The pandemic tends to make us immune to shock when it comes to extraordinary statistics. There was barely a murmur last week, for example, when it was announced that, during the financial year 2020/21, UK government borrowing topped £303 billion—a peacetime record of 14.5 per cent of GDP.

Nevertheless, another striking statistic recently caught my eye as indicating just how profound this period has been. Kip Viscusi, the renowned U.S economist, has calculated the direct mortality costs of Covid-19 across countries. His estimates now imply that the UK’s 127,410 Covid-19 deaths represent a collective loss in value of £717 billion—around a third of annual GDP.

Non-economists recoil at putting monetary “values on human lives” like this, while politicians prefer talking as if every life is infinitely valuable. But Viscusi follows Thomas Schelling, a Nobel prize-winning economist, in believing people’s willingness to pay to mitigate deaths risks, or to be paid to bear them, provide us with a reasonable guide to how much we should societally value mitigating mortality risks.

What does that mean in practice? People are often compensated to bear elevated mortality risks at work. Studies assessing these pay-risk trade-offs in dangerous jobs allow economists to calculate how much workers must be paid collectively for the statistical probability of any one death.

This represents a risk-specific “value of a statistical life” (VSL). For similar death risks to Covid-19, and adjusted for the UK’s income level, Viscusi estimates a population-wide £5.6 million VSL for Britain. Little wonder, then, given these stakes, that the public was willing to tolerate such costly disruption via lockdowns.

Lockdown sceptics, of course, dispute using population-wide VSLs. The age profile of death from this virus skews much older than most work risks, after all. Indeed, the largely unspoken message of those who continuously repeat that “the average age of death from Covid-19 is 80” is to support the “good innings” principle. Those most likely to die from Covid-19 are said to have “had a good run,” haven’t got long to live anyway, and would otherwise live in ill-health. Charles Walker explicitly said in Parliament “to compare the death of someone of 90 with the death of someone of 19 is not right.” When totting up the benefits of risk mitigation then, shouldn’t we adjust for age and quality of life?

Although valuing lives differently by age is itself ethically debatable, I agree that using a one-size-fits-all value is probably misguided. As I write in my own book, Economics In One Virus, the death risks faced by older people for Covid-19 are much, much higher than those faced in most workplace studies. As Walker’s constituency correspondence highlighted, at the very least the elderly are likely to have more varied preferences about bearing Covid-19 death risks than, say, miners with accidents. Importantly, we generally see a decline in the observed market value of a statistical lives derived from studies of the very elderly, at least after a peak in middle-age.

But the key point is that though we, societally, often assume young lives are “worth” much more, perhaps hundreds or thousands of times more, academic assessments of people’s behaviour in markets often finds values of statistical lives as high for someone aged 80 as someone aged 20. At most, they tend to differ from oldest to youngest by a factor of seven.

When pushed to adjust for age, Viscusi’s own work suggest that the average value of a statistical life given the age profile of Covid-19 deaths so far would still be between £2.7 million to £3.2 million per life. That would imply overall still huge mortality costs of around £344 billion to £408 billion. And that’s before we even try to account for the non-mortality risks associated with widespread disease, which people self-evidently are keen to avoid. Clearly, then, mitigating Covid-19 risks always offered potentially huge benefits.

There are two common reasons, I suspect, why people might wrongly presume that controlling for age would reduce these values more significantly. The first is misunderstanding life expectancy. That life expectancy at birth is 79.4 for males and 83.1 years for females doesn’t mean that someone at the average age of a Covid-19 death (80) is due to die imminently.

In fact, the UK’s National Life tables show someone who has reached 80 would, on average, live another eight and a half years if a man and 10 if a woman. That’s because life expectancy at birth averages across everyone, including those who will die much younger. So, even accounting for ill-health, a large proportion of Covid-19 deaths will still bring a significant loss of life years.

Second, and most importantly, despite our stated preferences, our revealed ones show the latter years of life are simply more highly valuable to people than commonly supposed. As the economist Jeremy Horpedahl argues, these are years in which people are generally at their wealthiest, with most time on their hands, and enjoying the pleasures of seeing their grandchildren develop. That we collectively spend so much on the elderly through the state pension and the NHS, in fact, shows we value the lives of our elders highly.

Why, then, are certain lockdown sceptics so keen to play down the loss of elderly lives? Aside from these errors, I suspect motivated reasoning is at play. After all, it makes no sense for someone who argues for “focused protection” of the elderly to then outline their lives aren’t worth much anyway.

Then again, many of these people celebrated the example of “no lockdown” South Dakota, which saw a larger proportion of elderly deaths in its total than many other U.S. states, despite a younger population. That suggests the ends (avoiding government lockdowns) are what justify the means (whether that’s disputing the worth of elderly lives in some contexts or arguing for focused protection in others).

My argument here should not be misconstrued: the fact that these market measures suggest the death costs of Covid-19 are massive isn’t a slam dunk for the action we’ve seen this past year. What matters when calculating the benefits of lockdowns, of course, is working out how many lives the policies have purportedly “saved,” not how many people died anyway. These then have to be weighed against the broader social costs of any restrictions on our lives, which have no doubt been very large too.

What Viscusi’s numbers do show, however, is that the decisions we make in markets imply we value our lives extremely highly—much more highly, it seems, than the sorts of threshold values used by the NHS’s rationing board, the National Institute for Health and Care Excellence. And those losses and potential losses of value are surely important in evaluating how we should have responded this past year.