PortfolioAI Macro Monitor · September 2, 2026
Recession Risk: Low Odds, Rising Data Sensitivity
The economy retains a labor-and-output cushion, but the margin for an adverse growth surprise is smaller than the calm headline probability implies.
Executive Summary
The U.S. recession case remains a tail-risk scenario rather than the base case. Real GDP increased at a 1.5% annualized pace in the second quarter, the unemployment rate was 4.1% in July, and the Sahm Rule reading was below zero. Those are not recession-confirming conditions. The market-implied probability of a recession by year-end was about 7.5% on September 2, with a 7¢ bid and 8¢ ask.
The more useful conclusion is conditional: a modest slowdown can coexist with these readings, but a weaker labor print, renewed credit stress, or a material retrenchment in consumer demand would matter more now because growth is positive but not rapid. This is a monitor for confirmation, not a call to abandon risk assets.
- Next 3 months
- Low risk
- Next 6 months
- Low–moderate
- Next 12 months
- Moderate
What the Dashboard Is Pricing
Probability gauges remain subdued
Latest available observations: FRED July 2026; Polymarket September 2, 2026. The measures use different definitions and should not be treated as interchangeable forecasts.
The labor cushion has improved
Unemployment edged down from 4.3% in May to 4.1% in July. The Sahm Rule has moved from 0.10 in May to -0.03 in July, well below its 0.50 trigger threshold.
Risk Table
| Indicator | Latest reading | Signal | Why it matters |
|---|---|---|---|
| Real GDP | 1.5% annualized, Q2 2026 | Supportive | Output is still expanding, though the pace leaves little room for a sustained demand shock. |
| Unemployment rate | 4.1%, July | Supportive | Labor-market slack has not broadened; a persistent reversal would be the clearest near-term deterioration signal. |
| Sahm Rule | -0.03, July | No trigger | The indicator remains far below the 0.50 level historically associated with the start of a recession. |
| Smoothed recession probability | 0.76%, July | Low | The historical model remains subdued, though it is a lagging confirmation tool rather than a trading signal. |
| Money-market fund assets | $8.29T, Q1 | Liquidity buffer | Large cash balances provide optionality, not a recession forecast; flows can reverse when risk appetite changes. |
| “Recession” searches | 1,346 daily, Aug. 29 | Cooling | Search interest was down 7.1% week over week and 35.8% month over month, reducing evidence of broadening public anxiety. |
| Year-end recession market | 7.5% Yes | Low | Prediction-market pricing is an aggregated wager, not a macro model, but it is consistent with a low-immediacy recession view. |
The Narrowing Margin
Why the base case still holds
A positive GDP print, stable-to-lower unemployment and an untriggered Sahm Rule form a coherent expansion signal. Search behavior and event-market pricing reinforce that the public and marginal forecaster are not treating a near-term contraction as central.
What would change the assessment
Watch for unemployment rising across several reports, a Sahm Rule acceleration toward 0.50, a credit-sensitive deterioration in business conditions, or negative real-GDP revisions. Any one print can be noisy; a cluster across labor, consumption and credit would be more meaningful.
Portfolio Implications
For diversified investors, the evidence favors balance over a binary recession trade. Retain exposure to companies with recurring demand, pricing power and durable free-cash-flow conversion, while keeping a watchlist of businesses whose earnings are most sensitive to a weakening consumer or tighter credit. The distinction matters: a slower economy is not automatically a recession, and defensive positioning should be sized to evidence rather than headlines.
Sources: Federal Reserve Economic Data (FRED); DailySearchVolume; Polymarket; NBER Business Cycle Dating Committee. Data availability varies by series.