Segment 1 — The Cold Open

Segment 2 — The Good Intention

William Stanley Jevons, a Manchester economist, watched this pattern unfold and tried to explain it in his 1865 book The Coal Question. He started from the same practical observation that had driven Watt: Britain’s factories, mines, and transport systems needed cheaper power if they were to keep expanding. Watt himself had set out to reduce the ruinous fuel cost of the Newcomen engine, which often consumed more than twenty-five pounds of coal per horsepower-hour. The separate condenser achieved that saving without sacrificing work output. At the time, every engineer and mine manager treated lower fuel use per engine as an unmixed benefit; it let marginal pits stay open and let mill owners run more machinery for the same outlay. Jevons simply asked what would happen once that cheaper power reached every corner of the economy, and he examined the sequence of decisions that followed each cost reduction rather than assuming the total quantity of coal would remain fixed.

Segment 3 — The Implementation

Watt and his partner Matthew Boulton began licensing the improved engine in the 1770s, first to Cornish tin mines and then to Lancashire cotton mills. By 1800 more than five hundred Boulton-Watt engines were at work. Early users reported fuel savings of sixty to seventy-five percent compared with the machines they replaced. Parliament extended Watt’s patent until 1800, and after it expired other builders produced still cheaper, simpler versions. Railways and steamships, which had barely existed in 1800, multiplied after 1830; each new line and each new vessel added fresh demand for coal that had never been burned before. Jevons collected the production figures in his book and showed that total British coal output had risen from roughly ten million tons in 1800 to more than ninety million tons by the mid-1860s. The rollout therefore did not replace one fixed stock of engines with more efficient ones; it multiplied the number of engines and extended their use into sectors that previously relied on water power, horses, or human labor.

Segment 4 — The Unintended Consequences

The efficiency gain lowered the cost of power so far that entirely new uses became profitable. A mill owner who once hesitated to install a steam engine because of coal bills now found it paid to run machines day and night, because the marginal cost of each additional hour fell below the value of the extra cloth or yarn produced. Mines that had been abandoned because drainage was too expensive reopened once Watt engines made pumping cheaper; the same engine that saved coal on an existing shaft also made it possible to sink new shafts in deeper seams that had previously been unreachable. The same logic scaled to railways: lower fuel cost per ton-mile let companies lay track into districts that could never have justified horse-drawn wagons or canals, and each new station then generated traffic in coal, iron, and manufactured goods that required still more power to produce and move. Jevons noted that the very industries that adopted the efficient engines grew fastest, so the saving on each individual engine was swamped by the increase in the number of engines. The causal chain ran from lower cost per unit of work, to wider adoption, to new industries built around cheap power, to higher total tonnage mined and burned. Contemporary observers who expected coal use to fall as engines improved were therefore surprised by the opposite result, because they had counted only the reduction inside existing operations and had not traced how the same cost change altered the boundary between what was worth doing and what was not.

Segment 5 — The Aftermath

Jevons’ book made him briefly famous; he was invited to give evidence before parliamentary committees on coal supplies. The immediate policy response was modest—some discussion of export duties and of encouraging economy in domestic fires—but the larger pattern continued. Later economists labeled the mechanism the Jevons paradox and tested it in lighting, aviation, and computing. In each case the same sequence appeared: an efficiency improvement reduced the resource needed for one task, lowered its price, and expanded total use. No single reversal of the original engines occurred; instead the insight entered economic textbooks as a standing caution about assuming that technical savings automatically reduce resource consumption. The pattern persisted because the underlying arithmetic—lower cost per unit of service leading to more units of service demanded—remained unchanged across technologies.

Segment 6 — The Lesson

When James Watt and Matthew Boulton measured success by fuel burned per horsepower-hour, they rewarded designs that made each engine cheaper to run and therefore worth installing in more places. That rule, applied across Britain’s growing industrial base, produced the measured rise in total coal tonnage that Jevons documented in 1865. Anyone evaluating an efficiency proposal can therefore ask which activities will become newly worthwhile once the cost per unit falls, and whether those new activities will be counted in the same ledger that records the original saving. The same question now faces data-center operators who expect lower energy per computation to reduce overall electricity demand, because the arithmetic that turned a fuel-saving engine into a coal-increasing system has not disappeared with the fuel itself.

💬 Reply to this email — Patrick reads every one.

Share: X · LinkedIn · WhatsApp

Forwarded this email? Subscribe here — it's free.