Are central banks running out of ammunition?
January 21, 2019
In his recent Presidential Lecture to the annual meeting of the American Economic Association, Ben Bernanke has made the case that the Federal Reserve is not short of ammunitions to face a serious slowdown or even a recession. He was more cautious in drawing conclusions for the ECB and the Bank of Japan. His forceful statement is one more addition to the messages that central bankers send these days. Are they fully convincing and reassuring?
Bernanke brings together a substantial amount of research to conclude that QE and Forward Guidance have clear, lasting and powerful effects on long-term interest rates. The literature is indeed reasonably large and increasingly sophisticated. The standard conclusion is that in the US, the so-called nonstandard policy instruments have pushed down long-term interest rates by about 1% (100 basis points in the finance jargon). This is sizeable and it allows Bernanke to conclude that when the short-term interest rate is very low, with little room for lowering it much further, central banks can still achieve their aims with their novel instruments.
At the same time, a number of economists, both from academia and financial institutions, have recently argued that this reassuring conclusion is not warranted. A first reason is that the literature remains somewhat controversial. There are doubts about the ways these effects are measured and about the precision of the estimated results. The 1% effect is not to be taken literally, it could be more, or less, and it could be zero. A second issue is whether increasing use of the instruments do not face declining returns; could they be initially powerful and gradually less so? Third, QE involves flooding the financial markets with cash, which could have adverse side-effects. The initial view that it could lead to runaway inflation has been proven wrong for one simple reason: most of the cash stays with banks, it is not lent out. Plausible side-effects include encouragement to risk taking in the search for higher returns and increases in stock prices that do not reflect higher productivity but simply discounting with low interest rates. It could be a desirable effect from the point of view of monetary policy effectiveness, but it raises the specter of a misuse of savings to finance investments in low-value projects that will reduce growth for years to come.
Finally, reducing long-term interest rates, possibly into negative territory is one thing, affecting the economy is another. Did the new instruments raise growth, reduce unemployment and bring inflation up to its target rate? Here again, recent research tends to detect an impact, but it is controversial and the results are highly imprecise. The main challenge is that many other things affect the economy, which requires disentangling them from monetary policy actions. Even if the correct answer is “maybe, somewhat”, that is not good enough to be fully reassured.
It does not take a sophisticated analysis to make the following observation. In unprecedented moves, over many years the balance sheets of many central banks have been multiplied by three, four or more, and have not declined since, or only partly so as in the US. In spite of the new paraphernalia, growth has been subdued in most developed economies since the great financial crisis and inflation has remained below target nearly all the time (the US performance has been a bit better, partly because fiscal policy has been expansionary).
The same applies to negative nominal interest rates, those that are quoted. Economists understand that zero is not a magic number and that real interest rates, which are what matters for policy effects, are often negative. In this view, concern about negative nominal rates is the result of an illusion. The public at large, however, fail to understand that it is odd to pay people to convince them to borrow.
As with QE, the evidence in favor of the effectiveness of negative interest rates is not firmly established. When cash is plentiful, and not borrowed in proportion with its abundance, it is unclear why necessarily limited forays in negative territory would succeed in significantly propping credit up. It is also surprising that the financial markets expect long-term interest rates to remain negative for well over a decade in Japan and the Eurozone. This is a bit disorienting. Households and firms, already deemed to be subject to illusion, may be worried enough to save more and borrow less than otherwise.
Forward guidance is yet another part of the nonstandard policies. It implies for central banks to make commitments – on the interest rate and on QE – for periods of one year, often quite longer in order to convince the financial markets of their intentions. Here the evidence is less controversial: commitments move financial markets. Whether the market reactions translate into sizeable effects is another question.
None of this means that the adoption of nonstandard policies has been a mistake. That the global financial crisis did not lead to a much-feared repeat of the Great Depression is enough to justify nonstandard policies. It does not mean, though, that monetary policy will be potent enough to deal with cyclical downswings when the short-term rate is low to start with. When the effective lower bound of the interest rate is reached – be it zero or below – nonstandard policies must become the only game in town. Yet, it seems improbable that they will increase their balance sheets four times over again after having being unable to bring them down, or that they will explore much more negative interest rates. Central banks are said to consider more tools, an implicit recognition that they are not as reassured as they claim.
This leaves us with the worrying impression that central banks are stuck in a very unpleasant corner. And so are all of us, because central banks have been successfully conducting counter-cyclical policies for a few decades by now. Governments are called upon, including by central banks, to take the burden of carrying out expansionary fiscal policies. Unfortunately, there are many good reasons why this task has been delegated to central banks. Governments are typically slow to react, fiscal policy is intensely political, and there is a tendency to open deficits when needed only to not closing them when warranted. Some progress is possible by making counter-cyclical fiscal policies automatic, beyond the weak automatic stabilizers, along with really binding debt constraints. Governments, however, strenuously resist being subject to automaticity and to binding constraints.