Did electricity companies really overcharge by $4.3b? Part 2

In a number of places, the Wolak report assumes that the demand for electricity in New Zealand is not sensitive to the spot price. This is certainly reasonable as a short-run assumption. In New Zealand, wholesale prices are set every 30 minutes. If something unexpected happens to supply (say a failure in a piece of the transmission network) causing prices to spike up, then there is no capacity for buyers to respond to that price spike within the half-hour to reduce the pressure on prices.

But over the course of a dry winter, when prices remain high for long periods, demand can and does adjust. This happens in various ways. First, a few large buyers who buy directly on the spot market reduce their demand to lower their exposure to high prices; second, other large institutional buyers who purchase in advance on forward contracts find it profitable to reduce their electricity use and sell those contracts back into the market at the spot price; finally, retail companies, who are buying at the spot price and selling to customers at a pre-set fixed price, offer financial incentives to their customers for reducing usage below normal levels.

What this means is that, even if, contrary to what I argued in my previous post, there were genuine excess capacity in the system in the dry years due to generation companies exercising market power, prices could not have been reduced right down to normal-year levels without generating excess demand and market failure.

The implicit assumption in the Wolak report is that there is so much excess capacity in our system that, even in dry years, we could meet all the country’s electricity demand without a significant increase in price, and without the need for public “conserve power” campaigns. Furthermore, it assumes that at price equal to the marginal cost of the most expensive thermal unit, generators would be able to earn a sufficient return to cover the investment costs of creating capacity that would only be used for a few months 1-2 times a decade. I simply don’t find this plausible.