Solar Industry on a Roller Coaster

Polysilicon spot prices have dipped below $40 per kg.

Yesterday, summarizing insights gained from last week’s Solar Power International (SPI) conference, Mercom Capital Group forecast a “Poly Price Collapse”.

Crisis? Or not?

While poly production may not yet be in crisis, red ink in downstream PV module production has become a critical industry issue.

Last week, Mehdi Hosseini of Susquehanna Financial Group observed “we expect the majority of module manufacturers to run into operating losses.” Underscoring the concern, Hosseini added that “the resale value of the installed (solar module manufacturing) equipment is, in our view, nil.”

Poly pricing a silver lining?

Are lower polysilicon prices a cause for alarm? Or, can poly prices drop further and provide some relief for the beleaguered module producers?

For months now, Credit Suisse has forecast polysilicon pricing of $30 per kg by year end. That threshold may yet be achieved.

At $40, polysilicon has returned to its decades-long, historic pricing levels. Fundamentally, current poly price levels should not be cause for alarm.

Even with all of the poly manufacturing expansion over the last few years, at least seven of today’s major poly producers are expected to be operating with positive operating margins. They are in a much stronger position than their customers find themselves.

Will lower poly prices provide any sort of buffer for the module manufacturing sector?

Lower polysilicon prices are definitely a benefit against thin film module competition.

There is no question that lower polysilicon pricing, along with ongoing increases crystalline module efficiency, have been the two jaws in the vice squeezing the prospects for thin film alternatives such as Solyndra and AMAT’s Sunfab. According to Susquehanna Capital, this week’s abrupt departure of First Solar CEO Rob Gillette may be at least indirectly related to First Solar’s less promising performance on cost cutting and efficiency improvements.

New poly expansions?

The drop in poly pricing may not yet herald the same negative margins as we are seeing in module manufacturing. Notably, much of the Asian manufacturing expansion in the PV industry has been driven by a relative flight-to-margin economic decision-making criteria.

The PV industry has offered relatively higher margins and much higher growth than manufacturing in old industrial sectors like hard goods and clothing. Margins in PV manufacturing have also compared favorably in newer sectors such as the computer and monitor industries. This is one of the economic perspective subtleties that help explain the PV industry expansion in Asia.

If similar decision-making criteria, such as relative operating margin, that produced the massive Asian manufacturing presence are applied to polysilicon, we may yet see even more polysilicon manufacturing newcomers.

Then again, another part of the investment justification for the expansion in module manufacturing was a belief that the manufacturing equipment would minimally retain its value. In other words, there was a belief that capital invested in manufacturing equipment could easily be fully recaptured. As Hosseini just observed, however, “installed equipment no matter how new has little value in the used market.” This is a new reality for the PV industry.

As with module manufacturing, further price pressure on polysilicon may begin to erode the underlying polysilicon equipment values.

What now for Polysilicon?

What are the anticipated impacts of lower pricing in polysilicon manufacturing?

With massive production expansions and the emergence of new, high-volume entrants in Asia, surely the prospect of a return to historic poly pricing norms was not unanticipated.

On the one hand, the major poly players still represent the profit center in the industry.

Strengthening their competitive position, as illustrated in The Economics of Efficiency in PV Systems, intense cost pressure on poly customers makes the value of efficiency – and fundamental product quality – much more pronounced.

The lower the pricing, marginal differences in efficiency make an even bigger impact on overall value. The highest quality poly from dependable producers is even more attractive now. The statement, “At these prices, why would you buy low grade poly?” is a common industry refrain.

On the other hand, there is an industry oversupply situation, intense cost pressure on poly customers, and there are many players – especially in China – with uneconomic processes (and with little industry knowledge and weak customer relationships). They should not survive.

Moreover, margin pressure will force many poly customers to look for the lowest cost, but highest quality, alternative. Small cost reductions are magnified when the value chain has such tight margins. Thus, there will be much more margin pressure and even more contract breaking as we recently saw with Suntech and MEMC.

Big producers will feel the pinch

Moreover, a soft market is always hardest on the largest producers as they require high capacity utilization to maintain favorable economics. Look for aggressive marketing in the months ahead, especially from the largest of the new players. The strongest initial reaction may be to try and flood the market in order to maintain market share. That is likely to backfire, however, especially in the current market.

Polysilicon manufacturing equipment providers may also feel increasing margin and sales pressure.

Even as the third-party suppliers have ramped up, the polysilicon manufacturing cost leaders are those companies, like Hemlock, Wacker and REC, that have their own proprietary equipment.

If cell and module manufacturing equipment has no balance sheet value, does the same fate lie ahead for third-party polysilicon equipment?

Is this good for UMG polysilicon?

Oddly, perhaps, the challenging economics may stimulate a renewed interest in alternate polysilicon production strategy such as UMG. Though there will be very little tolerance for processes that cannot offer realistic near term economic relief.

Long held as the disruptive innovation in polysilicon, can the current conditions actually provide the opportunity that UMG is looking for?

Recent reports from a variety of sources underscore that UMG can provide equivalent performance to conventional polysilicon. UMG has come a long way. With the right process, UMG can be much more cost-effective to produce. Some market players are going to be looking for the most affordable polysilicon supply. Moreover, the expansion costs can be much lower for UMG than high purity polysilicon.

Stimulated demand for UMG at this point could create an even larger capacity oversupply. UMG capacity can be added quickly, and the process can be scaled back to follow demand without the same economic consequences as the large Siemens process facilities of the incumbent polysilicon producers.

But, it is very clear that simple cost reduction and comparable performance may not be enough to compete against low-cost, ultra-pure polysilicon.

UMG producers will have to offer some degree of strategic or synergistic value to make any dent in the market, even if their economics are compelling.

Your turn

The industry remains dynamic. What are your thoughts?


7 Responses to “Another PV Solar Polysilicon Crisis Looming?”

  1. Gopala Krishna Murhy says:

    One scenario is that the major poly manufacturers will form a cartel, control production marketing and sales of poly at price which is remunerative to them just like OPEC. This cartel will ensure that all smaller manufacturers disappear soon because of cost factors and then start dictating costs of poly. In the ninetees, such a cartel ensured that a small polysilicon company was closed by offering rediculously low priced poly to it. After a few years, ensured that the small plant would not restart, the price of poly supplied to this company was hiked to normal values.

  2. Can you provide a reference to those “recent reports from a variety of sources underscore that UMG can provide equivalent performance to conventional polysilicon.” I have seen this, but I am not convinced given Q-Cells’ troubles.

  3. Hi ECD Fan,

    Good question. Your cited source, itself, provides some illumination:

    “2010 Slope of enlightenment main questions in utilisation of umg-Si clarified”

    “The efficiencies of solar cells are further following and approaching the efficiencies of cells made from poly-Si, in standard production lines (a comparable efficiency of) 16,4% can be obtained Umg-Si is therefore an interesting alternative to poly-Si as long as the quality is above and the price below a certain threshold”

    At Intersolar North America, where some of these results were discussed, there was also discussion of superior performance to regular poly in low light conditions. No explanation, however, was provided at the time.

  4. Is this good for UMG-Si? My guess is no, given that diminishing margins would seem to favor vertical integration versus introducing “low-cost” poly-Si into the mix. For project developers, if you can incredible pricing on Tier 1 bankable modules, why would they look at modules made with UMG-Si? I’d agree that UMG-Si needs to bring an additional value proposition to the market besides just price.

    • Hi Richard,

      On the one hand, you are probably right. This situation provides a more difficult challenge for UMG as it does for any new poly market entrant.

      On the other hand, there is an argument that UMG is further facilitated by this.

      Please consider the following If Conditions:

      1. If UMG can be used interchangeably with regular polysilicon with nominal performance impact or improvement
      2. If UMG can be produced at equal or lower costs than regular polysilicon
      3. If UMG can provide integration benefits such as more cost-effective recycling from ingoting and wafering operations


      4. If UMG capacity can be added at the same or lower cost, and in the same or faster timeframes

      Then UMG should have a compelling advantage as value-chain costs are being squeezed.

      If UMG meets If Condition #1, and at least one of the other conditions, it should have a competitive chance.

      While there is an argument regarding why not buy cheap, high quality polysilicon when prices are low, we have to consider the current economic environment and not the previous high priced environment. Cheap poly now, may still be expensive poly.

      When margins are being squeezed, margin relief is critical.

      There are negligible operating synergies between polysilicon production through chemical vapor deposition (‘Siemens’ process) or the FBR process. There are many lessons where integrated approaches suffer when each process is on a different technology improvement path.

      Why ‘integrate’ when there is no underlying process improvement or cost improvement benefit?

      Can any PV company really afford to continuously upgrade each and every process step in the value chain to stay abreast of the rapid pace of evolution? With depressed stock valuations and highly leveraged balance sheets, how are they going to raise the capital do accomplish this?

      It is much more difficult to optimize each element in an integrated operation, than to do that for a single activity. The auto industry offers a great example of increasing specialization into maturity.

      And, given the lack of operating synergy for conventional production, if integration is a key driver then the integration opportunities for UMG should strongly be considered.

      There are currently two commercial UMG producers of note. Both of these processes meet at least two of the above criteria.

      More importantly, both processes offer opportunities for #3 – process synergies. There could be a significant improvement in bottom-line performance from improved recycling for any company pursuing an integrated strategy with UMG production.

  5. Hi David

    “At these prices, why would you buy low grade poly?” Is this really a common industry refrain? You’d think they’d be looking under every stick and stone for ways to reduce costs.

    Couple questions. Is the purity of the new poly the same as the legacy poly. My understanding is that the refineries were built with the capability of nine nines but you could tune them so that SGS was produced. The idea behind the tuning is that you speed up the process and maybe reduce your production costs by sacrifing a few nines of purity that don’t necessarily hurt your cell efficiency too much. You can tell me if I’m way off base. I’m a big boy.

    Here’s something that looks interesting and really caught me by surprise. I’d think this OCF process would be a natural fit for slightly lower purity poly.

    Here I am imagining UMG put through a monticrystalline wafering process and then upgraded with an Optical Cavity Furnace. I wonder what kinda performance you’d get outta those cells.