The New York Times

April 4, 2002

When it comes to income inequality, more than just market forces are at work

Alan B. Krueger

NEARLY 50 years ago, Simon Kuznets, the father of national-income accounting and a Nobel laureate in economic science, predicted that income inequality would follow an inverted-U pattern over time -- ''widening in the early phases of economic growth when the transition from the preindustrial to the industrial society was most rapid; becoming stabilized for a while; and then narrowing in the later phases.''

New research that updates and extends this classic work, however, turns the Kuznets curve on its head.

Income disparities seem to follow no automatic pattern. Instead, for long stretches the degree of inequality appears to result from norms and social policy, especially taxes, at least as much as from economic forces of supply and demand. In the United States, the Kuznets curve -- the trend in inequality over time -- is better described by a U shape than an inverted U. In Britain and France, the curve is more like an L.

As Gary Fields of Cornell observed: ''The Kuznets curve is neither a law nor even a central tendency. The pattern is that there is no pattern.''

The latest blow to the Kuznets curve was struck by Thomas Piketty of École des Hautes Etudes en Sciences Sociales, in Paris, and Emmanuel Saez of Harvard. In a study, ''Income Inequality in the United States, 1913-1998'' (available at www.nber.org), they provide estimates of the share of total income going to top earners. As did Kuznets, they calculated the share of total personal income earned by top income groups from tax records. Because only a fraction of citizens were required to pay taxes early in the century, they compute total income from other sources.

They found that the share of income going to the top 10 percent of families grew from 40 percent at the start of the 1920's, to 45 percent, and remained near that level throughout the Depression. The top 10 percent's share fell sharply at the beginning of America's involvement in World War II, from 44 percent in 1940, to 32 percent in 1944.

The share of income going to the top 1 percent also fell a third from 1940 to 1944, to 11 percent from 16 percent.

The sudden drop in inequality during World War II, though inconsistent with Kuznets's gradual process, is easy to explain: wage controls during the war compressed pay. What is puzzling is that the more equal distribution that emerged under the controls persisted for 30 years before inequality began to rise again in the late 1970's.

If wage controls artificially compressed market forces, one would have expected inequality to grow sharply when the controls were lifted. It did not. Indeed, the share of income going to the top 1 percent and top 0.1 percent drifted downward for 20 years after the war, before stabilizing for a decade or so and then surging in the 1980's and 1990's.

Why? Professors Piketty and Saez suggest that the new wage patterns established during the war were difficult to reverse, even with competitive forces pushing in the opposite direction.

They also discovered something quite unexpected about the composition of income received by the very rich. Before the 1940's, the wealthiest Americans earned the bulk of their income from returns on capital; now their primary source is wages and salaries. In 1916, for example, the top 0.01 percent of ''tax units'' earned 70 percent of their income from capital, 24 percent from business ventures and only 6 percent from wages. By 1998, wages and salaries accounted for 45 percent of the very top group's income; business activities, 33 percent; and return on capital, 22 percent.

Because capital's share of national income has been stable, this shift must involve a more equal ownership of capital. Professors Piketty and Saez argue that the shift came about because of progressive taxation on capital income. In 1931, the top marginal rate was 20 percent. The Roosevelt administration increased the top rate during the Depression, eventually to 91 percent in 1944, so that the wealthy would shoulder more of the financial burden of the war. Top rates remained high until the 1980's. Professors Piketty and Saez stress that capital taxation has a cumulative effect on top incomes because it reduces the net return on today's wealth, which generates tomorrow's wealth.

So today the rich are not so different from the rest of us after all: they work for a living, too.

But they earn a lot more money. In 2001, the average chief executive of an industrial company with approximately $500 million in sales was paid $1.9 million, according to the Towers Perrin Worldwide Total Remuneration Report. The average N.B.A. player will earn more than $4 million this season, and the average Major League Baseball player more than $2.1 million.

After adjusting for inflation, salaries of chief executives grew about 6 percent a year in the 1980's and 90's, while those of basketball and baseball players grew more than 10 percent a year. Wage growth has been so strong at the high end that the top 1 percent of taxpayers have taken home 94 percent of the growth in total income since 1973.

Salaries at the top of the range seem to be particularly affected by prevailing norms and policy decisions. The pay of American chief executives is three times that of their counterparts in similar-size companies in Britain and four times those in France and Germany, according to Towers Perrin. The average chief executive is paid 24 times what the average production worker gets in the United States, but in Germany, only 8 times as much. It is hard to argue that their work or economic circumstances are so different to justify such large pay differences.

Further, Kevin Hallock of the University of Illinois has found that chief executives are paid more in the United States if they sit on one another's boards, suggesting that buddies matter.

The pay of professional athletes has grown rapidly in part because public subsidies for stadiums increase team revenues, and barriers to entry prevent competition.

So the capital-rich rentier class has been replaced by handsomely paid executives, athletes and entrepreneurs.

But Professor Saez predicts, ''If the estate tax goes away and top income rates are cut, a new generation of rentiers should reappear down the road.''