A Reply to Michael Grubb’s Growth-Decarbonization Optimism from Semieniuk et al

Hope for mitigating climate catastrophe may not be lost, but the scale of political change needed is no cause for optimism
In this piece, Gregor Semieniuk, Lance Taylor, and Armon Rezai offer a reply to Michael Grubb’s response to their paper.

We would like to thank Professor Michael Grubb for his comments on our paper in which he raises important points about logistic trajectories of technological change in the energy sector. This very much complements our discussion of the importance of mitigation via lowering the carbon intensity of energy. We agree that a fast replacement of fossil fuels with low-carbon energy sources must be a key plank in any mitigation strategy as opposed to relying too much on reductions in energy demand, and welcome his call for more research on “historical futures.” Yet, we also believe that it is very difficult for a change in the energy mix—from one dominated by fossil fuels to one primarily relying on low-carbon source—to do the heavy lifting in reducing CO2 emissions, while labor productivity grows exponentially. Grubb’s “conditional optimism” about mitigating climate change aims to reassure about managing future impacts of global warming. Besides the reasons for hope he advances, there are others. Unfortunately, all have their limits. To see why, it makes sense to review five points raised in our note.

First, humanity embarked on its contemporary energy use path three centuries ago. Now there is more or less direct proportionality between per capita output and energy use, locked into technology and embedded in social and economic relations (just think, for example, of the entrenched market power of coal and oil companies, grid-tied electric utilities, and the automobile industry). The energy mostly comes from fossil fuels which were used initially because of their high energy density—and hence low cost—in comparison to traditional sources.

Second, alternative technologies such as solar and wind power have only recently begun to emerge. As shown in our Table 1, the IPCC report presents modest projections for the use of non-fossil energy. To discuss this issue, Professor Grubb invokes the logistic curve which has been used since the mid-19th century to illustrate possible growth paths of population, the spread of species, economic phenomena such as the adoption of hybrid corn, etc.[1] He chooses a plausible value of 0.25 for the growth parameter to illustrate adoption of new energy technologies in his Figures 1 and 2. The problem is that with k = 0.25 their penetration reaches only 20% in 20 years, too low to offset the climate catastrophe envisaged by the IPCC.[2] The starting years in the diagrams vary, but the key point is that non-hydro renewables now provide only about two percent of energy worldwide. His logistic curves strongly suggest that it will take decades to arrive at a level of 20%.[3]

A final observation is that bringing in endogenous growth à la Paul Romer to resolve global warming is a dead end. This theory was an empirically unverifiable fad among mainstream economists starting in the 1980s. Interest in it expired before the turn of this century. Induced technological innovation driven by shifting costs (for example due to a carbon tax) is a far more relevant process. It undoubtedly played a role in setting up fossil fuel dominance in the first place. See Foley and Michl (1999), chapter 15.

Our third point is that fossil fuel technology is the driving force behind a feedback loop between an increase in CO2 emissions resulting from higher output on the one hand, and a bigger atmospheric stock of greenhouse gas cutting into output growth on the other. Severing this loop is essential to avoid a climate crisis. The only two ways to do so are to reduce the use of fossil fuels as sources of energy, or reduce the amount of energy needed to provide output.

Fourth, as already noted, the first option is not likely to be strong enough to offset a climate crisis in the near future. Our Figure 1 shows on historical grounds that the IPCC is unduly optimistic about prospects for slashing energy use. Simply put, the growth rate of labor productivity tends to be at least as big as growth of energy productivity, forcing the ratios of energy use and—so long as fossil fuels dominate the energy mix—CO2 emissions to population to be stable or to rise. For the next few decades, the IPCC unrealistically postulates that they will fall.

Finally, the only possible way to stave off a crisis is through immediate extensive mitigation (including more use of non-fossil energy sources as well as afforestation, carbon capture, insulation, replacement of internal combustion engines, and many other measures) to reduce net CO2 emissions. Hope may triumph because our Figure 3 shows that the required outlays are within the macroeconomic realm of possibility. But they will require widespread political backing to be put into place.


Foley, Duncan K., and Thomas R. Michl, 1999, Growth and Distribution, Cambridge MA: Harvard University Press

Hirsch, Morris W., Stephen Smale, and Robert L. Devaney, 2013, Differential Equations, Dynamical Systems, and an Introduction to Chaos (3rd edition), Waltham MA: Academic Press

IEA, 2018, World Energy Balances 2018 Overview, IEA, Paris.

Grubler, Arnulf, Charlie Wilson, and Gregory Nemet, 2016, Apples, oranges, and consistent comparisons of the temporal dynamics of energy transitions. Energy Research and Social Science 22, 18–25.

Semieniuk Gregor, and Mariana Mazzucato, 2018 Financing Green Growth. SOAS Department of Economics Working Paper No. 208 [Forthcoming in Fouquet, Roger 2019, Handbook of Green Growth, Cheltenham: Elgar]

Smil, Vaclav, 2016, Examining energy transitions: A dozen insights based on performance, Energy Research and Social Science 22, 194–197.

[1] Hirsch, Smale, and Devaney (2013) give a succinct summary of the differential equation mathematics which generates the logistic. If the relevant variable is initially small, its increase is proportional to size and its exponential growth rate is close to the growth parameter (k in Grubb’s notation). Its growth falls to zero at his MAX level.

[2] And to focus with Grubb on recent history, there is worryingly not a single year since the financial crisis when global primary fossil energy demand, coal plus gas plus crude oil, fell (IEA 2018).

[3] That energy supply transitions, especially at the global level, not just in leading ‘core’ countries take long is also well established in innovation literature (Grubler et al. 2016, Smil 2016). And even national transitions, heavily pushed and financed by governments have taken decades to unfold in the past (Semieniuk and Mazzucato 2018).

Share your perspective