“We believe the Federal Reserve's large-scale asset purchase plan (so-called "quantitative easing") should be reconsidered and discontinued. […] The planned asset purchases risk currency debasement and inflation, and we do not think they will achieve the Fed's objective of promoting employment.”
Letter from a group of 24 economists to Fed Chair Ben Bernanke, November 15, 2010.
A very simple version of 1960s monetarism has two elements. First, controlling money growth is necessary and sufficient to control inflation. Second, leaving aside a financial crisis, the monetary base―the sum of currency in circulation and commercial bank deposits at the central bank―determines the quantity of money. Putting those together means that, in order to control inflation, all central bankers need to do is ensure that their liabilities grow at the appropriate rate. Conversely, when the central bank’s balance sheet grows quickly, inflation inevitably follows.
This simple monetarist reasoning was still on display in 2010, when Ben Bernanke received the letter from which we quote. At that stage, the Federal Reserve’s assets exceeded 250% of their level in September 2008. Over just over two years, the Fed had purchased roughly $400 billion in Treasury securities and $1 trillion in federally guaranteed mortgage-backed securities. But, as Bernanke explained at the time, the purpose of these asset purchases was to aid the economy in recovering from the crisis-induced recession. Moreover, in contrast to prior norms, since October 2008 the Fed had been paying interest on reserves, raising the opportunity cost for banks to lend.
Subsequent experience proved the letter writers very wrong. The Fed’s balance sheet continued to grow, peaking at $4.5 trillion in early 2015. And, over the decade just ended, inflation (measured by the Fed’s preferred consumption expenditures price index) averaged 1.6%―below the central bank’s long-run goal of 2%. If anything, in recent years, and despite massive central bank balance sheet expansions, inflation both in the United States and in other advanced economies has been too low, not too high.
With central bank balance sheets now surging again, we recount this history in the hopes of blunting any inflation concerns, which we see as profoundly misguided. Over the six weeks ending April 22, the Fed’s assets have grown by the same amount as they did from September 2008 to March 2013. While this does raise some serious concerns, inflation is not high among them.
Explaining this conclusion requires us to explore the two relationships that form the foundation for the simple monetarist framework. First, how do changes in the monetary base influence the quantity of money? Second, how do changes in money influence prices? It turns out that neither relationship is stable, despite the decades-long quest by many to show otherwise. (For a survey, see here.)
To see the problem with using these relationships as the basis for policy actions, we can look at some data, starting with the evolution of the ratio of the quantity of money to the monetary base. (We used M2, a broad monetary aggregate that includes time deposits, but the results are similar with M1, a narrower measure.) This ratio—known as the M2 money multiplier—rose through the 1960s and 1970s, peaking in the mid-1980s. We can think of the money multiplier as indicating the willingness and ability of private banks to transform the central bank’s base money into the transactions medium that households and businesses routinely use.
With changes in financial regulation and the structure of the financial system―especially the rise of mutual funds offering daily redemption―conventional bank account balances fell through late 1980s and early 1990s. Then, as the Fed increased the supply of reserves following the September 2008 Lehman bankruptcy, the U.S. money multiplier plunged. At the height of the financial crisis, banks were content to increase their reserve holdings and they were loath to make new loans. Their loss of ability or willingness to lend as reserves surged short-circuited the usual mechanism that links base money creation to deposit creation. More recently, prior to the start of the COVID-19 pandemic, banks accumulated reserves to satisfy post-crisis liquidity requirements (see here). So, our first point is that massive increases in the monetary base need not lead to large increases in the quantity of money. And, over the last decade they have not. (Likely reflecting the sudden widespread takedown of pre-existing credit commitments as nonfinancial firms scrambled for cash, both M1 and M2 increased notably in recent weeks.)
M2 money multiplier (monthly and weekly), 1959-April 15, 2020
Note: The data are monthly through February 2020 and weekly thereafter. Source: FRED.
Second, with inflation at low levels since the mid-1980s, standard monetary aggregates (like M1 or M2) no longer exhibit any link to inflation. The following chart plots monthly data on inflation and M2 growth since 1962. The vertical axis is the 12-month percentage change in the personal consumption expenditures price index excluding food and energy—a measure of core (or trend) inflation. The horizontal axis is the 12-month growth rate in M2 lagged 24 months. The orange dots are for the period 1962 to 1984, the gray dots for 1985 to 2009, and the red triangles for 2010 to 2020.
Inflation and M2 money growth (monthly), 1962-March 2020
Source: FRED.
During the early period, when inflation temporarily rose above 10 percent, there was a positive relationship: increases in money growth were associated with higher inflation two years later. For the most recent decade, the period when the Fed’s balance sheet exploded, the two are nearly uncorrelated (see the red dashed line). In other words, at low levels of inflation, inflation appears to be unrelated to money growth. (While we do not show it here, another implication of this chart is that the velocity of money—the ratio of nominal GDP to money—is unstable.)
Third, there is virtually no relationship between the size of the Federal Reserve’s balance sheet and long-run inflation expectations. The following chart shows the evolution of five-year ahead five-year inflation expectations derived from financial markets (the measure is the difference between the five-year-forward five-year rates on a standard 10-year Treasury note and on a 10-year Treasury note with inflation protection). For reference, we include the Federal Reserve’s total assets in red (on the right scale). The challenge for policymakers is clear from this chart: compared to their target of 2%, long-term inflation expectations are too low, not too high!
Long-term inflation expectations and the Federal Reserve’s assets (monthly), 2003-April 22, 2020
Source: FRED. The final observations are for April 22, 2020.
Whether a simple monetarist framework ever provided reliable predictions is a subject for another debate. To be fair, in 1961, at the height of the theoretical and empirical battle, Milton Friedman famously noted “that monetary actions affect economic conditions only after a lag that is both long and variable.” Yet, at least since the 1980s, changes in the financial system have undermined simple policy prescriptions based on conventional monetary aggregates, narrow or broad.
To conclude, we share the Fed’s view that, if inflation is a concern today, the reason is that it has been below the FOMC’s target for years. Similarly, as we wrote in a recent post, the size of the Fed’s balance is not a problem per se. To be sure, there is one caveat: as Olivier Blanchard notes, should the U.S. government command the issuance of Federal Reserve liabilities to meet its funding needs—a case of fiscal dominance―then inflation becomes inevitable.
Consequently, so long as the Fed retains control of its balance sheet, our principal concern is with the central bank’s new role in providing credit directly to the nonfinancial sector. The new facilities to purchase municipal and corporate bonds from the issuers, as well as the ones to provide loans directly to firms, are turning the Fed into a state bank that directs credit to favored borrowers. Having taken over large swaths of the intermediation system, giving up these emergency tools is going to be very difficult.
We believe this is something that Congress could still remedy, but unfortunately, politicians seem to have little will to do so. The looming challenge for the Fed is not inflation; it is that by obligating the Fed to play the role of a state bank, Congress is putting the central bank’s monetary policy independence at great risk. In the end, the consequences for the U.S. economy of a Federal Reserve that is subservient to political interests could be just as severe as the inflation that the 2010 letter writers mistakenly feared.