Reading Lorenzo Fioramonti’s book The World After GDP recently, I was struck again at how simplistic Gross Domestic Product is. It remains our primary measure of national success because it’s now deeply embedded in our institutions, but it’s blind to what really matters. We urgently need our politicians to consider a wider set of metrics, and adopt a more nuanced view of progress. Here are some of the shortcomings of GDP, and you may well have your own examples:
- If you buy a meal at a restaurant, it counts towards GDP. The exact same meal cooked at home does not. All housework or domestic work is thus invisible to GDP, even though it accounts for over half of the work done in Britain. Since women do more housework than men, GDP has traditionally valued men’s work and ignored women’s contributions.
- Parents raising their own small children has no value to GDP, but dropping them off with a professional child-minder does. Caring for an elderly relative would be the same – growth if a professional carer does it, no growth if a family member helps out.
- There’s no way to measure averted problems. If I pull out of a junction carelessly and hit another car, GDP will be boosted by insurance and repair costs, and legal and healthcare expenses. If I see the other car and brake, there’s no benefit to GDP. From an economics point of view, it would have been better to have had the accident. At the macro level, this means that GDP can’t properly value preventative medicine, health and safety, community resilience, and a whole host of things we’d consider positive.
- All waste is good for GDP. If we stopped throwing away a third of the food we buy, that would be bad for growth. Insulating our homes would also be bad for growth.
- Leisure has little value in GDP terms, unless we’re doing a paid activity. Sleep is of no value whatsoever.
- Speaking of nocturnal activities, changes to the way we calculate GDP were introduced in 2014 to include illegal and black market transactions. That means that GDP only values sex if someone pays for it, even though that’s illegal. Sex in any other context is worthless.
- Natural assets aren’t counted, so a tree is worth nothing. Cut it down and turn it into wood, and it now registers on the GDP radar. Since the loss of the tree is not subtracted, chopping it down is recorded as pure gain. That logic applies in the world’s endangered rainforests, and is one of the main reasons why they are endangered.
- There’s no way to calculate the loss of something that is free. Pollution, aquifer depletion, or soil erosion aren’t subtracted from GDP.
- Conversely, the increase of natural resources is a form of growth that GDP doesn’t capture. Growing or recovering fish stocks, for example, are not accounted for. (Grow those same fish in a fish farm though, and they would count towards GDP.)
- Many vital services are invisible to GDP. Fioramonti highlights rainfall – it’s critical to agricultural production, but it’s a resource that doesn’t appear in any accounts of national wealth. A safe and stable global climate is perhaps the biggest oversight.
- GDP has no truck with inequality, as it’s just a raw measure of activity. All growth is positive, even if it all goes to a handful of people and society is falling apart all around them.
- All education is a problem for GDP, because it can count teaching, but not knowledge. The cost of providing education can be counted, but not education itself.
- Technological progress is a tricky one too. Someone buying solar panels today would be paying less for them and getting a better product than someone buying them five years ago. But since GDP can only count the actual price, the older and less efficient panels did more for GDP.
- On the subject of solar, renewable energy is a challenge to GDP. We can count coal or gas, but the wind and the sun are free. Beyond the initial set-up and ongoing maintenance, renewable energy has no fuel costs and makes a much lower contribution to GDP. On purely GDP terms, fossil fuels are much better.
- The internet is also leaving GDP behind. Buying a map or an atlas is positive for growth, and looking something up on Google Maps isn’t. The value of free services, including email, Wikipedia or social media, are impossible to quantify in GDP terms.
Some of these have been shortcomings from the start. Others are emerging problems. Together I think they signal the beginning of the end for GDP as the one metric to rule them all. There’s no other single metric to take it’s place, I should add. Instead, we’re going to need to draw on a wider set of measurements, balancing gains and losses, paid and unpaid work and services, and scoring non-monetary benefits.
I absolutely agree.
Also, war is good for GDP. Bomb everything to pieces then work like mad to rebuild; marvellous for GDP, not so good for the dead, mained and traumatised.
GDP was created, I think, for a specific purpose – to measure the delivery of the Marshall Plan at the end of WW11. Most people, including economists and politicians, forget it is a recent invention, and for a specific purpose too. Most 1940s technology has been replaced, so we need to do the same for our metrics.
Yes. Some of the highest GDP figures ever recorded have been in the context of post-war reconstruction.
The Marshall Plan made GDP international, and really embedded it in the world’s institutions. It originated a few years earlier, as a way of measuring whether stimulus was working during the great depression.
I do think you rather over egg GDP. Many other statistics are important for government: Inflation and unemployment are just two. And almost any basket of statistics you seek to replace it with will be either as poor in their own ways and/or much more delayed making them far less useful in decision making (what we were doing 3 months ago is sort of helpful, what we were doing 2 years ago is less so).
No statistic is going to be correct, so we should treat them all with care. Personally I think we should learn form Sir John Cowperthwaite’s policy of positive non-intervention. When he ran the Hong Kong civil service he refuse to collect statistics “for fear that I might be forced to do something about them”. That’s the problem with statistics, people think they can use them to interfere and they can never be smart enough of have enough good information to do better than the free market.
I agree with the second part of your comment, that statistics shape our actions and should always be approached with care. But I don’t think I’m the one with a problem when it comes to GDP.
“Growth is the number one priority of this government” as Gordon Brown put it, and Cameron and May have both said similar things. Our politicians have a narrow-minded focus on GDP figures, over-emphasize growth, and write it into policy in unhelpful ways – see Cameron’s insistence that every single department prioritise economic growth as a key objective.
I also take issue with the idea that a basket of statistics would necessarily be worse. It’s not a matter of replacing GDP, for one thing – I’d still be quite happy for us to keep measuring it, and GDP can be part of the overall picture. The important thing is to reduce the pressure on that single artificial metric, and recognise the trade-offs. Growth of what, and for whom, and at what price?
The thing with GDP is that while its wrong (as are all official statistics), and doesn’t completely measure what it sets out to it has survived because it is quite a good correlation to the things that voters, and by extension politicians, care about: Rising living standards, improved life expectancy, better public services, happier lives.
You might replace it with a confusing and hard to understand (for the average voter) basket of statistics that at best ends up tracking what GDP was telling us, just later. If we have 5 official metrics the government is trying to hit then there are going to be trade offs between them. And that is going to be much messier.