September 16, 2022
The August Consumer Price Index (CPI) came in higher than many of us expected. While the overall inflation rate for the month was just 0.1 percent, the core index rose 0.6 percent, up from a 0.3 percent increase in July. Food prices also had another sharp jump, rising 0.7 percent, pushing the year over year increase in store bought food to 13.5 percent. This was not a good inflation story.
The biggest items in the core story were a large jump in new vehicle prices, which rose 0.8 percent in August, and high inflation in rents. Inflation in the rental indexes had moderated slightly in July, with the inflation rate for both indexes edging down by 0.1 percentage points compared to the June rate. But both indexes showed 0.7 percent inflation in August, with the number for owners’ equivalent rent being 0.1 percentage point higher than the July rate.
Most of the other data in the report were also not encouraging. The medical services index rose 0.8 percent, driven largely by a 2.4 percent rise in the health insurance index. The household furnishings and supply index rose 1.1 percent, continuing a pattern of sharp price increases in a category that typically showed almost no inflation before the pandemic.
There were some price declines in the August report. Used car prices fell a modest 0.1 percent, but that was after a 0.4 percent drop in July. Even with the recent drop used car prices were still 7.8 percent above their year ago level and more than 50 percent higher than their pre-pandemic level. Appliance prices fell 1.2 percent, after a 0.6 percent drop in July. This is a category where it seems the supply chain problems have been resolved, but prices are still 13.9 percent higher than before the pandemic.
The Good News
It is hard to feel good about the course of future inflation based on the August CPI report. Fortunately, we got somewhat better news the next two days with the Producer Price Indexes (PPI) report and the Import/Export Price Indexes Report. Both of these reports showed widespread price declines in both overall and core measures.
The overall August PPI fell 0.1 percent in August. This follows a decline of 0.4 percent in July. The core index also showed inflation to be tame. The August rise was 0.2 percent, following increases of 0.1 percent in July and 0.3 percent in June.
There was higher inflation in the PPI earlier in the year, but the PPI typically leads the CPI by several months. This means that the slower inflation in the PPI likely means lower inflation in the CPI in the months ahead.
It is worth noting that rents do not appear in the PPI. Rent is a major factor in both the overall and core CPI, accounting for almost 31.0 percent of the overall index and nearly 40.0 percent of the core index. Rental inflation is likely to remain high for the rest of the year, although private indexes show the rate of rental inflation is slowing. This means that we are likely to see high inflation in the CPI for the next several months, even if most components show slow inflation or even falling prices.
It is important to remember that much of the story of rental inflation is that a much larger share of the workforce is now working from home. We had been seeing rents and house sale prices outpace inflation for about five years before the pandemic.
The story here is that after having a glut of housing as a result of a building boom during the housing bubble, we seriously underbuilt housing for a dozen years. Starts were running at an annual rate of more than 2.2 million units at the start of 2006. They then fell to under 600,000 in the Great Recession. We didn’t cross 1 million units again until 2015. Construction continued to inch up, but we were still under a 1.4 million unit rate when the pandemic hit.
As a result, there was a severe shortage of housing when millions of people suddenly decided that they needed bigger homes, and often in different places, since they were now working from home. They also had the money to pay for larger homes. The stimulus checks helped, as did extraordinarily low mortgage rates, but these families were saving thousands of dollars a year in commuting and other expenses related to working in an office. (These savings do not show up in the CPI.)
Anyhow, this rush to get bigger homes sent house prices soaring, with prices rising nationally by more than 30 percent, and considerably more in many areas. The Fed’s rate hikes have slowed these increases and led to declines in many areas. It will take some time for the full impact to be felt, in part because there is a long lead time between when a house sale is contracted and the house is sold and the sale price shows up in our data. But there is little doubt that the big pandemic price surge has come to an end.
However, it will take longer to address the underlying shortage of housing. Construction did rise sharply during the pandemic, with starts peaking at over 1.8 million this spring, before Fed rate hikes sent them sharply lower.
The picture on housing completions looks better. Even though starts had soared, completions had remained under 1.4 million. This was due to supply chain problems that prevented builders from finishing homes. These are now being resolved and completions are increasing in spite of the drop in starts.
A higher rate of completions will help, but it will still take some time to restore something resembling balance in the housing market. During this period, we are likely to see rental inflation continue to outpace the overall rate of inflation regardless of the state of the overall economy, even if it comes down from its current rate.
The Export-Import Price Indexes
The data on export and import prices was released on Wednesday. This also gave us some good news on inflation. The overall index dropped 1.0 percent in August following a 1.5 percent drop in July. This was not just the good news on gas and oil prices, the non-fuel index fell 0.2 percent, its fourth consecutive monthly decline.
The drop in import prices is also likely to show up in consumer prices, especially in areas like apparel and furniture, where most of demand is met by imports. While the CPI for apparel has increased only slightly more than the import price index, the furniture index has risen by 23.1 percent since the start of the pandemic, compared to 11.3 percent for the import price index. There is a similar story with motor vehicles, with the index for new vehicles in the CPI rising by 18.1 percent compared to 5.9 percent for the price index for imported vehicles.
The items in these indexes don’t correspond exactly, and there are other factors in the retail prices here, but it is hard to imagine the sort of discrepancies between the CPI price increases and the import price indexes persisting for long. In short, this is another factor pointing to good news in the months ahead on inflation.
The Labor Market
At the end of the day, the Fed’s rate hike agenda is focused on the labor market more than anything else. The story is simple, if we have persistently rapid wage growth, there is no way to avoid high inflation. To take the recent numbers, if the average hourly wage is rising at a 5.0 percent annual rate, then we are almost certainly looking at sustained inflation in a 3-4 percent range, well above the Fed’s 2.0 percent target.
The Fed’s rate hikes are aimed at slowing the rate of wage growth by weakening the labor market and reducing workers’ bargaining power. In other words, they want to increase the unemployment rate.
While the labor market was almost certainly too strong, earlier in the recovery, it is less clearly the case now. The most immediate measure of labor market strength is the number of weekly unemployment insurance claims. Weekly claims had fallen to under 170,000 at the start of April, a level not seen since the late 1960s, when the labor force was less than half its current size.
There is a good case that a labor market this strong is going to create major inflationary pressures, which cannot be sustained through time without seeing a serious inflationary spiral. However, weekly claims did rise sharply in the spring and summer, hitting a peak of 260,000 in mid-July. They have again edged back down, but the recent numbers of around 220,000 a week are above the levels seen in many weeks in 2019, before the pandemic.
On the other side, the number of continuing claims, which measures the people who remain unemployed and getting benefits, is still extraordinarily low. Just over 1.4 million workers were collecting benefits in the most recent week, this is roughly 180,000 less than the lowest figures before the pandemic.
This indicates that people who do lose a job are able to quickly find a new job, an indication of a very strong labor market. On the other hand, the fact that employers are willing to lay off workers at a more or less normal rate, indicates that they are not so stressed for workers that they keep workers on the payroll even when they have no immediate need for their labor.
In the same vein, the length of the average workweek has fallen back to its pre-pandemic level. The length of the workweek averaged 34.4 hours in 2019. It rose in the pandemic as employers struggled to get workers and had their existing workforce put in more hours. It peaked at 35.0 hours in January of 2021. If that sounds like a small increase, consider that it is an increase of 1.7 percent, the equivalent of more than 2.5 million jobs.
The average workweek stood at 34.5 hours in August, the same as for many months in 2018 and 2019. Most employers are not now seeing a need to get their workers to put in unusually long hours.
The share of unemployment due to voluntary quits, a measure of workers’ confidence in their labor market prospects, rose to 15.2 percent in August. This is slightly higher than peaks hit in 2000 and 2019, but these data are erratic. The share stood at just 12.8 percent as recently as April. If the figure remains above 15.0 percent, it would definitely be an indication of a very strong labor market, but we should not make too much of a single month’s data.
The data series that the inflation hawks point to as suggesting a seriously overheated labor is the job vacancy rate. This is at record highs (the series only goes back to December of 2000), with the July reading at 6.9 percent. That compares to a pre-pandemic peak of 4.8 percent.
This has to raise serious concerns about an excessively strong labor market, but even here the picture is not unambiguous. The vacancy rate peaked at 7.3 percent in March, so we have seen a substantial decline over four months that was not associated with any increase in the unemployment rate.
In some sectors the decline has been considerably more rapid. In hotels and restaurants, the vacancy rate dropped from 10.9 percent last August to 8.9 percent in July. In retail, the rate peaked at 7.1 percent last August, but now sits at 5.8 percent, below pre-pandemic peaks.
This raises the possibility that the extraordinarily high vacancy rates we are now seeing may fall back to more normal levels without a big jump in unemployment. They may also not have as clear a relationship with wage growth as they have in prior decades.
At the end of the day, inflation will depend on the pace of wage growth. We actually got good news on that front in the August employment report. The average hourly wage increased at just a 0.3 percent rate for the month, that would translate to a 3.8 percent annual rate, a pace not far out of bounds for the Fed’s 2.0 percent target inflation.
As I always point out, the monthly data are erratic and also subject to large revisions, so we can’t make much of the August number. However, we can say that wage growth is not accelerating from the pace seen at the end of 2021, as would be predicted by the record high vacancy rate. In fact, it has slowed slightly.
If wage growth has not been following the predicted pattern for the last year, it seems a stretch to insist it will follow it going forward. This could mean that we will see wage growth fall back to a pace consistent with more acceptable rates of inflation without a big jump in the unemployment rate.
Does the Fed Have to Keep Going Big?
It is virtually certain that the Fed will again raise rates at its meeting next week. The only question is by how much, with the consensus seeming to be on another 75 basis points hike, with some expecting 100 basis points.
The Fed must be concerned about inflation, but it also has a responsibility to maximize employment. Chair Powell has taken this responsibility far more seriously than any of his predecessors over the last eight decades. There are clearly forces pushing inflation lower, as the supply chain problems get resolved. We don’t know the extent to which they will tame inflation without further action from the Fed.
We also know that the full effect of past rate hikes has not been felt yet. There has been a massive whack to the housing market, which is clear in a variety of indicators. This is spilling over to other sectors, but we are just beginning to see the secondary impact. Higher interest rates will also have an impact over time on many other sectors. There is a serious cost to the Fed’s rate hikes.
The data on inflation expectations continue to show that they are headed lower rather than higher. This means that, at least for now, there is little basis for fearing a wage-price spiral driven by self-fulfilling expectations of higher inflation.
As was the case before this week’s data, the overall story on inflation remains mixed. There are real grounds for concern, but also evidence that inflation is headed downward without the need for sharp increases in the unemployment rate. In this situation, the Fed would be best advised to raise rates with caution.