RECENTLY, the U.S. Census Bureau released its annual report on income and poverty in the United States, including data on income distribution. As it does every year, the Census found a yawning income gap between the rich and the poor.
However, the report is a gross exaggeration; the income gap is a gully, not a canyon.
The Census Bureau measures income distribution by dividing society into fifths, or quintiles, and then calculating the share of total income received by each. In 1997, for example, the Census showed that the bottom quintile received 3.6 percent of income while the top quintile received 49.4 percent.
We analyzed these data and found a number of flaws that make the income gap seem bigger than it is. First, the Census quintiles contain unequal numbers of people. The lowest-income quintile is significantly under-populated, containing only 14.7 percent of the population, or 39.2 million people. The top quintile holds 24.3% of the population, or 64.2 million people.
This demographic imbalance -- not revealed in Census reports -- skews the apparent distribution of income. The quintiles are unequal because Census counts households, not population.
Low-income households are generally small, often single persons (retirees and young people just entering the work force). By contrast, high-income households are generally large with multiple wage earners (90 percent of the households in the top quintile are married-couple families, most with two or more wage earners). So even though the lowest quintile has the same number of households as the top quintile, it has fewer people. And it has little income in large part because it contains relatively fewer people to earn or receive income.
Second, the Census neglects to take into account differences in the amount of work performed between quintiles. Consider: The bottom quintile contains only 11.5 percent of all working-age (18 to 64) adults and only 5.6 percent of all hours of work performed in the economy.
By contrast, the top quintile contains 27.6 percent of all working-age adults and nearly one-third of all the hours of labor performed. There are nearly five hours of paid work in the top quintile for every one in the bottom.
In short, some of the "income gap" can be explained by the "labor gap." And third, the Census uses income figures that fail to count taxes and many types of cash and noncash income.
The Census should subtract taxes from the income of the top quintile and add welfare, food stamps, public housing and Medicaid to the income of the bottom.
That alone raises the bottom quintile's share of total income from 3.6 percent to 5.6 percent and reduces the top quintile's share from 49.4 percent to 45.3 percent.
When we analyzed Census data using a methodology that corrects for population as well, the income gap narrowed still further. The percentage of total income controlled by the bottom quintile more than doubled, from 3.6 percent to 9.4 percent.
These same adjustments reduce the income share of the top quintile from 49.4 percent to 39.7 percent. Looked at another way, official Census data show the top quintile of society earning $13.86 for every $1 earned by the bottom fifth.
Yet if income and taxes are counted more completely, and the quintiles are adjusted to contain equal numbers of people, that ratio drops to $4.23 to $1.
Actual differences in income in the United States are far smaller than Census figures suggest. Moreover, the income differences that do exist are the natural result of differences in behavior and ability between individuals.
People at high-income levels tend to be married, work more hours per year, have high levels of skill and productivity and provide higher levels of savings and investment to sustain the overall prosperity of the economy.
By contrast, individuals in the lowest income quintile tend to be unmarried, work less and have lower levels of skill and productivity. Of particular importance are the growing numbers of single-parent families in the bottom quintile, a direct result of the alarming fact that one-third of U.S. children are now born out of wedlock.
Those truly concerned with the well-being of lower-income Americans would do well to focus their efforts not on raising taxes on the most productive segments of society but on raising marriage and work rates among the least affluent.
Robert Rector is a senior research fellow and Rea Hederman is a policy analyst at the Heritage Foundation in Washington. This first appeared in the Los Angeles Times.
Pub Date: 10/12/99