The Costless Disinflation of 2022-24

Global inflation rose in 2021 and continued to rise well into 2022. The FRED chart below illustrates this rise in inflation for the U.S., euro area, United Kingdom and Canada. In the U.S., for example, personal consumption expenditures (PCE) price index inflation began rising in February 2021, peaking at a year-over-year level of 7.1% in June 2022, well above the Federal Open Market Committee’s (FOMC) 2% target for that inflation metric.

NOTE: The FRED chart illustrates inflation rates for the U.S., euro area, U.K. and Canada from 1970 through early 2024.

Analysts generally attributed such increases to factors related to economic consequences of the COVID-19 pandemic, including disruptions in global supply chains and labor markets, as well as fiscal and monetary policies designed to ameliorate the economic effects of pandemic-related job losses and business shutdowns.

To reduce inflationary pressures in the U.S. and pursue its price stability mandate, the FOMC began tightening monetary policy—raising the federal funds rate target range—in March 2022. The Federal Reserve was not alone. Governments commonly direct their central banks to stabilize prices, so many central banks also raised short-term interest rates to reduce inflation.

Reductions in trend inflation are called disinflations. These episodes have been fairly common, particularly in the 1970s and 1980s. The FRED chart above shows that inflation rose in the U.S., U.K. and Canada during the mid- and late-1970s before central banks in those countries began to reduce it again. Significant recessions—output losses and rising unemployment—accompanied these disinflations, however.

In 2022, during the start of the disinflation, two sets of prominent economists publicly discussed these potential costs. Olivier Blanchard, Alex Domash and Lawrence Summers very reasonably noted that substantial unemployment has historically accompanied large disinflations and that the then-current disinflation was likely to follow this pattern.See their July 2022 policy brief, “Bad News for the Fed from the Beveridge Space,” published by the Peterson Institute for International Economics. Andrew Figura and Fed Gov. Christopher Waller responded by suggesting that more sanguine effects on the labor market were possible, depending on the then-current properties of the labor market.See their July 29, 2022, FEDS Notes article, “What Does the Beveridge Curve Tell Us about the Likelihood of a Soft Landing?”

In fact, the U.S. economy did not experience large output losses or increases in unemployment during the 2022-24 disinflation, which began after inflation peaked in June 2022. This unusual behavior raises questions: What typically happens during a disinflation? How unusual was the U.S. experience in 2022-23—when the country saw the sharpest drops in monthly inflation—compared with historical patterns or other countries’ experiences during their recent episodes?

To answer these questions, I looked at monthly consumer price index (CPI) and unemployment data from the International Monetary Fund (IMF) for 25 Organization for Economic Cooperation and Development (OECD) countries from 1970 through early 2024.I used CPI rather than the Fed’s preferred PCE inflation metric to measure inflation consistently across countries. The OECD is a group of 38 democratic countries with market-oriented economies. I excluded 13 countries from the sample either because their price data was missing from the IMF dataset or because their maximal inflation was greater than 100%, suggesting that their experiences might be misleading. The countries excluded for high inflation were Chile, Costa Rica, Estonia, Iceland, Israel, Latvia, Lithuania, Mexico, Poland, Slovenia and Turkey, while Australia and New Zealand were excluded for lack of data. During this sample period, there were 105 episodes among these countries in which the underlying trend of inflation declined by at least 4 percentage points. Of these 105 episodes, 94 started prior to 2022 and 11 started in 2022 or later.I estimated the underlying inflation trend with a 13-month moving average that smooths out the wiggles in inflation. The results would be much the same without this smoothing.

To meaningfully average data from countries with different levels of inflation, I subtracted the value of the inflation level at the start of the disinflation from each data series around an episode. This means that the normalized inflation rate for all 105 disinflations will be zero at the start of the disinflations.

The following figure shows the cross-country average of inflation for the episodes beginning before 2022 (red line) and those that start on or after January 2022 (blue line), as well as for the U.S. experience in 2022-23. The fact that the blue line is below the red line in the figure shows that average inflation fell a bit faster and further in 2022-23 than in past disinflations, while U.S. inflation (black line) fell a bit more slowly than in past disinflations.

Change in Average CPI Inflation before and during Disinflation Episodes

SOURCES: International Monetary Fund’s International Financial Statistics and author’s calculations.

NOTES: The figure illustrates the cross-sectional averages of CPI inflation for 94 disinflation episodes prior to 2022 and 11 episodes that start on or after January 2022. The episodes are defined as reductions in trend inflation for 25 OECD countries, from 1970 through early 2024. The vertical axis shows percentage point changes in inflation around the value at the start of the disinflation. The horizontal axis shows months around the start of the disinflation, which is month zero. For each episode, the inflation rates are normalized by subtracting the value of the inflation rate at the start of the disinflation. Therefore, all inflation rates will equal zero at the start of the disinflation by construction.

The next figure shows cross-country average unemployment rates before and during the same two sets of disinflations. While in past disinflations (red line) average unemployment typically rose by nearly two percentage points over the 18 months following the start of the episode, the average did not rise at all during the most recent disinflation (blue line). This latter effect is peculiar, as there are many examples of central banks producing undesired recessions as collateral damage of disinflations.

Change in Average Unemployment Rates before and during Disinflation Episodes

SOURCES: International Monetary Fund’s International Financial Statistics and author’s calculations.

NOTES: The figure illustrates the cross-sectional averages of unemployment for 94 disinflation episodes prior to 2022 and 11 episodes that start on or after January 2022. The episodes are defined as reductions in trend inflation for 25 OECD countries, from 1970 through early 2024. The vertical axis shows percentage point changes in unemployment around the value at the start of the disinflation. The horizontal axis shows months around the start of the disinflation, which is month zero. For each episode, the unemployment rates are normalized by subtracting the value of the unemployment rate at the start of the disinflation. Therefore, all unemployment rates will equal zero at the start of the disinflation by construction.

Determining why unemployment didn’t increase during the 2022-23 disinflation would require much more research. It could be due to some combination of the source of the original inflationary shock, continuing fiscal stimulus during the disinflation, the credibility of OECD central banks in pursuing disinflation, and/or some other factor.

In examining these figures, we should be careful to remember that they just show how variables tend to move during disinflations. We should be careful in using correlations to infer causal relationships. For example, people carry umbrellas on rainy days, but that doesn’t mean umbrellas cause rain.

Notes

1. See their July 2022 policy brief, “Bad News for the Fed from the Beveridge Space,” published by the Peterson Institute for International Economics.
2. See their July 29, 2022, FEDS Notes article, “What Does the Beveridge Curve Tell Us about the Likelihood of a Soft Landing?”
3. I used CPI rather than the Fed’s preferred PCE inflation metric to measure inflation consistently across countries. The OECD is a group of 38 democratic countries with market-oriented economies. I excluded 13 countries from the sample either because their price data was missing from the IMF dataset or because their maximal inflation was greater than 100%, suggesting that their experiences might be misleading. The countries excluded for high inflation were Chile, Costa Rica, Estonia, Iceland, Israel, Latvia, Lithuania, Mexico, Poland, Slovenia and Turkey, while Australia and New Zealand were excluded for lack of data.
4. I estimated the underlying inflation trend with a 13-month moving average that smooths out the wiggles in inflation. The results would be much the same without this smoothing.