On Wednesday, the New York Times, Wall Street Journal, and Washington Post all ran above-the-fold, page-1 headlines touting recent gains in income for the typical American household. President Obama appeared at a political rally in Philadelphia and, after citing these gains and the existence of lower gas prices, proclaimed, "Thanks, Obama." (He really said this.) And, indeed, the top-line claim that real median household income rose 5.2 percent from one year to the next—"the first [statistically significant] annual increase in median household income" of the Obama presidency—is true. But the overall impression given by the press corps—that happy days are here again—is false.
First, the announced income gains weren't from 2015 to 2016. They were from 2014 to 2015. Thus far in 2016 (for which income figures aren't yet available), the real gross domestic product has grown at an anemic annual rate of 0.8 percent in the first quarter and 1.1 percent in the second quarter—according to the Obama administration's own Bureau of Economic Analysis. (That won't help boost Obama's last-place standing among postwar presidents in terms of economic growth on their watch.)
Secondly, the income figures in question come from the Current Population Survey (CPS), which the Census Bureau redesigned two years ago. That redesign was undertaken in hopes of better capturing all income, particularly retirement income and income from government transfer-payments. The Census Bureau expected that the redesign would boost income tallies by about 3.0 percent, and in fact it did almost exactly that—boosting them by 3.2 percent. (Compare "2013 1" and "2013 2" in table A-1 for the tallies with and without the redesign.)
Therefore, income figures for 2015 are 3.2 percent higher than they would have been without the CPS's redesign—not because of any actual increase in income, but simply because of a change in how income is now measured. That change doesn't affect comparisons from 2014 to 2015, as both years came after the redesign went into effect, but it does affect comparisons from pre-2014 to 2015. For example, when comparing median incomes from 2009 to 2015, either the 2009 figure needs to be adjusted upward by 3.2 percent to reflect the CPS's redesign, or else the 2015 figure needs to be adjusted downward by 3.2 percent. Otherwise, the comparison is one of apples to oranges.
Once that necessary adjustment is made, it becomes clear that real median household income has actually dropped since 2009, the last year of the most recent recession. The median income in 2009 was $56,731 ($54,988 increased by 3.2 percent to reflect the CPS's redesign). The median income in 2015, six years later, was $56,516—a drop of $215 (or 0.4 percent) during the "recovery."
Compare that result with results from six years after the end of each of the three prior recessions. From 2001, when the previous recession ended, to 2007, real median household income rose 1.6 percent under President George W. Bush—a weak, but still positive, tally. From 1991 (when the recession before that one ended) to 1997, median income rose by 6.6 percent under President Clinton. From 1982 (when the recession before that one ended) to 1988, median income rose by 10.2 percent—or about $5,000 per household in today's dollars—under President Reagan.
So far, the 21st century has actually been a time of income reduction. From 2000 to 2015, real median household income has fallen from $59,621 in 2000 ($57,790 increased by 3.2 percent to reflect the CPS's redesign) to $56,516 in 2015—a drop of more than $3,000, or 5.2 percent. Thanks, Obama (and Bush).
Really, however, the problems trace back even further. Incredibly, real median earnings (that's paychecks, as opposed to all income) for men working full-time were actually higher in 1972 than in 2015. (Women's earnings have increased noticeably since then, but not men's.) In 1972, during the final year of President Nixon's first term, the typical man made $51,708 (in 2015 dollars) for doing full-time work (see table A-4). In 2015, the typical man made $51,212 for doing full-time work—about $500 less than his counterpart made 43 years earlier.