By Larry Summers
Coming into Jackson Hole, economists are grappling with a major issue: Can central banking as we know it be the primary tool of macroeconomic stabilization in the industrial world over the next decade? In my forthcoming paper with Anna Stansbury, we argue that this is in doubt.
There is little room for interest rate cuts. In every US recession since the 1970s, the fed funds rate was cut >500bps. In most, the real rate fell >400bps below the neutral rate. Now, the maximum feasible cut is 200-300bps, bringing the real rate only 150-250bps below neutral.
This limited space for interest rate cuts is true of the US, which has the highest interest rates in the industrialized world. It is even more true of Europe and Japan.
QE and forward guidance have been tried on a substantial scale. We are living in a post QE and forward guidance world. It is hard to believe that changing adverbs here and there or altering the timing of press conferences or the mode of presenting projections is consequential.
We usually agree with Janet Yellen, but believed at the time of her 2016 Jackson Hole speech -and believe even more in today’s world of 150bp 10yr rates- that her optimism about the existing monetary policy toolkit is misplaced.
Black hole monetary economics – interest rates stuck at zero with no real prospect of escape – is now the confident market expectation in Europe and Japan, with essentially zero or negative yields over a generation. The United States is only one recession away from joining them.
Everywhere in the industrial world, the risks of a sharp upturn in unemployment appear greater than the risks of a sharp upturn in inflation (even though market expectations of inflation are clearly below 2 percent targets).
The one thing that was taught as axiomatic to economics students around the world was that monetary authorities could over the long term create as much inflation as they wanted through monetary policy. This proposition is now very much in doubt.
Many believe that events proved Alvin Hansen wrong about secular stagnation. On the contrary, the fact that it took WW2 to lift the world out of depression proves his point. Absent the military buildup, a liquidity trap deflation scenario would likely have persisted.
Call it the black hole problem, secular stagnation, or Japanification, this set of issues should be what central banks are worrying about.
We have come to agree with the point long stressed by Post Keynesian economists & recently emphasized by Palley that the role of specific frictions in economic fluctuations should be de-emphasized relative to a more fundamental lack of aggregate demand.
In our forthcoming paper, we argue that it minimizes our predicament to see it – as is current consensus – simply in terms of a falling neutral rate, low inflation, and the effective lower bound on nominal rates. Secular stagnation is a more profound issue.
Limited nominal GDP growth in the face of very low interest rates has been interpreted as evidence simply that the neutral rate has fallen substantially. There may well be more to it than that.
We believe it is at least equally plausible that the impact of interest rates on aggregate demand has declined sharply, and that the marginal impact falls off as rates fall.
It is even plausible that in some cases interest rate cuts may reduce aggregate demand: because of target saving behavior, reversal rate effects on financial intermediaries, option effects on irreversible investment and the arithmetic effect of lower rates on government deficits.
This can be illustrated using a textbook macroeconomic diagram with a very steep, non-linear or even backward-bending IS curve.
If the central problem for macroeconomic stabilization is a falling neutral real interest rate – what might be called “old new Keynesian” economics – monetary policy can achieve full employment if it can get the interest rate low enough.
In contrast under the secular stagnation view we have outlined – what might be called “new old Keynesian” economics – interest rate cuts, even if feasible, may be at best only weakly effective at stimulating aggregate demand and at worst counterproductive.
There is the further point that reducing interest rates may degrade future economic performance for any of the following reasons.
First, financial instability. The financial crisis had roots in bubbles and excessive leverage caused by efforts to maintain demand after the 2001 recession. Japan’s late 1980s bubble had roots in a low interest rate tight fiscal environment after the 1987 stock market crash.
Second, risks of zombification of firms. Firms that do not face debt service payments are like students who do not have to take tests. They can drift along complacently and ultimately unsuccessfully. And low rates may contribute to increased monopoly power and reduced dynamism.
Third, risks of bank failures. Low rates crowd bank profits and franchise value, making them more vulnerable to adverse shocks at any given level of regulatory capital.
Fourth, risks of further reducing monetary policy effectiveness. To the extent to which rate cuts now “borrow” demand from the future as firms and consumers bring forward investment and durable purchases, low rates now may imply less effective monetary policy in the future.
The right issue for macroeconomists to be focused on is assuring adequate aggregate demand. We believe it is dangerous for central bankers to suggest that they have this challenge under control – or that with their current toolkit they will be able to get it under control.
Obviously fiscal policy needs to be a major focus, especially given what low or negative interest rates mean for the sustainability of deficits.
The high inflation and high interest rates of the 1970s generated a revolution in macroeconomic thinking, policy and institutions. The low inflation, low interest rates and stagnation of the last decade has been longer and more serious and deserves at least an equal response.
We hope this will come out of Jackson Hole’s focus on “Challenges for Monetary Policy” – but we are not holding our breath.