So which forms of energy should we subsidize?

Natural gas fracking well in Louisiana, (c) 2013 Daniel Foster

I am often asked what forms of energy should be subsidized. Though much of my work documents how subsidies end up wasting money and don't achieve their stated goals, I don't universally reject the use of public resources to help implement important societal shifts. Clearly, we have a moral responsibility to use both public (and private) money to alleviate suffering in our country -- whether due to a lack of food, heat, or illness. The case for subsidizing systematic changes in our economy (such as to a lower carbon footprint) could be made as well, though not in the way it is usually argued in policy circles. 

Once one starts moving from the theoretical use of subsidies to actual policies that transfer large amounts of wealth, trouble begins.  Implementing a transfer policy that is efficient and effective in achieving the desired structural shifts is not easy.  Often, subsidies are put forth as a solution to problems that could actually be better addressed through desubsidization of competing goods or services. In many other cases, subsidy policy is gamed to the benefit of entrenched interests, and the policies can slow rather than accelerate the desired shifts.

Power LinesThus, I do not have a list of "most deserving" subsidy recipients.  Unlike Congress, I do not believe that I can reliably predict the most effective energy paths for the country.  Finding a cool new energy technology is one thing.  However, identifying the set of options that are actually going to work technically, that are economic to implement, that can gain market share quickly enough to make a difference, and that can be scaled to the magnitude needed without creating large ancillary problems is a different task altogether (see "Energy Transition: Mapping an Appropriate Role for Government").

Thus, I focus instead on getting the policy structure correct so that the best choices are more likely to win out in the marketplace.  This approach channels the dynamism and discipline of market forces, while allowing multiple pathways to be pursued at once.  Since there is attrition at every stage of new innovation, scale-up, and market adaption, having multiple pathways at the start is necessary if there are to be sufficient successes by the end.

For identifying how to allocate public support, something as simple as defining the real end-goal of the subsidies is a critical starting point, but can be surprisingly difficult to nail down.  Is it energy independence (proponents argue we should subsidize anything made in the US or tax oil imports)?  Counter-acting climate change (proponents argue for subsidizing lower-carbon fuels)?  Poverty reduction (where perhaps subsidies focus on fuels used by the poor)?

Political forces pass legislation by pooling together interest groups. In the subsidy arena, this means a "big-tent" process where everybody important (i.e., their support is needed to pass the bill) gets something.  The more powerful groups get more. The "real" objective of the policy often becomes secondary to shifing around piles of cash.

So here's my guidelines on policy pre-requisites if a subsidy program is to have any chance of success:

  • Do no harm. If you are trying to solve a problem, the first step should be to remove subsidies to behaviors that exacerbate it. Thus, if we don’t like high oil imports but still have reduced taxes on foreign subsidiaries, socialize the costs of oil stockpiling, and don't recover the costs of oil security from users -- policies that all make imported oil cheaper -- we ought to fix it.  Similarly, if we care about climate change, stop subsidizing ghg-intensive activities. Ending subsidies to coal mining is an obvious step, but there are many others.  Tax-exempt bonds for pollution control equipment on power plants (clean air is important, but coal has been a main beneficiary of the bonds and should be paying full price for the controls); landfill gas recovery (a favorite of EPA – but fugitive emissions remain high, the control costs should be internalized in tip fees, and the subsidies undermine climate friendlier reuse and recycling); and all sorts of land conversion and cropping patterns that drive up emissions.

  • Focus on outcomes, not inputs. If you decide offer subsidies, subsidize the outcome, not a Congressionally pre-defined pathway to that outcome. Otherwise, you get balkanized policies -- such as different rules for each possible approach to alternative-fueled vehicles; or that focus on lower carbon electricity when there are many cheaper ghg-reduction strategies that have nothing to do with power generation.  Interestingly, federal energy subsidies are nearly always directed towards the supply side.  Demand-side options to curb energy or raw materials use can be less expensive and faster to deploy, but are far too often ignored. This is a costly error. In fact, the outcome needs to be defined carefully to be neutral in its impact. Rather than “reduce gasoline use in the United States” it needs to be something like “reduce energy consumed per passenger-mile or vehicle-mile traveled." That way, you have all drive technologies competing against each other, while also opening the playing field to innovative ways to reduce the ghg-intensity of transport -- boosting load factors or shifting transport modes, for example. With something like energy for the poor, there is probably no way around some subsidies to fuel consumption. However, weighting total spending too heavily on supplying fuel to old and leaky homes – rather than financing insulation – is also self-defeating.  Addressing structural problems (for example, split incentives on energy-efficient capital upgrades for landlords and tenants that result in sub-optimal conservation investments in rental housing) can also be more important over the long-term.

  • Pool resources linked to outcomes; force alternative pathways to compete for funds on offer. Once you’ve defined the objective carefully, the next step is to put a single subsidy budget linked to that objective and to open it to competitive tender. Thus, rather than ten separate subsidies to alternative fuels (each carefully crafted by lobbyists to guarantee public money to their clients), you have a larger pool of money that both entices more activity and forces the alternative ways to meet the target objective to bid against each other. The lowest subsidy increment per objective delivered (e.g., per mt CO2e reduced) would win. This type of approach is at the core of many renewable portfolio standards.  However, in addition to restricting ghg reductions to just the power sector, the RPS' are often layered on top of all sorts of other subsidies, distorting the reasons particular resources are able to submit winning supply bids.

  • Even the good solutions have drawbacks; don't ignore them. Resist the temptation to cut deals for power sources you “just know” are the right path. For example, all of them should have to go through environmental permitting (MN exempted ethanol facilities some years back). All of them should have to pay for the water they use (virtually no thermal power plant in the country pays for cooling water, even though the sector uses as much as agriculture). All of them should have to have appropriate bonding or post-closure funds if they are leaving big changes to the landscape that will need to be dealt with by taxpayers should the firm go belly up. If these rules are applied evenly, they shouldn’t disadvantage the newer energy technologies. They may, however, force manufacturers to deal more carefully with problems ahead of time -- for example, large concrete pads at remote windmill sites; heavy metals in PV panel production; or massive land conversion for biofuels. Problems do arise if high standards are being set for emerging energy resources while conventional fuels have been exempted from dealing with many of these types of problems for decades (e.g., drilling wastes from oil wells or socializing many of the costs of nuclear waste management). The solution needs to be bringing conventional fuels into appropriate controls rather than exempting emerging fuels from them as well.

Here’s a graphical example of what happens all too often when subsidies are politicized.