Bioenergy, Geothermal, Hydropower, Solar

Capacity, prices, and all that solar jazz

Navigant Consulting’s Paula Mints examines trends in capacity growth and prices, and reiterates the industry’s need to wean itself from incentives, to suffer a short-term expense hit to reap long-term benefits.

by Paula Mints, Navigant Consulting

This article, excerpted from Navigant’s solar services program annual Manufacturer Shipments report, is dedicated to the pioneering photovoltaic industry in Japan.

March 21, 2011 – Despite anxiety over future FiT degressions, shipments in 2010 grew at 120% over 2009, from 7.9GWp to 17.4GWp. Europe consumed 82% of all shipped technology, for 14.2GWp. Europe will continue as the largest market for photovoltaic technologies in 2011, with rapidly decreasing tariff rates straining margins for manufacturers and system integrators.

In 2010, and continuing into 2011, there is significant crystalline silicon (c-Si) product available at prices as low as, or in some cases lower than, could be produced by in house manufacturing. Though it is not intuitive, manufacturers will add capacity even when it is not fully utilized, and even when outsourcing is cheaper. In many cases, manufacturers add capacity so that they can be ready to meet demand when it comes. In the case of PV, it is not build it and they will come, but build it for when they come.

Historically (and logically), manufacturers have increased capacity significantly following years of strong demand, and added capacity more slowly (sometimes changing plans to add capacity) after years of low capacity utilization. Examples are:

  • In 2000, capacity utilization was strong at 77%.

  • In 2001, manufacturers increased capacity by 54% and had utilization of 70%.

  • In 2002, after a year of strong utilization, manufacturers increased capacity by 45% and had utilization of 69%.

  • In 2003, with capacity utilization slowing slightly, manufacturers increased capacity by 38% and had utilization of 67%.

  • In 2004, demand began to increase strongly along with raw material concerns. Manufacturers increased capacity significantly by 46% and had 72% capacity utilization.

  • In 2005, following a strong 2004 with high utilization and very strong demand, manufacturers increased capacity by 58% and had utilization of 61%.

  • In 2006, manufacturers responded to low utilization the previous year by increasing capacity by 39% and had utilization of 62%.

  • In 2007, responding to continued strong demand — and despite low capacity utilization — manufacturers increased capacity by 70% to 5442.6MWp, and had utilization of 56%.

  • In 2008, total industry capacity increased by 82%; even in a year of significant shipment growth (79%), capacity utilization was extremely low at 56%.

  • In 2009, total industry capacity increased by 37%, with capacity utilization at 58%, a slight improvement over 2008.

  • In 2010, capacity increased by 64% over 2009, with utilization of 77% and shipments increasing by 120% to 17.4GWp. China and Taiwan held 53% of total industry capacity in 2010.

  • Given the strong growth of the past few years, and despite announced decreases in feed in tariff rates, capacity increases are already underway that will result in ~57% increase in total industry capacity, to >35GWp, from the current 22.3GWp. This expectation of significant over capacity will pressure cell and modules prices further down.

Figure 1 provides a picture of regional capacity growth from 1995 through 2010:

Figure 1: PV industry capacity increases by region, 1997-2010.



Capacity up, incentives down, prices down

It is important consider the cost of unused capacity, in general terms, to both thin-film and c-Si manufacturers. Unused capacity can be a significant cost for manufacturers, as it typically leaves insufficient revenue to cover fixed costs such as investments in capital equipment and real estate. How sensitive a manufacturer is to the cost of carrying unused capacity depends upon the cost structure of the facility. For an operation with a higher proportion of variable costs, such as labor and raw materials, total cost is more elastic and decreases with output. The reverse is true for operations with a high initial capital expense, because the costs of operating the facility do not decrease as significantly as output is reduced. The minimum utilization a facility must maintain to cover its fixed costs increases as the ratio of fixed to variable costs increases. Facilities with a higher proportion of fixed costs need to maintain a higher rate utilization to pay off capital investments.

Lecture sort of over — significant capacity, lower incentives, and a changing industry landscape are driving prices down in the near- and mid-term, likely leading to some interesting changes in the long-term.

First, the multi-megawatt installation (investment or utility owned systems) does force the participants, from manufacturers of technology through to buyers of electricity, into commoditization. Electricity is a commodity, and this is what multi-megawatt systems are selling. The future for this large segment of the PV industry is lower component costs (the module is a component of the system), and future cost reductions will come from system design innovations along with installing it all efficiently/quickly — that is, the faster and cheaper, the better. To serve this application, manufacturers of technology pursuing a vertical integration (system) strategy will have an advantage.

Second, photovoltaic systems sold to end-users (residential and small/medium commercial) may in the future be able to price accordingly. That also means the industry must learn how to market the joys of owning-the-means-of-electricity-production to these end-users, to whom the photovoltaic system is both an investment and an appliance.

Finally, when the building integrated PV (BIPV) application finally matures, PV for this segment will be another building material, and will need to adhere to the building industries expectations in terms of margin.


Et tu, price?

As in 2009, a highly competitive technology pricing continued in 2010, along with constrained margins. Prices ticked up in 3Q10 due to a combination of high demand and a non-significant shortage of polysilicon. In December, and continuing into 2011, prices decreased in response to slow demand (typical for the beginning of 1Q) and in response to rapidly decreasing tariff rates. The effect of lower tariffs and other incentives on prices in 2011 and going forward should not be underestimated, as this situation along with high levels of manufacturing capacity will pressure pricing downward, leading to an unprofitable trend.

A pricing summary for 2010:

  • Thin-film ASPs fell 4% from $1.65/Wp in 2009 to $1.58/Wp in 2010;

  • Cell ASPs were flat vs. 2009, at $1.25/Wp — even after an 8% decrease to an average of $1.15/Wp at the end of the year;

  • Module ASPs for large-quantity buyers decreased by 25%, from $2.18/Wp to $1.64/Wp;

  • Module ASPs for mid-level buyers decreased by 16% over 2009, from $2.82/Wp to $2.36/Wp;

  • Module ASPs for small-quantity buyers fell 21% from $3.68/Wp to $2.90/Wp.

Noting that regional and country average prices can hide significant nuance, average technology prices by region, based on revenues are:

  • US: $1.76/Wp

  • Japan: $2.40/Wp

  • Europe: $1.68/Wp

  • ROW: $2.05/Wp

  • China/Taiwan: $1.62/Wp

PV industry pricing has historically not been cost-based. In fact, there have been long stretches during which manufacturers priced technology at or below the cost of production. The reason for this seemingly non-rational behavior is that industry demand was (and is) incentive-driven. During the early years (the 1970s through the mid-2000s), demand was sporadic at best, even with incentives, and primarily limited to demonstration projects and early adopters. PV manufacturers were unprofitable until 2004. In the mid-2000s, the feed-in tariff incentive model — where utilities are mandated to pay a tariff for solar electricity that is fed into the electricity grid — began to change the demand paradigm from push (into the market) to pull (into the market). From 2005-2010, PV industry demand grew by a compound annual rate of 65%. Given the fact that incentives are still required, there is still not, technically, true pull in the market.

Unfortunately, incentives are extremely expensive to support, and, one way or another someone must pay the bill. Since 2008, the industry has experienced annual boom and bust cycles in its markets. In reaction to the boom cycles, governments have begun decreasing and rethinking the industry’s necessary incentive rates and how these programs are administered.

Figure 2 offers a graphical view of module pricing trends, with an estimate for 2011.

 
 Figure 2: Module pricing trends 1985-2010 (and estimate for 2011), in current $/Wp.


Lecture not quite over

Like it or not, as long as the PV industry remains incentive-driven it has little control over its future in terms of prices, margins and frankly, markets. Work continues to reduce manufacturing costs (more cushion in the margin) and increase incentives, and less publicized work also continues on balance-of-system (BoS), system design, and installation along with business model innovations. Call it “grid parity” — or more accurately, a confluence of innovations, incremental technology improvements, and business strategies — the PV industry needs to wean itself from incentives (necessary as these market tools are currently) and become self-sufficient so that it can control its own future and stave off destructive boom and bust cycles. And yes, since conventional energy technologies, and nuclear energy, enjoy significant incentives, it is not fair.

Finally, in the future, the energy production pie should show renewables as having the largest share. It is imperative that governments and energy end-users commit to this path even though it is an expensive one — the expense is short-term, while the benefits are long-term. This means solar, wind, small hydro, biomass, geothermal, and ocean energy. And, stepping out for once to make a personal statement, it does not mean clean coal or nuclear energy.


Paula Mints is principal analyst, PV Services Program, and director in the energy practice at Navigant Consulting. E-mail: [email protected].