Has fracking contributed to the decline in US carbon emissions?

Yesterday I came across an interesting but slightly perplexing paper in Nature Communications entitled 'Drivers of the US CO2 emissions 1997-2013'. The central claim of the paper seems to be that, contrary to some headlines, exploitation of shale gas from fracking has not been an important contributor to the recent decline in US carbon emissions:

Before 2007, rising emissions were primarily driven by economic growth. After 2007, decreasing emissions were largely a result of economic recession with changes in fuel mix (for example, substitution of natural gas for coal) playing a comparatively minor role.

(I say "seems", because on p. 4 they claim that "about half of the 2.1% decrease in emissions during 2011–2013 is related to changes in the fuel mix of the energy sector"). Incidentally, the reason why exploitation of shale could plausibly have contributed to a decline in US carbon emissions is that natural gas has much lower emissions per unit of energy than other fossil fuels, particularly coal.

The authors of the Nature Comm. paper carry out a rather complex decomposition analysis in which they partition recent changes in carbon emissions into contributions from six factors: population, energy intensity, consumption patterns, fuel mix, production structure, and consumption volume. (I'm not quite sure why they felt that such a complex analysis was necessary to make their point; but I'm of course well outside my area of expertise.)

In any case, I decided to conduct my own, simpler analysis using data from the Energy Information Administration. The chart below plots absolute change in energy consumption from different sources, along with absolute change in total energy consumption, for the period 2007-2014. Since 2007, there has been: a fall in total energy consumption, followed by a partial recovery since 2012; a protracted fall in consumption of coal and petroleum; a rise in consumption of natural gas since 2009; a smaller rise in consumption of wind and biomass; and little change in consumption of hydro, geothermal and solar. In fact, coal consumption has fallen by 4.8 quadrillion Btu, while natural gas consumption has risen by 3.9 quadrillion Btu. 

In order to examine how much the exploitation of shale gas (i.e., the rise in natural gas consumption) has contributed to the decline in US carbon emissions, I reconstructed changes in carbon emissions since 2007 under the assumption that there had been zero substitution of coal for natural gas. In other words, I assumed that natural gas consumption had remained constant, and that all the increase in energy consumption from natural gas (3.9 quadrillion Btu) had actually come from coal. To calculate carbon emissions in each year, I multiplied the quantity of energy consumed from each source by that source's life-cycle carbon emissions (taken from the IPCC), and then summed over all the sources. 

The chart below plots three series for total carbon emissions: actual emissions (taken from the EPA); reconstructed emissions; and reconstructed emissions, without shale gas (i.e., under the assumption of zero substitution). All three lines have a very similar profile to the thick black line for total energy consumption in the chart above, consistent with the Nature Comm. paper's finding: most of the decline in carbon emissions has been due to a fall in energy consumption caused by the recession. Comparison of the non-dashed blue and red lines reveals a remarkably good fit between actual emissions and reconstructed emissions (r = .99). Comparison of the non-dashed and dashed blue lines indicates that emissions are appreciably lower than they would have been if there hadn't been any rise in natural gas consumption. In fact, exploitation of shale gas accounts for just under a third (29%) of the difference in total carbon emissions between 2007 and 2014.

My reading of the evidence here is that most of the decline in US carbon emissions since 2007 is indeed attributable to a decline in total energy consumption caused by the recession, but that just under a third of it is due to exploitation of shale gas from fracking. As noted, I am well outside my area of expertise, so I would be interested to know if anyone with a better understanding of the area has any comments or criticisms.