Energy systems are subject to strong and long-lived path dependence, owing to technological, infrastructural, institutional and behavioural lock-ins. Yet, with the prospect of providing accessible cheap energy to stimulate economic development and reduce poverty, governments often invest in large engineering projects and subsidy policies. Here, I argue that while these may achieve their objectives, they risk locking their economies onto energy-intensive pathways. Thus, particularly when economies are industrializing, and their energy systems are being transformed and are not yet fully locked-in, policymakers should take care before directing their economies onto energy-intensive pathways that are likely to be detrimental to their long-run prosperity.
In the late 1980s, economists were offered a theoretical explanation for why markets can fail to move towards the socially optimal outcome, even in the long run1. Building on the classic example of the QWERTY keyboard and other case studies2, such as the dominance of inferior VHS videotapes, an explanation was given for how increasing economies of scale, as well as learning and network effects, could lead an economy (or system, more generally) to be faced with multiple potential outcomes, and how the eventual outcome depended on circumstances in the early history of particular technologies. In other words, history mattered, and, if the ‘wrong’ path was followed, the economy could be stuck in a socially sub-optimal outcome.
Shortly afterwards, an example was given of path dependence within an energy system that showed how water-pressurized nuclear reactors became the dominant technology through a series of historical coincidences3. Then, studies began to outline the implication of technological lock-ins for addressing climate change4–6. After a quieter decade on this front, the recent explosion in long-run and historical research has offered a number of new examples of path dependence.
A better understanding of path dependence in energy systems is critical and urgent for two reasons. First, current efforts to stabilize the climate require unlocking industrialized economies from their existing fossil-fuel energy systems6. Second, the period of global economic growth in the late 1990s and early 2000s was associated with a wave of industrialization in a number of developing economies. Industrialization is a time of heavy investment, creating lock-ins and ultimately path dependence. Thus, lock-ins and path dependence urgently need to be better understood, their implications identified, and strategies to deal with them formulated.
Given that the role of path dependence in unlocking from the current fossil-fuel-based system is receiving new attention7–9, the purpose of this Perspective is to pull together examples related to energy systems, and explore their interlinkages with economic development and the energy intensity of the economy, drawing attention to the potential burdens from locking-into an energy-intensive economy. My aim is to connect the micro-studies of path dependence with the macro-level implications. Because of the scale of the topic, this Perspective can offer only a few examples from a cross-national and historical viewpoint, in the hope of stimulating further research and debate.
Energy demand and economic development
Access to cheap energy is seen to be fundamental for economic development and for reducing poverty — especially with more than one billion people globally currently without access to electricity10,11. In parallel, the expansion of an energy-related physical infrastructure has frequently been critical to the provision of abundant cheap energy12. Thus, there tends to be a positive feedback between energy resources, infrastructure and industrial development, locking an economy into specific consumption patterns13,14,15.
Long-run demand for mobility (and its associated energy consumption) at a given level of per capita income is heavily determined by the urban and national transport network16,17. The expansion of transportation infrastructure is typically seen to generate additional ‘induced demand’, principally by initially lowering the cost of travel owing to shorter travel times18. It is associated with urban sprawl and often used as an argument against expanding road infrastructure. Certainly, in the United States, increased provision of roads appears unlikely to do much to relieve congestion, although it does increase total amount of travel19. Thus, all things being equal, greater transport infrastructure is likely to increase the energy intensity of the economy and lock it into a pathway of higher energy intensity20.
A form of induced demand is also likely to occur in other energy-related engineering projects, such as large-scale hydroelectric dams and nuclear power stations, which create sudden steep increases in power supply. Such energy projects can also leave long-term legacies of local poverty, distributional inequality and environmental damage (Box 1). Once completed, however, these megaprojects do tend to offer low marginal costs of energy production. For the economy's development, reducing the constraints on energy use and the associated prices is clearly seen as desirable. This drives up energy consumption, putting the economy on a new (if energy-intensive) path. Commentators have argued that mega-projects are often implemented to transform society in ambitious ways21,22.
Yet, many23–25 recommend against developing economies investing in energy mega-projects, because their costs are consistently under-estimated and, in many cases, they do not come out positively in cost–benefit analyses. Moreover, conventional cost–benefit analyses struggle to quantify the unintended and nonlinear costs and benefits of such projects21. Furthermore, the lure of cheap and abundant power can seduce economies into energy-intensive behaviour that eventually makes them vulnerable to energy shortages26,27 (see below).
A crucial point is that the role of energy services in economic development is likely to change at different phases of economic development28. For instance, during the Industrial Revolution, the influence of declining energy service prices on economic growth appears to have changed greatly over time29. Increases in energy use and improvements in energy efficiency were key sources of economic growth in the nineteenth and early twentieth centuries, but not in the second half of the twentieth century30. Thus, one lesson learned might be that, if timed and managed correctly (for example at the right phase of economic development, and tied in with policies promoting structural transformation — as occurred in South Korea in the 1960s and 1970s31), judicious infrastructure projects reducing energy prices can help kick-start the economy, but if at the wrong time or poorly managed, they will only feed through into inefficient energy consumption and large debts.
Locking-into energy-intensive systems
Certainly, there are signs of energy-intensive economic development today. Given that the efficiencies of energy technologies have improved enormously over the past 200 years32 and have played a role in declining energy intensities (that is, the ratio of energy use to gross domestic product, GDP)33,34, one study set out to identify whether currently developing economies are less energy-intensive than present-day OECD (Organisation for Economic Co-operation and Development) countries were, when they were at similar levels of economic development35. It identified three factors influencing their energy intensities: more efficient technologies today; more exporting in developing economies today; and more energy-intensive goods and services being consumed today. The first of these factors would drive down energy intensity, while the other two increase it. The study found evidence of increased energy intensity overall, arguing that these two latter factors have outweighed the first factor35. In other words, today's developing economies appear to be following energy-intensive pathways11,36, potentially associated with technological, infrastructural, behavioural and institutional lock-ins.
Although infrastructure is arguably the most powerful and long-lasting lock-in, defining the geography of a country and the behaviour of an economy for centuries and even millennia37,38, the most commonly referred-to lock-ins relate to technologies (Box 2). But evidence is also emerging of behavioural lock-ins associated with energy production and consumption. One study finds that path dependence (that is, persistent behaviour 60 years after conditions changed) is responsible for 60% of total coal-fired power station capacity in certain counties in the United States39. Two other examples of path dependence show that the proximity to nineteenth-century coal mines in the United States and the United Kingdom has been associated with less-developed entrepreneurial cultures today40,41. Another study indicates that temporary rationing policies can have long-term effects on behaviour42. In that work, extreme electricity shortages in the Brazilian southeast due to low rainfall were shown to force regional authorities to introduce strong demand-side management programmes that altered habits, evident 10 years later. In other words, factors (including policies) can drive consumption down, as well as up.
Often, policies can be influenced by the institutional and market structure, which becomes the source of the path dependence. Although more research is needed43, there is likely to be a correlation between the size of corporations, market power and energy system lock-ins. Large companies and more concentrated industries will have greater financial wealth and will be better coordinated to lobby governments (also known as regulatory capture or rent-seeking) to protect a particular energy system44. Certainly, where nuclear power is dominant, and the electricity industry is both highly concentrated and connected with related policy decisions, such as in France, it is harder to move towards liberalized and competitive markets and potentially different energy systems45.
Even more evident was the dominance of energy-intensive companies in the US$1.5 billion spent on lobbying associated with the failed US climate policy known as the Waxman–Markey Bill46, which would have aimed to cap greenhouse gas emissions. In 2014, eight of the top ten largest companies in the world (as measured by sales revenue) were oil or car companies47. In other words, there is (and has been for a long time) considerable financial and political power to support the fossil-fuel status quo48. More generally, the market power of energy companies can heavily influence the degree of lock-in of a particular energy system.
Subsidies, which are lobbied for by energy companies, play a critical role in placing economies on energy-intensive pathways — although they are often introduced to boost production and employment on the supply-side and reduce fuel poverty on the demand-side. As shown in Fig. 1, there is a close positive relationship between per capita subsidies and per capita consumption of petroleum, natural gas and coal (for more than 50 energy-producing developing and industrialized economies) — although causality certainly cannot be attributed, because of the complexity of disentangling the interaction between economic development, production, consumption and energy prices. Thus, the existence of US$4.6 trillion of global fossil-fuel subsidies (including the external costs of energy production, distribution and consumption49) in 2013 (or 6% of global GDP) is associated with higher per capita consumption, and may be linked to lock-ins favouring energy-intensive production and consumption. Thus, the full impact of removing subsidies will probably take many decades (if not centuries) to change.
As an example, in 2013, post-tax subsidies in the United States amounted to US$350 billion for petroleum products, US$78 billion for natural gas and US$178 billion for coal: thus, US$606 billion of subsidies on fossil fuels, equivalent to 3.75% of the GDP of the United States49. These subsidies have undoubtedly been in place for a long time (at least 100 years for the fossil-fuel industry in the United States50), locking the economy into an even higher level of energy intensity than would exist without subsidies. It has been argued that American policies have increased the US economy's energy intensity and done so at a high cost to the economy and society51.
The burden of locking-into energy-intensive systems
Most of the lock-ins mentioned have forced economies onto more energy-intensive pathways than might have occurred in ‘socially optimal’ market conditions. This implies that, if circumstances change, consumption patterns will fail to adjust fully for a very long time. Certainly, economies with higher energy intensities are more vulnerable to the impacts of an oil shock52. Although rising energy prices can stimulate improvements in energy efficiency53,54, these improvements are slow to be adopted55, and an ‘efficiency gap’ exists between the most efficient technology and what consumers use56. The inability to adjust in the long run, in part due to path dependence, creates a major vulnerability.
This long-term lock-in-induced vulnerability to energy price shocks implies market failures and potential costs to the economy and society. For instance, the disruption component of the social cost of oil in 2004 ranged from US$2 to US$8 per barrel of oil consumed by the United States and highlights the benefits to the American economy and society from reducing imports of oil57. As a result of these types of disruptions, many governments have sought to develop energy security policies.
Energy security policies can work on supply-side or demand-side58. In the United States, over the past 30 years, the annual costs of demand-side management projects have been between 0.01% and 0.04% of GDP59,60. Although there has been some debate about the estimated benefits of demand-side management projects61,62, they have achieved reductions in vulnerability to rising energy prices59. Nonetheless, despite these benefits, demand-side management efforts in the United States peaked in the early 1990s59, then remained low until 200860. Thus, such policies appear to be at their lowest when energy price hikes occur (such as in 1973, 1979 and 2008) and so tend to be reactive, rather than proactive. Furthermore, these policies rarely address the underlying energy system, particularly related to key infrastructure, and focus more on incremental improvements in the efficiency of energy technologies. Thus, demand-side management offers little real opportunity to place the economy on a less energy-intensive pathway.
Rather than using resources more efficiently, the history of economic development has tended to be based on dealing with resource scarcity by opening up new frontiers or exploiting new reserves63. In turn, supply-side energy security policies aggravate energy-intensive lock-ins. Naturally, some countries have been more aggressive in their supply-side energy policies than others. As an example, the military expenditure by the United States to ensure oil supplies from the Persian Gulf has been estimated. In 2004, the price tag for oil consumers, US oil companies and world oil-price stability was estimated to be between 0.2% and 0.6% of GDP64 — and this estimate has now been revised upwards by 300–600%65,66. But this example is not an isolated one; a study67 looking over more than 60 years and 600 conflicts found that, when a country has oil reserves very near the border, the probability of conflict is three times greater than if neither country has oil near the border (although strategic objectives on the production-side and associated revenue are likely to be a key driver of this finding, it is hard to exclude the influence of national objectives to meet energy demands). In other words, efforts to ensure the security of supply of oil and energy more generally — arguably to counteract the market failures due to path dependence in energy systems — have imposed substantial burdens on economies and societies.
The purpose of this Perspective is to discuss the implications of path dependence in energy systems for economic development, and stimulate further research and debate. The discussion should not discourage governments from seeking to use energy policies to assist objectives of economic development and poverty alleviation. Instead, it offers a reminder to policymakers that cheap energy is not a ‘silver bullet’ and can instead have a hidden price tag, especially in the long run.
Economic development needs access to cheap energy services, just as it needs cheap capital and labour. Infrastructure and other large engineering projects, as well as subsidies, can help to provide energy service access at low prices. If well directed, such policies may boost economic development and reduce poverty. Thus, these may be socially desirable, particularly at early phases of economic development.
However, the role of energy services in economic development changes at different income levels. For example, the potential large net benefits of such policies at early phases of industrialization may become net costs on the economy at later phases. So, care should be taken before embarking on large-scale projects and policies that leave an economy heavily in debt and offer little growth and development. Furthermore, cheap energy tends to lock economies into energy-intensive patterns (related to technologies, infrastructures, institutions and behaviour) that are likely to be detrimental to the prosperity of the economy in the long run, increasing the economy's vulnerability to energy price shocks, inflation, trade balance deficits, political pressures from energy companies and environmental pollution.
Once an economy is locked-into an energy system, the government rarely has opportunities to redirect it (Box 3). Thus, when an economy is industrializing, and its energy system is being formed (or transformed) and not yet fully locked-in, it is crucial for its long-run prosperity that an economy gets on the ‘right’ path. In addition, this is likely to reduce the costs of meeting global environmental regulation that will, no doubt, eventually be pressed on even the least developed economies. Indeed, the December 2015 agreement in Paris suggests that all economies will eventually need to unlock themselves from the fossil-fuel energy system and, therefore, that industrializing economies may want to avoid locking themselves into this antiquated energy system altogether.
Arthur, W. B. Competing technologies, increasing returns, and lock-in by historical events. Econ. J. 99, 116–131 (1989).
David, P. A. Clio and the economics of QWERTY. Am. Econ. Rev. 75, 332–337 (1985).
Cowan, R. Nuclear power reactors: a study in technological lock-in. J. Econ. Hist. 50, 541–567 (1990).
Cowan, R. & Hultén, S. Escaping lock-in: the case of the electric vehicle. Technol. Forecast. Social Change 53, 61–80 (1996).
Unruh, G. C. Understanding carbon lock-in. Energy Policy 28, 817–830 (2000).
Unruh, G. C. Escaping carbon lock-in. Energy Policy 30, 317–325 (2002).
Foxon, T. J., Pearson, P. J. G., Arapostathis, S., Carlsson-Hyslop, A. & Thornton, J. Branching points for transition pathways: assessing responses of actors to challenges on pathways to a low carbon future. Energy Policy 52, 146–158 (2013).
Acemoglu, D., Aghion, P., Bursztyn, L. & Hemous, D. The environment and directed technical change. Am. Econ. Rev. 102, 131–166 (2012).
Acemoglu, D., Akcigit, U., Hanley, D. & Kerr, W. Transition to clean technology. J. Polit. Econ. 124, 52–104 (2016).
Wolfram, C., Shelef, O. & Gertler, P. How will energy demand develop in the developing world? J. Econ. Persp. 26, 119–138 (2012).
IIASA Global Energy Assessment: Toward a Sustainable Future (eds Johansson, T. B. et al.) (Cambridge Univ. Press, 2012).
Donaldson, D. & Hornbeck, R. Railroads and American economic growth: a ‘market access’ approach. Q. J. Econ. 131, 799–858 (2016).
Hughes, T. P. Networks of Power: Electrification in Western Society, 1880–1930 (Johns Hopkins Univ. Press, 1983).
Wright, G. The origins of American industrial success, 1879–1940. Am. Econ. Rev. 80, 651–668 (1990).
Fernihough, A. & O'Rourke, K. H. Coal and the European Industrial Revolution NBER Working Paper 19802 (National Bureau of Economic Research, 2014).
Grubler, A. The Rise and Fall of Infrastructures: Dynamics of Evolution and Technological Change in Transport (Physica, 1990).
Goodwin, P. B. Empirical evidence on induced traffic, a review and synthesis. Transportation 23, 35–54 (1996).
Hymel, K., Small, K. & van Dender, K. Induced demand and rebound effects in road transport. Transport. Res. B 44, 1220–1241 (2010).
Duranton, G. & Turner, M. A. The fundamental law of road congestion: evidence from US cities. Am. Econ. Rev. 101, 2616–2652 (2011).
Nye, D. E. Consuming Power: A Social History of American Energies (MIT Press, 1998).
Hirschman, A. O. Development Projects Observed (Brookings Inst., 1967).
Flyvbjerg, B. What you should know about megaprojects and why: an overview. Project Managem. J. 45, 6–19 (2014).
Ansar, A. Flyvbjerg, B., Budzier, A. & Lunn, D. Should we build more large dams? The actual costs of hydropower mega project development. Energy Policy 69, 43–56 (2014).
Grubler, A. The costs of the French nuclear scale-up: a case of negative learning by doing. Energy Policy 38, 5174–5188 (2010).
Sovacool, B. K. & Cooper, C. J. The Governance of Energy Megaprojects: Politics, Hubris, and Energy Security (Edward Elgar, 2013).
Thomas, J. J. Kerala's industrial backwardness: a case of path dependence in industrialization? World Dev. 33, 763–783 (2005).
Kitchens, C. The role of publicly provided electricity in economic development: the experience of the Tennessee Valley Authority, 1929–1955. J. Econ. Hist. 74, 389–419 (2014).
Toman, M. T. & Jemelkova, B. Energy and economic development: an assessment of the state of knowledge. Energy J. 24, 93–112 (2003).
Fouquet, R. The role of energy technologies in long run economic growth. IAEE Energy Forum 8, 11–13 (2014).
Stern, D. I. & Kander, A. The role of energy in the Industrial Revolution and modern economic growth. Energy J. 33, 127–54 (2012).
Pearson, P. J. G. Energy transitions. New Palgrave Dictionary of Economics online edition (eds Durlauf, S. & Blume, L. ) (Palgrave Macmillan, 2016).
Fouquet, R. Divergences in long run trends in the prices of energy and energy services. Rev. Environ. Econ. Policy 5, 196–218 (2011).
Sue Wing, I. Explaining the declining energy intensity of the U.S. economy. Resour. Energy Econ. 30, 21–49 (2008).
Csereklyei, Z., Rubio Varas, M. d. M. & Stern, D. I. Energy and economic growth: the stylized facts. Energy J. 37, 223–255 (2016).
van Benthem, A. Energy leapfrogging. J. Assoc. Environ. Resour. Econ. 2, 93–132 (2015).
Stern, D. I. Modeling international trends in energy efficiency. Energy Econ. 34, 2200–2208 (2012).
Bleakley, H. & Lin, J. Portage and path dependence. Q. J. Econ. 127, 587–644 (2012).
Michaels, G. & Rauch, F. Resetting the Urban Network: 117–2012 Economics Series Working Papers 684 (Univ. Oxford Department of Economics, 2013).
Meng, K. C. Path dependence in U.S. coal-fired electricity. Am. Econ. Assoc. Annu. Meeting (4 January 2016).
Glaeser, E. L., Kerr, S. P. & Kerr, W. R. Entrepreneurship and urban growth: empirical assessment with historical mines. Rev. Econ. Stat. 97, 498–520 (2015).
Stuetzer, M. et al. Industry structure, entrepreneurship, and culture: an empirical analysis using historical coalfields. Eur. Econ. Rev. 86, 52–72 (2016).
Gerard, F. What Changes Energy Consumption, and for How Long? New Evidence from the 2001 Brazilian Electricity Crisis RFF Discussion Paper 13–06 (Resources for the Future, 2013).
Pezzey, J. C. V. The influence of lobbying on climate policies; or, why the world might fail. Environ. Dev. Econ. 19, 329–332 (2014).
Laffont, J. J. & Tirole, J. A Theory of Incentives in Procurement and Regulation (MIT Press, 1993).
Glachant, J. M. & Finon, D. A competitive fringe in the shadow of a state owned incumbent: the case of France. Energy J. 26, (European Electricity Liberalisation special issue) 181–204 (2005).
Meng, K. C. Using a Free Permit Rule to Forecast the Marginal Abatement Cost of Proposed Climate Policy NBER Working Paper 22255 (National Bureau of Economic Research, 2016).
The World's Biggest Public Companies (Forbes, accessed 29 November 2015); http://www.forbes.com/global2000/
Barbier, E. in Handbook on Energy and Climate Change (ed. Fouquet, R. ) 598–616 (Edward Elgar, 2013).
Coady, D, Parry, I., Sears, L. & Shang, B. How Large Are Global Energy Subsidies? IMF Working Paper WP/15/105 (IMF, 2015).
Pfund, N. & Healey, B. What Would Jefferson Do? The Historical Role of Federal Subsidies in Shaping America's Energy Future (DBL Investors, 2011).
Lipsey, R. G., Carlaw, K. I. & Bekar, C. T. Economic Transformations: General Purpose Technologies and Long Term Economic Growth 79 (Oxford Univ. Press, 2005).
Acurio-Vásconez, V., Giraud, G., McIsaac, F. & Pham, N. S. The effects of oil price shocks in a new-Keynesian framework with capital accumulation. Energy Policy 86, 844–854 (2015).
Popp, D. Innovation and energy prices. Am. Econ. Rev. 92, 160–180 (2002).
Newell, R. G., Jaffe, A. B. & Stavins, R. N. The induced innovation hypothesis and energy-saving technological change. Q. J. Econ. 114, 941–975 (1999).
Fowlie, M., Greenstone, M. & Wolfram, C. Are the non-monetary costs of energy efficiency investments large? Understanding low take-up of a free energy efficiency program. Am. Econ. Rev. 105, 201–204 (2015).
Gillingham, K. & Palmer, K. Bridging the energy efficiency gap: policy insights from economic theory and empirical analysis. Rev. Environ. Econ. Policy 8, 18–38 (2014).
Leiby, P. N. Estimating the Energy Security Benefits of Reduced U.S. Oil Imports ORNL/TM-2007/028 (Oak Ridge National Laboratory, 2007).
Goldthau, A. & Sovacool, B. K. The uniqueness of the energy security, justice, and governance problem. Energy Policy 41, 232–240 (2012).
Gillingham, K., Newell, R. & Palmer, K. Energy efficiency policies: a retrospective examination. Annu. Rev. Environ. Resour. 31, 193–237 (2006).
Demand-Side Management Program Direct and Indirect Costs (US Energy Information Administration, accessed 13 January 2016); http://go.nature.com/29kWJqJ
Auffhammer, M., Blumstein, C. & Fowlie, M. Demand side management and energy efficiency revisited. Energy J. 29, 91–104 (2008).
Fowlie, M., Greenstone, M. & Wolfram, C. Are the non-monetary costs of energy efficiency investments large? Understanding low take-up of a free energy efficiency program. Am. Econ. Rev. 105, 201–204 (2015).
Barbier, E. B. Scarcity and Frontiers: How Economies Have Developed Through Natural Resource Exploitation (Cambridge Univ. Press, 2011).
Delucchi, M. A. & Murphy, J. US military expenditures to protect the use of Persian-Gulf oil for motor vehicles. Energy Policy 36, 2253–2264 (2008).
Stiglitz, J. E. & Bilmes, L. J. The Three Trillion Dollar War: The True Cost of the Iraq Conflict (Norton, 2008).
Stiglitz, J. E. Rewriting the Rules of the American Economy Part 2 (2015); http://go.nature.com/28MnqWz
Caselli, F., Morelli, M. & Rohner, D. The geography of interstate resource wars. Q. J. Econ. 130, 267–315 (2015).
Statistical Review of World Energy 2015 (BP, 2015).
Duflo, E. & Pande, R. Dams. Q. J. Econ. 122, 601–646 (2007).
Aghion, P., Dechezlepretre, A., Hemous, D., Martin, R. & Van Reenen, J. Carbon taxes, path dependency and directed technical change: evidence from the auto industry. J. Polit. Econ. 124, 1–51 (2016).
Schmalensee, R. Lecture 5: Path Dependence in Energy Systems. MIT Open Courseware (Sloan School of Management, MIT, 2012); http://go.nature.com/28Kz5Au
Acemoglu, D. & Robinson, J. Why Nations Fail: The Origins of Power, Prosperity and Poverty (Crown Business, 2012).
I thank F. Green, A. Kopp and N. Stern for discussions. Financial support from the ESRC is gratefully acknowledged.
The author declares no competing financial interests.
About this article
Cite this article
Fouquet, R. Path dependence in energy systems and economic development. Nat Energy 1, 16098 (2016). https://doi.org/10.1038/nenergy.2016.98
Nature Climate Change (2021)
Environmental Science and Pollution Research (2021)
Nature Energy (2020)
International Journal of Community Well-Being (2020)
Landscape Ecology (2020)