Thank you, Mr. Chairman. This is an extremely important discussion, and I am glad that you have arranged a special session. I very much appreciate the comprehensive and outstanding memos from the staff. At our October meeting, we agreed to take whatever steps were necessary to support the recovery of the economy, and that principle guides my thinking on monetary policy at the zero bound. In most circumstances, I see few advantages to gradualism, and certainly whenever we approach the zero bound, I think the funds rate target should be quickly reduced toward zero.
As to the level of the lower bound, my default position is that we should move the target funds rate all the way to zero because that would provide the most macroeconomic stimulus. For example, every 25 basis point cut in the target typically takes about 25 basis points off the prime rate and associated borrowing rates. The institutional concerns about Treasury fails and Treasury-only money market funds merit consideration, but I don’t consider them serious enough to ban lowering the target to a very low level. Still, the surprising results we obtained from paying interest on reserves should make us chary about predicting the reactions of financial markets to new circumstances, so there may be some benefits from allowing the funds rate to trade between zero and, say, 25 basis points.
Let me now turn to the third set of questions on communication strategies. Looking ahead, I believe that there could be significant benefits to communicating effectively the FOMC’s intentions to hold the target funds rate at a very low level. The Japanese experience at the zero bound suggests that this is one channel that can work, and the evidence suggests that our own guidance that began in 2003 similarly influenced longer-term rates. We learned then, though, that it is hard to convey the conditionality of such intentions and the multiple influences on the optimal setting of the funds rate. Still, I favor trying to include forward-looking communication on policy expectations in future FOMC statements.
We could also consider using the FOMC minutes to provide quantitative information on our expectations. For example, we could reveal the funds rate projections that implicitly accompany our quarterly economic projections, publishing ranges and central tendencies of the federal funds rate along with those for GDP growth, unemployment, and inflation. The advantage of this approach is that it would provide a clear future path to the funds rate that is conditional on the economic environment pertaining to output and inflation relative to our goals. We did discuss this approach before, and I remember that a number of you were uncomfortable with it. But circumstances have changed, and there could be particular value now in adding the federal funds rate to our projections, and in fact, we could consider a trial run.
I believe that, in addition to providing guidance on the likely path of future interest rates, we should become more communicative about our longer-term inflation objective to avoid a decline in inflation expectations as inflation drops over the next few years below desirable levels. One way to accomplish this is to include quantitative information on our longer-term projections in the Summary of Economic Projections (SEP), as the Subcommittee on Communications has recommended. We could go even further to endorse a Committee-wide long-term inflation objective, although that is something that we would have to further consider carefully. We could supplement including longer-term projections in the SEP with language in the FOMC statement that is akin to that used in 2003, when the Committee referred to an unwelcome decline in inflation. Alternative B does take a step in that direction.
The bracketed language in alternative A goes further by specifying a medium-term Committee inflation target. This is a big step, and one that deserves thorough debate. There are theoretical papers demonstrating the potential benefits in a liquidity trap of committing to an inflation rate after the economy recovers that is higher than we would actually want ex post because raising inflation expectations lowers real rates, thereby stimulating the economy. In theory, by committing to more inflation than we actually want later on, we could generate extra stimulus now. But this strategy requires a strong commitment device because the Committee will have an incentive to renege later on when the economy has recovered. I do understand the attractions of such a strategy in theory, but I am not at all convinced that the benefits would exceed the costs in practice. It would be enormously difficult to explain and could harm the Fed’s overall credibility as an institution. Moreover, it is not only real rates but also nominal rates that influence housing demand, and any increase in longer-term nominal rates triggered by higher inflation expectations could adversely affect this key sector.
Let me now turn to the nonstandard policy tools that the FOMC and the Board have authorized. I wholeheartedly support the many actions that have been taken to increase liquidity in the financial system, as well as those designed to increase credit availability and lower borrowing costs. Going forward, I support both the purchase of agency debt and MBS by the System Open Market Account and purchases of long-term Treasury debt. Both existing evidence and the market response we just saw to the recent announcements and comments concerning such programs suggest to me that such purchases can push longer-term borrowing rates down. Other new programs—for example, to improve credit market functioning in A2/P2 commercial paper and in commercial and private-label residential-mortgage-backed securities—are well worthy of consideration. Naturally, the potential benefits and costs of each new facility or program need to be assessed before adoption. Formulating the guidance from the FOMC to the Desk regarding how these new programs should be described remains a challenge. I think a possible formulation could have the FOMC setting some objectives for levels or movements in Treasury yields or MBS spreads, but those open up thorny issues, and I think that this is something we really have to study further.
With respect to the FOMC’s operating regime going forward, I oppose switching from a regime based on targeting of the fed funds rate to one based on a quantitative target for the monetary base, excess reserves, or the overall size of our balance sheet. The Board or the FOMC or both, in my view, should consider the merits of each program on its own, without any presumption of PAYGO. Most of you probably recall that PAYGO was a budget device employed by the Congress to constrain the federal deficit to a target level. In our case, an overarching decision by the FOMC about the size of our balance sheet or the monetary base would force tradeoffs among our various programs to hit that total, similar to PAYGO.
Imagine, however, that the commercial paper market were to revive, allowing us to terminate the CPFF. Excess reserves would decline, but that decline would have no negative effect on economic activity, so there should be no presumption that some other program should be expanded to restore the monetary base to its previous level. Theory suggests that when the monetary base is increased by purchasing conventional SOMA assets, its expansion should have little or no effect on the behavior of banks or asset prices more generally after the zero bound has been reached. Abstracting from expectational effects, the evidence generally supports this view. While the quantity of money is surely linked to the price level in the very long run, most evidence suggests that variations in the base have only insignificant economic effects in the short or medium term under liquidity trap conditions. This makes the base an inappropriate operating instrument for monetary policy in a zero bound regime. As Japan found during its quantitative easing program, increasing the size of the monetary base above levels needed to provide ample liquidity to the banking system had no discernible economic effects aside from those associated with communicating the Bank of Japan’s commitment to the zero interest rate policy. I think my views on this mirror those that you expressed in your opening comments, Mr. Chairman.
With respect to the directive, the version proposed in the current Bluebook specifies the types and amounts of mortgage-related assets that the Desk should buy and the objective of these purchases—namely, to boost activity in the mortgage and housing markets. Language of this type is consistent with the policy approach I support, in which each credit facility program and asset purchase decision is judged on its own merits, according to whether it improves the availability of credit or lowers its cost, thus stimulating the economy. I support this approach to drafting the directive going forward. It is one way in which the Committee communicates the logic of monetary policy. But I think we do need to go further—as you emphasized, Mr. Chairman—in providing clear explanations to the public about the objectives of the various facilities, how they work, and why they are part of a coherent monetary policy strategy.
With respect to governance, I endorse the suggestion that you made, Mr. Chairman, about how we should proceed—that is, to work together collectively to forge and communicate a consensus view of the entire Committee to the public, while adhering to the particular responsibilities that the Board and the FOMC each have according to our governing legal document, which is the Federal Reserve Act.