Why electric utilities like coupling
Writing from the Eco:nomics conference last week, David noted that at least one utility CEO is pretty down on decoupling:
Michael Morris, CEO of American Electric Power … said “I’m not a decoupler. If my revenues go down, they go down.”
David appropriately questioned whether AEP is really so agnostic with respect to falling revenues. But Morris does raise a larger, quite accurate point. Namely, many electric utilities aren’t decouplers. Given the prominence that decoupling has come to play in many state and federal policies, it’s worth taking the time to understand why.
Decoupling is often framed as a way to get rid of the utility disincentive created by energy efficiency. With large fixed costs, small reductions in revenue can have big impacts on equity returns. This has historically made many utilities work really hard to incentivize inefficient use of their services, from special all-electric rates to exit fees, declining-block pricing schedules, and standby tariffs. (Don’t worry about the jargon — the unifying feature of all of the above is that they penalize any customer who has the temerity to invest in energy efficiency.) It has also made the regulated electricity industry the biggest opponent of sensible energy use.
Eliminate the “coupling” of revenues and equity returns — so the theory goes — and you eliminate utility hostility to efficiency.
The problem with that theory is most utilities like their coupling for the exact same reason that they don’t like energy efficiency: namely, it provides them with a way to dramatically increase their equity returns through load growth.
As creatures of their regulation, utility returns are theoretically set only to provide them with a fair return on capital. I emphasize the word theoretically because as a practical matter, it is not possible for rates to be re-calculated every year. So utility commissions will periodically hold a rate case (typically every 5-10 years) and calculate the rates based on (a) utilities’ approved capital expenditures and (b) their likely sales during a “test year”.
This test year is typically a recent one and is meant to capture the likely kWh sales that a utility will make in a probabilistically likely year. But it’s ultimately a guess. An educated guess, to be sure, but a guess — the output of which sets the rates that the utility gets to charge regardless of the actual kWh sales in a real year.
So let’s look at what this means mathematically, in a highly simplified but instructive case. Let’s assume the following utility:
- Total capital costs of $3900/kW for generation and wires;
- All fuel, operating and overhead costs are pass-throughs without markup.
- Rate-of-return based regulation set to provide an 11 percent return on invested capital.
- A “test year” with 6100 kWh of sales per installed kW of capacity.
Per these numbers, the utility would get a rate that included 8 cents/kWh to recover all of their capital costs, which in turn would provide them with an 11 percent return. (I’m calculating with an assumed 20 year return horizon, so this implies that the utility earns $125 per year for each $1000 of capital investment.)
One can quibble with the assumptions, but the point here is to be illustrative. Now let’s look at what happens if utility sales volumes move:
- A 10 percent reduction in kWh sales lowers their annuity to $112 per $1000 of capital investment, dropping their long-term return to just 9.4 percent (e.g., an 14 percent reduction in their anticipated equity return). “This is the kind of thing that tends to piss off shareholders.
- A 10 percent increase in kWh sales raises their annuity to $137 per $1000 of capital investment, raising their long-term return to 12.5 percent (e.g., a 14 percent increase in their anticipated equity return). This is the kind of thing that makes shareholders happy.
Utilities, like any business, are ultimately beholden to their shareholders. And so Mike Morris is simply fulfilling his fiduciary obligation when he objects to decoupling. That doesn’t make him a bad guy — but it does mean that as we think about how to use decoupling to remove utility disincentives, we not lose sight of the counter-balancing incentives that have shaped our modern “coupled” utilities.