November 15, 2013

Why the Fed Doesn't Matter: Blame the "Roving Cavaliers of Credit"

The name, in full, of the Fed is the Federal Reserve Bank of the United States. What does that mean? It's in the name: it was conceived as the reserve to which the banks would resort when they were experiencing a run.
In its first big test, the Great Depression, the Fed failed miserably. So badly did they fail that a new agency, the Reconstruction Finance Corporation, or RFC, was brought into existence under Hoover. It tried, but also came up short.
Finally, under FDR, a couple of legislators, Representative Steagall of Alabama, and Senator Vandenberg of Michigan, inserted into the Glass-Steagall Act the creation of the Federal Deposit Insurance Corporation.
The contrast in success between the Federal Reserve and the FDIC is summed up in these two quotes from John Kenneth Galbraith's Money: Whence it Came, Where it Went:

1. On the Federal Reserve:

2. On the FDIC:



Roosevelt merged the RFC into the FDIC, which continued to exist right through WWII.

To give it some credit, Bernanke did make some attempts at staving off a huge bank run in 2008, but once again came up short. Lehman, the instigator of that year's run, was neither a member of the Federal Reserve nor insured by the FDIC. Given 75 plus years to figure out how to get completely around the system meant to tie them down, the nation's financiers had, finally, succeeded. But 75 years is quite a long time to go between huge banking crises, and it took a partial repeal of Glass-Steagall (the parts that separated investment and commercial banking) PLUS the repeal of the Bank Holding Company Act of 1956 that prohibited interstate banking PLUS the repeal of usury laws, allowing unlimited profits to be made in finance, PLUS a truly monumental amount of ingenuity fueled by the hiring of hordes of rocket scientists deploying arcane formulas to hide the risk of the speculation they engaged in, to succeed in finally giving us a crisis worthy of standing up to a comparison to the one that started the Great Depression. That's how well the FDIC succeeded in stopping banking crises.

So, if the Fed can't stop bank runs in a truly big crisis, still, doesn't it control the money supply and therefore the rate of inflation and even employment?
No.
The idea that it does is based on the notion that the money supply can be controlled from a central source. The idea is that by controlling the reserves that a bank has to keep against outstanding loans you can restrict or expand credit. In this theory, a certain amount of money - called "base money" - would be created first by a central authority like the Fed. The banks put this in their reserves, and lend against it. Receiving, say, a million simoleons, they'd lend ten million, given a reserve requirement of 10%. If the system worked this way, you should see base money, aka the monetary base, expand first, and then credit expand afterwards. But as Aussie economist Steve Keen pointed out in Feb 2009, in a truly great essay on his blog titled The Roving Cavaliers of Credit (a quote from Marx) back when the crisis was still in full swing, the evidence directly contradicts the theory:



The essay goes into great detail, and if you're interested in this stuff, it's an excellent read. But the point of why this is so can be made very simply: if banking is a business, and its primary source of profit is the making of loans at a rate of interest that exceeds what it pays on deposits, then it will make as many loans as it can regardless of any reserve requirement and figure out how to get the money to report a good reserve requirement afterwards.
In real life, this is exactly what happens. Here in the US, banks are required to report the average level of their reserves once every two weeks, on Fed Wednesday, as it's called in the business. Note that: the average level of reserves. It's done this way because trying to minutely keep track of your reserves minute by minute is just unrealistic, even in this modern age of computing. As long as you only have to report an average, the reserve level can dip and rise as needed in the intervening period, firstly, and secondly, given that time lag, you can always get the money you need from other banks via the interbank lending market, which is the market that the Fed Funds rate is used for.
As Keen points out, the Fed and the other rich world central banks long ago gave up trying to regulate reserves in any sort of strict way, because they realized how unrealistic it was.

As for interest rates, the only rate the Fed directly controls is the Discount Rate. This is the rate at which member banks are made to borrow from the Federal Reserve. This is a ridiculously unimportant rate in normal times, because in normal times banks only get money from the Fed if they absolutely have to to cover their overnight cash needs. No one does it regularly, and it's always considered a big deal if you do, because it shows you haven't managed your money properly.
Banks get their money if they need it from the interbank market previously mentioned, the one where the Fed Funds rate is used as a benchmark. But the Fed doesn't directly control that one. It sets a "target", but that target is regularly missed in both directions.
Here's the graph of the target rate of Fed Funds:



…and the effective rate of Fed Funds, as in what the market rate actually was:



The effective shows more variation, but seems to follow the target rate.
But what if it's the other way around? What if the effective rate leads the target, rather than following it? Remember, the rate is set by traders in the interbank money market and at what rate they will actually lend to each other. If the Fed sets and enforces the target, rates should move towards the target after the Fed announces the target. But if rates are already moving in the direction the Fed was going to move the target to before it actually did, then it's reasonable to say that the Fed is following the market rather than leading it, and doesn't exert any meaningful control over it.
I tested this by taking the difference in the effective rate versus the average rate for the 21 days (there are on average 21 working days in a month) preceding a change in the target. Out of 150 instances, in 108 of those cases, the market was already moving in the direction the Fed moved the target rate. In other words, if the target was going to go up, the effective rate was already rising, and vice versa.
So, we have only two rates that the Fed has any real control over. Over its own rate for lending out money, the discount rate, it has absolute control, but that rate is rarely important. Over the only other rate it allegedly controls, it follows rather than leads moves in that market. Rather than the Fed setting the rate, the sequence in the overwhelming majority of the cases (108 out of 150, or 72% of the time) is that the market sets the rate, and the Fed follows it with an official announcement.

So, let's recap:

1 - The Fed was founded as the reserve of last resort for banks experiencing a run. It failed that test miserably first time it was really tested, and hasn't done much better since.
2 - Its reserve-setting function doesn't actually do anything. The Fed has admitted as much by relaxing the reserve requirement it allegedly enforces to such an extent it's basically irrelevant.
3 - In interest rates, it does control its own rate for lending out money, but that rate is mostly unused. The other rate it allegedly controls, it apparently doesn't, the evidence rather strongly suggests.

So, I like Janet Yellen. I'm sure she'll be useful as far as taking to the bully pulpit to get economic policy moving in the right direction. But for the reasons noted above, the Fed is mostly useless. It could - and in my opinion should - be merged into the FDIC, as the FDIC long ago proved better at its one truly vital function: stopping bank runs.