Wednesday, May 9, 2018

Wages and unemployment

A few people have been talking about wage growth and unemployment lately.  I have put together a few posts on the Phillips Curve. (This was the latest.)  David Henderson has a couple of posts about wage growth, building on a post from Paul Krugman.

I have an old chart that I haven't updated in a while, so I thought I'd take a new look at it.  This is a chart comparing real wage growth to the unemployment rate.  (Here, real wages are measured with Average Hourly Earnings of Production and Nonsupervisory Employees, deflated with core PCE prices.)

Now, over the past few years, wage growth has been somewhat flat, which is why it has been a topic of conversation.  But, it seems to me that it was slightly above the norm a few years ago and has moved slightly below the norm now.  But, it really hasn't been out of the ordinary.

The big mystery in this cycle is why was wage growth so strong when unemployment was shooting up to 10%.  I think Extended Unemployment Insurance might have something to do with that.  Maybe the two minimum wage hikes in 2008 and 2009 might have had something to do with it, too.

But, there is one other adjustment we should make.  I know, readers will be shocked to hear this, but I think this has to do with the housing shortage.

I have argued that inflation is much lower than it is currently understood to be.  That is because most inflation is due to rent inflation.  Rent inflation isn't high because the Fed is printing too much money.  It is too high because there is a shortage of housing.  And, high rents have little effect on other spending, because most rent is "owner-equivalent imputed rent", which involves no cash.

Now, I think this is a reasonable measure to use for estimating the cost of living, but it doesn't really have anything to do with cash, so I think it is more accurate to remove shelter inflation from the basket of goods if we are talking about monetary policy.  And, I think it makes sense to do that when we look at real wage growth, because really, rising rents are a tax imposed on workers that is then transferred to real estate owners (who are frequently the workers themselves.)  There is no reason that these high rents should be associated with rising wages.

So, if we deflate wages with CPI inflation less food, energy, and shelter, the chart looks like this.  And, with this measure, wages have been moving right along the 2nd degree polynomial trendline.


  1. Very thoughtful post, and yet I think perhaps the elephant feels different depending on where you stand.

    My guess is that living standards, and thus real wages, are lower in LA than 50 years ago, primarily due to soaring housing costs.

    A difficult apples-to-oranges comparison. Today we have smartphones and better medicine. Back then cheaper housing, but smog.

    But the housing is a killer. A young family could move to LA and buy a house, non-workng mom through the 1950s-- 1970s. My parents did, ordinary couple. That is not possible anymore.

    In the 1980s, young people could move to town and rent a unit. Now, they double up or triple up. I moved back to LA in the 1980s, and found affordable housing. I was a nobody (a status I maintained).

    Yes, a key today is whether you own your own place or not.

    But with an average house in Seattle now at $820,000---are people actually living better than when an ordinary Boeing employee could buy a house, back in the 1980s?

    If you are renting, I would say wages are lower in LA than 30-40-50 years ago. Probably the same up and down the West Coast.

    1. I think you're basically right. And that's why I say imputed rent is a reasonable measure to include for cost of living.

      Regarding wages, if we imagine stable 2% wage growth and 2% monetary inflation, and leaving all else equal, monetary policy is shifted to 3% inflation, then we would expect wage growth to shift to roughly 3%.

      If, on the other hand, we leave monetary policy stable, but we knock down enough homes in LA to push inflation up to 3%, wages will not move up to 3% because most of that rent inflation measures phantom rent transactions of owner-occupiers.

      Eventually, homeowners will liquidate their equity by selling or borrowing, and then it will raise nominal economic activity. Then everyone gets uppity about "the housing ATM" and living high off of borrowed money, and they demand tight money even though money had nothing to do with it.