TCS Daily

The New Political Economy?

By Arnold Kling - October 29, 2002 12:00 AM

"It's all too easy to see how we may become a country in which the big rewards are reserved for people with the right connections; in which ordinary people see little hope of advancement; in which political involvement seems pointless, because in the end the interests of the elite always get served."
-- Paul Krugman
Paul Krugman has introduced a novel theory of the political economy of income distribution, in which what matters is the ratio of the wealth at the top to the wealth in the middle. Buried under a mountain of melodramatic rhetoric, Krugman's argument boils down to two points:
  1. The ratio of the income of the highest earners to the median income is higher than it was thirty years ago; and

  2. This is a threat to society.
Let us begin with the simple facts. By extracting just a few figures from the bottom table of this Census Bureau data, we can tell the story. The data are for real incomes (that is, adjusted for inflation) for different segments of the income distribution. The ratio is the ratio of the income of the top 5 percent to the income of the middle 20 percent.

Year Income of middle 20 percent Income of top 5 percent Ratio
1970 $38,674 $136,602 3.5
2001 $51,538 $280,312 5.4

The facts support Krugman's first point, which is that the ratio of income at the top to income in the middle has increased dramatically. However, the facts also show a cumulative increase in real income in the middle of 33 percent.

I suspect that this increase in income for the middle quintile helps explain why the threat to society is not obvious to many people. In 1970, the income for the fourth quintile (from the 60th to the 80th percentile) was $52,168, which is roughly (after adjusting for inflation) where the middle quintile is today. In other words, if in 1970 you had defined affluence as an income over $50,000, then 40 percent of people were affluent then, while 60 percent are affluent today.

The Tax Burden

Krugman is concerned that the concentration of economic power will lead to a concentration of political power. That is, unless the government transfers more income from the rich to the middle class, the rich will accumulate political power, which will make it difficult to enact programs to transfer income from the rich to the middle class.

Once again, the facts are available. The table below compares the shares of adjusted gross income (taxable income) and income taxes paid for the top 5 percent of income-earners and the bottom 50 percent of income-earners.

Year Share of Taxable Income Share of Taxes Paid
Top 5 percent Bottom 50 percent Top 5 percent Bottom 50 percent
1980 21.0 17.7 37.1 7.0
1999 34.0 13.2 55.5 4.0

These shares of taxable income are consistent with the Census data, in that they show a trend toward greater concentration of income at the top. However, the shares of taxes paid also show a concentration at the top. It is difficult to argue from these data that the rich are not paying their fair share of income taxes. If anything, the trend indicates that the relative tax burden on the rich has been increasing.

On the other hand, the income tax is not the only tax that affects the distribution of after-tax income. The most important tax reinforcing the inequality of income is the payroll tax, used to fund Social Security and Medicare. We could switch to using the income tax to fund these programs. Doing so would successfully convert these programs into income transfers from the rich to the middle class. However, my impression is that older people would prefer to keep Social Security as it is, and younger people are more interested in using their own savings to take care of themselves when they retire.

Fluid or Static?
"Annual snapshots of the income distribution might deserve attention if we lived in a caste society, with rigid class lines determining who gets what share..." -- W. Michael Cox and Richard Alm, Myths of Rich and Poor, p. 72
The specter that Krugman raises is one of a static distribution of well-being, in which economic status is inherited and no one has hopes of rising. However, he does not cite any data indicating that the income distribution is becoming less fluid. He simply asserts, based on anecdotal evidence, that the 1930's and 1940's were a time of fewer class barriers.

I can think of a few counter-anecdotes. In the 1930's and 1940's, Barry Bonds could not have gotten rich playing baseball. In the 1930's and 1940's, Paul Krugman probably would not have obtained an Ivy League education, because admissions were based on social and religious criteria rather than on merit.

Until after World War II, many important businesses were family dynasties. Say what you will about today's corporate executives, at least they did not inherit their positions from their fathers.

Anecdotes aside, what is the picture of economic mobility in the United States? Cox and Alm cite the University of Michigan's Panel Study of Income Dynamics, or PSID. They compare the income quintile of people in 1975 with their income quintile in 1991. Here is a subset of the data:

Where they Began in 1975 Percent who made it to the top 40 percent in 1991
Lowest Fifth 59%
Second Fifth 52%
Middle Fifth 49%
Fourth Fifth 70%
Highest Fifth 86%

True, those who started in the higher income classes had a higher chance of ending up in the higher income classes. But even the lower income classes had about a 50 percent chance of ending up in the higher income classes. What is this - Lake Wobegon, where everyone is above average?

Obviously, the lower classes in 1991 were populated in large part by people who were not in the PSID in 1975. Presumably, these would be new immigrants as well as young people just starting their careers.

In fact, one could argue that these data simply show the importance of the "life-cycle" effect on income - as people get older, their income rises. This allows them to break into the top 40 percent and the top 20 percent.

It could be that there is less mobility at the very top of the earning distribution. Perhaps the top 1 percent is much more stable. I have not seen any data on that, and to my knowledge neither has Krugman. Cox and Alm did not look at the top 5 percent or the top 1 percent--their focus was more on the other end of the income scale.

Eliminate Wealth to Eliminate Poverty?

In fact, most economists who examine the income distribution do so because they worry about how to eliminate poverty, not how to eliminate wealth. In making the argument that disparities between the highest earners and median matter in a society with a prosperous middle class, Krugman is breaking new ground.

It is not clear how programs to lower the ratio of high income to median income would reduce poverty, unless one can show a "trickle-down" effect. In fact, it could even have perverse results. Some wealthy individuals are turning to venture philanthropy and others are providing assistance to the poor overseas. Krugman would instead redistribute income toward the American middle class.

The idea of focusing on the disparities between high-income and middle-income earners could be a new breakthrough in political economy. Now that Krugman has presented it in the New York Times, it will be interesting to see it discussed by the economics profession at large.

TCS Daily Archives