The money supply and the price level: a broken link?

This past week, The Association of Private Enterprise Education (APEE) held its annual conference in Las Vegas, Nevada. As always, it was a very interesting event and I would like to take this opportunity to encourage young scholars who are serious about academic research and interested in promoting economic freedom to attend.

At the first monetary policy session that I attended, Sound Money: Are Central Banks Necessary?, it was said that the link between the money supply and the price level is broken. This really stood out to me. While this view wasn’t articulated by all of the presenters, it was certainly held by many of the economists in attendance.

“A new model is needed, even if we don’t know what the model should look like yet.”

Although such a statement has come from the mouths of many notable economists, there are theoretical and economic reasons why I disagree with the notion that a new model must (or should) be developed.

The theoretical reason is that the inverse of the price level (1/P) is the price of money. Therefore, to maintain that the “link is broken” is to say that the supply and demand analysis does not hold to money anymore. However, as long as money is considered an economic good, the law of supply and demand still applies. The USD is still used to buy goods and services, which means it has a value; this in turn means that supply and demand is still the link between money and its price.

One question that was posed at the session was, “What does the price level depend on if the price of money does not depend on the supply and demand of money?” This is not a trivial question. After all, money is the economy’s most important good.

One cannot maintain that money is an economic good used to buy other goods and services while maintaining that the supply and demand model is broken.

The empirical reason for this is that even though base money (BM) increased significantly with the financial crisis, the M2 monetary aggregate did not. The reason for this is that the Federal Reserve is paying banks to not lend money to the market. In other words, the Fed is moving two variables at the same time: it is increasing BM at the same time that it is reducing the money multiplier. The result is the fairly constant evolution of M2 we observe in the data. This does not mean that the model is broken, or that we need a new one, but rather that the Fed is moving different variables of the model at the same time. It is true that since 2008 M2 has increase faster than the price level. But has been the case since the mid-1990s, not since 2008.

But there is another general issue that I find troubling. Semantics implicitly exculpate the Fed from any wrong doing. While I believe that this concept was partially explored by the panel, I am now speaking more broadly. Semantics, of course, prize the Fed for any positive outcomes. The implicit message sent to the public is that the Fed can only do good. For instance, to say that today the “money supply does not depend on the Federal Reserve, but on commercial banks” or that “the Fed has no power to set interest rates” sends the hidden message that the Fed is not the cause of the poor economic performance since the crisis because the Fed does not control the relevant variables.

Of course, banks are the source of the secondary creation of money, but the Federal Reserve remains the primary source of money supply. The Fed asked Congress for permission to pay interest on reserves, and is still doing so. To pay interest on reserves is like paying the banks to reduce the money multiplier. Surely, money creation stops at the banks, but this the result of the Fed’s policy and not a sudden change of behavior by the banks themselves. I would suggest that this policy was put in place for two reasons: First, to provide liquidity to banks while cleaning their balance sheets from “toxic assets,” and second, because the Fed is buying toxic assets, it cannot sell them back to the banks to keep base money on check. So, instead of taking reserves away from the balance of the banks, it’s paying the banks to hold them. Unless supply is vertical, if we start paying to keep reserves sitting at the Fed, then more reserves will be kept sitting at the Fed. This way the Fed avoids the high inflation that we would have seen if the expansion of base money would not have come hand-in-hand with a fall in the money multiplier.

Assume hypothetically that the Fed decides to stop paying any interest on reserves to the bank. Do you expect that all reserves would remain in the banks endlessly, or you expect that banks would start putting the money back in the market? If paying interest on reserves has no effect at all, why would the Fed continue to do this?

Another hypothetical situation to consider would be the result of placing this policy (plus the demand for USD by other central banks around the world) in a demand and supply chart. What we would see is a more horizontal demand of base money than we are used to. An increase in the supply of base money would have a minimal effect on the price level. Changing a parameter of the model (or the slope of demand) is very different than saying that the model is broken and that we need a new one. Supply and demand did not disappear– they changed their slopes. We don’t need a new model as much as we need to revise how policy makers are changing the model’s parameters.

Similar deviation of attention from the Fed’s effects on the economy can be perceived when we maintain that the Fed has no power to set interest rates. If that truly were the case, it would mean that the Fed has been targeting a useless variable for years.

Again, there is a grain of truth in such a statement, but it begs contextualization.

Surely, the Fed only has limited, if any, control over real interest rates. Eventually, changes in the money supply would affect the nominal interest rate through the Fisher effect. The federal funds rate is decided by banks through inter-bank lending, not by the Fed itself. But this doesn’t mean the Fed has no power to affect interest rates in the short-run– even if it did, the short-run might be long enough to produce significant distortions. The Federal funds rate depends on the supply and demand of federal funds. The Fed does not fix the federal funds rate the same way we think of fixing prices under a price control regime. What the Fed can do, however, is change the supply/ demand of federal funds until the federal funds rate is at the desired level. To say that the Fed has no control on interest rates in the short-run is like saying that the Fed has no power to change the price of bananas in the short-run, but can “print” as many bananas as it wants until the price of bananas is at the desired level. Then, when something bad happens in the market of bananas, we don’t think of the Fed because our semantics imply that the Fed has no power to affect the price of bananas.

Such logic is simply not sound.

The monetary situation of today was not caused by a broken model or link; it was caused by discretionary central bank policies around the world– especially those orchestrated by central bankers at the Federal Reserve. If the Federal Reserve, or any other central bank, is powerful enough to do so much good, it is also powerful enough to do great harm. We do a great disservice to the cause of sound money if we promote the idea that the Fed is unable to do harm to key monetary variables.