SMIC's 97.3% Utilization Faces a $48.8 Million Q4 Profit Test
SMIC has proved scarce capacity can produce bigger profits. Its next expansion is about whether those profits survive the cost of making more chips.
SMIC was using 97.3% of its wafer capacity in the third quarter of 2026. Almost every slot was filled. Management expects fourth-quarter revenue to rise and gross margin to fall. You would think scarce factory time lets a chipmaker send customers a bigger bill. Depreciation has other ideas.
SMIC has already demonstrated the attractive half of this business. Its gross profit rose from $760.640 million in the second quarter of 2026 to $953.190 million in the third, a gain of $953.190 million minus $760.640 million, or $192.550 million, in its 12 November 2026 earnings release. The harder question is whether new capacity preserves that gain as depreciation rises. The Scarcity Tax is the cost of expanding production that consumes the profit from stronger demand. Its test is fourth-quarter 2026 gross profit against the third quarter's $953.190 million.
The physical constraint is plain. SMIC's second-quarter 2026 release put utilization at 93.7% and monthly capacity at 1,096,500 eight-inch-equivalent wafers. In the third quarter, utilization reached 97.3% and capacity 1,116,750. Shipments rose from 2,869,495 to 3,033,683 wafers. Slack is disappearing. You are looking at demand using up the remaining slack faster than production space expands. At 97.3% utilization, reaching 100% gives just 100 divided by 97.3 minus one, or about 2.8% more output on existing capacity. Anything beyond that headroom needs added capacity.
The profit arithmetic gets awkward. SMIC guided fourth-quarter 2026 revenue up 4% to 6% sequentially and gross margin to 26% to 28%, down from 29.9% in the third quarter. In a midpoint scenario, $3,190.038 million of third-quarter revenue times 1.05 for growth times 0.27 for gross margin equals $904.376 million of fourth-quarter gross profit. Against the third quarter's $953.190 million, that is $953.190 million minus $904.376 million, or $48.814 million less. More sales. Fewer gross-profit dollars. Management guided revenue and margin; the profit calculation is a scenario, not a result.
The expense line is growing too. SMIC's depreciation and amortisation reached $1,270.420 million in the third quarter of 2026, up from $995.857 million a year earlier. Yet gross margin went from 22.0% to 29.9% over those same quarters. Depreciation within cost of sales was $945.759 million against the third quarter's $3,190.038 million in revenue. A heavier fixed-cost load has been spread over growing production and a changing product mix. That operating leverage deserves applause. The same machinery will demand its share before the next increment of capacity earns anything for shareholders.
Capacity spending has its own strange rhythm. The first-quarter 2026 release recorded $1,088.139 million of depreciation and amortisation. The second quarter brought $1,211.721 million; the third, $1,270.420 million. Capital expenditure, meanwhile, fell from $1,835.7 million in the second quarter to $1,094.3 million in the third. Why spending fell remains unexplained. Today's spending number gives you an incomplete view of tomorrow's costs. Capacity takes time to qualify, while depreciation keeps turning up in the accounts.
Cash complicates the return story. At 30 September 2026, SMIC carried $14,904.184 million of debt against $13,406.216 million in cash, leaving $1,497.968 million of net debt. At 30 June 2026 it had $162.342 million: $1,497.968 million minus $162.342 million equals $1,335.626 million of additional net debt in one quarter. Operating cash flow fell from $2,522.116 million in the second quarter of 2026 to $1,182.208 million in the third. You can have crowded factories and a growing financing bill at the same time. The balance sheet has less room.
The product mix supplies SMIC's best defense. SMIC's industrial and automotive share of wafer revenue rose from 16.5% in the second quarter of 2026 to 19.3% in the third. Its 12-inch wafers accounted for 78.9% of wafer revenue, against 78.2% the previous quarter. The release offers neither an AI revenue line nor a clean wafer-price series. A better mix can lift gross profit without requiring an across-the-board price rise. If you own SMIC, that is a perfectly good source of value. It also gives you an earnings driver other than scarcity itself.
SMIC's management makes the optimistic argument in its 13 August 2026 release. It pledged to "flexibly allocate existing capacity and accelerate the qualification of newly added capacity" to meet demand. At the time, second-quarter utilization stood at 93.7%; by the third quarter it was 97.3%. The company then beat the top of both its revenue and gross-margin guidance ranges for the second and third quarters of 2026. These are real accomplishments, and they give management's fourth-quarter caution a credibility problem. You would be foolish to treat a guided margin as money already lost.
TSMC offers the more powerful version of that argument. Its 2025 Form 20-F shows gross profit increasing from NT$1,624,354 million in 2024 to NT$2,281,294 million in 2025, while margin improved from 56.1% to 59.9%. Capital expenditure also increased, from NT$956,007 million to NT$1,272,411 million. Investment and better margins can coexist. TSMC has a different technology and customer mix, and its annual results cannot isolate the return on a newly added fab.
United Microelectronics showed what a more direct scarcity premium looks like. Its 25 January 2022 SEC exhibit put fourth-quarter 2021 utilization at 100% or more, revenue at NT$59.10 billion and gross margin at 39.1%. With shipments guided flat for the first quarter of 2022, UMC wrote, "ASP in USD: To increase by 5%", and guided gross margin to approximately 40%. That was guidance, rather than a realized price increase, and UMC served a different technology and customer mix. Still, its stated price lever is conspicuous beside SMIC's missing one.
What SMIC actually offers is a more demanding bargain. Even the high end of its fourth-quarter 2026 scenario, $3,190.038 million of third-quarter revenue times 1.06 times a 0.28 margin, yields $946.803 million of gross profit. That is below the third quarter's $953.190 million. SMIC has to outrun the Scarcity Tax in actual gross profit. The third-quarter improvement is real. So is the cost of earning the next dollar. You are buying both.
I would value SMIC for the earnings its crowded factories already produce and refuse to pay extra for scarcity until gross profit holds up through the next quarter. The test is money retained, rather than orders imagined. My falsifier for February 2027 is SMIC's fourth-quarter 2026 results release showing fourth-quarter 2026 gross profit of at least $953.190 million, its third-quarter 2026 level. That would overturn my call that the next sales increase brings a gross-profit squeeze. Until then, you are being asked to value the next factory on the strength of the last one.