Monetary Policy is Dead, Long Live Monetary Policy
Much has been written about the Federal Reserve in recent months. Indeed, much of the discourse has focused on expansion of the Fed’s balance sheet and the consequences it has for asset prices, purchasing power and the Dollar. These concerns are not without justification after all, with the Fed’s balance sheet having risen to 42% of GDP and the US budget deficit forecast to hit 18% of GDP this year. We are truly in uncharted waters.
As so many astutely observed, this new era of monetary policy can be elegantly summarized as: “Money printer go: Brrrr!”.
While the above generally misses the issue, it does underscore a fundamental point: monetary policy, as we understand it, is broken. Models of monetary policy that are taught in graduate school are radically outdated and it many respects entirely fanciful.
In this post I intend to highlight some of the more subtle aspects of Fed policy, monetary transmission and banking. Some of what follows will likely surprise you. Hopefully by the end you’ll have better context for how monetary policy functions and can add some nuance to any discussion about “money printing”.
1. There is No Link Between Money Supply Growth and Inflation
The common description of the “money printing” narrative is: “an increase in the money supply will necessarily lead to an increase in inflation, all else equal”. This depiction is intuitive and was plausibly true in Milton Friedman’s day when the monetarist view of economics was first developed. However, the past decade has witnessed a dramatic change in monetary policy and banking practices and the proposed linkage between balance sheet expansion and inflation has not played out in practice.
The below chart demonstrates that M2 growth has very low correlation with CPI. Put simply, an increase in the money supply does not impact prices symmetrically.
Part of this can be explained by a secular decline in M2 velocity. Money velocity is defined as follows:
“The velocity of money is the frequency at which one unit of currency is used to purchase domestically- produced goods and services within a given time period. In other words, it is the number of times one dollar is spent to buy goods and services per unit of time. If the velocity of money is increasing, then more transactions are occurring between individuals in an economy.”
–Federal Reserve Bank of St. Louis
More succinctly, velocity is a measure of turnover. The total stock of M2 has increased substantially, but the impact has been entirely offset by a decline in velocity.
When viewed as a percentage year-over-year change we observe that each time M2 has increased, it has been offset by a commensurate decrease in velocity:
The protracted decline in velocity remains something of an anomaly to economists. I have read no convincing explanation for why velocity has continued to decline and many have questioned its utility as a measure. Nonetheless, the view that an increase in the money supply leads to a decrease in purchasing power via increased inflation simply does not hold up under scrutiny.
2. Loans are Not Based on Reserves
The classical textbook description of banks is as an intermediary between surplus spending units and deficit spending units. Under this framework, banks take in deposits from savers and make loans to borrowers, subject to a reserve requirement (more on that in a moment). When the Fed increases the money supply it is intended to ease financial conditions and stimulate lending.
Unfortunately the savings-lending paradigm misses a crucial aspect of credit creation: most bank credit is based on collateral and not reserves.
The below chart shows total outstanding bank credit against reserves. Until very recently, banks were required to keep 10% of deposits on reserve in the event clients wished to withdraw money. The remaining 90% of deposits could be loaned out. However, as the chart demonstrates, there is little relation between this purported requirement and the actual amount of credit created.
The relationship changed significantly following the 2008 Financial Crisis so let’s concentrate on the before period for illustration. In 2007, total outstanding bank credit stood at ~$8T. Meanwhile, you can’t even see the graph for total reserves in 2007, but if you hover over the x-axis you find that reserves were only ~$45B which is not even 1%.
What’s going on here? The fact is that bank credit provisioning has evolved well beyond the era of reserve requirements. Your credit card is totally unsecured and your bank doesn’t need to hold any reserves back in order to extend credit. Other lending facilities like Home Equity Lines of Credit, collateralized credit lines, margin accounts, etc. also don’t require reserves.
Collateral lending is now king. Today, length of collateral chains and haircut rates-neither of which are determined by the Fed-define the upper bounds of the money supply, not base money and reserve requirements.
To be fair, collateralized lending is modulated at least in part by capital requirements imposed by the Basel Accords, but the fact remains that this transmission mechanism is wholly distinct from changes in the money supply.
3. There are No Reserve Requirements
This revelation is possibly the biggest blow to the textbook description of monetary policy.
In the last section I described how previously banks were thought to loan money out of the deposits they take in; usually 90% loans to 10% reserves. The reserve requirement formed the basis for the concept of the money multiplier. The idea behind the money multiplier is very simple: if banks are only required to keep 10% of cash on hand, then the total effective money supply is 10X greater than the monetary base. Formulaically:
This heuristic was a useful way to think about the effect of increasing the monetary base or changing the reserve ratio. Unfortunately, it is a fiction.
As the last section demonstrated, credit creation has very little to do with reserves. Moreover, in June 2019, the Federal Reserve formally recognized the outdated use of the reserve requirement and reduced the reserve ratio to 0%.
A 0% reserve ratio implies an infinitely high money supply; a notion which clearly makes so sense and puts to rest the Econ 101 depiction of monetary policy and the reserve requirement.
4. “Excess” Reserves
The elimination of the reserve ratio was at least partially motivated by the recognition that we live in an era of excess reserves. Excess reserves were defined as the money kept by banks in excess of the reserve mandate. Remember the chart of total reserves from Section 2? Again we plot total reserves, but now against excess reserves.
Prior to 2008, it was unheard of for banks to keep excess reserves. Excess reserves paid no interest and banks are loath to forgo interest on their cash. Following the Crisis however, we see that excess reserves are almost indistinct from total reserves. This implies that rather than lending it out, banks are content to keep large quantities of cash on hand.
To me, this chart is one of the most important to understand when discussing Fed policy and banking as it raises several questions:
Q: Why is the bank just holding extra cash then and not lending?
A: Because they are being paid to do so. The ability to pay interest on reserves was first introduced in 2006 as part of the Financial Services Regulatory Relief Act. The idea is to allow for a smooth implementation of monetary policy. By giving the Fed discretion to pay interest, the Fed can separate the monetary base from the real economy which is accomplished by paying a higher interest rate on reserves than the rate banks can get in Treasuries or in the Fed Funds market.
The advantage this presents is that it allows the Fed to reduce the supply of outstanding assets via QE, increase liquidity and not induce credit fueled inflation as banks reach for yield. Only once the economy stabilizes and truly profitable loan opportunities present themselves might a bank choose to forgo the risk free rate that they are getting and draw down their excess reserves as loans to individuals and businesses.
Q: What rate do banks get paid on excess reserves?
A: Take a look at the chart below. Generally, the rate on excess reserves has closely tracked the Federal Funds rate. You can see that the two are approximately the same today so holding reserves offers no distinct advantage over lending in the Fed Funds market. However, that wasn’t always the case. Observe that for a good portion of its existence the rate on reserves was decently more than the Federal Funds rates; 25 bps v. ~10 bps during the early-mid 2010's.
The higher rate allows the Fed to let funds “leak” out into the real economy as they see fit. Based on the rate today, they don’t seem to have any concerns around lending.
Q: If rates are so low today, then why aren’t banks lending?
A: And here is the crux of the problem. Even with rates at the zero-lower bound and the money supply expanding banks still aren’t lending. Monetary policy is only effective if it makes its way into the real economy. The Fed has not raised the interest rate on reserves to slow down credit creation, they are encouraging it, but banks still aren’t going for it. This tells us that banks are either very risk averse, the investment returns available don’t justify the risk or they have been cut out of the funding process; something tells me that it’s all of the above.
Concluding Remarks
All of this brings us back to “Money printer go: brrrr!”, which I personally find very funny. As an erudite quip about the current state of monetary policy it is basically accurate. We have truly entered a new stage of thought around the economy and money; one that is quite different from what you will find in an economics textbook or in the general discourse. My hope is that this post has given you a better understanding of monetary policy implementation and transmission.
Until next time, thanks for reading!
- Aric Lux.
Originally published at https://lightfinance.blog on January 24, 2021.
