kitchen table math, the sequel: not your father's bell curve

Monday, April 30, 2012

not your father's bell curve

Sticky wages in 2011:

The tall bar in the middle represents workers who had a wage increase of $0
People to the right of the bar had pay raises
People to the left of the bar had pay cuts
Why Has Wage Growth Stayed Strong?
By Mary Daly, Bart Hobijn, and Brian Lucking

I've been meaning to post this for a while now.

I find this chart fascinating. I have never, not once in my entire lifetime, seen a curve that looked like this -- although they must be out there.

Here's the San Francisco Fed write-up:
Researchers generally point to asymmetries in the distribution of observed wage changes among individual workers as evidence of nominal wage rigidities. Figure 2 plots an example of this type of wage change distribution in 2011. The dashed black line shows a symmetric normal distribution. The blue bars plot the actual distribution of nominal wages.

The figure’s most striking feature is the blue bar that spikes at zero, indicating the large number of workers who report no change in wages over a year. This spike stands out in the distribution of actual wage changes, suggesting that, rather than cutting pay [in line with reduced earnings], employers simply kept wages fixed over the year. This is supported by the large gap to the left of zero between the actual distribution of wage changes and the dashed black line representing the normal distribution. This gap suggests that the spike at zero is made up mostly of workers whose wages otherwise would have been cut.
Once you start thinking about sticky wages, you see them everywhere.

For instance, Ed and I were watching an episode of Friday Night Lights a couple of weekends back, and the story line, which spanned multiple episodes, revolved around budget cuts to the schools. First the cuts were rumored, then the cuts were announced, and then there was rending of clothes and tearing of hair and parents mobbing board meetings to demand that cuts happen to other people's programs, not theirs. (And, yes, this sequence of events does sound familiar).

It was high drama.

Teachers would be FIRED!

Football teams would be MERGED!

Fire and flood, death and despair!

Two years ago, right up to the moment I found out about  downward nominal wage rigidities over the business cycle, * this story line would have made perfect sense to me. Like everyone else on the planet (including pick-up farm workers and their employers in India, apparently), I simply took it as a given that people can be cut but wages can't. In hard times, fear and loss (and television drama) follow directly from this belief.

But once you know about the money illusion, a school budget crisis loses most of its oomph as dramatic premise. Watching the pandemonium onscreen, I was unmoved. I kept thinking, "How about a wage freeze? How about a furlough? How about a wage cut?"

"How about everyone sit down and do some arithmetic and, while you're at it, figure out that it's not like the Dillon School District is a family where the sole breadwinner just lost their** job. You've still got money coming in, you just don't have as much money coming in next year as you did this year. (Or, if Dillon is anything like Westchester County, you don't have as big an increase coming in next year as you did this year, but you've still got an increase.) So everyone's gonna have to make do with less, but nobody's gonna starve, not unless you insist on firing the young teachers so the old teachers don't have to take a cut."

No dice.

The story arc ends with teachers getting fired and football teams getting merged, and nobody says 'boo' about the possibility of 'shared sacrifice' and the like.

I realize that, in real life, employees do take freezes and furloughs to keep everyone on the job. But they don't do it often, as the chart reveals.

* I have now consumed so much macroeconomics that I can read formulations like downward nominal wage rigidities over the business cycle almost as fast as I can read twinkle, twinkle, little star.
**It pains me to write "the sole breadwinner just lost their job," as opposed to "the sole breadwinner just lost his job," but I think the time has come.


Catherine Johnson said...

I've just realized I forgot to explain the graph.

The tall bar in the middle is people who had a $0 wage increase in 2011.

People to the right of the tall bar received pay increases in 2011.

People to the left of the tall bar received pay cuts.

Presumably, the reason this isn't a normal bell curve is that a lot of the people who logically should have had pay cuts kept the same salary year-to-year.

lgm said...

Does that chart include the 60 million on social security benefits? They did not recieve an increase in 2011, but they did get a 3.6% increase in 2012.

lgm said...

I think I'd rather see a chart using figures from the IRS. This one is from a survey.

Regular poster but anonymous today said...

From the linked site, it appears that they covered the period from 2008 to 2011. That coincides with a recession and the beginnings of the recovery out of the recession.

The people who didn't get salary increases aren't necessarily the low-wage widget makers.

I'm in software development, in flyover country, at a mid-sized company where developers might be making anywhere from $60K to $160K. (I don't know the exact number, because I'm not in management, but that's probably in the ballpark.)

Our entire company was hit with a salary freeze and cessation of 401(k) matching for a couple of years during that period. I think we thawed out around November 2010, so we probably aren't included in that spike, but had they created a similar graph for 2009 or 2010, I and my six-digit salary would have been part of that spike.

Catherine Johnson said...

anonymous - thanks for posting -

lgm - I'm pretty sure this chart doesn't include people on social security. The article is specifically about wages I **think.** (I've only skimmed at this point.)

Scott Sumner cited data from Bentley the other day, which apparently gave everyone EXACTLY a 0% wage increase last year, I think it was.

Sumner pointed out that -.5% wage decreases are far less common than 0% or .5% increases.

Given everything I've seen here, I believe wages are crazy-sticky....

(And, as lgm knows all too well, here in NY, in the public sector, raises are sticky!)

Catherine Johnson said...

So it's 2008 to 2011 --- I better change that...

Catherine Johnson said...

btw, I gather there's a dispute over the question of whether non-sticky wages would help get an economy back out of a slump.

That is, if wages (and SS, etc) were pegged to NGDP, then wages would naturally go down as well as up, so people wouldn't have to lose their jobs during a depression. Everyone would get paid less, so everyone would keep on working. (I realize that omits companies that cease to exist -- but, on the other hand, a company that can reduce its wages down to zero or close to might not have to cease to exist .... )

Anyway, I haven't delved into all that, and I don't know what I would think if I did.

However, I feel pretty confident that if people generally expected to take pay cuts during recessions AND expected to keep their jobs as a result of those pay cuts, that would be psychologically better for a majority.