Lynne Kiesling
Tim Haab helpfully points out an article from Time about EnergyHub, a device for consumers to see more, and more timely, information about their energy consumption. I’ve written about EnergyHub here before, and honestly, they have not been among the most forward-looking or impressive of the products I’ve seen for providing consumers with both the information about their energy consumption and the ability to take action and automate changes in energy consumption behavior. Perhaps EnergyHub’s product has advanced since I last looked at them, but I do think Time christening EnergyHub’s smart thermostat one of the best inventions of 2009 is a bit of hyperbole, as well as being incorrect, since companies like Tendril have been working on more informative and communicative “smart thermostat” technology since 2007.
Still, having a publication like Time draw attention to the ability of homeowners to see better energy information and to respond to dynamic price signals autonomously is a positive step toward a more competitive and efficient retail electricity industry.
However, I have one nit to pick with Tim’s post. He states
Typical consumers get a bill at the end of the month reporting total consumption and the total bill. But efficient energy pricing requires the consumer to know the marginal cost of the next unit consumed–how much will it cost me to toast this frozen waffle?
Yes … and no. Think about the other products you consume — do you necessarily pay precisely the marginal cost for every single unit of every single product you consume? No. Electricity is no different from, say, your cell phone — do you know the marginal cost to Verizon of the next minute of mobile communication you consume? No, you don’t, and you don’t pay a price that directly reflects that cost. Why not? Because pricing also reflects preferences, not just costs, and there are differentiated products/service contracts.
However, what all other products you consume have that electricity does not is choice and product differentiation. In a competitive retail market, retailers would offer time-differentiated and quality-differentiated products, or bundled services, but these products and services do not necessarily all have to include real-time retail pricing to lead to efficient retail markets. Here’s an example: suppose I am risk-averse, so I do not want real-time prices, but I am willing to pay a peak-off peak time of use price. Suppose I contract with a retailer for a TOU contract. That retailer is essentially engaging in risk management, so the retailer will either buy from the wholesale market on long-term contracts or from the wholesale spot market (probably some combination of the two). In this case I do not observe the marginal cost of the last unit I consume, but I am choosing between peak and off peak. My retailer sees its marginal cost of the last unit I consume, though, in its engagement with the wholesale market. But the contract that I chose voluntarily reflects my willingness to bear price risk.
Put more simply: an efficient retail electricity market does not require that the retail price for all consumers precisely reflects the marginal cost to the supplier of the last unit consumed, but it does require that consumers have choices that enable them to express their diverse preferences over price risk, generation source, etc. It is also enabled by technology that allows consumers to see how much it’s going to cost them to toast that frozen waffle, and to make more sense of what a price per kilowatt-hour means in terms of actual consumption.
Tim is correct that electricity regulation has led to a world in which consumers pay a fixed, averaged price and only know how much they have consumed when they receive their bills at the end of the month. Given how much communication technology is prevalent in our lives, and how inexpensive it has become, that lack of information is appalling.
Lynne,
You make a good point. I keep reading smart grid articles that make it sound like the shiny new real-time communication infrastructure, along with smart thermostats, and new refrigerators, all must be in place in order to make TOU work. These reports leave people thinking that TOU will be part of some far away sci-fi future world.
IMO, static time-of-use pricing (maybe just 2 or 3 tiers as you suggest) could capture a good chunk of the demand elasticity that’s out there, and it would be an easy first step.
Jim
I think we’re conflating three market failures here.
1) Customer awareness of energy usage
2) Marginal pricing
3) lack of customer choice between products.
Tim’s advocacy of real time metering addresses #1, and lets people know how much power any activity is drawing. This can lead to energy savings because appliances that are drawing too much electricity due to lack of maintenance are likely to be maintained sooner and electronics with power hungry standby modes are turned off with power strips.
TOU Pricing addresses #2, and requires real time metering as a prerequisite.
#3 seems to be the improved efficiency that Lynne is addressing.
Any of these would increase the efficiency of the electric market, although I doubt even all of them would make the market truly efficient, because there are other barriers against true efficiency such as lack of consistent and informative labeling for electronics and appliances.
It’s interesting how some associate the marginal price with only the marginal consumption. If the power price is high enough that I need to be concerned about using the toaster (What, a kW-minute? Please, I just want my breakfast.) then perhaps I’d better start thinking about what else in the house is already consuming, like maybe the AC or strip heat. If the TOU rate isn’t tiered, then all consumption in the house is paying the same marginal rate.
Even at VOLL, a piece of toasted bread is pocket change.
It matters a lot more what refrigerator and TV you choose every 10 years than what you do in the next 10 minutes. I wonder if these things will have the impact to influence major purchases, or if by then they will have become background noise.