Brought to popular attention by Enkvist at McKinsey & Company, marginal abatement cost curves (MACCs or MAC curves) are a way of showing both the cost of saving carbon, and also the size of the potential saving from that source. I’ve mentioned them before in a post about low hanging fruit offered by refurbishment. They […]
I read recently on the Green Building Forum the counterintuitive claim that in the UK richer people use the least energy. I’m always interested in the relationship between income and environmental impact – and this is not the way I’d expect the relationship to work. As people have more disposable income, they spend more. They have bigger houses. They are less likely to share those houses. This is also supported by the REAP data which I used on a report for WWF-UK in which more affluent UK cities tended to have higher ecological footprints.
But the claim is apparently based on data from DECC (see the graph) so I thought it deserved a look at least. I don’t want to dismiss the idea out of hand so I went off to see what data was out there to test the theory.
There are of course counter-arguments to the standard view. Perhaps better-off households have more disposable income to invest in energy efficiency measures. Perhaps less well-off households are more likely to be renting and so have little control over how energy efficient their home is.
The data I was looking for is collated by the ONS at the level of something called a Middle Layer Super Output Area or MLSOA. There are figures available for electricity and gas use by each household (or each meter to be specific) . Each MLSOA contains in the region of 3,000 houses and 7,000 occupants.
Income data was harder but I found a file online, although I could only find it for 2004 .
I chose London as a test bed as I thought there would be a reasonable range of incomes while keeping the spreadsheet to a manageable size. As it turns out there are only two MLSOAs in the bottom range of income (<£336 weekly disposable household income before housing costs)
There were a few disparities between the number of meters and the number of households. This could be because of unmetered communal heating systems where one meter serves a number of homes. It could also be affected by new homes and/or demolitions in between data collection. In any case the total level of disagreement was fairly low, and could be expected to be randomly distributed.
Domestic gas and electricity use in London
Carbon emissions in London
So taking London as our first example we can see that in an urban setting domestic energy use is strongly correlated with income. Transport may be a different matter but unfortunately there isn’t data to test this at the same level of disaggregation.
I chose London as a test bed as I thought there would be a reasonable range of incomes while keeping the spreadsheet to a manageable size. As it turns out there are only two MLSOAs in the bottom range of income (<£336 weekly disposable household income before housing costs). I’ll expand this to cover a larger distribution of MLSOA types – rural as well as urban – in the future.
I’d love to track down the data source for SteamyTea’s graph above to see where the disagreement comes from. But I’m not about to pay £50 to register for the Green Building Forum just for that. If any GBF members are reading this and fancy asking I’d be very grateful. Edited: Thanks to Nick in the comments correcting my dodgy eyesight I’ve now stumped up my £5 (not £50) to join the Green Building Forum so will try and get hold of that data from SteamyTea.
Income data at MLSOA level (2004)