone revenue per year, more than the sum raised by the individual income tax.

The problem is that such expenditures do not reallocate income and resources effectively. True, the Earned Income Tax Credit assists the working poor, and other tax breaks, such as the Child Tax Credit (a reduction in taxes of up to $1,000 for each dependent child under the age of 17, which phases out at higher incomes), benefit many middle-income households. But the majority of the largest tax expenditures help the affluent the most. The Urban-Brookings Tax Policy Center estimates that in 2011, households with incomes in the top fifth of the income distribution received two-thirds of the benefits from tax expenditures, with the top one percent receiving a quarter of them. Much of this disparity has to do with the fact that only 30 percent of tax expenditures are credits, in which the size of the break is the same regardless of income. The other 70 percent of tax expenditures are deductions from taxable income, the value of which rise at higher incomes. For example, a family paying $5,000 in home mortgage interest would receive a $1,750 tax break if it belonged to the 35 percent tax bracket but only a $500 break if it belonged to the ten percent bracket.

In addition to containing tax breaks that favor the rich, the U.S. tax code includes many regressive and flat taxes, which tend to exact more from the poor. These include the Social Security and Medicare payroll taxes, income and sales taxes in many states, and many others. When such taxes are combined with the reduction of progressive taxes for high-income households, the overall tax system — federal, state, and local — becomes only slightly progressive. This means that the share of taxes paid by each income group essentially resembles the share of income that it receives, which would not be the case in a more progressive system. According to the Institute on Taxation and Economic Policy, a fiscal think tank, in 2011, the lowest fifth of earners received 3.4 percent of total income and paid 2.1 percent of total taxes, the middle fifth received 11.4 percent of income and paid 10.3 percent of taxes, and the top one percent received 21 percent of income and paid 21.6 percent of taxes.

The United States achieves comparatively little redistribution through social programs as well, devoting less than many other advanced countries to such services. Although Social Security and Medicare reduce poverty among the elderly, working-age people receive fewer universal benefits than those in other countries, from national health insurance to paid leave for new parents. Outside of Medicaid and food stamps, programs that have grown in recent years, most social assistance initiatives meant to aid the poor remain small and have fallen in real value over time. That is why, despite having nearly the highest per capita GDP in the world, the United States has the highest poverty rate among rich nations.


The United States’ fiscal state is not only small and minimally redistributive compared with those of other advanced nations; its tax code is also far more complicated. This complexity makes it much more difficult and costly for households and businesses to comply with the rules. And it undermines public trust in the system, with taxpayers fearing that those with better knowledge of how to navigate the system of loopholes, particularly the rich, get away with not paying their fair share.

To get a sense of the immense complexity of the U.S. system, consider that the Internal Revenue Code is almost 12 times as long as the New Testament. The 2012 instruction booklet for the 1040 individual tax form and the accompanying schedules is 188 pages long; the one for the “EZ” form is 43 pages. A 2003 IRS report estimated that individual taxpayers pay $18.8 billion, spending 25 hours and $149 per taxpayer, on compliance each year, mainly through payments made to accountants and agencies to help them sort out their taxes. Corporate taxes are even denser. According to the U.S. Treasury Department, large corporations spend over $40 billion annually on compliance.

The tax expenditure system is one major reason for this intricacy. Each exclusion, deduction, and credit adds another layer of complexity to the code and increases the time needed to prepare one’s taxes. Taxing families rather than individuals complicates the system even more, since this method makes it difficult to calculate the proper tax withholding for each partner. Another culprit is the Alternative Minimum Tax (AMT), which is meant to prevent high-income taxpayers from using exclusions and deductions to avoid paying the federal income tax. This system requires some taxpayers to calculate their tax liability twice, once under the regular income tax and then again under the AMT, and pay whichever is higher. Although meant for those in the highest tax brackets, the AMT now affects more ordinary taxpayers because the government did not peg its parameters to inflation, and some very high earners have tax rates outside the parameters of the AMT, allowing them to pay the regular tax instead.

As the economists Joel Slemrod and Jon Bakija have noted, many peers of the United States have put in place far less complex tax systems, making compliance much easier. Several dozen countries utilize a return-free system for most taxpayers. They can do this because they tax individuals rather than families and withhold taxes from both income in the form of wages and income in the form of interest. These two features allow countries to precisely calibrate withholdings to actual tax liabilities. It is also much easier to implement such a system when governments hand out tax breaks either as credits that are equal for all taxpayers or as deductions that are set at a flat rate. In the United Kingdom, for example, the equivalent of the U.S. home mortgage interest deduction is a flat 15 percent reduction on the interest paid on the mortgage applied directly by the bank. Meanwhile, several countries, such as Australia, the Netherlands, and the United Kingdom, require any changes in tax rules to include estimates of the costs of compliance.

To get another sense of the difference between the United States and other developed countries, consider the subsidization of the cost of raising children. Many advanced-country governments calculate and send allowances to families with children. In the United States, however, households with children must navigate and administer a complex system of tax breaks themselves, such the Child Tax Credit and the Earned Income Tax Credit. And if they file their returns incorrectly, the IRS may fine them. That is why even many low-income families rely on tax-preparation firms, which have lobbied against periodic attempts to simplify the tax-paying process, such as having the IRS mail out draft returns for households in straightforward situations. Perhaps it comes as no surprise that when a 2003 survey by National Public Radio, the Henry J. Kaiser Family Foundation, and Harvard University asked Americans what bothers them most about taxes, respondents were more than twice as likely to cite the complexity of the system as the amount they pay.


The central debate in U.S. politics is whether to keep taxes, particularly federal taxes, at their current levels in the long term or emulate other advanced nations and raise them. In making this choice, Americans will undoubtedly have to prioritize some values and programs over others. But the polarized conversation about this issue thus far has clouded understanding of the actual likely consequences of each course.

If the government fails to raise more revenue, it will no longer be able to afford programs that many Americans say they want, from Social Security to Medicare. A 2008 survey of the American National Election Studies, a biennial study of national election outcomes, showed that a majority of Americans supported increased spending in such areas as financial aid for college, support for the poor, and improved public education. Yet without greater financial flexibility, the U.S. government cannot meet those demands. The consequences of the current system are thus known: the need to slash many of the social safety programs that Americans have come to depend on.

Congressman Paul Ryan (R-Wis.) has spearheaded a movement of budget hawks proposing to address this budget crisis by reducing federal spending to 16 percent of GDP by 2050 (it was 24 percent in 2011). This would represent the lowest level of spending since World War II, before Congress introduced Medicare, Medicaid, the interstate highway system, and most federal aid for education. Although Ryan’s cuts might alleviate the revenue problem, low earners would bear the brunt. According to the Center on Budget and Policy Priorities, 62 percent of the spending reductions would affect low-income households, which would also face higher federal taxes due to a reduction in the Earned Income Tax Credit. Meanwhile, Ryan’s plan would give those with incomes over $1 million a tax cut of $265,000, and that is on top of the Bush tax cuts already in place.

President Barack Obama, by contrast, has suggested boosting tax revenues, proposing to raise the rates of the top two tax brackets. Another way to add revenue would be to introduce new tax brackets. This would allow the government to differentiate among extremely high levels of income better than it can with the current system, which sets the top bracket at $388,350 and does not distinguish between the merely affluent and the superrich. In 2007, the top one percent paid an average federal income tax rate of 22.4 percent. As the economists Peter Diamond and Saez have noted, increasing that rate to 29.4 percent would raise one percent of GDP in additional revenue; increasing it to 43.5 percent would raise three percent of GDP and still leave the share of after-tax income for these top earners at a level that is more than twice as high as in 1970.

What, then, are the consequences of changing the tax system? The first question is whether higher taxes would harm the economy. Taxation affects economic growth in complex ways. To be sure, at very high rates, taxes can drag down the economy, and government activity and investment, fueled by tax revenues, can drive it. But the connection between relative rates of taxation and economic prosperity may not be as strong as both supporters and opponents of tax hikes think. Slemrod and Bakija found little correlation across the OECD countries between taxes as a percentage of the economy and the size of the economy itself, as measured by per capita GDP. Nor, according to their research, is there a high correlation between taxes as a percentage of GDP and the annual rate of economic growth. There is some evidence of a similar lack of correlations across time in the United States. The U.S. economy grew at a faster pace before the Bush tax cuts rather than after, and it grew more in the 1950s and 1960s than it has recently, even though taxes as a percentage of the economy were the same then as they are today. Raising taxes moderately — perhaps by a few percentage points of GDP — would certainly provide the government with much-needed revenue. And it might not have a detrimental impact on the U.S. economy, perhaps even spurring it.

Similarly, many worry that higher taxes would cause people to work less productively or spend less money. Yet there is little evidence that tax rates affect the participation of either middle- or high-income individuals in the work force. And despite higher taxes, higher earners ultimately did not spend much less during the 1990s, since the total income of the top one percent during that decade rose. Somewhat higher taxes would likely alter behavior to some extent. But with the earnings of the top one percent mostly back to their pre-recession levels, past experience suggests that a tax hike today would not severely damage the economy, and productivity might even rise with the security and investments that government spending can provide.

In proposing all these options, from reducing spending to raising revenue, policymakers are confronting the reality of U.S. fiscal policy: compared with its counterparts among the advanced nations, the United States’ tax system collects little revenue, poorly redistributes that money across the population, and is mind-bogglingly complex. The decision of whether to change that system is a political one; whether and how to undertake tax reform is ultimately a referendum on the direction in which Americans would like to take their country.