Keepin’ it real on the growth slowdown: the first of many factoids from the 2017 ERP

December 20th, 2016 at 4:19 pm

President Obama’s Council of Economic Advisers just released their final Economic Report of the President and it is the perfect holiday gift for your favorite econo-nerd. (It’s also, I fear, the last cogent ERP we might see for a bit.)

So, for the next few days, I’ll highlight some of the arguments in there that really resonated with me and some of my colleagues.*

Episode 1:

Whenever people bemoan the slowdown in GDP growth in recent years, part of me moans with them but part of me doesn’t, because some of the growth deceleration is a function of slower population growth. Remember, growth is basically productivity plus labor input, and an aging population tends to slow the latter.

Thus, whenever you’re making long-term historical comparisons over periods where this population growth factor is in play, you must account for it, by looking not simply at real GDP growth, but at some measure of per-capita growth.

Think of it this way. Suppose GDP’s growing at 3 percent, and the population is growing at 1 percent. Thus, GDP/capita is growing at 2 percent. Now, suppose both slow half-a-percent. You could complain about slower aggregate growth, but on a per-capita basis, growth hasn’t changed at all. And that means there’s the same amount of income per person to go around (obviously, we’re not talking about inequality yet–that’s to come in later posts).

The figure below shows the sharp deceleration in the growth rate of the working-age population, meaning we’d expect overall growth to slow, just based on the decelerating trend of this input.

Source: 2017 ERP

Source: 2017 ERP

The next figure is one I made, but totally ripped off from the ERP (Fig 2-viii; I remade it so I could calculate the “slowdown” bar, which isn’t in their figure; the current cycle looks a little different in my figure, maybe because I used more recent data). Overall GDP growth slows by a very significant 1.9 percent in the first set of bars. But the deceleration is half that much if we use the working-age population, and about a third if we look at “per labor-force participant.” (I’d argue that the slowdown under this latter measure is a bit biased by weak labor force participation having to do more with insufficient demand than demographics; i.e., it’s “endogenous”.)

Source: My version of Fig 2-viii from 2017 ERP, (BEA, BLS)

Source: My version of Fig 2-viii from 2017 ERP, (BEA, BLS)

Trust me when I tell you that none of us is downplaying the importance of that slower growth rate under these working-age population-adjusted measures. If anything, those negative bars represent what I (and I believe CEA) consider our most pressing macroeconomic challenge: the slowdown in productivity growth (about which the ERP has many excellent figures which I’ll parade out soon).

But it is somewhere in between incomplete and misleading to complain about the slowdown in GDP growth without accounting for the sharp slowdown in the growth of the working-age population.

*Do not confuse this with my Best CBPP Charts of 2016, as that’s still to come. IKR!: an embarrassment of riches.

Print Friendly, PDF & Email

4 comments in reply to "Keepin’ it real on the growth slowdown: the first of many factoids from the 2017 ERP"

  1. William Meyer says:

    Could you explicate your “Slower growth partly due to aging” graph a bit more?

    What exactly are the units in the second and third set of bars? What is the definition of your terms “working age population” and “labor force participant”?

    Thanks for any clarity you can provide.

    Also, I would suggest you disaggregate the 1960-2007 bar. I tried a similar exercise; I calculated the average compound annual growth rate of each decade on a per capita basis since 1940. I found that U.S. per capita growth has been slowing each decade pretty much for that whole sequence, with the sole exception that the 1960s had a higher growth rate than the 1950s.

    This raises obvious questions. Why are economists so down on command-and-control economies? The U.S.’s experience with central planning during the 1940s was frankly a roaring success, economically speaking (strongest growth rates in U.S. history.) Did the turn away from Keynesianism after the 1970s make sense, given that the Keynesian years had higher per capita growth? Did Reagan’s deregulation empower anybody other than finance and private equity–it certainly didn’t fire up 1980s per capita growth rates higher than the awful stagflationary 1970s. If greater openness to trade and financialization (e.g., derivative markets) are so awesome, why has per capita growth slowed seriously since 1990 even as both trends have spiked? It looks to me as if ideology and financial industry profits have mesmerized a lot of people into ignoring their own lying eyes, including the economics profession.

    • Jared Bernstein says:

      In the second bar, the growth rate is for real GDP per working age (16-64) year-old person. In the third set, it’s for real GDP per labor force participant.

  2. SPENCER says:

    It may be worth pointing out that over the last 5 years while real GDP growth averaged 2.1% the real growth rate of non-farm business output averaged 2.6%. The 0.5 percent point difference reflects the direct impact of tight fiscal policy imposed by Congress. It also implies that Trump can easily get 3% real GDP growth simply by allowing government to grow..