Friday, May 8, 2020

What Is the Shape Of This Cyclel As A Letter: V, L, W, J, U, Or Maybe A Lazy J or Wiggly W?

For some time now it has become commonplace for people to describe business cycles by how they resemble one letter or another, although obviously this amounts to a lot of handwaving. But it does provide bright images.  Thus Trump and crew seem to believe that the US will experience a V recovery, one that will boom up as rapidly as it fell down, so the sooner we can reopen America the sooner he can get that boomy uptrun to guarantee his reelection. Somehow he fails to understand that if he gets it going too soon, the uptrun is more likely to get sideswiped by a serious Second Wave of the coronvirus tht would turn the upturn back into a downturn, making it into a W, or if, as I suspect, given that the douwnturn we have just seen pushing the unemployment rate upo 10% in two months is truly unprecedented so that the upturn will not be all that fast, the outcome will not be a neat W, but a Wiggly W.

Use of letters is recent, but debates on the shapes of macro fluctuations is very old, dating back into the 19th century.  Up until the 1930s, nearly all the discussion focused on "commercial crises" generating sharp downturns that then were asymmetrically followed by slower upturns.  This was still the view in 1913 when Wesley Clair Mitchell of the NBER and Columbia published his Business Cycles, coining that term.  I have been recently over on Econbrowser taken to describing such a pattern as a "Lazy J."  Think of a J but then tilt it to the right so its upslope gets flatter until it is flatter than the downturn.  I think this is what we are going to see now in the US, although it could turn into a Wiggly W.

In the 1930s, driven by Mitchell associated at the NBER, it became conmon to use sine curves to describe business cycles, so symmetric, and if one cuts one off at the inflection points going down and up, somethign that looks like a U.  Over on Econbrowser, Jeffrey Frankel recently posed that the current situation might end up looking like a U, although I doubt it.

After WW II it looks like quite a few fluctuations actually had more rapid upswings than downturns, asymmetric but in the opposite direction of a Lazy J, more an alert proper J, with a rapid upturn.  Over on Econbrowser David Papell posed something like this, although I really see no likelihood of a more rapid upturn to follow this massive downturn we have just seen.  This just seems out of the question.  My Lazy J looks more likely.

We have also seen some that do look like the V that the Trump gang hopes for.  Probably the most obvious and impressive candidate would be the fluctuation that bottomed out in 1982 under Reagan with an unemployment rate ovet 10%, higher even than the peak in 2009 during the Great Recesson.  After the deal was cut between Volcker and Baker for no more tax cuts in return Volcker would end his high interest rates after visiting and incoming Mexican Finance Minister Jesus da Silva threatened to default on Mexican debts to New York banks bearing those high interest rates, the evonomy zoomed back upwards through 1983, giving Reagan his "Morning in America" to run for reelection on, taking 49 states, certainly what is the hoping gleam in Trump's eye for now.

Then we have candidate L. Near as I can see the US has nver shown suvh a pattern, but recently John Cochrane was out there posing it as a possibility and claiming that was what we had during and aftwer the Great Recession.  We certainly had the sharp decline upfront, but an L suggests that the economy then does not grow at all, simply lies flat.  We may have seen such a pattern in Mexico after the 1982 debt crisis and maybe also in Japan in the 1990s after its debt crisis.  But in the US we saw growth ranging from 1.5 to 3% annually following the turnaround, certainly not a flat line, although the "no growth" story fits ongstanding GOP propaganda repeated by Trump regarding the performance of the economy under Obama, even though the growth rates after he came in were not noticeably superior to those under Obama.

As it is, what happened then looks like a throwback to the 1800s, and maybe also the most likely scenario now if we avoid another near term downturn, an asymmetric pattern of a sharp downturn followed by a much slower upturn, in short the Lazy J.

Barkley Rosser


Anonymous said...

August 4, 2014

Real per capita Gross Domestic Product for United States, 1929-1937

(Indexed to 1929)

August 4, 2014

Real per capita disposable personal income for United States, 1929-1937

(Indexed to 1929)

Anonymous said...

The dramatic upturn begins from the time Franklin Roosevelt enters the White House and the whirlwind 100 days of programs begin. Nutty conservatives have been denying the efficacy of the New Deal since Roosevelt came to office and they sought to stop any such deal before, but with Roosevelt comes the upturn of the V. said...

I did not comment on the Great Depression, Anonymous, but in fact that one was probably a V. I suspect that was part of why what noted about the move to supporting a theory of a symmetric cycle in the 30s became popular, although they favored the more U-like sine curve pattern, but both a U and a V are symmetric, their only difference having to do with the bottom and turnaround, a difficult fine point, a sharp versus more gradual turnaround. Don't thing data from 1933 accurate enough to distinguish those two, but it was certainly pretty symmetric, a pretty strong growth after FDR got in.

Jerry Brown said...

My name begins with a J, but I don't like to be reminded how lazy I am. Which is pretty lazy to be honest. Got an 'o' in my name also. Not sure what an o shaped recession/recovery would consist of though. What about 'B'?

Anonymous said...

The BLS noted that the last employment report was an underestimation; the current rate is about 20% which only corresponds to unemployment during the worst years of the Depression. What is relevant here is that while there was a V shaped recovery of per capita growth beginning in 1933, employment only gradually improved with the improvement largely relying on government New Deal programs:

May, 1975

The New Deal and the Great Depression

Rates of Unemployment

(Percent) *

1929 ( 3.2) Hoover, March

1930 ( 8.7)
1931 ( 15.9)
1932 ( 23.6)
1933 ( 24.0) Roosevelt, March
1934 ( 17.0)

1935 ( 15.2)
1936 ( 10.1)
1937 ( 9.2) Recession begins, May
1938 ( 12.5) Recession ends, June
1939 ( 11.3)

1940 ( 9.6)
1941 ( 6.0)
1942 ( 3.1)
1943 ( 1.8)
1944 ( 1.2)

1945 ( 1.9) Truman, April
1946 ( 3.9)
1947 ( 3.9)
1948 ( 3.8)

* Numbers including employment in New Deal programs.

-- Michael R. Darby

Anonymous said...

There may well be a sharp increase in per capita growth from quarter beginning in July, but employment by my understanding of recent recessions and the Depression is likely to be slow in improving unless there is express government hiring or another New Deal.

Anonymous said...

May 8, 2020

If the workers who were recorded as employed but absent from work due to "other reasons" (over and above the number absent for other reasons in a typical April) had been classified as unemployed on temporary layoff, the overall unemployment rate would have been almost 5 percentage points higher than reported (on a not seasonally adjusted basis)....

[ This means about 20% unemployment currently. The proper analogy is the Depression, especially in light of recent recessions showing long lags in employment recovery. ]

Anonymous said...

Paul Krugman @paulkrugman

Friday's employment report was devastating — and reality almost surely worse. The BLS itself says that the unemployment number was probably understated 1/

If the workers who were recorded as employed but absent from work due to "other reasons" (over and above the number absent for other reasons in a typical April) had been classified as unemployed on temporary layoff, the overall unemployment rate would have been almost 5 percentage points higher than reported (on a not seasonally adjusted basis)....

9:44 AM · May 10, 2020

True unemployment probably ~20%. If so, worse than all but the worst 2 years of the Great Depression, according to estimates that try to apply modern concepts to available historical data 2/

But how long does this go on? Short answer is nobody knows. Long answer, I think, is that it depends hugely on the course of the pandemic — which may seem like a "well, duh" proposition but a bit more to it than that 3/

First of all, maybe a good idea to stop talking about whether there will be a V-shaped recovery, because unclear what that means. Growth in 1934-6 was very fast, but not nearly enough to bring tolerable unemployment 4/

Reasonably certain that we will have something like a Nike swoosh — recovery much slower than initial descent, but question is how fast. 5/

Past 6 recessions show two very different kinds of recovery — rapid surges after pre-1990, extended "jobless recovery" in later cycles 6/

I think we understand the difference. Slumps like 1979-82 brought on by the Federal Reserve raising rates to squeeze down inflation; as soon as it relented demand sprang back. Later slumps were Minsky moments, when the private sector realized it had overreached. Much harder to restart 7/

So where does the Covid-19 slump fit? Despite myself, I'm (very) guardedly optimistic. The virus is arguably more like the shock of temporarily high interest rates than like an overhang of excess household debt. So fairly fast recovery seems possible 8/

The market doesn't seem to agree! Never mind stocks: medium-term interest rates suggest that the market expects policy rates to be near zero for a long time 9/

I ran that chart back to 2008 in order to show that markets can be wrong. In early stages of recovery from 2008 crisis markets effectively bet on a V-shaped recovery that never arrived. Making the same mistake in reverse now? 10/

Two big caveats: First, recovery depends on dramatic decline in social distancing, which depends on virus/fear of virus. If premature opening leads to extended plateau or 2nd wave, slump can go on and on 11/

Second, while structural problems didn't bring on the slump, slump may create problems that extend its shadow — eg state fiscal crises, widespread household and business bankruptcies etc 12/

Some political irony here: Trump/GOP demands for early opening and refusal to extend relief may be what prolongs the slump, so that people are still miserable in November 13/

Anonymous said... said...

So, looks like Krugman agrees with a Lazy J, even if he is not using the terminology (which I invented). I do not buy this square root one, which just looks like a sharp V on the front end.