Damned Lies, Statistics, and the Great Recession

Free Money Finance last week pointed me to an AP article from the end of September headlined “Income gap widens as poor take hit in recession”. (Your Soup Kitchen Crop local news outlet may have used a different title.)

The piece recounts how the recession has lowered incomes across the board but hit the deserving poor and ever-wholesome middle class harder than the fat cats on the top. Kinda ho-hum as mainstream media reports on the Great Recession go. The AP story was run in a few places (see link above and here and here too) but hardly made a splash. Why would it? It just confirms what everybody knows to be true.

But it isn’t true. Or, at least, the conclusions in the article are completely unsupported by the data it pretends to be based on.

The story was based on a release of data on incomes from the Census Bureau and an accompanying press release, both of which are apparently available only to members of the media. Your tax dollars at work.

So we are left to speculate whether the AP reporter came up with the Great Recession spin on her own or just summarized what the mandarins at the Census Bureau put out. Not that it really matters.

The big fly in the ointment is that the Census Bureau operates at a glacial speed that would never be tolerated outside of government. The “new” data released at the end of September 2009 is for 2008. And not December 31, 2008, mind you. It reflects monthly surveys taken throughout the year 2008, meaning that when released the data was between 9 and 21 months old.

And it gets worse. The income numbers the Census Bureau collects are not what survey respondents are currently making but what they actually made over the previous 12 months.  So if a respondent had been unemployed for a few weeks he would still put down a decent sounding, and out of date, income.

What we have then is average income data based on a time period that ranges from January 2007, the start of the period reported on by respondents in January 2008, and December 2008, the end of the period reported on in that month. The center of this averaging period is, of course, December 2007, which, coincidentally, is the month the recession started.

This isn’t a snapshot of the effects of the Great Recession, it is a snapshot, albeit a blurry one, of the crest of the Bush Boom. (A term I made up just now. Like it?)

The AP tells us that “Median income fell last year from $52,163 to $50,303″. Ignoring what is meant by “last year” for the moment, there is an important missing word here. It is “real”. Those are inflation adjusted numbers, a fact not mentioned anywhere in the 744 word story.

Boiled down, the census data tells us that median real incomes fell between 2006 and 2007 by 3.7%. Inflation, as measured by the CPI, was 4.1% in 2007. (I know it was long time ago, but you may dimly remember gas at $4 a gallon.) In nominal terms, median income actually went up. Now I would be the first person to point out that it’s real incomes, not nominal ones, that count. But real incomes declined in 2007 because of high energy prices, not because of a recession that hadn’t started yet.

Of course, the main thrust of the story is not that incomes declined overall, but that “The recession has hit middle-income and poor families hardest, widening the economic gap between the richest and poorest Americans….”

This sort of apocalyptic hand-wringing over how the rich get richer and the poor get poorer is nothing new. It dates back, at least, hundreds of years and may just be inherent to the human psyche. And there are some challenging issues around how you measure the disparity between rich and poor and some profound philosophical questions of justice involved, all which I don’t have the energy to get into here.

What I will say is that the college professors and op-ed columnists who concern themselves with this troubling trend always seem to present data like this stuff from the census in a near vacuum. They never mention that a widening gap between rich and poor is everywhere, and has always been, a long-term side effect of economic development. Far from being a new and disturbing turn for the worse, it’s something that has been going on, as far as we can tell, forever.

Of course, in the short term, from year to year, the gap between rich and poor can get larger or smaller. There’s a lot of random noise involved in the measurement of such things, but as a general principle, when the economy is doing well the gap gets bigger and when it is doing poorly it gets smaller. The reason for this isn’t terribly interesting. Higher incomes tend to be more variable and more tied to the economy.

So if the Great Recession increased the gap between the highest and lowest earners that would be news. In a few years, when the government gets around to releasing useful data, we will find out if this happened. In the meantime, what we know from this census data is that in the modestly growing economy of 2007, the ratio of the income of highest paid 10% and the poverty line went from 11.2 to 11.4. Golly.

Anyway, the point of this shockingly long post is just how many layers of ineptitude and innumeracy there are here. And remember, we’re not talking about a soft human interest story on how the Great Recession has affected the lives of a few interviewed people. This is a write-up of a release of government data.

The government needs nine months to collate data that would take nine hours in the private sector. Then it puts out a secret press release on it. The wire reporter assigned to cover it understands neither the basis for the data nor the larger context. Instead she jams it into the standard motif of how the Great Recession is hurting the least well off the most. Said piece is run in a few places and is quickly forgotten because, I assume, of its dog-bites-man non-news nature. Then a few weeks later a maladjusted blogger stumbles across it and writes a too-long post about what a total crock it is.

Such is the state of our money culture. Even on subjects that are primarily quantitative we tolerate media coverage that is driven by what we want to think is true rather than what is actually true.

In fact, let’s be honest, we don’t just tolerate it, we encourage it. Ambiguous complexity is such a bore. Just keep repeating those acceptable themes we know are true. Don’t try and tell me that house prices tend to go up with inflation but are subject to speculative bubbles like any capital good. That’s not the advice I want. It makes my brain hurt.

7 Comments

No comments yet.

RSS feed for comments on this post. TrackBack URI

Leave a comment

WordPress Themes