Health

The Relation between Health and GNP: India and China

Greg Mankiw points to a Google tool (from Gapminder) plotting per capita GNP and life expectancy of several countries over time. I presume it is supposed to show the strong correlation between economic growth and improvements in health, but I noticed some problems with.

Insufficient Correlation between Health and GNP

As a patriotic Indian I first compared the plots of India to China and then Brazil. I expected China’s plot to show significant improvements in life expectancy since 1980, when economic growth accelerated. Yet, China’s plot hardly moved north after 1980, despite large increases in GNP. Even for India, as economic growth accelerated, life expectancy increases seemed to slow down.

These observations seemed to contradict the belief that there is a direct correlation between the two. So, I extracted data from the plot to get the following table for India, China and Brazil.

	1975	1980	1985	1990	1995	2000	2004
India
Per capita GNP (USD)	1139	1178	1385	1701	1994	2415	2885
Life expectancy	52	54	57	59	61	63	63
GNP Growth		3.4%	17.6%	22.8%	17.2%	21.1%	19.5%
LE Growth		3.8%	5.6%	3.5%	3.4%	3.3%	0.0%
China
Per capita GNP (USD)	595	763	1181	1596	2734	3982	5364
Life expectancy	65	67	68	69	69	70	71
GNP Growth		28.2%	54.8%	35.1%	71.3%	45.6%	34.7%
LE Growth		3.1%	1.5%	1.5%	0.0%	1.4%	1.4%
Brazil
Per capita GNP (USD)	5502	6890	6640	6497	6940	7301	7531
Life expectancy	61	63	65	66	68	70	71
GNP Growth		25.2%	-3.6%	-2.2%	6.8%	5.2%	3.2%
LE Growth		3.3%	3.2%	1.5%	3.0%	2.9%	1.4%

For both India and China, the life expectancy improvements were highest in the periods prior to economic reforms. In India, this happened in the late 1980s and early 1990s, whereas in China that happened a decade earlier. After reforms, as per capita GNP increased the rate of life expectancy improvements actually decreased.

This may well be a manifestation of the law of diminishing returns. However, there are two other more likely explainations.

First, it could indicate that as GNP rises, the resulting income inequality may be impacting overall life expectancy. As fewer people earn more the GNP rises, but the large majority that gets relatively poorer are worse off.

Another ugly truth may well be that a market-oriented India and China provide far less for its people than did socialist India and China. It has been widely reported that public health systems in post-reform periods have deteriorated. This data shows that under the guise of reform, governments in both countries are not only withdrawing from the market but also from public services.

The Impact of HIV/AIDS

I cannot conclude without discussing the horrific impact of HIV/AIDS on Africa, which is best illustrated here. Playing the timeplot for South Africa, for instance, shows that while per capita GNP remained constant, life expectancy dropped from 55 to 45 years. Worse, the timeplot for Botswana takes a vertical plunge, showing that while GNP increased from USD 1548 to USD 9140 (possibly also because of a dying population), life expectancy dropped from 58 years to 35 years!

Update: The PSDBlog has a post about Gapminder and other visual tools, the challenge of badly presented data for development, and the opportunity for business.

No, not an ad for London’s underground, but a healthy reminder from Hans Rosling, co-founder of Gapminder of how the lack of common data standards and compelling visual display is hampering development work and perpetuating stereotypes.

© 2007 The Discomfort Zone.