Endogenous Price of Lighting Services

Another potential issue with the Tsao and Waide and Tsao et al. work on demand for lighting is that the price of lighting services is to some degree endogenous. If this is the case then estimates of a model of the sort suggested in my previous post will be subject to simultaneity bias. So why might the price be endogenous?

First, there are the usual reasons that we are not separating the effects of moves in the supply and demand curves here at all. Demand for lighting might affect the prices of the energy sources used to supply it and the prices of lamps. Lighting accounts for only a few percent of energy use so the effects on energy prices might not be that important. What seems to be a more fundamental problem here is that at many periods in time more than one lighting technology has been available. For example, at the moment there are incandescent bulbs, fluorescent lights, emerging solid state lighting etc. The efficiency with which energy is converted to light depends on the mix of these energy sources. Therefore, the cost of "lighting services" as defined by Tsao et al. depends on that mix of technologies. But the mix of technologies is chosen by consumers. That seems to be a problem I think?