Reading Comprehension

There has, of late, been no shortage of attacks on Gross Domestic Product as a measure of national progress, and the substance of these attacks is by now familiar. The metric counts goods and services exchanged for money and ignores those that are not; it treats environmental degradation as it treats environmental restoration, both as positive contributions if they generate paid activity; it remains silent on how the resulting income is distributed; and it celebrates the broken window whose replacement adds to output while ignoring the unbroken one. The catalogue is real and has been recited often enough that it now feels almost ceremonial.

What is less often considered is the question of what the alternative is supposed to be. The candidates that suggest themselves — measures of subjective wellbeing, composite human-development indices, ecological accounts — each have their own difficulties. Subjective wellbeing is volatile and culturally varied; composite indices require weights whose selection turns out to be a normative argument disguised as arithmetic; ecological accounts struggle to assign monetary value to losses for which money is the wrong unit. None of these difficulties refutes the alternatives, but each suggests that the comfort of a single headline number, of the sort GDP has provided for three quarters of a century, may not survive replacement.

This is not a defence of the status quo so much as a complication of the polemic. The polemic, in its strongest form, treats GDP as a kind of confidence trick that policy-makers should simply abandon. The more sober view treats it as one indicator among several, useful for the limited range of questions to which it gives an interpretable answer — the volume of recorded economic activity, the size of taxable income, the rough scale of an economy relative to others — and dangerous chiefly when asked to bear weight it was never designed to carry. The error is less in the metric than in the use; less in what GDP says than in what is asked of it.

The honest position grants the critics their substantive points while declining to share their cheerful confidence that anything better is yet ready to take its place. A pluralism of measures, each interpretable on its own terms and each presented alongside its own caveats, is probably what the next generation of national statistics will offer. Whether that pluralism actually changes policy, as the critics hope, will depend on something the critique itself does not supply: a political readiness to read several numbers at once where one has long sufficed.