There's two ways of looking at neo-classical economics. The first is as a coherent series of stories and assumptions about the way that the world works. I certainly hold that view. However, there's another, looser, sense which many more people would agree with. People like Dani Rodrik for example might not agree that those stories and assumptions explain the world (or at times that they are coherent) but neo-classical economics still provides the economist with a toolbox with which to explore the value of certain policies. In this view the value is in the methods rather than the assumptions.
The value of this can be seen in the current discussion of daylight savings time. It's long been an assumption that the move saves energy. Thus most governments promote it and there's even a move here in the UK to move to double summer time so as to save even more energy. The question we would really like the answer to though is whether summer time (or daylight savings time, same difference) does in fact save energy? Just because governments have been assuming it does for a long time doesn't make it so.
Which is what has just been done using the economists' toolbox: an empirical study of whether advancing the clocks saves energy. Something which is actually a little more difficult than you might think. We can't compare a year with savings with a year without as cloud cover and temperature could have varied making accurate comparison impossible. We can't compare areas with and those without for the same reasons. But Indiana had a strange system whereby some counties daylight saved and others did not and then it became statewide that they did. So we can compare the same places with and without and the same (almost) geographic areas with and without. The result?
A final component of our analysis is the calculation of costs associated with the estimated effect of DST. We find that the policy costs Indiana households an average of $8.6 million per year in increased electricity bills. We also estimate social costs of increased pollution emissions due to the residential response to be between $1.6 and $5.3 million per year.
DST leads to more energy use, not less. So I guess that idea of double DST as a way to curb climate change isn't going to work either then?