Lynne Kiesling
Last Friday National Public Radio ran a Planet Money story called “Do smart meters curb energy use?” (first link is to program listing, second is to story transcript) Members of the KP community will not be surprised by any of the content in the report, but it does provide a good introduction to the “information and energy efficiency” literature for the non-specialist.
The story refers to the work of Carnegie-Mellon behavioral economist George Loewenstein, who contends that smart meters could actually increase energy use by revealing how cheap electricity is:
In fact, some of the information that a smart meter would give you might actually worsen your behavior because, for example, electricity is really amazingly cheap. It’s amazingly cheap to air-condition your whole house for a few hours. And if the smart meter is giving you objective information about how much it’s costing you, you might be surprised at how cheap it is rather than surprised at how expensive it is.
I think he’s failing to take into account the probability that electricity prices will increase either (1) if fuel sources become more scarce, (2) if we implement a carbon policy that increases electricity costs, or (3) if we implement renewable energy mandates that increase electricity costs. He also doesn’t take into account the variations in costs of producing electricity over the course of the day, in which case what’s important is not the level of the price, but the variability in the price and how that relates to the variability in the cost of providing the service. Still, it’s a pretty good story.
One other element that is missing from the story, and from Loewenstein’s framing of the question, is the transactive capabilities of intelligent end-use devices that the smart meter enables. Loewenstein doesn’t offer an example of “devices doing the work for you” that is any more sophisticated than direct load control, in which the consumer signs a contract that allows the retailer or the distribution company to cycle off the air conditioner as system conditions warrant. The combination of intelligent end-use devices and dynamic pricing, enabled by a smart meter, creates the potential to be so much more innovative, clever, and effective in optimizing individual energy use, and to do so in a much more user-friendly and less intrusive manner. That’s where the Planet Money team should be focusing their attention, not on direct load control.
We need to move beyond direct load control, which is a top-down sledgehammer in the energy efficiency toolkit. Bottom-up technologies that integrate more directly and subtly into individual lifestyles, such as Direct Energy’s vision that I discussed the other day, are likely to be more effective, more robust, and more sustainable.
This is an important point, and one where I would tend agree with Lowenstein. There are a number of other factors involved where by knowing the price of electricity, consumption may in fact go up; KP’s other contributor had a series of posts last year about negative wholesale prices in Texas. Furthermore, if a state has certain policies that are designed to create a stable retail price, say by requiring IOUs to procure upwards of 90% of expected load to be contracted for a month in advance, tying price signals (and rates) to the wholesale market that is only dealing with the margins would show a potentially very low price (see CAISO since MRTU). Residential customers in Illinois last year that took service under their RTP had prices 30% lower than the retail rate, you don’t think they might have increased their consumption, even while the bill went down? To me the larger, and more important point is to save customers money. For regulators, how far are they willing to interfere with the price if customer behavior responds accordingly to prices, but not in accordance with other policy goals?
Well said, Lynne. Seems to me that Lowenstein’s arguments are rather nonsensical outside of an economics textbook. For the vast majority of Americans, even in the highest electric rate regions, the incremental cost imposed by an incremental decision to use a bit more or less energy is immaterial. In theory, everyone responds to price. In practice, very few go out of their way to save a buck an hour (about what you’d save if you were paying $0.20/kWh and upgraded your whole office lighting system to knock out 100 lightbulbs worth of electric load at 60 W/bulb. Needless to say, that argument is all the stronger if you’re only paying $0.10/kWh.
The argument for smart metering doesn’t seem to be what incremental decisions it will cause people to make, but rather (a) what changes it will cause them to make in their capital stock and (b) the degree to which the new appliances can respond in real time. E.g., the point isn’t to make YOU decide to run your dryer at night, but rather to make you decide to buy a dryer that automatically elects to run when the price signal is low. I’d argue that this is even true for big industrial users – operating modalities for existing energy equipment are slow to change, even when energy prices are very volatile; but once they upgrade their equipment, those changes are rapid.
Particularly if we implement a carbon policy, then if consumers are consuming less than they would if they knew how much it cost, an increase in consumption is welfare-improving. NPR listeners might not want to hear that, though.
I think it takes a Pollyanna-ish optimism to think that smart meter capabilities will be implement to enable consumers to take control. It is much more likely that they will be implement to allow the state to micromanage its subjects lives. It will be a great excuse for not investing in new generating capacity, so that resources can be used for bloated public payrolls.
Pingback: Smart Grid Blog » Blog Archive » Planet Money: Do smart meters curb energy use? « Knowledge Problem