Labor Market Woes: Keynesian v. Real Business Cycles?

David Altig received some heat from Brad DeLong:

Look at the what the figure tells us about the end of 1982. It tells us that at the end of 1982–in the middle of the deepest recession of the post-WWII period–after three years during which labor force participation had hardly grown at all after growing by 0.4% per year throughout the 1970s–labor force participation was not depressed below its long-term trend. If macroblog is going to say that the slowdown in growth in labor force participation between 1979 and 1982 is not a sign of a slack labor market but of “efficient changes in labor force participation” due to factors analogous to “summer vacations… winter holidays… every weekend” that cause fluctuations in labor force participation that we “would never think of calling… ‘gaps’–well, then, I wish it luck out there in the Gamma Quadrant.

He has also been receiving some heat from AB and Macroblog reader knzn (memo to Vanna White: we don’t need no stinking vowels). David replies:

My point is that I am skeptical about the position that sluggish job creation has been the result of old-fashioned deficient demand, amenable to being fixed by stimulative fiscal and monetary policies.

Even though I see this trend-cycle debate to be a bit of a distraction, I’d like to take on Brad’s comment by noting a 2001 paper by David T. Ellwood entitled The Sputtering Labor Force of the 21st Century: Can Social Policy Help? (with hat tip to Mark Thoma for emailing this to me):

In the last twenty years, the overall labor force grew by 35 percent and the so-called prime age workforce those aged 25-54 grew by a remarkable 54 percent. The number of college educated workers more than doubled, and increased as a fraction of the labor force from 22 percent of the total to over 30 percent. In the next twenty, there will be virtually no growth in the prime age workforce at all.

While Ellwood’s paper went well beyond our debate on the employment to population (EP) ratio, Ellwood documents the substantial rise in the labor force participation rate (LFP) from 1980 to 2000 – with much of this increase coming from an increase in the share of women who were in the workforce. While he predicts a fall in the growth rate of employment, his paper does not explicitly suggest a fall in the EP ratio. Ellwood noted forecasts that LFP for each subgroup (e.g., prime age workforce v. those 55 and older) would likely remain roughly the same over the 2001 to 2020 period.

Before addressing David’s suggestion of real business cycle effects, let’s tease out a couple of implications of Ellwood’s paper. I have provided three charts of LFP and EP with the first one being from 1975 to 1984, which covers the 1979 to 1982 downturn. The second one is from 1985 to 1994, which covers the Bush41 recession. During both periods, the overall trend for LFP was upwards so the cyclical efforts of the two recessionary periods showed up more in the EP ratio as well as the unemployment rate, which gave similar signals. These two signals have diverged during the latest recession and partial recovery (see the third chart covering 1995 to July 2005), which is the genesis of the current debate on LFP. But focusing only on LFP is sort of like looking at supply but not demand (more on that later).

Perhaps the best way to wrap up this discussion of trends v. cycles would be to note my own going musing with Bill Polley who noted:

Demographics will start working against us on this series very soon as well if some baby boomers start to retire early. Whether the e/p ratio can attain the level at its last peak is a serious question, just as valid today as in 1995. Like PGL, I would like e/p to get back to something more akin to full employment, but I also think there should be research into what that level is for the demographics.

If Ellwood is correct that the subgroup LFP will show no long-run trend, Bill is suggesting that as a larger share of the adult population comes from the 55 and over crowd – we will see a gradual decline in LFP. As I read Bill’s many very wise comments, I think he agrees with Brad that little of the recent decline in LFP is from demographics.

Let’s especially note that Tyler Cowen has joined David in reviving Kydland and Prescott’s real business cycle ideas. While knzn is arguing for a Keynesian explanation to the fall in the EP ratio, David is arguing for a classical demand and supply explanation. But not only do we need to think in terms of quantity variables (EP and LFP), we should also think in terms of real compensation, which have not kept pace with productivity, and real wages, which have been flat. The introduction of price as well as quantity variables is where I think Brad has David cold.

But to be fair, let’s consider four possibilities of a real business cycle starting with one that Bush supporters such as Prescott often mention:

(1) The reductions in marginal tax rates were supposed to induce higher after-tax real wages, which would encourage people to work more hours and greater entry into the labor force. But of course, we see little evidence of this effect even if the National Review keeps claiming employment has never been higher.

(2) David has mentioned the oil price shocks, which would tend to increase the cost of consumer goods relative to the price of goods domestically produced. Certainly, any such unfavorable shock would lower wages relative to consumer prices inducing a fall in LFP. Such an adverse terms of trade effect would also tend to lower profits for American companies, which is not consistent with what we have observed. Also note that the consumption price deflator has risen only 10.8% since 2000, while the GDP deflator has increased by 11.6%

(3) Kash and others have noted a third possible real business cycle effect – that being the increases in the cost of providing health care, which would explain why real compensation as traditionally measured has increased even if workers are get less in goods and services from their total compensation. In other words, the argument that has been so ably put forth by Kash and others would predict at least a modest reduction in the market clearing level of employment.

(4) I suspect these effects are minor as compared to the favorable productivity increase of the late 1990’s and its resulting investment boom. During the late 1990’s, real wages and LFP were both rising significantly. We witnessed an investment bust starting in 2001 (or even late 2000) – so might one argue we are in the flip side of the positive real business cycle shock of the late 1990’s?

I should mention one problem with my fourth and favorite real business cycle suggestion, which is the fact that productivity seems to continue to rise. Perhaps the investment bust was more along the animal spirits argument of Lord Keynes, which of course brings us back to the point knzn has been trying to make to David – as well as to me. Knzn – let me just suggest that you are preaching to the choir.