Economic progress and health improvement

The economic performance of countries is generally compared using the concept of gross domestic product (GDP). This is a measure of economic activity developed in 1940 by two British Nobel prize winning economists, James Meade and Richard Stone.1 The concept and the associated data framework—national accounts—were designed to manage the economy in wartime. The extent to which it has gained much wider currency is perhaps surprising.

Even with a very restricted notion of welfare, the link between GDP and welfare is not straightforward,2 and it has been recognised for many years that there is more to welfare than consumption.3 In 2008 President Sarkozy was concerned that France’s opinion of itself suffered because it compared badly with some other advanced countries on the basis of GDP per capita. A part of the reason for this “underperformance” is that France’s per capita labour supply is lower because the typical working week is shorter in France than in the United States and the average retirement age is lower. Sarkozy was keen to have a metric that took account of quality of life measures like this and set up a commission chaired by Professor Joe Stiglitz, a former chief economist at the World Bank and winner of the 2002 Nobel Prize, to report on the measurement of economic and social progress. The report was published in September 2009.4

The report is more useful as a consolidation of existing knowledge than as a source of new ideas; inevitably in some parts it has not kept abreast of the recent literature. But it does provide a valuable summary of what we currently know about non-economic aspects of wellbeing. Health obviously features prominently among this. To give an example, the GDP per capita of the US is nearly 40% higher than in France,5 but life expectancy in France is nearly three years longer than in the US.6

The report does not suggest any obvious way of producing a single indicator of welfare that reflects both differences in life expectancy and differences in output, either between two different countries (or regions) at the same point in time, or within any given country at two different times, and we should not expect it to. Economists often talk about the difficulties of producing a single measure that combines two different goods, and they resolve the problem by assuming that the market prices of the different goods provide a basis for making this combination. That assumption underlies the calculation of GDP. There is no obvious way of doing this for life itself. Nevertheless, governments have to allocate resources between spending on health services, which can save lives, and spending on other important things like education. This requires balancing benefits that might accrue from saving lives against other types of benefit and involves deciding how much value to place on each.

The National Institute for Health and Clinical Excellence (NICE) requires strong reasons for supporting an intervention that costs more than £30 000 (€33 300; $49 200) to deliver a year of good quality life.7 The Department of Transport generally puts a much higher value on life when deciding on measures to reduce the risk of road and rail deaths. This implies that lives could be saved with no extra overall expenditure by diverting resources from spending on road and rail safety to spending on medical interventions.8 More controversially still, some economists have argued that the value of life depends on what people can afford,9 with the inference being that life is worth more to the rich than to the poor, an approach that, despite the logic behind it, is unlikely to find widespread popular support. But a general point emerges that, even with a relatively modest value put on life, such as that proposed by NICE, the value of the increase in longevity over the 20th century is very substantial compared with the increased value of economic output.

If producing a measure of the value of life is controversial, at least survival rates can be measured clearly. A more complete measure of welfare would take into account not only how long people live but also their state of health. Measures of quality adjusted life years (QALYs) or disability adjusted life years (DALYs) of course exist. To produce these, people’s state of health needs to be measured in a coherent way, and—if a single indicator is produced—a means of comparing the different health states is needed, which once again requires some kind of valuation.

Aside from the production of aggregate indicators, the report stresses a separate question about how far social and economic inequalities should be taken into account—also a longstanding problem faced by economists.10 Again, it is hard to foresee a satisfactory answer. Overall, the report regards itself as opening rather than closing a discussion. If it helps policy makers and commentators to see that there is more to life than GDP it will have made a valuable contribution.