Skip to content
Heterodata An Arcanum Research project Leontief
Leontief

Understand the U.S. Economy
as a Web of Industries

Free, one-click-downloadable BEA input-output tables for 71 sectors across 28 years — plus an interactive explorer, progressive tutorials, and fully reproducible example studies.

28 Years of data
(1997–2024)
71 BEA sectors
(Summary level)
10 Reproducible
studies
10 Tutorials

What is an Input-Output Table?

An input-output (I-O) table is a grid that records how much each industry in an economy buys from every other industry — and how much goes to final demand (households, government, exports). Developed by Nobel laureate Leontief, I-O analysis answers questions like: if car sales rise by $1, how much more steel, glass, and electricity does the economy need to make?

The table below is a tiny three-sector economy. Read a row to see where a sector's output is sold; read a column to see what a sector buys to make its output.

A simple economy (all figures in $ millions)
sells ↓  /  buys → Farms Food factory Households
(final demand)		 Total output
Farms 0 60 40 100
Food factory 0 0 120 120
Value added
(wages + profit)		 100 60 — —
Total input 100 120 — —
The food factory (column) buys $60m of crops from farms and adds $60m of its own labor and profit to make $120m of food, all sold to households. Trace one more step and you can see how $1 of extra food demand pulls crops, then the inputs to those crops, and so on — that ripple is exactly what input-output analysis measures, across all 71 real U.S. sectors.

The U.S. Bureau of Economic Analysis (BEA) publishes annual I-O accounts at the 71-sector summary level. This site packages every table from 1997 to 2024 into machine-readable formats and derived matrices — all freely downloadable and reproducible.

Start the tutorials → Browse the studies →

Learn

A progressive tutorial track covering what an I-O table is, how to read it, and how to compute multipliers, linkages, structural decomposition, and more — with KaTeX math and embedded examples.

10 tutorials

Tables

Interactive I-O Table Explorer: pick a year (1997–2024) and matrix (Use, Supply, Leontief inverse, …), toggle 71↔15-sector aggregation, search sectors, and download in CSV, XLSX, or Parquet.

28 years × 7 matrices

Studies

Fully reproducible example analyses — key sectors, multipliers, supply-chain networks, structural decomposition, and more. Every chart links to the exact Python code that produced it.

10 studies

Data

Machine-readable catalog of every table and derived series, with provenance notes and bulk downloads for every year or the full dataset.

BEA U.S. only