The idea of “building back better” during economic recovery -e.g., having a “greener” economy with lower energy consumption and CO2 emissions after a crisis- has gained political momentum in recent years. However, it remains unclear whether it could or has ever worked. Looking at past crises, in the Baltic states, energy consumption and CO2 emissions more than halved in the first years of the 1990s during the crisis generated by the collapse of the Soviet Union and did not bounce back to the pre-crisis level during the recovery. Something similar happened in Spain and several other European countries during the global financial crisis in 2007. But how did it happen? Are these just isolated cases, or do economic changes during crises and the policy responses to them have long-lasting effects on emissions? This question which motivated our research, became more salient in recent years as policymakers must deal with the long-lasting climate crisis on the one hand and the short-term social and economic disruption originated by the Covid-19 pandemic and now the war in Ukraine on the other hand. Because we have and will certainly have new national or even global economic and energy crises, understanding their effects on the process of decarbonisation is essential to design more resilient climate mitigation policies in the pathway to carbon neutrality by 2050.
Do economic crises have any long-lasting effect on structural change towards decarbonisation?
Scholars from different disciplines have hypothesized the effects of economic crises on technological and structural change: like economists through the ideas of “creative destruction” and “Green Keynesianism”, political scientists who have investigated crises as “critical junctures” allowing fundamental shifts on policy trajectories, and transition scholars interpreting crises as “exogenous shocks” that may support the break-through of niche innovations.
In my recent article, co-written with Johan Lilliestam and Tim Tröndle, we tested some of these theoretical claims by analysing the impacts of crises on national economies. In particular, we investigated how past economic crises affected the energy systems in countries that have already achieved a peak in territorial CO2 emissions – a precondition to reaching the Paris agreement´s goals. We assessed all major crises of the last 50 years in the 45 countries that are part of the OECD or the G20 and account collectively for around 80% of global CO2 emissions. We analysed not just C02 emissions but its determinants to understand the mechanisms by which a crisis can bring about this change. We tracked changes in GDP per capita, population, energy intensity (the energy consumed per unit of GDP), and carbon intensity (the CO2 emitted per unit of energy consumed) before and after the crisis. The lasting effects on energy and especially carbon intensities are critical for the transition to carbon neutrality. We found that 26 out of 28 countries that reached an emissions peak between 1965 and 2019 did so before or during an economic crisis (Figure 1). That is explained by lower GDP growth and accelerated structural change, reflected by energy and/or carbon intensity improvements during and after the peak-related crisis. The paper then discussed three key mechanisms behind such improvements: (i) energy efficiency measures taken by governments and firms to respond to higher energy prices or deteriorating economic conditions; (ii) changes in economic structure due to the decline of energy/carbon-intensive industries and the rise of less intensive ones post-crisis; (ii) changes in the energy mix (e.g., coal to gas and renewables), leading to reduced carbon intensity, triggered by new market conditions or policy changes.
Figure 1: Number of countries that have and have not peaked CO2 emissions (the year with the highest rolling average of the five past years). OECD and G20 countries. The flag indicates the peak year of the respective country. Flags designed by OpenMoji. Data source: BP (2022).
Our findings suggest that crises can trigger structural change in specific countries -usually the ones more affected by the crisis. In many cases these benefits can spill over to other countries.
- The oil crises in the 70`, beyond accelerating the deployment of nuclear power, activated innovation in renewable energy. Years later these technologies became cost-competitive for power generation worldwide.
- During the deep recessions in the early 1990s related to the collapse of the ex-Soviet bloc, the innovation effects were weak, but direct economic effects were strong and they reduced CO2 Some of these effects -economic restructuring and gains in energy efficiency- remained after the crisis explaining why emissions didn’t return to pre-crisis levels, even if today they still are far too high from a Paris Agreement perspective.
- The global financial crisis (2007-2009) is associated with CO2 emissions peaks in many European countries, Japan, and the USA. Most of these countries were already (slowly) decarbonising their economies -mainly by improvements in energy efficiency and support for renewable energy- and the crisis accelerated such processes, while consumption of coal and oil decreased.
- Not all crises -and certainly not in all countries- lead to positive effects on decarbonisation. This was either because countries were only marginally affected by a crisis resulting in minor disruptive economic effects or because governments did not take meaningful measures to improve energy efficiency or change the energy mix.
In sum, crises do not automatically trigger structural change and emissions peaks but may strengthen ongoing decarbonisation trends through economic mechanisms and especially when the right policies are implemented before and after the crisis strikes.
Are the responses to the Covid-19 pandemic and to the current energy crisis delivering systemic change?
There are important differences with previous crises. First, under the Paris Agreement regime, most countries now explicitly recognise the transition to carbon neutrality as a fundamental and overarching policy goal. Second, many low-carbon or zero-carbon technologies are cost-competitive with fossil-fuel-based technologies, weakening the political dilemma of “environment protection vs economic recovery” that usually emerged during recessions in the past. Third, the Russian invasion of Ukraine and subsequent energy crisis made it clear that a faster transition away from fossil fuels offers a way to enhance energy security. Thus, these were good reasons to expect a strong “green” policy response from governments and a shift in business strategies from “brown” to “green” to address the current crisis. There are good examples of such developments, like the Inflation Reduction Act in the US, and the REPowerEU plan in Europe. Investment in low- and zero-carbon technologies is clearly accelerating in most regions as is R&D funding for new technologies. However, policies in the opposite direction have also been implemented. For instance, the crisis pushed fossil fuel consumption subsidies to an all-time high in 2022, rising above USD 1 trillion worldwide for the first time. Subsidies and similar measures tend to preserve existing energy regimes, even increasing fossil-fuel investment plans for the coming years.
Globally, the policy responses are mixed and certainly not strong enough from a Paris perspective. More decisive policy action is urgently needed, particularly in major coal and oil producer and consumer countries, to break their dependence on fossil fuels. The actual long-lasting effects of today´s crises will only be known in some years, but the window of opportunity for more transformative policies is rapidly narrowing. The current energy crisis is probably the last opportunity to reallocate public and private investment, aligning it with the goal of rapid and deep decarbonisation.
This work has received funding from the European Research Council (grant 715132) and the EU-funded Horizon 2020 TIPPING + project (grant 884565). I thank Dr Maria Apergi for her useful comments. Poster image: @RIFS/Felix Beger