In a recent post in Let’s Talk Development, my colleague Adam Wagstaff raised a number of interesting points about measuring inequality of opportunity in the context of human development indicators, such as access to health services or educational achievement variables. His three main arguments, as I read them, were:
- The Human Opportunity Index (HOI), first proposed in this LAC regional study, inevitably understates the true extent of unfair inequality in access to services among children. This is because the dissimilarity index – the measure of inequality of opportunity within the HOI – captures only some of the inequality in (say) access to water among children while, in Adam’s view, all of that inequality is unfair.
- The Index of Economic Opportunity (IEO), used in the second part of that regional study, is also likely to understate the share of inequality in, say, student achievement among 15 year-olds, for a similar reason: students are partitioned into groups with identical circumstances (family background, gender, place of birth, race or ethnicity, etc.), and inequality of opportunity is measured as the share of the overall variance in test scores that occurs between – rather than within - those groups. Adam feels this is too little, because some of the within-group inequality may also be caused by factors beyond the young person’s responsibility – such as feeling sick on the day of the test, for example.
- Finally (though not in this order), Adam questions whether the HOI really is of much use to policymakers, because “it makes it hard to dig down and see what’s going on”. He suggests, as an alternative, the between-component of the Gini coefficient or, I think more precisely, of a concentration index.