Journal Nature: COVID lockdowns are key to begin ‘personal carbon allowances’: ‘Restrictions on individuals…that were unthinkable only 1 year before’ have us ‘more prepared to accept tracking & limitations’ to ‘achieve a safer climate’

Here we discuss how personal carbon allowances (PCAs) could play a role in achieving ambitious climate mitigation targets. We argue that recent advances in AI for sustainable development, together with the need for a low-carbon recovery from the COVID-19 crisis, open a new window of opportunity for PCAs. Furthermore, we present design principles based on the Sustainable Development Goals for the future adoption of PCAs. We conclude that PCAs could be trialled in selected climate-conscious technologically advanced countries, mindful of potential issues around integration into the current policy mix, privacy concerns and distributional impacts.

Climate change could undermine the achievement of at least 72 Targets across the Sustainable Development Goals (SDGs)¹. The development of a just and equitable transition to a net-zero society is vital to avoiding the worst impacts of climate change¹. However, by May 2021, Climate Action Tracker² estimated that climate policies implemented across the world at present, including the effect of the pandemic, will lead to a temperature rise of 2.9 °C by the end of the century. Thus, although many countries have made pledges of net-zero emissions by 2050, implemented policies and pledges are insufficient to deliver the Paris Agreement ambition of limiting global warming to well below 2 °C (ref. ³). To take a national example, the United Kingdom has made strong progress in reducing carbon emissions, and was an early adopter of a net-zero by 2050 target. However, the government’s independent advisory climate body advises that policy steps taken so far “do not yet measure up to meet the size of the net-zero challenge“⁴.

In this context, the introduction of personal carbon allowances (PCAs), a mitigation policy proposal developed in the 1990s⁵, is ripe for revisitation. This policy aims to link personal action with global carbon reduction goals. A PCA scheme would entail all adults receiving an equal, tradable carbon allowance that reduces over time in line with national targets. In its original design, the allowance could cover around 40% of energy-related carbon emissions in high-income countries, encompassing individuals’ carbon emissions relating to travel, space heating, water heating and electricity⁶. Allowances were envisioned to be deducted from the personal budget with every payment for transport fuel, home-heating fuels and electricity bills. People in shortage would be able to purchase additional units in the personal carbon market from those with excess to sell. New, more ambitious PCA proposals include economy-wide emissions, encompassing food, services and consumption-related carbon emissions⁷, for example.

Several variations of mandatory PCAs or personal carbon-trading schemes have been proposed in the literature under different names⁸. For instance, centrally allocated and tradable PCAs have been examined by the UK government, looking at a design covering household energy and personal travel⁹. Electronic Tradable Energy Quotas (TEQs) were also proposed in the United Kingdom, covering the whole economy and divided among individuals (40%) and other energy users (60%)¹⁰. In Ireland, cap and share certificates covering the whole economy were proposed, giving all adults emission certificates for an equal share of national emissions. Such certificates were proposed to be sold by individuals via banks and post offices to fossil fuel companies¹¹. In California, household carbon trading was proposed for household energy, and managed by the utilities¹². In France, centrally managed tradable transport carbon permits were assessed related to private transport¹³. Scholars from the University of Groningen have proposed European Union (EU)-wide emissions trading for households and transport, embedded in the EU Emissions Trading Scheme (ETS) design. In this design, free carbon allowances are allocated to each category of small emitters on the basis of their historic emissions (grandfathering), then surrendered with the purchase of energy from distributors, which in turn give them up as they obtain fuel from fuel producers and importers, who then have to match with allowances their supply of fuel¹⁴. Furthermore, tradable consumption quotas have been proposed to cover all consumption emissions related to manufacturing processes¹⁵. The mandatory nation-wide designs described above are complemented by voluntary schemes, some of which have been trialled in several locations⁸.

The literature highlights the importance of economic incentives, cognitive awareness, prevailing social norms and education as drivers for pro-environmental decision-making and behaviour^16,17. Research indicates that behavioural change could be engendered by creating a direct and visible incentive to reduce carbon emissions^14,18. Studies show that people tend to adhere to the prevailing norm and that descriptive social norms and comparison with others influence decisions about electricity use^19,20 and mode of transport²¹. Building on this literature, PCAs are envisaged to deliver carbon-emissions-related behavioural change via three interlinked mechanisms: economic, cognitive and social²² (Fig. 1). Similar to a carbon tax, a policy with which it is often compared, the economic mechanism of PCAs is envisaged to influence decision-making by assigning a visible carbon price to the purchase and use of fossil-fuel-based energy in the first instance, and possibly also to consumption-related emissions in more advanced PCA designs. However, in addition to the economic mechanism, PCAs aim to influence energy and consumption behaviour by increasing carbon visibility, by evoking users’ cognitive awareness of carbon in their daily routines and by encouraging carbon budgeting. Moreover, the shared goal of emissions reduction and the equal-per-capita allocation of PCAs is envisaged to create a social norm of low-carbon behaviour. These three interlinked mechanisms are hypothesized to promote low-carbon lifestyles in a synergetic manner.

Key mechanisms through which PCAs encourage and promote a low-carbon lifestyle. Adapted from Parag and Strickland⁶⁴.

Full size image

Furthermore, end-user emission cap-and-trade schemes have been described in the literature as a means to rationalize individual engagement in sustainability activities, regulate voluntary offset markets, cap uncapped sectors such as the residential and transport sectors, and stimulate energy-efficiency interventions⁷.

In the 2000s, when the UK government explored the adoption of PCA scheme to reduce carbon emissions from households, the idea was rejected due to claimed low social acceptability, technological barriers and high implementation costs^8,9,23. PCAs were defined in the early 2010s as “a big idea that never took off”²⁴, and ‘“a policy ahead of its time”^5,9. No large-scale national programmes have so far investigated PCAs as a policy option. By 2021, arguably, the policy window of opportunity provided by the COVID-19 crisis²⁵, in combination with the need to address worsening climate and biodiversity crises²⁶, and by the advancements in information and communication technologies, particularly artificial intelligence (AI)²⁷, could improve the feasibility and attractiveness of PCAs to policymakers and the public.

The purpose of this Perspective is not to advocate for the widespread adoption of PCAs, but rather to restart a science and policy dialogue on a policy option that could help achieve climate mitigation goals by re-evaluating the attractiveness of PCA schemes in the 2020s and beyond. We first analyse the barriers that were recognized a decade ago to the widespread adoption of PCAs and reflect on recent social and technical changes that may increase the appeal of PCA schemes in the 2020s. We then develop SDG-based design principles for guiding future applications of PCAs, and present recommendations for the future exploration of PCAs. In our evaluation we are not referring to any specific PCA design; we consider PCAs as a national mandatory policy, with diverse potential designs depending on the local context. To limit the boundaries of this Perspective, PCAs are assessed here as a scheme for more developed countries—those with high per-capita emissions and the administrative capability to implement such policies.

In 2008, after concluding that involving households was critical to reach climate goals²⁸, the UK government commissioned a pre-feasibility study on PCAs. The study, developed by the Department for Environment, Food and Rural Affairs (DEFRA), investigated the effects of a mandatory household-level scheme with free equal-per-capita carbon credits for all UK adults. The study highlighted some substantial challenges with PCAs, which resulted in PCAs and trading being characterized as an “idea ahead of its time”⁹. Starting from that landmark assessment, and adding analysis from the subsequent literature, we identify the main barriers to the adoption of PCAs.

Political resistance and crowded policy landscapes

As mentioned above, at the time of consideration in the United Kingdom, PCAs were considered a radical approach for mitigation. This is still true: PCAs have been described as radical in more recent literature²⁹. There are clear political risks in advocating for challenging or radical policies, particularly if they have never been implemented elsewhere and there is no previous policy experience to learn from. Aside from the United Kingdom’s early interest, no European country has expressed clear political interest in examining, let alone adopting, PCAs⁷. Furthermore, existing climate and energy policies may be perceived as creating a barrier to the inclusion of PCAs. In particular, some argue that PCAs as a downstream measure combined with the existing EU ETS could result in double-pricing of certain emissions, if not properly planned^7,14,30. Although the need for a combination of policy instruments to address the multiple market failures that have led to the excessive generation of environmental pollutants has long been recognized in the literature³¹, and a policy mix is a normal characteristic of policy landscapes³², incorporating a radical policy that has never been implemented before into an existing policy landscape is nevertheless risky, and therefore challenging for politicians.

Technological barriers and high implementation costs

A key question about PCAs is how could they be implemented in practice? What technology is needed to manage carbon accounts? How will people keep track of their carbon allowances? And how would allowances be traded? In the 2000s, the vision was of carbon accounts, analogous to bank accounts, and a carbon card to which allowances would be charged and from which deductions would be made. This option was chosen as it was the most suitable given the existing technological capabilities and was perceived as the most appropriate for a public that was not very ‘carbon capable’³³. However, surveys indicated that the proposed system was perceived by the public as challenging and complex⁹. The DEFRA 2008 study evaluated and costed the option of assigning carbon credits in a national account system run by private sector organizations such as banks⁹. Costs were higher than other mitigation policy measures, such as the United Kingdom’s Climate Change Agreements⁹. Although lower cost estimations than the one in the 2008 DEFRA report for PCAs existed, all were higher than the cost of upstream schemes, mostly due to high administrative costs³⁰. As a result, it was concluded that significant cost reductions would be needed for PCAs to be economically feasible. As discussed later, advances in technology and increased awareness of carbon and climate change mean there are now different options available.

Low social acceptability

From its inception, there have been concerns about the social acceptability of PCAs and their potential to result in unfair distributional effects. Social acceptability was investigated by applying a range of methods including interviews, focus groups, questionnaires, choice experiments and modelling⁸. When the public perception of PCAs was evaluated through interviews in the United Kingdom in 2008, opinions ranged from quite positive to negative⁹. While interviewees were generally willing to accept some responsibility over their emissions, the perceived complexity and the central control over people’s activities were identified as key challenges⁹. Furthermore, surveys in other contexts suggest that the perceived complexity of a PCA scheme could limit its public acceptability³⁴.

Distributional impacts

Another factor that influences the social acceptability of PCAs is the need for them to be perceived as fair, such that certain groups are not being disproportionately affected. When a PCA scheme was evaluated in the United Kingdom in the 2000s, 71% of low-income households were identified as ‘winners’ and 55% of high-income households ‘losers’ from the policy⁹. In other words, due to the variation in energy use, most low-income households were likely to have more allowances than needed to cover their energy needs, and hence could sell excess allowances for money (winners), whereas most high-income households were likely to have fewer allowances relative to their energy needs, and therefore would need to buy extra units in the market (losers). However, a small percentage of low-income loser households were also identified, most of which were living in rural areas⁹. Public perceptions of fairness, as well as the distributional effects of PCAs, depend on how fairness is defined³⁵, on the detailed design of the PCAs scheme and on any associated compensatory policies.

Visible negative effects of the escalating climate and biodiversity crises on many sustainable development issues^1,36 have led to increased public concern over climate change, particularly by the young, as shown in the Fridays for Future movement and climate strikes around the globe. The global climate strike of 2019 was one of the largest events organized by environmental social movements so far³⁷. Recent evidence shows the significant impact of wide participation in these protests on political responsiveness, and on the dissatisfaction with current climate action among young adults and their families^38,39. Mounting public pressure may have played a part in the increasing number of countries and regions including the EU, the United States, the United Kingdom and China that by 2021 had presented pledges to have net-zero carbon emissions by 2050 or 2060. To achieve such pledges, mitigation policies have been put in place to reduce emissions through a wide array of interventions and programmes. However, as both energy and carbon are invisible, it remains difficult for individuals to estimate the contribution of their lifestyles and activities to the nations’ emissions. While energy prices contain some costs related to carbon (for example, the EU ETS, to the extent that this is passed on to energy consumers⁴⁰), and this may be expected to have some impact on consumers’ decision-making, the large participation in social movements demonstrates that many individuals also consider themselves as citizens with responsibilities to the environment and future generations. To this extent, PCAs may be effective as a ‘symbolic policy’—a practical measure that encapsulates a vision or story about a wider change, and signals and engages citizens in this wider vision and project⁴¹. If that is a good description of PCAs, then the route to political acceptability may be to show that it can deliver both practical and symbolic benefits. Given the public demand for more ambitious action and the political commitment to ambitious targets, PCAs could be of increased public and political interest.

PCAs should also be re-evaluated in the context of the COVID-19 experience and lessons that are being learned. Recent research has shown the pervasive negative effects of the pandemic on almost 90% of the SDG targets²⁵—drawing a strong parallel to the climate crisis, which in different ways may negatively influence a similar number of SDG targets¹. It was estimated that a low-carbon pandemic recovery could reduce carbon emissions in 2030 by 25% compared with pre-COVID-19 projections⁴². The aspiration of the international community for a ‘sustainable recovery’ from the COVID-19 pandemic, combined with heightened awareness of the effect of individuals’ actions on the spread of the pandemic, the global connectivity that means that people everywhere are affected by global problems, and the new behavioural and social norms formed during the pandemic, may favour PCAs.

In particular, during the COVID-19 pandemic, restrictions on individuals for the sake of public health, and forms of individual accountability and responsibility that were unthinkable only one year before, have been adopted by millions of people. People may be more prepared to accept the tracking and limitations related to PCAs to achieve a safer climate and the many other benefits (for example, reduced air pollution and improved public health) associated with addressing the climate crisis. Other lessons that could be drawn relate to the public acceptance in some countries of additional surveillance and control in exchange for greater safety. For instance, in many countries, mobile apps designed for COVID-19 infection tracking and tracing played an important part in limiting the spread of the pandemic. The deployment and testing of such apps provide technology advances and insights for the design of future apps for tracking personal emissions. Recent studies show how COVID-19 contact-tracing apps were successfully implemented with mandatory schemes in several East Asian countries, such as China, Taiwan and South Korea⁴³. In these countries, the apps assessed each user’s travel history and health status, playing a key role in tracking infections⁴³. These unique natural experiments give insights into possible strategies to use apps to track PCAs. For instance, the many digital contact-tracing algorithms that were developed and tested^43,44 provide initial valuable information for the design of future apps that—for example—estimate emissions on the basis of tracking the user’s movement history. However, the adoption of such apps also raised issues regarding the balance between data privacy concerns and public health⁴⁵. A recent review showed that only 16 of 50 reviewed contact-tracing apps explicitly state that the user’s data will be made anonymous, encrypted and secured and reported only in an aggregated format⁴⁶. Such a balance is also perceived differently in diverse countries. Initial evidence points to various issues related to adopting such schemes in liberal democracies such as in Europe and the United States—where data privacy, trust and ethical issues strongly limited participation in contact-tracing efforts during the COVID-19 pandemic⁴⁶. Such resistance itself also provides important lessons for future PCA-tracking apps. For instance, new regulations have been suggested to address data privacy concerns and security vulnerabilities when using these apps⁴³ and significant technological advances were made for privacy-preserving contact-tracing apps⁴⁴. These advances could help pave the way for the adoption of PCA schemes. However, citizen engagement and participatory approaches would be needed to design and implement PCA schemes that balance personal liberties with delivering climate aims in a socially acceptable manner.

Finally, advances in digitalization and AI for sustainable development²⁷ promise to shrink implementation costs and logistical challenges for PCAs—and to improve personalized feedback, information and advice. Recent advances in smarter home and transport options make it possible to easily track and manage a large share of individuals’ emissions. Evidence from the roll-out of smart meters and informative displays can be used to design feedback that is highly effective in engaging individuals to reduce their energy-related emissions⁴⁷. Furthermore, AI breakthroughs combined with very high ownership of smartphones will allow the low-cost development of new personalized apps to account for PCAs and trade personal emissions. For instance, machine-learning algorithms could be trained to automatically gather all the available information on someone’s emissions, and to fill data gaps and accurately estimate an individual’s carbon emissions on the basis of limited data inputs such as stops at petrol stations, check-ins at venues and travel histories. AI could be especially beneficial for PCA designs that also include food- and consumption-related emissions. Many voluntary smartphone apps can already capture personal travel and dietary behaviours for estimating carbon emissions and potential health consequences. Algorithms in those apps can intelligently understand the mode of transport on the basis of the user’s speed and trajectory, and can estimate food-related emissions on the basis of purchasing habits⁴⁸. More importantly, machine learning could also support our understanding of what information and advice are most effective for promoting behaviour change through PCAs. An ever-increasing number of decision-making tasks are being delegated to software systems⁴⁹, allowing the presentation of targeted personalized information to future users on their emissions patterns. The latest science on AI for learning, including the use of virtual agents^50,51, could help refine the type of information that users are shown to manage and reduce their carbon emissions. To the user, all of the above could be packaged in an easy-to-use smartphone app that presents tailored information and advice on personal carbon emissions and facilitates carbon savings.

Given the above, the adoption of PCA schemes to support climate action in the 2020s does not seem as challenging to implement⁵ as it did previously (Table 1).

Informed by recent methods assessing SDGs interlinkages^1,27,52, Table 2 explores how PCAs could interact with outcomes in various SDGs, to provide information for their future design.

Adopting PCAs at scale in any given region or country will be a challenging research and policy task. It is unlikely that the same PCAs design would work everywhere—or that PCAs are a suitable policy for all regions or countries⁵³. Climate-ambitious technologically advanced countries with high trust in the government would potentially have more success in implementing just and equality-based PCAs. Such countries would have to investigate how PCAs could be designed to work in their specific social, economic and geographical context, and how such a policy could be practically implemented and harmonized with existing climate policies^1,54 to reduce the risk of incompatibilities^55,56. Nevertheless, scholars argue that existing policies are unlikely to be effective in meeting emission targets⁵⁷ and therefore policymakers should use the full range of instruments⁵⁸. In the EU, insight could be gained from the way the EU ETS is linked to offset markets such as certified emission reductions and the Clean Development Mechanism⁷, and from proposals on how to harmonize PCAs with the EU ETS scheme¹⁴. This Perspective does not present an analysis of how PCAs would cohere with existing policy mixes; this analysis would need to be done at national level before implementation.

In terms of implementation platforms, while in the 2000s carbon allowances were expected to be managed by a card, in the 2020s high ownership would make smartphones the preferred option for accounting and trading (while providing alternative options for the few without smartphones). Innovative AI and machine-learning capabilities would facilitate the expansion of PCAs to include embedded emissions in goods and services, which are harder to calculate, and could help in providing individuals with tailored and timely advice on how to reduce their lifestyle emissions.

The SDG-based design principles for PCAs in Table 2 give an overview of the potential benefits, as well as challenges, that policymakers considering PCAs may encounter. PCAs could be designed to encompass only certain emissions (such as travel, or the household use of fossil fuel methane for heating) or be more comprehensive and cover the whole economy (for example, including all household direct and indirect emissions such as food- and other consumption-related emissions). Therefore, positive and negative impacts on the SDGs are likely to vary significantly.

Possible negative impacts of PCAs on vulnerable consumers will need to be carefully assessed to avoid situations in which they are negatively affected and do not have the means to change their emissions. The design of PCAs should strive to be fair, while acknowledging that it is not possible to have a policy with no losers. In particular, as people vary in their energy needs, an equal-per-capita allowance is not necessarily fair⁹, even if overall PCAs significantly reduce income inequality. Country-specific compensation⁵⁹ or additional policies (for example, initiatives to tackle under-occupancy or improve thermal performance in rural homes) are likely to be needed for some vulnerable loser groups⁹.

Technology-enabled PCA designs will need to consider issues around privacy, cybersecurity and digital ethics. Some lessons from the loss of privacy associated with the use of tracking apps during the COVID-19 pandemic⁴⁶ could provide initial insights into ethical and secure app design⁶⁰ (for example, new regulations and new algorithms for privacy-preserving apps^44,45).

The research community will need to step up to support a more detailed investigation of carbon allowances. Voluntary PCA initiatives and PCA-like schemes will be essential to trial various designs. Evidence from those trials should be incorporated into models that evaluate the impacts of various designs on different income groups. Participatory research methods and engagement with a wide range of stakeholders could help to advance the knowledge of this policy option.

With the world not on track to meet the objectives of the Paris Agreement using current policy tools, PCAs might offer a new approach. Although a PCA scheme would not be easy to design or implement, given the need for very ambitious reduction targets, climate-ambitious countries should ask: if not PCAs, what other scheme should be put in place to affect high-carbon behaviours in support of the objective of net-zero carbon emissions?