Why Does Using Less Sometimes Lead to Using More?

How Jevons Paradox explains why greater efficiency can increase consumption instead of reducing it.
Imagine a delivery company replaces its trucks with models that use half as much fuel. Deliveries become cheaper, profits improve, and the company expands into markets it previously couldn't afford to serve.
A year later, its fuel consumption has increased.
How can using less fuel per delivery result in burning more fuel overall?
The answer lies in a counterintuitive economic principle: making something more efficient can make it attractive to use much more of it.
This is Jevons Paradox, and understanding it requires examining not just what technology saves, but how people respond to those savings.
How Can Efficiency Increase Consumption?
The mechanism follows four steps:
- Efficiency lowers the effective cost. Less fuel, time, or money is required per activity.
- Previously expensive activities become affordable. More people and businesses can justify using the resource.
- Demand expands. Existing users consume more, while entirely new applications become economically viable.
- Total consumption can rise. Additional activity outweighs the resources saved per use.
Economists call the responsiveness of demand to price changes price elasticity of demand.
When costs fall, demand generally rises. But the magnitude depends on how strongly people respond.
Unmet demand matters too. Customers who previously couldn't afford a service can now participate. Entrepreneurs discover new applications, expanding the market beyond its original boundaries.
What Does Jevons Paradox Look Like in Practice?
Consider a hypothetical delivery fleet consuming 20,000 gallons of fuel annually.
New trucks double fuel efficiency, reducing consumption to 10,000 gallons if delivery volume remains unchanged.
But cheaper deliveries attract new customers. The company expands to 2.4 times its original delivery volume, raising annual fuel consumption to 24,000 gallons.
Despite becoming twice as efficient, it now consumes 20% more fuel than before.

Illustrative example: fuel consumption initially falls with greater efficiency, then exceeds its original level as demand expands.
The outcome depends on how much activity expands:
| Delivery volume | Annual fuel | Outcome |
|---|---|---|
| Original, old trucks | 20,000 gal. | Baseline |
| 1×, efficient trucks | 10,000 gal. | Full potential savings |
| 1.5× | 15,000 gal. | Partial rebound |
| 2× | 20,000 gal. | Full rebound |
| 2.4× | 24,000 gal. | Jevons Paradox |
With twice the fuel efficiency, doubling delivery volume is the break-even point. Any further increase pushes total consumption above its original level.
Economists call the erosion of expected efficiency savings a rebound effect. When additional consumption exceeds the original baseline, the outcome is called backfire, or Jevons Paradox.
Where Did Jevons Paradox Originate?
In 1865, British economist William Stanley Jevons published The Coal Question.
He observed that increasingly efficient steam engines made coal-powered production cheaper, encouraging industrial expansion rather than necessarily conserving coal.
British coal production rose from approximately 98.2 million long tons in 1865 to 225.2 million in 1900, according to the NBER Macrohistory Database, available through the Federal Reserve Bank of St. Louis.
That growth does not establish that efficiency caused rising consumption. Industrialization, expanding markets, and other economic forces also mattered. But it illustrates the historical pattern Jevons sought to explain.
His argument anticipated important ideas in modern research on rebound effects.
Why Is AI a Modern Test of Jevons Paradox?
Artificial intelligence provides a contemporary example of falling costs alongside expanding demand.
Stanford's 2025 AI Index reported that inference costs for GPT-3.5-level performance fell from $20 per million tokens in November 2022 to $0.07 by October 2024—a reduction of more than 280-fold.
Cheaper AI makes previously uneconomical applications viable, from customer support and software development to automated research.
Meanwhile, the International Energy Agency estimated that data centers consumed 415 terawatt-hours of electricity globally in 2024. Its 2025 report projected approximately 945 terawatt-hours by 2030, with AI the largest driver of growth.
These figures demonstrate falling computing costs alongside expanding electricity demand. They do not prove efficiency improvements caused the increase. More powerful models, new infrastructure, and wider adoption also contribute.
The central question is whether efficiency gains will outpace the additional demand they help create.
Does Greater Efficiency Always Increase Consumption?
No. Jevons Paradox describes a possible outcome, not a universal law.
A meta-analysis by Schütt and colleagues, first published in 2024, examined 43 household-level studies containing 1,118 rebound estimates. Its models estimated average total rebound effects of approximately 41%–52%, with substantial variation.
These estimates indicate that efficiency savings are frequently reduced by behavioral and economic responses, but not necessarily eliminated.
Three factors help explain why backfire does not always occur:
- Saturation: People eventually reach limits on how much they want to consume.
- Constraints: Time, infrastructure, and available resources restrict expansion.
- Policies and incentives: Pricing, regulations, and consumption limits can preserve savings.
The underlying logic also appears elsewhere, although these are analogies about expanding demand rather than strict energy rebound effects.
Adding highway capacity can encourage additional driving. Making email nearly free can increase message volume. Improving workplace productivity can increase output expectations rather than reduce working hours.
In each case, reducing the cost or effort of an activity changes behavior.
🧬QuestionStrings: Thinking Beyond Efficiency
Before assuming an improvement will save resources, work through five questions:
- What exactly are we saving?
- Who benefits from the lower cost?
- What will they do with the savings?
- What would make total consumption rise?
- What limit or incentive would preserve the savings?
This sequence shifts attention from the immediate improvement to its second-order consequences.
Efficiency remains valuable. It can reduce costs, expand access, and conserve resources. But improving efficiency and reducing total consumption are different objectives.
The consequences of an improvement depend not only on what it saves, but on the new behavior it makes possible.
The next time a technology promises to save time, money, or energy, ask:
What will people do with the resources they save?
You just thought more carefully than most people do all day.
One question like this, every morning. Free.
Start My Daily Practice →No password needed · Takes 30 seconds
Discover more from QuestionClass
Subscribe to get the latest posts sent to your email.