〔Day Digest〕Hike Odds Top 85% for September 16 to 3.75%-4.00% as 86 of 101 Economists Concur; Brent Jumps 3%+ on a Saudi Pipeline Halt and the 30Y Sits at 5.362% - Sell-Side Paths Now Span One to Six Hikes

1. The Hike Becomes the Base Case

• **[NEW] Reuters poll, decisive:** of 101 economists surveyed, 86 said the Fed will raise the federal funds rate to 3.75%-4.00% at the September 16 decision, and 37 of 70 said there will be at least two hikes by end-March 2027 ^[1][2]. Reuters framed the result as reversing a "fragile no-change consensus" that prevailed before Friday's firm inflation data ^[2].
• **[ESCALATED] Pricing is a band, not a point:** LSEG data implied 87% ^[3]; CME FedWatch was quoted above 90% ^[4]; one strategist note put it close to 90% ^[5]; a separate commentary put it above 85%, with later pricing at about 85% ^[6][7]. Quote the band.
• **[ONGOING] The political overlay:** the hike debate is being framed as a test of Chair Kevin Warsh's independence against President Trump ^[8], with one commentator arguing the market has already made the decision for him ^[4].

2. The Data: Gasoline Did the Heavy Lifting

• **[NEW] August CPI:** headline +0.4% m/m and 3.4% y/y, with July at just +0.1% m/m; core +0.3% m/m and 2.4% y/y, down 0.1pp on the year ^[6][4]. Both monthly readings came in slightly above expectations ^[4], and core m/m exceeded the 0.2% consensus ^[7].
• **[NEW] Internals:** gasoline rose 3.9% m/m and contributed more than a third of the headline 0.4%, while energy as a whole rose 2.1% ^[6]. With Brent and WTI back above $100 in September, gasoline, diesel, aviation and transport costs still have to feed through ^[6].
• **[NEW] Friday's equity relief came from oil:** the S&P 500 rose 0.9% and the Nasdaq 1% as Brent eased from near $110 intraday to $104.61 ^[6]. On Monday a Saudi pipeline used to bypass the Strait of Hormuz halted, oil rose more than 3% in Asian hours, and US overnight trading gave back much of Friday's gain ^[6].

3. Market Read: Front End Priced, Long End Split

• **[NEW] Rates:** the 2-year has risen 32bp from 4.34% since end-August and the 10-year touched 4.97% ^[4][6]. On Monday, the 2-year fell 3.3bp to 4.610% and the 10-year fell 1bp to 4.964%, while the 30-year rose 0.8bp to 5.362% ^[9].
• **[NEW] FX:** the dollar index rose 0.3% intraday to a high of 99.453, its strongest in 11 days; EUR/USD fell to 1.1549, a one-month low, and GBP/USD slipped below 1.35, down 0.2% on the day ^[3]. The Swiss franc traded near yearly lows on hike bets ^[10].
• **[ONGOING] Term premium:** after Warsh's hawkish Jackson Hole re-emphasis on the 2% target, the term premium measuring investor willingness to hold duration fell 20bp from its high ^[5].
• **[NEW] Doves' caveat:** Barclays said the CPI reading was read as basically locking in a 25bp hike, but that an unexpected hold, or guidance far below the roughly three hikes priced, could push the dollar persistently weaker ^[11].
• **[NEW] Dot-plot mechanics:** a Chicago Fed paper relayed by Nick Timiraos found that when the median SEP rate forecast sat 25bp above market expectations, market-implied rates (OIS forwards) moved only about 5bp that day, consistent with SEP dots being read as conditional assessments rather than commitments ^[12].

4. Sell-Side Dispersion: Two, Three, or as Many as Six

• **[NEW] Goldman Sachs:** chief US economist David Mericle wrote on September 11 that the bank now expects a 25bp hike at the two-day meeting ending September 16, from a prior call of no change, triggered by core CPI at +0.3% m/m versus 0.2% expected ^[7]. Mericle conceded the report only nudged the bank's August core PCE forecast to 0.26% and did not change its inflation view, attributing the shift to credibility concerns with hike odds near 90% ^[7].
• **[NEW] JPMorgan:** now expects hikes in both September and December 2026, versus only December previously ^[13][14].
• **[NEW] HSBC:** expects 25bp hikes in September and December 2026, having previously forecast no change ^[15].
• **[NEW] Deutsche Bank:** adds a March 2027 hike on top of its existing September and December 2026 calls ^[16].
• **[NEW] TD Securities:** expects a September hike to open a tightening cycle, with 75bp cumulative between September 2026 and Q1 2027 and follow-ups in October and January ^[9].
• **[NEW] BNP Paribas:** three hikes between September 2026 and January 2027, driven by healthy growth, persistent deficits and surging hyperscaler issuance, which it says should push long-end yields up and leave the curve sideways, with yields drifting modestly lower from Q2 2027 as Fed credibility questions are digested ^[17].
• **[NEW] The hawkish tail vs the doves:** BMO Capital Markets US rates strategy head Ian Lyngen expects hikes in September, October and December, taking the target range back to 4.25%-4.50% and effectively erasing the 2025 cuts, while Vanguard senior US economist Josh Hirt called three hikes a "fairly reasonable starting point" with a possible range of one to six ^[18]. Yicai commentator Tao Dong still puts the odds at 50/50, or 60/40 after the inflation data ^[4].

5. What Decides Next, and What Is Thin

• **[ONGOING] The decision is not the variable:** Jinshi Data flagged that the market is heavily positioned for a hike but has not resolved the disagreement over whether tightening continues, with oil and inflation expectations supporting the hawkish case and slowing growth constraining it, alongside this week's Bank of Japan decision ^[19][20][21].
• **[NEW] Falsifiable tests:** strategists flagged AI capex optimism and insurers' large private-credit holdings as the two main vulnerability exposures, with a move in the 10-year Treasury yield to 5% potentially triggering a new wave of selling ^[18]. A separate strategy note argues the hike may be paired with balance-sheet expansion, and favors oil chemicals, industrial metals, agriculture, oil shipping and large financials and real estate near term ^[22].
• **[ONGOING] Thin sourcing to flag:** the Swiss franc and "first hike since 2023" items are headline-only aggregations ^[10][23], the Financial Times item is headline-only ^[21], and the Caixin interviews with Miao Yanliang on who controls the Fed and why Warsh turned hawkish are headlined without detail here ^[24][25]. The 85%-90%-plus probabilities vary by instrument and timestamp, so the band, not a point, should be quoted ^[6][3][4][7].