The rest of 2026 will be decided by a single number that is already moving: the share of S&P 500 stocks trading above their 50-day moving average. It hit 25 percent in early October, near the worst reading of the bull market, and climbed to 37 percent within days.¹ If third-quarter earnings carry it back above half while the 10-year Treasury holds near 5 percent, the four-year-old bull gets a second engine. If it rolls back over, the index will still be leaning on a handful of AI names to do all the work.
Across more than a dozen pieces published on AdvisorAnalyst.com between October 12 and October 20, three forces explain almost everything the market is doing. The first is an earnings season with record expectations. The second is a bond market that has pushed long yields to levels last seen in 2002. The third is an AI spending boom that is both the market's main source of profits and its main source of concentration risk. These forces are pulling in different directions, and the outcome depends on which one blinks first.
Can Earnings Keep Outrunning Rates?
The bull case starts and ends with profits. Scott Rubner of Citadel Securities notes that consensus expects roughly 27 percent year-over-year S&P 500 earnings growth for the third quarter, among the highest pre-season expectations on record. Over the past four quarters, companies have raised guidance nearly three times for every cut, and Rubner observes that "earnings expectations have continued to move higher despite the recent weakness in equities."¹
Goldman Sachs Research puts the scale in context. S&P 500 earnings per share rose 51 percent year over year in the second quarter, and the forward multiple sits at 19 times, in line with its 10-year average. Goldman's team, led by Ben Snider, still sees a 12-month S&P 500 target of 8,700, driven by 11 percent earnings growth in each of 2027 and 2028 rather than any further multiple expansion.²
Hubert Marleau of Palos Management makes the same argument in plainer terms. The bull market is four years old and up 118.4 percent, and he sees this cycle as driven by earnings rather than the valuation expansion of 2000. He holds targets of 7,925 and 8,700, with one firm condition: "Take note that avoiding a recession is critical for a bull market."³ Jurrien Timmer of Fidelity adds the most striking valuation comparison: investors "are currently paying a similar multiple (19x) for earnings that are growing 30-40%." He expects third-quarter results could beat the roughly 25 percent consensus, writing, "I see no reason to not expect a repeat of the past few stellar quarters."⁴
To be clear, the headline growth number is narrower than it looks. Excluding the ten largest contributors, Rubner calculates third-quarter growth falls to about 11 percent, and Micron and Nvidia alone account for 36 percent of expected earnings growth.¹ Goldman flags a further wrinkle: mega-cap tech booked more than $150 billion of unrealized private equity gains in the second quarter, adding about 12 percent to S&P 500 earnings, and their absence next year would be an 8-point drag on 2027 growth.²
Is 5 Percent the New Normal for Yields?
Rates are the pressure, and the threshold everyone is watching has already been crossed. The 10-year Treasury pushed above 5 percent this month, reaching levels last seen in 2002,¹³ before easing to about 5.07 percent in the week ending October 16.³ The Fed's September rate hike was its first tightening move since the post-pandemic easing cycle.⁷
Jeff Schulze of Franklin Templeton argues the market is bracing for the wrong outcome. Futures price roughly 115 basis points of further hikes. "If that's accurate," Schulze says, "it's going to be the most modest hiking cycle in history." His recession dashboard shows a clear picture: "Out of those 12 individual indicators, 11 are green," which implies roughly a 15 percent chance of recession over the next year. His reminder to nervous clients is simple: "Bull markets don't die of old age."⁵
Professor Jeremy Siegel of WisdomTree sees a different driver behind long yields. In his view, demand for capital from the AI buildout now matters more to long-term rates than oil or the Fed's next move, which means yields rise when the buildout looks durable and fall quickly when investors doubt it. On policy, he is specific: "I still do not expect the Fed to raise rates in October," though he sees a December hike as much more likely.⁶
Higher rates are also changing which stocks win. AllianceBernstein's Avi Lavi and Snezhana Otto studied six Fed tightening cycles since 1980 and found that value stocks beat growth at the 3-, 6- and 12-month marks after each began. Their reasoning is mechanical: "Higher discount rates have a greater impact on distant cash flows." They advise broadening toward value selectively rather than rotating wholesale, with one caveat: "Statistical cheapness is not enough."⁷
Who Pays for the AI Buildout?
The third force is the largest capital spending wave in corporate history. Chris Konstantinos and Rod Smyth of RiverFront Investment Group report that the four largest hyperscalers guided to $720 to $745 billion of 2026 capital spending, about double the 2025 pace. Amazon's roughly $220 billion plan is not expected to provide enough capacity, Microsoft's Azure grew 43 percent last quarter, and TSMC says "our packaging capacity is so tight that now it's limiting our customers' growth." Their conclusion is that supply, not demand, is the binding constraint, and that adoption "has exceeded our optimistic estimates."⁸
Goldman's research team, led by Ryan Hammond, runs the payback math. With hyperscaler capex near $800 billion this year and consensus at $1.1 trillion for 2027, the infrastructure needs about $300 billion of annual revenue to break even and about $417 billion to earn a 10 percent return. Cloud revenue currently runs about $70 billion above its pre-AI trend: real progress, but well short of the full bill. The encouraging part is what is not priced in: "the equity market today does not appear to be pricing expectations of widespread productivity gains."⁹
The same spending inflates today's earnings in ways that will reverse. Snider's team estimates AI capex adds about 11 percentage points to 2026 earnings growth and turns into roughly a 1-point drag by 2028 as spending growth slows and depreciation rises. If capex were fully expensed, Goldman notes, "consensus 2027 S&P 500 EPS would be 22% lower than under current accounting rules."² RiverFront names four warning signs to watch: reported earnings diverging from free cash flow, pricing power eroding as compute costs fall, debt-financed deals among chipmakers, hyperscalers and AI labs, and electricity and materials inflation.⁸
Siegel sketches what a disappointment might look like. "I could certainly see a 20% to 30% correction in Nasdaq," he writes, yet he estimates the S&P 500 would fall only 5 to 10 percent as lower yields lift value stocks. He also draws a clear line from 2000: "I am not predicting anything resembling the technology collapse of 2000."⁶
The Market Inside the Index
All three forces meet in one place: concentration. DoubleLine Capital's Eric Dhall and Ryan Kimmel find that the S&P 500 returned 12.7 percent through the first nine months of 2026, with five stocks producing 52 percent of that return and the top 20 producing 87 percent. Information technology is now 40 percent of the index, which they describe as "a narrowness of market breadth not seen since the late 1990s." They are careful not to overstate it: "Market concentration alone does not necessarily signal an impending correction."¹⁰
Mike Dickson of Horizon Investments puts the gap in a single statistic. Only 50.2 percent of S&P 500 members trade above their 200-day moving average, down from 75 percent at the prior peak. "The stock market's breadth has weakened even as the overall market has marched higher," he says. Yet he sees the raw material for a broader rally: "Robust earnings growth has remained widespread across the market this year and is even accelerating."¹¹ Cole and Bill Smead of Smead Capital Management take the more cautious read, noting that 70 percent of S&P 500 stocks fell in September even as the index rose, and that concentration in the hottest sector has reached the roughly 40 percent level seen at past manias.¹²
Timmer frames the tension directly. Only about half of stocks are in uptrends, high-yield credit spreads are widening, and he acknowledges the 1998-2000 price analog has reached its peak. He is still constructive: "I am giving this cyclical (and secular) bull the benefit of the doubt."⁴
From Two Engines to a Broader Rally
The turn may already be starting. On October 7, the ratio of the equal-weight S&P 500 to the Nasdaq-100 hit its lowest level since April 2000. Since then, Rubner reports, equal weight has outperformed by 2.8 percent and the share of stocks above their 50-day average has risen from 25 percent to 37 percent. Retail investors bought a net $9.1 billion of U.S. equities in a single week, the largest in over a year, and a record $1.3 trillion of buyback authorizations is coming out of blackout through October and November. Rubner's caution is that fewer than half of similar rebounds have lasted, so he favors "selective single-stock exposure over simply adding index beta."¹
That is the arrival state the bulls are describing: a market where earnings growth outside the largest tech names becomes visible enough to pull the other 490 stocks along. Dickson expects strong third-quarter results from healthcare, consumer and industrial companies to do exactly that.¹¹ AllianceBernstein sees the AI buildout itself helping capital-intensive value companies in power, data centers and semiconductors.⁷
Where Things Stand, and What Would Change the Picture
The S&P 500 sits about 2 percent below its recent high at roughly 19 times forward earnings, supported by consensus expectations of 27 percent third-quarter earnings growth and an AI capex cycle that remains supply-constrained. The 10-year Treasury is holding near 5 percent, the Fed is likely to hike again in December, and half the index is still below its 200-day average.
The trigger is the breadth reading through the rest of earnings season. If the share of S&P 500 stocks above their 50-day average keeps climbing from 37 percent toward and past 50 percent, while the 10-year stays at or below about 5 percent, the rally broadens and the bull gains the second engine it has lacked all year. If breadth rolls back toward 25 percent, or a hyperscaler signals that its capex is outrunning its revenue, the market's dependence on a few AI leaders becomes its biggest vulnerability. When the facts change, the positioning should change too.
At a Glance: Who Said What
| Strategist / Firm | Date | Core View | Forward-Looking Signal | Portfolio Stance |
|---|---|---|---|---|
| Astoria Investment Management | Oct 20 | Q3 rally narrowed as yields hit 2002 levels | AI assumption now load-bearing for valuations | Valuation math holds for now |
| Lavi & Otto, AllianceBernstein | Oct 20 | Higher rates favour value over growth | Value beat growth after all six cycles since 1980 | Broaden toward value selectively |
| Ben Snider, Goldman Sachs | Oct 19 | Over-earning, but not an earnings bubble | AI adds 11 pts to 2026 EPS growth, 1-pt drag by 2028 | S&P 500 target 8,700 |
| Jeremy Siegel, WisdomTree | Oct 19 | AI capital demand now drives long yields | No October hike; December likely | Nasdaq -20% to -30% would mean S&P -5% to -10% |
| Mike Dickson, Horizon Investments | Oct 19 | Breadth weak, earnings breadth strong | 50.2% above 200-day vs. 75% at last peak | Q3 results could widen the rally |
| Hubert Marleau, Palos Management | Oct 18 | Earnings, not valuation, power the bull | Avoiding recession is the critical condition | Targets 7,925 and 8,700; growth and rate-sensitive barbell |
| Ryan Hammond, Goldman Sachs | Oct 15 | AI payback is early but real | $300B revenue to break even; cloud $70B above trend | Market not pricing broad productivity gains |
| Scott Rubner, Citadel Securities | Oct 15 | Constructive but selective into Q4 | Breadth 25% to 37% above 50-day; $1.3T buybacks | Single stocks over index beta |
| Cole & Bill Smead, Smead Capital | Oct 15 | Concentration near past mania levels | 70% of stocks fell in September | Favour free cash flow today |
| Jeff Schulze, Franklin Templeton | Oct 14 | Market is fighting the wrong fear | 11 of 12 recession indicators green; 115 bp priced | Bull intact through a modest hiking cycle |
| Konstantinos & Smyth, RiverFront | Oct 14 | Supply, not demand, limits AI | Capex $720B to $745B in 2026; four warning signs | Manage volatility, stay with the trend |
| Dhall & Kimmel, DoubleLine | Oct 14 | Five stocks drove 52% of the return | IT is 40% of the index | Add equal-weight and CAPE-based sector exposure |
| Jurrien Timmer, Fidelity | Oct 12 | Paying 19x for 30% to 40% growth | Q3 could beat; S&P above 8,000 possible | Benefit of the doubt; 60/20/20 |
Five Takeaways for Advisors
- Track the share of stocks above their 50-day average every week of earnings season. It moved from 25 percent to 37 percent in early October. A sustained move above 50 percent confirms the broadening thesis; a slide back toward 25 percent means the index is again running on a handful of names.
- Hold the 10-year at 5 percent as the line. It crossed above 5 percent this month and eased to about 5.07 percent. Siegel expects no October hike and a likely December move, so the December FOMC meeting is the next decision point for rate-sensitive positions.
- Look through the headline earnings number. Consensus third-quarter growth is about 27 percent, but roughly 11 percent excluding the ten largest contributors. Check how much of each client's equity exposure depends on the names doing the heavy lifting.
- Measure AI revenue against the $300 billion breakeven. Goldman's payback test gives advisors a scoreboard for hyperscaler earnings calls: cloud revenue running about $70 billion above trend is progress, and guidance that widens the gap between capex and revenue is the warning sign RiverFront and Goldman both flag.
- Broaden on purpose, not by accident. DoubleLine's equal-weight and CAPE-based approaches, AllianceBernstein's selective value tilt and Rubner's single-stock selectivity all point the same way. Diversify before leadership shifts, because the data on value after tightening cycles suggests the shift may already be under way.
AdvisorAnalyst.com Editorial Team, October 20, 2026
Notes
- Rubner, Scott. "Fundamentals in Focus: Rubner's Case for Selective Conviction." AdvisorAnalyst.com, Citadel Securities, 15 Oct. 2026, https://advisoranalyst.com/2026/10/15/fundamentals-in-focus-rubners-case-for-selective-conviction.html/.
- Snider, Ben, et al. "The Boom Is Borrowed." AdvisorAnalyst.com, Goldman Sachs Research, 19 Oct. 2026, https://advisoranalyst.com/2026/10/19/the-boom-is-borrowed.html/.
- Marleau, Hubert. "Aging Gracefully." AdvisorAnalyst.com, Palos Management, 18 Oct. 2026, https://advisoranalyst.com/2026/10/18/aging-gracefully.html/.
- Timmer, Jurrien. "Reflections of a Market Cycle." AdvisorAnalyst.com, Fidelity Investments, 12 Oct. 2026, https://advisoranalyst.com/2026/10/12/reflections-of-a-market-cycle-10-12-26.html/.
- Schulze, Jeff. "Not Every Hike Kills the Bull." AdvisorAnalyst.com, Franklin Templeton Institute, 14 Oct. 2026, https://advisoranalyst.com/2026/10/14/not-every-hike-kills-the-bull.html/.
- Siegel, Jeremy J. "AI Buildout Drives Yields and Market Rotation." AdvisorAnalyst.com, WisdomTree, 19 Oct. 2026, https://advisoranalyst.com/2026/10/19/ai-buildout-drives-yields-and-market-rotation.html/.
- Lavi, Avi, and Snezhana Otto. "Have Higher Rates Finally Tipped the Scales in Value's Favour?" AdvisorAnalyst.com, AllianceBernstein, 20 Oct. 2026, https://advisoranalyst.com/2026/10/20/have-higher-rates-finally-tipped-the-scales-in-values-favour.html/.
- Konstantinos, Chris, and Rod Smyth. "Supply Is the Ceiling." AdvisorAnalyst.com, RiverFront Investment Group, 14 Oct. 2026, https://advisoranalyst.com/2026/10/14/supply-is-the-ceiling.html/.
- Hammond, Ryan, et al. "AI Capex, Required Revenues, and What the Market Is Actually Pricing." AdvisorAnalyst.com, Goldman Sachs Global Investment Research, 15 Oct. 2026, https://advisoranalyst.com/2026/10/15/ai-capex-required-revenues-and-what-the-market-is-actually-pricing.html/.
- Dhall, Eric, and Ryan Kimmel. "The Illusion of Diversification: Why U.S. Equity Portfolios Are More Concentrated Than Investors Think." AdvisorAnalyst.com, DoubleLine Capital, 14 Oct. 2026, https://advisoranalyst.com/2026/10/14/the-illusion-of-diversification-why-u-s-equity-portfolios-are-more-concentrated-than-investors-think.html/.
- Dickson, Mike. "Half the Market Is Missing the Party." AdvisorAnalyst.com, Horizon Investments, 19 Oct. 2026, https://advisoranalyst.com/2026/10/19/half-the-market-is-missing-the-party.html/.
- Smead, Cole, and Bill Smead. "Breaking the Fever." AdvisorAnalyst.com, Smead Capital Management, 15 Oct. 2026, https://advisoranalyst.com/2026/10/15/breaking-the-fever.html/.
- Astoria Investment Management. "Rates Hit Multi-Decade Highs: Will Growth and Earnings Keep the Rally Going?" AdvisorAnalyst.com, 20 Oct. 2026, https://advisoranalyst.com/2026/10/20/rates-hit-multi-decade-highs-will-growth-and-earnings-keep-the-rally-going.html/.
