AI Policy Tightens as Central Banks Lean Hawkish

Central banks lean hawkish, and we remain underweight duration

Central banks dominate the calendar this week, with the U.S. Federal Reserve (Fed), the Bank of England (BoE), and the Bank of Japan (BoJ) all meeting. The European Central Bank (ECB) set the tone last week by raising rates by 25 basis points and with notably hawkish guidance. Markets now price another hike at the next meeting and three hikes in total by spring 2027.

Growth supports that stance. Outside of France, Purchasing Managers' Indexes (PMIs), which measures economic activity, point to solid expansion across Spain, Germany, and Italy, which allows the ECB to focus on its price stability mandate. Staff forecasts revised core inflation higher for next year, and they now show inflation still running above the 2% target in 2028. The BoE looks likely to hold this week, though we would watch the dissents closely after three at the last meeting, and markets lean toward a November hike. We expect the BoJ to hike this week as well, with five-year household inflation expectations near 10.8%, nominal wage growth at the strongest pace in three decades, and corporate profit margins at wide levels. Even so, the BoJ remains behind the curve.

We have argued for some time that the Fed's next move will be a hike rather than a cut, and last week's Consumer Price Index (CPI) report strengthened that case. The Fed sits close to its employment goal, so its dual mandate miss now falls almost entirely on inflation.

Breadth tells the same story. The share of CPI components growing at 5% or more on a one-month annualized basis has averaged 38% since the start of 2025 and sits at 44% today, against a 25% average in the years before the Covid pandemic (Figure 2). In our view, maintaining credibility on price stability requires a hike.

This chart shows the percentage share of CPI components with 5% +1-month annualized growth since 2014.

That view carries directly into bonds. Ten-year inflation compensation priced into Treasuries has held steady near 2.3%, while the realized headline CPI has averaged 3.3% over the past decade and continues to trend higher. Upside risks to inflation keep appearing, most recently through energy and transportation costs. If inflation does not return to target, we believe inflation compensation in Treasuries looks badly mispriced.

We are often asked when we would add back to duration and the question usually relates to the level of yields. We would frame it differently. The 10-year yield tracks the Fed funds rate, deficits, and Fed holdings of Treasuries very closely, and the funds rate is the most important driver. This means yields usually fall in a sustained way only when the policy rate declines significantly. That outcome typically requires a recession, and our probit recession metric, which draws on jobless claims, Institute for Supply Management (ISM) surveys, and the unemployment rate, puts the current odds of recession at only 2%. The business cycle matters more than the level of yields, so we remain underweight duration and favor short maturity, high-quality bonds.

Bottom line: Inflation runs hotter than target across most major central banks, and policy should stay biased toward higher rates. We would not treat a round number on policy rates or Treasury yields as a reason to shift duration positioning.

Seasonal volatility may create opportunity for underinvested investors

As mentioned in our Bulletin last week, September and October historically rank as two of the most volatile months of the year for markets, particularly when combined with midterm election cycles. Past patterns do not always repeat, and we would not position on seasonality alone. Yet if the headline volatility from the weekend persists while fundamentals remain robust, we believe this kind of volatility creates potential opportunity rather than risk.

Higher Treasury yields should not stand in the way. Investors often ask whether a move toward 5% on the 10-year Treasury would threaten equities, and history suggests otherwise. Since 1990, average forward S&P 500 returns when yields sat above 5% reached over 11% in a year, above the 7.1% recorded when yields ran between 4% and 5%. It was also higher than the 9.5% recorded between when yields were between 3-4% (Figure 3). Equity markets have tended to perform well in higher-yield regimes, not despite them.

This graph shows the average percentage of S&P 500 forward returns by 10yr yield level.

Sentiment gives us a useful guide for timing. A S&P 500 sentiment and positioning indicator we reference has moved away from complacency, which earlier this year correctly signaled stocks would chop around current levels. We would treat a further move toward our fear threshold as a potential opportunity to add to equity risk.

For investors who remain underinvested in equities or are holding excess cash, we believe it will be prudent to monitor a pullback during this seasonally volatile period, as we are inclined to add to risk in that event.

Bottom Line: A volatile fall in a midterm year presents a historically challenging stretch that has often rewarded investors who add rather than retreat. We would use weakness, particularly on a fear reading, to put cash to work in equities while staying short duration in fixed income.