The pace changed, the demand did not

‍

Executive snapshot

October’s index return hid a change in who was leading

Large technology and semiconductor stocks, the market’s leaders for most of the year, sold off hard. Utilities moved the other way and rose 10.2% as Treasury yields fell.

‍

Verdict

• Our view is unchanged after the November 5 sell-off, which we read as a repricing of policy risk and crowded positions, not of the economy.
• The equal-weighted S&P 500, where every company counts the same, rose 5.0%, while the Nasdaq 100 fell 6.7% and semiconductor stocks fell 15.4%.
• Third-quarter growth came in at 3.4% in the advance estimate, the October manufacturing survey read 55.5, and about 82% of the 265 companies that had reported third-quarter earnings by October 29 beat expectations.

‍

Core driver

• For the fourth quarter, analysts expect the other 493 companies to grow earnings 31.6% against 27.4% for the seven largest, which would be the first quarter since 2022 that the rest of the index out-grows the leaders.
• The utilities overweight helped, with our utilities and power holding up 4.1% while the S&P 500 fell 0.9%.
• Bonds rallied, and investment-grade corporate bonds returned 2.5%, so holding fewer bonds than a standard 60/40 cost us.

‍

Portfolio stance

• Equities: Slight Overweight. Earnings, not October’s price moves, set the overweight.
• Fixed Income: Underweight. We hold fewer bonds than a standard 60/40 and keep duration modestly short.
• Alternatives: Slight Overweight. Real assets hedge energy-driven inflation and fiscal risk.

‍

On the midterm results and the ensuing sell-off

On November 5, two days after the midterm elections, the S&P 500 fell 4.4% and the Nasdaq fell 5.5%, which left the Nasdaq in correction territory. It was the worst day for the S&P 500 since April 2025. AI and semiconductor stocks led the decline, and several large chipmakers fell between 10% and 17%. The trigger was policy. The election results raised questions, in the states and in Washington, about how quickly new data centers can be built.

We read the day as a repricing of policy uncertainty and of crowded positions, not of the economy. Professional investors had been trimming their AI holdings since mid-October, the same unwinding that drove October’s rotation, and the election gave them a reason to sell more at once. A trade that has run hot is healthier letting air out over weeks than bursting all at once. The jobs report the next morning pointed the other way, with hiring well ahead of expectations.

The risk is to the pace of AI building, not to the demand for it. New rules can slow how fast data centers are permitted, powered and connected, which pushes spending out rather than cancelling it. The largest firms may come out ahead. They have the balance sheets, the existing capacity and the freedom to choose where to build that smaller rivals lack, so slower approvals could widen their lead. Pennsylvania offers an early template. A state executive order asks developers to bring their own power, pay for the grid upgrades they need and win local approval. Other states could follow. Clear rules, even strict ones, let companies plan, and over time that could turn this headwind into a tailwind. What would change our view is weaker demand, such as slowing AI revenue or softer cloud usage. We have not seen either.

Days like this are part of owning stocks. Since 1928 the S&P 500 has averaged more than three pullbacks of 5% and about one correction of 10% a year. Since 1980 its average decline within a year has been about 14%, and it still finished higher in about three of every four years. After November 5 it sat about 6% below its high. The Nasdaq went through a correction in March and set new records in October.

The sell-off pressed on our technology and utilities overweights, since utilities are tied to the power demand from AI. Other parts of the book cushioned it. The equal-weighted S&P 500 exposure we added this year spreads the equity holding beyond the leaders, and bonds and commodities diversify. Treasuries rallied on the day. In taxable accounts we used the decline to harvest tax losses.

‍

Core driver

October’s index return hid a change in who was leading. Large technology and semiconductor stocks, the market’s leaders for most of the year, sold off hard. The Nasdaq 100 fell 6.7%, semiconductors fell 15.4% and emerging markets, which carry a large weight in chipmakers, fell 8.0%. Utilities moved the other way and rose 10.2% as Treasury yields fell. Small caps rose 0.6%. The equal-weighted S&P 500 beat the cap-weighted index by almost six percentage points in a single month. The sell-off after the midterm elections hit the same crowded trade.

The honest question is whether earnings are broadening too, and so far that is a forecast, not a fact. Over the past year the seven largest stocks grew earnings 12.4 percentage points faster than the rest of the index. For the fourth quarter, analysts expect the other 493 companies to grow earnings 31.6% against 27.4% for the seven largest, which would be the first quarter since 2022 that the rest of the index out-grows the leaders. Third-quarter reports point the same way. As of late October, the other 493 had reported faster third-quarter earnings growth than the seven, even though the seven still lead over the full past year. The same estimates put the seven back in front in early 2027. October priced a handoff in leadership ahead of any proof of one in reported profits. We read the rotation as a market spreading out its bets, not a verdict that the leaders’ earnings are failing. The coming quarters are the test, and estimate revisions are how we will judge it.

We would rather say plainly how the stance did. The high-yield underweight helped as spreads widened, with our high-yield fund up 0.4% against 2.2% for our core bond fund. The utilities overweight helped, with our utilities and power holding up 4.1% while the S&P 500 fell 0.9%. Commodities were about flat. Oil’s 15% gain was offset by falls in gold and copper, so our broad commodity fund slipped 0.5%. Against that, the technology overweight hurt, with our technology sector fund down 8.3%. The emerging markets overweight hurt, with our two emerging-market funds down 6.4% and 8.5%. Our gold holding fell 6.6%. Bonds rallied, and investment-grade corporate bonds returned 2.5%, so holding fewer bonds than a standard 60/40 cost us. On balance October did not pay the stance. We hold it because the reasons for it are intact, and one month of price is not a reason.

Here is what that means for each part of the book.

‍

Equities

We stay slightly overweight, with US large caps as the core and overweights in technology and utilities. Earnings, not October’s price action, set those views. Analysts have raised forward estimates by 5.8% on average over the past three months, and that is the gauge that would tell us to change the technology view. Small caps and international developed stocks stay neutral. We stay modestly overweight emerging markets.

‍

Fixed income

We stay underweight, with high yield the deepest underweight, and we keep duration modestly short. The 10-year Treasury yield fell from 5.3% to 4.9% in October, and almost all of the move came from real yields, the return investors demand after inflation, which fell from 2.9% to 2.6%. With inflation above target, futures pricing a December hike as likely and about one more by mid-2027, we are not extending duration into that rally.

‍

Real assets

We keep both legs of the sleeve. Commodities are the hedge against energy-driven inflation. They were about flat in October, as oil’s gain was offset by weaker metals. Gold is the hedge against fiscal risk, and it gave back 6.2% in a month when real yields fell, which is not what the usual relationship would predict. We accept that the two legs will not always move together.

‍

Macro thoughts

Growth: above trend, and hiring picked up

Growth has not slowed in any way that matters. The Bureau of Economic Analysis put third-quarter growth at 3.4% in its advance estimate, the Atlanta Fed’s current-quarter nowcast is 3.2%, and the Weekly Economic Index has averaged 2.8% this year against a trend near 2%. The October manufacturing survey read 55.5, well into expansion.

Hiring had been the soft spot. Payrolls rose only 29,000 in September and the unemployment rate was 4.2%, still low. With labor-force growth much slower, estimates of the breakeven pace, the job growth needed to hold unemployment steady, have fallen to roughly 25,000 to 50,000 a month, so it takes few new jobs to keep unemployment flat or falling. Our read is that slow hiring without layoffs fits an economy growing on output per worker rather than headcount, though one month cannot settle that. The October report, released November 6, eased the concern. Payrolls rose 165,000 against about 60,000 expected, unemployment fell to 4.1% and the two prior months were revised up. We will keep judging hiring through jobless claims and earnings revisions together, not any single print.

‍

Inflation: headline rising with oil, the Fed’s gauge above target

Headline CPI was 3.7% in September, and the nowcasts put it at 3.7% for October and 4.3% for November as oil rose. Core PCE, the gauge the Fed acts on, was 3.1% in September and is nowcast at 3.2% and then 3.3%. Both sit well above the 2% target.

Markets are not pricing an inflation scare. The 10-year breakeven, the market’s estimate of future inflation, barely moved in October, slipping by 0.03 percentage points to about 2.3%. The open question is whether energy reaches core prices. If it does, the Fed has less room to ease and our case for keeping duration short gets stronger.

‍

Financial conditions and the Fed: rates eased, credit widened

The Fed raised rates by a quarter point on September 16 and held at 3.75 to 4.00 percent on October 28, with two voters dissenting in favor of another hike. That is a Fed that has paused but has not changed its view of the direction. Futures price a December hike as likely, about three chances in four, and about one more by mid-2027, for 2.16 quarter-point hikes in total through December 2027. That is down from 3.60 a month ago, and that pullback is what let real yields fall.

Credit moved the other way. High-yield spreads widened from 311 to 340 basis points in October and investment-grade spreads from 80 to 90. The Chicago Fed financial conditions index moved from -0.55 to -0.49 and the St. Louis Fed stress index from -0.81 to -0.58. All four moved in the direction we flagged, and both broad gauges still sit well on the easy side of their histories. That is a warning, not a confirmation.

‍

Earnings and valuation: strong, still concentrated, priced a little more fairly

Earnings are the reason we hold the overweight. Trailing earnings growth for the S&P 500 is 25.5% and analysts expect 22.9% over the next year. Estimates have been revised up by 5.8% over three months, and about 82% of the 265 companies that had reported third-quarter results by October 29 beat expectations. Net profit margins are 15.6%, at the top of their history, so margins have little room left to add to growth.

The rotation has not yet shown up in reported profits. The seven largest stocks are still growing earnings 12.4 percentage points faster than the rest of the index, and the fourth quarter is the first in which analysts expect the rest to grow faster since 2022. Valuation eased with October’s pullback. The forward P/E is 18.65, from 19.08, and the equity risk premium, what owning stocks pays over a 10-year Treasury, is back above zero at 0.46 percentage points.

‍

Geopolitics: Hormuz traffic is low, but oil is getting through

Strait of Hormuz transits are running at 3.9% of their normal pace, and Kalshi, a prediction market, puts the odds that traffic returns to normal within 12 months at about 44%. The transit count is a less useful gauge than it was, because oil is still reaching buyers. US crude oil rose from $90.42 to $104.12 a barrel in October, a 15% gain, and settled above $100.

In our experience, geopolitical shocks have tended to be buying opportunities unless they reach credit or earnings. Higher oil prices are the route by which this one could still reach both, through inflation and borrowing costs, but with supply getting through we see that as a smaller risk than it was. We keep the commodity tilt as the hedge and watch credit most closely.

‍

The Savvy Macro Dashboard

Summary

‍

Data

‍

Question of the month

"If AI stocks sold off, why are you still overweight technology?"

Because earnings revisions, not price, set the overweight. Our technology fund fell 8.3% in October and semiconductors fell 15.4%, and we felt it. But analysts are still raising estimates, by 5.8% on average over the past three months, and about 82% of the 265 companies that had reported third-quarter results by October 29 beat expectations. A price drawdown with rising estimates makes the same earnings cheaper. What would change the view is estimates turning down, not a bad month. We also hold technology inside a broad large-cap core, alongside a utilities and power holding that rose 4.1% in the same month, so the book is not a single bet on the leaders.

The spending is also starting to pay. On third-quarter calls, large financial firms said the benefits of AI already exceed what they spent on it this year, and one expects those benefits to more than double next year. It is early, but that is demand, and the November 5 sell-off did not change it.

‍

Savvy Total Portfolios: tactical asset allocation positioning

No changes to positioning this month.

‍

Asset class views

‍

Sector views

‍

Risks we’re watching

1. Tighter credit slows growth or AI spending.

High-yield spreads widened from 311 to 340 basis points in October and investment-grade spreads from 80 to 90. If the broad financial-conditions gauges follow, tighter credit could slow hiring, growth and the borrowing behind AI investment. We would then reduce equity risk, upgrade quality, extend duration and take the commodity tilt to neutral, together. This is the risk that matters most to us.

2. Earnings estimates roll over.

Analysts have raised forward estimates by 5.8% over three months, but the margin of companies seeing estimates raised over cut has narrowed, from +0.26 to +0.20 in the month to October 30. The seven largest stocks are still growing earnings 12.4 percentage points faster than the rest. If revisions turn negative, or growth narrows further onto a few names, the case for the equity overweight weakens. Estimate revisions are the signal we watch.

3. Yields rebound with the Fed.

Futures price a December hike as likely and 2.16 quarter-point hikes in total through December 2027. A rebound in the 10-year real yield from 2.6% would reverse October’s relief and squeeze the equity risk premium, now only 0.46 percentage points. Our short duration is the hedge.

4. Data-center policy slows AI building.

After the midterm elections, state and federal rules could slow how fast new data centers are permitted and powered. That would delay spending at the technology and utilities companies we overweight, even if demand holds. We would judge it through capital-spending guidance on earnings calls, AI revenue and new state rules like Pennsylvania’s.

‍

Closing thoughts

October looked flat and was not. Leadership moved from the largest technology names toward the rest of the market, rates eased for real-yield reasons, and credit widened. The stance did not have a good month, and we have said where it cost us. We hold it because growth is above trend, earnings are rising and the signals that would change our minds have not confirmed.

November has already tested that. AI stocks sold off sharply after the midterm elections, and the jobs report the next day came in well ahead of expectations. Still ahead are the October CPI report in mid-November, Nvidia’s results on November 17 and the Fed’s meeting on December 8 and 9. We will not reposition around any one of those. Credit and earnings are the signals that would move us.

‍