Over at the Atlantic blog, Megan McArdle looks at who is getting loan guarantees from the Department on Energy. As you know, the rationale behind the government acting as a venture capitalist in the green business is that if it didn’t, no one would, since these investments are just too risky. She explains:
There’s a lot of talk about the manufacturing “Valley of Death“, where startup manufacturing firms may have difficulty getting capital to commercialize their prototypes. According to proponents of this theory, there’s plenty of money for early stage ventures, and plenty of bank loans for established firms, but no money for mass commercialization of new manufacturing ideas. (Hence the “valley”). This valley, they say, is especially wide for energy firms, because the capital costs for starting up are so high.
While she doesn’t buy the argument, she took a look at who is getting these loans to figure out “how well is the government doing in its role of VC/valley of death sherpa?” This is what she found:
I have highlighted what jumped out at me: most of the money has gone to enormous companies that should have no trouble accessing capital. Established utilities, large multinational auto manufacturers, a global warehouse owner. The bulk of these funds are not going to rectify some gap in the capital markets. They’re straight subsidies to huge corporations. Even some of the smaller firms/deals are owned by large corporations like Total SA.
Giving large, established companies extra-cheap loans to build power plants, run transmission lines, and fix up the roofs of their warehouses is, in the immortal words of P.J. O’Rourke, like paying a Dairy Queen owner to keep his ice cream freezers on.
This has implications for the default rates. The genuine startups seem to be shaky–it’s not just Solyndra, but also Nevada Geothermal and Brightsource. In other words, the firms that actually need the money are likely to experience a far higher default rate than the overall portfolio.Why does that matter? Because it skews our perceptions of the usefulness of the program. If we loan a bunch of money to firms that could easily get the money elsewhere, and a little bit of money to firms that are very risky, we can claim a high “success” rate even if all the risky firms fail. But we won’t have actually added much value, because the government wasn’t addressing a genuine market failure. It was just giving Ford and Nissan some extra-cheap money.
But if we’re not really filling a gap in the capital market, this is a terrible way to go about subsidizing clean energy.
The whole thing is here, with data and charts. McArdle does a great job at breaking the information into a format that helps you understand what’s going on, which in this case is that much of the money is going to larger, established companies that are perfectly capable of getting loans on their own, not risky projects that can’t get funding otherwise.
Of course, I don’t believe that the government should be in the VC business at all. I don’t believe it can ever do so effectively, no matter how deceptively simple and appealing the idea is, and no matter how good government intentions are. I am amazed that so many years after Friedrich Hayek published his famous insight about the use-of-knowledge problem, which effectively discredited the notion that governments and central planners can successfully identify which industries or companies are worth investing in (or can do it better than the private sector or decentralized planner), such loan-guarantees programs and industrial policies are still in place.
Obviously, I understand public choice theory and I know why it is the case. But call me an idealist, it still doesn’t make it right.
For more on the problems with the DOE loan program, go here.