Silicon material prices keep falling. Can photovoltaic industry "grab" a spring?

In the photovoltaic market of March, the chill has not yet dissipated, and the industry is still waiting for a true spring.

On March 19th, the Silicon Industry Branch released the price of silicon materials, showing that the prices of N-type refed materials and dense materials continued to decline, with week-on-week drops reaching 4.42% and 6.12% respectively. Among them, the lowest transaction price of N-type dense material has reached 39,000 yuan per ton, hitting a recent low. Specifically, the average transaction price of N-type refed materials was 43,200 yuan per ton, while the average transaction price of N-type granular silicon remained at 44,000 yuan per ton.

The silicon wafer segment is also under pressure. According to the data from Shanghai Metals Market on March 24th, the price of 18X silicon wafers ranges from 0.98 to 1.00 yuan per piece, that of 210RN silicon wafers is between 1.05 and 1.13 yuan per piece, and that of 210N silicon wafers is within the range of 1.25 to 1.33 yuan per piece. The selling prices of some products have fallen below the cash cost line.

When the price of silicon material drops sharply again and the upstream and downstream links are under continuous pressure, will the photovoltaic industry face a new round of more severe tests in such a predicament?

Why haven't downstream customers sold out despite the continuous decline in costs?

Recently, the price of silicon material has continued to decline, and the price of silver has also changed its high position and entered a downward channel. Moreover, due to the escalation of military conflicts in the Middle East, global energy prices have risen sharply. From the perspective of cost and substitution demand, this seems to be the window period for photovoltaic enterprises to experience a market boom, but the reality has not unfolded as expected.

The reason behind this should first be attributed to the persistently high inventory levels in the industry. The overcapacity accumulated from the disorderly expansion over the past two years is still being digested to this day. According to the analysis of the Silicon Industry Branch, there are currently 10 domestic polysilicon enterprises in production. The average operating rate of the industry has dropped to 35.5%, a year-on-year decrease of 6.5 percentage points.

Although the supply side has been passively contracted, in the face of high inventory pressure, the supporting effect of production cuts on prices has still been significantly weakened. Upstream enterprises are generally confronted with an awkward deadlock: they are not operating, and equipment depreciation and personnel costs are constantly being lost. However, starting production means incurring losses. Moreover, in a market where supply exceeds demand, to capture market share, selling at low prices and high volume seems to be the only way out at present.

When the predicament on the supply side is transmitted to the demand side, the situation is equally not optimistic. The downstream solar cell and module segments have been affected by the sluggish demand for terminal installations, and the purchasing intention has remained low. Most enterprises mainly focus on replenishing inventories for essential needs. The silicon wafer market has seen sluggish transactions and slow inventory consumption.

In the domestic market, the progress of large-scale ground-mounted photovoltaic power station projects has generally lagged behind this year. The delay in the issuance of grid connection indicators, the extension of the fund approval process, and the frequent occurrence of land compliance issues have led to the forced postponement of a large number of projects. The operating rate of some enterprises in the industrial chain remains at a low level.

The overseas market has also failed to provide effective support. From the perspective of export data, compared with the same period in 2025, the export scale of silicon materials, silicon wafers and batteries in the first two months of this year has not changed much, with the average monthly export value remaining basically the same or partially declining. In the component segment, the export value in January and February was 11.212 billion yuan and 11.268 billion yuan respectively, and the export volume was 37.9275 million and 34.1501 million respectively, both showing a significant decline.

On the one hand, the inventory accumulated in the European market in the early stage has not been fully digested, and importers have a weak willingness to replenish their stocks. On the other hand, in the channel of continuous price decline, the photovoltaic market has also shown a mentality of "buying when prices are rising but not when they are falling". Overseas purchasers tend to hold onto their cash and wait and see, waiting for a more favorable bargaining space.

Both domestic and international demand are sluggish, making it difficult for upstream suppliers to boost substantial procurement even if they lower their prices. As a result, the more the price of silicon material drops, the more the market waits and sees. The more the market wait-and-see, the more the price of silicon material drops. The upstream of the photovoltaic industry is trapped in an unbreakable cycle.

With the tax refund window approaching, can silicon materials stop falling?

According to the latest forecast by InfoLink, as March approaches, the production scheduling expectations for all links in the photovoltaic industry chain are generally on the rise. In terms of silicon material, the output in March is expected to increase slightly by approximately 12% compared to February. The total output is estimated at 97,300 tons, of which the domestic contribution is about 90,300 tons.

In the silicon wafer sector, the expected output in March is expected to significantly increase to approximately 51GW. The recovery in the component segment is particularly notable. The production schedule for March is expected to be between 44 and 45GW, representing a month-on-month increase of approximately 28% to 29%. Among them, domestic production schedules have risen to 32 to 33GW.

The most direct driving force behind this round of bullish production is the upcoming April 1st. At the beginning of this year, the state issued an announcement stating that as of April 1st, export rebates for photovoltaic products would be completely abolished. For photovoltaic enterprises, it has become an urgent task to send out the goods in the last window period.

However, this short-term increase in production scheduling is more of a self-driven effort by enterprises to seize the policy window rather than a substantial increase in terminal orders. The demand has not kept pace. Can the adjustment of production rhythm really stop the decline and stabilize the silicon material market? I'm afraid it's very difficult.

Another factor that cannot be ignored is that a special contract manufacturing model is quietly spreading from the bottom up in the industrial chain. The traditional contract manufacturing model usually involves component factories providing raw materials and entrusting battery or silicon wafer enterprises to process them, with the latter earning processing fees. In contrast, the currently popular reverse contract manufacturing is completely different. The contract manufacturer purchases raw materials on its own to complete production and then sells the finished products to component factories or cell factories. This model means that the risks of raw material price fluctuations and inventory overstock are completely transferred to the contract manufacturers.

Take a certain component enterprise as an example. It can outsource the silicon material inventory in its hands to be processed into silicon wafers, and then further consume it through its own solar cell production capacity, and finally make components for external sales. In this way, not only was the inventory transferred, but also the price drop losses caused by the price decline were completely avoided.

There is also a certain component enterprise that sells its silicon wafers to solar cell manufacturers, entrusting them to produce solar cells on their behalf, and then repurchases them. In this way, component enterprises not only reduce the occupation of cash flow but also do not need to bear the risk of inventory depreciation caused by the decline in silicon wafer prices.

As for those enterprises that are willing to take on OEM orders, it is mostly out of helplessness behind the scenes. During the industry's downturn, having orders means that the production line can operate, with revenue and turnover, and only then can there be an account on the books. Even if the profit is meager or as thin as a piece of paper, it is still much better than the situation of halting production. This model, to a certain extent, conceals the real contradiction between supply and demand and also deprives enterprises of the sense of urgency to voluntarily cut production.

Third, the fight against internal competition is a well-discussed topic, but the real implementation of it still has a long way to go

Recently, the National Development and Reform Commission released the "Report on the Implementation of the 2025 National Economic and Social Development Plan and the Draft of the 2026 National Economic and Social Development Plan", clearly stating that efforts should be intensified to rectify disorderly and irrational competition in key areas such as photovoltaic power, strengthen supervision and management of product quality, and promote the recovery of prices of products such as polysilicon and silicon wafers. This is yet another statement made by relevant departments against the "internal competition" in the photovoltaic industry since last year.

The policy signal has been clarified again, but the market's reaction to it has been relatively calm. Looking back on the past year, from the self-discipline production cut meetings organized by industry associations to the joint initiative of leading enterprises to control production capacity, and then to the repeated guidance from relevant departments, the call for "anti-involution" has never ceased. However, after each round of announcements, the most ideal situation is that the price briefly stabilizes and then continues to decline. And this also seems to have formed a kind of inertia: the more one shouts about stopping the decline, the more difficult it is to stop it.

Why is it that the "anti-involution" campaign is always all talk and no action? The root cause lies in the fact that the structural contradictions within the industry have yet to be resolved. In the past few years, the excessive inflow of capital has led to a huge volume of overcapacity. Even if leading enterprises take the lead in cutting production, second - and third-tier enterprises still choose to grit their teeth and produce in order to maintain cash flow and market share. The contraction on the supply side is difficult to form a synergy. No one is willing to be the first to withdraw, fearing that they will give up the market.

What is more crucial is that the focus of market competition nowadays is no longer how much profit can be made, but rather who can last longer. When the price drops below the cash cost, enterprises still choose to operate. This is not a rational business decision but is driven by multiple factors such as cash flow pressure, bank credit renewal, and local government employment assessment. Halting production means losing everything, while persisting in production at least still has the possibility of turning the situation around. This kind of prisoner's dilemma game makes any unilateral self-discipline seem extremely fragile.

The intervention at the policy level has undoubtedly injected a strong dose of confidence into the industry. But to truly achieve the stabilization and recovery of prices, what is needed is not only administrative instructions, but also a substantive reconstruction of the supply and demand relationship. Only when backward production capacity is accelerated to be eliminated, the industry concentration is further enhanced, and enterprises shift from competing on price to competing on technology and quality, can the photovoltaic industry truly break out of the vicious circle of the harder they work, the more losses they suffer.