Presented by Charles I. Plosser, President and Chief Executive Officer
Federal Reserve Bank of Philadelphia
Capital Markets in the Post-Crisis Environment
Global Interdependence Center and the Bank of Finland
June 6, 2011
PDF version (83 KB, 6 pages)
Good morning. It is indeed a pleasure for me to be with you today. This is my first visit to Finland and it has proved to be more beautiful than I imagined. So I want to thank the Global Interdependence Center and the Bank of Finland for their hospitality and the opportunity to participate in this conference. It is also a pleasure to join my central bank colleagues from France, Sweden, and, of course, Finland at such an event.
The theme of this session, central bank policy after the crisis, is both timely and extremely important. The financial crisis and arguably the worst global recession of the post-war era have challenged policymakers around the world. While for most of the hardest hit countries, the recession trough has passed and a moderate recovery is underway, the actions by fiscal authorities and central banks, intended to mitigate the damage from the crisis, have left a legacy of governments with huge budget deficits and central banks with record-low short-term interest rates and balance sheets of unprecedented magnitudes.
Yet even though the policy responses by central banks were similar and extraordinarily accommodative, a few have already moved toward normalization. As we are likely to hear shortly from Stefan Ingves, the Riksbank has raised its policy rate to 1¾ percent from just 25 basis points since the middle of last year. This tightening of policy was accompanied by a reduction in its lending to the Swedish banking system, which reached more than 300 billion Swedish krona at the depth of the crisis. Other central banks, including the Federal Reserve and the Bank of England, face some daunting challenges as they try to unwind from their extraordinary actions and seek to restore confidence and credibility in their monetary policy frameworks. Today I would like to share with you my perspective on how I think about the normalization of monetary policy in the U.S. I should note the usual disclaimer applies in that I speak for myself and not necessarily for the Federal Reserve System or my colleagues on the Federal Open Market Committee (FOMC).
My thoughts about normalization are driven by some fundamental principles that I believe should guide the way central bankers approach policy. I believe that policymakers should do their utmost to be clear about their goals and objectives and then communicate how policy will be conducted to achieve those goals. Committing to a systematic, or rule-like, approach to policy decisions makes policy more predictable and will generally improve economic outcomes. Policy choices should, of course, be dependent on changes in economic conditions, but the nature of the response should be communicated to the public in advance, so that they understand how policy is likely to react to an evolving economy.
Thus, the first step in thinking about how central banks should plan their exit from unprecedented policies is determining the goal or where you want to go. You simply do not start out on a trip without some idea of where you are headed. Put another way, if we are to design an appropriate exit strategy, we must know what the policy framework will look like at the end of the journey. Before describing where I think we should be headed, let me review the current state of affairs. During the last three years, the Federal Reserve, like many other central banks around the world, has taken extraordinary actions to mitigate the impact of the financial crisis and support the return of economic growth. Our traditional instrument of monetary policy, the federal funds rate, has been near zero for nearly two and a half years. Since it is infeasible to reduce the targeted federal funds rate below zero, the Fed implemented additional policy accommodation through changes in its balance sheet. As a result, the Fed’s balance sheet has grown more than threefold, from nearly $900 billion before the crisis to about $2.7 trillion today, and its asset composition has shifted significantly from mostly short- to medium-term Treasuries to longer-term Treasuries, mortgage-backed securities (MBS), and agency debt.
I believe that as it normalizes policy, the Fed should be headed toward a system in which, first, the federal funds rate is re-established as the primary instrument of monetary policy; second, the Fed’s balance sheet is much smaller, probably less than $1 trillion, so that the federal funds rate trades above the interest rate paid on reserves; and third, the balance sheet’s composition returns to predominantly shorter-term U.S. Treasuries. For those who get excited about the minutiae of central bank operating procedures — and I know there are some of you in the audience — this system is referred to as a corridor system and has been used by a number of central banks around the world, including the Riksbank and the ECB.
For the U.S., this decision of which system we choose is probably more of an issue than for many central banks because prior to the crisis, the Federal Reserve Board was not permitted to pay interest on reserves and now it is. The corridor system, which I prefer, calls for monetary policy to set the federal funds rate at a level within a corridor between the interest rate paid on reserves and the rate at which the Fed is willing to lend to banks through the discount window — or lending rate. For the policy rate to lie above the rate paid on reserves, the supply of reserves must be limited. If the supply of reserves is very large — as is currently the case — banks have no incentive to economize on reserves and the federal funds rate would equal the interest rate on reserves. As a consequence, the interest rate paid on reserves becomes the de facto policy rate and the fed funds rate becomes irrelevant. This environment is sometimes called a floor system.
I believe the choice of an operating framework is important. While we would be able to conduct monetary policy by simply setting the interest rate we pay on reserves and allowing the federal funds market to become irrelevant, the floor system is very troubling to me. In a floor system, the size of our balance sheet would become irrelevant. Thus, one could envision using it as an independent instrument of policy, perhaps increasing it or decreasing it as needed to provide emergency liquidity to the banking system. The problem is that the floor system offers no restrictions on how the balance sheet might be used. If our operating framework divorces our balance-sheet decisions from monetary policy, it becomes a tempting instrument for future policymakers inside or outside the central bank to use it for non-monetary-policy purposes. This could jeopardize the independence of the central bank and, if abused, would be a source of many unforeseen problems. For example, using the Fed’s balance sheet to conduct fiscal policy-like actions could prove detrimental to long-run economic stability. So I believe the corridor system, with its constraints on the balance sheet, is more likely to preserve central bank independence and limit the temptation to use the Fed’s balance sheet as an instrument of fiscal policy.
Once we know where we want to go, we can begin to map out a path that will take us there. By articulating a systematic plan that gets us to our objective, we improve communication with the public, reduce uncertainty in the marketplace, and lend credibility to the commitment that policymakers will follow through.
Any exit plan that is to lead us to a corridor-like operating system has to describe how the fed funds rate or interest rate on reserves will be increased and how the balance sheet will be reduced. There are many ways to do so, but in every case, we must remember that monetary policy decisions should be contingent on the evolution of the economy.
In late March 2011, I outlined a proposal for a systematic, rule-based approach that would involve the Fed’s selling assets from its portfolio as it increased its policy rate, with the pace of sales dependent on the state of the economy. The plan would get us back to a normal operating environment in a timely manner, with the Fed’s balance sheet reduced to a size that would again allow the federal funds rate to be the primary policy instrument.
I also suggested that the first step of the normalization process would be to increase the interest rate paid on reserves and the targeted federal funds rate away from the zero bound, perhaps to 50 basis points. The Federal Open Market Committee would also announce its intention to stop re-investing the proceeds from maturing agency and Treasury securities. These decisions would initiate the normalization process and begin to raise the funds rate and shrink the balance sheet from the start. How the process evolves from that point will depend on the plan and the state of the economy.
The Committee would also announce that at subsequent meetings it would regularly evaluate the policy rate based on economic conditions, as it always does. Sales of assets between meetings beyond those maturing would also be tied to economic conditions, just as policy rate decisions are. The Fed would announce a pace of sales, which would increase when the Committee chose to increase the policy rate. Since the amount of securities to be sold between meetings would be determined in advance, the public and markets would see the Fed shrinking its balance sheet in a predictable manner. For example, the Committee might set a sales pace of $20 billion of assets between FOMC meetings and announce that for each 25 basis points the funds rate target is raised, it would increase sales by, say, $125 billion of assets during the intermeeting period, which is about the pace at which we bought securities. If the policy rate was not raised there would be no additional sales. There is logic to selling assets faster when the policy rate rises. If the economy is growing then market demand for both risk and duration is also likely to expand. So selling assets is likely to be less disruptive. Note that asset sales would not be a separate policy decision but driven by the same monetary policy decision process used to change the policy rate. This plan is consistent with the Fed’s dual mandate of price stability and maximum employment growth. It recognizes that our policy actions will be dependent on the evolution of the economy and ties the shrinking of the balance sheet explicitly to those actions.
Of course, other plans could be adopted. For example, an even simpler alternative plan might be to announce in advance that we will shrink the balance by some fixed dollar quantity between each FOMC meeting or each month. That pace would not change unless something extraordinary occurred. Such a plan, in essence, puts the shrinking of the balance sheet on a predetermined glide path, which would allow the markets to predict how the Fed’s balance sheet and asset sales will evolve. The major difference between this plan and the first one I described is that asset sales would not be conditioned on the state of the economy but would proceed at a steady pace. As a result, the interest rate would become the primary policy instrument and vary with economic conditions. To make such a plan credible, I believe that the Committee would have to state clearly that the hurdle for deviating from the planned pace of sales would be a high one. Otherwise, the temptation to manipulate the pace of sales based on some unspecified criteria might arise. Indeed, the major thrust of either plan is to avoid just such a temptation. A good plan will provide clear, reliable guidance on how asset sales will proceed and the conditions under which they may or may not change.
Either of these plans represents a viable strategy for normalization. One makes asset sales state contingent in the same way that policy rate decisions are state contingent. The other puts sales on a predetermined path while the federal funds rate becomes the main instrument for implementing policy. Both are plausible plans, and it is not clear that one strategy dominates the other.
Perhaps more important than the details of any exit plan is the very establishment of a systematic plan itself — one that can be clearly communicated to the markets and the public in a way that reduces uncertainty. A systematic plan will help define for the public and the markets not only where we are headed but also how we will get there. Of course, monetary policy will always be dependent on the evolution of the economy, but the more we can articulate how we will conduct policy during the transition to normalization, the more we will reduce uncertainty and contribute to stability. Announcing an explicit plan in advance would also enhance the credibility of our commitment to normalization. If, for some reason, the Committee chose to deviate from the plan, it would be forced to explain its reasons to the public.
This brings me to one final but important consideration for the Fed’s normalization strategy. Unlike most of the major central banks around the world, including the ECB, the Bank of England, and the Riksbank, the Federal Reserve has not clearly announced and committed itself to an inflation target. Some form of inflation targeting is widely accepted as a best practice by the majority of central bankers and monetary economists around the globe. Given the extraordinary amount of liquidity present in the U.S. banking system, it is reasonable for the public to be concerned about the prospects for inflation down the road. If the public’s expectation of inflation continues to drift upward, whether it is caused by higher commodity prices, concerns over fiscal deficits, or the Fed’s large balance sheet, we could find our credibility under pressure.
It is noteworthy that just within the last year or so the public discussion has swung from concerns that the U.S. might face a period of sustained deflation to concerns that monetary policy and rising commodity prices could produce higher-than-desired inflation. It is troubling that the public’s inflation concerns can be so volatile. It suggests that there may be less confidence in the credibility of the Fed’s commitment to price stability than we might desire. Yet, if we are to exit from this period of extreme monetary accommodation in a way that neither leads to higher inflation nor risks undermining the recovery, the public’s confidence in the Fed’s commitment to price stability will be crucial. For this reason, it would be highly desirable for the Federal Reserve to publicly commit to a clear and explicit inflation objective.
The actions taken in response to the economic and financial crisis have left the Federal Reserve and a number of other central banks far from what most central bankers would consider a normal operating environment. One of the challenges policymakers must face is how to commit to a plan for normalizing the operating framework for monetary policy. There is no one right plan to achieve this outcome. However, it is very important that we think carefully about that process and seek ways to credibly commit to exit from the unprecedented policies.
I have tried to outline some of the features that I think will be important for the Federal Reserve as it seeks to unwind its large balance sheet and establishes a new operating framework for monetary policy. I believe we can be most successful if we communicate a systematic plan that describes where we are headed and how we will get there. Moreover, such a plan would be strengthened if the FOMC adopted an explicit numerical objective for inflation, which would help ensure that inflation expectations remain well anchored.