World Energy Outlook 2016 and the Rebound Effect

I've been asked to make some brief comments on the 2016 World Energy Outlook just published by the IEA at the ANU Energy Change Institute's 2016 Energy Update. It's a huge report, but I'll focus on the global projections for energy use and GHG emissions. I think that the IEA are still over-optimistic about the potential for energy intensity improvements and underestimate the future contribution of non-fossil energy. Under the "Current Policies" scenario they expect fossil fuels to have 79% of total energy in 2040 vs. 81% today. The current rapid growth of renewables under current policies makes me skeptical about that. The decline in world energy intensity is also more rapid than in recent decades.

Three main scenarios used throughout the report are summarized in the following Figure:

 

The "New Policies Scenario" includes policies from NDC's where the policy to implement the pledge appears to actually exist. The "450" Scenario is where policies that actually limit warming to 2 degrees C are implemented. Clearly, decarbonization is minimal under the current policies scenario and not that great under the new policies scenario. But the improvement in energy intensity is very large under all scenarios and does the vast majority of the work in reducing CO2 emissions. How plausible is this huge reduction in energy intensity? Here, I plot the historical global trend in energy intensity and the growth rates projected under the current and new policies scenarios:

The current policies scenario projects an increase in the rate of reduction in energy intensity relative to the 1990-2015 mean. This is possible, the rate of change might accelerate, but I am skeptical. Just looking at the data, we see that in the last few business cycles, energy intensity rose or fell slowly after recessions compared with later parts of boom periods. So, we seem likely to go through other cycles like these. Another issue is that the Chinese economy might have grown slower than the government admitted to in the recent couple of years. This would have exaggerated the global decline in energy intensity but probably not be a lot. The main reason, is that energy efficiency improvements do not translate one-for-one to reductions in energy intensity. The rebound effect, which we are researching in our ARC DP16 grant, means that improvements in energy efficiency lead to increases in the use of "energy services" - like heating, lighting, transport etc. which mean that energy use does not decrease as much as it would if all the efficiency improvement flowed through to energy consumption. At the micro-economic level this is simply because these energy services become cheaper as a result of the efficiency improvement. At the macro-level things are more complicated. I suspect that the IEA's model, which is driven by exogenous assumptions on things like the rate of economic growth, underestimates the economy-wide rebound effect.