GDP & National Income Accounting

Simon Kuznets, a Belarusian-American economist at the National Bureau of Economic Research, was commissioned by the US Senate in 1932 to develop a measure of national output. His January 1934 report introduced what would become Gross National Product (later GDP). The framework gave the first quantitative answer to how badly the Depression had hit — US national income had fallen from $87.8B in 1929 to $39.3B in 1932, a 55% collapse. Kuznets himself was deeply ambivalent: the welfare of a nation can scarcely be inferred from a measurement of national income. GDP measures production, not welfare; counts pollution-causing activity and the cleanup of it equally; values a divorce lawyer's billings as economic activity but household care work as zero. The framework was extended during World War II (Keynes's How to Pay for the War, 1940), formalized at Bretton Woods, and codified globally via the UN System of National Accounts.

Three routes lead to the same number, and their agreement is a matter of construction rather than discovery. You can add up what every firm produces, net of what it bought from other firms. You can add up what everyone earns — wages, profits, rents, and taxes on production. Or you can add up what is spent: consumption, investment, government purchases, and exports minus imports. The same transaction is somebody's output, somebody's income, and somebody's expenditure, so the three totals must agree by definition. Where published estimates diverge, the gap measures the quality of the statistics, not a fact about the economy. Two adjustments come before any comparison. Nominal figures mix real growth with price changes, so they are deflated to separate the two. And comparing countries at market exchange rates understates the poorer ones, because haircuts and bus rides cannot be shipped across borders and are therefore cheap wherever wages are low — which is why cross-country work uses purchasing-power parity instead. The deep question is not how to add but what to count, and the production boundary is where the politics live. The accounts exclude household production: the same meal counts when a restaurant sells it and vanishes when someone cooks it at home. Valued at market wages that omission is worth something like a quarter of measured output in rich countries, and it falls almost entirely on work women have traditionally done. The boundary treats the natural world the same way — felling a forest adds to output, the standing forest counts for nothing, and the cleanup afterwards adds again. None of this makes the measure useless; it makes it a measure of production, which is what it was built to be. The trouble begins when a production statistic is treated as a scoreboard for national success, because what gets measured attracts policy — and a country can raise the number while running down everything the number was never designed to see.