The Grid's Marketplace Isn't Working. Should We Fix It or Start Over?
-2%20(8).png)
Regional transmission organizations run the power markets serving most Americans. Their rules are now contributing to higher prices and discouraging investment.
Most people have never heard of a regional transmission organization. Yet RTOs coordinate more than two-thirds of the electricity Americans use, and decisions made inside them are now showing up on monthly power bills. In Regional Transmission Organizations: Problem or Solution?, NCEA Senior Fellows Jonathan Lesser and Brent Bennett explain how these institutions came to be, why their markets are struggling, and what a serious reform agenda could look like.
The authors don’t dismiss what RTOs have achieved. Competition among generators has improved efficiency and lowered costs for consumers. But wholesale power markets now suffer from structural, regulatory, and market failures, and repeated attempts to fix them have brought more uncertainty, higher wholesale prices, inadequate transmission, and reduced reliability.
From Power Pools to Policy Engines
RTOs grew out of something simple. Nearly a century ago, utilities began coordinating their power plants to reduce the risk that one company’s outage would leave customers in the dark. PJM, now the country’s largest RTO, started that way in 1927 and today covers all or part of 13 states and the District of Columbia.
Federal regulators later handed these organizations much larger and more complex jobs: oversee competitive wholesale markets, plan new high-voltage lines, and connect new generators. For years, that work drew little attention. Three things changed that. Electricity demand began climbing again with the growth of AI and data centers. Federal and state policies forced many fossil-fuel plants into retirement while subsidizing wind and solar. And prices started to rise.
Energy vs. Capacity, in Plain English
To see where the trouble lies, it helps to know that RTOs can run two very different kinds of markets. An energy market pays for electricity actually delivered, measured in megawatt-hours. A capacity market pays plants to be available to produce power in the future, measured in megawatts.
Think of it as the difference between paying for a taxi ride and paying a cab company to keep cars waiting outside your door in case you need one. Capacity markets exist to make sure enough “iron in the ground” is ready for the moments of highest demand.
The Missing-Money Problem
Why would a market need to pay plants for standing by? Because RTOs cap wholesale energy prices, in part because severe price spikes are politically untenable. Caps reduce the incentive to manipulate prices, but they also stop prices from rising enough during shortages to support plants that run only a few times a year. Economists call the shortfall “missing money.”
Capacity markets were designed to supply it. The problem is that they are artificial in a way energy markets are not. The RTO decides how much capacity is needed, draws the demand curve, sets the reference price, and typically caps the auction at 150% of that reference. Each of those choices is a regulatory judgment, not a market outcome. And because utilities must buy a fixed amount of capacity no matter the price, sellers have significant leverage. PJM’s own independent market monitor has concluded that structural market power is endemic to the capacity market.
A More Than Elevenfold Increase
The results are now impossible to miss. In PJM, the capacity price was about $29 per megawatt-day for the 2024–25 delivery year. It jumped to about $325 for 2026–27 and just over $333 for 2027–28. Retirements of conventional plants have outpaced the capacity credited to new wind and solar, and peak demand keeps growing. The estimated cost to retail customers is about $16 billion a year.
Subsidies compound the problem. Tax credits and state mandates for wind and solar push wholesale prices artificially low or even negative in some hours, squeezing nuclear and coal plants that can’t ramp up and down quickly. Then, when the wind drops or the sun sets, the same markets face more frequent scarcity and sharper, less predictable price spikes. Meanwhile, the cost of new transmission is often spread across ratepayers, indirectly subsidizing poorly sited wind and solar instead of prioritizing reliable generation near new demand.
Why RTOs Can’t Simply Fix It
RTOs have a surprisingly limited tool kit. An RTO that sees a coming shortfall can’t order anyone to build a power plant. Most can’t force a transmission line to be built without approval from federal regulators and the affected states. Every market redesign can be challenged, sometimes for years. And each round of rule changes, however well intended, adds uncertainty that makes investors less willing to commit capital.
The Reform Menu
The brief lays out a range of options, from incremental to sweeping:
Return to power pools. RTOs would operate the grid and stop running formal markets, leaving participants free to trade electricity on their own terms.
Lift energy price caps. Let scarcity prices do their job, paired with severe penalties for manipulation, eliminating the missing-money problem that capacity markets were built to solve.
Stop subsidies from distorting markets. Require subsidized wind and solar to meet reliability criteria, pay a reliability fee, add storage, or trade bilaterally outside RTO markets.
Make generators pay for the transmission they cause. Alberta charges new generators up-front, location-based fees: far-flung projects pay more than those near demand, and intermittent ones pay more than plants whose output can be scheduled.
Encourage private grids for large loads. Big new customers could build their own supply and connect to the wider system later.
PJM has floated its own alternatives, but the authors find them too narrow. None of the fixes will be easy, and policymakers will have to confront entrenched interests to make them. Yet the current system is failing at its two core jobs: keeping power reliable and keeping it affordable.
-2%20(14).png)
-2%20(10).png)
-2%20(9).png)
-2%20(7).png)
.webp)
-3%20(1).png)
-2%20(5).png)
-2%20(1).png)
-6%20(8).png)
-6%20(9).png)
-2%20(11).png)
-2%20(12).png)
-2%20(13).png)
-2%20(3).png)
.avif)
