Photo Credit: PJM
Clean energy advocates have historically fought hard for every single inch of progress made. As a result, there is a tendency to focus on the immediate opportunities in deployment, rather than thinking about some of the challenges that are coming further down the road.
The long view is important when considering how to adapt electricity markets for a renewable-dominant system. The key issue is determining which electricity market structure can best accommodate renewable development, compensate resources efficiently, and maintain reliability.
There are still a lot of unanswered questions about how exactly to transition to an electricity system where most (i.e., more than 80 percent) of the electricity comes from renewable energy sources. Much work has been done to verify that high penetrations of renewables are technically feasible, but relatively little has focused on what is needed from electricity markets.
One little-discussed way to meet these objectives is to create capacity markets that allow for bundled bids from renewables. This model establishes an architecture to allow renewables to grow without fear of hitting a ceiling, ensuring a steady revenue stream for clean energy while providing incentives to invest in baseload backup power.
Ceiling? What ceiling?
The current electricity market is not designed to handle high penetrations of renewables. Eventually, enough renewables will come on-line to spark market changes. The sooner the market shifts, the better; until then, progress will be moderate under a theoretical ceiling for development.
Here’s why renewables may hit a ceiling: Wholesale electricity prices are driven by marginal costs. This means that at any moment, the price of electricity is met by the most economically efficient power plant needed to meet demand.
Imagine a reverse auction where demand is 500 megawatts. Every power plant bids its marginal cost of generation into the auction. These costs are closely tied to fuel costs, meaning that fossil fuel plants are always higher in price than solar and wind (which are close to zero). The price of electricity is set by the last unit needed to meet demand, so even if you had 490 megawatts of demand met by low-cost wind, if the last 10 megawatts come from a coal plant with a $20 per megawatt-hour bid, the clearing price for all power plants becomes $20.
So what happens if demand is entirely met by renewables? The clearing price becomes zero dollars, and the power plants receive zero dollars for energy produced. This means there is no more revenue stream for the power plants, including renewables.
In the current system, renewables actually need fossil fuels to make money. There is already strong opposition to renewables because of their tendency to suppress prices. But once renewables start meeting most of the demand, they could drop prices to zero.
This also creates significant risks for investors. Why would a bank provide financing to a project that has such limited cash flow? Furthermore, why would anyone want to build baseload that may only garner revenue periodically? That means no backup power for the grid, a less-than-ideal outcome, particularly during the transition to significant amounts of renewables.
How capacity markets help
Capacity markets are intended to ensure reliability and provide long-term price signals to power plants in the electricity market. They provide revenue to power plants for their available capacity, rather than for the power that they actually produce.
Under a capacity market, the system operator holds a reverse auction several years in advance of the delivery date. Power plants bid in their total cost of operation (capital costs plus operating costs). A stack is created, with the last unit setting the clearing price for all the power plants.
Since the power plant bids are based on total cost of operation, the first units to clear are usually depreciated power plants. Renewable plants could be slightly less competitive in these markets, as the higher cost of capital required to finance them makes their total cost more expensive. However, experience in the PJM capacity markets shows that this hasn’t been prohibitive for renewable power plants -- in fact, they have actually tended to lower clearing prices.
There are a few benefits to this model. First, capacity payments provide a stable, long-term revenue source for renewable power plants based on the capacity they offer. This helps with getting financing at the front end, which is particularly important given the comparatively high capital costs of renewable projects. In a roundabout way, this puts the risks on consumers instead of investors, which is more akin to the traditional rate-of-return model.
Second, these payments give baseload generators an incentive to be a “standing reserve” in an energy market that will increasingly be dominated by variable resources. This creates a market flexible enough to meet fluctuating demand while ensuring reliability.
Getting the capacity market design right
There are a few changes that could be made to capacity markets to help them better serve renewables.
If demand is forecasted inaccurately, there may be excess capacity. This may have the effect of keeping inefficient fossil fuels on life support. This is the obvious downside to having a “standing-reserve” capacity fleet for reliability. But other than improving forecasts, what else needs to change?
One tweak that could make a major difference is allowing renewables to issue bundled bids. Currently, a power plant owner bids in the capacity it can ensure will be available to meet peak demand in the auction delivery year. Because wind and solar are variable, by definition they cannot guarantee capacity in the same way that a gas plant can. Instead, they bid in a percentage of what is called their “on-peak” capacity. This number reflects a portion of the capacity the resource has reliably provided during peak hours. This “on-peak” number is also used in reliability assessments, which the North American Reliability Corporation carries out annually.
So how would bundled bids work?
Imagine two dispersed 10-megawatt wind farms and two dispersed 10-megawatt solar farms that can only bid 40 percent and 20 percent of their capacity, respectively. The wind farms and the solar plants bid 12 megawatts out of 40 megawatts available.
However, during that “on-peak window,” the probability that one or the other of these geographically dispersed resources will be available is much higher. Even if there is cloud cover and low wind in one area, there is still a statistical chance that conditions will not be the same in the other region. Therefore, the amount of capacity that can be expected on peak from this bundle could be calculated to be as much as 40 percent. That would allow 16 megawatts to clear in the auction instead of 12 megawatts. Furthermore, you could factor demand response as a backstop in some of these bundled deals to ensure that some capacity is available in a worst-case scenario.
This may seem like small change. But if clearing prices are high enough, it could make quite a difference. And if the probability of on-peak capacity is higher (because the resources are located farther apart or there are more of them in the bundle), it could result in much higher returns.
This approach is already used in the hedging and derivatives world, a domain with which many energy traders are already familiar. It takes advantage of the fact that variability decreases as forecasting improves and as renewables are more spread out geographically. Continuing to allow demand response and energy efficiency to compete in capacity markets is also essential to this approach.
Creating capacity markets can help smooth the transition to a clean energy future, but they must be designed with a high-renewable scenario in mind.