Why so glum?
By Scott Sumner
America’s pretty much the same place as it was 6 years ago. If you drive around the country you see the same sorts of neighborhoods you saw 6 years ago. Incomes haven’t changed very much. But Kevin Drum has a new post that seems to show that we are nowhere near as rich as we used to be. Is that what it actually shows?
Kevin Drum points to a survey that suggests the number of self-identified upper middle class and rich has fallen from 21% to 14% over the past 6 years, an astounding 7 percentage point decline. I found that highly implausible. Not the fact that many fewer people self-identify as upper middle class, but the implication some might draw that many fewer people are upper-middle class. In other words, is this survey describing economic conditions or states of mind?
[In fairness Kevin reports the data accurately, as attitudes not actual incomes. However I worry that some might misinterpret the survey.]
The census data only goes up to 2012 income. So where do we start, 2006 or 2007? Since the 2007 income cutoff is slightly higher, I’ll use that number, which works in favor of Drum and against me. The actual drop in the top 20% cutoff was from $110,739 in 2007 to $104,096 in 2012 (all data is real, adjusted for changes in the CPI). Because inflation in 2013 was only 1.5% (against 2.1% in 2012), I believe the rise in the cutoff between 2012 and 2013 was almost certainly higher than the small rise between 2011 and 2012. So by using the 2012 cutoff instead of estimating the 2013 cutoff I am making another assumption that biases the results toward a bigger loss of upper middle class people than actually occurred over the past 6 years.
But even with those generous assumptions, that’s a very small drop in the cutoff real income for the top 20%. How many people lie between those two numbers? I don’t know, but I’m almost sure the answer is “close to 2%.” That’s because between the 60th percentile and the 80th percentile cutoffs, about 1% of the population drops off as income rises by $2000. Between the 80th and 95th percentile about 1% drops off for every $6000 increase. The lower range suggests about 3% would drop off between $104,096 and $110,739, and the upper range suggests about 1% would drop off. Since bell shaped distributions with 100,000,000 households are “super smooth,” it’s got to be close to 2%. If you don’t follow statistics, trust me on this.
So of the top 20% of households in 2007, the net loss in 2012 is about 2%, and I’d guess in 2013 it would be about 1.6%. But even a 2% drop in the upper middle and rich groups is tiny compared to the reported 7%. So why are Americans so glum? After all, the gains between 1997 and 1999 were bigger than the losses over the past 6 years, and no one was saying that America was a completely different country in 1999 as compared to 1997.
Perhaps what really matters is how we are doing relative to expectations. We’ve been a country of growing affluence for decades. The Bush/Obama economic policies have pushed us onto a slightly downward trajectory. The losses have been small, but the losses relative to expectations have been fairly large.
PS. Of course I’m joking about Bush/Obama, as other developed countries suffered the same real income declines. I’m not sure the cause of this decline (other than monetary policy, which plays a role), but it is almost certainly unrelated to which fool happens to be serving as King of America.
HT: Tyler Cowen.