Maybe Not So Volatile?

When Jacob Hacker's Great Risk Shift came out, many liberals were super excited, here's a brand new argument for a bunch of conclusions I already agree with. I had no quarrel with Hacker's data, showing a rise in economic instability over the past few decades, but I wasn't so excited about his book which struck me as an unduly esoteric argument on behalf of policies that are perfectly defensible on other grounds -- I think, for example, that a better-designed health care system would boost economic growth, improve public health, and enhance social justice and any consideration relating to volatility is putting us on track to debate a side issue.

Well, now here comes the CBO and its very credible director Peter Orszag to report that "In previous work released in 2007 ... report concluded ... earnings volatility had not increased" and in a new study "preliminary results suggest that household income is much less volatile than individual worker’s earnings, and that household income volatility has not increased over time — and perhaps even declined slightly." Now as I say, I still think the vast majority of Hacker's policy ideas are perfectly good whether or not the CBO is right about this but obviously this is going to be a problem for folks who've tried to hang progressive policies on Hackerian arguments so it's no surprise that Greg Mankiw's linking to the report.

Comments

"preliminary results suggest that household income is much less volatile than individual worker’s earnings, and that household income volatility has not increased over time — and perhaps even declined slightly."

Presumably this is due to a higher percentage of households with both people working, which decreases household volatility even if individual volatility is going up. Or, is it due to some other source of income outside of worker's earnings?

Posted by Jim W | January 17, 2008 11:47 AM

I'm not a statistics guru; here's a link to the only set of data referred to (one link beyond Matt's):

http://www.cbo.gov/ftpdocs/89xx/doc8914/01-05-2007-SGEPresentation.pdf

The interesting stuff starts at page 21.

As best I can tell, this looks pretty cherry-picked to me. For instance, at page 21, they lump 25% or less household income increases with 25% or less household income decreases. WTF's up with that? Or on page 22, what's the basis for choosing 1984, 1995, 1998 and 2002 to compare income declines? What makes those years magic?

Feh.

Posted by vokosigan1 | January 17, 2008 12:17 PM

Exactly what I was going to point out: individual income volatility is going up. It would take a very complex kind of stupidity to argue that this doesn't matter. I notice this report was released in '07 and therefore doesn't account for the real estate meltdown.

I couldn't finish Hacker's book. Not that I thought it was bad, but....just basically what Matt said.

Posted by jonnybutter | January 17, 2008 12:19 PM

Exactly what I was going to point out: individual income volatility is going up.

Is it? The report Matt linked to doesn't seem to say that.

It would take a very complex kind of stupidity to argue that this doesn't matter.

If it's true, it may matter to some degree, but household income volatility matters more.

Posted by Mixner | January 17, 2008 12:31 PM

matthew, if you would read economist's view, you would accomplish two things: a.) you'd discover reports like this when they were issued and not when greg mankiw notices them; b.) you wouldn't give any traffice to a dishonest hack like greg mankiw who sold his birthright as an economist for a mess of near-to-power bush administration potage.

Posted by howard | January 17, 2008 1:33 PM

The report Matt linked to doesn't seem to say [that individual income volatility is going or has gone up].

My mistake. I misread.

household income volatility matters more.

Why is that, please?

Posted by jonnybutter | January 17, 2008 1:41 PM