John Gibson: Lockdown Again
So New Zealanders find themselves in a strict lockdown again, with bureaucrats dictating what businesses are essential and what not. Our politicians and health bureaucrats seem to have learnt nothing in the last 18 months. They are still using the blunt, costly, and ultimately ineffective tool of locking down both the sick and the healthy to deal with a very age-specific disease with well-known comorbidities (particularly obesity).
Here is a quick review of what they should know about the virus and the disease:
· Covid is far less deadly than initially feared. Even after just a few months the Bulletin of the World Health Organization reported an estimated Infection-Fatality Rate (IFR) is just 0.23%, summarizing outcomes from many countries. For countries with below average Covid death rates, which includes New Zealand, the median IFR is 0.09%; equivalent to one death for every 1100 infections. This is much more like the 1957 and 1968 pandemics, yet the response to Covid has been massively more costly.
· The risk of dying from Covid-19 is very skewed towards the elderly. Compared to the Infection-Fatality Rate (IFR) for people age 0-34, the IFR for the 65-74 age group is 625 times higher, for the 75-84 age group it is 2125 times higher, and for 85+ age group it is 7075 times higher.
· Contrary to media fear-mongering, Delta is not more lethal than earlier variants. Data from Public Health England, on all sequenced and genotyped cases from February 1 2021 to August 15 2021, shows a Case-Fatality Rate (which is far higher than the IFR as many infections go unnoticed) of 1.1% for the Alpha (Kent) variant, and just 0.3% for the Delta variant. For the at-risk elderly group (age ≥ 50, as the only breakdown given) the CFR for the Alpha variant was 4.8%, and just 2.2% for Delta. Likewise, from India where Delta emerged, a recent study concludes (p.7) “the countrywide data show a continuous sharp decline in the fatality rate after an initial surge, thereby suggesting that despite the burst of mutations in the spike proteins of Indian isolates, the new variants on average have not severely affected the COVID situation…”
Although it is an ever-changing landscape, there is also something crucial that the politicians and health bureaucrats should know about the available vaccines. They are more like flu shots in that their efficacy wanes quickly and so can be quite low by the time needed.
· A recent paper by Pfizer’s own scientists shows vaccine efficacy (VE) for infection with the Delta variant falls by 10 percentage points per month, to be just 53% if the second jab was more than 4 months ago. Falling protection was almost as fast against other variants, declining by 8 percentage points per month. When they break out the results by age, VE for the at-risk elderly (≥ 65 years) starts low and wanes at a similar rate as for all-age groups: VE against infection (from all variants) is highest at 80%, within a month of the second jab, and by four months later VE is down to 43%.
· The efficacy after six months is unclear because the original trials were unblinded (the participants were told if they got the placebo or not and most control group members then got the vaccine) and the number of Covid-related deaths in the original trials was too small (two in the control group, one in the treatment group) to draw conclusions.
It was extremely misleading of the Prime Minister to claim that Covid-19 could be treated like measles, given that measles vaccines offer lifelong protection, with VE close to 100%. As long as elimination is the goal, lockdowns are likely to be imposed, irrespective of variant and of progress with vaccination. So what do we know about lockdowns after 18 months?
· Early studies claiming lockdowns work compared infections with modelled viral spread. There are two problems: infections data depend on how hard one looks, in terms of testing rates and PCR test cycles, and so are inconsistent over time and space. Even worse, these models ignore people privately altering their behaviour when facing the risk of Covid-19. The headline results from the (in)famous Imperial College paper of Ferguson et al assume an absence of “spontaneous changes in individual behaviour” (p.6). Yet empirical study of actual humans (e.g. their shopping patterns) finds almost 90% of response was private action, and so the models greatly overstate lockdown efficacy (and are too alarming about the no-lockdown option).
· Subsequent studies looked at Covid-19 deaths, as the outcome that really matters, and also because deaths data should be more reliable than infections data. Variation in the strength of lockdown (or lockdown versus no lockdown) is not related to variation in Covid-19 death rates. Yet the question of death with Covid versus death from Covid affects these studies, especially as jurisdictions with good public health data start to retrospectively revise Covid death totals, which were inflated by up to one-third.
· The most recent studies look at all-cause mortality, as these data have fewer biases (one is dead or not, irrespective of cause) and also show lockdown collateral damage. Across European countries, stricter lockdowns did not reduce excess mortality, while across 43 countries and all U.S. states excess mortality rose following the imposition of lockdowns. Local defenders of lockdown may note New Zealand had no excess mortality in calendar year 2020. However, an unprecedented post-lockdown surge in deaths that carried on through the summer into 2021 largely reverses that pattern if the full 12 months after the first lockdown is considered.
Putting all of the above together, a thorough review by Canadian economist Douglas Allen (familiar in NZ, as he has been an Erskine visitor at the University of Canterbury) concluded that lockdowns will go down as one of the greatest peacetime policy failures. As to why they do not work, SARS-CoV-2 spreads via aerosols but most messaging (e.g. surface cleaning, keep two metre distances) assumes droplet spread. Locking people up in poorly ventilated houses and discouraging outdoor activity might perversely increase spread of an aerosolized virus. Some prevention measures are subject to economies of scale, so once the nature of viral spread is understood, retro-fitted workplaces could become safer than houses.
While lockdowns have not worked to reduce deaths in the present, they almost certainly harm future life expectancy. This especially matters for New Zealand, due to the following:
· The long-run relationship between the real value of our economic activity (what we produce) and life expectancy is 50% higher than the OECD average. A 10% fall in real GDP, in the long-run, reduces NZ life expectancy 1.8% below what it otherwise would be. So these trade-offs should matter more here than elsewhere.
· Real GDP in 2020 was 5.2% below expectation (using the last 2019 fiscal update). Part of this fall was outside our control but much is from a “go hard” approach. With a Covid response stringency of the median OECD country our 2020 GDP growth rate would be 3 percentage points higher (lockdown stringency is unrelated to mortality so this is all pain no gain). The $14 billion of output not produced in 2020 is not shifted through time, it is a permanent loss. The same will be true for the 2021 lockdown. A share of future output also has to go on debt-servicing, as NZ tried to borrow her way out of this pandemic, and so is not available to fund improvements in life expectancy.
· New Zealand residents currently alive can expect 224 million more life years (based on the latest 2017-19 period life tables). If real GDP ultimately falls 10% below what was expected pre-Covid (e.g. if Level 4 lasts as long as in 2020), and if we apportion half of this to the unusually harsh NZ response (rather than to overseas factors), then our politicians and health bureaucrats will have presided over a fall in life expectancy where there are two million fewer life years than would otherwise be expected. If this loss was full concentrated on a select group, it is equivalent to 46,000 deaths.
· In case one doubts these calculations, note that the Treasury long-term fiscal forecasts released recently show future life expectancy is almost two years below what they had previously forecast in 2016.
Thus, the apparent kindness of locking down to limit Covid-19 deaths will, instead, be killing more people by making us poorer. Just as Douglas Allen concluded (for Canada), so too for New Zealand—lockdowns are one of our greatest peacetime policy failures.
John Gibson, Professor of Economics, teaches at the University of Waikato. He previously taught at the University of Canterbury and Williams College, was a research visitor at the Centre for the Study of African Economies, University of Oxford and is an Associate Researcher at the LICOS Centre for Institutions and Economic Performance at KU Leuven. He received his PhD from Stanford University and has since worked around the world in countries like Cambodia, China, India, Papua New Guinea, Russia, Samoa, Solomon Islands, Thailand, Tonga, Vanuatu, and Vietnam. He is a Fellow of the Royal Society of New Zealand and a Distinguished Fellow of the New Zealand Association of Economists and of the Australasian Agricultural and Resource Economics Society.