• Core CPI landed on the screws at 0.3…
  • ...but pandemic-era seasonal adjustments are overstating core inflation…
  • ...lending more support toward the Fed cutting next week
 
  • US CPI, m/m % change, November, SA:
  • Actual: 0.3 / 0.3
  • Scotia: 0.3 / 0.3
  • Consensus: 0.3 / 0.3
  • Prior: 0.2 / 0.3

US core CPI inflation landed on the screws, was likely weaker than reported if not for ongoing distortions to seasonal adjustment factors, but all summed up reinforces expectations for a quarter point cut from the Fed one week from today.

Headline and core CPI inflation was 0.31% m/m in November, matching Scotia’s estimate and the median consensus estimate while higher than the minority of forecasters expected. It has landed at this rate for four months in a row (chart 1).

Chart 1: US Core CPI Inflation

Before turning to details we need to consider the role of seasonal adjustment factors. Core CPI was likely lower than stated if not for an ongoing bias in the SA factors. It's no coincidence that the five biggest SA factors for core CPI have all been in the pandemic era (chart 2). SA factors are calculated with a recency bias that is slanted toward recent years. They always have been and there are lots of papers about this online that predate the pandemic. But it's more problematic when the recent period covers wild shutdowns and reopenings that throw all the SA factors into a period of turmoil.

Chart 2: Comparing US Core CPI SA Factors for All Months of November

This issue matters a lot. At any of the pre-pandemic SA factors, November’s core CPI would have been between 0% m/m and 0.2 with generous rounding up (chart 3). In plain English, SA factors are distorting core inflation to be too high of late.

Chart 3: US Core CPI Scenarios

The problem is that the Fed can't have its cake and eat it too. Powell can't turn more dovish and cut 50 in September when SA factors were underestimating inflation over the summer and then be dovish when SA factors are overestimating it. Powell in July had it right when he said you need to smooth the figures. Powell since then has gotten it wrong imo. SA factor issues overstated inflation at the start of the year, understate is in the summer months and then overstate it again.

Across components, shelter is showing some improvement. It was up 0.3% m/m SA, but primary rent was up by just 0.2 and so was owners’ equivalent rent. Maybe the Fed's long awaited disinflation from housing as market rents push through the inflation figures is finally coming to fruition.

But what Powell also once emphasized is that this housing disinflation story may be offset by core services stickiness. Core service prices (ex-housing and energy services) were up 4.2% m/m SAAR last month and remain sticky at elevated levels.

Among other components, gas prices were up by 0.6% m/m SA, core goods prices were up 0.3%, food away from home was up by 0.3% which is still warm, groceries (aka food at home) were up 0.4% which is hotter, and clothing prices were unspectacular at 0.2%.

Based on CPI components I would tentatively lean toward 0.2 for core PCE pending PPI that we get tomorrow. We won’t get PCE figures until next Friday, two days after the FOMC, but they’ll have a good idea of the number ahead of time through their own estimates.

More charts are available on the following pages. Also please see the detailed table showing components, other measures, and charts.

Chart 4: US Goods Inflation; Chart 5: US CPI Core Services Ex-Housing; Chart 6: Housing Inflation; Chart 7: US Rent Inflation
Chart 8: New vs Used Vehicle Inflation; Chart 9: US CPI: Gasoline; Chart 10: US Food Prices; Chart 11: US Airfare
Table: US Inflation Component Breakdown
Table: US Inflation Component Breakdown
Table: US Inflation Component Breakdown