GDP is flawed, but it matters

Suggestions to replace GDP shouldn't be a proxy for advancing actual policies

July 25, 2018
A factory in Hatfield. Photo: Daniel Leal-Olivas/PA Archive/PA Images
A factory in Hatfield. Photo: Daniel Leal-Olivas/PA Archive/PA Images

At the moment, many economists and non-economists alike have little confidence in Gross Domestic Product (GDP). It is regularly criticised for being a flawed and incomplete measure of economic welfare. Politicians mindlessly seek to maximise it, it is said, promoting economic growth to the detriment of other policy goals.

Many of these criticisms have resonance. They point to a deep sense of malaise with our economic management, to a feeling that we are undermining those aspects of human welfare which can’t be satisfied through our marketplace transactions. Social respect, a secure job, natural beauty, a liveable climate—issues such as these are felt to be squeezed out by the veneration of GDP growth.

But the insistence that we replace GDP to deal with this malaise can sometimes come across as a displacement activity in lieu of advocating a political programme.

The idea that technocratic adjustments to the way we measure our economy would go a long way to solving the malaise by themselves is illusory. Moreover, there is an elephant in the room—the very real connection between measured economic activity and social and political stability, which is often ignored.

Above all, an alternative measure would need to have credibility across a broad cross-section of society to take off; not just economists, civil servants and politicians, but also journalists, central bankers, businesspeople and investors. While alternatives to GDP have been discussed for 50 years, the lack of a politicalconsensus on potential reforms which could be made has prevented adoption and impact among anyone but a fairly small group of advocates.

In exploring these points further, it is useful to separate the criticisms into two groups.

The first is what we describe as the “technical” critique of GDP. This widely used measure of the size of the economy was developed in the 1930s and 1940s, when manufacturing—the production of things—was much more important in the west than it is now. GDP is held to be an increasingly flawed measure in economies which are increasingly dominated not just by services in general, but by internet-based services in particular.

Another technical critique is the question of how national accounts statisticians allow for quality improvement of existing goods and services and the emergence of new, innovative ones. They are not unaware of the difficulties. But they may very well be underestimating the impact of quality improvement on GDP. Others point to the fact the GDP growth can be vigorous when a country simply runs down its natural resources without investing in other sectors.

These arguments involve hard questions for economists and statisticians. They are not mere debating points. Some leading econometricians, for example, believe that the annual growth rate of real GDP (after stripping out inflation) is systematically underestimated by at least 1 percentage point.

It is the second critique which gains more attention in political circles. This is what we call the “policy” critique of GDP, arguments relating to the policy decisions which stem from its use.

A wide range of arguments is used in the policy critique, but in essence, they boil down to the following. The whole thrust of public policy is to mechanistically increase GDP, but this simply measures how much is being produced and consumed at a given point in time. There is much more to life than this purely short-term economic concept. The focus of public policy should be much wider than mere “economism.” It therefore requires indicators which should encompass everything which contributes to human welfare.

At one extreme, a few argue that GDP growth should be abandoned as an objective of public policy. An example is the assertion that it should be replaced with Gross Domestic Happiness. Most critics, however, want to retain the basic concept of GDP. Their aim is to widen its definition, often dramatically.

“Technocratic adjustments to the way we measure our economy would not solve the malaise”
Many such proposals have been put forward. Contrary to popular perception, economists have often been at the forefront of the debate. A famous—within economics at least—article by Nobel Laureate James Tobin and Bill Nordhaus was written in the early 1970s.

They anticipated many of the arguments which are made today. For example, allowance should be made for any depletion of natural resources which takes place in producing goods and services. GDP should be adjusted downward to recognise the costs imposed by pollution. Leisure time and housework should be valued. Deductions should be made for the implicit costs of “disamenities” such as congestion, and “regrettable” expenditures such as defence, which do not yield direct satisfaction to people.

Indeed, the economists who constructed the first standardised estimates of GDP way back in the 1930s and 1940s—Nobel Laureate Simon Kuznets was a prominent figure—appreciated at the time that factors such as these were being omitted from their calculations.

We can usefully ask why, given that these rather profound critiques have been around for a very long time, GDP has not been reformed in a rather dramatic way, even after superstar economists Joseph Stiglitz and Amartya Sen reviewed the question for the French government ten years ago.

A basic point, often overlooked in human affairs, is simple inertia. The conventions around the measurement of GDP are well understood, and it is a familiar concept.

But crucially, while GDP is very imperfect it's a good heuristic for two things governments (rightly) care about very much—livelihoods and revenues. When recessions happen, people lose businesses, jobs, homes and public services. Lives are disrupted and society is less politically stable. GDP’s imperfections should not detract from the fact that this really matters. In the financial crisis of the late 2000s, for example, United States GDP per capita fell by about 5 per cent, and while not the only driver of American’s current political turmoil, it was clearly a factor.

Politicians don’t seek to grow GDP because they think citizens are nothing but vapid consumers with no non-material needs. They do so because they fear the social discontent that comes with lost livelihoods and stagnant living standards. This is supported by the fact that an ostensibly Communist country like China sets targets for GDP—it is not the mere fancy of a society in thrall to free market individualism.

It was apparent in the 1930s that there had been a catastrophic drop in economic output in many western economies. Economists later estimated it at over 20 per cent. But policy makers did not at the time have available to them an accepted way of measuring the size of the economy. This was needed to meet an urgent and widely recognised policy aim of averting any repeat of this collapse.

The point here is that GDP emerged out of a widely shared political impetus and resulting consensus. The common imperative for policy makers in the 1930s was to measure the value of output in a market-oriented economy. No such broad-based common agreement exists now on reform.

Some might argue that the serious environmental problems which the world faces—from biodiversity loss to soil depletion to climate change—justify a new consensus. The idea that measures of economic progress incorporating environmental damage could help us find the right balance between economic activity and environmental protection is appealing.

But it should also be questioned. Our complex environmental problems are fundamentally political challenges which require political leadership and legwork.

If the economy is eroding things we value and over-providing things which are detrimental, citizens need to put pressure on their governments to intervene. Funding and resources need to go towards the things that are valued, and practical solutions found—or, where necessary, bold interventions made—to limit the things that are harmful.

Given that plastic pollution is one of the pressing environmental problems of our day, it’s a useful thought experiment to consider whether measuring GDP differently would stop a single turtle choking on a single plastic bag. Proponents of alternative measures might argue it would; policy-makers motivated to optimise their alternative measure will recognise the value of nature and invest in solutions.

This is fairly oblique. Coming up with an alternative measure actually subtle enough to pick up the effects of insidious pollution whose ecological impact is still very much in the process of being understood is technically extremely challenging. It would require both theoretical underpinnings and data collection.
“The degree to which policy is obdurately driven by GDP is often overstated”
The intellectual effort would be better spent on working out how to fund and implement the most cost-effective on-the-ground solutions, urgently; and making the political case for decisive action and international cooperation. We need not to get distracted rearranging deck chairs on the Titanic.

As another example, GDP could plausibly be adjusted downwards because pollution in cities is making people ill. Doing something about this requires an open political decision, such as policies which reduce car journeys, or induce people to drive less polluting cars. It should not shelter behind a technocratic number which few would understand in all its complexity.  People don’t tend to oppose such policies on the grounds that they would decrease GDP. Rather, they object to perceived curtailment of personal choice. A simple counter-narrative based on clear values that others can support, such as “I want people to be healthy” is far more straightforward than “if we do this adjusted GDP will go up.”

It is exactly in the democratic forum that such policies should be debated and decided. We should not be coy that sometimes we need to put up a fight for the policies and values that we support.

The adjustments and redefinitions of GDP which are suggested are inherently political demands. Expecting incontrovertibly right policies to flow from a grand unified measure of human welfare is a fundamentally technocratic perspective. A healthy democratic culture brokers practical change supported and understood by a broad constituency of citizens.

To be fair, it should be acknowledged that alternatives to GDP often focus more these days on providing supplementary information rather than providing a replacement single figure. But policy-makers already do have a wide array of statistics and indicators at their disposal. Indeed, in the UK, the Treasury’s Green Book already contains detailed guidance on how to incorporate social and environmental costs and benefits into formal policy appraisal.

It is absolutely worth articulating the extent to which a healthy natural world underpins human welfare and economic activity, and this is particularly valuable when it helps inform specific decisions with well-defined goals (see the UK’s natural capital approach for some of the most advanced thinking on this). But the degree to which policy is obdurately driven by GDP is often overstated, as evidenced by the gamut of policies in advanced democracies on everything from smoking cessation to protected areas to employment rights. These policies happen if people propose and win support for them.

So perhaps it’s time that we acknowledged that our problems require political vision, and not technical fixes. It is possible that measuring economic welfare differently would enhance our governance. But ultimately whether these ideas are adopted by policy-makers and wider society rests on a practical and political proof of concept which has been lacking to date.

Five things might help: more specifics on what these measures can really do for us in practical terms; more joining up with big political ideas; efficient use of information at the decision level; more recognition that what GDP does for us is not trivial; and more awareness that gathering and processing information is itself not cost-free and takes up policy resources.

Otherwise, the idea risks being viewed as yet another example of the concerted attempt to appropriate debate from the messy and imperfect forum of democracy to the “rational” world of the truly expert, in which everyone involved can agree.