In a comment earlier today to this post, David Pearson shocked me by quoting the following passage from the Financial Services Regulatory Relief Act of 2006:
Balances maintained at a Federal Reserve bank by or on behalf of a depository institution may receive earnings to be paid by the Federal Reserve bank at least once each calendar quarter, at a rate or rates not to exceed the general level of short-term interest rates.
As I said to David Pearson in my reply to his comment, I am flabbergasted by this. The Fed is now paying 0.25% interest on reserve balances while and the interest rate on 3-month T-bills is now 0.01%. Yet the statute states in black letters that the rate that the Fed may pay on reserves is “not to exceed the general level of short-term interest rates.” In fact, as can be easily seen on the Treasury’s Daily Yield Curve webpage, only on rare occasions was the 3-month T-bill rate as high as 0.25% in 2009 and it has been consistently less than 0.20% for most of 2009 and all of 2010 and 2011. Perhaps the definition of short-term interest rates is more than 3-months, but the yield even on a one-year Treasury has been in the neighborhood of 0.1% for months and has been below 0.25% since April. So can anyone explain to me by what authority the Federal Reserve System continues to pay banks 0.25% interest on their reserve balances held at the Fed?
In looking around to see if anyone else has noticed that the Fed seems to be violating the very statute that authorizes it to pay interest on reserves, I found the following post from April 2010 by Stephen Williamson on his blog.
The Federal Reserve Act specifies that decisions about the interest rate on reserves are made by the Board of Governors, not by the FOMC. Obviously Congress did not think through the issue properly when it amended the Act. Since the interest rate on reserves is now the key policy rate, decisions about how to set it would appropriately reside with the FOMC. An interesting section of the Act is this one:
Balances maintained at a Federal Reserve bank by or on behalf of a depository institution may receive earnings to be paid by the Federal Reserve bank at least once each calendar quarter, at a rate or rates not to exceed the general level of short-term interest r ates.
This passage may be vague, but 1-month T-bills are now trading at 0.139% and the interest rate on reserves is 0.25%. The problem is that the Fed cannot do its job and (apparently) conform to the law. The T-bill rate has to be lower now, as the marginal liquidity value of a T-bill is higher than for reserves.
So Williamson also believes that the Fed lacks the statutory authority to pay as high an interest on reserves as it is now paying banks, except that he believes that the Fed would not be discharging its other statutory responsibilities properly if it followed the letter of the law on the rate of interest it may pay on bank reserves. But I admit to being totally unable to understand his reasoning. How can he conclude that the marginal liquidity yield of a T-bill is higher than the liquidity yield on reserves? Presumably in a competitive equilibrium, the pecuniary yield plus the liquidity yield on alternative assets must be equalized. But if banks can earn a higher rate on reserves than they can on T-bills, they hold only reserves and no T-bills. Non-banks, on the other hand, are ineligible to hold interest-bearing reserves with the Federal Reserve System, and must hold lower-yielding, less-liquid T-bills. So the rates on T-bills and reserves held at the Fed are not consistent with competitive equilibrium, and no inference about liquidity yields, premised on the existence of competitive equilibrium, follows from current yields on reserves and T-bills.
Uneasy Money 
The law is brighter than the Fed.
But who would prosecute?
LikeLike
Good catch! But aren’t the Fed violating one part of their dual mandate too? Someone should make a citizen arrest!
LikeLike
you are confusing “interest rates” with t-bills, and “short-term” with 1-month or 3-month maturity.
6-month libor is a short term interest rate too (and no way near 0.25%).
LikeLike
3-month LIBOR never fell below 0.25% either.
LikeLike
David,
Why would Congress stipulate that the IOR should not be higher than market rates in the first place? I can’t think of a reason.
Also, banks should own no T-bills or repo’s (also at 0%) under current circumstances. That means that someone else has replaced banks as the holders of those securities (otherwise the T-bill yield would converge on the IOR). Possibly its money market funds and the GSE’s that are pushing the T-bill rate below the IOR. If the IOR were zero, therefore, these entities would arguably push the T-bill yield below zero. So its possible we have a structural divergence between the IOR and T-bill rates based on “shadow banks” that cannot access the IOR.
LikeLike
Why were (local) banks replaced as holders of T-bills? This may not be quite the same circumstance, but I checked on the status of a bond recently and was told that the banks can no longer process or hold bonds, they can’t even use them for door prizes (as mine was, almost 10 years ago). When I asked who processed them, I was astonished to find out that the Federal Reserve handles all this directly. Might the IOR just be a way to make amends to the local banks?
LikeLike
I there a hard and fast definition of short-term? If there is isn’t it 1 year or less?
LikeLike
Lorenzo, Peter, Alea, and JP, I was astonished when I read the statute as quoted by David Pearson. I did not accuses the Fed of breaking the law, but it seemed to me that it is legitimate to ask what exactly is going on. The relevant terms are undefined and the question is who gets to define them.
David, The obvious answer (to me at any rate) is that Congress wants to put some limit on the transfer from the Treasury (via the Fed) to the banks. I don’t know what the limitations are on the assets that can be held by MMFs and GSEs. Presumably it is more costly to hold currency than T-bills which is why the return on T-bills might be slightly less than zero.
Becky, I am just guessing, but aren’t T-bills now all computerized? There is no reason for banks to be involved in the transaction any more.
LikeLike
Seems to me that the interpretation of the law is such that the Fed is unable to pay IOR at a rate that exceeds the administered (target) rate, rather than the prevailing rate, which would explain why the wording “zero to 1/4 percent” still exists in Fed statements.
LikeLike
I wrote up a similar post yesterday, but was waiting to post it until I had a better sense of the law. But I think David Glasner is right, they are violating the law. Of course these laws aren’t really enforced, for instance the Fed isn’t suppose to have more than one board member from the same district. That’s why I think they can do negative IOR if they want, who’s going to stop them? If someone tries, it would be tied up in the courts for years, and by then the need for negative IOR would be gone.
Niklas, If you are right then the law would have no teeth. The Fed isn’t currently maintaining a 0.25% fed funds rate in the market. If they could always say; “our ‘official’ fed funds rate is equal to the IOR,” regardless of where the actual fed funds rate is, then the law would be meaningless. Presumably the Congress didn’t want the Fed subsidizing the banking industry, but they are doing it anyway.
LikeLike
The ‘ short-term interest rate ‘ will be the target interest rate. With interest on reserve balances acting as a floor-system means the central bank can supply reserves without driving market interest rates below the target rate.
David Pearson October 11, 2011 at 11:20 am
” Why would Congress stipulate that the IOR should not be higher than market rates in the first place? I can’t think of a reason. ”
If IOR rate was higher than the target rate institutions could easily engage in arbitrage. For example, a good credit risk could borrow from an institution who do not have access to interest on deposits at the central bank and arbitrage the difference. They can still do that using market power with lenders who have limited alternatives if market rates are below central bank rates. However, there is a potential penalty to be seen purchasing large amounts of overnight funds which could be perceived as raising questions about creditworthiness.
As long as there are institutions that borrow and lend overnight funds without access to the central bank. I think IOR being sometimes above the overnight market rate is unavoidable.
LikeLike
Niklas, You may be right. On the other hand, you may not be. But I think that an official explanation and justification would be in order.
Scott, I can’t be right, because I haven’t taken a position. I just find the current situation troubling and somewhat suspicious. But I am in no position to voice any legal opinion on what the statute actually means. That said, you make a persuasive argument for the prosecution.
Richard, Not sure I follow your last point. Isn’t your argument that IOR should not be above the overnight rate?
LikeLike
Bank liquidity portfolios must be 1 year or less.
LikeLike
From the standpoint of the entire commercial banking system, the member banks don’t loan out existing deposits (saved or other-wise). That said, IOeRs not only stop the flow of savings thru the non-banks, IOeRs induce dis-intermediation (an outflow of funds), among the non-banks (i.e., the intermediaries between savers & borrowers).
The non-banks are the most important financial sector (financial institutions where savings are actually matched with investment). Pre-Great Recession, the non-banks represented 82% of the lending market (Z.1 release).
The laws that the FED are breaking are economic, not legal.
LikeLike
flow5, Interesting. Could you go through the reasoning behind your disintermediation point in more detail, because I am not sure that I totally follow it.
LikeLike
If you understand why Alton Gilbert’s “Requiem for Regulation Q: What It Did and Why It Passed Away” is disinformation, you might also understanding why interest on reserves is contractive, causes dis-intermediation, retards the flow of savings into real investment, and hastens debt deflation.
There are contrary to the Keynesian economics, fundamental differences between the member banks & the non-banks. Lending by the member banks expands both the volume & the velocity of new money. Whereas lending by the non-banks simply results in the turnover of existing money within the CB system.
The final increase (tipping point), to 5 1/2 percent in REG Q ceilings on December 6, 1965 (applicable only to the commercial banking system), caused the non-banks to shrink in size, but allowed the member commercial banks to continue to grow.
This 1966 example is analogous to the .25% remuneration rate on excess reserves today (vis a’ vis other competitive financial instruments & yields), i.e., IOeRs @ .25% is higher than the daily Treasury yield curve up to 2 years out – .30% now on 10/23/11. This causes the short-end of the yield curve commonly referred to as the money market to be inverted.
This view is simplified & conceptual but nevertheless correct.
LikeLike