Want Less Inequality? Tax It

Courtesy of the Ramsey and Muspratt Collection

Arthur Cecil Pigou

For the last two decades of his life, Arthur Cecil Pigou didn’t get much respect. Once considered Britain’s leading economist, he had come under caustic attack from a colleague at Cambridge—his friend and famous protégé John Maynard Keynes—for insisting that the Great Depression would correct itself without strong government intervention. By the mid-1940s, before Keynes died, Pigou had capitulated. But the dispute between the two, and Pigou’s eventual acknowledgment that Keynes might have a point, seemed to consign his own work to the dustbin. His student Harry Johnson remembers him as “a tall, straight figure, eccentrically garbed, glimpsed occasionally walking about the countryside.” In the hall where Pigou lectured on economic principles, one unknown undergraduate had carved into a desk “Pigou mumbles.” 

After the 2008 crash, it seemed Pigou might make a comeback. Nobel laureate economist Amartya Sen, writing in The New York Review of Books, declared that the “present economic crises … demand a new understanding of older ideas, such as those of [Adam] Smith and, nearer our time, of Pigou, many of which have been sadly neglected.” The New Yorker’s John Cassidy, writing in The Wall Street Journal, nominated Pigou as the deceased economist whose work “best explains financial crises, global warming, and other pressing issues of today.” N. Gregory Mankiw, the Harvard economist who chaired George W. Bush’s Council of Economic Advisers (and has recently advised Mitt Romney), launched a Pigou Club in favor of a higher gasoline tax to combat air pollution, global warming, and road congestion. 

Yet the comeback hasn’t quite materialized, and Pigou is in danger of returning to obscurity. That would be a major loss. Pigou, a key bridge figure in the history of his field, was one of the earliest classical economists to notice that markets do not always produce the best possible social outcomes. The pollution generated by a factory imposes costs on those who live downstream or in the path of its airborne emissions. The risks assumed by banks leading up to the recent financial crisis imposed costs on just about everybody. Market transactions often generate what economists call “externalities”—side effects, sometimes positive but often negative, that affect people who do not participate in the transaction.

Pigou, having recognized the problem, was the first to propose a solution. Society should tax the negative externalities and subsidize the positive ones. This simple notion—if you want less of something, tax it—is why his ideas periodically bubble up in the service of combating a recognizable cost to society, like pollution. We think that his approach offers an answer to another great problem of our time: inequality.


Does the extreme degree of inequality in America today really create, as Pigou would put it, negative externalities? Does the fact that hedge-fund manager Mr. Jones rakes in 100 or 1,000 times what office manager Mrs. Smith earns impose costs on everybody else? Plenty of Americans think not. Defenders of our skewed income distribution point out that a free-enterprise system requires some inequality. Unequal rewards give people an incentive to work hard and acquire new skills. They encourage inventors to invent, entrepreneurs to start companies, investors to take risks. It’s fine in this view that some people get astronomically rich. As Mitt Romney likes to say, “I’m not going to apologize for being successful.” 

On the other side, many of us have a gut feeling that inequality has gone too far. Our times are reminiscent of the Gilded Age’s worst excesses. Hence the popularity of the Occupy Wall Street movement’s slogan, “We are the 99 percent.”

But the conventional strategy for fighting inequality—far higher taxes on the rich—usually rests on a foundation of fairness, and the question of what’s fair and what’s unfair turns out to cut different ways, depending on your point of view. You may find it unfair that the very rich take in so much more than others. But somebody else might wonder why the rich should be taxed so heavily. Don’t they already pay disproportionately more than everyone else? These arguments quickly hit a dead end. That may be why befuddled Democrats and Republicans in Congress wrangle fruitlessly over top tax rates—39 percent, 35 percent, or even lower—that have already demonstrated negligible effects in reducing inequality. No one even discusses top rates that might make a difference. (We’re referring, of course, to marginal rates, rates that apply only to income over a certain threshold, like $250,000 or $1 million.) 

Pigou may be able to help us cut through the confusion. Viewed through a Pigovian lens, today’s highly skewed income distribution imposes three sorts of costs on the rest of us.

As the economy grows, the rich get nearly all the gains. The United States enjoyed an economic Golden Age from the end of World War II to about 1973. During that time, despite the fact that the top marginal tax rate never dropped below 70 percent (and stood at more than 90 percent until the early 1960s), growth in real gross domestic product (GDP) per capita averaged close to 2.5 percent a year. There is a widely held—and mythical—belief that the United States has been in the midst of a great stagnation ever since. The minuscule increase in the median household income, about 0.4 percent per year, supports this myth. But the economy has not stagnated. Output per capita has risen more than 90 percent since 1975, or an average of 2.1 percent per year. That’s less than the figure for the Golden Age but not much less. The difference is that since the 1970s, virtually all the gains have gone to those at the top. The average income of the top 1 percent has tripled, while the median household income has increased by less than 20 percent. 

The skewing of the income distribution, in short, imposes a cost on the vast majority of people. From 1975 to 2008, the median income of U.S. households rose by a total of $8,000 (in today’s dollars). Had the distribution remained the same as it was in the postwar years, the increase would have been $40,000. Instead of its current level, $50,000, the median income of U.S. households today would be $86,400. 

Many talented people focus narrowly on getting richer, to the exclusion of activities that might be more beneficial to society. The greater the number of very rich people, and the bigger their share of the pie, the more it seems that getting very rich is both feasible and desirable. For young people, going where the money is becomes more attractive than a career in science, education, or public service. As late as 1986, only 18 percent of Harvard graduates planned any sort of business career. In 2011, the figure was 41 percent, including 17 percent going into finance. 

The new incentives affect the behavior of anyone with a decent shot at getting rich. CEOs of large corporations expect astronomical compensation packages. Wall Street executives pursue riskier lines of business in search of higher profits than they could earn through traditional banking or brokerage. Doctors and lawyers have an incentive to train for lucrative subspecialties, leaving a shortage of general practitioners and public defenders. All these brain drains impose costs on everyone else. 

As a group, the rich can use money to pursue political power and influence well beyond their numbers. Despite fundamental principles such as one person, one vote, the wealthiest Americans have always wielded disproportionate political influence. Today they can make unlimited, anonymous donations to “super” political action committees. They can play a huge role in financing nearly every campaign for federal office. Senators and representatives, faced with the need to raise an average $9 million (Senate) or $1.5 million (House) for each race, spend up to four hours every day cultivating the well-to-do. Nor is the problem confined to elections. Generously financed lobbyists do their best to ensure that regulatory and tax policies at every level of government don’t interfere with the opportunity to make—and keep—a ton of money. “Another vicious circle has been set in play,” writes Joseph Stiglitz in his new book The Price of Inequality. “Political rules of the game have not only directly benefited those at the top, ensuring that they have a disproportionate voice, but have also created a political process that indirectly gives them more power.” 

Little wonder, then, that disaffected citizens on the right, on the left, and in the middle—whatever their views on inequality—believe that the very rich and organized business interests have hijacked America’s political process. The danger, of course, is that the trend is self-perpetuating. In their book Why Nations Fail, Daron Acemoglu and James A. Robinson show that sustained economic growth depends on open and inclusive political institutions. But when a moneyed elite so dominates the political order, it can rewrite the rules to forestall change and maintain a perch on top of the economic ladder. Acemoglu and Robinson quote Woodrow Wilson, who said during the last period when inequality threatened democracy: “If there are men in this country big enough to own the government of the United States, they are going to own it.” 


Assuming that a highly skewed income distribution results in negative political externalities, Pigou’s response would be quick. Slap a tax on vastly unequal incomes, and watch the externalities shrink.

A Pigovian tax on the rich might look on the surface like any other progressive income tax, but there are at least three philosophical differences. First, its purpose is to reduce inequality, not to make everyone pay a “fair share.” This goal cuts through the complaint that the rich are already paying far more than their share in taxes. 

Second, a Pigovian tax seeks to change incentives. That is, its goal is to alter the pretax distribution of income, not just the after-tax distribution—to discourage people from seeking to earn exponentially more than their fellow citizens. With higher tax rates, a job in finance might lose some of its appeal to talented young people. U.S. chief executives might content themselves with salaries averaging 20 times the typical worker’s earnings—the ratio in 1965—instead of 351 times the average, as was the case in 2007. Incidentally, past experience suggests that even with much higher tax rates, plenty of people would pursue careers in finance, corporate management, sports, entertainment, specialized medicine and law, and other well-paid fields. These occupations would still offer a high standard of living and above-average nonmonetary rewards. The most successful practitioners might earn $100 million or more in salary and capital gains and—depending on rates—pay taxes of $50 million or more. In our view, that would still leave them with a sufficiently comfortable lifestyle.

Finally, just as a Pigovian tax on pollution is not designed to eliminate all pollution—which would require an unacceptable ban on most industrial production—a Pigovian tax on inequality is not designed to equalize outcomes. It seeks only to reduce inequality from a level that threatens democracy to a level compatible with democracy, while still encouraging plenty of work and innovation.

So there would still be super-rich Americans, just not quite as many. Underlining this emphasis on reasonable gradations, the Pigovian tax would need a lot of brackets at the upper end. Perhaps the marginal rate should be 40 percent at $1 million (only 1 point higher than it was under President Bill Clinton), 50 percent at $5 million, 75 percent at $10 million, and 90 percent at $20 million. Such rates would affect only 140,000 U.S. households, or 0.1 percent of the total. Though the resulting taxes would raise additional revenue, we should be clear that they would not put a large dent in the deficit. No matter—fixing the deficit is not this tax’s goal. 

In fact, we Americans used our progressive income tax very much as a Pigovian tax in the middle of the 20th century, during that hazily remembered Golden Age. The top rate remained at least 90 percent until the Kennedy-era tax cuts lowered it, but it never fell below 70 percent until Reagan took office. These rates applied only to exceptionally high incomes. In the 1950s, the top threshold was around $3 million in today’s dollars, or about 130 times the average income of the bottom 90 percent. High marginal tax rates were undoubtedly responsible for keeping the income share of the top 1 percent in check. This did not prevent the United States from enjoying the strongest quarter-century of growth in its history. 

Is there reason to think that a Pigovian tax today would have similar effects? Would it reduce the pretax income share of those at the top? The evidence from a century of experience in the United States—not just the
Golden Age—is striking. Look at the chart above. As the top rate rises, the share of pretax income going to the top 1 percent declines. When the top rate falls, the pretax share of the top 1 percent rises. So the provisional answer is, yes.

True, the chart shows only a correlation, not causation. A skeptic might argue that high marginal tax rates push the rich not to earn less but rather to earn more of their money in ways not subject to federal tax. We can’t disprove that possibility. But we can examine whether higher tax rates raise or lower overall income-tax revenue. The evidence shows a strong positive correlation between the marginal tax rate for the top 1 percent and total tax revenue but not such a strong relationship for the top 0.01 percent (those who may find it easiest to earn in ways that avoid taxes). Economists who specialize in these matters, including Thomas Piketty, Emmanuel Saez, and Peter Diamond, typically argue that a marginal rate around 70 percent maximizes revenue. 

So far, so good. But how might high marginal tax rates affect overall economic growth? The conventional wisdom holds that high rates are a threat to growth. The conventional wisdom is wrong. A plot of growth rates against top tax rates shows little or no correlation between the two. To see how this can be so, it helps to understand that economic life is a mix of different elements. In every economy there are positive-sum games, in which market interactions benefit both parties, and zero-sum games, in which one party loses and the other wins. Because positive-sum phenomena—say, gains from trade—play an indispensable role in making market economies productive, economists usually focus attention on them. But every real economy includes many zero-sum activities that by definition do not contribute to higher growth.

As a thought experiment, imagine a quilt factory. Everybody is paid according to how much he or she produces—say, $5 per quilt. Produce one quilt an hour, and your hourly pay is $5; produce ten quilts, and it’s $50. Now, imagine that the government raises marginal tax rates. How would you respond? Maybe you’d decide to produce fewer quilts, since the government is going to take more of your money. Or maybe you’d decide to work harder and make more quilts to maintain your take-home pay. The economists who measure such things usually say that for most people, these two effects roughly cancel each other out. People end up producing about the same amount, regardless of whether marginal tax rates go up or down. 

But for the very rich, the response is different. As marginal tax rates rise, the highest-income people generally earn a bit less even before their income is taxed. Mind you, if the majority of the top 1 percent had been engaged in a positive-sum activity—like making quilts—giving them an incentive to work less in this way should lower the rate of GDP growth. But it doesn’t.

One logical explanation is that the rich are not those who can quilt 351 times faster than the average quilter. Instead, there may be among them company owners who convince the government to subsidize quilt production, or lobbyists who persuade Congress to scrap regulations requiring labels showing what’s in the quilts, allowing the companies to lower their costs by substituting poor-quality, flammable stuffing. 

These are zero-sum activities, a sort of economic version of airborne carbon emissions, and over time they can compromise the economy’s efficiency. The owners’ gains don’t come without somebody else’s loss. Zero-sum actions include increasing corporate earnings by holding down wages, taking advantage of financial customers by creating complex or risky products that customers don’t understand, and lobbying Congress for protection from competition. As these examples suggest, some individuals in the top income group pursue quite a number of activities that others might be glad to see them spending less time on.

If these zero-sum activities are significant, a reduction in work should result in lower income for the top 1 percent but not a decline in overall GDP. High marginal rates could actually create greater income growth for the rest of us—say, the bottom 90 percent—as zero-summers begin to spend less effort on increasing income in ways that impose a cost on someone else. 


A higher tax on gasoline, a Pigovian anti-pollution strategy championed by Greg Mankiw, hasn’t gotten far in Congress. But it has managed to draw support from well-known figures on both sides of the aisle: from Paul Krugman to Alan Greenspan, Al Gore to South Carolina Senator Lindsey Graham, centrist columnist Thomas Friedman to conservative commentator Charles Krauthammer. Even New York City Mayor Michael Bloomberg, and Alan Mulally, CEO of Ford Motor Company, are on board. 

We are under no illusions that Congress will consider Pigovian-style income-tax rates in the foreseeable future. One might as well wish for a constitutional amendment banning large campaign contributions. What we can hope is that voters and public leaders of various stripes will begin to see the worsening distribution of income in the United States, and in particular the amounts going to those at the top, as an externality with broader negative side effects akin to pollution and hence as something that should be heavily taxed. We aspire to the same incipient bipartisan accord—at least among those who do not have to run for re-election every two years—that greets the idea of higher gasoline taxes. 

Perhaps some at the top will sign on as well. As a group, the rich don’t get much applause these days, but they’re not so different from the rest of us. It’s just that the lack of any limit to outsize economic rewards turns out to have a measurable cost, which Americans who aren’t so wealthy keep getting asked to pay. With a shift in incentives, we could change this. A Pigovian tax might push talented people to use their skills in ways far more beneficial to the country than inventing the next credit-default swap. Even for the many successful citizens who remain in the same line of business, a Pigovian tax might start to do what has lately seemed impossible: give the best-paid Americans an interest in common again with the men and women with whom they work, and remind us that we’re all in this together. 

You may also like