the inequality myth

The title of Brad Schiller's most recent essay in the WSJ...

Class warfare is once again a campaign theme. The Democratic candidates are railing against the "tax cuts for the rich," lamenting the stagnation of middle-class incomes, and decrying the deepening woes of the poor....

The annual release of census data on household incomes provides the foundation for the "two Americas" thesis. The latest figures tracked changes in incomes all the way back to 1967. Two observations grabbed the headlines. First, the data indicate that the top-earning 20% of households get half of all the income generated in the country, while the lowest-earning 20% of households get a meager 3.4%. That disparity has widened over time: In 1970, their respective shares were 43.3% and 4.1%. These income-share numbers buttress the popular notion that the "rich are getting richer while the poor are getting poorer."

The second observation in the Census reports relates to the well-being of the middle class. The median household income in 2006 was $48,201, just a trifle ahead of its 1998 level ($48,034). That seems to confirm middle-class stagnation.

While there is some substance to these fears of widening inequality and middle-class stagnation, the situation is not nearly as clear-cut. Demographic changes in the size and composition of U.S. households have distorted the statistics in important ways.

First, we can easily dismiss the notion that the poor are getting poorer. All the Census Bureau tells us is that the share of the pie consumed by the poor has been shrinking (to 3.4% in 2006 from 4.1% in 1970). But the "pie" has grown enormously. This year's real GDP of $14 trillion is three times that of 1970. So the absolute size of the slice received by the bottom 20% has increased to $476 billion from $181 billion. Allowing for population growth shows that the average income of people at the bottom of the income distribution has risen 36%.

They're not rich, but they're certainly not poorer. In reality, economic growth has raised incomes across the board.

The Census data originate from an annual survey of households. The data don't track individual households from year to year, but instead just take a snapshot of the households in existence in March of each year. From these annual snapshots, we try to infer what's happening to the typical household over time.

The "typical" household, however, keeps changing. Since 1970 there has been a dramatic rise in divorced, never-married and single-person households. Back in 1970, the married Ozzie and Harriet family was the norm: 71% of all U.S. households were two-parent families. Now the ratio is only 51%. In the process of this social revolution, the average household size has shrunk to 2.57 persons from 3.14 -- a drop of 18%. The meaning? Even a "stagnant" average household income implies a higher standard of living for the average household member.

Last year, the Census Bureau published a new set of income statistics that adjusted for changing household size and composition. In a single year (2006), this "equivalence-adjusted" computation increased the income share of the poor by 8% and reduced the standard measure of inequality (Gini coefficient) by 4%. Such "equivalency" adjustments would mute unadjusted inequality trends even more.

A closer look at household trends reveals that the percentage of one-person households has jumped to 27% from 17%. That's right: More than one out of four U.S. households now has only one occupant. Who are these people? Overwhelmingly, they are Generation Xers whose good jobs and high pay have permitted them to move out of their parental homes and establish their own residences. The rest are largely seniors who have enough savings and income to escape from their grandchildren and enjoy the serenity of an independent household. Both transitions are evidence of rising affluence, not increasing hardship. Yet this splintering of the extended family exerts strong downward statistical pressure on the average income of U.S. households....

The supposed decline of the poor and middle class is exaggerated even more by the dynamics of population growth. When people look at the "poor" in any two years, they think they're looking at the same people. That's rarely true, especially over longer periods of time.

Since 1998, the U.S. population has increased by over 20 million. Nearly half of that growth has come from immigration, legal and illegal. Overwhelmingly, these immigrants enter at the lowest rungs on the income ladder. Statistically, this immigrant surge not only reduces the income of the "average" household, but also changes the occupants of the lowest income classes.

To understand what's happening here, envision a line of people queued up for March Madness tickets. Individuals move up the line as tickets are purchased. But new people keep coming. So the line never gets shorter, even though individuals are advancing.

Something similar happens with the distribution of income. People keep entering the distribution line from the bottom. Even though individuals are moving up the line, the middle of the line never seems to move. Hence, an unchanged -- or even receding -- median marker could co-exist with individual advancement. The people who were at the middle marker before have moved up the distribution line. This is the kind of income mobility that has long characterized U.S. income dynamics....