Illustration by Hugh Frost
Energy Subsidies Stymie Wind, Solar Innovation: Nathan Myhrvold
This month, the U.S. Department of Commerce launched a formal investigation into complaints, lodged by the U.S. solar-cell manufacturers, that the government of China is funneling loan guarantees, grants and subsidies to its solar-cell companies.
Apparently, the Commerce Department is shocked, shocked to learn that a government would subsidize the solar industry.
A few days later, the New York Times described a “gold rush” under way in the U.S. as builders of wind and solar farms cash in on grants and loan guarantees offered by both the federal government and various states. These incentives effectively allow players at every level of the renewable-energy industry to lock in profits of 10 percent to 30 percent a year for the 20- to 30-year life of their plants -- not bad considering 10-year Treasury bonds are paying only 2 percent.
Both of these developments are symptoms of a larger problem with the world’s current approach to renewable energy. The range of prospects being tried now is dizzying -- from high-tech windmills to biofuels, from corn to algae, from silicon photovoltaic cells to boilers heated by thousands of reflected sunbeams. But they all share one thing that makes them appealing to investors: taxpayer dollars. One of the ugly secrets of the renewable-energy industry is that its products make no economic sense unless they are highly subsidized.
‘Levelized’ Energy Cost
To get a sense of just how deep these subsidies run, I looked up projections from the U.S. Energy Information Administration of the “levelized cost” of energy from various power sources. Let’s say you decide to build a new electricity plant that will come online five years from now. Before getting a loan, you’ll want to estimate how much money you will have to lay out for construction, operations and maintenance, fuel, interest payments, insurance and so forth over the 30-year lifetime of your plant. Divide that total investment by the amount of electricity the plant is likely to sell, and you get a break-even price per kilowatt-hour. That’s the levelized energy cost.
In its most recent projections, the EIA gave renewables every benefit of the doubt. Its projections assumed, with remarkable optimism, that regulators will impose what amounts to a carbon tax of $15 per ton of carbon dioxide emitted by power plants. And they ignored the costs imposed on the larger electrical system by the highly variable output from solar and wind farms, which thus require backup energy from gas, coal or nuclear plants. Even so, the EIA numbers show that renewable energy of every kind is still far from being economically viable.
At 6.3 cents a kilowatt-hour, the levelized cost of natural-gas generation comes out the lowest, mostly thanks to the surge in cheap gas recently unlocked by hydraulic fracturing. Coal is the next cheapest source, at 9.5 cents per kilowatt-hour.
Among renewable-energy sources, wind turbines on land are the most viable, with a levelized cost per kilowatt-hour averaging 9.7 cents, but ranging (under the EIA’s optimistic assumptions) from 8.2 cents at the very windiest sites to 11.5 cents where conditions are less favorable. So wind power should command a 54 percent premium to electricity from natural gas -- and that price difference widens to 73 percent if the putative carbon tax fails to materialize (see table).
Watts from offshore wind turbines are 290 percent higher in price than those made from natural gas; solar photovoltaic is 230 percent more expensive; and the cost premium for solar thermal soars to almost 400 percent. Clearly, outside of exceptional circumstances, only enormous subsidies from taxpayers can keep these technologies competitive.
So why are they being deployed at all? The main argument -- often made by renewable-energy entrepreneurs and others who think that the U.S. should lead the world in “green” jobs -- is that once we start to produce renewables in volume, their costs will come down. So governments should merely subsidize today’s projects enough to push the technologies over their short-term humps.
This learning-curve effect is well-known in the electronics industry. And it’s true that incremental improvements in the design and production of solar cells and wind turbines during the past decade have shaved costs and lifted efficiencies. The cost gap between renewables and fossil fuels used to be even wider than it is now.
A second argument often made in favor of subsidies stems from concern about climate change. We really ought to cut our greenhouse-gas emissions, this thinking goes, and although the best way to do that would be to either outlaw fossil-fuel use for certain purposes or to enact a high carbon tax, neither of those options is politically viable right now. Although they rarely say it explicitly, many environmentalists embrace government subsidies and mandates for renewables as a politically expedient, second-best alternative to a ban or tax on carbon emissions. Green-jobs advocates have given them a convenient argument that sells a lot better than a tax.
But these arguments ignore the other, more productive uses that could be made of the money spent on subsidies. Rather than accelerate deployment of technologies that we know are inefficient, wouldn’t it be better to invest in the research and development that are needed to come up with renewable technologies that are cheap as well as clean? Subsidies also reward inefficiency, cultivate dependency on government largess and promote a rush to manufacture before the hard work of perfecting a technology is done.
As for the learning-curve effect, the electronics industry has demonstrated that when a technology has true potential, subsidies aren’t needed. Each step of the way, electronics technology paid for itself -- as well as for the R&D that led to successive generations.
This was certainly true for microprocessors and memory chips, whose cost per transistor or per bit of storage space consistently fell by half every year to 18 months. Solar-cell makers and their champions love to talk about a similar “Moore’s law” for solar, but the physics of the situation argues against this. Yes, since 1985, solar-cell prices have dropped by roughly a factor of four, but during that same period, Moore’s law slashed the price of computational power by a factor of at least 130,000.
Better Technology Ahead
Solar cells being deployed today convert into electricity just 10 percent to 15 percent of the solar energy that strikes them. Laboratories have made cells that can reach 40 percent efficiency, a level where things start to get interesting. But these advanced cells are expensive and tricky to make. A lot more R&D must be done before they can be produced cheaply and in volume.
Unfortunately, to compete in the gold rush for government support, the solar-energy industry mainly spends its time making current-generation, inefficient cells. The argument that this will help accelerate the next generation just isn’t holding up. It’s a bit like the old joke that you can lose money on every sale but somehow make it all up on volume.
Which brings us back to that Commerce Department investigation into China’s alleged handouts to solar-panel makers. If the charges are true, think about where those cheap solar panels end up: Many of them are being sold to utilities in the U.S., which are building solar-power plants that are themselves subsidized by the U.S. government. So at worst, the net effect is that the Chinese government is helping American taxpayers get more power plants for their money. Why is that such a bad thing?
Some people fret that China will reap the green jobs of the future, but no economically viable green-energy product exists. It makes no sense for the U.S. to try to dominate a money-losing industry, especially by guaranteeing profits to inefficient power plants for 30 years.
The smarter strategy would be to boost spending on R&D toward a new generation of renewable-energy technologies that can compete in the marketplace without subsidies. A great deal of invention is required, so this won’t be easy or cheap. But it’s just the kind of challenge that American inventors and entrepreneurs have excelled at in the past.
(Nathan Myhrvold, the former chief strategist and chief technology officer at Microsoft Corp. and the founder and chief executive officer of Intellectual Ventures, is a Bloomberg View columnist. The opinions expressed are his own.)
To contact the writer of this article: Nathan Myhrvold at firstname.lastname@example.org
To contact the editor responsible for this article: Mary Duenwald at email@example.com
Bloomberg moderates all comments. Comments that are abusive or off-topic will not be posted to the site. Excessively long comments may be moderated as well. Bloomberg cannot facilitate requests to remove comments or explain individual moderation decisions.