Release the lions and the tariffs. Withold the food stamps.
There is little good news in the CPI metrics for August, but the numbers also did not surprise to the upside either. Energy was again just plain ugly, and the trends in Sept in Energy are offering ample cause to expect Sept CPI to be more biased to the upside. The economy steady and demand is likely to allow for more cost pass-throughs in price.
With a 3.4% headline CPI flat to July and Core CPI a tick lower from 2.5% to 2.4%, the high level numbers were little changed. The Sept noise in oil and downstream refined products are seeing the backdrop get worse – not better – with the Houthis now potentially taking the Saudi East-West pipeline out of action and potentially shutting down or severely disrupting the Red Sea oil channels. The Houthi actions could undermine a wider range of product flow across those shipping lanes. As we go to print, the Bab al-Mandab Strait is vulnerable to a new “choke point.”
The unfolding Canada trade war does not look to get better and could feed unit cost pressures across a wide range of US consumer and supplier chain markets. Steel, aluminum, and various manufactured products are feeling tariffs (buyer pays, no matter what Trump says). Canada does not appear inclined to leverage its natural resource edge to retaliate in that trade mix. The relative need for Canadian aluminum is high in the US even if not for many Canadian steel products. The desire to keep peace with the provinces (notably Alberta and Saskatchewan) could keep Canada from more aggressive export tax alternatives in retaliation. US suppliers will have higher pricing power now in the domestic market.
The above time series plots the headline CPI and Core CPI from the COVID plunge into the supply-demand imbalance year of 2021 and then into the Russian-Ukraine oil spike of 2022. Inflation spiked to a 9.1% headline CPI in June 2022 and 6.5% Core CPI in Feb 2022. That was edged out by the 6.6% Core of Sept 2022.
We see the 3.4% down from the 4.2% in May and the Core CPI on Aug 2024 back down to the Sept 2024 lows. The threat is still how tariffs flow into price structure and working capital management cycles. Next week’s FOMC decision shows high odds of tightening in FedWatch. The probability of a hike is now 86.5%.
In the aftermath of Iran, there has been no hiding from the direct inflationary effects of oil prices and the even more notable spikes in downstream products with a crisis underway in diesel, painful aviation fuel trends, the household cash flow hit from gasoline, and the wide and varied effects across raw materials, power costs, and freight and logistics fallout among the most obvious secondary and tertiary effects.
We already addressed some of the leading indicator price challenges in PPI yesterday (see PPI August 2026: Leading Indicator Signaling Struggling 9-10-26). We look at the data in charts further below, but the +16.3% YoY in Energy, the +28.0% in “Energy commodities,” and +27.4% for gasoline are obvious pressure points. The MoM for Energy of +2.1% follows two negative MoM months (-1.5% July, -5.7% June) with Sept already starting off badly.
The progress made in July and June often revolved around jawboning by Trump around empty threats of peace breaking out. Short-term trading opportunities abounded for friends and family.
The above chart updates the historical time series for fed funds vs. headline CPI. The latest headline of 3.4% for August is above Biden’s last full month, which was below the 3% line at 2.9%. The chart includes the long-term medians, which underscore that the current compression of CPI and fed funds is “not normal.”
The above chart frames the fed funds vs. CPI differential. “Fed funds minus CPI” is now +0.2% after it had turned negative in May at -0.6% or dramatically different than the long-term median of +0.6%. In its simplest terms, that relationship looks more like accommodation.
With FedWatch odds at over 86% for a tightening next week, we did not see much reaction in the 10Y UST to today’s CPI number after the 10Y got close to the Oct 2025 highs of 4.99% this week. We saw around 4.95% this week, but as we go to print the 10Y UST is 4.97%. That is bad news for mortgage rates also with the Mortgage News Daily survey rate just under 7.1% at 7.08%.
The customized special aggregate CPI indexes posted in Table 3 of the BLS release are troubling in historical context. Those detailed above are our favorites, but there is a long list worth reviewing.
Note: We had to be selective in the months included in the table in order to show more of the timeline. May 2026 (not shown) was an especially challenging month (see CPI May 2026: The 4% Rubicon 6-10-26).
“All items less shelter” has been our top pick across time as we have cited in prior research since “Shelter CPI” comes with too many asterisks and derived numbers that do not fit the typical household reality (or even close).
The chart shows “Services” still stubborn above the 3% line in what is a “services economy.” The Services line comprises 63.6% of the CPI index. Table 3 also drives home the stark differential between the Durables CPI (-0.3%) at a 10.6% weighting in the CPI and the much larger Nondurables line at +5.8% CPI for 25.8% of the CPI index. The ability (or desire) to pass through tariff costs has still not hit Durables CPI yet in a meaningful way.
The 3.6% for the “All items less shelter” metric in August is better than the 4.1% in April and the 4.7% of May with the energy spike. Even at the 3% handle now, that metric is still a material contrast with the sub-2% levels of 2024 including 1.9% in Dec 2024 (Biden’s last month). In mid-2023, the level was +0.7%. The FOMC was doing its job the right way by then, and the payroll additions of those 2023 days crushed what was posted on average in 2026.
The above table details our Big 5 subsectors for CPI. These roll up to around 75% of the CPI index, so overall this mix of 5 broad categories is the main event even if there are plenty of line items outside this group that matter to consumers.
We will not give much space to our long-held view on the low value of the lines that are “derived” and inconsistent with the household “checkbook experience” and household cash flow. That is notably the case with shelter and the steep deflationary number of -8.5% for health insurance.
Some of those lines speak for themselves at a time when such items as ACA premiums have soared and the deductible offerings are shifting the mix in the wrong direction. Too many people have had to drop coverage or see deductibles spike to find “affordable” health care. There are times when the measurement theories of the BLS econ teams do not work with the real-world consumer experience.
The total energy bucket of +2.1% MoM in August reverses the -1.5% MoM of July and is worse than the -5.7% energy relief of June but is a long way from the +10.9% MoM in March after the onset of the Iran War. The “Energy commodities” line was at +21.3% MoM in March and had set off a chain reaction with lag times.
Energy YoY metrics moved down sharply from +23.5% in May to +15.7% in June and +14.7% in July. The Energy CPI YoY line is back at +16.3% in August. Within Energy, the gasoline CPI of +24.6% in July is ugly but beats the +40.5% gasoline CPI of May. August CPI reports +27.4% YoY with the Strait of Hormuz “shut” again and the Red Sea now also in stress as noted in the bullets. The Iran War remains in chronic uncertainty on how long and how damaging it will be with such an erratic approach to policy. Handicapping odds of escalation or any other scenario is a mug’s game at this point.
We have to say that Hegseth (Trump to a lesser extent) highlighting the Iran war in his 9/11 speech at the Pentagon was a bit misleading in historical comparisons with 15 of the 19 terrorists on the planes from Saudi Arabia and zero from Iran (2 from UAE, 1 from Lebanon, 1 from Egypt). As someone living across the street on 9/11, the White House owes the public facts. Clearly, the Saudis and UAE have other virtues from Trump’s perspective ($$$). Confidence in foreign policy judgment in the Middle East is low when every statement has a domestic agenda to manipulate history.
As we go to print, WTI is over $100 per bbl vs. around $82 in mid-July and Brent is near $105 in contrast to the $88 at this time in last month’s CPI write-up. WTI had a $68 handle in the first week of July. That was a short reprieve.
Housing and healthcare anomalies…
The higher costs of housing are uglier when including financing and the monthly payment pain. CPI metrics do not capture that financing cost effect. Mortgages and the cost of financing do not get factored into the product line CPI. The UST curve and mortgages and auto financing have clearly been going in the wrong direction since 2025 and now well into 2026. 2027 is not looking any better and may well get worse.
With many households crushed by health care premiums (notably the ACA), the consumer is certainly not “feeling the deflation” in health insurance seen in the table at -8.5%. Those that struggle with health care coverage thus get held hostage to the rising costs of services as broken out. The 5.5% for “hospital and related services” is high by any measure with 5.4% for the outpatient subset. In other words, if you lose coverage, that is what awaits you with prices high in absolute terms. Trump’s lack of effort in health care is one of the worst consumer headwinds and notably in the lower part of “the K.”
The above table updates some of the lines near and dear to households. We see a split decision with 3 higher and 3 lower YoY. Airline fares at +23.4% in August are heavily tied to jet fuel costs. These lines are a mixed picture with divergences across lines within these buckets. We will see how much fallout comes from the fare spike in the peak travel season when airlines report Sept quarter earnings.
The Apparel CPI ticked lower from 3.9% to 3.6% after reaching 4.8% in May. Trump looked to inflict pain on the low-cost Asian countries, and we are surprised it is not higher. Trump lost on IEEPA, and that had heavily targeted Asia. Trump is now using the Section 301 “forced labor” strategy to raise tariffs on essentially all trade partners.
More importantly, Trump is now turning to a more flexible provision of the Smoot Hawley Act of 1930 (Section 338) that allows him free reign with no Congressional check. Trump (in theory) can play tariff dictator by using existing legislation on the books. That was not the case with IEEPA. When in doubt, turn to Hoover for sound economic policies. Right? That went so well after 1930.
See also:
Existing Home Sales Aug 2026: Headwinds Still Driving the Story 9-10-26
PPI August 2026: Leading Indicator Signaling Struggling 9-10-26
Weekly Tech Check: Semis vs. Software Push-Pull 9-7-26
Weekly Returns: Benchmarks and ETFs 9-7-26
Duration Daze: Post-ZIRP Bond Returns 9-7-26
UST Market: Curve Meets Knuckler 9-7-26
Payroll Aug 2026: Trouncing Low Expectations is “Positive” with an Asterisk 9-5-26
Market Commentary: The Curve 8-30-26
PCE July 2026: Mixed Consumption, Stubborn Inflation 8-26-26
2Q26 GDP: 2nd Estimate 8-26-26