While I hate to disagree with my esteemed colleague’s assessment of David Cay Johnston’s electricity restructuring article from the New York Times, I do disagree. Yes, it’s better than the first one, but the first one was a confused muddle of too many angles on a very complicated set of issues. This one is more coherent, true, but in most of the article he misses the point.
For one thing, he completely and utterly falls for the political canard that the uniform-price auction market design is to blame for high prices. This claim is simply incorrect. He is faulting the uniform-price structure for other design features that do not provide sufficient discipline of supplier incentives to raise prices. The alternative, a pay-as-bid auction, generally performs worse, and it shouldn’t take too much intuition to see why. If you are a seller and you know that you are only going to get paid exactly what you offer, are you going to offer exactly your marginal cost? Of course not! So you have an incentive to shade your offers upward, and so does every other supplier selling into that market.
What is the single most effective design change that policymakers could make that would discipline this ability for sellers to raise prices? Active demand and a double-sided market. Buyers submit bids and sellers submit offers simultaneously, and the interaction of those bids and offers determines the (uniform) market-clearing price that all buyers pay and all sellers receive. That’s the right answer, but that’s not what Johnston discussees in his article. I hope he does discuss active demand, double-sided markets, and the role of enabling digital technology (and the removal of regulatory barriers to its implementation) in his next article in the series.
Furthermore, he misrepresents the experimental research performed by my ICES colleagues at George Mason. Yes, the experiments show that people learn how to manipulate the rules. Welcome to human nature. What that means is that it is incumbent upon the policymaker, upon the regulator, to create processes that lead to robust, resilient market designs. That means lots of testing, and experimental testing with real humans is the best test. But what Johnston does not report here is that in the treatments in which active buyers are bidding against those sellers, they are unable to raise prices. Again, institutions matter, market design matters, the rules matter.
Yes, the existing market institutions in electric power are generally inferior. But the uniform-price auction structure is not the culprit. It’s the lack of active, engaged demand, because of the lack of integrated wholesale and retail markets. I hope Mr. Johnston points that out in his next article.