Recent optimistic news about the availability of a number of vaccines against the coronavirus needs to be tempered by the realisation that, even in the countries that are in the vanguard, it is likely to be the middle of next year before a large share of the population has been vaccinated. In the meantime, governments around the world are trying to calibrate policy interventions so as to keep the spread of the disease under control without crippling economic activity, in many cases with limited success as virus transmission has recently picked up again in several countries. This study uses country experience during the first phase of the pandemic to estimate the impact of different government interventions on both the reproduction rate of the virus, R, and on mobility, as a proxy for economic activity. The empirical results then inform a number of scenarios where the epidemic/economic trade-off of different policy packages is assessed.1
OECD Economic Outlook, Volume 2020 Issue 2
Issue Note 4. Walking the tightrope: Avoiding a lockdown while containing the virus
Explaining mobility as a proxy for economic activity
The containment policies implemented by governments to reduce the spread of the virus come at an economic cost, proxied here in terms of their effect on mobility. Data on mobility are made available by Google, based on the movement of people with Android-based smartphones and with “location history”.
Figure 2.14. Link between mobility and GDP forecast revisions at a quarterly frequency for the first and second quarters of 2020
Note: The vertical axis is the ratio of the latest GDP estimate (or official outturn) to the projected level for the corresponding quarter in the December 2019 OECD Economic Outlook. Each dot represents a country/quarter combination. The chart covers OECD and BRIICS countries.
Source: Google LLC, Google COVID-19 Community Mobility Reports, https://www.google.com/covid19/mobility; OECD Economic Outlook No. 106 and 108 databases; and OECD calculations.
The effect on mobility of containment policies
To represent government containment policies, the empirical analysis here relies on a set of variables maintained by the Oxford Blavatnik School of Government (Hale et al., 2020) distinguishing eight types of policy, which in their original form are scored according to the degree of stringency or comprehensiveness with which they are applied (Table 2.3, Figure 2.15).2, 3
Table 2.3. Scoring of different stringency levels of containment policies
Note: Scoring scheme used by the Oxford Covid-19 Government Response Tracker. Not shown in the table, but “No measures” or “No restrictions” are always scored 0.
Source: Hale, T., et al. (2020), “Oxford COVID-19 Government Response Tracker”, Blavatnik School of Government, Oxford University.
Empirical estimation suggests that seven of the eight types of containment policies have a negative effect on mobility (Figure 2.16).4 Also, the more stringent application of a particular policy tends to reduce mobility by more: for example, the most severe form of workplace closure (score of 3) has 9 times the effect on mobility of the mildest form (score of 1). Three forms of containment policies stand out as having a particularly large effect on mobility, namely workplace closures, stay-at-home requirements and school closures: the most stringent application of just these three policies is estimated to reduce mobility by more than 40%. Other policies such as the cancellation of public events and travel restrictions, have a significant but smaller effect on mobility, although in some cases the most limited application of a policy has no significant effect on mobility.
Figure 2.15. Percentage of countries at different stringency levels for containment policies
Note: Based on the Oxford Covid-19 Government Response Tracker. Stringency is evaluated as an ordinal index, with a higher number representing greater stringency. Each panel subtitle indicates the range for that category in square brackets. The charts are based on the source’s full country coverage for a given date, at most 185 countries.
Source: OECD calculations based on Hale, T., et al. (2020), “Oxford COVID-19 Government Response Tracker”, Blavatnik School of Government, Oxford University.
Figure 2.16. The estimated effect of containment policies and natural caution on mobility
Effect on mobility (in the absence of the virus and containment measures, ‘normal’ mobility is indexed at 1.0)
Note: The chart shows the estimated effect on mobility of different containment policies at varying degrees of stringency, as well as the effect of increased natural caution, proxied by the daily death rate. If a particular degree of policy stringency is not shown in the bar chart, then it either means there is no significant effect on mobility (this being the case when bar segments at a higher level of stringency are shown), or that it has the same effect as the policy at the previous level of stringency (when bar segments at a lower level of stringency are shown).
Source: OECD calculations based on estimations reported in OECD (2020), “Walking the tightrope: avoiding a lockdown while containing the virus”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris.
The effect on mobility of more cautious behaviour
The virus is likely to have an impact on reducing mobility as general awareness increases natural caution and so increases voluntary physical distancing, independently of government policies. This effect is proxied in the empirical analysis by the inclusion of the national daily death rate from the virus. A national daily death rate running at around 15 per million – similar to the rate experienced by some major OECD countries going into the lockdown in March – is estimated to reduce mobility by 10%, independently of any government-mandated polices.
Explaining the reproductive rate
The effective reproductive number (R) of a communicable disease is the average number of secondary cases per infectious case. As is now widely understood, to eliminate the virus, R has to be maintained consistently below unity. The median R estimate for a worldwide sample of approximately 70 countries fell from around 3 in February to around 1 in early May and has remained stable since (Figure 2.17, Panel A).5 This, however, hides considerable cross-country variation, with R nearing 1.5 in October in European countries, before a further set of major lockdown measures was implemented (Figure 2.17, Panel B). In order to explain this profile three categories of variables are used in the empirical analysis: containment policies which enforce physical distancing, as described above; other public health policies, including testing and tracing; and variables which proxy more cautious social behaviour of the general population as well as the effect of an increasing share of the population having been infected and so possibly being less likely to spread the virus in future.
An important feature of the estimated equation explaining R is that the preferred functional form for the dependent variable is logarithmic implying that any policy intervention will have a larger effect when R is initially high than when it is low, and underlines the merit of early policy interventions.
Figure 2.17. Median and interquartile range for the effective reproduction rate (R)
Source: OECD calculations of R for individual countries are derived from data on deaths and infections, see OECD (2020) for details. The chart summarises trends in R for a selection of worldwide (Panel A) or European (Panel B) countries for which R can be computed over the full sample period.
The effect of containment policies on R
In estimation, the coefficients on five containment policies ‑‑ workplace closures, restrictions on gatherings, stay-at-home requirements, international travel controls and school closures ‑‑ are found to have a significant effect in reducing R (Figure 2.18). The coefficient on school closures has the largest effect of any containment policies, but there is a degree of collinearity between school closures, stay-at-home requirements and workplace closures arising because such containment policies have often been imposed at the same time. Further testing suggests that while the sum of the coefficients on these three containment variables is a robust indication of the effect of a combined package, the coefficient on any one of them is less reliable as it is sensitive to the exclusion of the other variables. Similarly, the absence of any role for the closure of public events in the equation is likely related to its overlap with restrictions on the size of gatherings, which is included. The combined effect of applying all containment polices suggests that from an initial R0 value of about 3, a complete package of containment measures would nearly halve the effective reproduction number.6
A feature of these results with potentially important policy implications is that the full R reduction is often achieved well before the maximum level of stringency is reached: for example, a stringency score of 2 on the workplace closure variable (“for some sectors”) reduces R, but it is not possible to detect any additional effect on R from a further increase in the degree of stringency (“closure for all-but-essential workplaces”).
The effect of public health policies on R
Test and trace policies
To capture the effect of test-and-trace policies, the policy indicators from the Blavatnik School of Government at the University of Oxford (Hale et al., 2020) are used, which in their original form are scored according to the comprehensiveness of the policy (Table 2.4). They suggest there was a substantial improvement in the number of countries increasing the extent of their test and trace policies in the 2-3 months from March, but further increases since then have been modest (Figure 2.19).7 An additional variable, constructed by the OECD, considers the importance of specific testing in care homes (Table 2.5). However, an important limitation of these indicators is that none cover issues of timing, which can be key to a successful strategy: tests need to be done quickly and with a minimum delay before the results are available and then contacts need to be traced quickly. On the other hand, many issues relating to testing, including timing, may be easier when the level of infections is lower, and this can be readily tested in the empirical framework.
Figure 2.18. Effect of containment policies and public health policies on (logged) R
Note: This chart decomposes the effect on (logged) R from the different containment policies (left hand side red bars) and public health policies (right hand side blue bars) according to an estimated equation. The log specification means that the absolute effect of a policy on R is greater the larger the initial value of R: from an initial R value of 3.0 (1.5) the effect of a ‘Typical lockdown’, which reduces logged R by 0.3 in both cases, is to reduce R in absolute terms by 0.8 (0.4). The maximum effect of the containment policies is shown in the chart, corresponding to a degree of stringency greater or equal to n, denoted by the number (>=n) shown in brackets next to the relevant segment of the bar chart.
* The effects of school closures (>=2), stay-at-home requirements (>=1) and workplace closures (>=2) have been combined into one segment labelled ‘Typical lockdown’, this is both because such policies have often been imposed at the same time and, as discussed in the main text, because multicollinearity means that the sum of the coefficients on these three containment variables are more reliable than any of the individual coefficients.
Source: OECD calculations based on an equation estimated for a worldwide sample of countries reported in OECD (2020), “Walking the tightrope: avoiding a lockdown while containing the virus”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris.
Table 2.4. Oxford Covid-19 Government Response Tracker: Scoring of testing and contact tracing variables
Note: Scoring scheme used by the Oxford Covid-19 Government Response Tracker. (1): Testing variable relates to policies testing for infection (PCR test), not to policies testing for immunity (antibody tests).
Source: Hale, T., et al. (2020), “Oxford COVID-19 Government Response Tracker”, Blavatnik School of Government, Oxford University.
Empirical results suggest that test and trace policies can reduce the spread of the virus, although the most comprehensive form of test and trace policies are more than 2½ times as effective in reducing R than more limited forms. Test and trace polices are most effective when the infection rate is not too high (which in estimation is taken to be less than 10 new daily cases per million population, a rate which was well exceeded by many countries in March and April). A rather unsurprising finding, given the difficulties of tracking all contact persons in a timely manner if the system is overwhelmed with new cases. Overall, the effect of the most effective test and trace regime in an environment of low daily infection, is estimated to have a greater effect on reducing R than any other public health interventions and is 2-3 times more effective than most individual containment measures (Figure 2.19).
Figure 2.19. Percentage of countries at different stringency levels for testing and contact tracing
Note: Based on the Oxford Covid-19 Government Response Tracker.
Source: OECD calculations based on Hale, T., et al. (2020), “Oxford COVID-19 Government Response Tracker”, Blavatnik School of Government, Oxford University.
Shielding the elderly
The elderly population is especially vulnerable to COVID-19 with much higher mortality rates than other demographic groups. A particular concern is that mortality rates have been very high in care homes in some OECD countries (ECDC, 2020; Gandal et al. 2020). The current empirical work tests for the effectiveness of three types of government policies targeted at the elderly or care homes using variables constructed by the OECD (Table 2.5): firstly, recommendations to persuade the elderly to stay at home; secondly, restricting visits to care homes; and thirdly, testing of residents and/or staff of care homes. Measures to specifically protect the elderly were relatively rare in mid-March, but have become more common across countries since then.8
The empirical analysis provides evidence that each of these policies can play a role in shielding the elderly population. The combined effect of these polices on reducing R is estimated to exceed the effect of most individual containment measures (Figure 2.18).
Mandating mask-wearing
Evidence from both clinical trials (Raina et al., 2020) and empirical analysis of public policy pronouncements at the regional or country-level (Leffler et al., 2020; Hatzius et al., 2020; Mitze et al., 2020), increasingly suggests that face masks can provide protection against the transmission of the coronavirus. This is especially true in closed and densely packed spaces and because a considerable share of infected people show no symptoms but have a high viral load. In the current study, mask wearing is investigated using variables constructed by the OECD, which denote whether there is an obligation to wear masks in shops, public transport or more generally in closed spaces (Table 2.5). While few countries had mandatory mask-wearing in closed public spaces in mid-March, a majority of OECD countries had adopted such measures by end-July.
The empirical analysis suggests a negative effect on R from the introduction of mandatory mask wearing in all closed public spaces (Figure 2.18), although other results (not reported) suggest that extending mask wearing obligations to the outdoors does not appear to add much to reducing the reproduction rate.
Table 2.5. OECD scoring of additional public health measures
Source: Constructed by the OECD using text search in three COVID-19-related databases.
The effect on R of more cautious behaviour and moving towards herd immunity
In addition to the variables representing policy responses, the estimation also includes different measures of the death rate from the virus as explanatory variables. Both the national and global daily death rates are included to proxy for general awareness of the virus prompting more cautious behaviour, for example voluntary physical distancing and increased hand-washing. The importance of these variables is that they proxy for changes in behaviour that are likely regardless of government‑mandated restrictions.
Total national deaths attributed to the virus expressed as a share of the population are also separately included as a proxy for the share of the population that has been infected, with the expectation of a negative coefficient; as the share of the population that has been infected rises (and presumably becomes immune, though for an uncertain period), the speed with which the virus spreads should be reduced.
These variables are all statistically significant with the expected negative sign and their magnitudes imply they play an important role in the evolution of R.
The global daily death rate has fluctuated around 0.5 per million during the period considered which, from an initial value of R0 of 3, would be expected to reduce R by about 0.6.
The national daily death rate varies substantially, both across countries and over time, but for some OECD countries it was running at around 15 per million going into the lockdown in March, and this would reduce R by a further 0.6.
The total national death rate also varies substantially across countries and has been increasing relentlessly in most countries. It is used here to proxy the profile of the number of people that have already been infected (and so are subsequently immune), so helping to reduce R. In a number of major OECD countries (including the United Kingdom, Spain, Italy and France) the total death rate currently exceeds 400 per million, at which level R would be reduced from 3 to 2.5.
Scenario analysis
In order to draw out the policy implications of the estimation results described above, they are used to construct a number of stylised scenarios to follow the evolution of R and mobility from the first outbreak of the virus, through full lockdown measures, followed by a number of alternative containment strategies (Table 2.6, Figure 2.20).
At the first outbreak of the virus, for the typical country, the initial reproduction number R is estimated to be about 3 and, before the impact of the virus is felt on the economy, mobility is normal (represented by the red triangle at the top right-hand-side corner of Figure 2.20). Even before the implementation of government-mandated measures, awareness of the seriousness of the virus (represented by the daily death rate) is likely to reduce mobility and foster more cautious behaviour, leading to a fall in R, although it remains well above 1 (the red triangle-labelled “Pre-lockdown + natural caution” in Figure 2.20 which is calibrated on the daily death rates of a number of major OECD economies just prior to the lockdown).
Once the number of daily infections is high (here proxied by the high national daily death rate), the implementation of a wide range of containment measures will be essential to contain the spread of the virus. In the scenarios considered here, the implementation of full lockdown (FLD) measures, accompanied by a limited test-and-trace regime, reduces R to close to 1, but at the cost of a sharp fall in mobility (represented by the blue squares in Figure 2.20). The degree of stringency with which lockdown measures are applied will determine the extent of the fall in mobility, with two scenarios considered here: the first assumes that containment policies are applied with a degree of stringency which is typical of that followed by countries in March/April (calibrated on the response of the median country); the second assumes all containment policies to be applied to their maximum possible degree of stringency. Mobility falls by more than 40% in the former case and by more than 60% in the latter; however, the estimation results suggest there is little additional benefit in terms of lowering R from maximising the degree of stringency of containment policies (particularly with regard to workplace closures or stay-at-home requirements).
Even in the absence of further policy changes, R will evolve during a lockdown as the number of infections/deaths change. The fall in the daily death rate may tend to lower natural caution and so lead to some increase in R and mobility; on the other hand, as the total number of individuals that have already been infected and are immune rises, then this will tend to lower R. The estimation results and particular calibrations used in constructing these scenarios suggest these two effects roughly cancel each other out.
A number of strategies for avoiding or exiting full lockdowns are considered (represented by the green circles in Figure 2.20). The basic issue facing policymakers is how to prevent the need for a full set of containment policies while bringing or keeping R under control. The estimation results explaining R, summarised in Table 2.6, suggest that the implementation of a comprehensive test and trace policy together with a package of other public health measures can more than compensate for the removal of lockdown policies, so that their successful implementation would see a return to near normality of mobility, with R remaining below 1 (as represented by the green circle labelled “No LD + full health measures” in Figure 2.20).
A larger reduction in R would be achieved, if comprehensive public health measures were accompanied by maintaining some limited containment policies. Bearing in mind their impact on mobility, the containment policies which appear the most obvious candidates for being extended are:
Restrictions on international travel, including obligations to quarantine all arrivals from selected countries, which would reduce R significantly and may have only a small effect on mobility (although this may be because the mobility measure does not capture international mobility accurately).
Restrictions on gatherings has a substantial effect on reducing R, whereas the cancellation of public events (which would seem to be inevitably linked) has a relatively small effect on mobility. Such policies may be particularly effective because such large public gatherings may otherwise represent a risk of being so-called “superspreader” events.
Such a package of measures would generate a more decisive reduction in R below 1, although it would come at some cost to mobility (Partial LD + full health measures” in Figure 2.20).
In practice, implementing a full range of public health policies and a comprehensive test, trace and isolate regime may be difficult, especially if the daily infection rate begins to rise. Variant scenarios with “limited health measures” assume only a limited test-and-trace regime together with mandating mask-wearing in indoor public places, but no other public health policies targeted at the elderly or care homes. Such a combination of policies accompanied by a full relaxation of lockdown measures might see mobility initially return to just below normal levels (assuming the daily death rate has previously been reduced by the lockdown), but R will likely increase well above 1 (represented by the scenario labelled “No LD + limited health measures” in Figure 2.20). However, this situation would not represent a stable equilibrium, as with R above 1 there would be a subsequent pick-up in infections and deaths, which in turn would further reduce mobility, regardless of any further government action.
A limited set of health measures accompanied by maintaining the same limited containment policies, would come at a more immediate cost to mobility, but bring R down by more, although in the scenario considered here it would still remain above 1 (“Partial LD + limited health measures”), and so would not represent a sustainable situation.
Figure 2.20. Stylised scenarios: from the first outbreak of the virus, through lockdown and exit
Note: The points represent scenarios, each of which are generated from consistent combinations of the equations for R and mobility using assumptions for the explanatory variables that are summarised in Table 2.6. The red triangles denote the situation at the start of the virus outbreak, blue squares the situation following full lock-down policies, and the green circles represent various exit scenarios.
Table 2.6. Scenario assumptions and outcomes for R and mobility
Note: The assumptions here correspond to the scenarios illustrated in Figure 2.20.
Bibliography
ECDC (2020), “Surveillance of Covid-19 at long-term care facilities in the EU/EEA”, European Centre for Disease Prevention and Control Technical Report, 19 May.
Gandal, N., et al. (2020), “Long-term care facilities as a risk factor in death from Covid-19”, 13 July, published on VOX, CEPR Policy Portal (https://voxeu.org).
Hale, T., et al. (2020), “Oxford COVID-19 Government Response Tracker”, Blavatnik School of Government, Oxford University.
Hatzius, J., D. Struyven and I. Rosenberg (2020), “Face Masks and GDP”, Goldman Sachs Global Economics Analyst, 29 June.
Haug, N., et al. (2020), "Ranking the effectiveness of worldwide COVID-19 government interventions", Nature Human Behaviour.
Leffler, G., et al. (2020), “Association of country-wide coronavirus mortality with demographics, testing, lockdowns, and public wearing of masks”, mimeo.
Magnusson, K., et al. (2020), "Occupational risk of COVID-19 in the 1st vs 2nd wave of infection", medRxiv, 3 November.
Mitze, T., et al. (2020), “Face Masks Considerably Reduce Covid-19 Cases in Germany: A Synthetic Control Method Approach”, Covid Economics, No. 27, 9 June.
OECD (2020), “Walking the tightrope: avoiding a lockdown while containing the virus”, OECD Policy Responses to Coronavirus (COVID-19), OECD Publishing, Paris.
Raina, M., et al. (2020), “Human coronavirus data from four clinical trials of masks and respirators”, International Journal of Infectious Diseases, 96, 631-633.