• Core CPI was warmer than consensus, matching Scotia’s estimate...
  • ...but CPI and PPI translate into tame core PCE for a third straight month
  • Still, the Fed is likely to hike next week…
  • ...mainly because Warsh and the hawks boxed themselves in to markets
 
  • US CPI / core CPI, m/m %, SA, August:
  • Actual: 0.4 / 0.3
  • Scotia: 0.4 / 0.3
  • Consensus: 0.4 / 0.2
  • Prior: 0.1 / 0.2
  • US CPI / core CPI, y/y %, August:
  • Actual: 3.4 / 2.4
  • Scotia: 3.4 / 2.5
  • Consensus: 3.4 / 2.4
  • Prior: 3.4 / 2.5

US CPI inflation landed on Scotia’s house expectations in terms of headline inflation (0.4% m/m) and core CPI (0.3%). Core was above consensus at 0.29% m/m SA and 3.5% m/m SAAR (chart 1). I’ll explain why it shouldn’t matter to the Fed but probably will.

The Fed is likely to hike 25bps next week even though there remains a solid case for whiffing. Today’s CPI numbers and yesterday’s producer prices continue to point to tamer core PCE inflation over recent months after doing the proper conversions.

So why hike? The strongest argument is that Chair Warsh has probably boxed himself in with his high deference to markets. He is likely to take the roughly 21bps pricing and deliver on a 25bps hike and with enough support on the Committee. To do otherwise would curiously ease financial conditions and go against his hawkish-sounding narrative; if you don’t hike when it’s priced, then when? Yet I believe that hiking would be policy error.

Key is how these figures translate into what we used to understand as the Fed’s preferred inflation metric: core PCE. Our math shows core PCE inflation likely to land at 0.2% m/m SA when we get the figures which unfortunately won’t be until the end of the month.

That would mean core PCE inflation running at a 3moMA of 0.2% m/m SA (2.4% m/m SAAR) with gains of 0.15% m/m SA in June, 0.25% in July and 0.2% in August. It’s cooling.

Chart 3 shows the math in converting from core CPI to core PCE that takes account of differing weights on key components. There is no single smoking gun that says core PCE will come in lower than core CPI as opposed to the additive effects across the categories, but about half of the -0.1% effect of translating weights comes from housing.

Chart 4 shows the effects of pertinent categories in producer prices and how they translate into core PCE by having no influence this time. Chart 5 combines the effects.

Charts 6–22 show trends in key parts of the CPI basket. Charts 23–26 break down August’s numbers in y/y and m/m percentages, both unweighted and weighted contributions.

Also see my morning note for views on relative versus generalized inflation. Barring clandestine methods, there will be no reaction from the FOMC in blackout. As it hikes, the Fed will strongly risk becoming pre-midterm cannon fodder.