It can be hard to separate fact from fiction in the economic statements being tossed around lately.
Which of the following are true?
Social Security
A) The average income of people receiving Social Security benefits is higher than the average income of those paying into the program.
B) Most people receiving Social Security have incomes well below the national average.
U.S. versus Europe
A) Per capita GDP in the United States, at least on a “purchasing-power-parity” basis, is higher than in northwestern European countries with the exception of oil-exporting Norway.
B) Most households in such European nations have higher incomes than do most households in our own country.
Actually, all of the statements are true. In both cases, any apparent contradiction is due to the difference between “mean” and “median.”
Nearly everyone learns this difference in junior high school, but many apparently forget it. A mean is what most people know as an average. Take the total income of everyone in a country and divide it by the total number of persons or households. The result is the mean per-capita or per-household income.
The median income is what is received at exactly the middle of the group, ranked from highest to lowest. Half of the people or households receive more than this amount and half receive less.
Take the total income of everyone getting Social Security and divide it by the number in that group. This is the average or mean income of Social Security beneficiaries. Then do the same for the people paying in. You will find the average income of beneficiaries is indeed higher than of those paying FICA taxes. But does this mean Social Security redistributes money from poor to rich, or that it makes income disparities greater rather than smaller?
The answer is no, probably not. The mean income of recipients is skewed by a relatively small number of high-income people. Warren Buffett and George Soros both are eligible for Social Security, for example. But if you rank all recipients from high income to low and look for the person in the middle, you find that median recipient’s income is not only lower than the mean of the group, but also lower than the mean for the entire population. And it is lower than the median for those paying into the program.
Just as the top 10 percent or fewer of recipients skew the group’s mean income, so the bottom 10 percent of payers skews the mean income of their group. Many of these low-income FICA payers are high-school or college students working limited hours or retirees picking up a few extra dollars. The median for this group is higher than this skewed mean.
Because Social Security payments are capped, not that much money goes to rich retirees. And as one’s income rises, the return on FICA taxes one has paid falls. And while FICA withholdings take a sharp bite out of paychecks for low-income people, the total amount paid in by this group is not great. So the program does not result in an overall transfer of wealth from poor to rich. On the whole, Social Security continues to make income more evenly distributed.
Comparisons between the value of output or incomes in our country with European countries like Switzerland, Denmark, the Netherlands, Sweden and Germany are affected by the same mean-median confusion. Mean incomes in the United States, adjusted for purchasing power, are higher than in the countries just listed. But because income distribution is so highly skewed in the United States, with the top 20 percent getting most of the total increase in national income over the past 30 years, the median incomes for these countries are higher than in our own country. That is, a household in the middle of the income distribution in those European countries has a higher income than one in the middle of the income distribution here.
Moreover, that is true for most income tiers. The poorest 20 percent of households in these European countries have incomes well above the poorest 20 percent here. Ditto for the next tiers, the 20th to 40th percentiles and the 40th to 60th. One has to get past the 60th or 70th percentiles for incomes in the United States to be higher than in countries like Switzerland or the Netherlands. So while it is true that per capita output and mean per-capita income are both higher in our country than in high-income European countries, it is also true that more than half of households in these European countries have higher incomes than those at the same relative positions in the United States.
This mean-median distinction also is important in domestic income growth over time. From 1989 to 2009, mean household income, adjusted for inflation, grew 11 percent. But median incomes grew a sum total of only 3 percent over the same two decades. And while mean household earnings have grown 30 percent since 1973, as of 2009, the median earnings of men who work full-time, year-round never has regained the high point of $48,268 (in inflation-adjusted 2009 dollars) hit that year. But the reasons for that are another column.
© 2011 Edward Lotterman
Chanarambie Consulting, Inc.