On August 26, 2008, the Census Bureau released the report "Income, Poverty and Health Insurance Coverage in the United States: 2007," which announced that last year income inequality had dropped, median household income had increased, and the poverty rate had increased by a statistically insignificant amount. Real median income increased to above $50,000 for the first time since the 2001 recession. While this report contains good news about the economy, the news would be even better if the Census Bureau included in-kind benefits as a way to reveal the true state of poverty.
The Census Bureau found that in 2007:
- Income inequality fell to its lowest level since 2002;
- Real Median Income increased for the third year in a row, and only real median income in 1999 and 2000 exceeded the 2007 level; and
- The poverty rate was up slightly from 12.3 percent to 12.5 percent, but this increase was statistically insignificant.
The Census Bureau examines income inequality in an attempt to define how income is distributed throughout a population. Census uses two different measures: the Gini coefficient, which is a single measure that defines the entire distribution of income, and the share of overall income going to each quintile.
Both measures declined in 2007, showing a decrease in inequality. The top quintile's share of income fell 0.8 percent, to its smallest share of income since 2002. The third quintile increased its share of income by 0.3 percent, and the fourth quintile increased its share by 0.4 percent. The Gini coefficient declined from 0.47 to 0.463, a decrease of 1.5 percent.
For the first time, the Census Bureau attempted to account for differences in family size. Households in higher quintiles usually have more people than households in the bottom quintiles. An equivalence scale attempts to balance the fact that households with more people need more resources than households with fewer people and that children consume less than adults. Income inequality shrinks further with the use of this equivalence scale, with the share of income to the top quintile falling from 49.7 to 48.5 percent. Again, the lower quintiles increase their share of national income when the differences in household sizes smoothed.
Unfortunately, this report from the Census Bureau does not include anti-poverty programs such as the Earned Income Tax Credit (EITC), which would increase income to households in the bottom quintiles, or the tax code, which attempts to reduce income inequality with a progressive tax rate structure. A previous report issued in the spring indicates that income inequality would fall even further if the tax codes' progressivity were taken into account. For example, the Gini coefficient falls 10 percent by including the effects of income taxes and the EITC.
The strong income gains of the middle class and a fall in income in the top quintile is why income inequality declined. The ratio of income at the 95th percentile to the 50th percentile fell from 3.61 to 3.52. This is because the mean income of the third quintile increased from $49,591 to $49,968 and the income of the top quintile fell from $172,941 to $167,971. This is a case of the middle class getting richer while the top quintile got poorer. The middle class also pulled away from the bottom with the 20/50 ratio falling from 0.42 to 0.40.
The Census Bureau roughly defines poverty as having insufficient income to meet all basic needs. The size of a family and the age of family members is a key determinant on the income needed to determine whether a family is in poverty. In 2007, the poverty rate increased from 12.3 percent to 12.5 percent, a statistically insignificant change. The poverty rate did increase for children, rising from 17.4 percent to 18.0 percent. The poverty rate for married couples remains significantly below the national poverty rate at 4.9 percent, unchanged from 2006.
The poverty rate ticked upwards slightly because more people did not work at all in 2007 as compared to 2006, even as the unemployment rate remained at historically low levels in 2007. People who worked at least part time did not experience any increase in poverty over the last year, illustrating the importance of work. Workers had a poverty rate of only 5.7 percent, compared to non-workers at 21.5 percent.
Single motherhood continues to be a large factor in why so many children are in poverty. 43 percent of children in single-mother families were in poverty, compared to only 8.5 percent of children in married families.
This Census Bureau report, like that of income inequality, is incomplete since in-kind benefits such as food stamps, housing assistance, Medicaid, and the EITC are ignored. The EITC is the largest cash anti-poverty program, but according to this Census report, it does not reduce poverty at all. If means-tested transfer payments and these types of benefits are included in income and work expenses are subtracted, 3 million people would no longer be falsely counted as poor and the poverty rate would fall to under 12 percent. The Census Bureau will release numbers showing this poverty rate counting other types of benefits and expenses much later, but it should be released now to accurately show real poverty.
Not the Whole Story
The Census Bureau report on poverty and inequality continued to buttress the facts that work and married families are the best solution to poverty. The poverty rate increased because more Americans did not work in 2007 as compared to 2006. Income inequality fell due to the income gains of the middle class and a loss of income to the top quintile. The Census Bureau did an excellent job attempting to account for disparity in household size in the different quintiles. Income inequality is exacerbated by the fact that more married couples with two earners are in the top quintiles, compared to single earners and single-parent households in the bottom quintiles.
The Census Bureau should release the tables showing the effects of taxes and transfer payments on poverty and inequality with this report instead of waiting another year. This report paints an incomplete picture of poverty in this country by excluding specific anti-poverty programs such as the EITC.
Rea S. Hederman, Jr., is Assistant Director of and a Senior Policy Analyst in the Center for Data Analysis at The Heritage Foundation.
 A Gini coefficient of 0 indicates a perfect distribution of income while a coefficient of 1 indicates that one person or household has all the income.
 Robert Rector and Rea Hederman, Jr., "Two Americas: One Rich, One Poor? Understanding Income Inequality in the United States," Heritage Foundation Backgrounder No. 1791, at http://www.heritage.org/Research/Taxes/bg1791.cfm.