June 10, 2021
The consumer price index (CPI) rose 0.6 percent in May, with the core index (which excludes food and energy) rising at an even more rapid 0.7 percent. This brought the increases in the overall and core index over the last year to 5.0 and 3.8 percent, respectively.
Does this mean the inflation hawks were right? Did Biden’s recovery package throw too much wood on the fire and is now setting off an inflationary spiral?
A little closer look at the numbers indicates that caution is still advised. First of all, prices plunged last April and May of last year, as the economy was shutting down in response to the pandemic. This means that the year-over-year comparison is not very informative.
To get a more honest evaluation, we should look at the rate of change of prices since February of 2020, before the pandemic was having an impact. Using this as a base, the overall CPI has increased at a 3.0 percent annual rate, while the core has increased at a 2.6 percent annual rate.
The 2.6 percent rate is somewhat above the Fed’s 2.0 percent target, but the Fed targets the personal consumption expenditure deflator (PCE), not the CPI. For coverage and methodology reasons, the core PCE is generally 0.2-0.3 percentage points lower than the core CPI.
Also, as the Fed has stated explicitly, the 2.0 percent target is an average, not a ceiling. Given that inflation has consistently run well below 2.0 percent, to maintain a 2.0 percent average the inflation rate has to be above 2.0 percent on occasion, so there is nothing in the data to indicate that we have a problem, accepting the Fed’s target.
The Used Car Crisis
In fact, even this above 2.0 percent inflation figure can be a bit deceptive. Used car prices have soared in recent months, rising 7.3 percent in May and 10.0 percent in April. (These are monthly increases, not annual rates of increase.) If we take the period since February of 2020, used car prices have increased at a 23.2 percent annual rate.
The weight of used cars in the core CPI is 3.8 percent. This means that this jump in used car prices alone added almost 0.9 percentage points to the rate of inflation in the core index in the months since the pandemic began. That means that if we pull used cars out of the core index, it would have been increasing at just under a 1.8 percent annual rate since the pandemic began, well below the Fed’s target.
The reason for the jump in used car prices is not a mystery. The worldwide shortage of semi-conductors has slowed auto production, leading several assembly lines to shut down for a period of time. (Most are now back up and running.) The shortage of new cars led many people to turn to buying used cars, sending their prices soaring.
This shortage of cars is a problem. People need cars for transportation and rental car companies need cars to restock their fleets, which they sold off at the start of the pandemic to conserve cash. But this is clearly a temporary problem. Semi-conductor output will increase as existing plants add capacity and new ones come back on line. When that happens, we are likely to see the price of used cars return to something comparable to their pre-pandemic levels. New car prices, which have also risen rapidly, should also fall back in line with pre-pandemic trends.
If we take the car story out of the picture, there is not much of a story of run-away inflation. The prices of some items have been rising rapidly, but this is a bounce back from price declines at the start of the pandemic. Apparel prices jumped 1.2 percent in May, car insurance 0.7 percent, and air fares 7.0 percent. These indexes are respectively 2.2 percent, 0.2 percent, and 6.3 percent below the February 2020 level.
Inflation in the rent indexes remains well contained. The rent proper index rose 0.2 percent in May, while the owners equivalent rent index rose 0.3 percent. Over the last year, they are up 1.8 percent and 2.1 percent, respectively. The medical care index, which has been a major source of inflation, has risen at just a 1.7 percent annual rate since the pandemic began. The index for college tuition has risen at less than a 0.6 percent annual rate.
We will see more erratic price movements as the economy continues to reopen. There will be some spot shortages of different items and there will be cases where strong demand gives workers the bargaining power to raise their wages, but there is not a story of an inflation crisis in these data.
It is worth mentioning once again the importance of productivity growth in the inflation story. Productivity growth had been running at just a 1.0 percent annual rate in the decade before the pandemic. In the last year, it rose at a 4.1 percent rate. The data we have for the second quarter, indicates that we will see extraordinarily high productivity growth again in the current quarter.
While no one expects anything like a 4.1 percent rate of productivity growth to continue, we may well be seeing productivity growth on a faster track. Businesses were forced to find new ways to operate in the pandemic. In many cases this will lead to continuing gains in productivity. Even an increase to a 2.0 percent rate of productivity growth will hugely reduce the risks of inflation.
It is important to remember that the 1970s inflation was associated with a sharp drop in productivity growth. Productivity had been increasing at 3.0 percent annual rate in the long post-war boom from 1947 to 1973. It slowed to just over 1.0 percent from 1973 to 1980. This slowdown was a major cause of inflationary pressure during the decade. If productivity growth increases instead, it will act to alleviate inflation.
Anyhow, we will certainly need more quarters’ data before we can say anything definitive about productivity growth. The series is highly erratic and we could see sharp reversals of the recent increases. But if the growth over the last year turns out not to be fluke or due to measurement error, it’s hard to see how we can have too much to worry about with inflation.
 Not all the increases in efficiency will show up in productivity data. For example, the savings in time and commuting expenditures for people who can now work remotely, will not appear as an increase in productivity. To take another example, my wife had to have a consultation with a medical specialist half-way across the country. In the pre-pandemic era, this likely would have required a 3-day trip, with two flights each way and two nights in a hotel. Instead, she just had to put an hour aside for a Zoom meeting. This is an enormous savings in resources, but does not show up in the productivity data at all.