Why is the FOMC statement so important? Why does an occasional announcement from the Fed about what they think the short-term interest rate should be affect the market so much? To read some commentary, it shouldn’t matter because the market does its own thing. In this post I want to think through (and likely make some mistakes) Fed policy, policy implementation, and what the announcement means.

To begin with, the Fed can implement policy in three ways: (1) adjusting reserve requirements; (2) adjusting the target fed funds rate; (3) open market operations. Note that (3) isn’t necessarily the method of implementing (2), although I used to think that way. They are separate, and the target fed funds rate can be implemented through member bank lending and repo/reverse repo (which I’ll call the repo market).

Changing reserve requirements doesn’t happen much, so we can leave that aside. So, let’s talk about the target fed funds rate. The Fed actually specifies a range for the fed funds rate, which we can interpret as a spread that the Fed quotes to its member banks. If inter-bank lending, which is accomplished through repos and reverses, go outside that spread then member banks can go to the Fed to get lending from the discount window. But the Fed discourages this type of lending, so the Fed maintains orderly operations in the repo market through transactions with its own member-dealers. Essentially, we can think of the Fed as setting an “outside” spread, and letting market dealers do their own thing within this spread.

The #1 thing that puzzles me is the 0.25% interest on excess reserves held at the Fed. Obviously this is only available to member banks, but I don’t understand why that isn’t the floor on repo. It seems a member-dealer would borrow as much as possible in the repo market to arbitrage this, but that would push up overnight repo rates to basically 0.25% – transaction costs (essentially 0). So anyone’s insight on this element would be delightful.

Open market operations (OMOs) are about controlling the money supply. In non-crisis situations, the OMOs are used to temporarily insert or remove money from the economy. This temporary action is done through short-term repo. Further, as part of their “dual mandate” the Fed is supposed to target low, stable inflation. That implies a steady increase in money supply. And, typically, that’s what we see when we look at money supply measures (M1, M2, MZM, etc.). This is historically accomplished through increasing the Fed’s holdings of Treasury securities. Also, this was mostly done using short-term securities, especially bills. So basically the Fed is supposed to operate on the “short-rate” in all circumstances, and let the long-rate get figured out through arbitrage in the market.

OMOs are conducted with the Fed’s primary dealers. Those primary dealers are also the Fed’s conduit to the Treasury auction, because the Fed is not allowed to directly purchase from Treasury. Primary dealers get preferential treatment because they are expected to carry out the FOMC’s desires for market operations. Remember, though, how dealers make money – they are spread traders. They might carry some risk if their trade book isn’t perfect matched buys against sells – in other words if they carry excess inventory. But the main task of a dealer is to maintain market liquidity, and they capture a spread (ask-bid) for doing so. Being a dealer in Treasury securities is a good deal, since supply and demand are both high.

Now, during WW2 (the digression has a point, I promise), the Fed was essentially charged with facilitating Treasury’s financing of the war. So the Fed was active across all maturities of Treasury markets. In 1952, there was an internal study done and a report compiled regarding the methods of Fed market involvement, among other things. This study was released as part of a book called “The Federal Reserve System After 50 Years.” An important point made in the study was that involvement of the Fed in all maturities was detrimental to good market order. It was recommended that the Fed should focus on the shortest-term Treasury securities only. Now, I digressed like this to contextualize this graph:

What stands out to me is the heavy buying of long-date Treasury securities after the recession (grey-shaded area) ends. Short-dated securities are essentially zero after early 2012. It seems like, if I wanted to peg long rates, this is exactly what I would do. My understanding of this is that the Fed has entered long-date markets as an active buyer, not a dealer and not a broker. That means they are not setting the outside spread (lender of last resort prices). They are actively in the market, and of size that likely has an effect on the actual nominal interest rate. And remember, the Fed never has a liquidity or solvency problem, so they can do this all day.

I think this overview indicates that more investigation is warranted on this topic. I hope to assemble some data and dig into this further. But, if I’m missing something fundamentally important, please do share in the comments.