Impactism
日本語 Join Impactism
Manifesto The Impact Standard Foundations Objections Read For organizations 日本語
Join Impactism
A row of analogue pressure gauges and valves photographed in monochrome
Photo by Miguel Ángel Padriñán Alba on Unsplash

Essay

Alternatives to GDP, and the One Thing None of Them Measure

September 10, 202610 min read

Alternatives to GDP are measures built to hold what gross domestic product leaves out. The best known are the UN’s Human Development Index, the Genuine Progress Indicator, Bhutan’s Gross National Happiness Index, the OECD’s well-being dashboard and its Better Life Index, New Zealand’s wellbeing budget, and the European Union’s long “beyond GDP” programme. Some replace GDP with a rival number. Some correct GDP by adding what it misses and subtracting what it wrongly counts. Some refuse a single number and publish a dashboard instead. Some leave the statistics alone and change how a government decides where the money goes.

They are serious work by serious people, and we want to walk through them honestly, then say the thing that usually goes unsaid. Every one of them measures a country. None measures a life. Those are different objects, and the difference is not a technicality.

The man who built it said so first

The complaint about GDP is as old as GDP. Simon Kuznets led the study that produced the first official national income estimates for the United States, delivered to the Senate as Senate Document No. 124 of the 73rd Congress and referred to the Committee on Finance on 4 January 1934. Inside that report, in the section on concept, scope and method, he wrote the warning that every later reformer has quoted (Kuznets 1934):

The welfare of a nation can, therefore, scarcely be inferred from a measurement of national income as defined above.

He gave his reasons in the two sentences before it. Economic welfare cannot be judged without knowing how income is distributed. And no income measure attempts to count “the intensity and unpleasantness of effort going into the earning of income.” He then listed the abuses he expected: reading a 30 per cent fall in national income as a 30 per cent fall in a nation’s welfare, or treating a country with twice the income of another as twice as well off.

The author of the number told the Senate not to use it as a measure of national welfare, in the document that introduced it. It has been used that way ever since. That is worth sitting with before anyone proposes a better index, because it suggests the problem was never a missing statistic.

Replacing the number: the Human Development Index

The most successful alternative is the oldest of the modern ones. The UNDP’s Human Development Index combines life expectancy at birth, mean and expected years of schooling, and gross national income per capita, normalised and combined as a geometric mean (UNDP). It gave the world a second number that finance ministers could not ignore.

The UNDP is candid about the cost of that success. Its own page says the index “does not reflect on inequalities, poverty, human security, empowerment, etc.” and that it “simplifies and captures only part of what human development entails.” A single number bought attention by paying in resolution. We have written about the argument inside the HDI’s own founding, between an economist who wanted one number loud enough to be heard and one who thought compressing human development into a single figure was a mistake, in The Shoulders We Stand On.

Correcting the number: the Genuine Progress Indicator

The Genuine Progress Indicator takes a different route. Instead of building a rival index it starts from personal consumption expenditure, a large component of GDP, and adjusts it using 24 components. It subtracts things GDP counts as gains but people experience as losses, such as crime, pollution and the depletion of natural capital. It adds things GDP ignores, such as household work and volunteering. It weights the whole by income distribution.

The result is the most uncomfortable finding in this literature. Ida Kubiszewski, Robert Costanza and colleagues assembled GPI estimates for 17 countries covering 53 per cent of world population and 59 per cent of world GDP over 1950 to 2003. Global GDP per person rose throughout. Global GPI per person peaked in 1978 and fell afterwards. Past roughly $7,000 of GDP per capita, in their estimates, GPI stops rising at all (Kubiszewski et al. 2013).

The authors report the criticisms of their own method at length, and they are real: contested valuation techniques, the assumption that built capital can substitute for natural capital, a defensible but subjective choice of which components to include. GPI is an argument, not a reading off a dial. It is still the sharpest version of the claim that a growing economy and an improving one came apart in the late 1970s.

Asking people: Bhutan’s Gross National Happiness

Bhutan is the clearest case of a state binding itself to a measure other than GDP. The phrase came from the fourth king, Jigme Singye Wangchuck, in the late 1970s. The index itself was built much later by the Centre for Bhutan and GNH Studies using the Alkire-Foster method, and it covers nine domains: psychological wellbeing, health, time use and balance, education, cultural diversity and resilience, good governance, community vitality, ecological diversity and resilience, and living standards. Those hold 33 indicators. A person counts as happy if they have sufficiency in at least 66 per cent of the weighted indicators, and the population is sorted into four bands. The survey has been run in 2010 and 2022 (OPHI).

Notice the design. GNH does not average a nation’s mood. It asks whether each person has enough across nine kinds of life, then counts the people who do. That is closer to a census of sufficiency than to a score, and it is the most interesting thing any government has built here.

Refusing a single number: the OECD

The OECD went the other way and declined to compress. Its Well-being Framework, running since 2011, covers 11 dimensions of current well-being plus four kinds of resource for future well-being, operationalised through a database of more than 80 indicators updated quarterly (OECD).

The Better Life Index sits on top of it, and its design is the quiet radicalism in this whole field. There is no OECD ranking. The tool “lets you compare countries based on what matters most to you,” and you choose how much weight each dimension carries. The organisation supplies the evidence and leaves the verdict to you. That is a statement about who owns a yardstick, dressed up as an interface.

Japan runs its own version. The Cabinet Office has published a Survey on Satisfaction and Quality of Life since 2019, with a well-being dashboard of indicators alongside it, the eighth round drawing on about 10,000 respondents (内閣府).

Changing the budget: New Zealand

New Zealand did the thing that matters more than publishing an index. It changed what the money had to justify. The Wellbeing Budget of 30 May 2019 required every funding bid to carry a wellbeing analysis, assessed against priorities drawn from the Treasury’s Living Standards Framework and its four capitals: financial and physical, human, natural, social (New Zealand Treasury 2019).

The finance minister’s own framing is worth quoting, because it is unusually plain for a budget document: “Success has been declared on the basis of a narrow range of indicators, like GDP growth. But New Zealanders have questioned that claim of success when they have seen other things that we hold dear … getting steadily worse.”

Writing it into law: the European Union

The EU has worked at this for nearly two decades. In November 2007 the Commission, with the European Parliament, the Club of Rome, WWF and the OECD, held the Beyond GDP conference. On 20 August 2009 the Commission adopted the communication “GDP and beyond: measuring progress in a changing world,” COM(2009) 433 final, which is unusually even-handed about the thing it is criticising (European Commission 2009):

For all of its shortcomings, it is still the best single measure of how the market economy is performing. But GDP is not meant to be an accurate gauge of longer term economic and social progress.

The communication set out five actions, from complementing GDP with environmental and social indicators to extending the national accounts themselves. The work is still running: the EU’s Joint Research Centre now describes a beyond-GDP dashboard called for by the Eighth Environment Action Programme (JRC).

What none of them is for

Line these up and the family resemblance is obvious. Every one of them answers the same question. How is this country doing, and where should a government put the next euro.

That is a good question, and none of these tools claims to answer another. It is worth being clear about that before criticising them, because the usual complaint about wellbeing indices is that they fail to tell you how to live, which is like complaining that a thermometer will not hold a door open.

But the question of how to live is the one that has to be answered by someone, and a national statistic cannot reach it. A country’s HDI does not tell a nurse in that country whether her work counted. A rising GPI does not tell the person who cared for a dying parent that the years were well spent. The measures are population averages, published annually, used by ministries. A life is singular, lived once, and judged by the people around it.

Why the two yardsticks are different objects

The difference is not one of scale, as if a personal measure were a national one divided by population. They differ in what they are for and in how they fail.

A national indicator exists to allocate. An institution uses it to settle a budget line, and what it ranks is a government’s performance. Governments are supposed to be ranked, so the thing can be central, published and singular without doing much harm. None of those three properties is safe once the subject is a person. Regard is handed over by a colleague or a neighbour, and putting a published central number in the middle of that builds a scoreboard for people. It is what every serious critic of measurement warns about, and what we spend most of our writing refusing.

The gaming runs differently too. A national index is gamed by the ministries that report into it. What suffers is the statistic, and a statistic can be repaired by fixing the method. Under a yardstick for a life, everyone living beneath it joins in, and what suffers is not a statistic.

There is no opting out of the second one. A wellbeing dashboard can sit unread for a decade without consequence, but the question of what makes a life worth having never sits unanswered, because a person answers it every day by how the day is spent. If nobody picks the answer on purpose, one arrives by default, and we have argued at length that money took the seat first and attention is taking it now: There Is Always a Yardstick.

Where we sit

Impactism is not another index. We are not proposing an eighth number to put beside GDP, and if we ever publish a global score, we will have become the thing we set out to avoid.

What we want is a change in what a society honours. That is a cultural claim, not a statistical one. Capitalism is the engine and we want it running; GDP is a fine instrument for watching the engine. The compass is a separate question, and it is answered by what people admire out loud.

Which is why the design rules are strict. Peers confer the recognition rather than an authority issuing it. It stays plural across many communities and it decays, so nothing accumulates into a single running figure. Money cannot buy it at any rate. There is no central score, and nobody has to pass a test.

The one place the national and the personal literatures meet is on the money question, and both sides are messier than the slogans. The claim that income stops improving life past a threshold is weaker than it is usually reported to be, and we have taken that apart honestly in The Log Curve.

What we are not claiming

We are not claiming these measures failed. HDI moved the conversation. GPI produced a finding nobody has comfortably refuted. GNH asks about sufficiency person by person across nine domains, which is harder than averaging a national mood. New Zealand changed a budget process, which is harder than publishing an index.

Nor are we claiming a personal yardstick has been shown to work. No society has run the version we describe. Whether contribution can be recognised without hardening into a gameable score is an open problem, and it is the one that has sunk most attempts to make non-monetary value real. We are proposing a direction and marking it as one.

What we are claiming is narrow. Ninety years of work on alternatives to GDP has been aimed at governments, and it should be. The question of what a single life is measured against was never on that agenda, and it will not be answered by a better national statistic, however good.

Make impact, not money, the measure of a life.

Sources

← All essays

Impactism is free — no paywall, no membership fee, and it will stay that way. If this argument was worth your time, donations are what pay for it. One-time or monthly, any amount.

Support the argument