THE federal government has decided to substantially ease the national lockdown imposed to suppress the outbreak of Covid-19. This is easily among the most difficult, consequential as well as contentious decisions any government has had to take in Pakistan’s recent history.
Governments around the world, including the most developed ones, have struggled to make timely (as well as even the right) decisions with regard to the complex issues thrown up by the Covid-19 pandemic. At the heart of the complexity are social choice questions surrounding the ‘lives versus livelihoods’ debate. In a nutshell, two important ones are inter alia:
1) Should the economic livelihood and generational prosperity of a younger, healthier cohort within society be sacrificed to protect the remaining years of a relatively smaller (though still sizeable in many countries) older cohort — one that may be afflicted by co-morbidities?
2) With a low population-wide case mortality rate, and the possibility of ‘herd immunity’ being developed with exposure of the broader population over time, does shutting down the entire economy for an extended period of time to suppress the virus make sense, especially given the economic structure and realities that face millions (informal workers/daily wage earners in particular)?
Flawed reasoning and muddled maths are informing government action.
Arriving at clear-cut decisions based on scientific ‘evidence’ is clouded by the added complexity provided by the novelty of this strain of coronavirus, where some of the underlying assumptions in the arguments are far from settled (such as the questions of herd immunity or lower susceptibility of the younger population).
In Pakistan, the federal government has generally muddled along in its response since late February, with its lack of clarity reflected in mixed messaging (and weak enforcement) viz the lockdown. As the economic costs and consequences mount in magnitude and significance, the government is finally at the end of its tether, and is relying on a combination of genuine humanitarian concern and untenable logic (mixed with muddled maths) to justify its predisposition to a partial, short duration lockdown.
The national coordinator responsible for leading the government response has compared the current mortality rate from Covid-19 to deaths from road accidents each year, and arrived at the conclusion that since the latter kill more Pakistanis each year and yet road traffic is not banned, the same logic should apply to Covid-19. The flaw in his reasoning? Unlike Covid-19, road accidents are not known to be contagious — and will follow a linear path over a given period of time for a country. The even more dangerous part of his argument? The muddled maths.
Ignoring the exponential nature of the mathematics associated with virus pandemics, the minister has extrapolated linearly from the average number of daily deaths so far (the early phase of the outbreak) to a monthly total of 720. However, if on day one (‘t’) there are 24 deaths, with a case fatality rate (CFR) of 2.3 per cent, it implies total confirmed cases of 1,043. With a basic reproduction number (R-zero) of around two, the total confirmed cases will spike to over one million by day six. Within less than two days of reaching that milestone, Pakistan’s entire population (and more) will have been infected.
Of course, the actual trajectory the virus is following is considerably different — and lower — so far, reflecting the unknowns regarding the actual number of infected cases in the absence of sufficient testing, the infectiousness rate (R0) etc. Nonetheless, it does not detract from the fact that mathematical epidemiology is exponential, not linear. So far, the federal government’s policy response appears to be a case of policy-led evidence, rather than evidence-led policy.
Another prop the government has used is a recent influential Yale University study, which uses a ‘Value of Statistical Life’, or VSL, framework to monetise the trade-off between saving lives and protecting livelihoods. While applying VSL to quantify the benefits of mortality risk reduction in cost-benefit analysis of public policies is a popular method, it suffers from limitations. Individuals are likely to misperceive mortality risk, especially in developing country settings. Altruistic concerns are ignored or not captured in the computation of VSL (for example, saving the lives of relatives or others) — as are distributive distortions.
The most critical, and flawed, assumption of the Yale study is that the Covid-19 outbreak has no effect/impact on the functioning of the economy. In suggesting a strategy for lifting the lockdown in poor countries, it assumes that the economy will resume functioning even while a major epidemic is going on nationally. Hence, it overstates the economic benefit of lifting restrictions.
On the other hand, as the virus rages and the limited critical-care facilities in the country are swamped, it will create a national panic, with fearful customers/buyers, absentee workers, etc — leading quickly to the need to reintroduce harsh suppression measures (including tighter lockdowns). The economy will be negatively impacted, and the country will end up with the worst of both outcomes: a higher death toll as well as lower output and economic welfare. The net benefit is very likely to be negative in welfare terms. (A better policy framework is provided by Debraj Ray et al, on the Vox CEPR policy portal (April 8, 2020)).
Going forward, the only practical option to opening up the economy in a phased manner is for a surge in the capacity of the public health system, especially its critical care facilities. The government needs to introduce frequent mass testing, which is both mandatory as well as free of cost, for those elements of the workforce with the greatest public contact (and hence, most at risk). And self-isolation should be enforced for the at-risk population.
Unfortunately, the atrophy of state capacity means that the government is left with few good options. It cannot provide social safety nets for a large part of the population, nor can it ramp up quickly enough the country’s critical care health facilities. The same applies to its testing, surveillance, contact tracing etc capacity. In this backdrop, it is preparing to ease the lockdown without essential prerequisites in place.
The writer is a former member of the prime minister’s economic advisory council, and heads a macroeconomic consultancy based in Islamabad.
Published in Dawn, May 8th, 2020