Elon Musk is wrong about GDP
Of all the dubious claims uttered recently by Elon Musk, I have yet to see a more interesting one than his tweet asserting that “a more accurate measure of GDP would exclude government spending. Otherwise, you can scale GDP artificially high by spending money on things that don’t make people’s lives better.”
Whether or not he is serious, the idea is worth a closer look, because it explicitly expresses one instructive misunderstanding and heavily implies a second. The explicit misunderstanding is that an accurate measure of GDP — gross domestic product — would include all and only the things that make people’s lives better. It wouldn’t. The implicit one is that GDP needs to be accurate because governments try to maximise it. They don’t. (Look around.)
I was surprised to learn from Diane Coyle’s new book, The Measure of Progress, that when GDP measures were first being hammered out, the Musk view nearly triumphed. Government spending was included in GDP only after a vigorous debate. Part of the reason for its inclusion was crude politics: the second world war was raging and governments didn’t want their military spending to be ignored.
The more solid justification for including government spending in GDP was theoretical: the economist John Maynard Keynes had formulated a theory of macroeconomics based on aggregate demand, including government demand. A measure of GDP that pretended government did not exist would be useless for trying to understand and stabilise the economy.
GDP is the value of all goods and services produced in an economy. That definition raises an immediate question: value according to whom? The answer is — sort of — value according to the market. If you grow wheat, its value is defined as whatever you sold it for, less whatever it cost you to grow. But what if the good or service is not sold in a market transaction? If it is still valuable, then, in principle, it should still be included in GDP. In practice, it may not be.
For example, if you pay someone to cook, clean and look after your children, these market transactions are included in GDP. If you do your own laundry and look after your own children, the same activity is overlooked. (A famous critique of GDP, “When a man marries his housekeeper, GDP falls”, somehow manages to package a feminist insight as chauvinism.)
Another example: education could be bought as a product, or it could be provided free at the taxpayer’s expense. How should we value a purely state-funded education system? We could estimate its market value. Or assume that the value of teaching is whatever the teachers happen to be paid. Or we could take Musk’s suggestion and declare that if the schools are paid for by the taxpayer, it would be “more accurate” to assume that they have no value at all. Simple, but silly.
Musk is not the only person to have said silly things about GDP. Senator Robert Kennedy gave a famous speech in 1968 in which he lamented that GNP (a close cousin of GDP) “counts air pollution and cigarette advertising . . . special locks for our doors and jails for the people who break them.” Yet, he added, it does not include “the health of children . . . the beauty of our poetry or the strength of our marriages, the intelligence of our public debate or the integrity of our public officials.”
US GDP has vastly increased since 1968, whether or not the same is true for the intelligence of America’s public debate and the integrity of its public officials. But to see why the speech was silly, imagine pulling together a brains trust and asking them, “how can we help our children to be healthier, our marriages to be stronger, and our poetry to be more beautiful?” I can think of many possible answers to these questions. None of them is “redefine GDP”.
GDP measures the productive output of the economy, and sometimes it does not measure it well. Coyle’s book points out that alongside the long-standing difficulty in capturing the value of unpaid work and of government services, our statistical infrastructure also struggles to measure financial services, quality improvements and services that are bundled together with physical goods. There is plenty of room here for improvement, but the goal of that improvement should be to do a better job of measuring economic output.
What, then, about the planet? What about life, liberty and the pursuit of happiness? What about all that beautiful poetry? If these things matter then they need to be supported by good policies, informed by good statistics. But when you start to think seriously about what those statistics might be, you realise that they have very little to do with GDP. If we actually wanted to measure things that “make people’s lives better”, it’s pointless to try to modify GDP. That’s just a category error, like trying to get a measure of your life expectancy by starting with the market value of your car and making a few adjustments from there.
To measure the quality of people’s lives, ask them. Curious readers can take a look at the recently published World Happiness Report for one example of such an effort. Similarly, saving the planet does not require the redefinition of some grand goal. It requires a myriad of well-crafted policies, each informed by carefully collected data.
The real problem with GDP is not that it includes government spending, or omits household chores, or that it fails to measure the quality of poetry. It is that it tempts us to view all human progress as an amorphous score, which we can start maximising just as soon as we fix whatever we think is wrong with how that score is calculated. That’s a mirage. Musk is said to love winning computer games. Human flourishing is not a computer game.
Written for and first published in the Financial Times on 28 March 2025.
Loyal readers might enjoy the book that started it all, The Undercover Economist.
I’ve set up a storefront on Bookshop in the United States and the United Kingdom. Links to Bookshop and Amazon may generate referral fees.
