I've always had my suspicions about the way that energy efficiency is presented as an easy way of reducing carbon emissions. For a start there's the Jevons paradox: the observation that efficiency gains tend to lead consumers towards enhanced performance. In other words, as houses become more efficient we tend to keep them much warmer. Being someone who lives in a cool (or even cold house) and wears jumpers all the time, I find modern houses stiflingly hot, but most people are much happier to wear shorts and t-shirts indoors.
But even leaving this kind of thing aside, whenever I have done the sums on my own house I've always come to the conclusion that investment in energy efficiency is not going to provide a good return. It's therefore interesting to see that my back-of-a-fag-packet calculations seem valid across the board. A new, and by the looks of it carefully controlled study of homes in the USA has found that the much touted gains from energy efficiency measures are actually relatively small and certainly less than the cost of installation