Who Actually Makes Money?

In an earlier post, I described a contradiction between popular economic theory and statistical data.

In this post, I will describe true premises upon which we can develop more accurate theories.

Popular Assumptions

Reserves held at the Federal Reserve somehow act as money — so goes popular assumption. When the Fed adds dollars to bank reserves, that simultaneously adds an equal number of dollars to the quantity of money. For a long time, empirical evidence seemed to confirm that theory.

Around 1973 evidence began to appear that contradicted this popular assumption. Since that time, the money supply has grown much more rapidly than have bank reserves.

In 2008 the assumed connection between bank reserves and the quantity of money completely broke down. The quantity of bank reserves skyrocketed while the money supply chugged along at about the same historical rate.

This contradiction exposed a critical question. Where did the error exist? — In the data? Or in the theory?

Flawed Premises

The error existed in the theory, for the theory started with a couple of false premises:

  • dollars = dollars
  • dollars always play the role of money

When a person selects one dollar bill over another from their wallet, they provide evidence that they do not value two, seemingly identical, dollars the same. People frequently confuse the idea of fungibility with that of equal value. Fungibility means that in a practical application, two nearly identical commodities can substitute for each other. On the other hand, no two individual commodities have equal value, for the simple fact that no two items are identical in all respects, and no individual will value them the same.

Every dollar is different from every other dollar. They have different serial numbers; some are new and crisp and others not; some dollars exist as Federal Reserve notes; some dollars exist as coins; some dollars even exist as negotiable checks. Thus, the idea that every dollar equals every other dollar either physically or in value proves false.

Every dollar does not play the role of money. The dollars with which you pay for your latte act as money. Any dollar you use in any exchange acts as money. Some dollars, however, do not act as money.

I once ate in a restaurant in which the walls were covered with dollar bills. People had written on those bills, laminated them, and hung them on the walls. These dollars did not act as money.

Dollars held in the accounts of banks at the Federal Reserve also do not act as money. For dollars to act as money, the public must be allowed to own and use those dollars for indirect exchange. The public cannot hold (or have a claim on) dollars held as bank reserves. We cannot, therefore, consider bank reserves as a form of money.

The Role of Bank Reserves

Bank reserves do not act as a form of money. They never have. When people use gold as a form of money, and banks had 100% reserves, banks would hold gold to back the claims represented by banknotes or checks. While held as reserves, gold no longer played the role of money. Only after the holders of banknotes and checks exercise their claim on gold (at which time the holders claim on the gold would be canceled), with the gold return to its former role as money. Gold and banknotes/checks would never play the role of money at the same time.

In the modern banking system, in which we have fractional reserve banking and reserves created by the Federal Reserve, this relationship has not changed. Bank reserves still do not play a role as money. They act more like a control mechanism on the ability of banks to create money.

In an hypothetical situation in which banks must hold a reserve equal to 50% of their deposit liabilities, a bank with $1,000 in reserves (2,000 reserve dollars) could create 2,000 money dollars (I use the phrase money dollars to distinguish them from dollars held as bank reserves). If the Federal Reserve created another thousand dollars in reserves, to pay for assets they acquired from the bank, the bank would have no obligation to create another 2,000 money dollars.

Historically, because of the relatively low level of bank reserves, when the Fed increased the quantity of reserves, banks would have sufficient demand for money that they would buy an adequate number of notes to fully use the capacity of their moneymaking capability. With the massive run-up in bank reserves at the time of the financial crisis in 2008, banks no longer had access to a sufficient number of quality notes for which to use their full moneymaking capacity.

Conclusion

The premise that the Federal Reserve adds directly to the money supply when it increases bank reserves contained flaws from the beginning. Bank reserves have never acted as a form of money.

When the Federal Reserve does increase the amount of bank reserves (under a fractional reserve banking system), banks have no obligation to use their full moneymaking capacity.

Thus, when we understand the distinction between dollars held as bank reserves (reserve dollars) and dollars held by the public for indirect exchange (money dollars), we have a more accurate premise from which to develop a theory about the relationship between bank reserves in the money supply. It makes perfectly good sense for bank reserves and the money supply to grow and contract at somewhat different rates.

 

Money Supply and Bank Reserves

Introduction

The fact that we perpetuate false or inaccurate theories causes at least one problem for understanding money and its role.

For years economic classes have taught that the Federal Reserve controls the supply of money. When they buy government bonds from banks, they pump reserves into the banks’ reserve accounts, and that addition to reserves somehow causes an expansion in the supply of money.

We continue to talk in these terms even though (thanks to Ben Bernanke) we now have evidence that that might not be the case. And possibly never was the case.

Contradiction

This chart shows the money supply (the Blue Line — scale on the left) rising continually from April 2014 to April 2019.

During that same period, bank reserves (the reddish line — scale on the right) declines continuously. (The thin green line at the bottom shows required reserves during that same period.)[1]

Figure 1 -2014-04-01 to 2019-04-01[2]

According to popular theory, this behavior should not happen. When bank reserves decline, the money supply should also decline.

Real behavior, at least for this period, seems to contradict popular theory. The data and the theory both cannot be true. Either the data contains an error (or an unknown influence), or the theory contains an error.

An Apparent Connection

If we look at historical data from past periods, we can see how people made a connection between bank reserves and the money supply.

During the period from 1963 to 1973, both bank reserves and the money supply traced similar patterns. Only in the latter part of those ten years did the patterns diverge.

Figure 2 – 1963-01-01 to 1973-01-01

In the chart below, as bank reserves rose from 1985 to roughly 1989, the money supply followed a similar pattern. Only at the end of this period from 1989 into 1990 did the patterns of bank reserves and the money supply diverge.

Figure 3 – 1985-04-01 to 1990-04-01

The two periods represented by these two charts seem to indicate a reasonably close connection between the rise of bank reserves and the increase of the money supply. They seem to confirm the popular theory that the Federal Reserve drives the growth of the money supply.

Can we reconcile the apparent contradiction between this evidence and that shown in the first chart?

Sign of Broken Connection

When we look at a longer period — from 1985 to 2008, we have evidence that no connection exists between bank reserves and the quantity of money.

Figure 4 – 1985-04-01 to 2008-04-01

Could this chart, and the first chart, provide signs that a pre-existing connection between bank reserves and the money supply had broken?

If there was a connection that no longer exists, that calls the original theory (or the data) into question.

Let’s look at a longer period of time to see if it gives evidence as to which to question — the data or the theory.

A Break From the Theory

If we look at a chart for an extended period, from 1985 to 2019, we see that the data seems to have had a fairly long break from the theory.

Figure 5 – 1985-04-01 to 2019-04-01

During that period bank reserves rose at a relatively insignificant rate up until 2008. During that same period, the money supply grew at a much faster pace.

Then, in 2008 — in response to the financial crisis of that same year, bank reserves grew at a phenomenal rate until roughly 2015 at which time the quantity of bank reserves began to decline. During this entire period, from 1985 two 2019, the money supply grew at a fairly consistent rate, never in real correlation with the level of bank reserves.

Can we draw any firm conclusions from the evidence that I have given above?

Conclusion

We need to find some way to reconcile the apparent conflict between popular theory and the data that I have provided in the charts above.

If the theory contains no flaws, a very powerful influence must exist to cause the data to diverge so far from what the theory would dictate.

If, on the other hand, the data reflects reality, we must conduct a thorough examination of the popular theory.

In future posts, I will explain why the popular theory contains fatal flaws. A more accurate theory will explain and predict results consistent with the data given.


Footnotes:

  1. I divided the Reserve Balances Required, which were stated in millions of dollars, by 1000 to convert the quantities into billions of dollars, making the figures uniformly comparable between total reserves and required reserves.
  2. The Federal Reserve Economic Data (FRED) site at the St. Louis reserve bank generated all the charts used in this article.

 

Fed Funds Rate; Who Cares?

People who care about financial markets have speculated about the effects of the Fed “raising” the fed funds rate. Would it cause the stock market to decline? Would the economy return to recession? Would mortgage rates start to rise and kill the housing recovery?

A person does not need a plethora of charts and graphs to gain a basic understanding of relationships between fed actions, fed funds rates, financial markets, and economic activity.

First, the Fed does not “set” interest rates in any market, including the fed funds market. Interest rates consist of the ratio of future money to current money. As a ratio— calculated from two independent variables — an interest rate is a dependent variable. It cannot be set or determined directly. One or more of the dependent variables must change in order for an interest rate to change.

Second, participation in the fed funds market is limited to financial institutions that have accounts with the Fed. As a closed market the total number of dollars borrowed equals the total number of dollars lent. Net lending in the fed funds market equals zero. Thus, the level of total excess reserves does not affect rates. It’s the interbank imbalances in reserve accounts that cause fed funds borrowing. So what determines interest rates?

Third, because of the closed nature of the fed funds market, the Fed funds rate is determined by the relative levels of excess reserves between banks in the system. To discover the determinants of fed funds rates, you need to examine the factors that affect those relative excess reserve balances. Those factors can include: Fed open market activities, which either increase or decrease excess reserves in individual banks; the demand for bank funds (i.e. deposit liabilities); or the willingness of banks to create more deposits for the purchase of notes.

Fourth, with a banking system awash in excess reserves, what would cause rates to change? How many securities would the Fed need to sell to have the slightest effect on the fed funds rate? A lot. Simply announcing a rate hike does not change that. The Fed funds rate is zero because banks don’t have a need for reserves. Fed action has not changed that.

I will address, in more detail, questions raised by my comments above in future posts.

Bye