#121 1990 · US Environmental Protection Agency · Environmental regulation / energy
Instead of telling 3,200 power plants how to cut pollution, Washington capped the total and let plants sell each other the right to pollute
the problem
Forcing every regulated party to hit the same fixed target wastes money on the parties for whom compliance is expensive, while parties who could cut cheaply have no incentive to cut more than required
background
Sulfur dioxide from coal-fired power plants was the primary driver of acid rain damaging forests, lakes and buildings across the eastern US and Canada by the 1980s, and the standard regulatory tool for pollution was command-and-control: mandate a specific technology (a scrubber) or a specific per-plant emissions limit at every regulated facility, regardless of how cheap or expensive that particular plant found it to comply. That approach forces expensive fixes on plants where cutting emissions is genuinely hard, while plants that could cut emissions cheaply have no incentive to do more than the mandated minimum.
The 1990 Clean Air Act Amendments needed to cut national SO2 emissions by 10 million tons below 1980 levels across roughly 3,200 coal-fired generating units with wildly different costs of abatement — some could switch to lower-sulfur coal cheaply, others would need expensive scrubber retrofits — and Congress opted for a fundamentally different mechanism than any prior US pollution law had used at this scale.
what everyone would do
The standard regulatory answer was command-and-control: mandate the same technology (a scrubber) or the same per-plant emissions limit at every facility, regardless of what compliance actually costs there. That forces expensive fixes on plants where cutting sulfur is genuinely hard while giving plants that could cut cheaply no reason to do more than the mandated minimum — the aggregate cut gets bought at whatever the most expensive plant's cost happens to be, applied uniformly to all 3,200 units.
what they saw
Policymakers saw that what actually mattered for acid rain wasn't which specific plant cut which specific ton — it was the national total. Once the goal is reframed as a fixed aggregate rather than a uniform per-unit requirement, the question of WHERE the cuts happen becomes a pure cost-minimization problem, and a market of tradable permits can solve that allocation problem automatically, without EPA ever having to know each plant's abatement cost itself.
the move
Title IV of the 1990 Clean Air Act Amendments created the Acid Rain Program: EPA set a hard national cap on total SO2 emissions and issued each generating unit a fixed number of tradable allowances, each permitting one ton of SO2, based on historic output. A plant that cut emissions below its allocation could sell its surplus allowances to a plant for which cutting was more expensive, or bank them for later — letting the market find the cheapest combination of cuts across all 3,200 units to hit the same national total, rather than mandating an identical action from every plant.
why it works
Issuing each unit a fixed number of tradable allowances and letting them buy and sell freely means a plant that can cut emissions cheaply has an incentive to cut more than required and sell its surplus, while a plant facing expensive abatement can buy allowances instead of installing costly equipment — so the market routes the actual cutting toward whichever units can do it cheapest, without any regulator having to know or set each plant's individual cost curve. Because the national cap is fixed regardless of who does the cutting, the same total reduction is achieved at the aggregate cost of the cheapest combination of cuts rather than the cost of the most expensive plant multiplied across everyone, which is the mechanical reason economic analyses found the program achieved its target for a fraction of a uniform mandate's cost.
the payoff
Power-plant SO2 emissions fell 94% from 1990 to 2019, and EPA's original 1990 cost estimate of $6.1 billion for the program proved far higher than what allowance trading actually delivered — independent economic analyses found the trading approach saved on the order of 15% to as much as 90% compared to a command-and-control alternative achieving the same emissions target, while EPA has estimated the program's annual health benefits, from reduced acid rain and airborne particulates, at well over $50 billion by 2010.
where it breaks
The mechanism depends on a hard, credible cap — allowances only ration scarcity if the total is genuinely fixed and enforced; a cap set too loosely, or one riddled with exemptions, lets total pollution rise even as trading looks active. It also needs a large enough number of participants with genuinely different abatement costs for gains from trade to exist at all — a market with too few participants, or one where every unit faces similar costs, has little for trading to improve on. And it requires emissions to be reliably measurable and attributable per unit (continuous monitoring, in this program's case); a pollutant that's hard to meter accurately at the source can be gamed by underreporting, which breaks the entire allowance-accounting system the trades depend on.
what came after
The Acid Rain Program is credited as the first large-scale pollutant cap-and-trade system in the world and became the direct template for later carbon cap-and-trade systems, including the EU Emissions Trading System and California's cap-and-trade program, cited across environmental economics literature as the proof that market-based emissions trading can achieve a regulatory target at dramatically lower cost than mandating uniform technology or limits across every regulated party.
references
- [1]Acid Rain ProgramUS Environmental Protection Agency, 2024epa.gov
- [2]Cap and Trade Curbed Acid Rain: 7 Reasons Why It Can Do The Same For Climate ChangeForbes, 2012forbes.com