A Free-Market Energy Blog

Radioactive Corporate Welfare

By Jerry Taylor -- February 18, 2010

A good default proposition regarding the government’s role in the economy would state that the government should not loan money to an enterprise if the enterprise in question cannot find one single market actor anywhere in the universe to loan said enterprise a single red cent.  It might suggest – I don’t know – that the investment is rather … dubious.

Alas, like all good propositions regarding the government’s role in the economy, this one is being left by the roadside by the Obama administration.  Unfortunately, the only complaint being made by a not insubstantial segment of the political Right – frequently, the political crowd that is busy decrying “Bailout Nation” – is that the loan guarantees are not fat enough.

I write, of course, about the $8.3 billion federal loan guarantee announced by President Obama this week for Southern Company to build two new nuclear power plants.  The money will be used to guarantee the loans being made by the federal government (via the Federal Financing Bank) to partially cover the cost of Southern’s projected $14 billion nuclear construction project at their Vogtle plant near Waynesboro, Georgia

The loan guarantees were authorized by Congress in the 2005 Energy Policy Act and, we are told, are the first installment on a total package of $54 billion that the President would like to hand out to facilitate the construction of 7-10 new nuclear power plants (Congress, however, has only authorized $18.5 billion to this point). 

The claim being made by some – that the loan guarantees are necessary to jump-start investor interest in new nuclear power plant construction – is not quite correct.  Even these lavish loan guarantees aren’t enough to do that.  In a letter to the U.S. Department of Energy dated July 2, 2007, six of Wall Street’s s then-largest investment banks – Citigroup, Credit Suisse, Goldman Sachs, Lehman Brothers, Merrill Lynch, and Morgan Stanley – informed the administration that, contrary to the government’s expectations, anything short of a 100 percent unconditional guarantee would be insufficient to induce private lending. 

Why is it risky to build nuclear power plants?  Because new nuclear projects tie up more capital for longer periods of time than its main competitor, natural-gas fired generation.  Nuclear power makes economic sense only if natural gas prices are very high.  Then, over time, the high initial costs of nuclear power would be offset by nuclear power’s lower fuel costs.  Moreover, as noted by Moody’s in an analysis published in July of last year, there is uncertainty associated with construction costs, regulatory oversight, technological developments that might reduce the cost of rival technologies, and the ability of utilities to recover costs and make a profit over the lifetime of the plant – a risk tied up in the economic prospects of the region being served by the plant.  And those risks have been increasing, not decreasing, as time has gone on. 

In short, even during the go-go days prior to the September 2008 crash – a time when Wall Street was allegedly throwing around money left and right to all sorts of dubious borrowers – the banks that stand accused of recklessly endangering their shareholders on other fronts were telling utility companies that they would not loan them anything for new nuclear power plant construction unless the feds unconditionally guaranteed every last penny of those loans.  That’s how risky market actors think it is to build nuclear power plants.

And it’s not as if the federal government disagrees completely.  The Congressional Budget Office pegs the chance of default (program-wide) at 50 percent or better and the Government Accountability Office likewise thinks that default risks are quite high.  Energy Secretary Stephen Chu says that he thinks the chance of default is much lower.  We can only speculate about who’s right given that no one has tried to build a nuclear power plant in the United States for over 30 years.    

Regardless of what the risk actually is, the loan guarantees do not reduce that risk.  They simply transfer the risk from the bank to taxpayer.  In this particular case, however, the loan guarantee transfers risk from one arm of the state to the other, so it doesn’t really count.  But if such loan guarantees were ever to induce actual private lending for plant construction, that’s how it would work.

Plenty of arguments have been offered to justify these loan guarantees.  Most of them are flimsy on their face.

For instance, we’re often told that we “need” new power plants.  But with electricity demand declining over the past couple of years, it is unclear when that need might arise. 

Regardless, when the market “needs” more electricity, that need will be manifested in price signals that will carry with them profit opportunities.  Profit-hungry investors will be willing and able to meet that need without the help of government.  Of course, if market conditions don’t radically change, those needs will be met with gas-fired power plants, but hey, if that bothers you, take it up with someone else.

Others argue that we need the jobs that will be produced by new nuclear power plants.  Well, building big new reactors will certainly employ a lot of (largely unionized) construction workers.  But that’s one reason why building a nuclear power plant is not very economic compared to building gas-fired generators.  If creating jobs is the idea whether the project makes any economic sense or not, then let’s just ban food imports and farm equipment and put everyone to work with hand plows and scythes.     

Two somewhat more serious arguments have been offered to justify these loan guarantees.  Neither of them stands up to much scrutiny either. 

The first argument – the argument most often heard from the nuclear power industry and some segments of the political Left – is that we need nuclear power to reduce greenhouse gas emissions.  Of course, the best (that is, most efficient) way to reduce greenhouse gas emissions is to internalize the cost of greenhouse gas emissions in the retail price of electricity and then allow market actors to adjust their production and consumption decisions accordingly.  That price internalization exercise, however (whether directly through a carbon tax or indirectly through a cap-and-trade program), does not appear to be in the cards in the foreseeable future.  Hence, the loan guarantees are advanced as a “second-best” solution, one that will get us the technology and economic efficiency that would be delivered by a properly crafted carbon tax or cap-and-trade program without the retail price increases associated with either.

One of several problems with this argument is that it would take one hell of a carbon tax – or one hell of an onerous cap-and-trade program – to get anyone interested in building nuclear power plants.  If natural gas prices remained roughly where they are at present (that is, if they were to remain at historical norms) then it would take a $90 per ton carbon tax or a cap-and-trade program that delivered carbon emission credits at $90 per ton on the open market to induce investment in nuclear power plants.  Few economists who study climate policy believe that a carbon tax of that size is defensible given what we know about climate change. 

And that’s if construction costs are as low as advertised.  Were they to double (as they did from 2003 to 2009) – either because of endogenous increases in the cost of capital, labor, or construction-related resources or because of cost overruns – then it would take at least a $150 per ton carbon tax (or a cap-and-trade program that delivered $150 carbon credits to the market) to induce investment.     

You might ask yourself what the historical relationship is between final nuclear power plant construction costs in the United States and construction costs as projected at the onset of the project.  Happily, the CBO has done your homework for you.  They found that final construction costs averaged 207 percent of projected costs.  Hence, a doubling of construction costs is probably more likely than not once a project is underway … if past is prologue. 

The upshot is that there are more efficient ways to respond to greenhouse gas emission constraints than to go on a nuclear power bender.  This observation is heresy on the Right, but almost every credible analysis of the matter backs up that observation. 

The second argument one hears to justify federal loan guarantees is that they are necessary to counter-balance the excessive regulatory costs associated with new plant construction.  Now, put aside the fact that the Nuclear Energy Institute – the trade association of the nuclear power industry – has often expressed near-total satisfaction with the current federal regulatory regime.  If the regulatory regime is truly “bad” and, accordingly, is imposing steep and unnecessary costs on the industry, then the correct remedy is to improve said regulatory regime, not to subsidize the industry.  

The counter-complaint that positive regulatory reforms are impossible is difficult to swallow.  After all, if there is sufficient political will to bestow tens of billions of dollars worth of tax money on this industry, then surely there is enough political will to reform the bad and unnecessarily costly regulations allegedly bedeviling the object of those very same legislative affections.

I will confess to being skeptical about the argument that high construction costs are largely the fault of regulators.  Building a light water breeder reactor is a technologically challenging and costly undertaking whether regulators are on the scene or not.  Moreover, it is not obvious to me that the regulations that are in place are a priori objectionable from a libertarian perspective

One rarely if ever hears of particulars in this bill of complaint offered about nuclear regulation.  But if a persuasive bill of complaints is ever presented, then the appropriate response is regulatory reform and then to leave the decision to build or not to build to markets.      

In the course of announcing these loan guarantees, President Obama said this week that “The fact is, changing the ways we produce and use energy requires us to think anew. It requires us to act anew.”  Well, there’s nothing new about throwing subsidies at nuclear power.  Economist Douglas Koplow calculates that federal nuclear subsidies have totaled $178 billion from 1947-1999.  The promise of a nuclear economy with rates too cheap to meter has been made for over half a century.  What would be new would be a policy of “just saying no” to industries with their hands out in Washington.


  1. Jon Boone  

    Curious minds want to know what form of energy production is not subsidized by government. The issue for me is not so much the fact of the subsidy but the benefit that will likely accrue from it. In this case, what society will reap, for dozens of years to come, is rather limitless capacity value with an incredibly high capacity factor approaching 95%.

    What people should object to vigorously is government investment in energy scams that provide no capacity and obligate everyone and everything around it to work much harder just to stand still.


  2. jtaylor  

    Curious minds should click on the link I provided (http://www.eia.doe.gov/oiaf/servicerpt/subsidy2/pdf/subsidy08.pdf) to a study recently published on that matter by the Energy Information Administration.

    Jon, we’ve subsidized the nuclear energy industry for years. Where’s my too-cheap-to-meter electricity? The most recent MIT study on power costs that I linked to in my piece shows that this whiz-bang technology is still the most expensive source of conventional electricity on the grid.

    The fact that others industries get subsidy is no reason to refill the trough so that every one else can have another taxpayer-funded drink. And that’s particulary the case given that nukes have received well more in the subsidy department than most other technologies.

    Not that it matters. Especially to some conservatives, who’s commitment to free markets is clearly situational.


  3. Radioactive Corporate Welfare | Think Tank West  

    […] [Cross-posted at MasterResource] […]


  4. Tom Tanton  

    Jerry, while nukes have been the recipient of largess, they’ve also been fairly productive. That is not to justify subsidization, just an observation. On the other hand, other technologies, like wind, have received largess yet produced nothing to speak of. Their $subsidy per useful output is 14-15 times, or more, that of any other technology (within the energy realm.) While I agree that eliminating eliminating all subsidies is the better alternative, one step in the right direction would be to reduce the more-wasteful subsidies where nothing good is produced. As to the often stated “infant industry” argument for wind, how “infant” is an industry that have recieved such preferential treatment for four decades?


  5. Jon Boone  

    Thanks for the EIA listing ad nauseam showing some of the cavernous subsidies available in the sweet land of liberty. There may be less stars in the galaxy than national government “incentive” programs. If memory serves, they got rolling under the first Clinton with the Erie Canal, and caught fire with the railroads, the stockyards, steel–and we’ve never looked back.

    Tom Tanton captures what I was really getting at. Within the electricity sector, let’s at least agree to proscribe subsidies for enterprise that clearly cannot accomplish the goals that justify the subsidy, ones wrought with a hope and a prayer and a contribution to the politics of dumb.


  6. Jerry Taylor  

    What’s all this about nuclear being “productive”? The only metric that matters is whether nuclear power can turn a profit absent government assistance. It cannot. It is far more expensive to produce a kilowatt of electricity with a nuclear reactor than with a coal-fired or gas-fired generator. By far. The rest is secondary. Hence, the subsidies produce nothing of value on balance. The program does net harm to the economy. Full stop.

    If nuclear power were economically attractive, then no subsidy would be necessary. If it is not, then no subsidy will make it so.


  7. Soylent  

    Jerry; nuclear is “unproductive” because coal gets to dump its waste straight into the atmosphere. Coal particulates kill ~15 000-45 000 americans per year(range for 95% CL, that corresponds to somewhere between 7 and 50 Chernobyls per year in case you where wondering).

    Nuclear is “unproductive” because a government beaurocrat can destroy billions in investment at the stroke of a pen.

    Nuclear is “unproductive” because the NRC adds 6 years and a couple of hundred million dollars to the 4 year construction time. It’s a lot harder to predict market conditions 10 years out than it is 4 years out; there’s a huge risk associated with this.

    Nuclear has the lowest operational cost, bar none. It’s not really any more technologically complex to build a nuclear reactor than a coal plant.


  8. Jerry Taylor  

    “Nuclear has the lowest operational cost, bar none. ” That is probably not correct. I suspect that hydropower has lower operational costs, particularly were we to strip away the uranium enrichment subsidies.

    “It’s not really any more technologically complex to build a nuclear reactor than a coal plant.” The history of those technologies certainly suggests that you couldn’t be more wrong. Anyway, what matters for our purposes is construction cost where the differences between coal and nuke are quite large. See the recent MIT study I linked to in my post for a thorough analysis.

    I find it telling that whenever one attacks subsidies to nuclear power in front of conservatives, they tie themselves in knots defending those subsidies. It is impossible to find any who will say, “Yes, you’re right, they should not get government help … period.” Believers in free markets cannot just oppose helping hands for techs they don’t like while finding a million excuses for giving handouts for industries they are fond of without appearing intellectually dishonest.

    The only difference between liberals and conservatives when it comes to energy policy seems to be (i) the rhetoric (cover stories) being employed to justify intervention, and (ii) manner in which the market is to be rigged, not whether the market is to be rigged in the first place.



  9. The Mistake of Nuclear Subsidies « Intercon  

    […] came across a fantastic article by Jerry Taylor on energy blog, MasterResource. As Taylor notes, the administration’s […]


  10. Donald Hertzmark  

    Jerry, Jon, Tom,
    We may be getting way ahead of ourselves on nuclear. After all, none of the loan provisions become effective until the plant is licensed by the Nuclear Regulatory Commission. If, as the cynic in me believes, this offer of a loan guarantee is just a ploy to gain some “R” votes on Cap and Tax and wind subsidies, then it will end up worse than useless by encouraging even more unproductive energy subsidies.

    On the other hand, if the Republicans do not cooperate with the administration, then the NRC can deny the construction or operating permits, the USG can withdraw the loan guarantee and we shall all revert to the status quo ante.


  11. Tom Tanton  

    Jerry, by “productive” I was using the physical and engineering terminology, not so much the economics one. You’re right wrt to tuning a profit–but that is an issue of the finer distinction between economics and finance. In any event, all I meant was the unit subsidy ($/kwh) is WAY fatter for some technologies than for nuclear, because nuclear has actually produced something. I agree that subsidies in general (well in total) ought to be eliminated, but suggest one strong step in that direction is to first get everybody on the same (and lowest) subsidy per unit production. It is simply a function of “one miracle at a time” not a distinction of the end game. Similarly, technology favoring regulations (prescriptive) should be eliminated and replaced with performance based regulations.


  12. Gary Moran  

    “The upshot is that there are more efficient ways to respond to greenhouse gas emission constraints than to go on a nuclear power bender. ”

    Like what?

    As far as I’m aware there is only one technology that currently can reduce CO2 emissions by any significant degree and that is nuclear. Replacing coal with gas will reduce CO2, but not at the scale we are told we need to start contemplating. Wind and solar, don’t get me started.


  13. Industry Observer  

    Funny how CATO’s outrage meter only seems to spike when government involvement in the nuclear sector comes up.

    Could it be that Taylor is really an anti-nuclear advocate hiding behind the same old free-market frippery?

    There are a lot of government subsidies in the non-nuclear energy sector that should really get him upset.

    Market failure (i.e., the difficulty in getting a new nuclear power plant built today, for multiple reasons) seems a perfectly good basis for the loan guarantees, a very limited approach to government support.

    Unlike wind generators that may shut down at the end of an accelerated depreciation schedule or when PTC runs out, nuclear power plants are operating full out and will keep doing so for decades longer. The owners of these nuclear plants are making large profits and would really like to find a way to get more nuclear power, hence the almost universal move to expand the output (uprate) the existing units and to extend their operating lives (license renewal).

    The EPAct of 2005 benefits, including the loan guarantees, were and are, meant to get the nuclear industry back into operation by providing benefits to the first cohort of new power plants.

    By about 2025, the NRC licensing process will be sorted, a few standard nuclear plant designs will be seen as global market winners and a few new nuclear power plants will be completed and operational in the US (and a lot more in China, the UAE and elsewhere). We may also get enough courage by then to force fossil power plants to internalize the high cost of their emissions, just as nuclear power plants now do.

    When all this happens, nuclear power plants will not need any government assistance.


  14. JamesG  

    The nuclear lobbies are out in force here. Well a good test case for nuclear power is in transport by ship. After all these small pwr’s are turnkey and relatively simple to manufacture. We also have years of experience with them. Yet every attempt to have a nuclear transport ship has failed due to ongoing costs – not construction costs. When i see a commercial nuclear ship I’ll believe it’s a viable technology.

    There is however a huge amount of lobbying money swimming about from the nuclear lobby, desperately trying to greenwash itself. Face it folks, the public don’t want it, the banks don’t like it, nobody can make it viable without government support and the fabled new generation of reactor only exist in concept.

    If it’s for energy security then let’s go ahead but then don’t pretend to be free market heroes. Wind and solar energy, unlike nuclear energy DO attract private investments. Go figure! Incidentally the Candu is a better bet because it doesn’t need enriched uranium and doesn’t need to go offline at any time, but then those jobs would go to Canada.


  15. Tom Tanton  

    JamesG–wind and solar attract private investment SOLELY because of the tax subsidies, not any thing else. Where else can you get a government gurenteed +50% ROR? To use your analogy of transport ships, I see no commercial shipping done by windjammers any more.


  16. Jerry Taylor  

    Gary, if a $150 per ton carbon tax is necessary to induce investment in nuclear power, then a carbon tax of >$150 per ton will induce some other response. Energy conservation is one response. Substituting gas-fired power for coal-fired power is another. There’s a large literature on this subject. Regardless, market actors should determine how best to allocate capital in a carbon constrained world, not politicians or bureaucrats.

    Industry observer, if you go to the Cato website, you’ll find plenty of articles from me decrying subsidies to the oil and gas sector, to the renewable energy industry, etc. You’ll also find plenty of explicit calls to eliminate everyone’s governmental preferences and to allow markets to let the best tech win.

    Regardless, the argument that subsidies to x justify subsidies to y is illogical. If subsidies to x are unwise, then they should be eliminated. Throwing around moreeconomically inefficient subsidies simply compounds the economic damage.


  17. The Mistake of Nuclear Subsidies | Intercon  

    […] came across a fantastic article by Jerry Taylor on energy blog, MasterResource. As Taylor notes, the administration’s […]


Leave a Reply