Is anyone familiar with this subject? I went to a really interesting seminar on it last year, although my libertarian (but mostly consequentialist) friend who also went was a bit scornful of what he saw as leftwing economists looking for a way to counter the apparent success of GDP-based policies.

That seems unconvincing to me so far - the basic criticism seems to be that it's hard to measure happiness, which isn't an argument for giving up the effort to do so and arbitrarily assigning merit to something that we can measure more easily. Also, from reading that Wikipedia piece, there are some (a fewer proportion, perhaps? - it's not clear) libertarian happiness economists.

In any case, there are several interesting links from that piece - maybe worth making individual threads on if anyone finds them particularly interesting?

I particularly liked the study on quality of life in the Americas (PDF).

It's not just that it's hard to measure happiness. If you use it politically, people will tend to interpret it as what they're doing is making people happy.

DanielLC wrote:It's not just that it's hard to measure happiness. If you use it politically, people will tend to interpret it as what they're doing is making people happy.

That's true of any measure, including money - the incumbent govt takes credit for the economy if it's strong and gets blamed by the opposition if it's weak, and most of the time neither seems to be fair.

True, but with GDP, you can at least be fairly certain which of two countries produced more, and you have a pretty good idea of how a country's production is changing. It still has problems, but they're fairly small. If you try to measure happiness, you can practically get any answer. There is a huge number of different variables that correlate with happiness, but none of them correlate to it perfectly, and, by extension, they don't completely correlate to each other. Depending on how you weight those variables, you can get it to say anything you want. If unemployment is higher in one country, and crime is higher in another, which has happier citizens.

Of course, separately measuring GDP, unemployment, inflation, taxes, etc. and having people guess which is better isn't really all that different, but at least you know what they report, and allowing a large group of people to come to their own conclusions is probably better than using the conclusions of a few.

By the way, do they just guess? It occurs to me that taking a series of statistics about each country, finding the correlation between them and happiness reported in surveys, and weighting them accordingly would probably result in a fairly objective measurement. Of course, it's still less objective then the survey, and surveys aren't known for being objective, but I don't think there's a more objective way to measure happiness.

I believe it's done by surveys that usually use the 7-point scale, whatever it's called. If the same survey was used worldwide and from year to year, I don't think bias would be a huge issue, since the issue is relative happiness.

Calling the problems of measuring GDP small seems hopelessly optimistic to me. It completely ignores externalities - the thalidomide and asbestos sagas wouldn't have registered as a blip, for eg.

It also ignores the idea of sustainability - by GDP accounts we keep getting better until we crash (either financially as last year, or more seriously when we deplete key resources (peak oil being the most worrying eg) or when feedback effects from climate [url]change[/url] make the planet incapable of supporting existing industries.

Happiness economics also has this problem to some degree, but unlike GDP, happiness isn't required to rise every year as a measure of success, so it doesn't become a self-defeating benchmark if you extend it into the future.

Lastly, using GDP as a benchmark fails to account for a) the diminishing utility returns on higher incomes (I have a better, albeit unpublished, account by Avner Offer if you want me to email it) or b) the possibility of inequality decreasing happiness (source for in the same Offer piece, source against here - though I haven't read the latter yet).

DanielLC wrote:True, but with GDP, you can at least be fairly certain which of two countries produced more, and you have a pretty good idea of how a country's production is changing. It still has problems, but they're fairly small. If you try to measure happiness, you can practically get any answer. There is a huge number of different variables that correlate with happiness, but none of them correlate to it perfectly, and, by extension, they don't completely correlate to each other. Depending on how you weight those variables, you can get it to say anything you want. If unemployment is higher in one country, and crime is higher in another, which has happier citizens.

Of course, separately measuring GDP, unemployment, inflation, taxes, etc. and having people guess which is better isn't really all that different, but at least you know what they report, and allowing a large group of people to come to their own conclusions is probably better than using the conclusions of a few.

By the way, do they just guess? It occurs to me that taking a series of statistics about each country, finding the correlation between them and happiness reported in surveys, and weighting them accordingly would probably result in a fairly objective measurement. Of course, it's still less objective then the survey, and surveys aren't known for being objective, but I don't think there's a more objective way to measure happiness.

If it's done by surveys, I'm okay with it. If it's done by weighting a huge number of variables, I'm not. I can't tell what it is.

When I called the problems with measuring GDP small, I meant that it was reasonably objective. Not that it was good indicator of happiness. I just think it's better to have millions of people guessing at the happiness than one.

I don't mean that GDP alone should be used. I just mean that people should individually look at GDP, unemployment, healthcare, equality, surveys, etc. and make their own decision.