Yves here. Get a cup of coffee. This is a very readable, information-rich post on what the Fed does and does not do, with emphasis on the nitty-gritty of monetary policy. If you are time-pressed, read the last item in the FAQ first, which is a terse item-by-item debunking of widely made, inaccurate statements about the Fed.
By Eric Tymoigne. Originally published at New Economic Perspectives
Previous posts studied the balance sheet of the Fed, definitions and relation to the balance sheet of the Fed, and monetary-policy implementation. In this post, I will answer some FAQs about monetary policy and central banking. Each of them can be read independently.
Q1: Does the Fed target/control/set the quantity of reserves and the quantity of money?
The Fed does not set the quantity of reserves and does not control the money supply (M1). It sets the cost of reserves; that’s it.
In terms of reserves, the Fed was created to provide an “elastic currency,” i.e. to provide monetary base according to the needs of the economic system in normal times and panic times. It would be against this purpose to implement monetary policy by unilaterally setting the monetary base without any regards for the daily needs of the economy system.
In terms of the money supply, the Fed has no direct influence. Even Federal Reserve notes (FRNs) are supplied through private banks, and banks supply only if customers ask for cash. The Fed does not force feed FRNs to the public, i.e. FRNs can’t be “helicopter dropped” via monetary policy. If the Fed did this, not only would it operate in a way that goes against the Federal Reserve Act, but also it would lead people to take the FRNs, bring them to banks, banks would have more reserves, FFR would fall, Fed would remove excess reserves to bring FFR back up—back to square one.
The Fed may have an indirect influence on the money supply through changes in its FFR target because changes in the cost of credit may change the willingness of economic units to go into debt, but the link is tenuous (see Q10).
Q2: Did the Volcker experiment not show that targeting reserves is possible?
In the 1970s, the Monetarist school of thought gained some influence in policy circles. Monetarists argued that there is a close relation between the quantity of reserves and the money supply, and that a central bank’s role is to control the quantity of reserves in order to influence the money supply and ultimately inflation. The Fed, under the leadership of Paul Volcker, tried to implement a monetary-policy procedure that aimed at more closely targeting reserves and monetary aggregates, with the hope of taming a double-digit inflation rate.
Practically what the Fed did is change interest-rate targeting from a narrow range to a wide range, which rose the level and volatility of the FFR dramatically (Figure 1). The Fed did NOT allow the FFR to move freely as targeting a total quantity of reserves would imply.
There is a debate about how truly “Monetarist” the Volcker experiment was, and this debate is reflected in the FOMC discussions of the time
- ROOS. Well, if the level of borrowing comes in higher than we would anticipate, [can’t] you reduce the level of the nonborrowed reserves path accordingly? Can’t you adjust your open market operations for the unexpected bulge in borrowing or the unexpectedly low borrowing if you ignore the effect on the fed funds market? Can’t you just supply or withdraw reserves to compensate for what has happened?
CHAIRMAN VOLCKER. The Desk can’t [adjust] in the short run. It’s fixed. In a sense they could do it over time if people are borrowing more, as they may be now. They seem to be borrowing more than we would expect, given the differential from the discount rate. But in any particular week it is fixed.
- ROOS. Do we have to supply the reserves?
CHAIRMAN VOLCKER. We have to supply the reserves.
- ROOS. [Why] do we have to supply the reserves? If we did not supply those reserves, we’d force the commercial banks to borrow or to buy fed funds, which would move the fed funds rate up.
(FOMC meeting, September 1980, page 6)
Mr. Roos was deeply dissatisfied with the Fed still having a FFR target, albeit in the form of a wide range. While the Fed had a total-reserve growth target related to the 3-month growth rate of M1, if banks needed more than what was targeted, the Fed would supply extra reserves in order to relieve pressures on the FFR. Roos argued against this lenient reserve targeting and was for a total abandonment of any FFR targeting and a strict targeting of total reserves. Here he is in 1981:
I think there’s a very basic contradiction in trying to control interest rates explicitly or implicitly and achieving our monetary target objectives. And I would express myself as favoring the total elimination of any specification regarding interest rates.
(Roos, FOMC meeting, February 1981, page 54)
Most Federal Open Market Committee members were against that position, on the ground that the role of the Fed is precisely to promote an elastic monetary base. The Fed was not created to dictate what the quantity of reserves ought to be but too eliminate liquidity problems through smooth interbank debt clearing at par, lender of last resort, and interest-rate targeting. In addition, banks mostly hold reserves because they are required to, so if the Fed does not supply enough reserves to meet the requirements then banks will break the law.
The experiment was a monetary-policy failure. The Fed was never able to reach its reserve or money targets and the experiment contributed to massive financial instability and a double-deep recession. On may even doubt that it contributed significantly to the fall in long-term inflation, which had more to do with the downward trend in oil prices, oil-saving policies, and greater international labor competition:
May I remind you that we shouldn’t take too much credit for the price easing? I never thought we were totally at fault for the price increases that we suffered from OPEC and food; and I don’t think the fact that OPEC and food have calmed down has a great deal to do with monetary policy per se, except in the very long run.
(Teeters, FOMC meeting, July 1981, page 46)
The Volcker experiment was, however, a public-relation success. Most FOMC members knew that reserve targeting was not possible but, it allowed them to claim that they were not responsible for the high interest rates of the period:
I do think that the monetary aggregates provided a very good political shelter for us to do the things we probably couldn’t have done otherwise.
(Teeters, FOMC meeting, February 1983, page 26)
I think the important argument, and really the reason why we went to this procedure, was basically a political one. We were afraid that we could not move the federal funds rate as much as we really felt we ought to, unless we obfuscated in some way: We’re not really moving the federal funds rate, we’re targeting reserves and the markets have driven the funds rate up. That may have had some validity at the time, and I had some sympathy for it. But as time goes on, I’ve become more and more concerned about a procedure that really involves trying to fool the public and the Congress and the markets, and at times fooling ourselves in the process.
(Black, FOMC meeting, March 1988, page 12)
Of course the high and volatile FFR was precisely the result of the change in monetary-policy procedures. If needed, some FOMC members were willing to do the same thing in the future:
Well, I have only a little to add to all of this. I think Tom Melzer is probably right: We’re going to need to shift the focus to some measure or measures of the money supply as we proceed here if we can, both for substantive reasons and also because that has some political advantages as well, as we go forward.
(Stern, FOMC meeting, December 1989, page 50)
Q3: Is targeting the FFR inflationary?
With the failure of the Volcker experiment, the FOMC entered a period of operational uncertainty until the mid-1990s. The Fed was back on a tight FFR target procedure (there was still a wide range until 1991 but the Fed targeted the middle of the range) and this was deeply unsatisfactory to FOMC members.
We’ve advanced from pragmatic monetarism to full-blown eclecticism.
(Corrigan, FOMC meeting, October 1985, page 33)
No, I would say that we have a specific operational problem that we have to find a way of resolving. Just to be locked in on the federal funds rate is to me simplistic monetary policy: it doesn’t work.
(Greenspan, FOMC meeting, October 1990, pages 55–56)
In a world where we do not have monetary aggregates to guide us as to the thrust of monetary policy actions, we are kind of groping around just trying to characterize where the stance is.
(Jordan, FOMC meeting, March 1994, page 49)
The Fed was unwilling to disclose that it was targeting the FFR, and continued to announce a targeted growth ranges for monetary aggregates even though it did not use them for policy purpose.
[In response] to talk that says we can significantly influence this – or as the phraseology goes that if we lower rates, we will move M2 up into the range – I say “garbage.” Having said all of that, I then ask myself: ‘What should we be doing?’ Well, we have a statute out there. If we didn’t have the statute, I would argue that we ought to forget the whole thing. If it doesn’t have any policy purpose, why are we doing it? By law [we have] to make such forecasts. And if we are to do so, I suggest that we do them in a context which does us the least harm, if I may put it that way.
(Greenspan, FOMC meeting, February 1993, page 39)
We do not, in fact, discuss monetary policy in terms of the Ms between Humphrey-Hawkins meetings. Don Kohn dutifully mentions them because he thinks he ought to, but that is not the way we think about monetary policy.
(Rivlin, FOMC meeting, February 1998, page 91)
To announce to the general public that the FOMC was targeting the level of the FFR would be going against all the Monetarist principles (reserve targeting, money multiplier theory, quantity theory of money). By targeting an interest rate, and so having an elastic supply of reserve (horizontal supply at a given FFRT), it seemed to indicate that the Fed was no longer able to control the money supply and so inflation. FOMC members themselves believed that was the case:
Many analysts, both inside and outside the Fed, argued that using the Federal funds rate as the operational target had encouraged repeated over-shooting of the monetary objectives. (Meulendyke 1998 49)
Talking about the FFR target became a taboo and “the Committee deliberately avoided explicit announced federal funds targets and explicit narrow ranges for movements in the funds rate” (Kohn, FOMC Transcripts, March 1991, page 1):
I must say I’m still quite reluctant to cave in, if you will, on this question that we can do nothing but target the federal funds rate.
(Greenspan, FOMC transcripts, March 1991, page 2)
As a practical matter we are on a fed funds targeting regime now. We have chosen not to say that to the world. I think it’s bad public relations, basically, to say that that is what we are doing, and I think it’s right not to; but internally we all recognize that that’s what we are doing.
(Melzer, FOMC transcripts, March 1991, page 4)
We will see later why the entire Monetarist logic is flawed. Monetary policy is always about providing an elastic supply at a given interest rate. There is nothing intrinsically inflationary about this. Having an “elastic currency” usually just means supplying whatever amount of reserves banks want, and usually banks don’t want much.
While all this was very well understood by many economists long before Volcker experiment (see for example Kaldor), it took the FOMC member until the mid-1990s to get comfortable with FFR targeting.
Q4: What are other tools at the disposition of the Fed?
Monetary policy is always about setting at least one interest rate. While today the Fed operates mostly through the Fed Funds market, it has other tools at its disposition to help influence interest rates.
One is the (re)discount rate, the rate at which banks can obtain borrowed reserves. This interest rate is now higher than the FFR target but from the mid 1960 until 2003 the discount rate was usually below the FFR target. The Fed decided to put the discount rate above the FFR target to put a ceiling on the FFR and so limit upward volatility in FFR.
Reserve requirement ratios can also be used to help target the FFR. These ratios state how much total reserves banks must keep on their balance sheet in proportion of the checking account they issue. This is what the ratios look like today in the US
If a bank issued less than $15.2 of checking accounts it does not have to have any reserves, 3% between 15.2 to $110.2 million worth of outstanding checking accounts, 10% beyond that.
While these ratios are often discussed in relation to the ability of banks to create checking accounts, their actual purpose is once again to help target the FFR. By raising reserve requirement ratios, the demand for reserves becomes more predictable given that a greater proportion of the reserves available must now be held by banks. With a more stable demand for reserves comes less volatility in FFR (see this for more)
Q5: What is the link between QE and asset prices?
The link can’t be one where banks have excess reserves that they use to buy asset prices in the secondary market. As we saw in a previous post, banks can’t do this as a whole. They could buy from one another but if they all have reserves they want to get rid of, then this is not possible because no bank wants to sell assets for reserves.
The link goes through the following channels:
- The Fed buys large amounts of securities from banks which raises their prices and so lower their yield
- As yields on these securities fall, economic units seek assets that provide a higher yield. They will look for assets with large expected capital gains, especially so knowing that others are experiencing the same problems and are “searching for yield.” They will continue to buy these securities, which will raise their prices (and so lower their yield), until all rates of return are equalized once adjusted for risks.
Financial companies have to do the second step because they have to meet target rate of returns that they promised to their stakeholders. Pensioners expect a substantial rate of return from their pension funds, wealthy individuals expect a substantial rate of return from hedge funds, mutual funds shareholders wants a substantial rate of return…and they all check every quarter if financial companies stay on course to meet the promised target. Long-term treasuries used to provide a safe and simple way to meet this promise, no longer so.
Bill Gross (a well-known portfolio manager who specialized in bond trading) brought the point forward very clearly when he hoped that the Fed would raise the FFR to make it easier to reach targeted rates of return. He noted that low FFR prevents savers to earn enough to pay for healthcare, retirement and other costs because return on financial assets are so low compared to the expected 7-8%. If the Fed does not help by raising the FFR, financial-market participants will take large risks on their asset side (speculative, high credit risk, and structured securities) and liability side (high leverage) to try to reach their targeted rate of return.
There is a broad problem though. In an economy in which the growth of standard of living is low, why should anyone expect that demanding 7-8% be sustainable? Those can only be achieved through capital gains and leverage and the combination of these two is highly toxic (we will see why later in a post about financial instability). Instead of the Fed raising its FFR, it should be the rentiers who should reduce their expectations of rate of return. An economy that growth at 2% per year cannot sustainably provide a real rate of return higher than 2%, even that is a stretch. Other means must be used to meet the challenge of an aging economy than increasing the financialization of the economy.
Q6: How and when will the level of reserves go back to pre-crisis level? The “Normalization” Policy
Normalization policy means the willingness of the Fed to do two things: 1-to raise the FFR to more normal level 2- to reduce the size of its balance sheet in order to return the proportion of excess reserve to pre-crisis levels. The previous blog showed how this would be done for interest rate. Regarding reserve balances, a previous blog showed that their dollar amount is determined as follows:
L2 = Assets of the Fed – (L1 + L3 + L4 + L5)
Most of L2 is now composed of excess reserves, which is unusual. A graphical representation of this balance-sheet identity is Figure 2.
The implication of this balance-sheet identity is that reserves balances will fall either when assets of the Fed decline given other liabilities, or when the other liabilities of the Fed rise given assets. Let’s look at each case in turn.
Given other liabilities than reserves balances, reserves won’t go back down until the following happens to the securities held by the Fed:
- Securities ISSUED BY NON-FED-ACCCOUNT HOLDERS mature (let’s call them “private securities” to simplify)
- The Fed decides to sell some securities to banks.
If the Fed let treasuries mature the account of the Treasury (L3), not reserve balances, will be debited:
When private securities held by the Fed, currently agency-guaranteed MBS, mature the following will occur:
If all MBS held by the Fed matured at once, that would reduce reserve balances by almost $2 trillion (the Fed does not plan to sell most of the MBS it holds).
Given assets, reserve balances will go down if banks need to make payments to other account holders or if banks convert reserve balances into cash. For example, if the Treasury ran fiscal surpluses reserve balances would fall as funds would move from L2 to L3. While the Fed may ask, and has asked, for the Treasury’s help in managing monetary policy (more on this in the next blog), the Fed has mostly no control over what happens to liabilities that impact reserve balances.
One may note to conclude that, besides changes in assets and liabilities, another way to reduce excess reserves without reducing the amount of reserves is to raise reserve requirement ratios. As to when the reserves will be back to their usual level, nobody knows. The pace of decline will change as the economic environment change.
Q7: Is there a zero-lower bound?
This question is so 2013! There is no lower bound and I have an old post that explains all this in more details. Remember that the Fed sets the price of reserves. It can set it wherever it wants: “It is the Fed’s way or the highway.”
Which interest rate can be below zero? Any of them as long as the Fed is committed to do so.
The interest rates used in the corridor framework (DWR, FFR, IOR) are under total control of the Fed so they are easy to make negative. The Swedish central bank shows us how this is done under a corridor framework (Figure 4). The Swiss central bank shows us how it is done with a negative overnight interbank rate range (a negative FFR range) (Figure 3).