2. Energy Transitions in History
A brief history of energy transitions
An energy transition occurs when an economy switches from one main source of energy to another.5 There are many examples in history, and the phenomenon has been much studied (Fouquet, 2010; Smil, 2010; Sovacool, 2016; Cherif, Hasanov and Pande, 2017). In the past, energy transitions have tended to be largely national or occasionally regional developments. However, in today’s globalized world, the latest energy transition may well affect numerous nations simultaneously.
Two case studies
Two examples illustrate the essential elements of energy transitions: that seen in the US between 1865 and 1900, and the transition in France between 1973 and 1987.
United States, 1865–1900
Figure 1: Primary energy consumption by fuel/generation source in the US, 1860–2000
— Source: Darmstader, Teitelbaum and Polach, 1971; BP, 2018a.
Figure 1: Primary energy consumption by fuel/generation source in the US, 1860–2000
At the end of the American civil war in 1865, wood accounted for 80 per cent of primary energy consumption in the US and coal for 20 per cent. By 1900, coal accounted for 75 per cent and wood for around 20 per cent. The trigger for this change was a growing shortage of commercial wood6 as a result of the dramatic increase in demand. This was driven by rapid economic development – described by one source as ‘the greatest economic expansion in history’ – whereby ‘abundant wood was substituted for scarce labor’ (Maurice and Smithson, 1984: p. 47). In particular, rapid growth in wood use was due to the expansion of the railway system – from the 1870s to 1900, the US railways consumed as much as a quarter of timber production: ‘everything except the rails, spikes, car wheels and locomotives were made of wood’ (ibid.: p. 50).
The shift to coal began to produce many technological changes, designed in part to reduce wood use (through greater energy efficiency and wood preservation measures) and also to lower the cost of producing coal.
As the price of wood rose to reflect the growing shortages,7 consumers looked for alternatives, and in terms of an energy source, coal was the obvious choice. The shift to coal began to produce many technological changes, designed in part to reduce wood use (through greater energy efficiency and wood preservation measures) and also to lower the cost of producing coal. Thus the relative prices of wood and coal changed, with coal increasingly becoming the preferred option. Even as this change was happening, the pace of transition started to accelerate. The process was almost entirely driven by allowing market forces to operate with minimal interference from government.8
As Figure 1 illustrates, there was a second energy transition in the US from the early 1900s to 1970, when coal was gradually replaced by oil as the main source of primary energy. This was triggered by changes in vehicle manufacturing, specifically the development of the assembly line process pioneered by Henry Ford and his Model T.
France, 1973–87
The French transition is illustrated in Figure 2. We can see that the first oil shock of 1973–74 generated huge concerns in France about the security of energy supply. In response, the government made a strategic decision to increase the use of nuclear power in electricity generation, at the expense of oil,9 and to increase the role of electricity across society (leading to the increased use of electricity for heating). The emphasis was on developing a standard reactor design in an effort to lower the costs of building and operating nuclear plants.10
Figure 2: France’s primary energy consumption, 1965–2015
— Source: BP, 2018a.
Figure 2: France’s primary energy consumption, 1965–2015
The result was that oil was increasingly superseded by nuclear as a source of power, a process entirely driven by government intervention.
The lessons from history
From these case studies, and indeed many others, certain lessons can be learnt that are relevant to the current energy transition.
The first is that any transition has triggers. Once the trigger has been activated, various reinforcing factors come into play. These may be the result of markets working through adjustments in relative prices, government intervention, or a combination of the two. The process, once triggered, is very much driven by technological change, although other factors can play a key role. Fouquet (2010) identifies three major changes that have historically featured in the transition: changes to the supply network; changes to the energy source; and changes to the energy service provided. In all cases, the key to the transition has been the creation and delivery of better or different services. This has normally meant that the energy price has needed to fall, and/or that the technology’s efficiency has needed to improve. This has allowed ‘the diffusion of the technology and the new energy source through the broader market’ (ibid.: p. 6592). The transition is invariably complex, involving many different services and sectors.
Any transition has triggers. Once the trigger has been activated, various reinforcing factors come into play. These may be the result of markets working through adjustments in relative prices, government intervention, or a combination of the two.
The second lesson is that the time frame can vary enormously. For example, the transition in Great Britain in domestic heating – i.e. moving from wood to coal – took more than 200 years (Fouquet, 2010). In contrast, the French experience of moving to nuclear power, as described above, took only around 15 years. An illustration of the time dimension can be seen in the concept of ‘phases’ developed by Grubler, in which a core or innovation phase is followed by expansion to early adopters (‘the rim’), and then by another phase of uptake by late adopters (‘the periphery’) (Grubler, 2012). His research suggests that the duration of each phase is shortening. That said, ‘the historical record does seemingly support the mainstream view that energy transitions all take time’ (Sovacool, 2016: p. 205). However, in support of the main thesis of this paper, that the current transition will happen much faster than many realize, recent evidence and experience suggest that ‘some energy transitions can occur much more quickly than commonly believed’ (ibid.: p. 203).
A third lesson is that the current transition is also more complex than many originally believed. In the words of a recent study of the subject by the Saudi Arabia-based King Abdullah Petroleum Studies and Research Center (KAPSARC): ‘As awareness of the complexity of the energy transition increases, many scenarios are raising the number of possible pathways, a phenomenon that serves to reduce, rather than increase, clarity’ (KAPSARC, 2018: p. 4).
A fourth lesson is that the pattern of energy transitions has been one of moving from low-energy-density resources to higher-energy-density resources. It was always the energy density of oil that made it an attractive option, especially in the transport sector. This could raise doubts over the speed of the current transition, suggesting that it might proceed more slowly than the evidence in this paper indicates.
5 In the literature, the definitions of an energy transition are much more complicated and sophisticated. For example, for a discussion of definitions in the academic literature, see Sovacool, 2016. However, for the purposes of this paper – to stimulate discussion on the views of the ‘energy establishment’ – a simple definition will suffice.
6 Wood is a high-volume, low-value commodity. Therefore there are strict geographic limitations on the distances over which wood can be transported commercially.
7 Between 1896 and 1907, the price of crosstie stumpage (‘railway sleepers’ in English) increased by between 500 and 800 per cent (Maurice and Smithson, 1984).
8 Interestingly, the US Forest Service did provide advice that often was unhelpful since it only advocated supply-side solutions, i.e. increased planting and growing. It was left to the private users of wood, notably the railway companies, to try to develop wood-preserving techniques and improved designs that used less wood (Maurice and Smithson, 1984).
9 In the period leading up to the first oil shock, oil had been the fuel of choice for power generation, not least because as the 1960s proceeded oil prices were falling in real terms and were expected to continue to do so.
10 It has been suggested to the author that the reason the French were enthusiastic about nuclear power was because of the role that Marie Curie – a French national – played in researching radioactivity.