post-autistic economics review
Issue no. 20, 3 June 2003
article 3



      issue 20 contents                            PAE Review index                               home page




Is GDP a good measure of economic progress?*

Olivier Vaury   (École Normale Supérieure, Paris)

© Copyright 2003 Olivier Vaury


Every year, or even every quarter, economic growth figures are anticipated and scrutinised to assess the economic health of a country. In spite of abundant commentary in the media by politicians and economists, the very notion of economic growth remains elusive: who really knows what it really measures ? Yet the level of GDP (or GDP growth) is probably the most widely used indicator for piloting economic policies around the world and for making international comparisons.

When one says that “GDP growth reached 3% in 2002”, what does that mean ? Broadly speaking, GDP measures the amount of goods and services produced in a given place (a country, a region, etc.), in a given period of time (a year, a quarter, etc.). All goods and services ? Well, that’s where the whole issue lies!

Our point is that GDP includes goods and services that do not increase a country’s economic wealth, and, furthermore, excludes goods and services that do. Thus, the use of GDP as an indicator of economic progress is flawed and results in biases in international comparisons.


What GDP forgets

“Marry your cleaning person, and you will make GDP drop!”. This weird remark, made by the famous French economist Alfred Sauvy, points to the fact that GDP excludes (or significantly underestimates) goods and services produced outside the official market economy. The bits forgotten by GDP can be divided into three categories:


-          Household production: marrying the cleaning person means transforming a standard marketed activity (house cleaning, paid at a given rate, etc.) into domestic work, not accounted for in GDP as there is no way to measure the value added by this service (no price paid). A priori, this change does not alter the level of service enjoyed by the newly married consumer (in both cases, house cleaning was made, by the same person; this example is used to point out the insufficiencies of GDP, not to advocate more domestic work…). Is that a problem? Yes, as by any measure domestic production represents a large part of goods and services. For instance, one estimates that people spend 17% more time in household production (cleaning, cooking, childcare, etc.) than in paid activities. According to various studies carried out in France, domestic production could represent as much as 75% of standard GDP.


-          Voluntary work: a bicycle repaired by a friend makes GDP fall if the work used to be done by (and paid to) a professional. Thus, a society where voluntary work is widespread will enjoy a higher level of economic well-being but not necessarily a higher GDP (in fact, even a lower GDP).



-          Public administration: in public accounting rules, GDP is the sum of values added by all economic entities, i.e. the difference between the value of production and consumed inputs (energy costs, raw materials, etc.). Thus, value added is in fact constituted by two main parts: wages and profits. But for public administration, no value of production is available as public services are generally not bought by anyone on a market (think of public gardens maintenance or tax collection). To include them in GDP, accountants decided to measure value added as wage costs (we exclude capital depreciation from our analysis). As a result, the contribution of public services to GDP is always underestimated. By the same token, a free service resulting from a past public investment (a road, a fountain, a public park or a public sport facility) will not appear in GDP, contrary to its private equivalent (priced road, private sport facility, etc.).


What GDP should forget


“Burn Paris and you will make GDP grow!”. Another weird remark, another problem with GDP : GDP only includes positive values: if something is destroyed, then rebuilt by a private company, GDP goes up while economic well-being is unchanged. Those who use GDP as a good measure of economic progress forget that production is closely linked to destruction. In two ways:


-          Production as measured by GDP is often just compensation for a previous destruction (think of booming activity after a cyclone or of most legal and medical activities). If lawyers prosper because there are more crimes and more offences, does that mean the country is richer?


-          Production is, by definition, destruction: destruction of human and natural capital. What about two countries achieving the same level of standard GDP but, for one of them, through the exhaustion of its natural and human resources? It reminds us of those companies who report profits only by under-reporting depreciation of assets. The case is not just theoretical: Britain and France have roughly the same GDP but British workers work 25% more.



The Obesity Connection


The remarks made above can be illustrated by the example of obesity. Obesity is not just a pathology, it is also a formidable way to create standard economic wealth without increasing well-being (figures below can be found at, the website of an American think tank, Redefining Progress):

-          in the US alone, food companies spend around $20bn in advertising to convince consumers to eat more;

-          unfortunately, their efforts prove successful as each day, one in every four Americans eats in a fast-food restaurant, for a total expenditure of $110bn a year;

-          consequently, Americans are increasingly obese (over the past 30 years, the number of Americans not able to use a standard airplane seat – due to overweight, not to financial difficulties… – has risen by 350%);

-          obese people (and others) spend around $30-50bn a year in weight-losing products, gym facilities, etc.;

-          but most do not manage to lose the weight “gained” in fast-foods, which feeds medical expenditure linked to obesity (obesity often leads to diabetes, increases the likelihood of heart attacks, etc.), at around $50bn per year.


These expenditures boost GDP, but their contribution to economic well being is at best questionable. As asked by one of Redefining Progress’s leaders: “if GDP is up, why is America down?”.


Flawed international comparisons


The arguments presented above cast doubt on the usefulness of GDP as the main “pilot” of economic policy. If the thermometer is wrong, then the policy based on it should be wrong too. The use of GDP produces biases in favour of a particular set of political choices, consisting in marketisation of economic activity. This also means that international comparisons based on GDP are fundamentally flawed. In two ways, following the two problems associated with GDP:

-          two countries with different levels of “marketisation” cannot be compared on the basis of GDP, as GDP will not include the same activities in the two countries (part of economic activity will be excluded from the comparison);

-          two countries with different levels of destruction cannot be compared on the basis of GDP as these very destructions are not taken into account. Peaceful societies, characterised by a lower level of crime (and consequently legal activities, private prisons, etc.) are penalised in terms of GDP! As are countries with a healthy way of life!


And all this is without even taking into account the technical difficulties: how to compare GDPs measured in different currencies? If we use simple exchange rates, we run the risk to have very volatile results (the euro lost 30% of its value against the dollar from January 1999 to October 2000, but it would have been stupid to conclude that GDP dropped in the same proportion). The purchasing power parity method, used by economists to take into account the fact that currencies do not buy the same amount of goods and services, is not without problems, too. Is it relevant to compare the price of the same basket of goods and services in different countries (to derive purchasing powers) when the structures of the studied economies vary. Finally, statistical methods used to measure outputs and prices differ significantly across the world. Thus, cross-country comparisons create many more difficulties than the already problematic national GDP calculation.


Paradoxically, it is mostly for poor countries that alternatives to standard GDP have been developed, though they are badly needed for rich countries too. For instance, the United Nations Development Programme (UNDP) calculates a Human Development Index (HDI), which includes GDP per capita, but also the literacy rate, life expectancy at birth and school enrolment ratios. A report to the French ministry for Social Economy (abolished by the current government) suggested that the human development report (now limited to poor regions) be extended to cover Europe. Others have tried to build more relevant measures of GDP, by removing from standard GDP values added that do not increase well-being or that contribute to the destruction of natural resources, and by adding domestic and voluntary work. One such example is the Genuine Progress Indicator, calculated by the US NGO Redefining Progress (see above) for the US, and which has been stagnating since the early 1980s (while standard GDP almost doubled over the same period).


Yet, these interesting experiments are not without drawbacks : besides numerous technical difficulties, similar to those identified for the measure of standard GDP, the choice to include or exclude economic activities from the new index can easily be arbitrary. For instance, should we exclude health care linked to avoidable diseases on the ground that it merely reflects a fundamentally unhealthy way of life (the counterpart of other parts of GDP such as fast-food activities, or tobacco, so as not to be counted twice…). Or should we include it because it makes people live better, everything else being equal…? Should we exclude legal activities linked to divorces because divorce is a sad thing which reflects the disintegration of families in contemporary societies, or include them as they allow women to enjoy more independence?


In my opinion, we should give up trying to compete in terms of GDP (an aggregated indicator). Economic development is always something with many dimensions. Therefore, an economic system should be judged on its ability to provide individuals with what they need to achieve well-being : food, health, leisure, clean air, a high life expectancy, means of communication, etc. For each of these sub-categories, it is possible to build indicators that reflect to what extent the population enjoys access to these resources. Not an average indicator but one that takes inequality into account : two people with a phone each is better than one people with two phones and one without any. Thus, we would be able (in fact we already are, but these figures are never publicized) to compare two countries by comparing their ability to provide these essential goods to the largest possible part of their population. We would certainly have some surprises, such as that the World Health Organisation ranks the US health system as 37th in the world, while France ranks 1st and Portugal 12th, two countries with a much lower GDP per capita.



*This article originally appeared in Les Éconoclastes, Petit breviaire des idées reçues en économie, Paris: Éditions la découverte, 2003.



Olivier Vaury, “Is GDP a good measure of economic progress?“, post-autistic economics review, issue no. 20,  3 June 2003, article 3,