Will 2018 Be the Year of Solar or Coal?

Happy New Year!

At Forbes Brian Potts asks a pithy question: will 2018 be the year of solar or coal? He starts with an observation familiar to most who have a passing familiarity with energy innovation in the past decade: the decline of coal as an electricity generation fuel source is a consequence of the dramatically increased supply (and hence lower price) of natural gas, with its attendant approximately 50% reduction in greenhouse gases as an added benefit. Solar is not yet economical for energy generation on a levelized-cost basis (meaning if you factor in the fixed and variable costs and capacity utilization for the life of the asset and compare them) in many regions of the US.

Coal’s losing market share in the power industry, notwithstanding these eased regulatory burdens, because natural gas prices have stayed low and utility-scale solar costs have declined to the point where solar is, in many places, the cheapest way for a utility to add new capacity.

This latter development has been a game changer. What seemed impossible just five years ago is starting to happen now with some regularity: utilities are retiring their coal plants and announcing plans to replace them with large solar arrays.

In most parts of the country, solar still isn’t even close to being the cheapest way to provide energy to homes and businesses. That honor goes to wind turbines and large combined-cycle natural gas power plants. These are the cheapest sources of energy right now, and that isn’t likely to change in 2018.

But Potts delves in to some really important and nuanced aspects of generation technology choice that can get overlooked. One aspect is the availability of generation capacity for reliability purposes. Utilities have regulatory reliability requirements; if they are in competitive/restructured states that means they have to contract for capacity commitments consistent with reliability, and if they are in traditional vertically integrated states they have to have sufficient generation capacity consistent with reliability. In many states as coal plants age and depreciate that capacity is being replaced by utility-scale solar.

There are three regulatory policy dimensions of this technology choice: two federal and one state (but common to all states). At the state level, distribution utilities face service quality regulations, and one form such regulation takes is reliability regulation, where utilities must meet reliability standards. Reliability measures include SAIDI (system average interruption duration index, or how long outages are), SAIFI (system average interruption frequency index, or how frequently outages happen), and CAIDI (=SAIDI/SAIFI, or how long the average outage is). It’s been a remarkably fast evolution of utility-scale solar technology to get to a point where solar is the go-to technology for building new capacity for reliability in peak hours — this clearly makes the most sense in states with high insolation, like California, Nevada, and Arizona.

Another policy dimension here is the availability of federal tax credits for construction of new solar capacity. At the margin these credits reduce the relative cost of solar, even relative to gas plants that are cheaper than ever to run but for which the fuel cost is high compared to other technologies. These technology-targeted subsidies are justifiably controversial — would this technology have matured this quickly without them? How much more slowly would it have been commercialized without them? Those are important counterfactuals, and I’d like to see an analysis that estimates them.

Finally, in October the Department of Energy proposed a regulatory change that would increase the value of existing coal-fired power plants for reliability purposes, and has asked the Federal Energy Regulatory Commission to implement such a proposal. FERC has issued this proposal for public comment (known as “the NOPR”, for Notice of Proposed Rulemaking), and has received a large number of comments, most of them in opposition to such a proposal. If you want to see more about how the regulatory sausage is made, this EE News article summarizes this procedure and the general nature of the comments FERC has received. FERC is scheduled to issue its decision on January 10.

Potts’ article requires getting in to the regulatory weeds, and the question he asks is an important and complicated one without simple yes/no answers.