“Median household income under eight years of Obama was up $975. That’s eight years. Under eight years of President Bush — we had President Obama, $975. President Bush, $400. Eight years, eight years, $400, $975. Me, two and a half years, it’s up $7,000. $5,000 plus $2,000 for the tax cut that everybody got.”
The president has a shiny new talking point, courtesy of some opinion articles by Stephen Moore, a former campaign economic adviser whom he briefly nominated for a seat on the Federal Reserve Board. Moore argued that the median household income gain under Bush was a little over $400 and under Obama about $1,000, compared with $5,000 under Trump. He also added in the supposed tax cut savings.
But as regular readers know, The Fact Checker frowns on slicing and dicing economic data by presidential terms. Despite how much presidents like to brag about a successful economy under their watch, the nation’s economy is derived from many factors that do not involve presidential decisions — and are far beyond a president’s control.
In this case, the time periods are especially misleading. (Note: Contrary to what Trump says, Moore works for the Heritage Foundation, not Moody’s, and the numbers are already adjusted for inflation, so the president can’t say the gain under Bush would be “wiped out by inflation.”)
That’s what greeted Obama when he became president. The National Bureau of Economic Research deems that recession to have ended June 2009, though the Great Recession was so brutal that the economy struggled for months afterward to recover.
So the numbers for both men are dramatically affected by those recessions.
Moore actually dated the gain in median income under Obama from June 2009, not the start of Obama’s presidency, telling The Fact Checker that he didn’t think it was fair to tag Obama with a recession that he inherited from a previous president. That, to us, underscores why dating economic presidential terms is so artificial. Median household income tends to keep declining for a period after the official end of a recession, so Moore’s method of starting the clock still does not account for that phenomenon.
Moore also acknowledged that Bush’s numbers were affected by two recessions and said he should have accounted for that. (While the first recession under Bush started a couple of months after he became president, it was certainly not the result of anything Bush did; economists attribute that recession to the burst of the bubble in technology stocks under Bill Clinton.)
Now let’s explore where the numbers come from, because the official census numbers do not show nearly this kind of gain under Trump. The Census Bureau said median household income in 2018 was $63,179 (in 2018 dollars), a gain of $1,380 since 2016 but not statistically different than in 2017. That’s a far cry from the $5,000 gain in income claimed by Trump and Moore.
(Not to get too much in the weeds, but the Census Bureau redesigned its income questions in 2014 and its processing system in 2018, making comparisons to previous years difficult. The bureau says that when adjustments are made to account for the differences, 2018′s annual median household income would be lower than 2007, 2000 and 1999, after adjusting for inflation.)
Moore instead is using data from a private firm called Sentier Research, run by Gordon Green and John Coder, who had been senior officials at the Census Bureau. They examine data from the Current Population Survey, which is used to calculate the unemployment rate, and come up with an estimate for a monthly median household income number. They also adjust the numbers each month to account for inflation, so the most recent figures are in August 2019 dollars, dating to 2000.
The advantage of this approach is that Sentier is able to spot trends sooner than the annual release of the official census figure. The disadvantage is that the figures are based on a relatively small survey, and so the numbers can bounce around a lot. But over time the trend line appears to track the official number, which is an annual average. Politicians often have cited Sentier research, as has The Washington Post.
Green, who at one time was chief of the income statistics division at the Census Bureau, provided The Fact Checker with a spreadsheet of the Sentier data and we produced some charts with it.
The first chart goes back to 2000, and it clearly shows that the trend under Trump is a continuation of the increase in median income that started under Obama after median income bottomed out at $55,828 (August 2019 dollars) in June 2011. Notice that is two years after the official end of the recession, indicating the severity of the Great Recession, though Moore argues that this is the fault of Obama’s policies.
“The problem with counting Obama from the bottom of the income slide is that this rewards Obama for the dismal performance in his first three years," Moore said. "The income gains are only making up for the rotten performance under the fiscal stimulus.”
When we compare the last 31 months of Obama to the first 31 months of Trump, the trend is even clearer. In those 31-month periods, median household income rose 5.6 percent under Obama and 7.1 percent under Trump, respectively. That gives a slight edge to Trump but remember these monthly numbers bounce around a lot, so it’s unclear whether the recent pace under Trump can be maintained. In any case, the economic record of the two presidents is far closer than suggested when including the effect of the recession under Obama’s numbers.
Since June 2011 through the end of Obama’s term, median household income rose 10.3 percent. Trump, in effect, is piggybacking off that growth.
But Sentier’s research does suggest that the Census Bureau annual figure for 2019 could show an impressive increase.
“My guess is that much of the jump in the last three months is error, but as the labor market has continued to tighten, wage growth has picked up steam, so I wouldn’t surprise me if we see some acceleration in income growth from the Obama years,” said Dean Baker of the Center for Economic Policy and Research.
(As for Trump’s statement that one could add $2,000 because of his tax cuts, the effect of his tax cut is widely scattered across income groups, with the biggest gains going to the biggest wage earners. Data released by the Ways and Means Committee, showing the impact on median household income by congressional district, suggests that $1,500 to $2,000 is in the ballpark for congressional districts with incomes that mirror the national average for household median income. The Tax Policy Center found an average tax reduction of just under $1,000 for the middle quintile of taxpayers.)
The Pinocchio Test
Trump may have reason to crow about the trends in the growth of household median income, but his framing is misleading, especially when he compares himself with Obama. In general, the gains documented are mostly a continuation of a trend started under Obama. Trump earns Two Pinocchios.
Send us facts to check by filling out this form