Next week, the US rate decision, Big Tech earnings and inflation data all come out at once. Depending on the results, which way could my stocks move, and what should I prepare now?
2026-10-25
Three tests packed into 48 hours
Reuters called this week a "pivotal test for the rickety AI rally" (Reuters, 2026-10-23). The degree of overlap in the schedule is unusual.
| US time | Korea time | Event |
|---|---|---|
| 10/28 (Wed) 2 p.m. | 10/29 (Thu) 3 a.m. | FOMC decision, Chair Warsh press conference |
| 10/28 (Wed) after the close | 10/29 (Thu) early morning | Alphabet and Microsoft earnings |
| 10/29 (Thu) 8:30 a.m. | 10/29 (Thu) 9:30 p.m. | US September PCE |
| 10/29 (Thu) after the close | 10/30 (Fri) early morning | Amazon earnings |
The order matters. The rate signal comes first, earnings a few hours later. If a hawkish signal comes first, the market will judge the same earnings by a stricter standard. For Korean investors, the 10/29 (Thu) session is the first to reflect the FOMC and Alphabet/MS, and the 10/30 (Fri) session the first to reflect PCE and Amazon.
Premises the market has already priced in
- An October hold is nearly certain. The futures market puts the probability of an October hike below 10% (Reuters, 2026-10-23). Investors see a hike one week before the midterms as unlikely
- A December hike is the baseline. Futures price the probability of a December hike at about 80% (Reuters, 2026-10-23), versus about 71% on CME FedWatch, so 70-80% depending on the source
- Chair Warsh said in a speech in Bangkok on 10/16 that he would not prejudge the remaining meetings, calling October and December the moments of decision (investingLive, 2026-10-16)
- Hyperscaler capex is expected to reach about $800 billion this year, and S&P 500 Q3 earnings are on track to grow more than 32% year over year (Reuters, citing LSEG, 2026-10-23)
This means a December hike and strong earnings growth are already priced in. So this week's variables are not "whether there is a hike" but "the strength of the hike signal" and "proof of returns on capex."
The reaction pattern the market has shown this year
This year the market has not welcomed Big Tech capex increases. When Alphabet raised its AI investment plans in late July, Alphabet and Tesla plunged and major indices fell for the week (Reuters, 2026-07-26). On 10/22, Meta drove the Nasdaq down -2.60% for the same reason. The formula "higher capex = positive for chips" has broken; now only "capex proven by revenue" works.
"What we need is just another round of really strong earnings." — King Lip, BakerAvenue Wealth Management (Reuters, 2026-10-23)
Nick Giorgi of Alpine Macro also called the market's current refusal to give companies spending heavily on AI the benefit of the doubt healthy, while saying it comes at the price of "short-term pain" (Reuters, 2026-10-23).
Four scenarios
They are divided by the combination of the rate signal (hawkish/dovish) and earnings (proof of returns/funding burden). No probabilities are given. The focus is on the conditions that open each path.
| Scenario | Conditions | Expected path | Relative winners | Relative losers |
|---|---|---|---|---|
| ① Double blow | Warsh strongly signals a December hike; Big Tech announces higher capex together with deteriorating cash flow | 10-year back above 5%, further declines in chips | Ultra-short Treasuries, energy | Chips, KOSPI, small-cap growth |
| ② Earnings hold up despite rate pressure | Hawkish signal; cloud growth proves returns on capex | Rates rise but tech is relatively strong | Hyperscalers, memory | Rate-sensitive defensives, real estate |
| ③ Relief rally | Weaker December hike signal, solid earnings | Falling rates and a joint rebound in tech and small caps | Chips, Russell 2000, KOSPI | Energy (relative) |
| ④ Growth scare | Dovish signal, disappointing earnings | Repeat of 10/22. Treasuries strong, tech weak | Long Treasuries, defensives | Chips, AI infrastructure |
Indicators that will separate the scenarios
- FOMC: How strongly the statement and press conference refer to "December," and the number of dissents against the hold. At the September meeting, the market was shaken more by Warsh's tone than by the hike itself (investingLive, 2026-09)
- Big Tech: Whether the size of the increase in next year's capex guidance and cloud revenue growth move in the same direction. If only capex rises and cloud slows, it's ①; if both rise, it's ② or ③
- PCE: July PCE was 3.7% year over year and core 3.3% (BEA, 2026-08). With September CPI beating expectations at 3.7%, a hot PCE would raise the odds of scenarios ① and ②
So what should investors do?
- Cut leverage before the events. Two of the four scenarios (① and ④) are negative for chips. It is not too late to get back into leveraged chip ETFs or margin purchases after seeing the results
- Make your decision in two steps. The FOMC and Alphabet/MS will be known in the 10/29 (Thu) session, and PCE and Amazon by the 10/30 (Fri) session. Rather than betting on a direction all at once, adjusting half your allocation after the first results and the rest after the second keeps the cost of a misjudgment smaller
- Park waiting cash in ultra-short Treasuries. With the 13-week T-Bill above 4%, the cost of waiting is small. It is the asset with the lowest risk of loss in any scenario
- Korean chip holders should look at US Big Tech results first. Even if SK hynix (10/28) and Samsung Electronics (10/29) post good earnings, foreign selling could continue if US Big Tech reveals capex burdens at the same time. They are the end customers for memory demand