Laura Tyson and I gave a presentation to the IMF Fiscal Forum on Wednesday the 17th. A draft of our paper. I talked about changing perceptions of fiscal policy. Laura talked about changing perceptions of multipliers.
Here is my part of our talk:
For the past five years, each year I have begun talks by saying that I thought in the previous year that the crisis was at its peak, and that each year I turned out to be wrong, as the crisis revealed itself as even deeper and more of a crisis than it had been the year before.
We are now six years in. The natural thing to do, therefore, is to compare 1929-1935 with 2007-2013. 1929-1933 saw much larger output declines and unemployment increases than 2007-2011--between two and three times as large. But by 1935 the tide had turned nearly everywhere except France, which was still on the gold standard (but about to elect a Popular Front government and drop its link). There had been substantial gap closing between 1931 or 1933 and 1935. The Great Depression was clearly not over by 1935, but the global economy was clearly on the mend.
By contrast, today we have had a slower decline. But we have seen no signs of anything we could call gap-closing, no sign of any returns back to what we used to think of as normal levels of activity. Here at the IMF we speak of a "three speed recovery". In the North Atlantic plus Japan, at least, the speeds are "reverse" in Europe--the proportion of the population at work is dropping--and "neutral" in America--the proportion of the population is stable. Thus it is no longer clear that, when this is over, people will look back and classify it as a smaller worldwide event than 1929-1939. Another five years of the last two years' trends, and it will be 2007- rather than 1929-1939 that will seize the name of "The Great Depression".
Only the third speed, the emerging market speed, is driving the global economy forward. That is the first bright spot, and it is a very bright spot. In the 1930s the emerging markets shared the Great Depression. Today they do not share in anything like the North Atlantic's degree of macroeconomic distress. That is indeed a blessing.
There is a second bright spot--and this second is perhaps the most important bright spot of all. As of 1935 we were down to 14 democracies in the world, and we were about to lose Spain, which was about to join Italy and Germany among the fascist powers. The fact that political democracy has remained so strong and so powerful and robust in the face of the macroeconomic shocks of the past six years is a very bright spot indeed. Perhaps it is the facet of the situation that most justify our being able to sleep at night.
There is a third bright spot. The third bright spot is that we are here, sitting here in the house that Harry Dexter White and John Maynard Keynes built. The world had no counterpart in the 1930s. It had no talk shop in which to try to figure out what to do, and there is indeed a great deal of figuring out what to do that needs to be done. To quote Keynes's critique of Leon Trotsky, who:
assumes that the moral and intellectual problems of the transformation of Society have been already solved--that a plan exists, and that nothing remains except to put it into operation…. [But w]e lack more than usual a coherent scheme of progress, a tangible ideal. All the political parties alike have their origins in past ideas and not in new ideas…. It is not necessary to debate the subtleties of what justifies a man in promoting his gospel by force; for no one has a gospel. The next move is with the head…
For us, too, the next move must be with the head--as opposed to putting a plan into action. We need to evolve a plan, check the plan, and make sure it’s the right plan. And we have this organization to serve as the medium-term informal planning department of the human race. In the 1930s we did not. So we should be grateful, and turn to our task.
As your chief economist Olivier Blanchard said a week and a half ago, a great deal of humility is in order. Virtually none of us back in 2007 thought we would be anywhere like where we are today.
So let me be humble.
Start with what we thought about fiscal policy back in 2007. Back then, there was near-consensus agreement with John Taylor’s (2000) position that fiscal policy should be set on classical cost-benefit terms, with aggregate demand management the exclusive or the near-exclusive province of central banks.
There were five reasons for this. The first was the problem of legislative confusion. Given the demands on legislators' time in our complicated world, legislatures that are asked both to worry about long-run balance and about the management of the business cycle are more likely to do both jobs badly than those that just focus on long-run balance and the proper size of government. The second was the problem of legislative process. Legislatures are designed to be inertial and act slowly for good reasons. They cannot turn on a dime. Markets do move--you can think that the long run is a decade off one day, and the next day find that asset prices have shifted, the long run has shown up at the door, and started unpacking its luggage. And central banks can turn on a dime.
The third reason was the problem of implementation. Central banks by changing asset prices induce every private-sector agent in the economy to start calculating how to change their behavior. You thus get a rapid and decentralized implementation of a policy shift. By contrast, a government’s abilities to start and stop infrastructure projects quickly and on a large scale are much, much limited. Fourth comes the problem of rent seeking. Whenever you have large amounts of money sloshing around a government, you will have lobbyists trying to direct government policy to their clients. They will entrench themselves. Your successors will then have a very difficult time trying to undo what was done in a hurry. Keeping the legislature out of the aggregate-demand management business is a way to reduce those "act in haste, regret at leisure" difficulties against which all governments and international organizations in the world struggle.
And fifth, of course, there was the believed-to-be fact of superfluity. Fiscal policy was simply not needed. Monetary policy was thought to be strong enough to do the job. There were some worries about what would happen if economies ever hit the zero nominal lower bound, but it was expected that any time spent at that limit would be short. And, once you have space for conventional monetary policy, why not use it--rather than opening the fiscal policy cans of worms?
Today we think differently. Today we find central bankers--like Federal Reserve chair Ben Bernanke--asking for help from fiscal authorities, and saying they can’t do it alone.
What caused the shift in thinking?
First and most important, the superfluity point turned out to be wrong. Standard monetary policies have proven inadequate to the task of restoring and stabilizing aggregate demand anywhere near potential capacity. Unconventional monetary policies are unproven, and perhaps risky. Moreover, the required scale of them turns out to be orders of magnitude higher than people who were optimistic about monetary policy envisioned even half a decade ago thought.
Furthermore, there is an increasing body of evidence that when there is excess capacity and when short term interest rates are at or near their zero lower bound, the effectiveness of discretionary fiscal policies as measured by Keynesian multipliers is significantly larger than in other times, or than had been expected. So the big question that remains is: What are the limits on the appropriate use of discretionary fiscal policy in order to rebalance aggregate demand in situations like this? That is where we need the hard thought. That is where we must move with our heads.
As I see it, since 2007, not a near consensus but, rather, the centroid of an increasing divergence of economists has evolved away from the Taylor (2000) belief that fiscal policy was an unneeded extra wheel in six stages.
On the first stage there was fairly general agreement. When the zero lower bound was reached in 2008-9, that in combination with worries about nonstandard monetary policies led to a belief that the fiscal policy toolbox should be opened up and used in coordination with experiments on nonstandard monetary policies to try to quickly rebalance the economy.
As 2009 turned into 2010, develeraging and price adjustment seemed to be slower than expected. What has supposed to have been a very short fiscal policy impetus became a longer one, or at least there was discussion about whether it should be longer. "No 1937s" was the catchword in the United States[[no repetitions of the premature turn to fiscal austerity in the US in the 1930s. This second stage, however, was accompanied in 2009-10 by concerns that discretionary fiscal policies could lead to higher long-term interest rates and widening interest rate spreads. At the start of 2010 these were concerns. By the middle of 2010 these were realities for peripheral European countries. Runs on government bond markets and sharp increases in interest rate spreads validated these concerns--at least for sovereigns that were not able to control their own money stocks, and for sovereigns that worried about maintaining exchange rates near normal levels.
People then tried to get their minds around the fact that macroeconomics for the sovereigns that were thought of as “credit worthy”--Germany, France for most of the time Britain, the United States, and Japan--seemed to be very different than macroeconomics for other countries. As 2010 turned into 2011, concerns that delaying fiscal consolidation in even the most credit worthy could raise risks of inflation, financial repression and/or slower potential growth came to the for, highlighted by the eloquent triple of Carmen Reinhart, Vince Reinhart, and Kenneth Rogoff, who argued that we were venturing into truly unknown peacetime debt territory, and that the historical record showed signs of a slow-growth high-debt correlation--although it did not provide clear and convincing evidence of a strong causal mechanism running from the second to the first.
On the other hand, as 2011 proceeded to 2012, there was a recognition that the collapse of risk tolerance in the private sector seemed likely to persist. Private agents seemed to have an enormous demand for safe nominal assets--things rated triple AAA. The disappearance of $8T in private sector assets that had been rated AAA seemed to expose a safe asset shortage. That shortage might be playing the same role in depressing economic activity that liquidity shortage had played in previous business cycles. That provided an extra argument for the government to borrow--at least to the extent that the government could borrow and create more of the AAA assets that the private market wants to hold, without cracking its own status as a safe debtor. What will crack that status? Who can know?
The last of the six stages was the recognition, in large part pushed by work done in this building, that indeed Keynesian multipliers do appear significantly larger than the numbers all of us had kept in the back of our minds during the first four years of the crisis.
- J. Bradford DeLong and Laura d'A. Tyson (2013): Discretionary Fiscal Policy as a Stabilization Policy Tool: What Do We Think Now That We Did Not Think in 2007?