Will we have a depression?
By Scott Sumner
Tim Duy has a recent post that looks at the risk of an economic depression. I mostly agree with his comments, but would like to slightly reframe a few of his points:
This isn’t a Great Depression yet but instead a Great Suppression. There was nothing “broken” in the economy in February in the sense of massive imbalances that threatened to be unwound over the course of years. With nothing broken, the economy wants to get back to work and will as soon as we let it. You may not like the level the economy is at when that happens, but market participants will like seeing the light at the end of the tunnel.
Some readers might infer from this comment that the Great Depression was preceded by massive imbalances. That was not the case. The US economy was extremely well balanced in mid-1929, right before it plunged into the worst depression in history. So then why did the Great Depression occur? Because of policy mistakes beginning in late 1929—mostly tight money, but also Smoot-Hawley, income tax increases, and Hoover’s high wage policy.
Didn’t the bank failures of the 1930s expose the fact that banks were overextended in 1929? Not really. The bank failures reflected America’s flawed unit banking laws and, much more importantly, the foolishness of a monetary policy that cut NGDP in half. Could the boom and bust in stock prices during 1929 have caused a depression? No. The quite similar boom and bust in 1987 didn’t even produce a tiny recession, much less a Great Depression. We also had a budget and trade surplus in 1929. There was no inflation. NGDP growth during the 1920s was a tiny bit above average, but not unusual. There were no imbalances worth mentioning.
Unfortunately, this means that we should not take much reassurance from the lack of imbalances. And yet, Duy might be partly correct. I suspect Duy was thinking about the imbalances that preceded the 2008 recession. Even there I believe monetary policy played a bigger role than has been acknowledged, but you could argue that the collapse of the real estate boom depressed the equilibrium interest rate, making the Fed’s job somewhat more difficult. If the equilibrium interest rate returns to 2019 levels after the epidemic ends, then a fast recovery might be possible.
But this rosy scenario depends on the current slump not permanently reducing the equilibrium interest rate. And that depends on monetary policy. If the Fed allows inflation and/or NGDP growth to slow too much, it can depress the equilibrium interest rate so much that unconventional methods are required to generate recovery. Duy also shares this concern:
A key reason the economy hasn’t slipped into a depression is that enhanced unemployment benefits keep pumping money into people’s pockets even with the unemployment rate surging to 14.7%. The stimulus checks and payroll protection program are providing a boost as well. The economy won’t have time to heal before those enhanced benefits expire; an extension is needed to smooth the transition into recovery (aid to state and local governments is also critical). We might not – and probably won’t – get the same firms coming back as we had going into the economy, but if households have money to spend, they will support a new, post-virus range of businesses. If they don’t have money to spend, the economy will get locked into a substantially lower equilibrium. (emphasis in original)
I basically agree with the final sentence, although I’d put relatively more weight on monetary policy (especially level targeting) and less on fiscal stimulus. But I’d slightly quibble with the first sentence. Unemployment rose from 3.5% in February to 14.7% in April, and many expect it to hit 20% in May. You could easily argue that the economy has already in fact “slipped into depression”.
Again, Duy may be on to something here if we slightly reframe his argument. There are a couple data points that differentiate the current slump from a garden-variety depression. First, the drop in spending is being driven by social distancing, not an unwillingness to spend money. People are unable to spend, or perhaps afraid to spend. As a corollary, firms like Walmart are actually booming, which is not typical in a depression. Second, stock prices are only modestly depressed, which suggests that traders believe the economy will bounce back after the pandemic ends. In a 1930s-style depression, stocks would almost certainly be lower right now. So we haven’t fallen into a typical demand side depression.
That doesn’t mean that all’s well. Part of the strength in the stock market is due to lower long-term interest rates. Cash flows are being discounted at a lower interest rate. And TIPS spreads are still quite low, indicating that inflation is likely to stay below 2%. We could easily have a sub-par recovery.
Some of this may be semantics. Economists have never precisely defined the term “depression”, and indeed even “recession” is somewhat vague. (The 2001 recession did not see two consecutive quarters of negative growth.) I’d say that 20% unemployment is enough to qualify as a “depression”, but perhaps not if unemployment falls below 6% by the fall, as Lars Christensen has forecast.
I’d rather not focus too much of the issue of how to label this slump, which seems less important than thinking about what sort of problem we are facing, and what sort of policies are appropriate. In my view, the best policy today is identical to the best policy in 2019—level target NGDP along a 4% growth trend line, with policy focused on stabilizing expectations of NGDP 12 months out.