Monday, March 06, 2006
Happiness and Choice
Given the ongoing theme of the relationship between more choice (or more income) and happiness, I feel it is past time that I discussed the basic process and limitations of such research.
Let's start with the obvious measurement problem. How do we quantify happiness? Typically, I ask you a question, "How happy are you?" or something like that (although I am sure someone has tried to approach this more cleverly). Does this measure how happy you are?
Not really. At best, it captures how happy you are right now. That is, it measures an emotional state (not necessarily your welfare) at one point in time. I would be surprised, however, if someone hasn't asked the same people how happy they are repeatedly (i.e., in a panel dataset) and looked at average happiness levels at several points in time. I would be further surprised if these averages were not at least somewhat correlated with individual point-estimates. So while one needs to keep measurement issues in mind, it is not obvious that there is bias in the measurment (just noise).
Whether or not we actually care about happiness (or some other outcome) is a deeper question that I haven't figured out the answer to myself, although I know a lot of economists do not believe that individuals try and maximize happiness (nor do they think that people should maximize happiness). Thus, it is not obvious how strongly we should care about happiness as an outcome.
A question that those unconvinved by happiness metrics frequently ask is if we took a large sample of supposedly happy poor people and offered them the opportunity to become rich, unhappy people (i.e., we told them that people who are rich are on average unhappier then they are currently) how many would take the offer? When given the choice to have more choice, people almost always reveal that they prefer more choice. If large numbers of people systematically choose to make themselves unhappy, can we really believe that they are actually worse off?
Further (assuming that concern about happiness makes sense), determining causal relationships between the variable of interest and the happiness outcome is essential. Without controlled experiments or natural experiments, establishing causal links between happiness and other things is very difficult, and you should not draw strong conclusions from research which lacks a credible identification strategy.
Finally, when presented with interesting relationships between choice or income and happiness, think hard about what to take away from the findings. If someone tells you that more money (or more choice) doesn't increase happiness on average, does this mean that you should not try to earn more money (or try and increase you choice set)? I am enough of an straight economist that it is hard for me to deny the logic that more choice necessarily improves your welfare. If considering some set of choices makes you worse off, I say figure out how to ignore them (but don't get rid of the choices because you may really regret not having them at some later point). Some people (i.e., the satisficers discussed in the previous post and elsewhere in the happiness literature) have apparently already figured out how to cope with choice. As such, others probably can learn. We may just need to help train people how to do this more successfully (although whether or not we can train ourselves or others to handle more choices without regret, etc. is an empirical issue that should be addressed (if it hasn't been addressed already)).
Barry Schwartz offers a few sensible guidelines in his article "The Tyranny of Choice":
1) Choose when to choose -- We can decide to restrict our options when the decision is not crucial. For example, make a rule to visit no more than two stores when shopping for clothing.
2) Learn to accept "good enough" -- Settle for a choice that meets your core requirements rather than searching for the elusive "best." Then stop thinking about it.
3) Don't worry about what you're missing -- Consciously limit how much you ponder the seemingly attractive features of options you reject. Teach yourself to focus on the positive
parts of the selection you make.
4) Control expectations -- "Don't expect too much, and you won't be disappointed" is a cliche. But that advice is sensible if you want to be more satisfied with life.
Let's start with the obvious measurement problem. How do we quantify happiness? Typically, I ask you a question, "How happy are you?" or something like that (although I am sure someone has tried to approach this more cleverly). Does this measure how happy you are?
Not really. At best, it captures how happy you are right now. That is, it measures an emotional state (not necessarily your welfare) at one point in time. I would be surprised, however, if someone hasn't asked the same people how happy they are repeatedly (i.e., in a panel dataset) and looked at average happiness levels at several points in time. I would be further surprised if these averages were not at least somewhat correlated with individual point-estimates. So while one needs to keep measurement issues in mind, it is not obvious that there is bias in the measurment (just noise).
Whether or not we actually care about happiness (or some other outcome) is a deeper question that I haven't figured out the answer to myself, although I know a lot of economists do not believe that individuals try and maximize happiness (nor do they think that people should maximize happiness). Thus, it is not obvious how strongly we should care about happiness as an outcome.
A question that those unconvinved by happiness metrics frequently ask is if we took a large sample of supposedly happy poor people and offered them the opportunity to become rich, unhappy people (i.e., we told them that people who are rich are on average unhappier then they are currently) how many would take the offer? When given the choice to have more choice, people almost always reveal that they prefer more choice. If large numbers of people systematically choose to make themselves unhappy, can we really believe that they are actually worse off?
Further (assuming that concern about happiness makes sense), determining causal relationships between the variable of interest and the happiness outcome is essential. Without controlled experiments or natural experiments, establishing causal links between happiness and other things is very difficult, and you should not draw strong conclusions from research which lacks a credible identification strategy.
Finally, when presented with interesting relationships between choice or income and happiness, think hard about what to take away from the findings. If someone tells you that more money (or more choice) doesn't increase happiness on average, does this mean that you should not try to earn more money (or try and increase you choice set)? I am enough of an straight economist that it is hard for me to deny the logic that more choice necessarily improves your welfare. If considering some set of choices makes you worse off, I say figure out how to ignore them (but don't get rid of the choices because you may really regret not having them at some later point). Some people (i.e., the satisficers discussed in the previous post and elsewhere in the happiness literature) have apparently already figured out how to cope with choice. As such, others probably can learn. We may just need to help train people how to do this more successfully (although whether or not we can train ourselves or others to handle more choices without regret, etc. is an empirical issue that should be addressed (if it hasn't been addressed already)).
Barry Schwartz offers a few sensible guidelines in his article "The Tyranny of Choice":
1) Choose when to choose -- We can decide to restrict our options when the decision is not crucial. For example, make a rule to visit no more than two stores when shopping for clothing.
2) Learn to accept "good enough" -- Settle for a choice that meets your core requirements rather than searching for the elusive "best." Then stop thinking about it.
3) Don't worry about what you're missing -- Consciously limit how much you ponder the seemingly attractive features of options you reject. Teach yourself to focus on the positive
parts of the selection you make.
4) Control expectations -- "Don't expect too much, and you won't be disappointed" is a cliche. But that advice is sensible if you want to be more satisfied with life.
// posted by BW @ 11:34 PM 
Subscribe to Posts [Atom]