Why energy prices alone don't build enough power plants.
Electricity markets have a structural problem known as the missing money problem. In competitive markets, price normally signals scarcity — when there's not enough supply, prices rise, attracting new investment. But electricity markets cap real-time prices at regulatory limits (typically $1,000–$3,500/MWh, depending on the RTO). During the small number of hours each year when the system is truly stressed, prices can't rise high enough to cover the full cost of a peaker plant that only runs 50–100 hours annually.
The math is brutal. A gas combustion turbine costs roughly $800,000/MW to build. To recover that investment over 20 years at a 9% discount rate, you need about $100,000/MW/year in revenue. If the plant runs 100 hours and earns $300/MWh (well above average real-time prices), it earns $30,000/MW — only 30% of what it needs. Where does the other 70% come from?
In theory, it comes from extreme price spikes during scarcity hours. In practice, price caps prevent this from happening. The "missing money" — the revenue that a competitive market cannot deliver — means that, left to their own devices, energy-only markets tend to underinvest in peaking capacity.
Most RTOs have responded to the missing money problem with capacity markets — separate markets where generators are paid not for energy, but for the commitment to be available during peak periods. Load-serving entities are required to procure a certain amount of capacity — typically 15–17% above their peak demand forecast — one to three years in advance. Generators bid their capacity into an auction, and the market clears at the price needed to attract enough capacity to meet the requirement.
PJM's capacity market — the Reliability Pricing Model — is the largest in the world, clearing billions of dollars per year. MISO, ISO-NE, and NYISO also operate capacity markets. CAISO and ERCOT do not (though CAISO has a resource adequacy program that functions similarly in some respects).
Capacity markets are among the most controversial features of organized electricity markets. Critics argue that:
Defenders argue that without capacity markets, price caps would inevitably produce chronic under-investment in reliability resources — the "missing money" would translate to missing megawatts at exactly the wrong moment.
The debate continues. Some states have experimented with alternative mechanisms. ERCOT has tried higher energy price caps (the "energy-only" model). The jury remains out on which approach produces better long-run outcomes.
PJM, May 2007 — the first Reliability Pricing Model auction. PJM's Reliability Pricing Model launched as a direct response to a decade of debate about whether energy-only market revenues were sufficient to attract investment in new generation. The first Base Residual Auction cleared capacity three years forward, with plants bidding to guarantee their availability during peak demand periods in exchange for a predictable annual payment. The auction made the "missing money" problem concrete and visible: some marginal generators that had been economically uncertain suddenly had a revenue stream; others that had counted on clearing the auction did not. Within a few years, RPM was clearing billions of dollars annually — more capacity revenue than any other market mechanism in the world. But the debate the auction was meant to settle instead intensified: generators argued clearing prices were still too low to finance new investment; consumer advocates argued the market was transferring billions to resources that would have run anyway.
PJM — Reliability Pricing Model Overview Monitoring Analytics — PJM State of the Market Reports