24 June 2012
Posted in News - SPVI news
If anyone asked the top three Chinese solar module companies for a good source of cost reduction, they would all speak of polysilicon
Ever since the United States Department of Commerce issued preliminary anti-dumping duties against the China-made cells and countervailing duties against companies that manufacture them, the industry has been expecting a move from China to retaliate. China did complain recently to WTO about subsidies given to seven American projects, addressing the issue of countervailing by the US government. Now it is time to find a similar scenario for the anti-dumping action, and polysilicon manufacturing seems to be in the crosshairs.
For the second time in two months the media is describing an apparent petition from three Chinese polysilicon companies, GCL-Poly, LDK Solar (NYSE: LDK) and Daqo New Energy (NYSE: DQ) requesting action against the US and Korean polysilicon producers. It seems that the Chinese Ministry of Commerce has been reviewing the situation with greater caution, instead of embracing it in the way its American counterpart did the petition made by SolarWorld. For one, cheap polysilicon represents a driving force behind the low cost of Chinese modules. Jinko Solar (NYSE: JKS), which has reported weak sales and an even weaker balance sheet a few days back, announced poly costs of $0.16 per watt, the lowest cost for polysilicon out of any US-listed companies to date. Naturally, cheap poly allows Jinko to stay close to the leading solar firms, supporting the company even while it is faced with pressurized financials. Polysilicon is literally an obstacle to breaking even for every single company, starting with Suntech (NYSE: STP) and ending on Hanwha SolarOne (NASDAQ: HSOL). A Korean supplier, OCI Chem, ships to at least four Chinese tier ones: Suntech, Trina (NYSE: TSL), JA Solar (NASDAQ: JASO) and Yingli (NYSE: YGE). While the first three have close ties with GCL-Poly, Yingli has been almost exclusively connected to OCI Chem for its poly needs. The decision to add duties to those shipments would create more cost, causing Yingli to lose its competitive edge. The company had walked away from its own poly plant, Fine Silicon at least financially, in Q1 2012, and counting on it would offer no relief. The costs are so far out in comparison to the current market dynamic that added duties would not make a difference. This condition is shared by the majority of Chinese enterprises. Built with old equipment into medium-scale plants averaging 3,000 to 5,000MT, many tried to run efficiently but failed. Now they are shut down, as ASP is below their cost of production.
The global market sees 30 to 35GW of shipments in 2012. This translates to around 200,000MT of poly. Counting the top four producers, GCL Poly (65K MT), Wacker Chemie (47K MT), Hemlock Semi (46K MT) and OCI Chem (42K MT) can supply for this demand. The world’s production capability is around 400,000MT of poly, with 100,000MT produced for around $25 per kg average, making the rest produce below the spot ASP.
If anyone asked the top three Chinese solar module companies for a good source of cost reduction, they would all speak of polysilicon. Yingli, which had a rather excellent quarter, paid $0.26 per watt in polysilicon. Trina paid $0.21. Almost all tier one Chinese companies are bound to long-term contracts made with foreign suppliers. Currently, every quarter, prices are negotiated down to get closer to spot market pricing. It is unlikely that polysilicon firms will absorb duties, so prices would go up, consequently pushing long-term pricing up. Faced with higher costs of poly from abroad, would the Chinese then forgo existing contracts, faced with almost certain legal action? Finally, in the background of negativity caused by American levies, which already divided the industry, would the Chinese punish their current allies, who had condemned SolarWorld’s petition, promising work towards healthy competition without national or corporate protectionism?
The direct impact of rising polysilicon cost would wipe out any benefits made to date in processing costs. While companies experienced polysilicon drops since $40s from November 2011, aiding costs reduction, polysilicon can still go down to meet the $22 spot pricing in long-term contract negotiations. A reversal of the trend would kill the margin recovery as pricing mechanisms for modules would not trigger the increase. Polysilicon pricing joined the ASP reduction almost 10 months after the first modules started to come down in price, forcing module makers to absorb costs internally for the whole period. The same would happen here.
GCL Poly is a giant of a company with 65,000 MT of poly production. Fourth in the world, OCI Chem had backed off its plans for a 44,000MT expansion, staying with 42,000MT at the end of 2011. Many saw poly global glut as the cause for the decision, but one of the outcomes is OCI’s inability to cut costs, now without new technology and added scale. While OCI struggles with costs ranging between of $25 to $30 per kg, GCL announced expected costs of $18.50 for Q2 of this year. Certainly at any point of this debacle GCL is clearly making money. LDK and Daqo are definitely another story. While in Q4 LDK paid for its own poly at $42 per kg, it is unclear what costs the company had at the time. Using the example of Daqo, those are at $30 plus per watt, so both companies are simply outside of the competitive range and chances are they will never reach GCL’s levels. Taiwanese news outlet “Digitimes” quoted industry sources who expect to see a price of $25 per kg; after the initial shock of polysilicon duties, this would certainly keep both Daqo and LDK outside the benefit of duties.
Both US companies Hemlock Semi and REC USA had been very vocal against American antidumping duties. Outside of the common-sense approach on global competitive edge and free trade on solar energy, both had feared a possible loss of the largest market for their poly due to retaliation caused by the “irresponsible” decision made by the US. If both companies are selling cheap poly to China, they are helping the industry. Low spot pricing is essential in negotiating low contractual pricing and this is essential for low production costs. Using the examples of Trina's and Yingli’s costs, they have not reached desirable levels. European supplier Wacker Chemie, with 47,000MT, has contracts with China as well. The company’s 18,000MT plant in Charleston, Tennessee, could be drawing duties by the time of its ramp up in late 2013, derailing possible production cost benefits.
The Chinese government has been supportive of the solar industry and has been very sensitive to ensuring that leading companies are able to lead. The same government established standard rules for polysilicon production by naming fewer than 30 companies as the benchmark to follow. All three alleged petitioners are on the list. However, there is no doubt that Suntech, Trina and Yingli have at least as much, if not more attention.
In this case, unlike the American decision – which was political in nature ̶ the Chinese government will make a decision based on economies of scale and sustainable benefits for China’s solar industry. Therefore it seems that the decision is rather simple. If GCL production costs are that low, duties will have no difference to GCL, and if the industry sources quoted by Digitimes have it right: they would not save Daqo or LDK. However, when enforced, it would add another crippling effect to costs for everyone else, something which nobody who cares for the industry would like to see done.