August CPI Pushes the Fed Towards Can’t Not Hike Territory
- After some early summer softness (World Cup whipsaws, residual seasonality issues) August’s inflation data was not dovishly accommodating and likely hot enough to push the Fed to a September hike (> the Waller test). Market pricing at 85-90% also now makes the risks of a dovish surprise likely too much in the current backdrop, even if some officials still may not want to hike.
- Core CPI came in above expectations with core at 0.29%. It was the second hottest print of the year for CPI core services ex housing and core CPI ex shelter and used autos.
- The topline print was not catastrophically hot and had some more dovish features one could point to, particularly the narrowness of the services impulse. The basic message remains one of corporate pricing power in the face of serial supply shocks amid solid demand growth and a stable labor market. There were distinct shocks higher in communication services, airfares, and hotels but we’re well past the point of idiosyncratic or exclusionary logic being particularly persuasive, esp. when some of those same series recent softness was pointed to in a dovish way earlier in the summer.
- The market reaction seemed to suggest that the data being just hot enough to induce a Fed hike(s), may well be a good thing when we’ve been worrying about excessively accommodative or inflationary policy choices. The recent move higher in oil prices has built on the pressures higher from the AI boom and robust consumer spending.
After a few brief months of deflation, core goods ex used auto returned to mildly positive inflationary territory again. Most forecasts seem to embed an eventual steady return to mild deflation here. That seems overly optimistic given the global manufacturing cycle’s upturn, the impacts of the war (which are still building for many goods even before the recent ramp higher in diesel prices), and the inefficiencies and disruptions brought on by tariffs. I do not expect particular heat over time here but if that embedded deflationary assumption doesn’t hold it amounts to either a bit of a persistent forecast miss or, if strict 2% is the goal, a pressure which must otherwise be countered by housing or core services ex housing.
The largest ever jump in communications services prices was a clear one-off shock higher that accounted for much of the upside surprise in core services ex housing. This will be dropped from the trimmed measures when they are released later today. Airfares and hotel prices also moved notably higher after a few months of dovish signaling. There was too much dovish extrapolation of these downward surprises amid World Cup and jet fuel price whipsaws and now that volatility moved higher again. In a broader sense this is decent evidence that amid fairly strong consumption growth (particularly in nominal terms) and a global cycle that’s broadening out, the skews to inflationary shocks remain positive. Judging by the inflation and aggregate margin data firms (staples and some particularly tariff hit retailers perhaps a bit of an exception) seem to have pricing power.
Shelter inflation seems to be slowly troughing. Amid the noise brought into the data by last year’s government shutdown, shelter’s disinflationary impulse has slowed notably. While it may keep drifting down a bit more, the upturn in marginal rent indicators suggests that this disinflationary impulse is wrapping up.
Domestic discretionary services inflation (dining out, hotels, concerts, etc) was the most benign looking part of the report. From a forward-looking perspective this is a good thing and suggests that while inflation remains too hot, we are at least so far not seeing substantial additional inflation from the most discretionary and cyclical parts of the economy. The overall pace of inflation here remains somewhat too hot but it’s absence of a post-war reacceleration is so far encouraging for eventual disinflationary trends, even if that trend may be somewhat above target without some policy restrictiveness.
