Mark D. White
In this morning's Wall Street Journal, James Bovard pokes a little fun at the US government's plans for measuring gross domesic happiness (of which Nicolas Sarkozy was a leading advocate), pointing to how well they currently measure the myriad economic statistics regarding things that aren't entirely subjective. Many economists take this very seriously, however; as it happens, I'm currently working on several projects that, to some extent, deal with this issue (as is Deirdre McCloskey, if I remember correctly). The literature is recent but already vast: a marvelous summary and critique can be found in Daniel Hausman's 2010 Economics and Philosophy article, "Hedonism and Welfare Economics."
My take, in a nutshell, is that measuring happiness is neither feasible nor desirable. It is not feasible because happiness is irreconciliably multifaceted (along several dimensions) and inescapably subjective. It is not desirable because any official focus on happiness violates ideals of liberal neutrality and personal autonomy regarding persons' individual pursuit of the "good life," and the measurement of such can only lead to excessive government manipulation (if not paternalism) towards that imposed end (as Bovard describes). Rather, institutions should be established and maintained to ensure that person have the maximal capacity to make choices in pursuit of their own interests consistent with all others doing the same. Only such a system can ensure respect for persons' own interests and the choices they make towards them.