Quantitative Input and Output Relations in the Economic System of the United States
Every industry's output is another's input — so map the whole economy as one grid of give-and-take, and solve it.
To bake a loaf you need flour; to make flour you need a mill; to build a mill you need steel — Leontief found the arithmetic that follows every one of those threads at once.
The idea, unpacked
An economy is a web in which almost everything is both a product and an ingredient. Electricity lights the steel mill; steel builds the trucks; trucks deliver the grain; grain feeds the workers who run the power station. Pull on any one thread and the whole web twitches.
Leontief's insight was that you can write this web as a single table — every industry a row and a column — and turn it into arithmetic. Tell the table how much of each final good people actually want, and it computes how much every industry must produce in total, once you count not just the obvious direct orders but all the hidden inputs behind them, and the inputs behind those.
Where it came from
Wassily Leontief was a Russian-born economist who settled in the United States and, at Harvard in the 1930s, set out to do something economists had mostly only talked about: measure the entire interlocking economy. Earlier thinkers — Quesnay in 18th-century France, Walras in the 19th — had pictured the economy as one great system of mutual dependence, but as theory on paper. Leontief made it real, painstakingly assembling the first input–output table of the American economy from scratch.
The calculations were enormous — inverting a big table by hand is punishing — and his project became one of the first serious customers for early computers. The work earned him the 1973 Nobel Prize, and the tables he pioneered are now built by governments all over the world.
Why it mattered
For the first time a planner could ask a precise question and get a number: if the country wants this many houses and this many cars next year, exactly how much steel, electricity, glass, and labour must every industry produce to deliver them? Because the method captures the indirect demands — the steel inside the car, the coal inside the steel — it sees costs a simpler tally misses. That made it indispensable for wartime production, for building national economic statistics, and, today, for measuring things like the carbon footprint hidden inside everything we buy.
A way to picture it
Think of a recipe book in which every recipe's ingredients are themselves dishes with their own recipes. To throw a dinner for ten you don't just cook ten plates — you cook the sauces those plates need, and the stocks those sauces need, down and down. Leontief's inverse is the master shopping list that adds up every ingredient at every level, so you know the true total you must produce. The tool below lets you place the order and watch the hidden layers pile up.
Where it sits
Leontief turned Walras's abstract general equilibrium (also in this Library) into something you could compute with real numbers, descending from Quesnay's centuries-old dream of charting the economy's circulation. His table is a cousin of other ways of finding the steady state of a system that feeds back on itself — the same spirit as Markov chains, or the link-counting behind modern web search.