Correction: The first paragraph of this blog post has been updated with the correct overall consumer price index (CPI) in March 2021 of 2.6% and “core” measure of the CPI of 1.6%. The initial analysis had accidentally switched the two numbers. The numbers in Figure 1 remain the same.
Today, the Bureau of Labor Statistics (BLS) reported that the overall consumer price index (CPI) in March 2021 was 2.6% higher than in March 2020, while the “core” measure of the CPI (which excludes volatile food and energy prices) was 1.6% higher than a year ago. Given that these are noticeable (if modest) increases over recent months’ year-over-year inflation rates, some might be tempted to argue that this data should make policymakers worry about economic “overheating” stemming from “too much” fiscal support provided in recent recovery legislation. This is clearly wrong, for a number of reasons:
- The data released today do not show that prices have risen rapidly since recovery legislation passed—instead they just show that prices plummeted during the near-total shutdown of large swaths of the economy a year ago in response to the COVID-19 shock.
- Measured on an annualized basis from February 2020—before the COVID-19 economic shock—inflation in March 2021 was running at just 1.5%.
- Measured since October—shortly before the $2.8 trillion in additional relief spending provided by legislation in December 2020 and the American Rescue Plan (ARP) in March—inflation is running at an annualized rate of 1.3%.
- Non-price measures continue to show enormous degrees of economic slack, so there is no conceivable way that the U.S. economy was overheating in March 2021.
- Employment remains 8.4 million jobs below what prevailed in February 2020. If we add in growth in the working-age population, the jobs gap remains over 10 million—larger than it was during the worst parts of the Great Recession of 2008-09.
- Nominal wage growth (adjusting for composition) remains very tame. True “overheating” has to come from the labor market and this isn’t happening.
- Nominal consumption spending in the U.S. economy is slightly lower than it was a year ago, before the COVID-19 shock.
- Even in coming years, and even if growth in economy-wide spending outstrips growth in the economy’s productive capacity for a long stretch of time, damaging inflation that would require policymakers’ response is highly unlikely to appear.
- Even several years of inflation above 2% would still not make up for years of too-slow price growth over the past 12 years, and making up for these years of too-slow price growth would go a long way to reestablishing the credibility of the Federal Reserve inflation target. This is an opportunity, not a threat.
- Counterbalancing recent expansionary fiscal actions like the ARP is the steady and significant drag on spending growth imposed by the enormous rise in inequality in recent decades. As inequality has redistributed income from households that spend most of their income to households that save much more, this puts a steady anchor keeping demand growth depressed. Until this structural headwind to growth is addressed, persistent overheating is highly unlikely to happen.
- Long-lasting upward spirals of wages and price inflation that may call for action from policymakers require relatively balanced bargaining positions of workers and employers to sustain. This does not describe today’s U.S. economy and so such spirals will not happen in coming years.
Today’s data are mostly about what happened in March 2020, not March 2021
As a number of useful explainers have pointed out, the uptick in year-over-year inflation reported in today’s data is mostly a function of the sharp drop in the price level that happened a year ago as the COVID-19 economic shock hit. To show that today’s inflation numbers are not driven by “overheating” following the economic recovery packages passed in December 2020 and last month, in Figure 1 below we show the year-over-year “core” inflation measure reported by the BLS today, along with two alternative measures: an annualized rate of inflation based on price growth since February 2020, and the annualized change in prices since October 2020.
The first alternative measure (measuring inflation since February 2020) includes data from the period of the extreme COVID-19 shock, but it doesn’t use it as the base in the inflation formula. The second measure is more volatile since it’s measuring a shorter period of time, but the latest data point (all the way on the right) would pick up any inflationary pressure stemming from the surge of fiscal spending that began at the end of last year to now. As can be seen in the figure, even today’s inflation measure—the one that allows the base period to be set in the very unusual months when large swaths of the U.S. economy shut down completely in March 2020—is not really close at all to even reaching the 2-2.5% comfort zone of the Federal Reserve. (Wonky aside: the widely-reported 2% inflation target of the Fed is based on another inflation measure that consistently runs below the CPI data in this figure, so to translate their target to this figure requires raising it a bit.)
In short, today’s slight uptick in year-over-year inflation rates (but still showing inflation beneath the Fed’s 2-2.5% target) captures something unusual that happened a year ago, not anything unusual that is happening currently. This same logic will hold for the next few months of inflation data, which will be using a “base” period that includes the near-total shutdown of large swaths of the U.S. economy that happened a year ago.
Non-price measures show lots of economic slack
Inflation driven by genuine “overheating” would be accompanied by non-price measures showing very little slack in the economy—either a very low unemployment rate or high levels of domestic demand (spending by households, businesses, and governments). But, as of March 2021, there are 8.4 million fewer jobs in the U.S. economy than in February 2020. Add in the growth of the working-age population since February 2020 who will need to find employment, and the jobs gap is over 10 million—larger than what prevailed during the worst parts of the Great Recession in 2008-09.
Similarly, if one compares nominal consumption spending in February 2021 to where it was in February 2020, it is actually still lower than a year ago. It seems hard indeed to see how slightly lower nominal spending by consumers—by far the largest component of gross domestic product (GDP)—could somehow cause the economy to overheat.
Finally, overheating caused by excessive demand growth must come through a tight labor market empowering workers enough to demand fast wage increases that drive up employers’ costs and feed into price inflation. The massive jobs gap highlighted above should make us suspicious that workers feel confident and empowered enough to demand such wage increases. And, measures of wage growth that remove the distorting compositional effect of COVID-19 wage losses being concentrated overwhelmingly on low-wage workers confirm this. The Atlanta Federal Reserve Bank’s wage growth tracker looks at just workers who have remained continuously employed over the past year. For this group, there is no jump in wage growth in recent months that would signal the economy was overheating.
Inflation is not a near-future concern
The evidence above is clear that the U.S. economy is not overheating and that today’s data only reflect extraordinary events of a year ago. But is there a worry for coming years that inflation could get too high and reining it back in will be a struggle for policymakers and U.S. households? This worry is extremely slight.
For one, the danger of future damage would actually stem from not seeing inflation perk up noticeably in coming years. Since 2008, the economy has consistently undershot the 2% inflation target set by the Federal Reserve. This constant undershoot likely slowed recovery from the Great Recession—and could weigh on recovery going forward. Essentially, every year U.S. households are promised 2% inflation and it comes in less than this is a year when the real (inflation-adjusted) burden of household debt rises. Household debt has been considerably reduced relative to what prevailed pre-Great Recession, but it’s still not low in historical terms, and this effect of too-slow inflation boosting its burden on households is not trivial. The large shortfall of growth in prices since 2008 implies that restoring the credibility of the Fed’s inflation target by returning it to an average 2% since then would require 4% inflation for 3.5 years going forward, or 3% inflation for 7.5 years. A few months of modest inflation is too-little price growth, not evidence of damaging overheating.
Additionally, while many know about the large fiscal boost to aggregate demand provided in recent months by the rescue packages, not many know about the large structural headwind to demand growth that so far remains mostly unaddressed by policy: the large upward redistribution of income in recent decades. This transfer of income from households with low savings rates to higher-income households who save much more has put pronounced downward pressure on demand growth. When the effect of the fiscal rescue package fades (which will be in the relative short run, in the case of the package passed in December and the ARP), the structural headwind of inequality will remain. Until this is addressed, the prospect of damaging inflation driven by overheating over the medium term is hard to imagine.
Finally (and related), a large temporary fiscal boost is unlikely to spur a long-lasting upward spiral of wages and price inflation in an environment where U.S. workers have been so disempowered by policy over decades. The root cause of the iconic 1970s episode of inflation that has left far too large a mark on policymakers’ perception of inflationary risk was a distributional conflict between workers and capital owners. When large exogenous shocks hit (oil price inflation and a productivity slowdown in the 1970s), workers and capital owners tried to avoid bearing the burden of the shock by demanding nominal wage increases or price increases that preserved profit margins, essentially passing the economic cost of the shock back and forth.
Because a number of historically sui generis features of the early 1970s economy gave workers an unusual degree of power in protecting their wages against exogenous shocks, the exogenous shocks propagated rapidly and for a relatively prolonged period, with wages and prices chasing each other higher. Workers today do not have this power to easily wrest wage gains from employers. Evidence of that can clearly be seen in the fact that even in the late stages of the expansion following the Great Recession, with unemployment well under 4%, wage growth barely accelerated and never threatened to run ahead of economy-wide productivity growth.
For years following the Great Recession, some economists and policymakers pointed to the large expansion of the Federal Reserve balance sheet and made confident predictions that inflation was right around the corner. They were proven badly wrong. Others of us argued in real time against their predictions because we knew they failed to take the full economic context into account. This time around, people are making nearly the same mistake, assuming that a large fiscal expansion will lead to prolonged and damaging inflation. This prediction is highly likely to be proven wrong as well, for all the same reasons.