September 15, 2011
I have been dismissive of the claim that the financial crisis has condemned the country to 8-10 years of high unemployment and low growth. This claim is derived from the book by Carmen Reinhart and Ken Rogoff, “This Time is Different.”
Reinhart and Rogoff perform a valuable exercise in recounting the history of financial crises over the last six centuries. They note that these crises have been followed by a long period of adjustment in which economies are subject to weak growth and high unemployment. Based on this history, they and others have argued that the United States is condemned to a prolonged period of stagnation. The moral of the story is to stop your whining and live with it.
I find this to be a case of incredibly bad induction. If we had looked at the probability that newborns would live to age five, examining random 20-year intervals in different countries over the last six centuries, we would find that in most of these intervals, most newborns do not live to age five. If we therefore concluded that we should expect children born today to die before the age of five, we would be utterly crazy. The advances in health care, nutrition and sanitation over this period make it possible for the vast majority of children almost everywhere to survive to adulthood.
I would argue that the story with economics is similar. We do know more about the economy today than we did 500 years ago or even 100 years ago. We do know how to get out of severe downturn like the present one, the real question is whether we have the political will to make it different.
In this spirit, I was prompted by a question from Ezra Klein to lay out the case as to how exactly it could have been different. So, now that I have the reins of government fiscal policy and monetary policy, let me show exactly how we would have done it differently.
First, we should recognize the size of the gap. My unit of measure is jobs, since the loss of jobs is what makes this a crisis. Germany has had a comparable loss of output to the U.S., but its unemployment rate is 0.5 percentage points below the pre-recession level. As a result, there is not the same suffering there as here.
For some rough numbers, payroll employment is down by almost 6 million from its pre-recession level. On top of that, we should have expected to see the economy generate roughly 1 million jobs since the recession began in December of 2007. This gives us a gap of 10 million jobs.
Okay, step one is work sharing. We encourage firms to reduce work hours rather than lay off workers. This can be done through the unemployment insurance (UI) system. Instead of requiring that someone be completely unemployed to get benefits (i.e. they lose their job), we allow people to get benefits for being partially unemployed in the sense that they have their hours reduced.
For example, if they get their hours reduced by 20 percent, then they may have half of this loss made up by payments through the UI system. This leaves them working 20 percent fewer hours, for 10 percent less pay. That’s not ideal, but most workers would probably view it as much better than being unemployed.
In terms of a realistic target, suppose that we could get an average reduction in work time of 2 percent (roughly 0.7 hours a week). This would translate into 3 million additional jobs.
If we think of this by sector, clearly manufacturing is the sector where shorter workweeks could be most easily adopted. If we reduced the average workweek in manufacturing by 10 percent (@ 4 hours) then it would have saved 1.5 million jobs. This would mean that in the rest of the economy we would need a reduction in average hours of approximately 1 percent to find our other 1.5 million jobs. That seems doable.
Next we turn to the government sector. Employment in this sector is down by 415,000 since the start of the recession, almost all of it at the state and local level. This is due to recession-induced austerity. In the three years prior to the downturn the government sector was creating jobs at the rate of 160,000 a year. If we had continued this rate of job growth, rather than seen job loss in this sector, we would have 1 million more people today employed in the public sector.
Now we have youth job programs. This is make-work stuff. We have a teen unemployment rate of more than 25 percent. The unemployment rate for African American teens is 46.5 percent. You’ve got a problem with make-work stuff?
Okay, so we are handing people rakes, shovels, hammers and nails. They are going to clean up streets, parks, board up abandoned buildings. They get paid minimum wage or perhaps a bit more. Ideally we could get some decent training thrown in, but the point is to keep the story quick and simple.
Employment among teens is down by 1.9 million from pre-recession levels. Employment for the age 20-24 cohort is down by 1.1 million. Let’s suppose we get half (1.5 million) of these people employed in this jobs program.
Now we go to the Fed. Paul Krugman suggested that the Fed target a higher inflation rate in the range of 3-4 percent. In his pre-Federal Reserve Board days as a Princeton professor, Ben Bernanke advocated the same thing for Japan’s central bank. Let’s suppose the Fed went this route and maintained an inflation target of 4.0 percent since the start of the downturn.
With the Fed keeping its foot on the accelerator and the short-term rate near zero, this implies a real interest rate of -4.0 percent. Suppose this raised investment in equipment and software by 10 percent from its current levels. That gets us 0.7 percentage points of GDP or roughly 1 million jobs.
Let’s imagine that it also pushed down the real value of the dollar by 5 percent relative to the currencies of our trading partners. This should be sufficient to lower imports and raise exports enough to buy us at least 1 percentage point of GDP off the trade deficit. That gets another 1.4 million jobs.
Finally, the higher inflation rate would increase home equity. If we had seen 4 percent annual inflation since the start of the downturn, and house prices rose in step with inflation, then they would be approximately 7 percent higher today. This implies another $1 trillion in housing wealth, which translates into roughly $60 billion a year (0.4 percent of GDP) in additional consumption. That gets us another 600,000 jobs.
Okay, so here’s the tally so far:
| Work sharing | –3 million jobs |
| Youth employment | –1.5 million jobs |
| Sub-total | –4.8 million jobs |
| Government employment | — 1 million jobs |
| Additional investment | — 1 million jobs |
| Lower trade deficit | –1.4 million jobs |
| Additional consumption | –0.6 million jobs |
| Sub-total | –4 million jobs |
| Induced jobs (25%) | –1 million jobs |
| Total | –9.8 million jobs |
The latter four categories are lumped together since we can expect a substantial multiplier effect on the spending from these jobs. (Work sharing redistributes wages and the youth employment jobs are relatively low-paying.) If we assume that the spending by these workers generates one additional job for every four jobs directly created, then we get one million jobs generated through this re-spending.
Summing up, we have 9.8 million jobs of our 10 million jobs. This is before we add in the green jobs that would be generated by my more generous tax credit for retrofitting homes, businesses and public buildings. This also doesn’t count my big infrastructure project which would get us things like high-speed rail in our lifetime. Nor does it include my grand scheme to have the government pick up the tab (@$70 billion a year) to give us free inner-city bus service.
The moral of the story is that this time it could have been different. We now return control of economic policy to President Obama, the Republican Congress, and Ben Bernanke.
