The Race to Fight COVID-19: On the Desirability of Regulatory Competition

Jurisdictions worldwide struggle with the COVID-19 pandemic. The two key policy goals are, first, to contain the spread of the virus to protect lives and, second, to minimize the damage to the economy to protect livelihoods. These policy goals are in potential conflict: drastic measures to contain the pandemic, such as a lockdown of private and commercial activities, will inflict huge costs on the economy. At the same time, it is also clear that reducing the spread of the virus does not only save lives. It has economic benefits as well by, for example, reducing healthcare costs and maintaining the size and health of the workforce and consumption levels.  

Striking the right balance between pandemic control and containing the economic fallout is a delicate exercise, as is identifying the right measures that will (hopefully) achieve the specified regulatory goals. The Oxford COVID-19 Government Response Tracker (OxCGRT) records the unfolding government responses in a rigorous, consistent way across countries and across time. It also defines a ‘Government Response Stringency Index’, which ranges from 0 to 100, and plots it against the number of COVID-19 cases in a jurisdiction. The emerging picture is striking: some countries do nothing (0), some have brought public life almost to a standstill (100), and we see case numbers widely dispersed around a regression line, which is only very slightly upward sloping around a stringency of 60.

Is this regulatory diversity a cause for concern?  Some believe that the pandemic requires a (strictly) coordinated response. The President of the European Commission, Ursula von der Leyen, for example, demanded on 13 March 2020 that ‘each Member State [of the European Union] needs to live up to its full responsibility. And the EU as a whole needs to be determined, coordinated and united’. Similarly, in a statement at the Extraordinary G20 Leaders’ Summit on 26 March 2020, Chinese President Xi Jinping demanded ‘a collective response for control and treatment at the international level. … Countries need to leverage and coordinate their macro policies’.

Yet, regulatory diversity and regulatory competition between jurisdictions in responding to the COVID-19 pandemic may be inevitable.  We should not expect much of a planned collective response based on multilateral agreements. Rather, experimentation, diversity of approaches, and imitation will characterize the regulatory landscape, and for good reason. In a nutshell, against a backdrop of uncertainty about the optimal policy, and with differing priorities and contexts across jurisdictions, policy diversity and regulatory competition allow a rapid and productive learning process. This is a better ‘treatment’ for the pandemic than clumsy efforts at harmonising the policy responses across countries. 

Various factors limit the degree of government coordination and collective action.  First, beyond the nation-state, collective action depends on ad hoc consent, which is difficult to achieve and fragile. Even within nation-states, constitutional and political constraints may prevent federal governments from implementing specific policies. The European Union (EU) is a case in point. It has significant competencies with respect to the internal market.  These are important during the crisis to maintain the free flow of goods, including medical supplies, across borders. But on health issues its formal role is all but nominal.  Under Article 168 of the Treaty on the Functioning of the European Union, it is the EU countries that define and deliver their national health services and medical care.  The EU is only empowered to complement national policies by means of its Health Strategy.  Even within many of its Member States such as, for example, Germany and Italy, health issues are not for the federal government to decide, either. Rather, regional authorities or local governments are empowered to adopt and implement the policies and measures to contain the crisis. Local politicians want to be seen as energetically tackling the problem and addressing the concerns of the people in the jurisdiction in which they have been voted into office. When immediate action is required, they are likely to view engaging in difficult and time-consuming coordination efforts as a second-order concern.

Second, states are hit by the pandemic at different points in time, and with different intensity. In Europe, for example, cases surged first in Italy. The virus then spread to neighbouring countries such as Switzerland, Austria and Germany, before affecting many other states, including the UK. In terms of the number of confirmed cases, for many weeks Germany has been lagging behind Italy by roughly one week, and the UK by two. It is unlikely that the same policies and measures should be implemented in jurisdictions regardless of when and how the virus affects them. 

Third, countries differ—they have unique histories, cultures and traditions. People in different countries differ, too. Their preferences, for example with respect to the trade-off between health/safety and economic prosperity, their appetite or tolerance for risk, or their acceptance of curtailments of individual liberties, including their sensitivity to privacy issues, may significantly diverge. Different preferences will lead to different government responses to the COVID-19 pandemic. One can see this as an interesting new application of the ‘Varieties of Capitalism’ debate in economic/political theory (see Hall/Soskice (eds.), Varieties of Capitalism, 2001). The Oxford COVID-19 Government Response Tracker tracks this distinction, showing that coordinated-economy jurisdictions such as continental European ones have taken stricter measures than the ‘liberal’, market-friendly ones, such as the US and the UK.

Consequently, it is unrealistic to expect much planned, multilateral coordination in the struggle to contain the spread of the virus and to mitigate the associated economic fallout.  Even more importantly, though, this deficit in planned coordination may not be as great a cause for concern as it may appear. It may even be beneficial.

First, as explained above, even if it were theoretically possible to identify the optimal ‘treatment’ for the pandemic given the specific state of affairs in a particular jurisdiction at a particular point in time, such ‘treatment’ would always depend on the local preferences of the people in that jurisdiction. What is best—or tolerable—for the US is not necessarily best for France, South Korea or Japan, for example. There is no global optimum.        

Second, and more importantly, identifying and implementing an optimal ‘treatment’ for the COVID-19 pandemic is illusory. Societies worldwide operate under a huge information deficit as regards how best to tackle the disease and its consequences for humans and economies. Scientists learn more and more about the virus, how it spreads, how it affects our health etc. Of course, we know that observing rules of basic hygiene and isolating carriers of the virus helps. However, we know very little about the effects of (marginally) different policies regarding, for example, forms of social distancing or quarantine on infection rates or economic losses. Even where we do not operate under a (huge) information deficit, implementing effective policies may be made difficult because of scarcity of resources. For example, mass testing is key, but testing kits are scarce and expensive.  

Hence, it is not surprising that we see divergence in responses from various jurisdictions.  As mentioned, the variance between countries in the Oxford COVID-19 Government Response Tracker in terms of adopted policies for each number of confirmed COVID-19 cases is significant. To put it differently: The data tells us little (if anything) on the (marginal) effectiveness of different policies on combating the pandemic. Similarly, we have little knowledge on the marginal effect of different specific policies on key economic factors such as GDP, frequency of business insolvencies or employment levels.

These are the kinds of conditions under which it appears reasonable to experiment and learn from the experiences that governments and policy-makers make in different countries. Some things work, others do not. Regulatory competition between jurisdictions has the potential to initiate a discovery process (Hayek) for the best or most efficient law. Different countries are adopting different measures, study the responses of other governments, and learn from the experience. The UK, for example, started out with relatively few restrictions on people’s behaviour. However, after cases surged, and having the benefit of fresh data on the effects of more stringent measures by continental European states such as Austria or Germany, it adopted a more stringent line. The ‘race’ here is clearly to the top, in the sense of minimising the death toll.  People in different jurisdictions want to ‘win the war’ against the virus with the tools their systems offer and by learning from others. Witness, for example, the interest and analysis surrounding the low death rates in Germany compared to the total number of cases: ‘Germany’s low coronavirus mortality rate intrigues experts’, according to The Guardian. 

The economic policies implemented to moderate the economic damage inflicted by the pandemic can be assessed similarly. Since long, Germany and the Netherlands, for example, had in place a social insurance system known as ‘Kurzarbeitergeld’ (wages for reduced working hours), whereby employees receive a significant portion of their wages if employers cut working hours in a crisis. ‘Kurzarbeit: a German export most of Europe wants to buy’ titled the Financial Times in an article ahead of the introduction of similar measures in other European countries, including in the UK.

I strongly doubt whether the same convergence could or would have been achieved based on multilateral negotiations between countries with the goal to agree on effective policy-measures. Experience tells that such coordination efforts frequently come to nothing—after precious time has passed and significant coordination costs have been sunk. If the efforts are successful, we might get a minimum consent, which might well distract from or even prevent the implementation of rapid and efficient policies on the local/nation-state level. This would reduce the level of experimentation, mutual learning and coordination through imitation (which is quite different from planned coordination).

An obvious counterargument to this analysis are potential externalities: is there not the risk that, if a large state such as the US ‘gets it wrong’, the negative consequences will be felt by people all over the world, similar to the financial and economic crisis in 2008-2009?  The risk of externalities must be taken seriously. At the same time, the COVID-19 pandemic is different from the global financial and economic crisis in 2008-2009. One should distinguish between ‘exporting’ the virus (and the negative health and economic consequences that come with it) and directly ‘exporting’ negative financial and economic costs. Most borders are closed by now, so the risk that a country (re-)exports the virus to others appears to be relatively low. Reopening the borders can be calibrated by states—or supra-national bodies like the EU—to account for the specific regional risk of a (new) surge in cases. As far as economic and/or financial externalities are concerned, the current crisis is primarily a crisis of the real economy, not a crisis of ‘systemically important’ financial institutions or the financial system. The externalities imposed on the real economy are limited in scope. Global supply chains are still working reasonably well. Firms are in trouble primarily because the retail markets have been shut down in many countries. Hence, again, the risk of contagion appears to be relatively low—lower, at least, than in 2008-2009.

The stakes in the fight to contain the pandemic, minimise the death toll, and to mitigate the economic fallout could not be higher. Effective coordination of government policies by multilateral agreement appears to be a first-order concern. But such coordination is not a realistic prospect. Moreover, there are reasons to believe that the search for a worldwide ‘optimal treatment’ of the pandemic is not only illusory but distinctly harmful. Regulatory experiments on the level of nation-states or even local communities, rapid learning from the experience of others, and coordination through imitation may be the best way to win the war against the virus and to contain the associated economic fallout.

Horst Eidenmüller is Professor of Commercial Law at the University of Oxford.

I gratefully acknowledge the perceptive comments of my colleagues Luca Enriques and Thom Wetzer on earlier drafts of this post. The usual disclaimer applies.