Wind power and real-time pricing: Mutual benefits

Michael Giberson

The marginal cost of generation from a wind power generator is essentially zero, which means once the generation is installed you pretty much want to use every bit of wind power generated. A problem, of course, is that wind-based generation is not particularly dispatchable. You don’t tell it when to run, you just try to use as much of it as you can while it is available.

A further wrinkle is that, at least for some wind power locations, the winds are strongest overnight and early morning when the demand for power is lowest. And, when wind power is generated far from the consumers who’d like access to cheap power, it requires adequate transmission capability to move the power to the people. If the transmission system is limited in its ability to move all of the power available, then some of the wind generation capacity will be wasted.

Ramteen Sioshansi and Walter Short observed that there may be mutual benefits available for a power system with a large share of wind generating capacity from also implementing real-time pricing for ultimate consumers. In a paper they describe simulations they ran using data from Texas to examine the potential benefits of real-time pricing for use of wind power resources.

They find useful synergies:

  • Real-time pricing tends to smooth out the normal ups and downs in consumption, because consumers tend to decrease consumption in high cost hours (which are the high demand hours), and to increase consumption in low cost hours. Smoothing out consumption means that the transmission system is less likely to be congested and therefore it is less likely that distant wind generation will be shut in by transmission limits.
  • In addition, because wind power comes at a low marginal cost, whenever it is plentiful it will drive down electric prices. That effect encourages consumers to adapt their consumption to the patterns of wind power availability.

It is true that the effects can be small, but even with some conservative assumptions in the simulations, the authors found that usage of wind power could be increased by more than 80 percent. Don’t get hung up on that number, which very much relies upon the assumptions going into the simulation. Take away the idea that a series of small marginal adjustments in consumption, responding only to price signals, can have significant effects on the use of low marginal cost renewable (or any other form of) power generation.

The authors also note that many political analysts object to the idea of exposing consumers to highly variable real-time power prices. They re-ran the simulations limiting the real-time pricing regime solely to commercial and industrial consumers. Commercial and industrial consumers represented a little over 60 percent of the demand in their historical data, and the new runs of the simulation showed that just about 60 percent of the benefits were retained with the more limited application of real-time prices.

(Like the Hogan talk mentioned yesterday, the Sioshansi and Short paper was presented at “The Economics of Energy Markets” conference at IDEI.)