All happy macro-economies are alike; all unhappy macro-economies are unhappy in their own way.
This post, based on some important new work by Johns Hopkins Professor Larry Ball, will present a typology of those unhappy economies in a moment, but first, let’s talk about “hysteresis,” a problem that sounds as bad as it is. It’s when a cyclical shortfall morphs into a structural one. Under hysteresis, key inputs like capital investment or the labor force fall in recession, but they keep falling or fail to recovery much in the upturn. This, in turn, generates potentially large “output gaps”—lost GDP and jobs—and, if it lasts long enough, can even slow the economy’s potential growth rate even when it’s fully back on its feet.
It’s like a patient whose long illness has reduced their baseline health, even upon full recovery. Or sticking with Tolstoy (and building on the econo-literary tradition introduced by Thomas Piketty), had Anna and Vronsky had a quick fling, no hysteresis. Like a short, V-shaped recession, their brief affair wouldn’t have necessarily altered the course of their history. By keeping it going, they invoked all kinds of trouble.
Unfortunately, as we’ve learned all too well in recent decades, economic cycles have been less V-like than…well, let us count the ways.
From the hysteresis perspective, the bounceback, or V-shaped recovery, is the most benign. It’s your grandfather’s recession: the economy hits a snag, a bunch of folks get furloughed from the factory, snag plays out, whistle blows, and they’re back to work. Demand, stored up and untapped in the downturn, quickly returns, output gaps are closed and the pre-recession growth path is rejoined.
By contrast, the last three recoveries have been more backward-leaning L’s than V’s. We even had to invent a name for them: “jobless recoveries.” If you’re scratching your head asking how we can have a bona-fide recovery without much job growth, well…welcome to my world.
To document the extent of hysteresis in the wake of the Great Recession, Professor Ball examined the damage in 23 countries. While a few countries escaped unscathed, the costs of the protracted downturn are steep (economically large losses in real income and jobs) and lasting (hysteresis).
To read his key results, you need to understand three lines, all of which track real GDP growth. The actual line just shows the path of real GDP. The other two-lines are trends showing the underlying potential growth path of these various countries’ GDP. One line shows pre-recession potential (PRE) and the other shows post-recession potential, or POST (the latter is indexed to zero in 2000 in all figures).
Here, for example, is the US figure. The loss of potential output between the PRE- and POST-recession trends is about 5% of GDP. That’s a loss of about $900 billion, or about seven million lost jobs, or about $7,500 less income per household per year.
Ball’s figures (see his Figure 2) show that in most countries, the recession led to large and persistent output gaps. Worse, in many counties, it also looks to have lowered the economy’s potential growth rate, well after their economies recovered. That is, in some cases, it’s not just that POST is below PRE, as in the US. It’s that the slope of POST is below that of PRE (also slightly the case in the US, but I’ll show much worse cases below). The patient may be recovered, but she’s moving a lot more slowly than before her illness.
Sticking with Tolstoy’s theme of differences among unhappy economies, in Ireland, the growth rate of POST is far below PRE, meaning the great recession was far more damaging to future growth there than here. Greece, predictably, looks even worse.
Conversely, Germany came away largely unscathed: PRE, POST, and Actual are all close to coincident.
This implies a rich area of future research into the different policy architectures that generated these different results. Can we identify a policy set that insulated certain countries from the shampoo cycle—bubble, bust, repeat—that’s been so pervasive of late? I suspect financial market oversight, including that of housing finance, along with independent currencies, and non-austere monetary and fiscal reactions to recessions will surface as key factors.
Aggregating across all the countries in his sample, Ball finds that over $4 trillion in potential output is lost. And remember, relative to the pre-recession trends in these countries, that’s an annual loss that just keeps on losing.
Those are the economic magnitudes you should be thinking about when you hear certain economists complain about the policy mistakes we’ve made both in allowing bubbles to form and not aggressively mopping up the damage when they implode.
The question is: must these losses be permanent? Can hysteresis be reversed such that output gaps are closed and the post-recession trend accelerates to something that approaches the pre-recession growth rate? I think in some cases it can, but far more importantly, so does Fed Chair Janet Yellen, who recently said, regarding the decline in labor force participation, “…some “retirements” are not voluntary, and some of these workers may rejoin the labor force in a stronger economy…a significant amount of the decline in participation during the recovery is due to slack…”
The longer we ignore hysteresis, the worse, as in the more entrenched and thus harder to reverse, it becomes. A significant program to repair and rebuild our national infrastructure, with an emphasis on hiring some of the long-term unemployed to do the work would help, as would direct job creation programs for people who want work but have given up looking for it. But to continue to do nothing is to accept what ought to be unacceptable, and to do so at a very high cost indeed.