The New Zealand government has claimed to be adopting a scientific and risk-based approach to the introduction of policies and regulations to implement its objective of eliminating Covid-19 from New Zealand. A suite of policies, including strict controls on who can cross the border, associated with stringent quarantine and testing, and physical isolation of any individual testing positive for the virus (and their close contacts if necessary) have resulted in extremely few cases of community transmission being identified. When community transmission has been identified, a first-rate testing and contact-tracing system combined with the use of localised lockdowns has led to rapid containment of infection breakouts. In the most recent example (February 2021 – the Auckland February cluster), a breakout first reported on February 13, leading to a Level 3 lockdown in Auckland, and level 2 for the rest of the country, had been sufficiently satisfactorily contained for the government to relax the restrictions from February 18. However, the subsequent emergence on February 27 of a case with apparently only an indirect association with the Auckland February cluster resulted in the reinstatement of Level 3 status in Auckland and Level 2 for the rest of the country, for seven days from February 28.
Risk and uncertainty
When announcing the most recent lockdowns, the Prime Minister stressed that the moves were consistent with practicing an “abundance of caution”. It is not the purpose of this comment to pass judgement on whether these decisions were warranted. It is presumed that the circumstances surrounding the emergence of the February 13 case constituted an increase in the risk of community transmission escalating, and this is why the change in lockdown levels was imposed. Likewise, it is presumed that when the lockdowns were relaxed from February 18, this was because the risk (or perhaps more correctly, perceptions of the uncertainty about the level of risk) had reduced. The rest of the country returning to Level 1 suggests that the risk in that territory had returned to the pre-February 13 level. That Auckland was reduced to Level 2, pending a revision on February 22, indicates the risk there had fallen, but not yet to the pre-February 13 levels.
The reinstatement of lockdowns from February 28 reflects the abundance of caution. There was an increase in risk of virus transmission, as the new case had not isolated after being tested, as expected in the Covid management plan. This is distinct from the uncertainty surrounding the success of the 3-day lockdown in containing virus spread, which some have claimed was ended prematurely, because there may have been an over-estimation of the reduction of risk of virus transmission. The “surprise” subsequently revealed that not all contact information had been shared with contact tracers by the cases interviewed earlier in the development of the cluster illustrates how a lack of information (or in this case, incomplete information), itself creates a risk if decisions are made under the false presumption that the information available is all there is to know about a situation. However, that is not the substantive point of this discussion.
Risk, decision-making and signalling
What this comment does seek to address is the risk-based justification for the policy announced on February 17, to apply from February 18 2021, that the wearing of masks by passengers on public transport in all areas under Level 1 lockdown be mandatory. Mask-wearing on public transport was made mandatory at Level 2 on August 30 2020. From November 19 November 2020 it was mandated for Auckland under Level 1, presumably due to the higher population density and larger population in that area meaning transmission risk, if the virus was present, was higher there than elsewhere in the country.
Under a risk-based management system, the introduction of a new constraint would be justified under two circumstances: new information coming to light that the proposed constraint would lead to a demonstrable reduction in risk in the specific area in which it was applied; and a demonstrated increase in risk in the relevant area that could be effectively mitigated with the measure.
The question is raised therefore, why the requirement for mandatory mask wearing on public transport across the whole country was implemented, when it was. It does not appear to meet either of the above rational justifications. The WHO document on mask-wearing policies linked on the Ministry of Health website to support its decision was published on 1 December 2020. In the fast-moving world of Covid-19 management, this hardly counts as “new information”. Neither does there appear to have been a demonstrable increase in risk of transmission – indeed, the contemporaneous lowering of the lockdown levels was, as indicated above, the consequence of a demonstrable (or at least perceived to be so by the decision-makers) reduction in risk of community virus transmission.
At best, then, the mask-wearing mandate appears inconsistent with the risk-based approach for managing Covid-19. Should citizens infer from the government’s actions that the risk of community transmission has reduced (as per the lockdown decision) or increased (as per the mask decision)? The inconsistency is not trivial. The uncertainty itself creates a risk in the minds of citizens: some observers at least have been left wondering if the government “knows something” about the risk that it is not revealing.
From the perspective of Covid transmission, the WHO document states there is “only limited and inconsistent scientific evidence to support the effectiveness of masking of healthy people in the community to prevent infection with respiratory viruses, including SARS-CoV-2 (75).” To the extent that there are benefits, it appears to be in scenarios where there is “known or suspected community or cluster transmission of SARS-CoV-2” – that is, Auckland during the recent lockdown. Where there is “known or suspected sporadic transmission, or no documented SARSCoV-2 transmission”, then a risk-based approach to mask-wearing is recommended. As there has been no demonstrated community transmission of Covid-19 over most of the country since June 2020, then not only is the relative risk unchanged by the recent events in Auckland, but the absolute risk too is extremely low.
So following the science leads to the conclusion that there was no justification.
However, decisions such as the mask-wearing one do not occur in isolation. Managers and policy-makers must balance a wide range of objectives when making decisions. In addition to the physical risks to the population of Covid-19, the government must also manage the risks associated with its reputation and perception amongst voters. When uncertainty prevails, individuals tend to react by favouring actions being taken (even if they are unjustified or just plain wrong) over inaction (even when it is the result of a reasoned analysis), which can be considered as a weak or indecisive response. The mere act of “doing something” can convey a sense of being in control, even though it has no effect on the outcome (for example, panic-buying of toilet paper). For politicians in a time of crisis, being seen to do something is the natural response to manage the risk to their political reputations in the face of the new uncertainties amongst voters.
In that light, the mask-wearing mandate does come from a risk-based approach to decision-making – but where the risk concerned is a political one and not a Covid-19 related one. The constituency being protected is the government, and not the population of New Zealand. It may be dressed up to look like the actions of a caring government, but as the above analysis shows, it is not consistent with that objective. The government must take care though – in protecting itself in the short-term, it has created new uncertainties from the inconsistencies in its risk-based Covid-19 policies that might ultimately feed back into negative reputational effects.
 There is much debate in the literature about the distinction (or coalescence) between risk and uncertainty. For the purposes of this comment, as uncertainty creates the perception of risk, they will be treated as equivalent.
 John Kay and Mervyn King (2020). Radical Uncertainty: Decision-Making Beyond the Numbers. London: The Bridge Street Press. Daniel Kahneman. (2011). Thinking Fast and Slow.