Why the Fed Should Not Increase Rates

After Removing a Major Flaw in the CPI Index, August Core CPI Inflation was just 2.1%

Summary: As I wrote in previous posts here, here, and here, there is a logical flaw in the methodology used to construct the Consumer Price Index that distorts the inflation picture. In spite of the fact that the CPI is advertised as a measure of the out-of-pocket cost of living, it contains a wholly made-up concoction called Owners' Equivalent Rent (OER) that makes up 1/4 of the All Items CPI and 1/3 of the Core CPI. OER is a made-up number that purports to measure the hypothetical amount that I, as a homeowner, would have to pay every month to rent my own house from myself. It is a terrible idea.

OER is a classic example of economists at the BLS confusing stocks (the value of the existing housing stock) with flows (payments made to other people for goods and services produced this year.) The fact that I own my home means I have already purchased and stockpiled a lifetime of home services-I will never have to pay for housing services again. The fact that I will pay other costs associated with the house, e.g.,the cost of insuring and maintaining the house over time-is already included elsewhere in the index.

OER enters the CPI index with a weight equal to 26.9% of the total index even though the amount actually paid as rent by people who write a check to an actual landlord every month is just 7.7%. The appropriate weight for OER is zero.

OER should be completely removed from the index. As I show below, doing so would correct August Core CPI to just 2.1%, virtually at the Fed's 2% target. Given the difficulty of getting a clear picture of the price level due to the spike in oil prices caused by the ongoing Iran war and the impact of AI capital spending on construction costs and communications equipment, if I were running the Fed, I would delay any action on interest rates until we have better visibility on demand-driven inflation, the only kind of inflation the Fed can actually control.

Base Policy on What We Do Know, Not What We Don't Know

There are two reasons why the Fed should not raise rates at this time. First, there is so much noise in the data caused by irregular events that it is not possible to accurately know how much of current inflation is caused by rising demand, which is the Fed's job to control, and how much is caused by supply factors, like the shutdown of oil shipments on both sides of the Arabian peninsula, gyrations in tariffs, the Ukraine war, the el niño in the Eastern Pacific, and the massive buildout of AI infrastructure that has produced spikes in construction costs and the prices of all products that contain semiconductors.

More simply, you should base policy on things that you do know, not on things that you don't know. And right now, the Fed doesn't know enough to make a reasoned decision.

Second, once we correct for the OER bias I described above, the inflation numbers are pretty close to the Fed's announced 2% target. In the graphic, above, I have calculated what both All Items CPI and Core CPI would have been in August after removing the distorting influence of OER. As it turns out, August Core CPI X OER was just 2.1%.

The table, above, contains the data I used to make the calculations for each of the past three months. The calculations are easy. All Items CPI Inflation is just the amount that I due to OER (the OER weight times OER inflation) and the amount that is not due to OER. (I know, duh!) And the same is true for Core CPI after adjusting for the fact that OER represents a larger share (32.8%) of the Core CPI index (25.9%) than it does for the All Items Index. You can do the rest of the math yourself.

After removing OER, Core CPI inflation in June, July, and August was 2.2%, 2.2%, and 2.1%. All are low enough to make a case for keeping rates right where they are for the time being. And before you ask, I know that the Fed puts more weight on PCE inflation than on the CPI. I will post the corrected PCE numbers when the new PCE report comes out next week.

Before I do, here are two things to keep in mind about the PCE. First, the weight attached to OER is smaller in the PCE (about 12%) than in the CPI (about 26%) so the impact of the correction will be somewhat smaller. Second, you need to know that the dreaded OER issue has also polluted the Personal Income and Personal Consumption Expenditure data too as well as all of the numbers in the GDP report. That means the PCE report will overstate expenditures by the amount of imaginary rent built into OER and overstate Personal Income, along with Disposable Income and Per. Capita Income by the amount of the imaginary rent collected in OER. More on that in a later report.

Finally, I realize that there is all sorts of political and market pressure on the FOMC. President Trump is demanding lower rates. The bond market and the financial press are demanding higher rates. Neither is a good reason to make an important policy decision.

Follow the numbers. Leave the rates right where they are.

Dr. John