The Institute for Public Policy Research (IPPR) has got together with some green energy consultants to discuss whether wind energy is a Good Thing. Given the nature of the co-authors, I'm sure you can guess their conclusions.
The authors propose what they call a steady state model of the electricity grid.
Adding wind energy to electricity supply without altering demand will displace or push out an equivalent amount of supply from the marginal plant.
Having established a simple model of the electricity system, in order to estimate the carbon impact of adding wind energy we need to first establish which generation type is the ‘marginal plant’ and secondly how much CO2 it emits. In the UK there are two fossil-fuel candidates for the role of the most common ‘marginal plant’ – coal and gas.
Now this is a bit sneaky. The authors start by discussing the "marginal plant" and then move on to discussing not the candidates for that marginal plant but the "fossil fuel candidates". How did that pesky modifier make it into the sentence. As Gordon Hughes points out in his report "the key margin in the UK is
between wind and nuclear power".
If you make nuclear uneconomic by replacing part of its output with subsised wind power, you will lose all of its output and will have to make up the shortfall with inefficient open-cycle gas generation. You have replaced completely carbon-free generation with a mixture of carbon-free and carbon-intense generation. So depending on how much wind you have in the system and how the rest of your electricity is generated it is certainly possible to increase your carbon emissions.
To their credit the authors mention the system effects in their report, although it is left out of the conclusions, which are based solely on the naive "steady state" model. Their justification for this presumably relates to a draft paper they cite which found that carbon emissions have been reduced on US electricity grids. Given what we have said about the importance of the relative proportions of different generation types in the system, it's not clear to me that the US experience translates to the UK, but the paper is certainly worth a look.
Colin McInnes analyses the IPPR report (H/T Omnologos) and finds that the implied cost of saving carbon through wind power is £130/tonne, which makes it absurdly expensive.