19 April 2013
Revisiting the High Tax Rates of the 1950s
In the heated political debate that Americans are having about federal spending and revenue, advocates of higher taxes often cite the 1950s as a Golden Age. Then, it is claimed, the wealthy paid higher federal taxes and the system was fairer. A closer look at the facts, however, does not support this assertion.
In the 1950s, very few people paid the very high income-tax rates aimed at the wealthiest.
Claims that wealthy people paid more taxes rest instead on the assumption that the rich, as stock owners, bore the entire burden of higher corporate taxes of that era. There are good reasons to doubt this assumption about corporate taxes.
Even if we leave these assumptions unchallenged, the economy of the 1950s was so different from our own that its tax structure cannot be reproduced today.
- The most plausible viable paths to higher taxes in today’s economy would render the tax system less fair, not more so.
* * *
In the current political debate over the United States government’s finances, an influential and oft-heard argument is that raising taxes—specifically current levels of income tax—would not inhibit economic growth. A key piece of evidence for this contention stems from an analysis of the tax regime of the 1950s, during which the U.S. saw robust growth, even as top marginal tax rates were high. Indeed, the question of the significance and relevance of the 1950s tax regime arises with some regularity in contemporary policy debate.
One cogent advocate of interpreting the 1950s as a high-tax regime is New York Times columnist and Princeton economist Paul Krugman, who summarized it this way:
[In] the 1950s incomes in the top bracket faced a marginal tax rate of 91, that’s right, 91 percent, while taxes on corporate profits were twice as large, relative to national income, as in recent years. The best estimates suggest that circa 1960 the top 0.01 percent of Americans paid an effective federal tax rate of more than 70 percent, twice what they pay today.
This paper examines this reading of the evidence from the 1950s, and argues against the interpretation favored by Krugman and those who agree with him.
The notion that the 1950s combined robust economic growth with a higher tax burden on the rich finds prominent support in the pioneering work of Thomas Piketty and Emmanuel Saez, economists at the University of California at Berkeley. Though often cited by advocates of higher taxes, their work, when examined in detail, does not clearly support the pro-tax position.
First, Piketty and Saez did not find that the era’s high tax burdens were due to high marginal tax rates. Very few people paid the high-end official rates. The figure below, from a key paper by Piketty and Saez, shows average tax rates for different income percentiles (with the top end of the income distribution spread out to focus on their specific role) divided into different types of taxes.
These charts show clearly that average income-tax rates remained fairly unchanged from the 1950s to 2004, even after the Bush tax cuts took effect. Tax rates of the era were designed to ensure that the code was “progressive”—a technical (not political) term, meaning simply that a citizen’s tax rate increases as his income goes up. On this point, they were very aggressive, with a top-end marginal tax rate of 91 percent until the Kennedy tax cuts of 1964. Yet very few people earned enough to meet the top rate threshold (the threshold for that rate for married couples was $400,000, which in today’s dollars would be more than $3 million). Even among those tax filers with the very highest incomes, capable of paying those rates, the actual average tax rates were in fact comparable with today’s. As Piketty and Saez note:
Interestingly, the larger progressivity in 1960 is not mainly due to the individual income tax. The average individual income tax rate in 1960 reached an average rate of 31 percent at the very top, only slightly above the 25 percent average rate at the very top in 2004. Within the 1960 version of the individual income tax, lower rates on realized capital gains, as well as deductions for interest payments and charitable contributions, reduced dramatically what otherwise looked like an extremely progressive tax schedule, with a top marginal tax rate on individual income of 91 percent. (emphasis added)
In other words, loopholes and deductions left the 1950s with an income-tax rate structure fairly similar to today’s. Why, then, did the researchers conclude that wealthy Americans paid more in federal tax in that era?
The answer, they write, is that corporate and estate taxes were much higher. The indirect effect of non-income taxes was principally responsible for high tax burdens on the rich:
The greater progressivity of federal taxes in 1960, in contrast to 2004, stems from the corporate income tax and the estate tax. The corporate tax collected about 6.5 percent of total personal income in 1960 and only around 2.5 percent of total income today. Because capital income is very concentrated, it generated a substantial burden on top income groups. The estate tax has also decreased from 0.8 percent of total personal income in 1960 to about 0.35 percent of total income today. As a result, the burden of the estate tax relative to income has declined very sharply since 1960 in the top income groups… (emphasis added)
The reduction in top marginal individual income tax rates has contributed only marginally to the decline of progressivity of the federal tax system, because with various deductions and exemptions, along with favored treatment for capital gains, the average tax rate paid by those with very high income levels has changed much less over time than the top marginal rates. (emphasis added)
It is because they claim that the burden of those non-income taxes fell heavily on the rich that Piketty and Saez conclude that average tax rates in the 1950s were far higher than today’s, especially for the very highest incomes.
Calculating the impact of estate taxes is relatively simple, since these can be easily reckoned for individuals. Estimating the incidence or burden of corporate taxes, however, is more difficult. In theory, shareholders lose income because the taxed company pays lower returns on capital. And workers lose out on wages and benefits the company cannot pay. In practice, however, it is a major challenge to model the precise mix of people who pay for a company’s tax burden.
Piketty and Saez solve this problem in two ways. First, they postulate that the entire burden of corporate tax rates fall on capital. Second, they also assume that the individual income against which these taxes should be counted is well approximated by realized capital gains.
Consider the prototypical American company of the 1950s, General Motors (subject of the famous 1953 remark by its chief executive, “What is good for the country is good for General Motors, and what’s good for General Motors is good for the country”). For GM, as for other major firms, taxable corporate earnings were substantial during the 1950s and 1960s (indeed, taxable corporate earnings averaged 9 percent of GDP in the 1960s). Therefore corporate taxes were a substantial stream of revenue for the federal government.
In accounting for the economic consequences of corporate taxation, Piketty and Saez assumed that the entire burden of these corporate taxes fell on stockholders in the form of lower returns. To calculate the era’s effective tax rates, they compute the total amount of corporate taxes and divide by all the income that shareholders made by selling shares—the realized capital gains in stock. Since stocks were predominantly held by wealthy individuals, Piketty and Saez estimate that the overall tax burden on the extremely rich was high. This is how they arrive at the 70 percent figure touted by Krugman and other advocates of higher taxes today.
It is important to note, then, that the notion of the 1950s and 1960s as a high-tax era does not come from direct examination of tax returns. It is, instead, an estimate created by a complex set of calculations that presumes the impact of high corporate taxes via stock ownership.
While the assumptions made by Piketty and Saez were certainly defensible, a more careful analysis of their work suggests four important qualifiers to pro-tax conventional wisdom about the 1950s.
1. In modeling how corporate tax rates affected individuals, Piketty and Saez were incorporating a seemingly obvious fact into their analysis: corporate taxes do not fall on bodiless entities, but are in fact paid by real individuals. This point is quite significant. It counters a common defense of high tax rates: the assertion that investors in capital pay unusually low tax rates today (lower even than those of their secretaries, as Warren Buffett has famously asserted).
That claim is based, of course, on the difference between federal tax rates on wages and the lower rate for capital gains. It does not take into account the effect on investors of corporate taxes. Yet that effect is essential to the Piketty and Saez analysis. It is the assumption that all the burden of a corporate income tax falls on the owners of capital (such as Warren Buffett) that supports their claim about 1950s tax rates.
To cite this claim in defense of high taxes, then, is to be inconsistent. It is to say that the 1950s were an era of harmless high taxation (by counting corporate taxes that were in effect paid by wealthy shareholders) but that today’s is an era of low taxation (by ignoring the effect of those same corporate taxes and counting only capital-gains rates).
2. It is not at all certain that Piketty and Saez were correct in assuming that the entire burden of corporate tax rates falls on investors.
In particular, as we’ve mentioned, the burden of corporate taxes may be felt by another important constituency—employees—in the form of lower wages. In fact, most economists agree that labor bears at least some of the burden of higher corporate tax rates. To put it simply, what is captured by the federal government in taxes becomes unavailable for transfer to workers. This may have been particularly important in the 1950s, when improvements in wages and benefits were tied to high union activity.
Furthermore, looking only at realized capital gains—actual income from stock sales—may miscount the full extent of corporate income. Higher stock prices in general certainly represent an increase in overall wealth, but not all of that gain is captured by stock trades. Consider a wealthy individual for whom the benefit of higher earnings was to boost the value of his portfolio—but who never sold stock. In the Piketty and Saez framework, this person is assumed to bear the burden of the corporate tax rate. He has, however, realized no capital-gains income and thus paid zero in taxes.
There is no perfect technique for measuring where the burden of corporate taxes fell in the 1950s, so assumptions and simplifications must be a part of any model. This should not preclude an analysis of the weaknesses of the models on offer. In the case of the analysis by Piketty and Saez, there is ample room to question both their assumption that the burden of corporate tax rates falls on capital alone, and their assumption that realized capital-gains taxes can fully represent the effect of taxes on capital. One potential pitfall is clear: these assumptions, when combined, might well have spurred Piketty and Saez to estimate that the rich in the 1950s were paying more in taxes than they really were.
3. Even if we were to set aside questions about the validity of the assumptions made by Piketty and Saez, there remains a third problem with any attempt to make their work into an argument for higher taxes today. Their results are largely an artifact of the nature of growth in the1950s.
The collapse of the global economy after World War II and the nature of postwar industrial capitalism, created a period of high corporate earnings in the United States. American firms did not vie then, as they do now, with competitors on every inhabited continent. Both law and convention supported large, monolithic corporations in an environment in which disruption was rare. Capital was relatively immobile, and corporate profits were high—boosting redistribution in the forms of union activity (resulting in higher wages and benefits for workers) and government taxation.
Half a century later, the nature of global capitalism has drastically changed. Though the U.S. still has a high statutory corporate tax rate by developed-country standards, corporate tax revenue today is far lower as a percentage of GDP. Greater competition within industries, the spread of corporate tax loopholes, and the global spread of business and capital mean that domestic capital and corporate earnings are no longer a “captive” source of revenue that can be easily taxed away. Additionally, the holders of capital have diversified. They now include pension funds and ordinary investors. Therefore capital taxes no longer fall so sharply on the very top end of incomes.
In other words, the trouble for high-tax advocates is not just that Americans in the 1950s might not have been as heavily taxed as commonly supposed. Even if they were, there is no returning to the economy of that era.
How, then, could supposedly better higher tax rates be levied today? One avenue might be higher income taxes. But history shows that different marginal income tax rates produce a remarkably stable average income tax paid (as people respond to rate changes by moving to protect their assets as best they can). It is not at all obvious, then, that an income-focused strategy would be successful.
4. Even accepting the Piketty and Saez analysis entirely will yield the conclusion that the very richest Americans in the 1950s lived with higher rates of taxation. This is far from proof that, overall, the tax system was more progressive in that era.
One important measure of a tax code’s progressivity is its effect on the share of income of the bottom 90 percent of the population before and after taxes. Piketty and Saez estimate that in 1970 the bottom 90 percent of Americans took home 67.6 percent of pre-tax income, but had 70.5 percent of post-tax income. In other words, the net impact of taxation on the distribution of income was to raise the income share of anyone not in the richest 10 percent of the population, by 4.3 percent.
In 2004, however, the bottom 90 percent saw their income share rise 6.6 percent after the impact of taxation. By this metric, the supposedly progressive French tax code this same year had much less impact. There, the bottom 90 percent only saw their income rise by 1.8 percent. There are, of course, many available ways to judge a tax code’s progressivity. But this metric is compelling in its focus on the direct impact of taxation on earnings. Applying it does not support the argument that the current tax code is exceptional in U.S. history or in comparison to other nations. Indeed, by this measure, today’s U.S. tax code remains progressive.
Though advocates of higher tax rates often cite European nations as models of a fairer approach, the fact is that many European tax systems are not particularly progressive. Instead, in their need for high revenues to finance the typical European welfare state, many governments have turned to broad-based taxes that reduce progressivity. The value-added tax, for example, is common in Europe. It taxes all forms of consumption, which means it hits low-earners (who tend to consume a higher share of income) harder than the wealthy (for whom consumption is a smaller proportion of income).
These four lines of reasoning all lead to the same end point: It is potentially misleading to imagine that U.S. taxes in the 1950s can serve as a model for a better approach in 2013. Income tax rates actually paid in the U.S. have remained stable for decades. Corporate taxes may have played a role in pushing up the total tax burden for the rich during the 1950s, but this is not as clear-cut as is claimed. And even if high corporate tax rates did lead to high tax burdens on the rich in the past, it is unlikely that we can replicate that experience today. Meanwhile, the European example does not teach us that higher tax rates can be levied with little effect while enhancing progressivity. Rather, it teaches that the most reliable way to raise taxes in our time would be a broad-based approach that would be the opposite of progressive. The solution to the United States’ revenue debate cannot be found in past.
Arpit Gupta is a doctoral candidate in finance and economics at Columbia University.
1. See David Brooks: “High taxes will produce long-term changes in social norms, behavior and growth. Edward Prescott, a winner of the Nobel Memorial Prize in economics, found that, in the 1950s when their taxes were low, Europeans worked more hours per capita than Americans. Then their taxes went up, reducing the incentives to work and increasing the incentives to relax. Over the next decades, Europe saw a nearly 30 percent decline in work hours.” See also, Timothy Noah: “[The] lesson of the 1950s is that you can eliminate tax loopholes and raise rates well above their level today, and still end up with a healthier economy than the one we’ve got today.”
2. Paul Krugman, “The Twinkie Manifesto,” The New York Times, November 19, 2012, A19, accessed on March 21, 2013, http://www.nytimes.com/2012/11/19/opinion/krugman-the-twinkie-manifesto.html?_r=0.
3. Thomas Piketty and Emmanuel Saez, “How Progressive Is the U.S. Federal Tax System? A Historical and International Perspective,” Journal of Economic Perspectives (21:1), Winter 2007, 3-24, accessed March 21, 2013, http://elsa.berkeley.edu/~saez/piketty-saezJEP07taxprog.pdf.