Here’s the value of your life, according to the Australian government: $213,000 a year. A bit under $600 a day.
As the coronavirus cuts a swathe through the globe, and as Australia weighs the horrible trade-offs between the virus and the economy, the value of a life is a useful statistic. It works as an input to thinking about how we balance lockdown and death.
The figure above for the value of a statistical life year was published by the Australian government in 2019, updating from 2014 when we valued a year of life at $182,000 (inflation has been at work!).
The poets in our midst may be aghast, but placing a dollar value on human life is necessary.
Without it we would not know how much to spend on cancer screening, how high to set speed limits, how much to tax cigarettes. When health bureaucrats make their choices about allocating resources, they must weigh up years of life saved against expenses.
Lives can’t be worth infinite amounts, but they must be worth something. Decision makers need a number.
So how do we decide on the value of a life? One way to find out is to ask people. Pollsters regularly poll people on how they value another year of life.
Another way to determine the value of a life is to observe the decisions people make. How much extra will people pay for a safer car that reduces the chance of death by 1%? How much extra pay do we demand to do a risky job that increases our chance of death by 1%?
The answers to these questions can be extrapolated to find out how much we value our lives.
The figures these methods elicit are controversial. Some people think they are way too high. And it is true that the sums in question are much more money than most people earn in a year. But of course, life is not valued by your paid work alone.
Still, the high value of a life poses problems, according to Dr Peter Abelson writing for Applied Economics. He argues that the marginal value a person might pay to save their own life for mass decision-making might cause us to place too much emphasis on safety:
If we are concerned only with marginal public decisions, it does not matter that the value of a statistical life exceeds individual budget constraints. But, in aggregate, groups of individuals cannot spend more than their total income on saving lives. Given the high proportion of GDP spent on health and safety, the aggregate budget implications and feasibility of basing all public policy for safety and health on individual marginal valuations may require consideration.
This is an obvious concern in a mass fatality situation, such as a pandemic. But the story is not over yet.
In the US, some interventions undertaken by the Environmental Protection Agency save lives at a cost far higher than the value of a statistical life. This is popular. It points to a key problem with the value of a statistical life. We care not only about how long we live. We care how we die.
People would be angry about the injustice of being poisoned by a polluting factory. But at the same time people are often willing to take risks to have fun. We are willing to pay far more to avoid death by pollution than to avoid death in a cycling accident.
This is very relevant to the case of coronavirus. The idea of dying in a crowded hospital where we can’t access a ventilator is frightening. That’s the sort of death we will pay a lot to avoid.
One question for decision makers is whether they perceive this as an irrational fear or a rational one. If we truly valued all deaths equally, perhaps we would be willing to let the coronavirus run rampant.
To do so might be to let the assumption of rational economic man determine policy too much.
Another objection to the estimated value of a statistical life also reflects the way it fails to focus on death enough. The value is determined by asking the person doing the dying — it does not fully take into account the impact of deaths on others.
This in turn has been studied, and according to some the value of a life should be adjusted upward by 10% to 40% to account for grieving relatives.
My modelling of the value of Australia’s coronavirus lockdown — using an 80% infection rate, and fatality rates taken from a study on Wuhan published in The Lancet — suggests the lockdown is saving 187,912 Australian lives, around 90% of which are of people over 50. The value of the life years saved is $635 billion.
That’s a lot. Nearly 30% of GDP.
Can we use this to say that the lockdown has been good value or alternatively not worth it?
Not exactly. The model implicitly relies on a somewhat artificial counterfactual — one where both government and individuals did almost nothing as the virus tore through 80% of the population.
A counterfactual is vital for comparison and the reality is that even if government didn’t react, people would have changed their behaviour. That would have caused some economic damage and saved some lives. The true value of the lockdown is in the lives saved compared to that scenario.
In an unprecedented scenario like a pandemic, finding the right counterfactual for comparisons is hard. This is not like making policy on smoking rates where the data tells you what will happen without an intervention.
Australia has had great success at suppressing viral transmission, which does hint at the idea that a slightly gentler lockdown might still have been successful. Indeed, we might have been able to avoid some economic damage and still saved a lot of lives.
But it is easy to play Monday morning quarterback. In February and March, the government was making decisions under extreme uncertainty and the use of the precautionary principle was appropriate.
One last point about using the value of statistical life years to assess a response to a pandemic. It assumes that the deaths don’t disrupt society at large.
In the case of a pandemic that might be inaccurate. Polls suggest people wanted a lockdown. Undermining people’s faith that society works for them might have been very costly indeed.