What changed
Households have now met one of the two conditions that would switch our framework to its stagflation playbook, so whether that playbook takes over comes down to a single release: August core PCE, due Sep 30. The playbook needs both halves at once: the University of Michigan’s one-year inflation expectation at or above 4.5%, and core PCE inflation running at 3.3% or faster on a three-month annualized basis.
The household half arrived with Michigan’s final September survey on Sep 25. One-year inflation expectations rose to 4.6%, clearing the line. The longer 5-10 year reading came in at 3.4%; the size of its change is not yet confirmed, but its position above the 3.0% warning line is. Consumer sentiment slid to 48.1, which headlines described as the second-lowest reading on record.
Market prices pointed the other way. VIX, the options-implied gauge of expected S&P 500 swings, fell below 15, the level we treat as a calm tape. Stock index futures gained in the night session. In Treasuries the two-year yield did not move while the long end edged higher, and crude oil fell. The table collects the moves.
| What moved | Latest | Where it sits |
|---|---|---|
| Michigan 1-year inflation expectations (Sep final) | 4.0% → 4.6% | Above the 4.5% playbook line |
| Michigan 5-10 year inflation expectations | 3.4% | Above the 3.0% warning line |
| Michigan consumer sentiment (Sep final) | 48.1 (prior 51.7) | Slightly above the 47.6 consensus |
| VIX | 14.9 (−5.1%) | Below 15, the calm threshold |
| 10-year / 30-year Treasury yield | 5.18% (+2bp) / 5.5% (+4bp) | 2-year unchanged (0bp) — only the long end rose |
| S&P 500 / Nasdaq 100 futures | +0.5% / +0.4% | Night session |
| WTI crude | $91.4 (−2.6%) | Below the first warning line at $95 |
| High-yield spread | 280bp | 70bp inside the 350bp caution line |
The new inflation alarm, in other words, came from a household survey. Stocks and oil moved toward relief, and the fear in this market lives in surveys and headlines more than in prices.
What hasn’t changed
The household survey moved, but the hard data that decides the other half of the playbook did not.
- Core PCE, three-month annualized: 3.05% — the July figure released Aug 26 is still the latest; the August update lands Sep 30.
- Payrolls 162K, unemployment 4.1% — from the Sep 4 report; the next one is due Oct 2.
- Effective fed funds rate 3.88% — the next FOMC meeting is Oct 28.
With core PCE and payrolls standing still, the only macro variable that moved is what households expect prices to do. That does not overturn our existing read. It makes Sep 30 heavier instead, because both halves must be met before the playbook changes, and only one release now stands between them.
Our read
A heavier Sep 30 still does not let us call the result early. Reflation — firm private demand, core inflation near 3% and a Fed that keeps hiking — remains our most likely scenario at 34%, but stagflation trails by only 3pp at 31%. Neither earns high conviction before the Sep 30 and Oct 2 releases.
On the surface, VIX back in its calm zone suggests the market is comfortable with a close race. Underneath, protection on individual stocks is thin and stocks and bonds are moving together.
| On the surface | Underneath |
|---|---|
| VIX 14.87 — calm zone (below 15), in steep contango against three-month VIX | CBOE Skew 144.91 — demand for far out-of-the-money puts in its elevated 140–150 band |
| Equity put/call 0.52 — the data feed labels it normal | 0.068 below the scenario-weighted floor of about 0.588 — single stocks are underhedged |
| High-yield spread 280bp and the NFCI financial conditions index at −0.555 — credit is calm, conditions loose | SPY-TLT 60-day correlation +0.477 — 0.023 from the +0.50 stagflation line |
| Stock index futures higher | Equal-weight ratio (RSP/SPY) 0.2737, a new low · small-cap ratio (IWM/SPY) 0.3656 |
Put simply, the index and the credit market are calm, but the stocks beneath the index and the usual stock-bond cushion have already weakened.
Calm VIX is the first thing to question. With spot VIX below three-month VIX — a steep contango — the market is pricing quiet now and risk later. That “later” starts on Sep 30, when core PCE with its annual revision, the GDP update, quarter-end and MU earnings all land on the same day. Move statistics built from single events likely understate a day with four catalysts stacked together.
History offers a counterweight. When Skew sat in its current 140–150 band (2021–2026), the S&P 500 was up 1.3% on average 30 days later. Those same windows carried an average maximum drawdown of −3.3%, so a dip before the gain was the common path.
As long as VIX stays low, the layer of protection beneath the index stays thin. Meeting a stacked event day in that state means accounts holding single stocks without hedges take the first hit. A slow grind lower hurts them most, because the index protection others already bought only pays out on a bigger move, so no one is forced to cover.
The more useful question today is how household inflation expectations travel through the Fed and the bond market into stocks after Sep 30, rather than what VIX has priced.
Households and markets already disagree on inflation. The gap between Michigan’s 5-10 year expectation and market breakevens measures about 1.1 percentage points: markets trust the Fed, households do not. That gap is why stagflation holds near 30% instead of sliding into the low 20s.
Rising household expectations strengthen the case for further hikes, and FedWatch already puts October hike odds at 66.6%. The problem is that stocks and bonds are moving together, with their 60-day correlation sitting just under the stagflation line. A hot inflation print could then push both lower at once instead of bonds cushioning stocks.
The counter-evidence comes from how the three-month rate is calculated. The Sep 30 release drops May’s hot +0.36% from the window, so even a 0.3% monthly print would pull the three-month annualized rate down to about 2.8%. Reaching 3.3% takes a monthly reading of roughly 0.42% or more. The caveat is the annual revision released the same day, which could reset the May–July base and change that arithmetic.
Equities carry their own rebuttal. One-month S&P 500 earnings-per-share revisions are +3.0%, so the case that earnings are outrunning rates is still alive.
So even with the household half met, the bar for the other half is higher than it looks. Yet the assumption that bonds cushion stock declines is what breaks hardest on the day that bar is cleared. The nominal equity risk premium — stocks’ earnings yield minus the 10-year yield — is just +0.03 percentage points, leaving almost no room if rates rise further. With three-month bills paying 4.07%, holding the index is a bet on future earnings growth rather than on what stocks earn today.
Options positioning
- Institutional flow: net premium was negative, but the signal strength was neutral, and the largest unusual trades were institutional in size without a clear direction.
- Index downside protection: demand is concentrated in far out-of-the-money index puts, while protection on single stocks is thin.
- Dealer positioning: dealers’ hedging amplified moves in SPY in 7 of 10 sessions and in QQQ in half of them; the data carries no date stamp, so we treat it as supporting context only.
Key dates
The same Sep 30 that brings core PCE also brings the GDP third estimate with its annual update, and MU earnings are reported for that day as well.
| Date | Event | Why it matters |
|---|---|---|
| Sep 30 (Wed) | August core PCE price index | A monthly reading of about 0.42% or more lifts the three-month annualized rate to 3.3% or above, raising stagflation to 41% and completing both halves of the playbook alongside the 4.6% household reading. A reading of 0.20% or below adds 4pp to goldilocks |
| Sep 30 (Wed) | Q2 GDP third estimate and annual update | A downward revision alongside sticky core PCE is the stagflation combination. A changed May–July base also changes the core PCE arithmetic |
| Sep 30 (date not confirmed) | MU earnings | EPS consensus is $32.32. SOXX holding above $578.51 afterward would revive the demand-led reflation case |
| Oct 2 (Fri) | Nonfarm payrolls | Payrolls below 50K, unemployment at 4.3% or higher and a four-week claims average at 240K or higher together move 10pp toward recession if core PCE’s three-month rate is below 3.0%, otherwise toward stagflation. Above 100K with unemployment at 4.1% or lower revives demand-led reflation |
| Oct 28 (Wed) | FOMC | FedWatch: hike 66.6%, hold 33.4%. Household expectations that keep rising strengthen the case for a hike |
- What to watch: whether August core PCE on Sep 30 prints above 0.42% month over month
What it means for investors
Ahead of Sep 30 core PCE, we judge whether any setup deserves new weight by how many independent checks point the same way.
| Setup | Independent checks in support | Grade | Status |
|---|---|---|---|
| QQQ upside | 0/6 on the formal strategy check (an alternative count of 5/6 conflicts, so the stricter reading applies) | Low | No signal |
| SPY upside | 2/7 (29%) | Low | No signal |
| SPY downside-protection overlay | 5/7 (71%) | Medium | Monitoring — a defensive tool, not a directional call |
| Weaker long bonds | 3/7 (43%) | Medium | Below the halfway bar |
| Options flow | Neutral | — | No directional evidence |
We call a setup a fat pitch when most independent checks line up behind it; that is rare by design, and none reaches the bar today. Our overall risk grade is calm (GREEN), which permits taking risk but does not call for adding it.
- Dollar-cost-averaging index investors — scheduled contributions have no reason to change on today’s data. For a lump sum, the Sep 30 core PCE print and the Oct 2 payrolls report arrive within days, and a cautious reader would frame the timing around them.
- Concentrated and semiconductor-heavy holders — MU earnings are the swing factor. SOXX, the semiconductor ETF, carried the market’s leadership, but Friday’s gain came on 65% of average volume, and a China-Taiwan flare-up would hit semiconductors first. A cautious reader in this group would see how SOXX handles the results before reading much into that leadership.
- Multi-asset and allocation-minded investors — for readers who want inflation protection, TIPS are the asset tied to the survey-versus-breakeven gap behind today’s household reading, though rising real yields hurt them too, which is why they rank behind short-term bills in our framework. The assumption that long bonds cushion equity drawdowns does not hold while the two move together.
- Short-term and leveraged traders — SPY is in a squeeze, with its Bollinger bands inside the Keltner channel, so a break either way can be large. If dealers’ hedging keeps amplifying moves, leveraged products absorb that amplification in full, and adding leverage before the direction resolves has poor odds.
With household inflation expectations already past their line, VIX below 15 reads less as a reason for comfort and more as a quiet stretch before Sep 30 core PCE.
Glossary
- Contango (VIX) — near-term implied volatility priced below later months; the market is pricing calm now and risk later.
- CBOE Skew — the price of far-out downside protection on the S&P 500; the 140–150 band marks elevated demand for tail hedges.
- Scenario-weighted floor (equity put/call) — the normal lower bound under reflation and under stagflation, blended by each scenario’s probability; readings below it mean single stocks are underhedged.
- Three-month annualized rate — the last three months of inflation scaled to a yearly pace; it jumps whenever a month enters or leaves the window.
- Breakevens — the inflation rate implied by the gap between regular and inflation-protected Treasury yields.
- Keltner channel — a band drawn around the average price by the average daily range; Bollinger bands inside it signal compressed volatility.