Twisting

There are some things in the minutes of the September 20-21 FOMC meeting that seem worth discussing.

Bernanke and other Fed officials like to tell us about the large toolbox they have available for fixing what ails us, and the meeting began with a discussion of the available tools that could be used to supply more accommodation. The choice was framed as Goldilocks would see it. There is "too cold," "just right," and "too hot," in that order, and you know at the outset that the committee will choose just right.

Too cold would involve a change in how the proceeds from principal payments on its holdings of agency securities would be revinvested. Policy before the last meeting was to hold the size of the Fed's balance sheet constant, and to take the proceeds from agency securities and mortgage-backed securities (MBS) that run off, and reinvest that in Treasury securities with a particular average maturity. The proposal was to simply lengthen that average maturity. Just right was Operation Twist - lengthen the average maturity of the Fed's portfolio by selling short-maturity Treasury bonds and buying long Treasuries, while holding the size of the balance sheet constant. Too hot was a repeat of QE3 - an increase in the size of the Fed's balance sheet through purchases of long Treasuries.

At this point in the meeting, there is some discussion. People on the committee who are in an accommodative mood are thinking they will want to keep trying if whatever the committee decides to do now does not work. Just right is seen as a one-time intervention - clearly you can't keep increasing the average maturity of the Fed's portfolio indefinitely. Some people raise some objections: maybe none of these interventions will have much of an effect, if any; maybe we will get too much inflation. A proposal is introduced which was not heretofore on the table (and which ultimately the committee will go for), which is to reinvest the proceeds from maturing agency securities and MBS in more MBS, rather than in Treasuries.

Then, there is a discussion about transparency. In particular:

Most participants indicated that they favored taking steps to increase further the transparency of monetary policy, including providing more information about the Committee's longer-run policy objectives and about the factors that influence the Committee's policy decisions.

It's hard to know what to make of this without more specifics. Maybe what the committee members had in mind was in line with what Evans talks about here. If so, it's wrongheaded, and some people on the committee seem to think so too:

a number of participants expressed concerns about the conceptual issues associated with establishing and communicating explicit longer-run objectives for the unemployment rate or other measures of labor market conditions, inasmuch as the long-run equilibrium levels of such measures are influenced importantly by nonmonetary factors, are subject to change over time, and are estimated with considerable uncertainty. In contrast, participants noted that the long-run level of inflation is determined primarily by monetary policy.

The committee also considered the possibility of lowering the interest rate on reserves (IOR), presumably to 0% from 0.25%. It is quite important that this discussion appears in the FOMC minutes, as decisions about changes in the IOR actually rest with the Board of Governors, not the FOMC. In discussing this at the FOMC meeting, the Board is recognizing that the committee should have a role in the decision, though that role was not given to it by Congress. In my view, a change in the IOR, or language that tells us about the future path of the IOR, is the only relevant element of Fed decisionmaking currently. None of the quantitative interventions actually matter, under current circumstances.

Here is a useful piece of information from the IOR discussion:

a recent change in deposit insurance assessments had the effect of significantly reducing the net return that many banks receive from holding reserve balances.

There are some seemingly puzzling things going on with respect to the behavior of the fed funds rate relative to the IOR. One might expect that the IOR would place a lower bound on the fed funds rate, much as in any channel system (Canada, Australia, ECB, for example). But this is not the case, as the fed funds rate is currently less than the IOR, and has even decreased since late 2008. The GSEs (Fannie Mae, Freddie Mac) do not receive interest on reserves, and commercial banks do not arbitrage away the difference between zero and 0.25%, for reasons that are in part unexplained. However, a contributing factor to the lack of arbitrage has been the change in deposit insurance assessments. Banks are now charged the assessment based on total assets, not deposits. Thus, if I am a bank and attempt to arbitrage the difference between the fed funds rate and the IOR by borrowing on the fed funds market and holding what I borrow as reserves, I increase what I pay to the FDIC. As well, there seems to be some effect of the total quantity of reserves in the system on the IOR-fed funds rate differential (higher reserves increases the differential), but this is just a correlation with no theory backing it up.

Otherwise, the IOR discussion is a bit murky. For example:

Moreover, many participants voiced concerns that reducing the IOR rate risked costly disruptions to money markets and to the intermediation of credit, and that the magnitude of such effects would be difficult to predict in advance. In addition, the federal funds market could contract as a result and the effective federal funds rate could become less reliably linked to other short-term interest rates.

The "disruptions to money markets" may refer to effects that might arise because of the rules governing money market mutual funds, but I'm not sure. I'm not sure why anyone is concerned with activity on the fed funds market. Most of the activity on this market currently must simply be commercial banks borrowing from GSEs. Why do we care if that goes away?

The FOMC of course settled on a policy involving a swap of $400 billion in short-term Treasuries for an equal quantity of long-term Treasuries ("Operation Twist"), and a policy of reinvesting principal repayments on agency securities and MBS in new MBS. The committee members voting for these measures seem to think this will result in decreases in long-term interest rates, and a reduction in the margin between mortgage rates and other long-term interest rates.

There were of course three dissenting votes. Fisher's objections were:

Mr. Fisher saw a maturity extension program as providing few, if any, benefits in support of job creation or economic growth, while it could potentially constrain or complicate the timely removal of policy accommodation. In his view, any reduction in long-term Treasury rates resulting from this policy action would likely lead to further hoarding by savers, with counterproductive results on business and consumer confidence and spending behaviors. He felt that policymakers should instead focus their attention on improving the monetary policy transmission mechanism, particularly with regard to the activity of community banks, which are vital to small business lending and job creation.

The first sentence makes sense, but I can't decipher the rest. Hoarding by savers and improvements in transmission through community banks? What's that about? We know Fisher is not an economist, but he has economists briefing him. Does he not listen? Does it not sink in? Are the briefers bad at their jobs? Who knows?

Kocherlakota says:

Mr. Kocherlakota's perspective on the policy decision was again shaped by his view that in November 2010, the Committee had chosen a level of accommodation that was well calibrated for the condition of the economy. Since November, inflation, and the one-year-ahead forecast for inflation, had risen, while unemployment, and the one-year-ahead forecast for unemployment, had fallen. He did not believe that providing more monetary accommodation was the appropriate response to those changes in the economy, given the current policy framework.

This is basically identical to Kocherlakota's objection at the previous meeting. Finally, Plosser:

Mr. Plosser felt that a maturity extension program would do little to improve near-term growth or employment, in light of the ongoing structural adjustments and fiscal challenges both in the United States and abroad. Moreover, in his view, with inflation continuing to run above earlier forecasts, such a program could risk adding unwanted inflationary pressures and complicate the eventual exit from the period of extraordinarily accommodative monetary policy.

I think that is basically correct, but the maturity swap is essentially irrelevant for inflation, and does not further complicate exit, given that the size of the balance sheet is being held constant.

So, the majority of voting members on the FOMC seem to think that they can actually do things that are more "accommodative" currently. Even the people who are objecting (particularly Plosser) seem to think that the Operation Twist maturity swap, and the QE2 swap of reserves for long Treasuries, actually matter for long-term interest rates. If the Fed can in fact move long-term interest rates at will, then in fact they should be able to target long-term nominal interest interest rates. Indeed, if we believe what the FOMC says, the Fed should be able to determine the whole nominal term structure of interest rates by intervening sufficiently. Why then are these unusual interventions specified not as interest rate targets but in terms of the quantities of assets purchased? You know why. If they thought they could hit the interest rate targets, they would announce it that way. But they know they cannot; basically it doesn't work.