While median family income in the United States fell for over a decade prior to the beginning of the recession that started in late 2007, per capita incomes continued to grow during that time period. From 2000 to the beginning of 2008, American per capita GDP, after adjustment for inflation, grew at a decent pace at about 1 ½ percent per year. Per capita real consumption of both durables and non-durables, an important measure of the real incomes of individuals, also grew at a good rate until the recession. These measures suggest sizable improvements in the welfare of the average person during most of this decade, yet the median household income cited by Posner suggests the opposite. How can these conflicting conclusions be reconciled?
The data I cite are arithmetical averages, while, as Posner indicates, his data on household incomes are medians, or midpoints, of the distribution of household incomes. Trends in arithmetical average incomes and median household incomes have differed in the US for a variety of reasons. One obvious factor is changes in the degree of inequality in the distribution of earnings. Earnings inequality of full time workers did rise rapidly in the United States during the 1980s, and more slowly but significantly during the 1990s. However, inequality among different education and skill groups increased much more modestly during the first seven years of this decade than in the prior two decades.
As Posner indicates, incomes at the very high end, especially in the financial sector, did grow rapidly during the years leading up to the recession. That contributed in a modest way to the difference between the trends of average and median incomes. Other factors that helped raise inequality in median household incomes during the past 30 years include the growth in the number of households headed by a single parent to about 9% of all households. Also important has been the decline in the labor force participation of males at the lower end of the skill distribution, especially of African-American males, and the growth in earnings from the underground economy, due to illegal immigration, taxes, and regulations. Earnings from the underground economy are not included in most statistics on earnings. These and other factors explain why the number of Americans with incomes below the official poverty line seemed to increased by 15% between 2000-2006, and why by the end of 2008 over 30 million workers appeared to earn less than $10 per hour.
The best longer-term solution to the inequality problem is to reduce the fraction of Americans who dropout of high school, a theme I have continued to emphasize in various postings on our blog. This drop out fraction has been stagnant for the past several decades at about 30% for males, and a somewhat lower but still high percent for females. This is almost surely the highest fraction of high school dropouts among rich countries, and is heavily concentrated among children from African-Americans and Hispanic families. In large cities, often less than half of all the children enrolled in public schools end up graduating.
In the first half of the 20th century, dropping out of high school reduced economic prospects, but not by so much since many good jobs were available to persons with limited education. Prospects for high school drop outs began to fade after World War II, and really fell during the past 30 years because of technological changes toward greater demand for skilled workers, shifts of the economy toward health, education, and other services that use more skilled workers, and globalization that reduced jobs in America for low skilled workers. High school drop outs now face a dismal future not only in the form of low earnings, but also poor health, less likelihood of both marrying and staying married, and worse outcomes on practically all other aspects of life in the modern world.
Probably the most fundamental cause of the continuing high drop out rates has been the large deterioration in family stability during the past 50 years. This has led, among other things, to the growth of low-income single parent families that contain about ¼ of all young children, and to the absence of the influence of fathers on the discipline and motivation of children in these households. Family stability is difficult to improve, particularly in the shorter run (although see my blog post of April 4).
However, schools can be improved relatively quickly by holding teachers and administrators up to higher standards. President Bush took important steps in this direction, and President Obama has been adding to and improving these changes by requiring even better school performance. I addition, studies have shown that charter schools, vouchers, and other ways to raise competition among schools that cater to children from poorer families contribute in important ways to better achievement scores of these children. That is why the caps placed by many states and localities on the number of charter schools-mainly under pressure from teachers unions- should be removed, as some cities are doing.
In trying to use public policies to reduce inequality, the goal should be to achieve this without reducing the efficiency of the economy. When efficiency declines, incomes of poor as well as better off households tend to fall. The great attraction of reducing the high school drop out rate, and other improvements in the school performance of children from lower income and lower educated families, is that they are likely to increase the economy's efficiency while at the same time raising incomes at the lower end of the earnings distribution.