September 30, 2009

Central Bank Exit Policies

Vice Chairman Donald L. Kohn

At the Cato Institute's Shadow Open Market Committee Meeting, Washington, D.C.

I am pleased to be on this panel on exiting from the unusual policies the Federal Reserve and other central banks have put in place to ameliorate the effects of the financial turmoil of the past two years. Chairman Bernanke has made a concerted effort to explain the thinking of the Federal Reserve in this regard, because it is so important that the public understand we have the means to meet our objectives of fostering stable prices and high employment. I will briefly underline some aspects of the Federal Reserve's framework for exiting that I believe to be especially critical to that understanding.1

Conditions for Exit
In its most important aspects, the decision about when to begin exiting from the unusual policies is not materially different from any decision to start tightening monetary policy. We will need to begin to remove the extraordinary degree of accommodation in its various dimensions when we judge that exiting from the current stance of policy will be necessary to preserve price stability as the economy returns to higher levels of resource utilization. Because it takes people time to adjust their spending and pricing decisions in response to a change in interest rates or other aspects of financial conditions, like other monetary policy decisions, that judgment will need to be based on a forecast of economic developments, not on current conditions. So we must begin to withdraw accommodation well before aggregate spending threatens to press against potential supply, and well before inflation as well as inflation expectations rise above levels consistent with price stability.

I cannot give you a small list of variables that will trigger an exit; as always, our forecasts will use all available sources of information. And I can't predict how rapidly we will have to raise short-term interest rates from around zero or remove other forms of accommodation; that too depends on how the economy seems to be recovering and the outlook for inflation. Clearly, the present degree of accommodation--as gauged by nominal and real short-term interest rates and the size of our balance sheet--is extraordinary, and we will have to take account of how that is influencing spending and inflation expectations when deciding when and how fast to tighten.

Tools for Exit
We have the framework to exit from these policies when we need to do so. And the tools at our disposal will allow us to do so at the pace and in the sequence we judge will best meet our objectives.

Most importantly, our ability to pay interest on reserves will enable us to raise short-term interest rates even while the quantity of assets we hold is still quite elevated and while the reserve base of the banking system is extraordinarily high. The opportunity for banks to earn interest on a highly liquid risk-free deposit at the Federal Reserve should put a reasonably firm floor under short-term rates, including the federal funds rate. To date, that floor has been somewhat soft, perhaps because not all participants in the federal funds market can hold deposits at the Federal Reserve, and because banks have been reluctant to allocate the needed capital to arbitrage a few basis points. But I am confident that when we begin to raise our deposit rate, it will put upward pressure on the rates on competing assets, increasing actual and expected short-term interest rates with the usual types of effects on other interest rates and asset prices. As banks become more comfortable with their capital levels, they will be more willing to undertake the arbitrage to tighten the link between the rate on deposits and short-term market interest rates.

Still, draining reserves at some point also will be an aspect of exiting. The large volume of reserves is contributing to the loose relationship of our deposit rate and market rates. In addition, although to date the high volume of reserves evidently has not increased bank lending or reduced spreads of rates on bank loans or other assets relative to, say, Treasury rates, it could begin to do so if banks start to perceive the risk-adjusted returns on loans as superior to our deposit rate. An increase in lending and narrowing of spreads on bank loans is a necessary and desirable aspect of the return to better-functioning markets and intermediation to promote economic growth. But spreads eventually could become narrower than what would be consistent with underlying risk, and lending could grow more quickly than appropriate for price stability if very high levels of reserves remain in place. We are developing new techniques for draining reserves, including reverse repurchase agreements against mortgage-backed securities and time deposits for banks at the Federal Reserve. And, of course, we retain the option to sell securities from our portfolio on an outright basis. The range of tools will permit us to drain large volumes of reserves if necessary to achieve the policy stance that fosters our macroeconomic objectives.

Our lending programs were designed to wind themselves down as market conditions improve, and they are doing so. When conditions are no longer unusual and exigent, those programs not focused on depository institutions will be terminated, and, with a few exceptions such as the Term Asset-Backed Securities Loan Facility, they will leave no residual on our balance sheet.2

The long-term securities we are buying will not run off so rapidly. But the effects of our holdings of mortgage-backed and agency securities on spreads nonetheless should decline even if they remain on our balance sheet. For one, some of the spread compression may result from the flow of our purchases, as well as our stock of holdings, and as we already announced, we will be tapering down our purchases. Moreover, the stock of assets we own will become an ever smaller share of a growing market. Finally, as confidence returns, asset demands will become less focused on particular classes of highly safe and liquid assets and more sensitive to relative interest rates, and private participants will arbitrage away at least some of any remaining spread distortions. Nonetheless, if, in the course of removing accommodation, the Federal Open Market Committee (FOMC) perceives spreads to be distorted or longer-term interest rates to be not responding appropriately, it could consider sales of these assets.

Communication about Exit
The unusual nature of our actions and the uncertainty about when and how they will be unwound suggest an even greater payoff than usual from being as clear as possible in our communications with the public. I already noted the importance of the public's understanding that we can and will exit from these policies when that is necessary to achieve our objectives for stable prices and maximum employment. In addition, we will need to explain especially carefully--in our policy announcements, the minutes of our meetings, and our quarterly forecasts--the evolution of our assessment of the economic situation and the risks associated with achieving our goals. Finally, we will need to be sure our rate guidance evolves along with our assessment of the probability that the exit is drawing closer.

Conclusion
Although economic conditions have apparently begun to improve--partly in response to the extraordinary steps the Federal Reserve and other authorities have taken--resource utilization is quite low, inflation is subdued, and continuing restraints on credit are likely to constrain the speed of recovery. For that reason, as the FOMC stated last week, exceptionally low interest rates are likely to be warranted for an extended period. Given the highly unusual economic and financial circumstances, judging when the time is appropriate to remove policy accommodation, and then calibrating that removal, will be challenging. Still, we need to be ready to take the necessary actions when the time comes, and we will be.


Footnotes

1. The views expressed are my own and not necessarily those of my colleagues on the Federal Open Market Committee. Return to text

2. Lending in support of the orderly resolution of individual systemically important institutions will run down more slowly. The Administration has agreed to seek to remove the so-called Maiden Lane facilities from the Federal Reserve's balance sheet.Return to text

Last Update: September 30, 2009