Facing the Risk in Fossil Fueled Electricity
I recently wrote about how our risk tolerance/aversion powerfully affects our estimation of the social cost of carbon, but obviously that’s not the only place that risk shows up in our energy systems. Fossil fuel based electricity is also exposed to a much more prosaic kind of risk: the possibility that fuel prices will increase over time.
Building a new coal or gas plant is a wager that fuel will continue to be available at a reasonable price over the lifetime of the plant, a lifetime measured in decades. Unfortunately, nobody has a particularly good record with long term energy system predictions so this is a fairly risky bet, unless you can get somebody to sign a long term fuel contract with a known price. That doesn’t really get rid of the risk, it just shifts it onto your fuel supplier. They take on the risk that they won’t make as much money as they could have, if they’d been able to sell the fuel at (higher) market rates. If the consumer is worried about rising prices, and the producer is worried about falling prices, then sometimes this can be a mutually beneficial arrangement.
In many cases, these kinds of “hedging” arrangements are worked out with speculators as the middlemen, via the commodity futures and derivatives markets. But there are many risks that these markets aren’t interested in betting on, as Mark Bolinger at Lawrence Berkeley Labs pointed out in a report last fall on the hedging value of wind resources in an era of cheap natural gas (which we also wrote about in January):
Although short-term gas price risk can be effectively hedged using conventional instruments like futures, options, and bilateral physical supply contracts, these instruments come up short when one tries to lock in prices over longer terms – e.g. greater than five or ten years.
Coal faces a similar problem, in that production costs have been escalating, while many mining operations, especially in Wyoming’s Powder River basin, are faced with making large capital investments to continue or expand production — investments which would take decades to pay off. Their credit ratings are being downgraded, increasing their cost of capital, not coincidentally at exactly the same time as the future of demand for their product is being called into question. This is a death spiral not so different from the one faced by the utilities that make up the coal industry’s entire customer base.
The result is that it has become hard to maintain confidence in a long term supply of attractively priced fuels for electricity generation.
A better solution for utilities.
Now, if you’re a monopoly regulated utility, you’ve got alternatives that are much better than any futures contract. Instead of paying a premium to lock in reasonable fuel prices for the long haul, you can get permission from the local utility commission to make your (captive) customers pay for the fuel, no matter how much it ends up costing. Historically, this is exactly how regulated utilities have worked, and one of the main reasons that they have long been seen as extremely safe investments — they are able to offload their unpredictable expenses onto their customers, who have no choice but to pay. In Colorado, this is done via the Electric Commodity Adjustment, which today makes up about 30% of our bills (in the parts of the state served by Xcel Energy, anyway).
Most regulated utilities in the US make their profits from capital investments — things like power plants which require a lot of up-front financing. The profit they receive is closely linked to their cost of capital — the returns they offer to investors in the debt (bond) and equity (stock) markets, in exchange for access to mountains of cash, which can be used to build said power plants.
Fuel isn’t a capital cost, and utilities don’t really earn a return on the money they spend to buy it. They don’t care one way or another about the fuel costs, so long as they get permission to build expensive power plants. This creates a perverse incentive: since nobody can really predict fuel prices, and they don’t actually care what the fuel prices end up being, they can get away with giving the public utilities commission whatever fuel price forecast allows them to justify their preferred capital investment program. The fuel price forecast will end up being wrong, but that’s okay — for the most part everyone else’s forecasts will also end up being wrong, so they won’t look bad. Their investors get a return on the capital deployed to build power plants. The customers get stuck with the fuel bill, but most of them don’t really understand how the system works and don’t have any other options for their electricity, and so just they just suck it up and write a big fat check every month, without realizing there might be another way to do things.
It doesn’t have to be this way.
What we have above is a classic example of a split incentive or principal-agent problem: the person with the power to make decisions that end up determining costs (aka the agent, in this case the utility) is not the same person who bears those costs (aka the principal, in this case the customers). The utility wants to maximize capital investments — which they earn a return on — while minimizing their own exposure to risk. Their customers would presumably like to strike a balance between minimizing energy costs and minimizing the risk of future rate increases.
Once upon a time, when there weren’t any options besides burning stuff (unless you were lucky and had some local hydro resources) this split incentive wasn’t a huge problem. But today, renewable electricity and energy efficiency have very different cost and risk profiles than fuel based power. If we (foolishly) continue to ignore externalized costs, or if we decide as a civilization that we are indifferent to climate risks and/or don’t value the future, then renewable electricity is often more expensive per kWh than fossil fueled power. But you can sign a power purchase agreement (PPA) for wind or solar with fixed costs that runs for 15-25 years. Investments in energy efficiency and demand response (giving utilities the ability to reduce electricity demand on command) also have predictable costs, last for decades, and displace much more expensive and unpredictably priced fossil infrastructure. You can’t effectively hedge fuel costs in the commodity markets on those timescales, so if you have any kind of risk aversion it’s worthwhile to choose modestly higher fixed costs from renewables and efficiency over slightly lower but potentially very volatile fossil fuel fired electricity costs.
There are two obvious ways to fix the split incentive, and reap the benefits of investing in modestly and predictably priced options for our future electricity. We need to unify the decision making power and the financial consequences. We can put both the decisions and the consequences in either the hands of the utilities, or in the hands of the customers.
Giving customers both the decisions and the consequences is a little bit complex, but potentially doable. Because the electrical infrastructure lasts a long time, their decisions would need to be durable — they’d have to stick for decades. When the generation facilities are owned by the customers — as in the case of rooftop solar PV, or shares of community solar projects, ownership of the generation assets would be clear and presumably transferable, and the market price of those shares or the rooftop systems would reflect the overall cost effectiveness and riskiness of the resource. If they stopped short of defecting from the grid entirely, customers would still need to cover the cost of backup power, integration, and distribution services that the grid would provide. The benefit of this approach is that different individuals and institutions with different risk tolerances would be able to choose the mix of price level and price stability that was best for them. The problem with this approach is that most individuals and institutions have no idea what that means, and probably don’t particularly want to have to figure it out (in exactly the same way it’s like pulling teeth to get people to plan and invest for their own retirement.). It would also mean exchanging the risk of fuel prices going up for the risk of renewable energy prices going down. But that’s a risk that we actually want people to take on… so offering a modest, relatively safe return on such capital investments is reasonable. This is exactly what a feed -in-tariff does.
Creating a risk-aware utility.
Giving utilities both the decisions and the consequences would probably be done at the state level. Note that implementing this option need not mean excluding the customer choice/feed-in-tariff model. Rather, the creation of a risk-aware utility would change the default price/risk allocation that customers are exposed to. This would almost certainly be a more equitable deal than the one we’ve got now, where customers unwittingly accept all of the risk resulting from the utility’s self-interested resource planning proposals.
Today utilities have an incentive to forecast low fuel prices since it allows them to justify building power plants, from which they make their profits. Then they recover the fuel costs from their customers quarter by quarter, pretty much as they buy it. In Colorado, those costs show up on your electric bill in a line item called the Electric Commodity Adjustment, which fluctuates along with fuel costs.
Shifting all the fuel price risks to the utility companies would mean requiring them to commit up front to a fixed fuel cost recovery rate for each power plant. In order to guard themselves against the possibility of future fuel price increases, this would mean the fixed rate would need to be much higher than current rates. Whereas in the current system, customers have virtually all of the risk, in this hypothetical fixed-cost electricity market, the utility has internalized all of the risk, and would be expected to pass the cost of accepting that risk through to the rate paying public. If they were not permitted to pass the costs through, then the nature of investing in utility companies would change — and the rate of return their investors demand would probably go up, increasing their cost of capital… which would in turn increase the prices they’d need to charge.
It’s important to notice that in the above we haven’t actually changed the amount of risk in the system — all we’ve done here is change who is exposed to the risk, and put a price on it — all of it. This is known as the certainty equivalent price — the price you pay to have absolute knowledge of future costs.
But there’s no reason to think that this is the most cost-effective solution. Rather than simply pricing the risks inherent in a fuel-dependent energy system, we should be looking at how the energy system can be changed to mitigate fuel risk.
A better solution for utility customers.
Electricity systems with less financial risk are possible. Instead of simply demanding very high fixed rates to protect themselves against possible fuel price increases, a utility could make investments that reduce dependence on fuel altogether — curtailing overall energy consumption by improving end use energy efficiency and building a portfolio of renewable resources whose energy costs are well known at the time of construction.
NREL put out a study last year trying to quantify the hedging benefits of integrating renewables into the grid. They considered three renewable mixes (mostly wind, mostly solar, and equal parts of each), as well as varying total renewable penetration from 10% to more than 50%. They also looked at how the risk mitigation benefits were different in coal dominated vs. natural gas dominated grids, and under three different natural gas price scenarios.
Unsurprisingly, they found hedging benefits were greatest when:
- the grid is dominated by natural gas, which can easily ramp up and down to accommodate variable wind and solar.
- natural gas prices are high.
- the renewable portfolio weights wind and solar equally.
In a system where fuel-based power is dominated by natural gas, the hedging benefit of renewable energy was roughly linear, at least up to the maximum penetration of 45% that they examined.
Annualized variable cost of electricity with RE penetration for the 50:50 solar-wind case and coal- and natural gas-dominated fossil scenarios. From this NREL report
If we assume that we are going to successfully decarbonize our electricity supply, then this is an important combination of circumstances to consider, since the first thing we should probably do is shut down our remaining coal-fired power plants. If replace that capacity with natural gas, we’re committing to both significant fuel price volatility, and decades of still substantial carbon emissions. If we replace that capacity with a diverse portfolio of renewables, we can get much less volatile energy prices and vastly lower carbon emissions, while we tackle the challenges of integrating massive amounts of variable generation and (eventually) scalable electricity storage. Will this electricity be more expensive than what we’ve got now? That will depend in no small part on how quickly and how far renewable energy costs continue to decline. But with continuing consolidation in the industry, and increasing economies of scale, who knows. And again, all these cost assessments depend powerfully on your risk tolerance and valuation of the future. If you would like to be fairly certain that we’ll make it through the next century without inducing catastrophic climate change, and you think the future is worth preserving, then renewable energy and efficiency probably look like a good deal, and carbon emissions probably look fabulously expensive.
Which brings us back to that other less financial kind of risk…
As things stand today in the world of regulated monopoly utilities, it’s not entirely clear how they would respond to a price on carbon. One could certainly imagine a scenario in which the cost of carbon is simply passed through to customers, in the same way that fuel costs are passed through today. This would result in exactly the same kind of gigantic split incentive I outlined above. If the utility, which has most of the decision making power when it comes to what kind of generation facilities to build, isn’t ultimately on the hook for the carbon costs, then they will have little incentive to reduce the carbon intensity of the electricity they produce. Instead, the price of electricity would simply rise, sparking ire from customers..
This is actually how we came to the issue of how fuel costs and risks are passed through to utility customers — we were looking at attempting to price carbon in Colorado, and it became clear that a price alone (politically challenging though it would be) probably wasn’t going to be enough to get the job done. We also need to re-examine how costs and risks are allocated more generally in our energy systems. We need to have regulatory mechanisms that allow us to transition away from sunk costs when they’re no longer cost-effective, or when they expose us to unacceptable risks — financial and otherwise. In an ideal world, the Public Utility Commissions would be well aware of these issues, and would seriously consider them in their oversight of the utility industry.
Unfortunately, that doesn’t seem to be where we’re at today.
Featured image of a solar farm in St. George, UT Creative Commons licensed by Carl Berger on Flickr.
Zane Selvans is a planetary geophysicist turned energy and climate policy wonk. In graduate school he studied the climate history of Mars and tectonics on Jupiter's moon Europa at the University of Colorado's Laboratory for Atmospheric and Space Physics, the California Institute of Technology, and NASA's Jet Propulsion Laboratory.
After receiving his Ph.D. in 2009, Zane decided to re-focus his ...
Other Posts by Zane Selvans
The Energy Collective
- Rod Adams
- Scott Edward Anderson
- a b
- Charles Barton
- Barry Brook
- Steven Cohen
- Dick DeBlasio
- Senator Pete Domenici
- Simon Donner
- Big Gav
- Michael Giberson
- Kirsty Gogan
- James Greenberger
- Lou Grinzo
- Tyler Hamilton
- Christine Hertzog
- David Hone
- Gary Hunt
- Jesse Jenkins
- Sonita Lontoh
- Rebecca Lutzy
- Jesse Parent
- Jim Pierobon
- Vicky Portwain
- Willem Post
- Tom Raftery
- Joseph Romm
- Robert Stavins
- Robert Stowe
- Geoffrey Styles
- Alex Trembath
- Gernot Wagner