The there-is-no-trade-off-between-health-and-the-economy brigade are getting very shrill.
There is an op-ed that appeared online today and will probably appear in print tomorrow that is very interesting.
Economists in Australia and around the world have come to a near-unanimous conclusion about how best to handle the coronavirus pandemic: we must get infections under control and keep the reproduction rate below 1.
Now you don’t need to be an economist to come to that conclusion.
In a pandemic, what is good for public health is also good for the economy.
I’m more agnostic on this point. Many people are somewhat surprised by my view here. This seems to me to be an empirical question. It may well turn out that this is the case, or is may not. What has surprised me is the extent to which anyone who has wanted to debate the point has been vilified (see here and here).
That said, a small – but vocal – contingent of dissenters have argued that legislators should use a seemingly scientific method to calculate the economic value of different lives. Were they correct, this grim calculus would have us evaluate a policy by weighing the costs of, say, a 25-year old’s reduced lifetime job prospects against the benefits of an 80-year old staying alive for the remainder of their life expectancy.
Notice the framing?
Were they correct, this grim calculus would have us evaluate a policy by weighing the costs of, say, a 25-year old’s reduced lifetime job prospects against the benefits of an 80-year old staying alive for the remainder of their life expectancy.
That should be:
Were they correct, this grim calculus would have us evaluate a policy by weighing the benefits of, say, a 25-year old’s
reducedlifetime job prospects against the costs of an 80-year old staying alive for the remainder of their life expectancy.
Notice also the change in terminology – dissenters – not economists, but dissenters. Now they aren’t picking on economist Gigi Foster, but rather the Vice Chancellor of Melbourne University. Duncan Maskell actually specialises in bacterial infectious diseases. To be fair – COVID19 is a virus, but you’d imagine that there is a lot of commonality around infectious diseases. He was recently interviewed in The Age:
Professor Maskell says decision-makers must consider the role of quality-adjusted life year (QALY), a unit of measurement used by economists to predict and assess the impact of health policies. In simple terms, it assumes that a life near its end, whether because of disease or advanced age, is empirically different to a healthy life closer to its beginning.
“We have to look at this as an overall picture. My personal view is there should be some form of sensible, public health, QALY-based analysis done and tough calls made. It boils down to a basic but very hard moral philosophy: What is the value of a 90-year-old’s life versus the value of the continuing livelihood and happiness of a 25-year-old?’’ he said.
Notice the switch. He says, “livelihood and happiness of a 25-year-old”. They say, “lifetime job prospects” i.e. livelihood only.
That is a very tough call to make, but our intrepid economists are happy to quibble.
Like many debaters with a dubious argument, this group often relies on appeals to authority.
Remember that; “appeals to authority”.
Medical professionals and health economists “use QALYs all the time”, they say. And yes, QALYs are frequently used in those fields – but for narrowly defined purposes that have nothing to do with making interpersonal comparisons of the value of lives.
I’m going to call “Bullshit” on this point. Maybe not in the academic literature, but I reckon that medical professionals do use QALYs for making interpersonal comparisons of the value of lives. Here is Dominic Lawson explaining how the NHS in the UK does triage.
As directed by the National Institute for Health and Care Excellence (Nice), the NHS will agree to fund non-palliative treatment for the seriously ill only if the drugs provided (for example) do not cost more than £30,000 in providing a further year of good-quality life. This is known as a Qaly (quality-adjusted life year). Since those with potentially terminal illnesses don’t have a good quality of life, that £30,000 limit would typically be cut to £15,000 per annum in their case — two of their years being regarded as of the same monetary value (in health accounting terms) as one year for an otherwise fit person.
As Sir David Spiegelhalter (who for decades ran the Medical Research Council’s Biostatistics Unit) put it to me: “It’s a myth that all lives are considered equally valuable in the NHS. The system already values years of remaining life given by treatment. So triage would be normal.” What Spiegelhalter means is that if the coronavirus crisis leads to a situation in which an older person is essentially left to die, in order to provide a scarce ventilator to someone with a greater life expectancy, that would be a graphic demonstration of the working model of the NHS.
So QALYs are used in practice in precisely the way our economists tell us they are not being used. But moving along … .
The interesting question is whether QALYs should be used like that?
Proponents of applying QALYs to pandemic policy often justify this huge logical leap by invoking the notion that assigning value to lives in this way is justified by economic theory.
I’m not entirely sure that economic theory says this, but okay …
In fact, that’s very far from the truth. As (excellent) economists Han Bleichrodt and John Quiggin pointed out 20 years ago: “QALYs have no foundation in economic welfare theory.”
So much for appealing to authority and dubious debaters. Anyway, I went and read the actual paper being cited. I decided to examine the precise context in which the statement “QALYs have no foundation in economic welfare theory” appeared. Please note the ” ” marks – that tends to suggest a direct quote. So imagine my surprise when I discovered that the quote did not appear in the paper. The closest expression I could find was:
Notwithstanding the fact that it is currently the most common tool in the economic evaluation of health care, cost-effectiveness analysis, unlike cost–benefit analysis, has no foundation in economic welfare theory.
That is a bit naughty. Now I understand that words are at a premium etc. in an op-ed, but manufacturing quotes, while appealing to authority is not the done thing. But what else do Bleichrodt and Quiggin say?
It is not our intention to resolve the controversy about the role of cost-effectiveness analysis. We only observe that there exists a perception of cost-effectiveness analysis which requires a foundation of cost-effectiveness analysis in welfare economics. This perception provides the rationale for this paper.
The aim of our axiomatic analysis is to reveal the conditions under which cost-effectiveness analysis is equivalent to cost–benefit analysis. Let us emphasize that we do not intend to argue that these conditions have descriptive or normative force.
They are focused on making very precise theoretical points only. It isn’t clear to me that they intended to weaponise their argument.
Back to the present and our current argument. Our economists now argue:
But the most dangerous claim that proponents of QALYs make is that they allow one to make principled comparisons among different groups of people. That is, make a reasoned and reasonable tradeoff between young people and old people, or between sick people and healthy people.
I do have some sympathy for this argument. How principled or unprincipled comparisons are is itself a value judgement. Rule based comparisons, however, are far more likely to be principled.
But it is here that our economist friends go overboard:
QALY-boosters say that by converting life years into dollar values it is easy to make such comparisons. But they have either forgotten – or never learnt – their basic first-year microeconomic theory. We could go into excruciating detail, but the next time some would-be economist tries to bamboozle you with such statements just ask them if they recall that a von Neumann-Morgenstern expected utility function is only defined up to a positive affine transformation.
Wow. There is a lot in that paragraph.
I think here they are having a go at either Gigi Foster or Adam Creighton of The Australian.
Now I admit, it has been 35 years since I did first year micro, 34 years since second year micro, and 33 years since third year micro. So please, folks, please go into excruciating detail explaining your argument.
Let’s play a game:
the next time some would-be economist tries to bamboozle you [insert claim] just ask them if they recall that a von Neumann-Morgenstern expected utility function is only defined up to a positive affine transformation.
the next time some would-be economist tries to bamboozle you [by claiming progressive taxation is economically efficient] just ask them if they recall that a von Neumann-Morgenstern expected utility function is only defined up to a positive affine transformation.
Talk about “appeals to authority” and trying to bamboozle people?
But wait, there is more.
The idea that QALYs allow interpersonal welfare comparisons is a profoundly dangerous and misguided claim. It is unsafe at any speed. Economics has nothing—absolutely nothing—to say about how to make interpersonal “welfare” or “utility” comparisons.
When you mark exams students engage in some deceptive practices. So you combine a true statement with a dodgy statement and hope the examiner won’t notice. But, of course, after many years of practice this sort of thing jumps out at you.
Economics has nothing—absolutely nothing—to say about how to make interpersonal “welfare” or “utility” comparisons.
Now that I do remember from my student days. Positive economics has nothing to say about interpersonal welfare comparisons. James Buchanan used to invoke a Rawlsian veil of ignorance to try get around that problem. But generally that statement is true. Making those sorts of comparisons is something that economists qua economists should not do. That is the role for ethics, morals, politics and so on. It is not that these comparisons cannot be made at all, but that economics, strictly defined as an academic disciple, gives no guidance as to how those comparisons should be made.
So this part of their argument is misleading:
The idea that QALYs allow interpersonal welfare comparisons is a profoundly dangerous and misguided claim. It is unsafe at any speed.
Yes. It is unsafe. It is fraught. But nobody is suggesting that economists make those comparisons. Here is the Melbourne VC:
Though his research expertise is in infectious diseases and microbiology, he says that ultimately, it is a question that can only be addressed by our political leaders, taking into account public sentiment.
Now you’d think we’re all in agreement because this is what our economists say:
Do politicians and policymakers make such judgments every day? Sure they do. It is part of what they are elected to do, and they are accountable to voters for such judgments.
But no. They carry on:
But it is intellectual malpractice to pretend that judgments about how to value different impacts of the pandemic on different individuals is an Excel-spreadsheet-level exercise that doesn’t force us to grapple with weighty moral choices that should give us all pause.
Anyone who pretends otherwise should be ashamed of themselves.
Intellectual malpractice! Take that Melbourne University VC – you should be ashamed of yourself.
I don’t think anyone is understating the moral choices that need be made. Accusing your intellectual opponents of intellectual malpractice and the like is not serious debate.