The COVID-19 crisis has brought us many lessons. Perhaps the most important is that we were unprepared to face it. In Europe and across the world, there was a shortage of products and services essential to deal with the pandemic (face masks, hand sanitisers, tests, protective garments, ventilators, tracking capacity, and so on).1 Presumably, the strategy (if any) or hope was that – should the need arise – a surge in demand would be met with imports, and/or with a timely supply response. But neither worked. European production suffered from the disruption of supply chains. In several cases, certain goods could not be produced simply because an essential input was no longer available due to other countries’ export restrictions.
Now that we are aware of the consequences of such shortages and disruptions, we have learnt that it pays to be prepared for the next crisis – be it another pandemic or a shock of very different nature (including disasters triggered by natural hazards, a nuclear accident, or a cyber attack, to name just a few). The key question is how to design a resilience strategy that helps prevent, detect, and mitigate its consequences. While one size does not necessarily fit all, certain solutions might be valuable to respond to all sorts of events. Beyond the provision of essential goods, a common ground for a resilient strategy should include a robust primary health care system; a strong logistics network; emergency decision-making bodies that combine legitimacy (elected politicians) with expertise (expert advice); and institutions that promote cooperation among countries or regions.2
Read CEPR Policy Insight No. 106 “Preparing for the next crisis: How to secure the supply of essential goods and services” here.
All such solutions require the state to play a key role by its intrinsic duty, which is to protect its citizens. Private incentives alone are insufficient to prepare for rare events with large externalities.
How to prevent, detect, and mitigate future crises
Preparation for catastrophic events should consist of an articulated strategy, which goes from having an adequate research infrastructure and human capital, to investing in prevention, to early detection (in order to mitigate their impact), to building the ability of reacting should such events materialise (National Academy of Medicine, 2016).
General research infrastructure and human capital
Having an appropriate physical infrastructure to address health issues and a highly prepared labour force will generally help in all stages of a resilience strategy. Arguably, during the COVID-19 crisis, Germany’s relative success presumably rested on its dense network of healthcare facilities and chemical and bio-chemical labs, together with its large number of intensive care units (ICUs). Likewise, the recent crisis has made manifest the importance of having genetic research centres and pharmaceutical laboratories, so as to be able to sequence the virus, monitor its evolution, develop a vaccine and find an appropriate medication.3 The public good nature of research, and research infrastructure, implies that it might suffer from under-provision, thus making public support needed. Tighter coordination at the EU level in this area would also be important – for example, the European Research Council (or a similar body) could identify the areas where research on possible catastrophic events is most needed and fund labs on the basis of merit.
By their very nature, crises trigger externalities that spread across individuals and beyond country boundaries. Thus, public institutions, and the coordination across them, are likely to play key role in any resilience strategy.
When it comes to individuals, it is crucial that government regulation is in place so as to prevent risk-taking behaviour (which will generally have consequences beyond the individual taking the action). And when it comes to countries, two types of institutions seem to be needed: (i) independent authorities that are entrusted with the task of investing in preventive measures, possibly endowed with a budget which is not conditioned to political changes (thereby avoiding, as much as possible, the political economy problems indicated above); and (ii) well-functioning supra-, multi-, or international bodies which may be relied upon to coordinate actions. Precautionary measures may then be publicly procured (and privately provided) or carried out by the different states, but in a coordinated manner.
Prevention and capacity to react to events
Early detection tools and stress tests to critical sectors (such as pharmaceuticals, medical supplies, utilities) should be part of a set of prevention measures to allow public authorities to anticipate the outbreaks and to make sure that society is ready to face them, providing the essential goods that people need during a crisis. For this purpose, a resilience strategy should contain a combination of different elements, consisting of (i) the precautionary accumulation of essential goods; (ii) clear rationing protocols for those goods and periods in which production is not sufficient to meet demand; and (iii) measures to guarantee that when catastrophic events occur, supply will be ramped up (if – as is likely – the stock of goods is insufficient).
For storable goods and inputs, the question is what fraction should be stored and what fraction could be provided through ramped-up production capabilities. Regarding the part of essential goods that is to be stored, the government has the option to do this itself or it can procure this service from private parties. Since this service is provided continuously, standard procurement practices can be used, accompanied by appropriate monitoring to make sure that the contracting parties are compliant. The part of the essential goods and inputs that are not stored, but instead have to be produced at the onset or during a crisis, requires maintaining excess capacity up to some quantity within a certain time window. This can be unused capacity or capacity that can be quickly converted from its use in normal times to the production of the specified essential good.4 Since this issue has been debated in electricity markets for a long time, it is worth looking at their experience to possibly draw broader lessons.
Learning from the experience in electricity markets
Electricity markets provide an extreme example of the need to have excess capacity. Storing electricity in large quantities is either not feasible or highly expensive. Yet, electricity demand and supply have to be matched at all times, even at times when renewable resources are not available or when demand peaks due to seasonal fluctuations or unexpected events. Demand rationing – as we have recently seen in California as skyrocketing temperatures have pushed up the demand for air conditioning – is a highly inefficient solution, and it is often not feasible (for example, if the shocks to demand or supply are unexpected). Given the difficulties in balancing demand and supply through active demand response, electricity regulators have traditionally addressed security of supply concerns by putting in place regulatory mechanisms (the so-called capacity mechanisms) that ensure that excess generation capacity is built, even if some of it would only be used under rare events. The public good nature of security of supply implies that market forces alone would not address it efficiently, giving rise to under-provision. The parallel with the necessity to guarantee the supply of certain goods in situations of emergency is evident, as it is the fact that failing to provide such goods would have large additional social costs on top of private costs.
In practice, capacity mechanisms differ along various dimensions, but they all have one common characteristic: the choice of capacity is not left to the market. Rather, the regulator decides how much capacity needs to be made available, and a competitive mechanism determines which firms will provide such capacity and the rewards for doing so. A critical issue is determining how much capacity needs to be secured. In electricity markets, just as in many other contexts, this question can be formulated as a principal-agent problem as the agent in charge of it might be biased towards too much or too little provision. Furthermore, predicting the future needs is not an easy task. In electricity, it requires forecasting future electricity demand and supply (which are ‘known unknowns’), but in our context it requires assessing the likelihood of future crises (which may well be ‘unknown unknowns’).
An important design dimension is whether capacity markets should be centralised or decentralised. Under centralised mechanisms, the regulator sets up a central auction that serves to determine the plants that will commit to provide security of supply, and the price at which they will do so. Under decentralised mechanisms, the regulator imposes a capacity obligation on the electricity retailers, who need to buy capacity credits from capacity providers, either bilaterally or through exchanges. Failure to do so is penalised through fines. Think of its parallel for the case of flu vaccines. A centralised mechanism would have the government procuring all the vaccines needed for the whole population, while a decentralised mechanism would have the regulator imposing this obligation on the hospitals according to the share of the population in their catchment areas. Typically, buying power and risk-sharing considerations would recommend centralised mechanisms. In contrast, information issues might favour the more decentralised approach. A drawback of decentralised systems is the need to put in place a system of penalties and monitoring (in cases in which violations may be detected ex ante) to avoid non-compliance. Yet, penalties might not always be credible given the nature of the agents involved (for example, would it be credible to penalise a hospital that fails to provide essential supplies during a pandemic?) and given the scale of the penalties that in some cases might be needed to make the incentive system work.
Another important design dimension is whether capacity mechanisms should be market-wide (i.e. all existing firms are entitled to receive capacity payments) or whether they should be targeted to a subset of firms (only new plants, plants located in a certain region, certain technologies, etc.). The main advantage of market-wide mechanisms is that they are better at selecting the firms that are more efficient in providing capacity. However, market-wide capacity mechanisms need not provide the least costly way of securing capacity from the consumers’ perspective.
Based on the experience with electricity capacity markets, and taking into account that the array of potential crises is very broad, we believe that as a general principle, centralised and targeted mechanisms – such as a system of strategic reserves to be procured trough competitive mechanisms – might help guarantee the supply of essential goods and services.
Overall, while catastrophic events are difficult to avoid, society should strive to be better prepared for the next crisis. This is a challenging task, but one which should be priority for all members of society. Our role as economists should be to shed light on how to achieve that goal at least cost to society.
Authors’ note: This column summarises and freely borrows from CEPR Policy Insight 106 (Fabra et al. 2020).
Aghion, P, S Amaral-Garcia, M Dewatripont, and M Goldman (2020), “How to strengthen European industries’ leadership in vaccine research and innovation”, VoxEU.org, 1 September.
Bénassy-Quéré, A, R Marimon, P Martin, J Pisani-Ferry, L Reichlin, D Schoenmaker, and B Weder di Mauro (2020), “Repair and reconstruct: A recovery initiative”, VoxEU.org, 20 April.
Fabra, N (2018), “A primer on capacity mechanisms”, Energy Economics 75: 323-335.
Fabra, N, M Motta, and M Peitz (2020), “Preparing for the next crisis: How to secure the supply of essential goods and services”, CEPR Policy Insight No. 106.
Fuchs-Schündeln, N, M Kuhn, and M Tertilt (2020), “The short-run macro implications of school and childcare closures”, VoxEU.org, 30 May.
National Academy of Medicine (2016), The Neglected Dimension of Global Security: A Framework to Counter Infectious Disease Crises, The National Academies Press.
Torrejón Pérez, S, M Fana, I González-Vázquez and E Fernández-Macías (2020), “The asymmetric impact of COVID-19 confinement measures on EU labour markets”, VoxEU.org, 9 May.
World Health Organization (2020), World in Disorder, Global Preparedness Monitoring Board Annual Report 2020.
1 As the World Health Organization writes, “The COVID-19 pandemic has highlighted preparedness gaps across levels around data and information sharing, communication and messaging, medical equipment (e.g. personal protective equipment) distribution, skilled personnel, managing travel restrictions, and testing and tracking cases.” (World Health Organization 2020: 33).
2 While we focus on the direct (and more immediate) effects we acknowledge that a resilient strategy should not only look at direct but also indirect effects. In the COVID-19 pandemic, “COVID-19 affects all spheres of life. It therefore requires a comprehensive response, encompassing health, mental health and psychosocial support, education, and other aspects of the social and economic sectors. Many sectors lacked plans to mitigate not only the public health risks but also the potential socioeconomic impact of the pandemic … This lack of multisectoral preparedness left many societies scrambling to figure out how to maintain essential services and mitigate economic disruptions. The potential impact of pandemics on non-health sectors and the private sector has been a known risk for years.” (World Health Organization, 2020: 33). Several VoxEU columns take a look at indirect effects (e.g. Fuchs-Schündeln et al. 2020, Torrejón Pérez et al. 2020).
3 Aghion et al. (2020) give details on how the US Biomedical Advanced Research and Development Authority (BARDA) supports vaccine research. They also advocate the creation of a European BARDA.
4 As the National Academy of Medicine (2016: 77) puts it in case of pandemics, “spare manufacturing capacities may be needed to accommodate mass manufacturing of products, as well as testing investigational products.”
5 For a formal analysis that provides a rationale for capacity mechanisms, see Fabra (2018).
6 See also Bénassy-Quéré et al. (2020).