Since the inception of this website, we have been positive on Trina Solar (TSL). We have also noticed that it trades at a significant discount to most of its peers, and we could not really come up with any convincing reasons why this should be the case. But we keep looking.

We had an earlier attempt here. Now, we come across another article that could contain negative implications for TSL see two other potential candidates.

The article, although not mentioning TSL, is arguing that the expected price declines in polysilicon, the crucial ingredient for the mainstream cell makers, might not be happening to the extent (or within the time frame) expected.

We immediately have a problem with that article, as it does not say a lot beyond that prices might rise 5-10% this year (exactly which prices the article is talking about is actually pretty interesting, see below). For a start, there are few who expected the price declines this year, and in addition to that, TSL itself has expressed similar expectations.

However, the article is touting LDK, another solar player, for building its own polysilicon plant. The argument tries to negate earlier criticism saying this plan would not be wise because when the plant starts producing, polysilicon prices would have become more plentiful and started their expected descend.

Well, TSL had similar plans, and they were equally criticized for it. In fact, when they abandoned them earlier this year the shares ‘suffered’ a relief rally, as these same worries (having a polysilicon plant start production just when prices start tumbling) plagued TSL shares.

You will also have to consider the fact that TSL’s polysilicon plant plans were a lot less concrete (it was just that, a plan,), hence much more into the future compared to LDK’s (they are already building it), which would have made it much more likely for the TSL plant to start producing only at a time when polysilicon is already plentiful and a lot cheaper than today, eroding much of the rationale for building the plant in the first place.

An added worry was of course the needed finance (approximately $1B, by no means a negligible amount, and a multiple of those short-term debts another scary article worries about and which we’ll get into in the near future).

Now, if the market starts to think that building a polysilicon plant is a good idea after all this could matter. The only way it could reasonably change its mind is if polysilicon is indeed remaining in short supply, and hence prices remain sky-high.

Building its own plant would not only sit well with the integrated model that TSL is pursuing, as it is basically the only step in the value chain (apart from constructing the pannels on roofs) that it is not involved with (although it does also recycle it out of scrap material).

It could also be that as they were counting on getting their own polysilicon plant until recently, they might not have secured enough polysilicon supplies as a result, so TSL would be extra vulnerable to continued polysilicon shortage and resulting high prices . This reason rapidly disappears when you realize that the poly plant they were planning would start production… somewhere in 2012…

By most accounts, that really is a time when the polysilicon plants that are now in the planning/construction phases will have come online, easing the shortages and lowering prices.

More importantly, the author seems to have made a pretty crucial mistake, as we wondered about rising cost as a reason for rising polysilicon prices. That seems odd. Why? Well, polysilicon is a commodity, and scarcity is such that the price of producing it is a fraction of the production cost (in the order of 10%). It’s not dissimilar to the difference between the cost of pumping Saudi oil and the prices it fetches on the markets.

So, how come ‘cost increases’ are such that they have to be compensated by higher sales prices?

The original source is an EETimes article does not, in fact, talk about polysilicon, but about waver suppliers increasing prices due to cost. Now, those rising cost could still mean polysilicon, but no, when you read the article, these do not seem the rising cost that are paraded as a reason for the price hikes of the waver suppliers.

What they are talking about here are:

rising costs of materials – such as steel wire –and electricity which has been driven up by the         hikes in oil prices.

So, in fact, the weak point the article was making about rising polysilicon cost is actually completely disappearing. By the way, good job TSL produces these wavers in-house, one of the advantages of it’s integrated model that involves every step bar polysilicon production.