The U.S. Department of Energy recently offered up a “Grid Resiliency Pricing Rule.” Identifying the possibility that some generators might not be able to get fuel during cold snaps, the rule would require that electricity suppliers or grid operators obtain backup power from generators that can keep 90 days of fuel onsite, paying a price sufficient to cover all of their costs. Many commentators view this as a barely disguised attempt to sustain business for coal and nuclear plants, since generators running on natural gas (the dominant fuel) can’t as a practical matter keep that much gas on hand.

Their suspicions certainly mesh with candidate Trump’s campaign pledge to revive the moribund domestic coal industry at a time when falling natural gas prices are primarily responsible for coal’s decline. Since the Trump administration is hardly likely to impede fracking — which is largely responsible for the increased U.S. supply of natural gas and the decline in its price — the DOE’s back-door mandate may be the best option left to boost coal’s prospects.

However well founded, though, those suspicions don’t necessarily mean the proposal is without merit. The real question is whether there is a genuine problem here that it could fix, at least cost to electricity users. To answer that, we need to step back and look at three crucial aspects that combine to make the production and distribution of electricity unique.

Start with the reality that because mass storage of electricity is not practical (though this may be changing), the machines for generating extra electricity have to be available to meet surges in demand at a moment’s notice. Some of that capacity will be used only a tiny fraction of time — as little as 1 hour in 200 across the year — making it very expensive per kilowatt generated.

In principle, each user of electricity (residential or commercial) could decide whether that expense is worth it by choosing a supplier of electricity based on how much backup capacity it maintains. But the second special aspect of electric power — that electricity cannot be routed to specific customers on a shared distribution network — rules out this approach. As described in detail in my recent Resources for the Future post, if one generating company cannot supply all the electricity its customers demand, all users on the common grid are affected. This has long made reserve adequacy a legitimate concern of public policy.

The most straightforward way to ensure reserve adequacy is for a regulator to mandate the minimum generation capacity that all electricity suppliers must have on hand. Generators could then set electricity prices high enough to reflect the extra cost of meeting demand during those rare times when supplies are tight. As noted above, this would be expensive. Regulators are thus reluctant to allow power producers to cover the costs of maintaining reserve capacity by sharply increasing prices during periods of shortage, because of both the political consequences of exposing buyers to high prices and the concern generators might withhold output to raise prices even more. This leads to the third special aspect: regulators aren’t likely to allow producers to recover the full cost of maintaining reserve capacity by charging an arm and a leg during those rare times when the market is tight. In the jargon of power producers, this is known as “missing money.”

To provide the missing money, most regional electricity grid operators have adopted “capacity markets” in which generators can get payments on top of what they receive from selling the electricity itself. The capacity market approach also fits with the grid administrators’ need to cover the costs of generation capacity held in reserve in case of a summer heat wave or the unplanned shutdown of a major supplier to the grid. The growth of wind and solar power has led to a second reliability-related justification: to provide electricity on very short notice when the wind stops blowing or clouds block the sun.

DOE’s Notice of Proposed Rulemaking posits a third reliability rationale — the one it directly ties to the extra-fuel-on-site mandate. The department asserts that during severe cold snaps, natural gas generators may lack access to adequate supplies because regulators give priority to home heating. In such circumstances, the economist’s glib “just get the prices right” in the market for gas may not work. On top of reluctance to let electricity prices rise, officials are likely loath to let the price of natural gas for home heating rise to the point at which low-income households are forced to choose between freezing and starving.

DOE may have done a public service in identifying this problem, but that doesn’t get us to its specific proposal. The first step toward ensuring that generation companies’ promises to make capacity available are binding is to set stiff penalties for those who fail to deliver what they’ve promised.  

The next step is to let markets work to minimize the cost of ensuring reliability. DOE should follow a long-standing principle of technological neutrality rather than prescribing the type of capacity needed to meet binding commitments during cold snaps — or any other time. If it turns out to be less expensive for gas-based power providers to insure against gas shortages by building their own storage tanks or for generators that rely primarily on renewables to install batteries than by keeping coal-fired units in operation, so be it. Moreover, with binding delivery commitments, the capacity market, not DOE, should determine how much backup is needed — whether it is 90 days, a week or perhaps less.

By the same token, DOE should not be in the business of dictating the price to be paid for this backup capacity, and certainly not in the business of setting the price high enough to cover the costs of coal plants. If the point is to improve reliability — and not simply to redistribute cash from electricity users to coal-based suppliers — the market will set the price just high enough to induce the appropriate supply of backup capacity.  

The coal industry would have valid complaints if providers of renewables-based power were allowed to participate in capacity markets even though they lack the ability to deliver when energy is needed. But regulators should always keep in mind that low prices resulting from greater competition and declining costs are generally indicators that a market is working well, not poorly.

Two other points are worth noting in evaluating DOE’s proposal for making grids more resilient. The DOE NOPR has next to nothing to do with most of the blackouts we experience, which follow from damage to local distribution lines. Federal electricity policy only addresses the rare region-wide outages, such as the Northeast blackouts of 1965 and 2003. 

The second point is political and legal. I do not dispute the legitimacy of the Trump or any administration’s wish to deliver on campaign promises. But every administration is obligated to play by the rules set by Congress and the courts. Hence, regulations, including DOE’s grid resiliency proposals, cannot be “arbitrary and capricious.” The staff of the Federal Energy Regulatory Commission, which would be charged with writing and enforcing this rule, is now asking the right questions. Let us hope that good economics and the rule of law carry the day.