mead cohen berger shevtsova garfinkle michta grygiel blankenhorn
Propped Up
New Solar Subsidies Come with Some Major Problems

Solar power can’t yet compete with fossil fuels on its own merit, at least to the degree that most greens want it to—it’s simply too expensive—so to kickstart the fledgling renewable energy source governments have to rely on subsidy schemes. Feed-in tariffs are one of the most popular options here; they involve guaranteeing solar producers long-term above-market rates for the electricity they contribute to grids. The problem with this route, as Germany has found to its citizens’ detriment, is that it’s expensive. Consumers typically eat the costs of these subsidies (in the form of a green surcharge on power bills in Germany’s case), and these price hikes have a tendency to disproportionately affect the poor.

There’s another form of government support that’s picking up traction, though, which involves auctioning off solar production contracts to the lowest bidder. India has aggressively pursued this strategy, but has seen companies eager to gain a foothold in what will eventually be a lucrative market go too far in this race to the bottom, raising doubts about the profitability of the auctioned-off projects. As Bloomberg reports, this isn’t an issue unique to the south Asian country:

While the new mechanism has produced a bonanza of contracts for well-capitalized developers, industry executives are concerned that many of the projects won’t make money or get built, endangering company finances and national green targets.

“You don’t want to end up in a situation where companies go bust and you have a non-sustainable way of setting the right price level in the industry,” Samuel Leupold, vice president of Denmark’s biggest utility Dong Energy A/S, said in an interview in London. […]

Michael Polsky, chief executive of one of the largest private clean-energy companies in North America, Invenergy LLC, bid unsuccessfully at Peru’s auction in March. He doubts the winning companies can profit from their offers. “None of them are financially viable,” Polsky said.

Rather than tripping over themselves trying to devise complicated new schemes to get renewables up off of the ground, policymakers would be far better served devoting their attention and their governments’ dollars towards the research and development of more efficient solar panel (or wind turbine) technologies. Greens place a high premium on sustainability but seem to forget what that word means when it comes to industry, so here’s a timely reminder: propping up energy sources incapable of expanding their market share on their own is not a formula for lasting success.

Features Icon
show comments
  • Fat_Man

    “dollars towards the research and development of more efficient solar panel (or wind turbine) technologies”

    Please stop. This canard had its wings clipped years ago. The problems with solar and wind are not efficiency. Wind is very close to as good as it can be. Solar is reasonable, were it not for the real problem, which is that the sun sets every day. No amount of research is going to change that.

    • Jim__L

      If you put your panels where the Earth’s shadow doesn’t fall on them (i.e., in orbit), you solve that problem.

    • Ofer Imanuel

      You can pump water upstream with wind / solar power when they are available. That adds significantly to the price. If you can get the total combined price down to compete with other sources of base load, you will be good to go.

      • SLEcoman

        It is impossible to get meaningfully large pumped storage projects permitted in the US. The regulatory trend has been to put more downstream flow and reservoir pool height change restrictions on existing dams, making it harder for these dams to provide energy storage services to support grid stability.

        • Ofer Imanuel

          Yuck. I was looking at it from a technical feasibility perspective, not from US regulatory nonsense. Thanks for pointing this out, though.

  • Andrew Allison

    How many nuclear power plants could have been built, or conventional ones converted to natural gas, with the subsidies thrown at solar and wind? As Fat_Man points out the latter have insurmountable consistency problems (in addition to, in the case of wind, killing a lot of birds).

    • TheRadicalModerate

      In 2013, total subsidies for renewable electricity were about $13.2B, about $8.6B in 2010,, about $5B in 2007, and a little under $2B in 1999. Linearizing those data points (two different slopes for pre- and post-Obama administration), converting to 2016 constant dollars for the past 20 years, and using a 3.5% discount rate gives you a net present value of about $78B in 2016 dollars.

      The last overnight cost data for nukes we have in the US was about $5000/W in 2016 dollars. But, since we’re imagining a world where we didn’t go full-blown crazy after Three Mile Island, let’s use the current South Korean number of about $2900/kW. So, for $86B, you could build out about 27 GW of nuclear capacity if you’d just invested the money up-front in nukes. Let’s take a 90% capacity factor and call it 24 GW online. Note that, even with magical South Korean construction efficiency, this is overnight cost; carrying costs for nuclear construction are considerably higher than renewables due to the length of construction.

      Based on a couple of tables here, it looks like we’ve added about 73 GW of wind and 13 GW of solar “nameplate” capacity. Bias that by 30% capacity factor for wind and 23% for solar, and you get an online equivalent of about 25 GW.

      In other words, nukes vs. wind/solar is about a wash.

      I’m strongly pro-nuclear, but the numbers just aren’t there for it right now. Show me a technology with an overnight cost of $1000/kW, or maybe even $1500/kW, and I’ll start getting excited. That might happen with some of the gen IV technologies, or it might not. It will almost certainly happen with fusion, but we have no idea how long it will be before that can be commercialized. (I tend to think that it’s less time than most people think, but the technology risk is huge.) Meanwhile, renewables are winning quite handily–certainly from a public popularity perspective, but the economics are pretty good, too.

      On the natural gas conversion issue, I think you’re on much firmer ground, but I’ve done as much figurin’ today as I’m willing to do.

      • Andrew Allison

        Thank you very much for answering my question! Am I right in thinking that if occasionally-on wind/solar and always-on nuclear power are a wash cost-wise, the choice is obvious?

        • TheRadicalModerate

          That’s a really hard question to answer, because it relies on the business model of the power producer. Nukes are great for baseload power, but they’re terrible as “dispatchable” resources, which provide peaking power. (They can’t change their output level quickly enough.) Generally, solar/wind aren’t considered dispatchable, because you can’t necessarily buy the power on demand if it’s dark or calm. But they’re sufficiently reliable in a lot of regions that a retail power company (I’m spacing on the technical name for this, but the company who sells power to end users) will use them as a peaker-of-first-resort.

          If everything were baseload all the time, then nukes would have a clear advantage. But a lot of the disadvantages of solar/wind can be smoothed out with software, which is very cheap. And even relatively modest amounts of storage can make renewables much lower risk.

          Note also that I was being incredibly generous with my overnight costs for nukes. On the other hand, based on the nature of your question, I punted on the whole issue of how much non-subsidized capital is at risk in a solar/wind installation. At an arm-wave, the generosities vaguely cancel each other out.

          • Andrew Allison

            Thanks again. But light and boiling water reactors do have load-following capability ( Is there enough gas turbine capacity installed to act as demand followers for the older nuclear plants. Seems as though power that be turned on regardless of whether the Sun is shining or wind blowing makes more sense. I realize we’re doing a lot of arm waving, just trying to get a rough idea of the alternatives.

          • TheRadicalModerate

            Found this, which describes how nukes do load-following in France and Germany. Note that, in the US, automatic load-following for nukes is forbidden in the license, because power output changes require a human in the loop.

            There are significant issues associated with deep reductions in nuclear output, which can cause reductions in design life (nukes are designed for a specific number of cycles in power output), and have to be considered carefully to stay inside the safety envelope.

            Nuclear power can’t just be “turned on”, so the “regardless of whether the sun is shining or the wind is blowing” part doesn’t necessarily follow.

            I’m hugely skeptical of renewables without storage to go with them. Even with storage, I’m skeptical that they can be built out as fast as we could build nukes with the proper technology and a regulatory regime designed to complement that technology. But the sad fact is that we don’t have that technology today, and even when we get it (sometime in the next 15 years, hopefully), it’s unlikely that the regulatory regime will be reformed until we have a lot of best-practice experience in operating it, which will take at least another 15 years.

            Meanwhile, because renewables don’t have a fundamental safety issue (and nukes do, even if that issue is eminently manageable), there’s an awful lot of experimenting that can be done with management practices to get them to be more reliable. Until fusion comes along–which changes everything–I suspect that that gives renewables an overwhelming market advantage over fission nukes.

          • Andrew Allison

            Thanks again for providing all this background.

          • SLEcoman

            The idea that nuclear generating units cannot follow load is not based on physics. US Navy nuclear powered aircraft carriers and submarines routinely make huge changes in power output in extremely short periods of time.

            The fact that civilian nuclear reactors cannot change load as fast is a function of design parameters, not fundamental physics. If rapid load following was important, reactors could be designed to follow load.

            On the other hand, the fact that wind and solar power cannot follow load is driven by physics, not design choices. At least solar has the advantage that peak summer demand occurs on sunny days, so peak solar power generation tends to occur when there is peak demand, contributing to grid stability. Of course, for areas where the winter peak is close to, or exceeds, the summer peak, (e.g. much of the U.S.) then solar is worthless for grid support since winter peaks typically occur from 4-8 pm. Peak wind power generation has no relationship to peak power demand so wind power tends undermine, rather than enhance, grid stability. For example, in Texas peak demand typically occurs 3-7 pm on weekdays in the summer and wind power generation is usually <20% of installed capacity at that time.

          • Andrew Allison

            I pointed out that (some) nuclear generating units are perfectly capable of load following a while back. That wind and solar are subject to the vagaries of the weather has also been discussed at length. It’s perfectly clear, to all but greens, that nuclear is the green choice. What’s not clear is whether it’s economically viable in a world awash in cheap natural gas.

          • SLEcoman

            For most of the world, coal fired generation technology is the biggest competitor to nuclear power. Even with today’s extremely low US natural gas prices, which are not sustainable, LNG on a spot basis is still $4.20-4.30/MMBtu (~$25/boe) into Japan. If US natural gas prices stabilize @ $3.50/MMBtu (~$18/boe), then all in cost (including capital recovery, taxes, and debt service) for LNG into Asia after it has been regasified will be ~$10/MMBtu (~$60/boe). Currently, Aussie steam coal delivers into Japan @ ~$2.30/MMBtu, and Indonesian coal is less expensive.

          • Andrew Allison

            Agreed, but the green choice is clearly between gas and nuclear. Of course, the developing world cares about cost, not CO2, so coal is going to remain a major part of the equation. Any idea of the relative cost of exported US coal vis a vis the others? Let me add how very much I appreciate the input from commentators who actually have useful information to share.

          • SLEcoman

            US coal exports are a tiny part of global seaborne coal trade – about 2% of thermal (steam) coal and ~7% of metallurgical (coking) coal. US coal has a hard time competing in Asia (the primary market for seaborne coal) because it is at large transportation disadvantage versus coal exports from Indonesia, Australia, Russia, and South Africa. Another factor that tends to make US coal uncompetitive is that most US coal reserves are located inland so there are usually significant costs to transport US coal to export ports. Many US coals mine mouth costs are very competitive with international mine mouth coal costs, but transportation costs trump coal costs. Its very possible for total transportation and logistics cost to be 80+% of the delivered cost of US steam coal into international destinations.

          • SLEcoman

            The US reformed its nuclear licensing process over a decade ago. Third generation Westinghouse pressurized water reactor design has been certified by the Nuclear Regulatory Commission. Several of these reactors are currently under construction. The first of these reactors will be completed in China.

            Four nuclear generating units have been permitted and are under construction in the US. After the 2012 election when President Obama touted his ‘all of the above’ energy policy, the Obama administration put a halt to licensing any additional new nuclear generating units until the nuclear waste storage issue is resolved.

            By law signed in 1983-4, the federal government is required to provide a nuclear waste storage facility by February 1, 1999. The law also required nuclear power plant owners to sign contracts with the federal government and make payments to the DOE to fund the construction of the waste disposal site. The federal government took the payments even after it knew that the waste disposal facility (Yucca Mountain) was not going to be completed anywhere near on time. Outside of government that kind of action is considered fraud and companies end up paying restitution, fines, and sometimes company executives serve prison time. In the case of the DOE, Bill Richardson subsequently became governor of New Mexico.

      • CaliforniaStark

        Interesting calculations.

        You “nameplate” capacity for renewable seems a bit on the high side, and may be from a government source. Government capacity estimates often only take into account new wind or solar. Once a wind turbine is operational, its production steadily declines. In the U.K. one study showed new land turbines at 24% capacity; in 10 years they fell to 15% capacity; then in 15 years 11%. In Denmark, one offshore wind farm went from 39% to 15% in 10 years — and building offshore turbines is not cheap. A general study of wind turbines in Germany found their capacity at about 14.88% percent — to which a wind advocate retorted that they should not have counted the “old wind turbines.” Solar has similar issues, as shown by the massive Ivanpah solar plant which is badly under-performing.

        • TheRadicalModerate

          I didn’t vet the Wikipedia numbers, but they’re from the EIA, which is indeed a government source. As a separate exercise, I worked backwards from the actual energy delivered (included as one of the Wkipedia tables), and derived nameplate numbers using the capacity factors. They came out a little lower (66 vs. 73 GW wind, 10 vs. 13 solar), but they’re still in the ballpark.

          The more serious flaw is that I didn’t even attempt to look at the full capitalized costs of the renewables vs. their subsidies, but I figured that, since I was using only about 75% of the actual US nuke overnight costs (Vogtle 3&4 in Georgia had overnight costs close to $3900/kW), it was still pretty close to a wash.

          I can’t comment on capacity factors, other than to say that operational experience tends to improve them over time. IIRC, Ivanpah’s most recent couple of quarters had dramatically better capacity factors, even though they appear to have figured things out a little too late. But best practices are likely to diffuse throughout the industry.

          Do you happen to know the reason for the tail-off in wind capacity? The PV degradations are well-understood, but it doesn’t make much sense for wind.

          • CaliforniaStark

            Your numbers appear sound. Suspected the capacity numbers were from the EIA, and reflect newly installed capacity. Agree with your conclusion, the economics between nuclear and renewable could be a wash right now.

            Had someone once explain to me why there is a tail off in wind capacity; it was too technical for me to repeat. Basically they wear out with use. It hits all wind turbines. An added cost that might not have been in your calculations involves the expense of taking down abandoned wind turbines. It took an effort to find funding and the party responsible for taking out the old wind turbines in Tehachapi, California and Kamaoa, Hawaii.

            Ivanpah is doing better, although they recently had a fire in one tower that set them back. However, it will never produce the amount of power originally projected; and it is using a significant amount of natural gas now. One reason for the lower output was the original estimated output was based on solar power being generated in the early morning hours, when the sun was not high enough above the horizon to hit the mirrors. It was not physically possible; and so they now power the turbines during the early morning with natural gas. Its embarrassing when your marque solar thermal plant needs to buy carbon credits. For the cost of Ivanpah, they likely could have built four natural gas combined cycle plants, which would generate three or four times the electricity of Ivanpah, and may cumulatively have generated less carbon emissions.

      • SLEcoman

        The problem with this analysis is that there is no value placed on dependability, and dependability is the most important factor in evaluating power generating source economics because loss of electricity supply, especially sudden loss of electric supply, has huge societal costs.

        Electric utility service is different from other energy because electricity cannot be economically stored in commercially practical quantities.

        Just as the firefighting equipment and firefighters that are available at the moment when a major fire occurs are all that count, it is the power generating and electric transmission capacity that is available at the moment that determines if the electric grid will remain stable or fail.

        We know that there are huge societal costs if there is inadequate firefighting capability when a major fire occurs. Similarly, sudden loss of electric service almost immediately causes significant societal costs (e.g. if someone who is trapped in an elevator due to lack of power has a medical emergency, they may die; traffic lights out of service; increased crime; etc.).

        We are comforted when we see idle fire trucks and ambulances because we know they are available if we personally experienced an emergency. Yet, somehow, people do not seem to understand unused (spare) power generating and electric transmission capacity is required to ensure dependable electric service.

        By analogy, when one says that 3,000 MW of wind power can replace 1,000 MW nuclear power, it is the equivalent of saying a city is just as safe if it replaces each of its diesel driven fire trucks with three wind powered fire trucks.

  • Matt_Thullen

    Rather than researching wind turbine and solar panel efficiencies, we should be pouring money into battery research. Once lightweight batteries that can hold large charges are produced, solar and wind will be a lot more useful as energy sources.

    • TheRadicalModerate

      The cost of storage for mature pumped hydro technology and compressed air aren’t quite to the point where they’re profitable, and battery storage isn’t even close yet.

      This is also a useful reference.

      PS: TAI is back to using stupid colors for links again. There are two of them in here.

      • CB

        “The cost of storage for mature pumped hydro technology and compressed air aren’t quite to the point where they’re profitable”

        You are correct that energy storage is currently fairly expensive.

        As a boutique industry, it would be surprising if it weren’t.

        We can make very serious reductions in emissions without storage, though storage would definitely be necessary to go 100% renewable…

        “Our results show that when using future anticipated costs for wind and solar, carbon dioxide emissions from the US electricity sector can be reduced by up to 80% relative to 1990 levels, without an increase in the levelized cost of electricity. The reductions are possible with current technologies and without electrical storage.”

        I also agree with the author that picking winners and losers with subsidies is a bad idea. It would be far better to strip companies of their subsidies and charge a carbon tax to companies that produce it.

        “Fossil fuel subsidies: Estimated to cost between US$455 billion and US$485 billion.”

        • TheRadicalModerate

          There’s more to technical maturity than just deployment at scale. IIRC, the Lazard white paper takes some stabs at scalability and things are still fairly grim for the dispatchable grid power cases. I suspect that that won’t always be true, but it is right now.

          I don’t have huge problems replacing subsidies with carbon taxes, but my guess is that the subsidies, while less efficient, may minimize the pain. Of course, what’s really attractive about them is that they’re opaque to voters. Carbon taxes–not so much.

          • CB

            “the subsidies, while less efficient, may minimize the pain. Of course, what’s really attractive about them is that they’re opaque to voters.”

            Unfortunately, I think you’ve hit on precisely why they are more common…

  • Jim__L

    If these companies go out of business because the contracts aren’t profitable, then whoever buys their capital goods at fire-sale prices will have a shot at completing the contracts for at least break-even.

    The people who will get taken to the cleaners are the early investors in solar projects, including “green” funds, and those foolish enough to invest in them.

    Which is as it should be.

  • Jacksonian_Libertarian

    The Free Market should be left to develop energy on its own, as it is the only institution competent at making these decisions.

© The American Interest LLC 2005-2016 About Us Masthead Submissions Advertise Customer Service