Daqo New Energy's second quarter marked a decisive inflection in the company's navigation of the deepest polysilicon downturn in over a decade. After a prolonged standoff in which management withheld inventory rather than sell below production cost, the company resumed sales in June, driving a 235 percent sequential revenue increase to $62.7 million and narrowing the gross loss from $139.4 million to $82.7 million. The quarter crystallized two competing forces: a structural supply overhang that has kept polysilicon prices below cash cost for multiple quarters, and a gathering policy response from Beijing that is beginning to enforce discipline across the value chain.
Revenue of $62.7 million came on sales volume of 15,190 metric tons at an average selling price of $4.04 per kilogram, compared to 4,482 metric tons at $5.96 per kilogram in the first quarter. Total production cost held flat at $5.95 per kilogram while cash cost edged down to $4.57 per kilogram, meaning every kilogram sold still loses approximately $0.50 at the cash level. The gross margin improved to negative 132 percent from negative 521.5 percent, primarily because inventory impairment provisions fell to $55.7 million from $98.9 million. Net loss attributable to shareholders narrowed to $81.2 million from $88.4 million, with loss per basic ADS of $1.20 versus $1.31. EBITDA improved to negative $29.3 million from negative $83.1 million. The balance sheet remains a fortress: $1.9 billion in cash, short-term investments, bank notes receivable, held-to-maturity investments, and fixed-term deposits against zero debt, providing roughly 2.3 years of runway at the current quarterly cash burn rate.
The investment thesis rests on three variables. First, whether the anti-involution policy cascade (energy consumption standards effective January 2027, CPIA cost accounting principles, SAMR price compliance guidance, and the August 6 joint industry pledge) creates a durable floor under polysilicon pricing that allows Daqo's cash cost of $4.57 per kilogram to become profitable. Second, whether the company's diversification into AI data center power infrastructure (energy storage, solid-state transformers, solid-state circuit breakers) can generate a meaningful second earnings stream before the polysilicon cycle turns. Third, whether the balance sheet can absorb continued cash burn ($276 million operating cash outflow in the first half) without compromising the zero-debt position or forcing equity dilution. The market appears to be pricing a high probability of prolonged losses, with the stock trading at 0.21 times book value and an enterprise value of negative $920 million. Confirmation of the thesis requires polysilicon spot prices sustaining above $5.00 per kilogram for two consecutive quarters and the AIDC venture reaching pilot production by mid-2027. The thesis breaks if prices remain below $4.50 per kilogram through year-end 2026, forcing further inventory write-downs, or if the AIDC investment exceeds $500 million without visible customer traction.