Terrified Shelley Duvall in lobby card for the film ‘The Shining’, 1980. (Photo by Warner … [+]
On June 10, 2021 (tomorrow, as of this writing), at precisely 8:30 in the morning, a data “bomb” will go off – and the financial media will go crazy.
The Bureau of Labor Statistics will release the Consumer Price Index (CPI) for May.
The “headline number” (as it is called) will provoke dozens of… well, alarming headlines. There will be plenty of op-ed pieces, much cluck-clucking and hand-wringing. Experts will make dire predictions. They will excoriate the Federal Reserve. Congressmen will go on record to tell us they told us so. Investors will be advised to prepare for the end of happy days.
What Is Real
We know, or should know, with virtual certainty, that:
- The headline number will be high
- It will be wrong
The CPI is notoriously inaccurate, almost always on the high side. But due to special circumstances related to the impact of the pandemic on the economy, the distortion this month will be greatly magnified. The headline “inflation” figure we will see in tomorrow’s news will be overstated by something like half its true value.
Economists know this. They have known for decades that the headline CPI distorts the true picture of price trends in the economy. The effects of this inaccuracy are very significant. Because it is tied to cost-of-living adjustments for many entitlement programs (e.g., Social Security), a Congressional Commission 25 years ago found that the bias in the CPI (which they called “overindexing”) blows up the federal deficit. The study estimated this bloat at more than $1 trillion over a dozen years. Nevertheless, in the furor that will follow the announcement tomorrow, many commentators will fail to remember this. Or, should we say, they will fail in their duty as journalists to understand and explain it properly.
It is likely that the May CPI increase will be reported at well over 4% year-over-year. It may be as high as 5% – which will surely cause many pundits to lose their composure. Expect to hear that “your father’s inflation” (pre-Volcker) is just around the corner, that it will destroy the bond market, blow up the stock market (your investments!), cause the dollar to lose its reserve currency status, and speed the decline of the American empire.
We will all be forced to endure another month of inflation “panic porn.”
The Distortions, A Brief Recap
The sources of these distortions are laid out in detail in a previous column to which the reader may refer for charts and numbers. Here we will summarize.
1. Excess volatility in certain components
It has long been recognized that some components of the index – e.g., food and energy – are highly volatile, and are normally removed from the CPI to produce what is called the Core CPI. Over the long term, the CPI and the Core CPI average out to about the same number, but the CPI is several times more volatile, which makes it unsuitable for many policy-making questions (eg., what should the Fed do now?). Lately the CPI has been running much higher than the Core, and it will be higher again tomorrow. (In April, the Core CPI increase was 2.96% compared to 4.15% for the unadjusted CPI.)
2. The Fed Prefers a Different Measure, Which is Always Lower
“Inflation” is not a well-defined concept. There are many ways to define and measure price changes in the economy and to construct a metric to which the label of “inflation” may be attached. But we should not be fooled into thinking that it is as clear and definitive a metric as, say, the temperature of the water in the Potomac River today.
In fact, there are a great many measures of “inflation” – including at least a dozen of them which are “official,” in the sense that they are employed in designing and managing economic and monetary policy. (A future column will inventory the most prominent of these.)
The Federal Reserve decided over 20 years ago that there was a much better measure than the CPI – which is called the Personal Consumption Expenditure Index (PCE). It is prepared not by the Dept of Labor, but by the Bureau of Economic Analysis at the Department of Commerce. It is considered superior and in a future column I will go into the reasons why.
The PCE is almost always lower than the CPI. In April, the CPI was 4.15% and the PCE was 3.6%. The core PCE (same concept as the Core CPI) was 3.1%.
3. The Base Effect
This is a big problem for the year-over-year comparisons. The Base Effect occurs when the year-ago numbers deviated in an extraordinary fashion from the trend line. That skews the 12-month comparison.
As it happens, May is going to be the month with the biggest Base Effect – nearly 1.4% will be added to the headline number, by my calculations.
May Base Effect
This is a problem that every economist and every statistician understands. It is often misunderstood by headline writers, and if it finds its way into the articles tomorrow it will often be buried in paragraph 18.
Summary
There are other problems with the CPI, which have to do with the composition of the “basket” of goods that underly the index. There are decisions about weightings, and substitutions, and how to handle statistical “outliers.” For example, the price of used cars jumped 21% year-over-year in April, relative to very depressed prices in the Spring of 2020 (when driven miles dropped by more than 40% overnight, which eviscerated the used car market). Used cars are a relatively highly-weighted component of the CPI – but the April increase cannot reasonably be seen as anything but a bottleneck effect.
For all these reasons, the May CPI number will overstate the actual inflation trend. It will be at least 1.5-2% too high.
More important than the short-term skew is whether this monthly jump means anything for the longer term.
It does not.
The base effect will start to correct in June and will be gone by September.
The bottlenecks will start to clear, as supply rises to meet demand. A year ago, there were shortages of toilet paper, flour, cleaning supplies. You couldn’t find hand sanitizer to save your life (pardon the over-apt expression). Now there are barrels of hand sanitizer in every aisle at the supermarket, and prices reflect a glut.
If we keep our eye on the PCE and the Core measures, we will have a better sense of the true pricing trends.
The Federal Reserve is correct to be patient, and to avoid reacting to a short-term anomaly.
