February 11, 2019

The Cost of a 70 Percent Marginal Tax Rate

Proposals for raising the maximum statutory tax rate to 70 percent, with dual objectives of generating additional revenue for new spending and curtailing inequality, are gaining attention among both academics and media commentators.

While the idea might be superficially attractive, there is little reason to believe that a 70 percent marginal tax rate will generate greater revenue or help lower-income Americans—or that it will accomplish these goals with minimal effect on the economy. On the contrary, similar marginal tax increases in neighboring Canada in 2015 actually produced less revenue: high-income earners paid billions less owing to behavioral adjustment. Higher tax rates, broadly speaking, discourage people from acquiring skills that could increase income and discourage businesses from expanding. In addition, research on the effects of these tax rates shows they will lower wage growth for lower-income American workers. Comparisons to historic norms about tax rates ignore the past four decades of globalization and international tax changes. Lastly, in the context of new spending proposals, this tax change is neither sufficient nor desirable.

Problems with Using a "Soak the Rich" Optimal Tax Rate

Economists such as Paul Krugman were quick to defend dramatically hiking the marginal tax rate for high earners as a smart and sensible idea. Krugman stated “that we shouldn’t care what a policy does to the incomes of the very rich. A policy that makes the rich a bit poorer will affect only a handful of people, and will barely affect their life satisfaction.” In his praise of the proposed tax hike, Krugman and others have alluded to the work of Peter Diamond and Emmanuel Saez, who have made the case for exceptionally high marginal tax rates in academic journals since the Great Recession.

The 73 percent optimal marginal tax rate suggested by Diamond and Saez is based on the assumption of a short-term elasticity of taxable income of 0.25, which means a reduction in declared taxable income of just 0.25 percent resulting from a 1 percent reduction in the “net of tax” rate (one minus the marginal tax rate). The tax rate is determined to be optimal when the level of tax maximizes revenues. The assumed elasticity of 0.25 is far too low and completely ignores the effects an increase in the top marginal income tax rate has on the choice of schooling, occupation, entrepreneurship, and business development of income earners. Given the range of estimates of taxable income elasticity, assuming a median value of the short-term elasticity of 0.25, as Diamond and Saez do, does not seem prudent when comparing to existing empirical studies. Other studies focusing on high-income individuals tend to find much higher estimates of the short-term elasticity, with a range of estimates averaging around 1 to 1.3.

Furthermore, the 73 percent optimal tax figure in the Diamond and Saez study fails to take state and local taxes into account. When these additional taxes are accounted for in a study by Jon Gruber and Emmanuel Saez, which also assumes a slightly more prudent elasticity, the optimal federal tax rate gets adjusted down to 48 percent—not too far above the present rate of 37 percent.

Other studies of more progressive tax systems have shown that progressivity discourages skill investment and thereby reduces pretax wages as well as labor supply. Even though wages do reflect skill investment choices, tax progressivity is not an effective way to compress inequality in pretax wages because greater relative scarcity of high-skill workers increases the difference between the wages of skilled and unskilled workers.

The Effect on Decisions in Education and Career Aspiration

Changes in marginal tax rates not only lead to changes in federal revenues, but also have a broader effect on human capital accumulation and labor market efficiency. Aparna Mathur, Sita Slavov, and Michael Strain have explained that in the long term, a more progressive tax system reduces the incentives to accumulate human capital:

"Earlier decisions such as education and career choices affect later earnings opportunities. It is conceivable that a more progressive tax system could reduce incentives to accumulate human capital in the first place. At 70% tax rates, a high school student may choose not to pursue his dream of becoming an engineer, or a small business owner may choose not to expand their business."

To put it another way, proposals to impose high marginal tax rates do not account for the incentive to enter high-earning occupations. Making career choices includes weighing the expected return of educational investments and the cost of forgoing alternative opportunities. A 70 percent marginal tax rate would, therefore, act as a disincentive for people to pursue high-earning occupations. Michael Strain notes that “a young person interested in health care might decide to become a nurse rather than a surgeon, because much of the income gained from being a surgeon will be taken by the government. . . . In this case, our hypothetical young person isn’t deciding to be a nurse because that’s her preference—instead, she’s making the decision because the top rate is so high.” There is no doubt that this sort of tax-driven inefficiency makes society worse off.

Higher Marginal Tax Rates Would Equate to Lower Wage Growth for Everyone

Another feature of high marginal tax rates that is often overlooked is the adverse effects on labor income and investment income that drive down labor productivity. Higher marginal tax rates on profits and investment would reduce savings, lower rates of investment, and thereby reduce the stock of productive capital, and therefore wage growth.

Capital investment and total factor productivity (TFP) growth are essential components of the productive potential of the economy. Robert Barro decomposed the growth rate of real GDP from the growth rates of capital, labor, and TFP and found that for the period 1947–1973, TFP growth accounted for 34 percent of GDP growth, while capital accounted for 43 percent of GDP growth.

As high marginal taxes drive down capital investment and labor productivity, workers’ wages ultimately grow at a slower rate, as the growth in GDP per hour worked will be lower. Some studies find that an increase in marginal tax rates reduces the growth rate of GDP in Organisation for Economic Co-operation and Development (OECD) countries—these results are consistent with endogenous growth theories. Other studies have observed the relationship between lower productivity and pay (compensation). Larry Summers and coauthors found substantial evidence of linkage between productivity and compensation: from 1973 to 2016, 1 percentage point higher productivity growth was associated with 0.7 to 1.0 percentage points higher median and average compensation growth.

An unintended consequence of a 70 percent marginal tax rate on high-income workers would be lower wage growth for lower-income American workers as a consequence of reduced investment in productive capital.

How Much Revenue Could this Tax Change Raise?

Estimates of the additional revenue that could be raised by the proposed tax changes vary significantly, from as much as $72 billion a year claimed by the Urban Institute, to a low of $5.1 billion by the Tax Foundation.

Brian Riedl, senior fellow at the Manhattan Institute, analyzed IRS data to calculate how much revenue a 70 percent marginal tax rate would raise. Riedl found that this tax change would raise about $50 billion annually, although this figure assumes that tax loopholes and overseas tax havens do not exist.

According to Reidl, “The top-earning 5 percent of families and pass-through businesses currently account for 30 percent of all income. . . . Furthermore, the top 5 percent already pays 42 percent of all federal taxes, including 61 percent of all federal income taxes, which leaves less room for additional taxes.” He explains further, “Popular liberal proposals to impose a 30 percent minimum tax on ‘millionaires’ and more aggressively tax banks, hedge-fund managers, and oil and gas companies would raise a combined 0.1 percent of GDP” ($21 billion) (emphasis in original). “Doubling corporate tax rates would add 0.7 percent of GDP,” which he calculates using the Congressional Budget Office (CBO) report Options for Reducing the Deficit: 2017 to 2026.

The Tax Foundation has also estimated revenues from the proposed changes in the top marginal tax rate: If an eighth tax bracket at 70 percent were applied to ordinary taxable income greater than $10 million, revenues would rise by about $291 billion over a decade. If the same tax bracket were applied to ordinary taxable income and investment income greater than $10 million, then revenues would rise by only $51.4 billion over a decade.

A 70 percent tax on all income would raise far less revenue than a tax on only ordinary income owing to the way taxpayers would react to a drastic increase in the marginal tax rate on capital gains. As capital gains taxes are paid only when capital gains are “realized” (assets are sold), taxpayers can choose when they want to pay the tax, so capital gains are very responsive to changes in the tax rate.

Recent tax changes in neighboring Canada may provide some evidence of how increases in marginal tax rates affect tax revenues. In 2015, Prime Minister Trudeau increased the top marginal tax rate from 29 percent to 33 percent, with the intention of raising $3 billion in additional revenue in the first year. However, in 2016 the tax increase failed to produce the extra revenue; instead high-income earners paid $4.6 billion less in federal taxes. In reaction to the tax changes, 30,340 fewer Canadians reported incomes in the upper tax bracket in 2016 compared to the year before. The top earners have greatest access to sophisticated tax advisers and are easily able to shift their assets to lower-tax jurisdictions. Studies tend to demonstrate that top earners are far more sensitive to changes in tax rates than lower-income taxpayers and will often react to tax increases by lowering their taxable income.

The Truth About Tax Rates in the 1950s

Proponents of high top statutory tax rates often argue that America used to have very high tax rates on the rich and that the economy during this time boomed.

Commentators who are supportive of the proposal hark back to the days when the top marginal tax rate in the United States was 91 percent. Bloomberg opinion columnist Noah Smith noted that the proposed plan “would thus be a return to the 20th century norm.” This argument ignores the four decades of globalization and international tax competition that have dramatically changed the economy since the 1970s. Top marginal income tax rates have declined significantly around the world, from an average rate of 68 percent in 1980 to 47 percent in 2017 among OECD countries.

Comparing tax rates across time is not an apples-to-apples comparison. Over the past 50 years, most countries have moved to lower tax rates and broader tax bases, following the conclusions of decades of economic analysis. While the headline rates are lower, more types of income are being taxed. In fact, effective tax rates on the wealthy were 36.4 percent in 2014, only slightly below where they were in the 1950s. As the Tax Foundation reports,

"The nonprofit Tax Foundation estimates that in the 1950s, for instance, when the top statutory rate was 92 percent, the top 1 percent of taxpayers wrote so much income off in the calculation that they paid an effective average tax rate to the federal government of about 17 percent."

Historical data from the Office of Management and Budget show that, overall, today’s 8.2 percent of GDP federal income tax revenues exceeds that of the 1950s (7.2 percent), 1960s (7.6 percent), and 1970s (7.9 percent). The 1950s were not a tax-the-rich utopia. In fact, by arguing to raise the marginal tax rates for those earning more than $10 million, proponents are essentially arguing for a cut in the effective rate of tax for the highest-income earners, as history shows that excessively high marginal rates lead to lower effective rates for those earners.

Funding Proposed Programs Would Require Broader-Base Tax Increases

Assessing the need for additional revenues should be done in the context of the additional spending proposals that policymakers who are supportive of such tax changes have advocated for in the coming decade. These proposals include a “Green New Deal” ($7–$10 trillion), single-payer healthcare ($32 trillion), student loan forgiveness ($1.4 trillion), a federal jobs guarantee ($6.8 trillion), infrastructure ($1 trillion), paid family leave ($270 billion) and Social Security expansion ($188 billion). Increasing federal revenue by 21 percent of GDP to fund these proposed program expansions—even after slashing defense—would require either a 55 percent payroll tax increase or 115 percent value-added tax, according to CBO data. In the context of $49–52 trillion in additional spending over the coming decade, raising $50–70 billion a year through a 70 percent marginal income tax doesn’t seem like any solution at all.

Today there is no country in the OECD with a top tax bracket of 70 percent. In fact, the top combined income and payroll tax rate in the United States currently exceeds that of Germany, the United Kingdom, and Norway, and is only 7 points behind that of France. Unlike the United States, Europe finances its generous welfare programs through very high value-added taxes that are paid by the entire population.

In European countries with significant revenue collections as a percentage of GDP, taxes are generally high for everyone. Denmark, Sweden, and Finland have broad-based value-added taxes at 24–25 percent and high payroll taxes. As the Tax Foundation’s Kyle Pomerleau notes, “Their income taxes have high marginal tax rates, but they apply to much more personal income than what Ocasio-Cortez is proposing. Denmark’s top marginal tax rate of 60 percent starts applying at 1.2 times the average income. If that were enacted here, all taxable income over $56,500 would face a 60 percent tax rate.”

It is time to acknowledge the facts, and immediately end any public flirtation with endless spending proposals and program expansions. Commentators need to move beyond these “soak the rich” talking points and instead lay out a comprehensive plan for controlling our spiraling deficit spending.