(This article was published in Spanish in Cotizalia on June 23rd 2011)
Solar stocks are the worst performers again, despite the Fukushima hype and extremely generous tariff support. The solar sector (January to June) is down 10% versus the renewables sector down 5.8%, and some of the most abrupt moves in solar are in Q-Cells -33%, REC -41.5%, Phoenix Solar -26%, Renesola -49%, E-ton Solar -50%, China Sunergy -56%.
The reason why the sector continues to collapse like a falling knife is that it has not proven how they can make any money unless there is dreamtime-like expansion.
We tend to believe that tariffs that support volume growth but cut prices more rapidly will benefit the industry (and equities) as a whole more, but hurt higher cost participants sooner. It is wrong. Because the sector starts from overcapacity and excessive cost and even with costs falling 39%, and subsidies that guarantee on average €180/MWh they still lose money.
Developers in regional markets in some cases would prefer capacity caps with higher tariffs, thus enjoying the spoils of overcapacity-driven solar ASP (average selling price) pressures from their suppliers. The Asian manufacturers have zero discipline in capex or growth, they need only a 3-month price signal and they further expand, which virtuously keeps driving down the cost. Something like 10GW of new manufacturing capacity (40%) is coming online during 2011, while demand (global) is expected to rise about 15% at best.
The big ‘if’ is if Italy imposes an annual 2-GW/yr cap on installations; if they do, most of the marginal players in Asia will have nowhere to put their last 20% of production and we have a rapid fall in prices, margins and profits which then will lead to the proverbial ‘boom/bust’. But before then, we would expect margins to compress by 50% or more in some cases in the food chain, as we saw in wind–could see stocks fall 20-40% over the summer.
If the industry were logical and wanted to lower the cost of solar – what should it do? If you assume 1600 kWh/kW capacity factor, €35c/kWh FiT (Italy), and assume that the installations are made from 2012-2016, the €6bb/year economic burden by 2016 translates to ~2.1GW/year from 2012-16.
If the industry instead accepted a €25c/kWh FiT (~30% cut), the industry could support a 3GW/year cap at the same economic burden. If the industry accepted a €20c/kWh FiT (~40-45% cut), the industry could support 3.75GW/year. If the industry accepted €15c/kW FiT (55-60% FiT cut), the industry could support 5GW/year. If the industry accepted a 55-60% FIT cut and used that as a basis to support a higher economic burden of €6.9bb/year, it could support a 5.8GW/year limit for PV.
Will they?. I doubt it. The Chinese companies have 50GW of capacity for a 20GW market. Companies are unable to achieve IRRs of 10% with subsidies of $140/MWh (US), €230/MWh (Germany) or, brace yourselves, €400/MWh (Spain), all PV solar. This is, on average between 4 to 10x the average price of power in any of those contries… and companies still are unable to make more than 10-12% IRR despite a fall in costs of 38-40%!!
In the first quarter of 2011, the average selling price (ASP) of solar modules fell again like a rock. ASPs fell c45% globally while costs fell a slower ~37% pace. So in essence, prices fall more than costs and, as overcapacity grows, the downward trend is exacerbated by working capital requirements. A race to zero? probably, unless capacity retirements really start and companies streamline their debt. But capacity cannot be slashed when the main competitors in all parts of the chain, Chinese in particular, but Germans too, have a capital allocation policy based on growth at any cost.
At the heart of this we have an over-leveraged industry: 80-85% at project level is already unsustainable, but the companies are leveraged also at the holding level, making total gearing to the tune of 5-6x Net Debt to EBITDA. Overcapacity in all parts of the value chain, from wafers to poly-sylicon and development stands at more than 50-60% on average, and competitors (not just Chinese, Germans too) show low capital discipline and imperialistic market share aspirations. So at the minimum price signal (a tariff announcement, a government renewable plan confirmation), capacity grows way above demand. And despite the fall in costs, it is astonishing to see the companies unable to get their head above water. Working capital requirements eat any equity IRRs and debt obliterates the profitability. Add to this overcapacity and no wonder you ave seen 40-50% falls in stock prices in solar.
The Asian build more and more capacity, however some part of the production chain are much easier (from a technology and timing stand point) and cheaper to add than others. To simplify, the lower in the value, the easier/cheaper it is to add capacity. As a result, the value chain today would look like an inverse pyramid with more capacity downstream than upstream:
From upstream to downstream the value chain is:
As cell capacity grew larger than wafer, demand for wafer was tight and wafer ASP shot up, squeezing cell margins. Then with this positive price signal at the wafer level, Asians have added wafer capacity. As a results, there were more wafer capacity than polysilicon available. Wafer guys margin got squeezed and polysilicon guys reaped all the profit. In Q1 11, the CAPEX for polysilicon for Asian companies could be recouped within one year as spot price for polysilicon was so high (gross margin >50%). Now overcapacity looms, gross margin collapsed and we are at the tipping point where polysilicon companies still need to lower their margins otherwise the whole industry will come to a halt. Presently, wafer and cells producers are barely earning a gross profit.
If you wanted to be cynical, you could say that the Germans and Spaniards actually tricked the Chinese by sending positive price signals for a few years, enticing the Asian the build, build and build, and now that there is massive overcapacity, cost of PV is likely to be close to grid parity. But companies are in dire straits and making poor returns, so any cut in tariffs obliterates them.
As the ability of governments to keep those supernormal subsidies dies (85% of subsidies for solar are in Europe, mostly Germany), the industry is condemned to a massive re-structuring.
Solar returns are collapsing DUE to the big subsidies. Subsidies created this industry and are now part of its demise. They created artificial signals for demand and then ultimately supply. Now, supply is in trouble due to overbuilding/expansions. Is it temporary? Maybe, and if “temporary” means 5 years it will be death for many, but margin compression is not.
Solar energy is not a problem as a concept. Greed fuelled by government-led price signals, that fade as quickly as they come, added to unsustainable debt and undisciplined capital allocation is the problem.
One thought on “The Debacle of Solar and ‘Tariff Support’ Models”
Julio 2012: módulos debajo de 45c€/W, EPC financiable (plantas >1MW) en torno a 0.9 – 1.1 €/MW. Resumen: LCDE en la horquilla de 35 – 65 €/MWh para sitios con radiaciones normales, con esto ¿quién quiere ser suministrado por una comercializadora pagando el doble, y subiendo?, porque no comprar solo a tu distribuidora las horas que te falten?. Lo mismo le puedo decir en casos de mini-cogeneración (igual si solo hablo de solar deja de leer el comentario). El caso es que las tecnologías de generación mejoran sus prestaciones y bajan los precios espectacularmente, en el lado opuesto las utilities son dependientes de un modelo de negocio que se basa en su cohabitación con los estados por lo que no les queda más remedio que pasar por el aro de tener precios de mercado inflados a impuestos (caso español 50 – 50 coste fijo del sistema – pool). El horizonte de precios que se puede esperar del sistema “tradicional” es precios al alza (impuestos + combustibles), sin embargo en el caso de sistemas de generación distribuida es el caso contrario. Llámeme iluso, ¿estamos ante un cambio de paradigma?, seguramente no en su totalidad (la sociedad avanza en complejidad, porque aquí iba a ser menos), pero si en partes. Así por lo menos lo piensa algún centro decisor en EEUU, se lo aseguro.