CPI Index Far Superior to PCE

PCE inflation as reported is highly distorted. Over 30% of the PCE index uses imputed prices vs. market prices, which creates massive distortions such as portfolio management services increasing 21% over the last 12 months. In addition, the BEA imputes a cost of checking accounts without interest that shows an increase in the cost of checking accounts when interest rates rise. Consequently, if the Fed raises interest rates, financial services inflation would rise not fall.

Also, by including a large swath of consumption not paid by consumers, such as government and employer health care payments, the weighting of other services becomes distorted. The net result is an index that does not reflect what consumers actually pay and more importantly an index that the Fed has less control over. Shelter, which is the primary sector the Fed controls, is only approximately 15% of PCE vs. 35% of CPI, and healthcare, which is driven by secular factors vs. cyclical, is over 20% vs. only 10% in CPI.

The Fed’s rationale for using the PCE makes no sense. They cite the fact that PCE is a chain weighted index, which could be superior as it constantly updates weightings of the index to reflect consumer responses to price changes. The problem with that rationale is that with 30% of the weightings imputed, there is no consumer response to those imputed prices, and the initial weightings are so distorted by imputed consumption that making minor adjustments over time is irrelevant.

CPI core is currently tracking at 2.6% Y/Y which is in line with the Fed’s 2% target. Our CPI-R index is tracking at less than 1% using real time pricing for shelter and financial services. PCE Y/Y is tracking at 3.3%, but that number is heavily distorted by the deeply flawed financial services calculation which estimates over 9% inflation in financial services and uses the 2-year delayed shelter component. We publish PCE-R that uses real-time prices instead of imputed prices and that index is tracking in line with the Fed’s target of 2%.

The Fed should change its preferred inflation gauge to CPI core from PCE as that index tracks what consumers actually pay for vs. the PCE’s imputed consumption tracking. In addition, Fed policy almost exclusively impacts shelter and other interest sensitive goods and has little impact on secular increases in healthcare services prices.

We continue to believe that a rate increase at this time would be a historic mistake in line with the “transitory” policy error of 2021. Fed policy is currently very tight with housing and construction in recession due to the Fed Funds rate being .75-1% over the neutral rate. CPI Core is in line with the Fed’s 2% arbitrary target and over 1% below target if adjusted for the inaccurate shelter and financial services components.

The Fed should tighten rates when excessive monetary growth causes inflation in interest rate sensitive sectors such as housing and capital goods as was occurring in 2021 with the money supply exploding by 70%, the Fed Funds rate 3% below neutral and housing prices rising over 20% Y/Y. Now we have the opposite situation with the money supply shrinking by over 5% Y/Y, the Fed Funds rate up to 1% above neutral and housing prices flat Y/Y.