Wednesday, December 12, 2007

Energy Optimization

To contrast with Mike Huckabee's idiocy on energy policy, a good idea:

Pepco is about to start sending personal e-mail messages to Jonathan and Lauren Schwabish every few hours that could determine when they do the dishes, wash the baby's clothes or turn on the air conditioner.

The couple will learn when the price of electricity for their old Capitol Hill home will spike the next day because Washington's winter chill or its steamy summer is nudging up the demand for power.

If they wait to turn on the washing machine or they turn off the air conditioner when the sun beats down, they'll be rewarded with a credit on their utility bill that could reach hundreds of dollars a year. Other D.C. residents have agreed to pay rates eight times the average if they use their appliances at peak times but rates well below it at off-peak hours, as part of a pilot program starting next month.

"Lexus lanes" are coming to the electricity grid. Energy conservation programs that died when the power market switched from regulation to competition are back, but with new technology and aggressive demands from government regulators facing anger over rising prices.

Just as long-awaited high-occupancy toll lanes will charge drivers a fee to travel at rush hours, electricity customers will pay more when the grid is congested and less when it's not. If the strategies succeed, customers will not only slash their bills but also reduce pollution from coal-fired generating plants.

And within a few years, energy experts predict that Washingtonians will live like the cartoon Jetsons, their homes powered by computer chips that shut down washing machines and dishwashers when electricity prices soar.


I've thought this was a good idea since 2004 when my final project at Waterloo was the exact device mentioned in that last paragraph. Right now, this type of idea is most plausible in deregulated markets like California and Ontario where electricity price is established through a wholesale market on an hourly basis. If customers decrease the system demand load at peak times, then not only do those customers save money, but all other customers do as well. This is because the least expensive generation units are used for baseline loads, and more expensive ones for peak loads. If demand is flattened over the course of a day, then there is less need for those more expensive units to respond to peaking demand. In many jurisdictions, those units also tend to be the higher polluting units, like coal burning plants. So its a win-win situation.

There are a few issues that need ironing out in these types of plans. First, the retail energy markets needs to become "efficient". That is, sufficient information must be relayed to consumers to be able to incorporate the real costs of energy into their decision making. This plan is using emails a day in advance. Ideally, an automated decision making mechanism would be better. Software should be designed that can take into account energy prices and demand forecasts for the next day, and optimize a house's energy consumption so that energy is used at low-peak, low-price times.

In the absence of an automated system, it remains to be seen how elastic energy consumption is. How much do consumers react to price signals in the energy market? To what extent do consumers actually change energy consumption behavior in order to incorporate pricing information?

I had a lecture in my public utilities regulation class by the director of customer care for BC Hydro, who was conducting trials with volunteers in Fort St. John in order to answer that very question. In BC, energy prices do not change hourly, but are set by the regulator at a rate sufficient to recoup the operating costs of providing service to customers. They are looking at a more simplistic version of the plan in that article whereby the first X amount of electricity used in a month is charged at a given rate, and then any electricity used beyond that amount is charged at a higher rate. They found consumers to be quite elastic. I think the average conservation by the participants was in the neighbourhood of 12%. Although, these were volunteers who tended to be more motivated to conserve than the average joe. This is called conservation driven rate-design, and it's something I think we should be doing much more of. Canada is one of the most inefficient consumers of energy in the world. All that is needed to help drive conservation is the economic insentive to do so.

The article raises a valid point about the burden this may place on fixed income consumers, like the elderly. This is the same question I put to the woman from BC Hydro and my prof. And I tend to agree with the response. Utilities regulators are fundamentally economic regulators. They have no expertise in social policy, and so should not be involved in implementing social policy. Energy consumption through rate design and protecting vulnerable consumers from high energy prices don't have to be mutually exclusive. But it would be simpler and more effective to just have something like an energy tax credit for low income and fixed income customers.

No comments: