The Nasdaq fell this week, but the small-cap index rose 2%. Will the flow of money out of large-cap tech and into small caps continue next week?
2026-W43
Small caps rose even on the Nasdaq's worst day
The Russell 2000 rose +2.00% for the week and the Nasdaq fell -0.71%, a gap of 2.71 percentage points. Breaking it down day by day shows the character of this week's moves. The figures below are calculated directly from weekly closing prices.
| Date | Russell 2000 | Nasdaq | Dow |
|---|---|---|---|
| 10-19 | +0.25% | +1.11% | -0.07% |
| 10-20 | +1.65% | -0.02% | +1.02% |
| 10-21 | +0.27% | +1.05% | +0.69% |
| 10-22 | +0.11% | -2.60% | +0.02% |
| 10-23 | -0.29% | -0.21% | -1.23% |
On 10-22, when the Nasdaq posted its biggest drop since June, the Russell 2000 and the Dow held flat. Money did not leave the market. It moved to a different part of it. Still, most of the week's gain came on 10-20 alone, and small caps also fell on 10-23 when oil jumped.
Two forces drove this rotation. One is money leaving large-cap tech as doubts about recouping AI spending grew after Meta. The other is the 10-year yield falling below 5%, which eased the interest burden on small caps. Next week, Big Tech earnings test the first and the FOMC tests the second, on the same day.
Small caps are far more sensitive to rates
The report noted only the Russell 2000's gain and did not address how much that gain depends on rates.
- About 30% of Russell 2000 companies' debt is floating-rate, versus about 7% for S&P 500 companies (247 Wall St., 2026-07-07 and others, citing Bloomberg and Apollo data).
- Interest expense is 31% of EBITDA, the highest in at least six years, versus about 6.7% for the S&P 500. One estimate puts the share of Russell 2000 companies that cannot cover interest from operating income at 41-46% (same reports).
These are secondary figures recycled from the same source data across several articles, so their precision is low. The direction is clear, though. If the Fed strongly signals a December hike, large-cap tech only faces a higher discount rate, while small caps see interest costs rise immediately. That has already happened this year. The Russell 2000 was up +18.8% year to date at the end of July, well ahead of the S&P 500 (+8.3%) and the Nasdaq (+7.3%), but the rotation stalled in March-April when the Iran conflict pushed oil and Treasury yields higher (citing AP, ChannelChek, 2026).
There are two precedents
- July 2024, a short-lived rotation. The Russell 2000 gained +10.2% in July, its best July since 2022 (Royce, ETF Trends 2024). Royce assessed it as a brief shift in leadership, and small caps gave back a large part of it in early August during the yen-driven unwinding. The rotation rested on a single factor, expectations of rate cuts.
- 2000, a multi-year rotation. In the 2000-2002 bear market the Nasdaq fell -80% and the S&P 500 fell -50%. By 2006 the Russell 2000 had far surpassed its 2000 peak, while the Nasdaq still needed to double to regain its high (StockCharts, 2006). When large-cap tech's investment payback actually collapsed, the rotation lasted years.
We are somewhere in between. There is no evidence that the AI payback has collapsed, only growing doubt. Next week's numbers will be the first to say which path we are on.
Paths for next week
The odds of a December hike are already priced high, at 75-94% depending on the report. So the FOMC variable is whether the statement confirms a hike or reduces the room for one.
| FOMC statement | Big Tech explains spending with revenue | Spending rises with thin explanation |
|---|---|---|
| Reduces room for a December hike | Both rise, but large-cap tech's rebound is bigger, so the rotation pauses | Rotation continues. Small-cap outperformance widens |
| Strongly confirms a December hike | Money returns to large-cap tech and small caps give back gains. An August 2024 pattern | Both fall. Interest-burdened small caps cannot hold up either. A March-April 2026 pattern |
Only one of the four cells has the rotation continuing. We have not calculated probabilities. What is clear is that for the rotation to continue, both conditions, rates and Big Tech, must line up at once.
What should investors do
- Do not chase the Russell 2000 now. More than half of the weekly gain came on 10-20 alone, and three of the four paths next week are unfavorable for small caps.
- If you add, do it after 10-28 and favor profitable small caps. The S&P SmallCap 600 only includes companies with positive recent earnings, which screens out many of the firms that cannot cover interest. Buying the whole Russell 2000 means buying that 41-46% as well.
- If you hold a large AI-stock weighting, do not switch. Just rebalance. Use this week's rotation only as a chance to bring large-cap tech back to its target weight. Betting everything on the rotation is vulnerable to an August 2024 pattern.
- Record relative performance on the two dates. If the Russell 2000 falls more than the Nasdaq on FOMC day, the rate channel is working; if the two move in opposite directions the day after Big Tech earnings, the spending channel is working. If both point toward small caps, that is when to raise the weighting.