Fed Speakers Counter Market’s Reaction to US CPI

• US CPI inflation lands on the screws…

• ...with very little breadth in key services inflation...

• ...driving a dovish market reaction…

• ...while Fed officials continue to guide further hikes including 25bps next time

• The uncertain path forward keeps inflation risk slanted upward

• US CPI, m/m % SA // y/y %, December:
• Actual: -0.1 / 6.5
• Scotia: 0.0 / 6.5
• Consensus: -0.1 / 6.5
• Prior: 0.1 / 7.1

• US core CPI, m/m % SA // y/y %, December:
  
• Actual: 0.3 / 5.7
• Scotia: 0.4 / 5.8
• Consensus: 0.3 / 5.7
• Prior: 0.2 / 6.0

US CPI landed on the screws at -0.1% m/m SA for headline inflation and +0.3% m/m for core CPI which on both counts was just a tick beneath my estimates and further below nowcasts. The annualized m/m rate of change in core CPI accelerated to 3.7% from 2.4% in November which puts the three-month moving average at 3.1% (chart 1). That is cooler than earlier periods but still persistently too warm for the Fed’s liking and that may be what the Fed will emphasize especially as evidenced in Powell's speech in November and his communications in December. Breadth, however, was generally rather soft. Forward-looking risks probably remain more skewed toward upside than downside risk.

MARKET REACTION

Markets reacted favourably as the two-year Treasury yield fell by about 3–4bps post-data, fed funds futures shaved 3–4bps off of pricing for the February 1^st meeting that is now basically pricing a 25bps hike and also reduced terminal rate pricing by just 1–2bps with still another 50+bps of tightening priced that would raise the fed funds upper limit to about 5% versus the FOMC’s 5.25% median estimate.

FED REACTION—STICKING TO THEIR GUNS

We’ve already heard from Philly’s Harker (voting) who said he leans toward downshifting to a 25bps pace and a terminal rate a little above 5%, so presumably he’s one of the 5¼% dots. Bullard said he still favours getting above 5% “as soon as possible.” and that there is too much optimism in markets that inflation will fall. Barkin next (12:40pmET, nonvoting).

HOW LOW CAN IT GO?

Where to from here for CPI? Chart 2 shows three scenarios for the year-over-year rate. The lowest scenario is derived by only allowing year-ago base effects to work through while keeping the CPI index level unchanged on a monthly basis over the next year; it shows how much of a deceleration would be driven by shifting year-ago comparisons. The next lowest line comes from combining this shift in year-ago base effects with average monthly seasonality in price pressures drawn from before the pandemic; it shows that inflation decelerates but bottoms at just under 2% y/y. The third line is our Scotia Economics forecast for CPI that expects it to end the year at 3.8% y/y (consensus 3%) by forecasting greater trend pressures than average seasonality would suggest.

There is a lot of uncertainty around all CPI views going forward, but I still think Powell will see more upside risk than downside risk to inflation going forward given evolving developments. One is how tight the job market is and what that may mean to real wage pressures going forward and his connection to core service prices defined as within PCE (not CPI) and by stripping out shelter and energy. Two is that pandemic developments might exert renewed pressure on supply chains and upward risk to prices via China’s abrupt ending of Covid Zero policies and the XBB.1.5 variant in the US. Markets and some economists may therefore be more focused upon the latest inflation print than the Fed may be as it sticks to its tightening path and leaning against easing bets.

DETAILS—LACK OF BREADTH EITHER UP OR DOWN

Core goods price inflation was a little less weak last month than the prior month, but prices still fell (chart 3). Core services (ex– energy services) inflation picked up a touch but was flat ex-shelter. 

Charts 4 and 5 (next pages) illustrate unweighted m/m changes in CPI components and their weighted contributions to m/m CPI respectively. It was basically a matter of housing versus gasoline with a few other small effects that overall point to narrow breadth in weighted movements.

Charts 6 and 7 do the same thing for year-over-year rates of inflation by component.

Among individual components, shelter was up 0.8 m/m with rent of primary residence at 0.8% and OER at 0.8%. There is so far zippo cooling evidence in terms of housing with all of that still ahead of us later in the year and gradually so (see the Global Week Ahead for how the lags work). Also recall that PCE attaches half the weight to shelter that CPI does.

Food was up 0.3% led by 'away from home' as at home was up 0.2 Chart 8. Gasoline fell 9.4% and shaved 0.4 ppts off of headline CPI SA.

Medical care services were up 0.1 and ditto for medical care commodities.

Airfare was down 3.1% m/m and has fallen in six of the past seven months (chart 9). Lodging was up by 1.5% m/m, car rentals fell by 1.6%.

New and used vehicle prices had minimal effects in weighted terms (chart 10). 

Apparel prices were up by 0.5% m/m SA which at a 2.4% weight meant nothing in terms of weighted contributions to CPI (chart 11).

Please also see the accompanying detailed table including micro-charts and z-scores of deviations from historical norms for price changes.