Hold the champagne

Posted: 20 September 2016 in Uncategorized
Tags: , , ,

us-median

Last week, to judge by the commentary on the latest Census Bureau report, Income and Poverty in the United States: 2015 (pdf), you’d think the fountain of broadly shared economic prosperity had just been discovered.

Binyamin Appelbaum is a good example:

Americans last year reaped the largest economic gains in nearly a generation as poverty fell, health insurance coverage spread and incomes rose sharply for households on every rung of the economic ladder, ending years of stagnation.

The median household’s income in 2015 was $56,500, up 5.2 percent from the previous year — the largest single-year increase since record-keeping began in 1967, the Census Bureau said on Tuesday. The share of Americans living in poverty also posted the sharpest decline in decades.

The gains were an important milestone for the economic expansion that began in 2009. For the first time in recent years, the benefits of renewed prosperity are spreading broadly.

And while the 5.2-percent increase is nothing to sneeze at (certainly not for the average American whose income, even now, remains below that of 2007 and even 1999), we need to keep things in perspective.

First, as is clear from the chart above, the gap between the average incomes of the top 1 percent and median income continues to grow. The ratio between the two has dramatically increased over time—from 8 in 1984 to 15.7 in 1999 and 19 in 2007—and remained very high (at 18.6) in 2015.

That’s not a story of broadly shared prosperity.

fredgraph

Second, while the 2015 increase in median household income was dramatic, it followed a year when median income actually fell (by 1.5 percent), after a previous increase (of 3.5 percent in 2013) and years of negative growth (from 2008 to 2012).

I hate to spoil the party. But, me, I’d keep the champagne on ice until we actually see sustained, broadly shared prosperity in the United States. And, to judge by recent years and indeed decades, that may be a very long wait.

Update

Good question, Bruce. I actually started by comparing the change in median income and real earnings—finding that there was a correlation during some periods but not in others.

fredgraph

What I found particularly interesting is that in 2008 and 2009 the change in real earnings was positive while median income fell. And then, in 2013, real earnings fell while median income rose. The interpretation? I think a lot has to do with unemployment—or, if you prefer, the Reserve Army. Real earnings rose in 2008 and 2009 for those workers who were employed (because of low inflation/deflation) but, of course, many workers were thrown out of work. The result? Median household income fell. Exactly the opposite in 2013: real earnings barely changed but the increase in employment raised median household income. Make sense?

Comments
  1. Bruce says:

    Can you give me an interpretation of the last graph (time series showing the rate of change of real median household income)?
    From that graph, it seems that, emerging from recession, the median household initially does better, but then, as expansionary growth proceeds, there comes a time when the median household does worse, and the economy then goes into recession.
    Curious how you understand this, and how it jives with the share, and share ratio, data, and the patterns they show? Why does real median household income seem to track cyclic profits better than cyclic real wages?

  2. David F. Ruccio says:

    Bruce, see my update to the original post. . .

    • Bruce says:

      Yeah, the dynamics of aggregate employment/unemployment do seem to play a role, as you say — rising employment could, probably does, explain the coexistence of stagnant earnings with a rising median (as in 2013).

      But I’m particularly interested in the periods directly preceding recessions, especially the last two (1999-2001 and 2006-08). In both cases, earnings growth is positive (or at least non-negative), which makes sense in a (relatively) “tight” labor market situation, where employment is still rising, or at least holding, and the job losses that come with the onset of recession haven’t yet begun. But in these same periods, the growth of real median household income drops precipitously, and in fact goes deeply negative — this, even though earnings growth remains non-negative and job losses aren’t yet occurring. So, if earnings and jobs are not falling (yet), why does median income fall through the floor?

      We know that in both these periods, profits were falling, in the typical end-of-expansion manner. I find it hard to believe that the median household income would be directly driven by profitability to that extent. So I guess I would want to know more about what exactly is being measured by the “real household median income” series. Why does this measure seem to collapse *before* job losses occur?

      • Bruce says:

        Hmmmm…. Taking the time to look at the underlying data helps me see it more clearly. The median household income series is strictly annual, so while the 2007 data point as shown (in your FRED graph) seems to be prior to the recession, the 2007 value is clearly picking up the recession effect of job losses (since the unemployment rate starts to rise pretty steadily from March 2007 onwards). So the “puzzle” I thought I saw for that period is less puzzling.

        But the “lead-in” to the prior recession still seems at least somewhat puzzling. Real median household income declines a bit in 2000, compared to 1999, even though unemployment continues to fall (to a low 3.9% for the entire Q4 0f 2000). So there was already a downward trend for real median income before recessionary job losses occurred, even with positive wage earnings growth. I guess that reflects erosion from inflation, though it’s not obvious in the monthly inflation numbers.

        I guess I conclude that the annual observation for median income doesn’t tell us much that’s useful in understanding cyclic turning points.

  3. […] question, Bruce. I actually started by comparing the change in median income and real earnings—finding that there […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s