26 Nov 2015

Wrong Questions. Wrong Answers. Wrong Policies.

Samuel writes*

In 2009, Joseph Stiglitz, Amartya Sen and a team of economists established that we need to measure ‘quality of life,’ given the limitations of measures such as GDP and HDI. ‘Quality of life’ should be partially subjective. They suggested that surveys are conducted, focusing on people’s evaluations of their life, and their prevailing emotions. Ideally, after aggregating with other approaches, they would end up with a number: your ‘quality of life.’ We can use an example to investigate their approach.

Take a ‘strict’ Buddhist village in Myanmar. The village could be based around a temple. The village would only trade where necessary and could strive towards self-sufficiency, consuming only basic foods. Due to many monks, unemployment would be very high. Villagers would also need to travel for education or health services. We would conclude that the village’s GDP/capita is exceptionally low, due to lack of production. IHDI would be low due to lack of education opportunities and health provisions. Here, Stiglitz et al.’s ‘quality of life measure’ should extrapolate on what GDP overlooks.

Say, hypothetically, villagers are asked ‘how would you evaluate your personal achievements,’ with the options ‘v. high, high, moderately, low, v. low.’ A Buddhist could be more inclined to answer low, due to ontological disagreements with ‘personal’ achievements. The same findings could exist for relationships and occupations. Buddhist villagers could also, due to a low intensity lifestyle, be much less likely to experience very positive emotions. Due to low life evaluations, simplistic goods and a lack of positive emotions, we would conclude that the village had an exceptionally low quality of life. The village would be in dire need of development.

However, a contextual analysis of the village, investigating both the discourse and culture, could show contentedness. More importantly, it could show different development needs, such as better transport to hospitals in other villages, which would potentially not affect ‘quality of life’ responses. The problem is not, necessarily, that Stiglitz et al. picked the wrong measure for quality of life; the problem is broader. Through quantitative research, we cannot understand development. Development needs rich contextual understanding, and this cannot be either numerically coded or aggregated. When we aggregate and code, we lose valuable information about what development is, what it means to different communities and how aid can be provided. Stiglitz et al. establish that the subjective dimension is important, but they use closed-ended questions; only open-ended questions could have provided us with useful information about the village.

Quantitative measures are attractive for policymakers for the same reason that they are flawed: they make the complex, simplistic. And, potentially, global development is not a simplistic, aggregate-able concept. It may be time to consider more mainstream macro-qualitative research, at least in the form of triangulation, in development studies.

Bottom Line: We are trying to condense and simplify development for policymakers. However, in most cases development is not simple, nor can it be aggregated. Oversimplification results in improper measures and improper measures result in striving for the wrong goals, with the wrong policies.
* Please comment on these posts from my microeconomics / growth & development economics students, to help them with unclear analysis, other perspectives, data sources, etc.