How (and who) will pay for our energy infrastructure?
You may have heard that the Federal highway trust fund is running out of money because, darn it, people aren’t using enough gasoline. The transformation of our energy system is rapidly accelerating the need to confront a long-standing problem with how we pay for our transportation and utility infrastructure. For the most part we have paid for this infrastructure through taxes or surcharges on the fuels associated with them. Highway construction and maintenance is supported primarily by a tax on gasoline. Natural gas and Electricity distribution infrastructure is funded through volumetric charges on energy usage.
While at first blush this may seem like a logical, even efficient, approach the problem is that the costs of this infrastructure do not scale with the consumption of these fuels. A BMW 7 series may tear up the same pavement as a Tesla S class, but only the Beamer is going to be chipping in for the repairs. What was once a tolerable and subtle cross-subsidy has turned into a serious funding problem. The somewhat paradoxical problem is that, as our energy consumption gets more efficient, we contribute less to these infrastructure costs. Since the costs don’t go down, we’re left with a funding shortfall.
On the electricity side, the infrastructure is funded through rates that are linked to the volume of kilo-watt hours (KWh) consumed. Its an even more extreme version of the gas-tax problem. Imagine a world where your fourth tank fill-up in a month costs 3 times as much as your first fill-up. That’s a typical electric rate in California. 
This figure illustrates the rate structure for PG&E in my area. The more you consume the higher the per-kWh price gets. On the fourth step (tier) prices rise to over 32 cents/kWh. The problem is that the kWh I consume cost PG&E around 10 cents/kWh (Even that 10 cents includes a bunch of fixed costs like wholesale grid charges, and the costs of funding the CAISO and the CPUC). The rest pays for transmission, distribution, and other “system” costs – including bond payments for the electricity crisis – that don’t go down when I reduce my consumption. PG&E recovers these fixed costs through a usage-based, per kWh fee.
While this can be a powerful incentive for conservation (and for installing solar panels), it creates a direct conflict between the goals of promoting efficiency and keeping our infrastructure funded. The less gas or electricity I consume, the more those costs need to be recovered through higher taxes or electricity rates. You could argue that these “mark-ups” of energy costs help to offset another market failure, the environmental cost of energy consumption, but in California we are already pricing the CO2 – at least somewhat. You could make the case that gasoline should still cost more, but on the electricity side, its pretty tough to justify a 32 cent/kWh price based on environmental damages alone.
These funding systems can also create a situation where individual consumers make choices driven by opportunities to shift costs onto others. The guy who came to my house last week – yes we have door-to-door solar salesman in Davis – offered to install solar panels for a contract starting around 17 cents a kWh (by the way, sales guy, if you’re reading this call me). If I take this offer, I’m swapping out power that costs 17 cents for power that cost 10 cents – a losing proposition, right? Except that I’m paying 32 cents for that power and take that other 22 cents or so on the last tier of my electric rate and shift those costs over to other PG&E customers. So I’m saving 15 cents a kWh (again, call me…) but total costs, thanks to the contributions of other PG&E customers, have gone up by 7 cents.
On a small scale, we’ve lived with this for many years, but we are rapidly reaching the point where raising rates on those without solar to recover these infrastructure costs won’t be viable any longer. That leaves us with three choices
- Stick utility shareholders with the costs.
- Reduce the money we spend on infrastructure – perhaps dramatically.
- Disconnect the recovery of these costs from the usage of fuel.
Many jurisdictions have been searching, sometimes clumsily, for ways to implement option C. We have the infamous “prius tax” in South Carolina and the solar tax in Arizona. Here in California, we have a proposed road user fee that would start to recover some highway costs on a per-car, rather than per gallon of gas, basis.
In utility regulation the phrase “revenue decoupling” has been around for a long-time. The concept was originally pursued as a means to stop utilities from trying to expand the volumes of product that they sell, and to help incentivize them to embrace energy efficiency. At its core, revenue decoupling means finding a way to allow utilities to recover their prudent fixed costs in a way that’s not linked to consumption. Some kind of user fee that’s not based on volume is the definition of revenue decoupling. Distributed generation and improved efficiency aren’t likely to reduce infrastructure costs dramatically anytime soon, so I see more user fees in our future.
Categories
Energy Institute Blog 
As Severin’s earlier post points out, fixed charges (user fees) don’t have much basis in economics. Another approach to ensuring adequate fixed cost recovery while minimizing cost shifts and trying to better align incentives is proposed in this paper, http://nccleantech.ncsu.edu/rethinking-standby-and-fixed-cost-charges-regulatory-and-rate-design-pathways-to-deeper-solar-cost-reductions/.
And slides: http://votesolar.org/wp-content/uploads/2015/01/Jim_Kennerly_FINAL.pdf.
Sean Gallagher
Vice President, State Affairs
Solar Energy Industries Association
P: (415) 261—7198
sgallagher@seia.org
The proposed road user fee in California is a fixed fee per car independent of miles driven or actual road wear and tear caused. Alternatively, look at this pilot program from Oregon DOT that ties revenue to miles driven:
Click to access RUCPP%20Final%20Report.pdf
On the electricity side, this is perhaps analogous to a demand charge, mentioned already by Mark Lively. You need to pay something for (roads or wires) that is distinct and different from (kWh or gasoline). The latter is volumetric, but is the former? What is better: a fixed charge per customer, or a charge per kW?
Wait a minute. “Stick utility shareholders with the costs”? The shareholders have been sticking it to the public with the externalities from their power generation. Air quality causing health issues, emergency room visits, higher infant mortality, and that’s before we talk about climate change. Three are huge costs not born by the shareholders or other rate payers. We should be saying thank you for stepping up to homeowners that invest 10 to 40 thousand dollars (directly or indirectly).
As i don’t have solar power at my home, and I rely on the grid, I do say thanks to those making the investment to improve air quality and our climate future and I hope many join you.
This is a case of using the wrong instrument (progressive pricing) to chase after equity. For reasons suggested here and by Borenstein, there are sizable efficiency losses due to deviating from efficiency pricing. There must be less costly ways of helping low-income households (Kaplow and Shavell). As suggested by Vernon Smith in the late 70s, the long run marginal cost of providing electricity is rising in the relevant range. That means you only need two blocks. The marginal block is set at full marginal cost. The inframarginal rate is set at a lower rate and used to avoid generating a revenue surplus. We still end up w/ increasing block pricing, but we have a more rational basis for setting the block prices and quantities. The other part of the mess is subsidizing renewables instead of pricing the externality in the first place. This seems to stem from a misinformed notion of sustainability. (Heal: Optimal is sustainable but not the other way around.)
Policymakers tend to ignore the old admonition to “be careful what you ask for”. In the case of electricity, they’ve determined with some prompting by certain advocacy groups that promoting adoption of rooftop solar is good public policy. Unfortunately it appears they never gave any thought to the other implications. At the moment it isn’t cost-effective for most customers to become self-sufficient because storage is still very expensive, which means the degree of cost-shifting and the amortized cost of providing electric infrastructure remain manageable. However the moment storage becomes really cheap is when policymakers will have to deal with the impact on those customers who generally can least afford high electric rates and self-sufficiency. My guess is that customers who remain connected to the grid will pay a lot more and everyone else will be taxed to fund subsidies for low income consumers. More than anything else it is the prospect of being caught on the wrong end of the cost-shifting shell game that drives me away from considering rooftop solar, but if storage ever became attractively priced, I might consider cutting the cord solely to make a point.
Infrastructure is not very different from national defense .. it is ‘required’. In both cases there are differences on how individuals might feel about how wide, where, etc roads to build and how many wars to engage in. Ideally Congress should do right by the country; practically the highest paying noisiest lobbyists get their way.
Road costs should be shared by society [general funds/ taxes] and by users [drive more pay more]. It is not too difficult to make auto taxes mileage+weight based; gasoline cars pay their share of the CO2 effect via CO2 taxes in the price of gasoline, and electric car drivers pay via a CO2 component in the price of electricity.
Usage-based price of electricity should be based on the marginal cost of production — if I use a lot of electricity at night, when the cost of production is lower [due to lower demand] I shouldn’t get bumped in to a higher tier. These legacy pricing mechanism are left over form the old manual meter/ monthly reading times.
Time-of-use [variable-cost-of-production component] and peak-demand [capacity] would be central components of a pricing model. People living farther away from the sub-stations would have to pay a higher ‘transmission’ component, and that can be set at the time of the connection when distances are determined.
Great article. The legislature will announce new climate change legislation today that will propose to move California’s renewable portfolio standard (or use of renewables for energy production) from 33% to 50% by 2050. This article illustrates a great point: We need to understand the true fully burdened delivered to load cost of all of these well meaning policy initiatives that have been mandated. Failure to do an honest review will result in what this article is highlighting. How will we pay for all of these policies? How will the confluence of policies impact our ability to actually achieve the goals/mandates that we are proposing to achieve? How will California’s economy be impacted? Who pays for it? Who will suffer? I hear words from legislators that claim to want to hear about these potential impacts — but I don’t see any movement to really want to understand them. Dr. Bushnell, maybe you could convince them to really make an effort to tackle the question. I hope so.
There are big differences between highways and electric wires, primarily that the wear and tear on the highways are roughly proportional to the miles that cars drive on them. The wear and tear on electric wires have almost nothing to do with the energy that goes through them.
But highways and electric wires both have a similar cost driver, their installed costs are driven by the anticipated peak demands, whether rush hour on the highway or the daily peaks on the electric system. Fortunately for electricity there is a way to charge for such peak usage, a demand charge, as I discuss in both
“Curing the Death Spiral,” with Lori Cifuentes (Tampa Electric Company), Public Utilities Fortnightly, 2014 August.
https://www.fortnightly.com/fortnightly/2014/08/curing-death-spiral?authkey=54d8da5efd3f76661023d122f3e538b4b3db8c8d5bf97a65bc58a3dd55bb8672
http://bit.ly/1oF2twM
and in
“Demand a Better Utility Charge During Era of Renewables: Getting Renewable Incentives Correct With Residential Demand Charges,” Dialogue, United States Association for Energy Economics, 2015 January, http://dialog.usaee.org/index.php/volume-23-number-1-2015/271-lively
I made this point last week in a presentation to the staff of the U.S. Department of Energy Solar Energy Technologies Office in “The True Impediment to More Rooftop Solar: Revenue Stability for Wires Companies,” DOE Sunshot Discussion Club, Washington, D.C., 2015 February 3
Electric utilities have used demand charges for years, including residential demand charges. I first saw a residential demand charge in the Burbank, CA, tariff almost 40 years ago. About a dozen U.S. utilities have residential demand charges, and their use is growing.
One other way to look at it is that these costs are doing exactly what we want – driving people from coal-fired grid to solar, or to reduce the use of gasoline (through public transit, or more efficient vehicles). If we raise the tax on gasoline to pay for the infrastructure then hopefully that will push more people to fuel-efficient vehicles (increase the Hybrid:SUV ratio). With solar, its interesting that your option #1 is wored “Stick utility shareholders with the costs”. In any other sector, if a businesses’ operators had ignored the trend towards renewables, and spent their time and money battling its adoption rather than embracing it and getting out ahead then we would expect the utility shareholders to bear the costs, why should it be different for utilities ?
One challenge of course with any other solution is that it also may be fighting against a rising tide of alternatives, for example if the annual cost of an electricity hookup is raised, then as storage and efficiency becomes cheaper it becomes more cost-effective to disconnect entirely.
Made a similar comment in the discussion about redesigning residential rates to incorporate customer charges–why is a discussion about shareholders bearing more of the risk from changing technologies apparently off the table? The utility ratemaking model was set up in anticipation of a technologically static world–that’s no longer the case (and was it ever?) Without facing risk, utilities have much less incentive to adapt. They won’t necessarily be “stuck with the bill” if they adapt their management practices. This is not a zero sum game, and the total sum will be smaller unless utilities are given the right incentives.