The Monthly Regime Report — August 2026
Manufacturing just printed a four-year high, factory employment expanded for the first time since January 2025, and the late-July selloff has fully reversed to new index highs. The score jumps to +9.
Start with what I got wrong — and right
On 30 July I published a report describing a disorderly de-risking, with the index 11.6% off its high and semiconductors down 24.9%. I wrote that the level that mattered was 644–649 on the Nasdaq 100, where the 200-day moving averages converged, and that I'd rather watch how price behaved there than guess in advance.
It held, and the reversal has been violent:
| Instrument | 30 Jul | 4 Aug | Change |
|---|---|---|---|
| S&P 500 (SPY) | 729.46 | 771.33 | +5.7% 52w high |
| Nasdaq 100 (QQQ) | 661.73 | 723.85 | +9.4% |
| Semiconductors (SMH) | 504.22 | 575.71 | +14.2% |
| Nvidia | 190.01 | 211.94 | +11.5% |
| Industrials (XLI) | 176.66 | 186.40 | +5.5% 52w high |
| Financials (XLF) | 56.68 | 57.88 | +2.1% 52w high |
| VIX | 20.65 | 16.49 | −20.1% |
| Long bonds (TLT) | 82.85 | 82.82 | unchanged |
What actually changed: the ISM print
The 1 August manufacturing release was the month's genuine surprise, and a large one.
Three things in that release matter more than the headline:
1. Employment crossed 50 for the first time since January 2025
This is the single most important number in the report. My July score downgraded labour to bearish because June payrolls came in at +57k against ~115k consensus, firing a pre-set rule. Factory employment expanding at 52.8 is the first evidence that the hiring freeze is thawing — and survey data leads payroll data. Labour moves from bearish to neutral, and would move to bullish on a payrolls confirmation.
2. Customer inventories at 40.7 are the mechanism
Deeply "too low" customer inventories are a forward-looking demand signal that gets ignored because it isn't a headline. Depleted customer stockpiles mean orders must keep flowing simply to rebuild them, largely regardless of end demand. Combined with backlogs at 55.0, this points to production strength persisting for at least a quarter — which is a more durable reason to be constructive than the price action is.
3. Prices are decelerating while activity accelerates
The prices index fell to 71.1 from 73.0, a third consecutive decline, while production surged. Growth accelerating and input costs decelerating simultaneously is the most equity-friendly combination in the macro toolkit — it expands margins without forcing a policy response. The caveat: at 71.1, prices remain firmly in expansion, and raw materials have now risen for 22 consecutive months. This is deceleration, not disinflation.
The one signal that didn't participate
Long bonds are the dog that didn't bark. TLT is unchanged over the period and still sits near 52-week lows, meaning long yields have not fallen at all despite the equity recovery and the softer prices data.
That is worth taking seriously. A rally in equities accompanied by falling yields is a re-rating. A rally accompanied by static, elevated yields is an earnings-and-growth story — which is fine while growth data cooperates, but leaves no valuation cushion if it stops. It also means the discount-rate problem I described in July hasn't been solved; it has been outrun.
Because of that, rates stays bearish and I'm not scoring this as a clean all-clear. A +9 with one hard bearish signal in the most important asset class in the world is a constructive reading, not a complacent one.
Positioning and the tests ahead
The immediate test is payrolls on 7 August. The framework's rule is symmetrical to the one that fired in July: a print above roughly 125k with positive revisions upgrades labour to bullish and takes the score toward +11. A second consecutive sub-75k print, in the face of an expanding employment sub-index, would be a genuine contradiction between survey and hard data — and when those two disagree, hard data has historically won.
The second test is whether long yields eventually follow the prices data lower. If they do, the last bearish signal clears. If equities keep rising while yields stay pinned, the market is spending its margin of safety.
On my own positioning: I have been heavily in cash since November and that stance has now cost me a great deal, including this rally. The framework turning to +9 is the mechanical case for reducing that. I'd rather state plainly that the caution was expensive than quietly re-write it.
For educational purposes only. Mechanical output of a personal framework, shared to show process. Market data as at 4 August 2026 via TradingView; ISM figures from the July 2026 Manufacturing Report On Business. Not investment advice.