Disinflation Noise vs. Persistent Signal
Trying to parse the available information to get a sense of where we are.
While the broadest measures of inflation have slowed sharply since last summer—and may well continue to decelerate in the months ahead as lease renewals are incorporated into rental price indices—the best estimates of underlying inflation remain several percentage points faster than the Federal Reserve’s 2% yearly goal. Rather than returning to pre-pandemic norms, inflation seems to be converging to a new, faster baseline of around 4-5%. That would help explain the willingness of many Fed officials to look through the recent ructions in the banking sector and signal openness to further increases in short-term interest rates.
The Analytical Challenge of Figuring Out What’s Actually Going On
The majority of the inflation that Americans experienced over the past two years can be attributed to the disruptions of the pandemic and, to a lesser extent, Russia’s most recent invasion of Ukraine. Dramatic changes in what we wanted to buy collided with significant impairments in our ability to produce, leading to shortages of some items, gluts of others, and massive swings in relative prices.
Businesses and consumers adapted to these challenges remarkably well, all things considered, while governments around the world attempted to minimize the risk of unnecessary financial distress by disbursing trillions of dollars to support incomes. As mass vaccinations enabled economies to reopen in the first half of 2021, many of the changes that had been necessary in 2020 were reversed, leading to another round of disruption and relative price shocks as businesses, workers, and consumers once again adjusted. Modest inflation—the level of the Consumer Price Index (CPI) was about 8-10% higher in April 2023 than what Fed officials in 2019 would have preferred—was unpleasant, but it facilitated these repeated adjustments and helped prevent far worse outcomes.
In the first year the pandemic, inflation was slower than the Fed’s goal, although not necessarily by so much as to be concerning by itself. Once the economy began to reopen thanks to mass vaccination, inflation began to accelerate. At first, this seemed to simply reflect the return to prior trends, as prices that had been depressed rapidly normalized. Then from September 2021-June 2022, the CPI rose faster than 10% a year, while the Personal Consumption Expenditures (PCE) Price Index preferred by the Fed rose about 8% a year. Over the past few months, both CPI and PCE inflation have come in around 3.5% annualized—faster than the Fed’s goal, but not very much.
Throughout all this, the challenge has been distinguishing between price increases that were clearly attributable to temporary disruptions from longer-term changes in inflation. The risk was always that those sources of price increases (or decreases) could persist even after normalcy had otherwise been restored.
In the beginning, it was more or less impossible for anyone to make this distinction. As I put it in September 2020:
Inflation indicates whether there is too much—or too little—money and credit relative to society’s productive capacity. But inflation is a useful signal only if the mix of goods and services people consume doesn’t change that much from month to month or year to year. After all, changes in the price of any individual good or service (or asset, for that matter) move around all the time for a wide range of reasons that have nothing to do with the state of the broader economy. From this perspective, it’s clear that the coronavirus pandemic has made the aggregate inflation data mostly useless.
As I noted then, the longer-term inflation outlook would ultimately depend on the permanence and magnitude of the post-pandemic adjustment:
If everything goes back to normal in a few months thanks to a vaccine, these shifts could quickly reverse…If, however, the virus leads to longer-term changes in how we work and consume, labor and capital would have to get reallocated across the economy. Workers would have to learn new skills, which at least in the short term would cause productivity to fall. Many businesses would have to close, and even those that survived would be stuck with at least some assets worth far less than in the pre-pandemic world. At the same time, businesses would need to make new investments to meet demand for a decidedly different mix of goods and services than existed before. In that scenario, a bit of inflation would be helpful. Broadly rising prices and wages would make it easier to endure losses on bad assets without going into bankruptcy. Higher prices would also help encourage needed capital spending and get workers to move to new industries and to new places.
As is often the case, we ended up in some intermediate state between those two hypothetical scenarios. That has made it harder to determine how much of what we were experiencing was a one-off change in the overall price level, as opposed to a persistent change in the underlying trend of price increases.
In many respects, the pandemic has not led to persistent changes in consumer behavior, although it took a while for that to become fully apparent. The surge and collapse in the stock price of Zoom Video Communications—currently just below where it was in January 2020—is perhaps the most obvious demonstration of this. The stock peaked in October 2020, just before news of effective vaccines came out, but the vertiginous descent did not really begin until the end of 2021. Other changes have been longer-lasting, such as the shift in consumer spending from services to durable goods.
Given all this, the question is: how can we cut through the noise to get a sense of where inflation is relative to the Fed’s goal (or any other target)?
I do not claim to have a definitive answer, but I have relied thus far on two complementary approaches:
Strip out the specific components from the broader price indices that reflect idiosyncratic factors, so that whatever remains is a better indicator of the underlying trend
Look at changes in nominal spending power (wealth+income+credit), how those sources of financing are translating into actual spending, and how much that spending might deviate from some rough estimate of real productive potential
Both approaches imply that America’s underlying inflationary trend is currently about 2 percentage points faster than before the pandemic—little changed since last summer.