April 3, 2019 Reading Time: 4 minutes

Since the financial crisis of 2007–8, the Federal Reserve has increased its holdings of securities from less than $1 trillion to $4 trillion. This policy is called quantitative easing. The purchases of these assets after the Financial Crisis of 2007-2008 were justified as a way to encourage home buying and long-term investment by firms. The empirical evidence indicates that effects on long-term interest rates and mortgage rates were 20 to 40 basis points (0.2 to 0.4 percentage points) and may have been only temporary. Even if these interest rates were permanently lower by 20 to 40 basis points, it is doubtful that such small changes in interest rates would have much effect. The increase in the Federal Reserve’s holdings in securities is dramatic, but perhaps the costs are minimal. Or maybe not.

A theory floating around in an obscure part of the economics firmament has gained prominence among some non-academic audiences, even being cited by some politicians as showing the way to pay for new government programs. This theory goes by the name “modern monetary theory,” or MMT for short.

MMT is at least partly the basis for some politician’s recent claims that the federal government need not raise taxes to finance a substantial increase in government benefits. Instead, the Federal Reserve can buy new government debt and the funds can be used to finance government purchases.

On its face, this claim is preposterous. There are few settled propositions in monetary economics, but one well settled proposition is that “Inflation is always and everywhere a monetary phenomenon.” High inflation invariably is associated with large increases in banks’ reserves and the nominal quantity of money. Conversely, large increases in the nominal quantity of money are associated with high inflation. It is not necessary to look at ancient history to see this theory at work. Venezuela is current evidence of exactly this proposition, with the government financing its spending by printing money and the consequence being inflation of about 100 percent per month. Venezuelans are not alone. Zimbabwe suffered the same fate in 2007 and 2008 with inflation peaking at 79.6 billion percent per month. For a time after that inflation, Zimbabweans used U.S. dollars and had U.S. inflation, but the government recently returned to printing Zimbabwean dollars and increasing banks’ reserves to pay for government spending. Inflation was about 60 percent per month in February 2019.

Are these politicians advocating 60 percent monthly inflation? No doubt they do not think so. MMT provides some cover for the claim that the Federal Reserve can finance government spending. Banks might hold large increases in reserves if provided and the nominal quantity of money would not increase. Besides, if inflation did increase, reserves supposedly would be reduced by federal-government taxation.

Purchases of trillions of dollars worth of assets by the Federal Reserve seems to provide more direct reason to think that the Federal Reserve could provide funds to finance government spending. If the Federal Reserve can decide to finance purchases of over $3 trillion of U.S. government bonds and private mortgage-backed securities, why not finance some spending on health care this way? The Federal Reserve could purchase another $10 trillion or so of government bonds and there could be more health care and apparently little other effect.

Despite it seeming so, the Federal Reserve cannot purchase an unlimited amount of government securities with no effect. The demand for excess reserves is the basis for the Fed’s ability to increase reserves. There are two major sources of this demand currently and both are limited.

U.S. branches of foreign — primarily European — banks are holding large amount of reserves in the Federal Reserve. Reserves at the Federal Reserve can satisfy regulatory requirements that these foreign banks hold liquid assets. It might seem more natural for banks to satisfy these liquidity requirements with reserves at their own central bank, the European Central Bank (ECB), but the Federal Reserve is paying interest on reserves and the ECB is charging interest on reserves. As of March 2019, the Federal Reserve pays 2.40 percent on reserves and the ECB charges 40 basis points. That is a difference of 2.80 percentage points per year. There is exchange risk due to fluctuations of the value of a dollar relative to a euro, but banks apparently are willing to bear that risk to collect the interest differential. This demand clearly has its limits though because banks are required to a minimum amount of liquidity relative to their assets, not an unlimited amount.

In addition, U.S. banks hold large amounts of excess reserves. These are held for two reasons. One is to satisfy the regulations concerning liquidity. In addition, since 2008, excess reserves typically have paid more than very short-term Treasury securities. Reserves can be a good risk-free investment. If there were nothing limiting U.S. banks’ demand for excess reserves, it would in fact be true that the demand for reserves would be virtually unlimited. U.S. banks, though, are required to hold equity capital against their assets. A major part of this requirement is called the leverage ratio and currently is 5 percent of assets for a bank to be considered well-capitalized. The leverage ratio is higher for the largest banks that are regarded as most important for the financial system, called “globally systemically important banks” (G-SIBs). Acquiring more reserves increases a bank’s assets and eventually requires increasing equity capital. While the increase in capital is a fraction of a bank’s increase in excess reserves, the low interest rate on reserves guarantees that a bank will not increase its assets without limit

In short, the demand for reserves is limited, notwithstanding appearances to the contrary after quantitative easing. Continuing expenses on benefits such as government health care can be funded by increasing reserves at the Federal Reserve only by inflation such as suffered by Venezuela and Zimbabwe. Few would think these countries are good role models for the United States.

Gerald P. Dwyer


Gerald P. Dwyer is a Professor and BB&T Scholar at Clemson University. From 1997 to 2012, he served as Director of the Center for Financial Innovation and Stability and Vice President at the Federal Reserve Bank of Atlanta. Dwyer’s research has appeared in leading economics and finance journals, as well as publications by the Federal Reserve Banks of Atlanta and St. Louis. He serves on the editorial boards of the Journal of Financial Stability, Economic Inquiry, and Finance Research Letters. He is a past President and member of the Executive Committee of the Association of Private Enterprise Education. He is also a founding member of the Society for Nonlinear Dynamics and Econometrics, an organization for which he served as President and Treasurer.

Dwyer earned his Ph.D. in Economics at the University of Chicago, his M.A. in Economics at the University of Tennessee, and his B.B.A. in Business, Government, and Society at the University of Washington.

Get notified of new articles from Gerald P. Dwyer and AIER.