The surprisingly strong jobs report has pushed the probability of a September rate hike by the Federal Reserve to 60%, placing Fed Chair Warsh in a dilemma between disappointing the market or disappointing President Trump.
A September 8 report from SYWG Research highlights a historical pattern: once market rate hike expectations exceed 40%, a hike has never failed to materialize. If expectations are dashed this time, term premiums could rise sharply, leaving the market exposed to a "backlash" risk.
The August non-farm payroll report, released on September 4, showed 162,000 new jobs added, far exceeding the market forecast of 55,000, which directly triggered a dramatic repricing of expectations for a September Fed hike. Meanwhile, the rebound in oil prices and sticky AI-driven inflation mean the probability of a sharp drop in August CPI is only about 10.6%, suggesting that elevated market expectations are unlikely to fade after the CPI release.
Against this backdrop, SYWG believes that a September hike may be the Fed's "least bad option" at present — the cost of not hiking could be rising term premiums and market backlash, while the impact of a hike would be relatively limited if it does not lead to a substantial upward revision of the rate path.
Non-farm payrolls beat expectations, keeping hike expectations elevated
Following the release of the August jobs data, market pricing for a September Fed hike quickly intensified. On September 3, after Fed Governor Christopher Waller's speech, the probability of a September hike briefly fell to 50%; but after the jobs report, it rebounded to around 60%.
The August CPI report will be the final key variable before the September FOMC meeting. Historical analysis shows that only a CPI reading significantly below expectations can lead to a substantial downward revision in market rate hike expectations. According to SYWG's calculations, since 2015, when inflation comes in below expectations, market pricing for a rate hike at the next meeting falls by an average of only 6 percentage points on the day of the CPI release. There have been only 13 instances since 2015 where CPI releases led to an intraday downward revision of more than 10 percentage points in hike expectations, and only 3 of those occurred when inflation was in line with or slightly above expectations, all accompanied by external shocks such as the pandemic, unexpected dovish comments from Fed officials, or banking crises.
Currently, August CPI faces dual pressures from energy and structural inflation. Escalating US-Iran conflict has disrupted passage through the Strait of Hormuz, driving oil prices higher, with US Gulf Coast crack spreads rising to $67.9 per barrel; AI-related service prices are also showing structural upward trends. SYWG conducted 10,000 Monte Carlo simulations based on four institutional forecasts, showing that the probability of a significantly below-expectation CPI reading is only about 10.6%.
Expectations above 40% have never been unmet; disappointment could trigger term premium "backlash"
Internal divisions within the Fed are extremely pronounced at present. After the July FOMC meeting produced a 9-3 vote, divisions have further intensified. Based on recent comments, Beth Hammack, Neel Kashkari, and Lorie Logan lean hawkish, continuously advocating for rate hikes; Christopher Waller, John Williams, and others lean relatively dovish; while Warsh himself, in remarks on August 28, stated there is "more work to do" if core inflation does not clearly improve, signaling a shift toward hawkishness.
Historical data carries important implications for Warsh's decision-making. According to SYWG's analysis, across 92 FOMC meetings since 2015, whenever market rate hike expectations exceeded 40% within the 10 trading days prior to a meeting, the hike never failed to materialize — all 20 such instances were delivered as expected or exceeded expectations. There were only 5 instances where expectations ranged between 30% and 40% but ultimately unmet, occurring in September 2015, September 2016, May 2018, November 2018, and July 2026.
Among these 5 unmet cases, the 2015 and 2016 situations were accepted by the market due to global risks and weak economic data. However, the May 2018 and July 2026 cases are particularly noteworthy — both were meetings during the framework-establishment phase early in the tenures of Powell and Warsh, respectively. After rate hike expectations were unmet in those instances, long-end term premiums rose sharply in a "backlash": within 10 trading days after the disappointment, 10-year term premiums rose by 5.0 basis points and 6.2 basis points, respectively.
Political pressure from Trump is also impossible to ignore. As of September 3, Polymarket data showed the probability of Democrats taking control of the Senate at 51%, with both Democrats and Republicans projected at 50 seats, leaving Trump facing enormous pressure from potential midterm election losses. However, SYWG points out that historical review shows that since 1983, rate hikes in September of midterm election years or years when an incumbent president seeks re-election have occurred 3 times — no less frequently than in "politically insensitive" years. In 2018, Powell, newly nominated by Trump, also withstood political pressure to deliver consecutive hikes. Taken together, market pressure may tilt Warsh's balance toward hiking.
Impact of a delivered hike is limited; the key lies in whether the path is revised upward
If a September hike materializes as expected, historical patterns suggest its impact on asset prices is relatively limited in duration. According to SYWG's review of 51 post-hike episodes since 1990: US equities typically show a pattern of short-term pullback followed by medium-term recovery, with cyclical stocks performing relatively weaker in structure; the 10-year Treasury yield trends higher but term premiums decline notably.
The divergence in post-hike asset performance depends mainly on two factors: whether the hike exceeds expectations, and whether the forward path is revised upward after the hike. Taking the 10-year Treasury as an example, when a hike is delivered above expectations, bond yields fall by an average of 9 basis points over the following 20 trading days; conversely, when the hike comes in below expectations, yields rise by an average of 28 basis points. When the forward rate path shifts notably upward, 10-year Treasury yields rise by an average of 35 basis points over the next 20 trading days; when the path stays roughly flat, yields fall by an average of 5 basis points.
SYWG believes that if the September hike is delivered slightly above expectations, the market may interpret it as a front-loading of rate increases for the coming year, which does not necessarily translate into a substantial upward revision of the rate path. The reasoning is twofold: first, the August jobs data is heavily distorted by seasonal adjustment factors; considering low hiring rates, low layoff rates, and low labor force participation, the US job market remains in a "weak balance." Second, wage growth has not shown a clear upward trend, and current inflation is more structural than broad-based, casting doubt on the necessity of multiple consecutive hikes. If the September dot plot guidance does not signal a substantial upward revision of the rate path, the impact of a hike on markets could be relatively limited, with limited impact on the short end of Treasury yields, and term premiums may even modestly decline.
Comments