Kenny, Charles (2005) “Why Are We Worried About Income? Nearly Everything that Matters is Converging” in World Development Vol. 33, No. 1, pp. 1–19

[1] Summary. — Convergence of national GDP/capita numbers is a common, but narrow, measure of global success or failure in development. This paper takes a broader range of quality of life variables covering health, education, rights and infrastructure and examines if they are converging across countries. It finds that these measures are converging as a rule and (where we have data) that they have been converging for some time. The paper turns to a discussion of what might be driving convergence in quality of life even as incomes diverge, and what this might mean for the donor community.

Everyone is interested in economic convergence; it would represent the catching up of the poor world with its rich contemporary. Basically, everywhere started poor, but now some places are wealthy. Income has been the sole, or at least overriding criteria, for some time and this stems from a humanitarian impulse to see improvements in the global standard of living. Kenny quotes Lucas:

[I]s there some action a government could take that would lead the Indian economy to grow like Indonesia s or Egypts? If so, what exactly? If not, what is it about the ‘‘nature of India’’ that makes it so? The consequences for human welfare involved in questions like this are simply staggering: once one starts to think about them, it is hard to think about anything else (Lucas, 1988).

[2] This interest in income comes from the linking of income to most if not all quality of life measures. Even if income isn’t your main focus in a humanitarian sense, an increasing income in a poor country will probably help you achieve your aim. As income has diverged many have inferred that quality of life was diverged as well. However, while income has diverged, quality of life measures are converging almost across the board.

The link between the quality of life an income

GDP per capita is an incomplete measure of wellbeing, this is one of the reasons HDI was introduced by the UN. In fact, from infant mortality to life expectancy to war deaths per capita, there is very little correlation between income, income growth and other quality of life indicators. If you have been seeking to increase someone’s life, then increasing their income may not be the easiest way to do so.

Evidence for the convergence in measures of the quality of life

This paper builds on the work of Crafts, Ram and Ingram.

HDI of poor and rich countries are converging, despite relative inequality and massive poverty remaining common. [2-3] Calorific intake, primary enrolment and urbanisation are all converging.

Methods of measuring convergence and methodological issues

 There is a discussion of different measures of convergence and divergence here which kinda goes over my head. I will revisit if necessary.

Data quality is an issue, with income and all other quality of life data.

[4] Weighting is also problematic, should China count the same as Sierra Leone, or should measures be weighted to reflect China’s massive population? Using individual data rather than national per capita data can massively change our analysis.

Which quality of life measures to use is also a  question which is important. There seems little scientific way to select which measure to use and how much importance to give it. However, effort has been made to use measures for which there is good coverage in area and time.


The results presented in Appendix Tables 11–13 suggest almost every potential quality of life variable shows significant variation across countries. In turn, this suggests that, either throughout history some quality of life indicators have been higher in some parts of the world than others, or that, in some point in the past, there must have been divergence. The available evidence suggests elements of both stories, although with a predominance of the second. The evidence also suggests that more recently (for most of the 20th century) the story is reversed—it is one of convergence [my emphasis].

Maddison shows that the long term and continuing pattern in income is one of divergence. In 1000 AD Western Europe and Africa had roughly equal GDP per capita ($400), but by 1998 Africa had reached the income Europe had in the early 19th C whereas Europe’s income was now 13 times that. Since 1950 inequality has not so much continued to diverge as stagnate.

[5] However, apart income GDP, and depending on which measure of convergence you use, all or nearly all other measures have converged since the industrial revolution worldwide.


There is a historical minimum life expectancy of around 24 (younger than this and presumably societies just collapse). Divergence occurred from the early modern period until the end of the 19th C and then convergence in life expectancy began.  This convergence in life expectancy has been driven by a convergence in infant mortality. Calorific intake improved in many places and this has in part driven the convergence, although this factor is unlikely to have been enough to enough to explain all the convergence.


Divergence in literacy levels can be traced back before the 18th C. By 1913 literacy in India was 13% and in the UK around 96%. Global literacy in the rich world has reached ~100% and the poor world’s literacy has improved too, closing the gap. Between 1950-99 global literacy rose from 52% to 81%, driven largely by improvements in the poor world.

A driving factor in this is the increased availability of primary education. Tertiary education has also become more common in the developing world.

Social Indicators

Female literacy as a percentage of male literacy (an important measure of economic potential and gender equality), has converged since the 1970s from 59% to 80%. The percentage of children not in the global labour force has also decreased from 76% to 90%.

War deaths per capita is the least encouraging measure. The world remained a war like place in the developing world throughout the end of the 20th C.

Other, more lighthearted but important, measures have also converged. Beer production per capita has nearly doubled since 1950, representing an increase in “non-necessary” production.

What is going on?

Why have we seen convergence in quality of life measures accompany divergence in income?

There  could be large returns to small increases in income for the very poorest people. A little extra food can hugely improve the immune system for example.

How do we explain the performance of Africa over the last 50 years? GDP per capita has increased from $477 to just $561 over the 40 years (1960–99), falling from 4.8% to 1.9% of the average for a high-income country. Compare this to an under-five survival rate which has risen from 746 to 839 per 1,000 live births over the same period—or 77% of the high income survival rate to 84% of that rate. In terms of infant survival (86–91%), life expectancy (57–60%, despite the impact of the AIDS crisis reducing life expectancy by three years 1992– 99) and gross primary enrollment (35–70%), the trend is also one of convergence (all figures from World Bank, 2000). While Africa remains far behind, it is catching up on these measures, which is more than can be said for its performance on income.

Ingram argues that income tends to have a declining marginal impact on quality of life. Small increases in income reap massive rewards.

However, there appears to be more going on (even if the above argument appears very important). Mozambique saw it’s per capita income decline over the period 1950-99 but its life expectancy, literacy and primary enrollment all increased.

Looking at the data they show us that it takes one tenth of the income that it did in 1870 to live the same amount of time. Life expectancy for countries with a GDP per capita today of $300 have the same life expectancy as countries of 1870 with a GDP per capita income of $3000. Life expectancy has also become far more egalitarian than it was in 1870.

Better immunisation plays a role. Increase in primary enrollment and literacy have helped the efficacy of other public health campaigns, who can now use posters, where they before had to rely on word of mouth. 

To back up this “public health policy not just income helps improve quality of life” argument it has been observed that urban mortality is lower than rural mortality – something only achieved in Europe and the US after extensive public health campaigns.


Different income measures yield different degrees of divergence. For example, the proportion of people living in extreme poverty (less than $1 a day) has decreased tremendously; a massive convergence with the rich world. However, income has diverged in more general terms.

While aid, the Washington Consensus or globalisation have had mixed outcomes with respect to income, by other measures they should surely deserve some degree of credit for the almost across the board improvement in quality of life measures.

The debate today rests on the assumption that there has been a failure in the developing world, because there has been so little improvement in income, perhaps this pessimism is misplaced.


About Left Outside
I blog, I drink, I study at the LSE, I work at a wine shop.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: