ENGTMK Verified Portfolio ↗

Growth is not the whole story: the market decline meets TSMC's margin test

16 Oct 2026 $TSM
A silicon wafer on a brass $TSM plinth pulled between revenue-growth papers and a model fab labelled ramp costs. Dated 16 October 2026, the conceptual illustration contrasts sales strength with cost pressure.

By @engtmk · Research as of 16 October 2026

My Wednesday inflation question now meets an earnings test: how much reflects changing demand expectations, and how much reflects the price investors are willing to pay for growth?

This expands today’s eToro post. The distinction matters because a business can grow while its shares disappoint. Revenue, profitability and the price investors pay are three separate tests.

A rebound does not explain the decline

The S&P 500 closed at 7,805.45 on October 13, 7,718.94 on October 14 and 7,744.06 on October 15. Those daily closes imply a 1.11% fall followed by a 0.33% rebound, calculated from the preceding close. The source is S&P Dow Jones Indices data distributed by FRED. These are price-index changes, not total returns, and not an October 16 intraday quote.

The rebound does not identify the cause of the preceding decline. Nor do headline index moves establish market breadth. Demonstrating multiple compression would require comparable valuation and earnings-estimate evidence. This article does not have that decomposition and does not claim it.

TSMC: stronger sales, a separate margin test

TSMC’s October 15 results put Q3 revenue at US$46.72bn. Its Q4 guidance is US$52.3–54.3bn. The US$53.3bn midpoint is about 14.1% above Q3, calculated as 53.3 divided by 46.72, minus one. This is sequential growth in company revenue, not AI-only revenue or unit growth.

Gross margin was 66.4% in Q3; the Q4 guide is 65%–67%. Its 66% midpoint is 0.4 percentage points, or 40 basis points, lower. The range includes outcomes above and below Q3, so the midpoint is not a guaranteed decline. The guidance assumes US$1 equals NT$31.8.

In the officially linked earnings transcript, management attributes the expected midpoint pressure to the 2-nanometer ramp and overseas fabs, with productivity and other cost improvements partly offsetting it.

Management also describes challenges in consumer and price-sensitive end markets. That complicates the picture: strength in leading-edge technology is not proof of universally strong demand.

The outlook challenges a blanket demand-collapse narrative. It does not establish that TSM shares are attractively valued. A lower gross-margin percentage also does not necessarily mean lower gross-profit dollars when sales rise; neither measure alone settles earnings or cash generation.

October 16’s releases add tension

BLS released September import/export prices at 8:30 a.m. ET on October 16. Nonfuel import prices rose 0.7% from August and 6.0% from September 2025. Overall import prices rose 1.1% monthly and 8.1% annually. These price indexes are not seasonally adjusted, as explained in the technical note.

The Federal Reserve’s October 16 initial September estimates show industrial production up 0.7% monthly, including manufacturing up 0.3%, mining up 0.2% and utilities up 3.5%. These figures are seasonally adjusted. Manufacturing was 1.1% above a year earlier; overall capacity utilization was 76.9%, 2.5 percentage points below its long-run average.

The releases show output growth alongside price pressure, with different coverage and adjustment methods. They cannot be subtracted to calculate a margin. U.S. import prices are not a measurement of TSMC’s cost base, and these macro data do not prove what caused a particular share-price move.

No verified primary-source consensus benchmark or current, comparable before/after policy-probability series is used here; no surprise or repricing claim is made.

What would change the interpretation?

  • Demand resilience: delivery against Q4 revenue guidance would support it; weaker orders or lower subsequent guidance would challenge it.
  • Profitability resilience: gross margin within the guided range, alongside operating profit and cash generation, would matter more than sales alone. Cost overruns would weaken that case.
  • Valuation reassessment: comparable earnings estimates and valuation multiples would be needed to distinguish changing profit expectations from changing willingness to pay. A price decline alone is insufficient.

These are conditional tests, not assigned probabilities or a buy signal. The next scheduled FOMC meeting is October 27–28. Q4 results will test TSMC’s delivery; no unverified earnings date is asserted.

Following the existing question

Wednesday’s research examined inflation and the return available elsewhere. This article takes that question into a business: strong demand must still become profit, and profit must be weighed against valuation.

TSMC’s outlook challenges a demand-collapse explanation for its business; it cannot resolve the cause of the broader market decline.

What would reassure you more: stronger sales or resilient margins?

My portfolio and views, not investment advice.

@engtmk engtmk.com