Electricity markets are weird: why a carbon price isn't enough
If power companies have to pay for their CO2 emissions, what will happen to the price of electricity? The answer isn’t as obvious as you think.
The linkage between costs and prices — increase the first and you raise in the second — is predicated on the notion of efficient markets. Yet it would be hard to find a market less efficient and more distorted than the U.S. electric sector, which is responsible for over 40 percent of America’s CO2 emissions. If we want to use a price on CO2 to drive changes in energy use, we better start with an understanding of how the electric industry actually works.
Electricity markets are not Pareto efficient
Pareto efficiency is one of those things that we all learned in freshman economics and subsequently forgot. A refresher course: markets are said to be Pareto-efficient when no party can be made better off without making another party worse off. When they’re not Pareto efficient, changes in price may not lead to changes in supply/demand. Northwestern economist Lynne Kiesling points out that this issue is particularly important for new markets seeking to monetize environmental externalities. She uses a flower garden as an example and asks how much you’d be willing to pay your neighbor for the vicarious pleasure (aka, market externality) you realize from their flowers:
… it’s a safe generalization that your neighbor-gardeners have more intense preferences over their gardening decisions than you do over their decisions… [this] implies that even if you did pay them as compensation to internalize your benefit, if your marginal benefit is small relative to theirs, your payment is unlikely to change their decision at the margin of how much gardening to do.
In other words, a market that seeks to internalize the positive externality of nearby flowers is unlikely to arrive at a price that leads to more daisies, because Pareto-inefficiencies cause prices to clear at levels too low to affect the gardeners’ behavior. This is a huge problem in electric industry.
In any capital-intensive business, there is a huge barrier to entry established just by existing. An existing generator that sells power for $0.01 more than its marginal costs will earn a profit, but a new power plant won’t get built unless it can secure a contract high enough to recover marginal costs and provide an attractive return on capital. That tends to cause prices to fall to levels that keep competition out, even while preserving (admittedly, depressed) profits for established players.
We’ve all got direct experience with this. Do you want to save money on gasoline but can’t afford a more fuel-efficient car? Do you wish you had a shorter commute but can’t afford to move? In both cases, the high up-front (capital) cost blocks investment in lower long-term (marginal) cost activities. Note the Pareto inefficiency: gasoline prices could conceivably rise to a level that caused you to stop driving your Escalade, but not high enough for you to buy a Prius.
Good CO2 policy should bring about both a reduction in energy consumption and investment in lower-carbon energy supplies. Unfortunately, the capital-intensivity of this sector means that new investment requires a higher CO2 price than reduction in consumption. This raises equity issues, since the poor will be the first to curtail demand in response to higher energy prices. A good policy therefore needs to ensure that new low-CO2 (typically also lower-cost) generation comes forward. Of late, we seem to dedicate much more time trying to ameliorate the regressive outcomes innate to bad policies.
U.S. environmental policy gives economic advantage to dirty power
The U.S. regulates pollution by forcing the next pollution source to come up to modern standards but allowing existing pollution sources to operate in perpetuity at yesterday’s standards. David Roberts gave a nice summary of those problems here. It’s virtually impossible to get a new coal plant permitted today (thank goodness), which has led to a dramatic shift towards gas-fired generation. But why do we then allow old coal plants to sell their dirty product at a price set by new (more expensive) gas-fired facilities? We might as well impose life-sentences on tomorrow’s petty crooks and give their contraband to yesterday’s (free) serial-killer.
So now let’s assume that a CO2 price is applied uniformly to all emitters. Even in that idealized case, utilities will only shift their operations towards lower-CO2 sources to the degree that the economic advantage gained exceeds the economic disadvantage forced on the cleaner sources by environmental policy. Another Pareto inefficiency.
A regulated monopoly isn’t a market
By definition and design, a sector dominated by regulated monopolies isn’t a competitive market. So why do we assume a linkage between costs and prices?
Even in “deregulated” states, the distribution utility isn’t. A PV array on your house interfaces not with a market but with a for-profit monopoly who (with their regulator) dictates all aspects of your economic interaction. If you planted a vegetable garden in your yard and required Cargill’s approval to harvest, eat, transport, set prices for, and/or sell your crop, you wouldn’t call it a functioning market — but that’s life for a rooftop PV investment in a “deregulated” electric market.
And that’s the best case! Now consider the 27 states that remain fully regulated. Electric utilities in those states are to normal businesses what a flying fish is to a blue jay. Normal businesses allocate resources in an effort to balance investors’ desire for dividends with customers’ desire for savings and competitors’ desire to steal market share. In a regulated utility, the regulator is responsible for that balance. To understand the impossibility of that job, pretend you are a regulator and answer the following questions:
- What level/structure of executive compensation is sufficient to attract talent without over-charging consumers?
- If maintenance is deferred to keep costs (and prices) low, should customers or shareholders bear the liability for subsequent problems? What if maintenance is too high? Should rate-payers bear the cost of gold-plated operations? How do you identify the appropriate level?
- If you make a regulatory mistake that compromises grid reliability, who should pay for the clean up costs?
These are hard issues, unique to the regulated enterprise, and they create an inherent disconnect between cause and effect, cost and price. Now imagine we place a cost on CO2 emissions. Should those costs be recovered through higher electricity prices or through a reduction in shareholder dividends? At one extreme, the utility has no incentive to change their behavior. At the other, the customer has no incentive to change theirs. And in all cases, there is no competitive pressure to draw customers (or investors) away. Which do you choose?
Deregulated markets lack cost/price causality
So what about the deregulated markets? Surely the power plants selling into those markets will rationally allocate their capital in response to changing costs, right? Not really.
For starters, those markets are dominated by “spot” rather than long-term contracts. Big capital investments require some degree of certainty about long-term revenue, but you’re hard pressed to get a contract on wholesale power markets for more than 5 years. So unless you have
a very high risk tolerance, wholesale markets are effectively useless when it comes to encouraging investment in new generation.
Indeed, the generation investments that have been made since deregulation have predominantly secured revenue streams external to wholesale markets: regulated utilities have built generation underwritten by the state/ratepayer, renewables have been built on the back of bilateral power purchase agreements with utilities, and some consumers have built on-site generation that could to displace their retail purchase. Unfortunately, while none of those investments have participated in wholesale markets, they still affect them. A utility that enters into a long-term RPS contract for biomass power no longer has to procure that power on the spot market. The wholesale market — where price is set by the balance of supply and demand — is thus short on demand and prices have been depressed accordingly, well below the level required to build new generation.
These are structural problems that CO2 pricing won’t fix. Absent that reform, wholesale markets will remain unable to provide a meaningful price signal to encourage new investment (clean or otherwise). As a result, the economic signal sent by CO2 pricing will be received primarily to the least economically-efficient parts of the electric sector.
So what should we do?
Who pays? Who do they pay? How does CO2 policy interface with other elements of state and federal regulation? Consider all those issues carefully and it is entirely possible to craft CO2 regulation in the electric sector that works in spite of these challenges. Effective CO2 policy ought to include the following:
- Clean Air Act reform
So long as old dirty plants retain an economic advantage against new clean ones, efforts to bring low-CO2 generation face unnecessary regulatory headwinds.
- Output-based emissions standards
Current emissions rules discourage investments in efficiency, inadvertently mandating greater CO2 release. This can be fixed in the law or in the courts, but we cannot sustain a conflict between two elements of our environmental policy.
- Explicit incentives to balance penalties
The carbon price required to shut down dirty generation is not sufficient to bring clean generation on line. A system that provides direct economic links between CO2 sources and sinks without regulatory intermediation will be critical. A stick is not a negative carrot.
- Capital cost incentives
CO2 pricing can only affect the variable cost of operation. Given the capital-intensivity of electricity generation, regulation may need to include some type of additional inducement to deploy capital.
- Technology neutrality
The goal of CO2 regulation is to reduce emissions of CO2, not to deploy technology X or bankrupt company Y. Any policy that confuses paths and goals introduces new Pareto inefficiencies into a market that already has plenty.