What Firm Means Now
Capacity value against nameplate, curtailment, and the difference between energy and the ability to deliver it at 6pm in January.
Words the industry uses are marked like this at the sentence that defines them, so you can tell a term you will hear on a desk from a phrase this book happens to be using.
Firm used to be a contractual word. Firm power was power the seller was obliged to deliver, with penalties for failure, as against interruptible power sold at a discount to buyers who accepted being cut off.
That word, firm, has become one of the most contested terms in power, because much of the recent capacity from renewables and batteries is not as reliable around the clock as legacy thermal capacity.
Nameplate stopped meaning anything
Chapter 16 introduced effective load carrying capability because the industry needed a number that nameplate capacity no longer supplied. A 300 megawatt gas turbine contributes close to 300 megawatts of reliability. A 300 megawatt solar farm contributes nothing on a January evening. A four-hour battery contributes fully to a four-hour shortfall and very little to a longer one.
Ranges across US regions. Solar spans single digits to over 80% depending on when a region actually has its shortfalls, and the values move as the portfolio changes, since adding storage raises solar's contribution and adding more of any resource lowers its own.
So the same unit of measurement, the megawatt, now describes quantities that are not interchangeable, and every capacity market, planning study and reliability standard has had to build machinery to translate between them.
A generation ago, counting megawatts told you whether the system would hold. It no longer does, and no single replacement number has emerged.
The three questions, restated
Part Three set out the three choices every system makes, and after Part Four they can be restated as what they actually decide.
Who dispatches decides whose judgement is trusted, and whether a bad investment is paid for by shareholders or by captive customers.
How location is priced decides whether the cost of the network is visible to the people whose decisions create it, or arrives later as a charge nobody can respond to.
How capacity is paid for decides where an unavoidable administrative judgement about the value of reliability is made, and how visibly it fails.
Every system in this book is a combination of those three answers, chosen at a particular moment for reasons that made sense to the people writing the rules.
What is actually changing
Three things run through every chapter.
The first is that the fleet's costs moved from the fuel to the capital. A system of nat gas and coal plants had marginal costs that varied with fuel, which is exactly what a marginal-cost auction prices well. A system of wind, solar and nuclear has almost no marginal cost and enormous fixed cost, which a marginal-cost auction prices badly. Europe and China both reached for contracts for difference within a year of each other because the arithmetic gave them no alternative.
The second is that services which used to arrive free with the energy now have to be bought. Inertia, voltage support and fault current came bundled with spinning machines nobody had to procure separately. Ireland, Britain and now Spain buy them explicitly, and Chapter 13 argued that the boundary between what a system buys and what it merely requires is unsettled and moves by event rather than argument.
The third is that demand started growing again in developed countries after fifteen flat years of outsourcing growth to Asia, so Western institutions built to allocate a fixed quantity are being asked to expand one.
Back to the first chapter
Everything in this book descends from the constraint in Chapter 1. Supply and demand are the same quantity on an electrical system, continuously, because the product cannot be stored easily.
Prices go negative for a thousand hours a year because surplus cannot be stored. Prices reach thousands of dollars because shortage cannot be buffered. The same megawatt-hour is worth different amounts a few miles apart because it cannot be moved freely. Regulators argue about paying for plant that runs forty hours a year because nobody can put those hours in a tank. Nodal against zonal, energy-only against capacity markets, obligation against procurement: every one of those arguments is a different proposal for the same problem, which is arranging in advance, by contract, for exactly the right quantity of an unstorable product to exist at exactly the right instant in exactly the right place.
Firm, in the end, means somebody has taken responsibility for that arrangement and can be held to it. The reason the word keeps being renegotiated is that the machines which used to make the promise easy are being replaced by machines that make it hard, and contracts have not caught up.
A reader who finishes here with the three questions in mind can walk into any electricity market in the world, ask who dispatches, how location is priced and how capacity is paid for, and understand most of what its prices are doing before looking at a single number.