The Real Reason Behind Coal’s Cycling Woes

Integrating renewables into the current mix of resources sure does get a lot of attention these days. Sadly, the issue has been thrown up as an unnecessary barrier to the development of wind and solar. One of the most pervasive arguments I’ve heard suggests integrating variable resources (like wind and solar) is costly and sometimes physically impossible. But data recently analyzed by UCS adds to the growing body of work that undercuts such arguments.

In this blog, we’ll explore why the premise of this argument is not only deeply flawed but its entire framing ignores the basic principles and intentions as to why states joined wholesale power pools in the first place. It is also worth noting that we do need flexible resources and that fossil fuels aren’t the best positioned to provide that flexibility.

Before we get into all that, I want to take a special moment to thank UCS Associate Analyst Ashtin Massie for her indispensable contributions to this analysis.

A deeply flawed premise

Coal-fired power plants are ramping up and down over the course of the day, but some analysts seem content with presuming that wind and solar are the culprits to coal cycling.

The latest research from UCS looks at the Midcontinent Independent System Operator (MISO) which spans from North Dakota to Michigan and down to the Gulf states of Louisiana and Mississippi. There is a fair amount of wind in MISO, with more coming online every year.

The variability in output from wind is far less than the variability in demand, which in any given moment has to be met with the mix of resources available on the grid. This relationship has long pre-dated wind or solar adoption and will continue (at least until we are better able to store electricity at scale).

The below graphic represents the 24-hours of an “average” day in MISO in 2018 for each of the four seasons. The sum-total of all resources in a given hour reflects the electricity demand in that hour.

Credit: Ashtin Massie. Processed using data made available by S&P Global Market Intelligence

In MISO, on an average summer day, coal-fired and gas-fired power plants ramp up and down over the course of the day in fairly substantial ways. The data confirms that coal-fired power plants’ output cycles over the course of the day.

However, a close examination of the data shows it’s not in response to wind.

Coal (and gas) aren’t cycling to follow the output of wind; they are cycling to follow load!

The variance in coal (or gas) output over the course of the day far exceeds that of wind, which wouldn’t be the case if coal was only cycling to back up the variability of wind power. The coal and gas output cycle corresponds to large swings in demand. In the summer, that comes from the air conditioning load to keep homes, office buildings, and industrial facilities cool. In the future, that demand could very easily be met with solar.

In MISO today, coal is no longer a baseload resource and it is this economic reality that is driving many changes in coal plant operations. Demand drives prices, and when coal plants do end up cycling, it is out of response to those price signals. In the future, flexible demand will also be able to respond to price and will work in concert with renewables and storage to meet our reliability needs.

Improper framing, we want to minimize system costs, not individual costs

The data show that at least for grids with wind adoption at the levels MISO is presently seeing, wind integration really isn’t a problem for coal-plant operators. But both wind and solar adoption are on the rise. Nearly 90% of new capacity in MISO is going to be from wind, solar, or storage. It isn’t impossible to imagine a future where wind integration might create challenges someday.

Yet some already assert that integrating renewables has become an “inefficiency.” Such an assertion only appears valid if you look at this issue from the glasses of the plant manager. Not even the lenses of the utility or system operator would produce such a myopic vision.

What the grid operators are supposed to optimize around is the lowest system costs, not the lowest unit costs. And sometimes unit cost minimization doesn’t result in system cost minimization.

There is a fundamental problem with framing wind integration as a “cost” or “inefficiency.” When the owner of a unit won’t shut it down (out of a desire to avoid shut down costs) the system is deprived of lower cost, more efficient resources. What UCS, Sierra Club, and many others have shown: turning down or off coal-fired power plants actually lower costs and increasing efficiency.

For example, an NREL study found that increased wind adoption in the west might increases costs at some individual power plants but that for every $1 in ‘integration costs’ there would be $45-$200 of system benefits.

Arguably the introduction of new resources has allowed coal plant operators to go long periods of time without relying on coal. See this glorious visual from a guest blog by one of our Stanford fellows last summer.

Credit: Emma Spellman, Union of Concerned Scientists

In many parts of the country, coal-plants are really only economic for a few months of the year, and the continued addition of wind, solar, storage, and flexible demand will allow utilities to avoid the costly shutdowns by operating coal plants seasonally.

Need for market rules that are fuel neutral

At the end of the day, we do need an electric grid that is flexible. Demand for electricity is not constant and so we need a grid that is nimble enough to meet those needs. Competitive markets, that often dictate the economics of our grid, might very well need market products for flexibility, but those market rules should be available on an equal basis to all resources and not set up the way capacity markets were set up that resulted in fairly substantial biases.

If a competitive market was developed for a flexibility product, and it was done agnostically, coal isn’t going to be able to offer that service at a lower cost than other resources. Wind, solar, storage, flexible demand, and gas will all beat out coal.

A manufactured problem

Trying to solve a problem before it gets out of hand is admirable but manufacturing a problem out of fear of the unknown, not so much. Simply put: the current rhetoric of renewables scapegoating is unjustified. We’ve been ramping up on renewables for some time now and every year we seem to learn new ways to cost-effectively integrate higher levels of wind and solar. Deployment of clean energy technologies has led to innovation which has led to more deployment and then more innovation. I’m skeptical that the virtuous cycle is suddenly going to stop.

Originally published in Union of Concerned Scientists.