Tuesday, May 31, 2011

Does the OECD's 'better life index' sound like fun?

I am not sure the OECD’s better life index is meant to be fun. But I have had some fun playing with it. The index is interactive. The fun comes from giving different weight to 11 different criteria (or topics as they are described by the OECD) and then observing how this affects rankings of well-being of OECD countries.


The criteria used in the index are: housing, income, jobs, community (individuals’ perceptions of the quality of their support networks), education, environment (air pollution by tiny particulate matter), governance (voting and transparency), health, life satisfaction, safety (assaults and homicide) and work-life balance (working mothers, total hours worked and leisure).

Under the default setting, with all criteria being given equal weight, the countries that come out on top are Australia, New Zealand, Canada and Sweden. If you suppress all criteria other than income, Luxembourg is a long way ahead of the field, followed by the United States and Switzerland. The income measure used in the study (reflecting household financial income and wealth) has Australia in 14th place and New Zealand in 25th place.

The substantial difference between the outcomes of these weighting systems is interesting. In a previous post I observed that all well-being indicators tend to tell similar stories about well-being levels in different countries. The two observations are actually consistent. My research covered a larger number of countries, including many poor countries as well as the wealthy democracies of the OECD. Well-being indicators tend to tell a similar story when wealthy countries are compared with poor countries, but can tell different stories when wealthy countries are compared to each other.

Equal weighting of a range of indicators and a focus on income alone seems to me to be equally arbitrary approaches to well-being comparisons. Well-being is obviously affected by factors other than income, but it would be difficult to argue that all relevant factors are equally important. Value judgements have to be made to determine appropriate weights. An appropriate weighting system might be derived by conducting surveys to obtain weights reflecting the values of people in different countries. Alternatively, surveys could be used to obtain weights reflecting the values of people with different political views in particular countries, or across the whole of the OECD.

In the absence of such survey evidence, I have looked at the rankings for three somewhat extreme political groups drawn from my own imagination: Scrooges, Socioholics and Warm Fuzzies. As I imagine them, all three groups perceive governance and safety as being important to well-being. The Scrooges add income as the only additional factor. The Socioholics add housing, jobs, education and health in addition to income. The Warm Fuzzies exclude income and all the additional factors added by the Socioholics, but replace those factors with community, environment, life satisfaction and work-life balance.

So, which countries come out on top of the welfare rankings according to the values of these three political groups?

Scrooges: The countries that come out on top are Australia, Luxembourg and the United States. New Zealand is placed about 8th, behind Sweden, Austria, Canada and UK.

Socioholics: Australia and Canada come out on top, followed by New Zealand and the United States.

Warm Fuzzies: Australia, Denmark and Sweden are on top, followed by New Zealand, Canada and Norway.

What do I get out of this? My main observation is that Australia seems to come out fairly well, whatever coloured political lenses you use. The well-being of New Zealanders also looks fairly good, particularly if you adopt either a Socioholic or Warm Fuzzy perspective.

Having had some fun, the more serious question that comes to mind is whether a focus on the OECD’s well-being indicators (and other similar constructions) is likely to distract political attention away from much-needed economic reforms to improve the economic strength of some economies. For example, if well-being indicators suggest that people in some lovely country (New Zealand comes to mind) tend to enjoy living standards substantially higher than other countries with comparable per capita GDP levels, there may be a tendency for the government of that country to become complacent about establishing conditions more favourable to further improvement of living standards.

Postscript

Roger Kerr, executive director of the New Zealand Business Roundtable, has commented on the OECD's well-being index here (with a reference to this blog).

My subsequent posts on the OECD's well-being index are:
Do well-being indicators all tend to tell similar stories about OECD countries?
and
How could the OECD's well-being indicators be improved?

I would also like to draw attention to posts by David Giles on his 'Econometrics Beat' blog: here and here.

4 comments:

Shona said...

I'm surprised that the Scandinavian countries didn’t rate highly in the socioholics.

Winton Bates said...

I was surprised too, Shona. The Socioholics would rank Sweden and Denmark just behind New Zealand and the US.

A comparison of Sweden and the US is informative. They both have similar rankings in terms of governance and safety. The US ranks a lot higher in terms of income and housing and Sweden ranks higher on health.

So the US performs better in terms of providing the private goods that the Socioholics like and Sweden performs better regarding health.

My Socioholics like some things that are not provided by welfare states.

Dave Giles said...

I've given a statistical analysis of some aspects of the BLI and its weighting scheme on my blog at davegiles.blogspot.com. See the posts titles "It's a Wonderful Life", and "An Even Better Life".

Winton Bates said...

Thanks for drawing my attention to your posts, Dave.

I have linked to them in the postscript.