Covid Treatment Lottery

“For patients with similar prognosis, who cannot be separated in other ways, a random allocation, such as a lottery, may be used”, says the protocol.

So says a Report in the Daily Telegraph (UK) of 5th January by Paul Nuki titled “Covid ‘lottery’: Doctors draw up triage protocol in the event treatment has to be rationed” (Telegraph usually paywalled, but this seems open-access)

It refers to a paper in J Medical Ethics “Development of a structured process for fair allocation of critical care resources in the setting of insufficient capacity: a discussion paper” also accessible f.o.c.

This is circulating in NHS hospitals as a proposed protocol.

The protocol – drafted by medical, legal and palliative care specialists at the Royal United Hospital Bath NHS Trust – is the most sophisticated attempt yet to devise an ethical system for rationing care in the event that there are insufficient resources to treat everyone.

Now this is exciting! But it is not new. Right from the start of organ transplantation (1960s) such moralistic contentions were weighed up.

In Seattle the so-called ‘God committee’ was set up to make these difficult choices (reported in Calabresi & Bobbit (1978) Tragic Choices). The committee eventually found that it was too agonising to make these choices, and passed the task back to the medical practitioners. In the end it was felt that only medical  factors should be taken into account. Even if no overt rules on social merit were in place, we should not be surprised if the doctor, genuinely uncertain on medical grounds,  was to pick the ‘nicer’ of the two patients.

A secret lottery?

Elster (1989) in his masterly ‘Solomonic Choices’ gives the example of child custody cases, where the judge is frequently unable (in his own mind) to give a clear-cut decision. Yet decide he must, so he goes ahead, dressing up the verdict with trappings of rationality.

This, claims Elster, satisfies both parties, the winner praising the wisdom of the judge, the loser cursing his bias. No doubt a similar process might go on when a medical doctor decides, even if partly randomly and in secret, between her two patients: So long as both patients believe that their case is decided clinically by an expert, then both winner and loser may find it acceptable.

The doctor herself may even be a bit cognitively dissonant—convincing herself that she is doing the right thing for the right reason, exercising judgement based on intuition  rather than validated knowledge. This form of fudging may be acceptable all round, but it is fraught with dangers.

If fakery is suspected, patients rapidly lose their trust in their professionals. Unwitting discrimination seems inevitable. True expertise will fail to develop unless its limits are acknowledged.

Against a lottery is Greely (1977) who suggests that if recipients can argue about any allocation, they feel more satisfied. Anand was also interested in what is called ‘voice’—that one of the reasons a coin-toss was thought to be unfair is that it deprived customers of a say in the decision.

In favour of a visible act of coin-tossing Calabresi & Bobbit explain that it draws attention to the fact that resources are limited. Edgeworth (1888) suggested another benefit would be that the public, seeing a random drawing take place, would be alerted to the ‘aleatory nature’ of the decision. Bureaucrats might not like having such attention focussed on this shortage of resources and their uncertain knowledge.

[This was part of my 2006 thesis Who Gets The Prize. It can be viewed in full on my website]

Wang Shaoguang and Yves Sintomer on sortition

A 2019 hour-long discussion on sortition at Shandong University with Wang Shaoguang and Yves Sintomer was recently published by its moderator, Daniel A. Bell.

This is a rather wide ranging discussion, and its lack of focus is somewhat of a flaw, in my opinion. Ideas on various matters are expressed. Many of those are well-hashed ideas, and the discussants are content to simply repeat them rather than examine them critically.

One idea that I think is relatively novel is briefly offered by Sintomer toward the end when Bell asks for proposals for applying sortition:

I would give the power to citizen juries randomly selected to judge politicians, when they are accused of misbehavior. Because I don’t trust other politicians to do this, as in Brazil or in the USA, where the impeachment is voted by the Congress. I think it’s a bad setting. And I don’t trust judges for judging politicians. Because judges are a very specific, professional body, and very often, a highly conservative body. I trust more randomly selected citizens to judge politicians when they are accused of misbehavior.

Senior: The tyranny of the credentialed

Jennifer Senior writes in the New York Times:

95 Percent of Representatives Have a Degree. Look Where That’s Got Us.

All these credentials haven’t led to better results.

Over the last few decades, Congress has diversified in important ways. It has gotten less white, less male, less straight — all positive developments. But as I was staring at one of the many recent Senate hearings, filled with the usual magisterial blustering and self-important yada yada, it dawned on me that there’s a way that Congress has moved in a wrong direction, and become quite brazenly unrepresentative.

No, it’s not that the place seethes with millionaires, though there’s that problem too.

It’s that members of Congress are credentialed out the wazoo. An astonishing number have a small kite of extra initials fluttering after their names.

According to the Congressional Research Service, more than one third of the House and more than half the Senate have law degrees. Roughly a fifth of senators and representatives have their master’s. Four senators and 21 House members have M.D.s, and an identical number in each body (four, 21) have some kind of doctoral degree, whether it’s a Ph.D., a D.Phil., an Ed.D., or a D. Min.

But perhaps most fundamentally, 95 percent of today’s House members and 100 percent of the Senate’s have a bachelor’s degree or higher. Yet just a bit more than one-third of Americans do.

“This means that the credentialed few govern the uncredentialed many,” writes the political philosopher Michael J. Sandel in “The Tyranny of Merit,” published this fall.
Continue reading

Galland and Schnapper: Citizen conventions and representative democracy, Part 2/2

The second part of an article in Telos by Olivier Galland, sociologist at the CNRS, and Dominique Schnapper, researcher at the EHESS and an honorary member of the Constitutional Council. The first part is here.

2. The choice of those of those responsible for the organization of the discussions, for informing the people convened by selecting the “experts”, for helping them to form an opinion, for writing and disseminating the conclusions must meet specific conditions as well. Who will select the people in those roles and what will be their legitimacy for making choices which may guide the conclusions of the deliberations? On this issue, it is important to distinguish clearly, when organizing the deliberations, between testimonies of scientific experts and those of other actors – activists, representatives of the state, trade unionists and business people. A distinction must be made consistently and meticulously between objective data – even if it is controversial – and opinions or beliefs which are a matter of ideological or political convictions or personal or group interests. Such a distinction is necessary so that the members of the conventions would be able to form judgement which is as informed as possible, especially in an era where mistrust of science has increased dangerously.

The members of this type of conventions can make political choices, but when they do so, they must be fully aware of the reasons for their choices and of their full implications. It is also necessary to shield them from pressure and influence that they may be subject to by activists and lobbies outside the convention. The experience of the CCC seems to show that risk is very real. A trial jury must be protected from outside influences in order to make its judgement impartial, but how can “a citizen convention” be protected against pressures originating from activists and interest groups?

3. However, the most decisive is the definition of these “conventions”, which are unmentioned by the constitutional texts and the democratic tradition: except for their makeup and their function, what is their purview, what is their legitimacy?
Continue reading

Galland and Schnapper: Citizen conventions and representative democracy, Part 1

Olivier Galland, sociologist at the CNRS, and Dominique Schnapper, researcher at the EHESS and an honorary member of the Constitutional Council, write in Telos.

Parliamentary institutions are the only legitimate institutions for enacting legislation and for government oversight. Under what conditions could those institutions be complemented by the involvement in the public space of groups of citizens which would work for a period of time in order to come to know in an informed and open-minded way the dimensions of a political problem and which would publish the results of their deliberations?

This question is part of a general ambition for some form of democratization to which the institutions of the Republic are responding poorly at the moment. Dominique Rousseau advocates a “continuous democracy” while Rosanvallon advocates for a “counter-democracy”. But neither of them poses the problem in a way that appears to us just or practicable.

It may be accepted that low turnout rates are an indicator of the weakening of the legitimacy of the parliament and of government, or, put differently, that we are experiencing a crisis of representation. It is thus not out of the question to reflect on forms of citizen consultation which would inform the public discussion between elections. These would share the space that is now left solely to the media, to social networks and to unaccountable citizens who are not particularly informed and who make their opinion known, for example, wherever information is transmitted or even through the device of opinion polls.
Continue reading

Ostfeld: The Case for Sortition in America

Jacob Ostfeld makes a radical, uncompromising argument for sortition in the Harvard Political Review. Some excerpts:

The political realities of 2020 have laid bare that these flaws are structural to American democracy itself and have existed since its founding. Our system is not broken; it is functioning exactly as was intended. The system was always built around undemocratic institutions. The Electoral College, which allowed President Trump to be elected despite losing the popular vote, was created to protect the interests of slaveholding aristocrats in the South. Members of Congress are able to sustain decades-long careers in Congress despite consistently low approval ratings because of millions of dollars in lawful donations from Wall Street firms — donations which were made legal in the first place by a 5-4 decision from the nine lifetime-appointed justices on the Supreme Court. None of the undemocratic systems governing us today are subversions of the Constitution. On the contrary, they are all perfectly legal.

How, then, do we save American democracy? Sortition.

In simplest terms, sortition means appointment by lottery. In America, sortition would mean replacing Congress with assemblies made up of randomly chosen American citizens; elected representatives are entirely eliminated. Almost every responsibility of the legislative branch is delegated to a randomly subset of the population. Laws are written, discussed, and passed by ordinary people. Federal judges are interviewed and confirmed by ordinary people.

Continue reading

The French Citizen Climate Convention: a provisional analysis

It has been about 5 months since the French Citizen Climate Convention has published its proposals, and with acrimony setting in about the de-facto shelving of much of its work, various conclusions are being drawn about the CCC process. As usual, the conclusions almost invariably confirm the existing notions of the analyst. My analysis is no different in this sense: it seems to me that to a large extent each party to the process has played its expected role and thus the outcomes are quite predictable. I will highlight however two points that have been established empirically that should not have been taken for granted regarding how things would turn out.

Here are points about the CCC process that in my opinion are worth noting:

1. The process was launched as a government response to the Gilets Jaunes, a mass movement whose agenda was not just anti-government but also anti-electoralist. A popular initiative process (Referendum d’initiative citoyenne, or RIC) and to a lesser extent sortition were a major part of the discourse of the Gilets Jaunes. The rise of the Gilets Jaunes movement was triggered by what the government presented as environmentalist policy – increasing the gas tax. Thus having a non-electoralist process for generating environmental policy proposals was a direct capitulation to GJ demands. This origin of the body as a direct, stop-gap response to mass protest is very different from the origins of other allotted bodies, such as the Irish constitutional conventions. Such bodies, even if they were in some way a response to public disaffection with the status quo, were constituted in a much more carefully controlled manner by established power.
Continue reading

Carolan: Ireland’s Constitutional Convention: Behind the hype about citizen-led constitutional change

A 2015 paper by Eoin Carolan, Professor and Director of the Centre for Constitutional Studies at University College Dublin, takes a skeptical look at the conventional claims around Ireland’s Constitutional Convention which led to the legalization of same-sex marriage. (Note that later there was also a different allotted body constituted in 2016 which was called a “Citizens’ Assembly” and which led to the legalization of abortion.)

The article suffers from the standard pro-status-quo bias of showing no recognition of the urgency of the need to address the problems with the existing system. As usual, recognition of problems with the established system is phrased in terms of “public perception”, “disenchantment”, “disillusionment” and a “crisis of confidence”, rather than in terms of the facts of ongoing consistent systemic atrocious policy. Thus, while the paper rightly subjects the Convention process to a series of critical examinations, it seems to assume that the status quo is a legitimate default alternative. That said, I find that the article asks good questions, makes good observations and is generally very useful.


Ireland’s Constitutional Convention is one of a number of recent examples of ordinary citizens becoming involved in constitution-making processes. These participatory experiments are often praised by democratic scholars. That has been the case with the Convention, which has already been cited as an example for any future process of constitutional change in Britain. This article argues that the Irish experience has been oversold. The process in fact suffered from a number of serious limitations that undermine its claims to either representative or deliberative legitimacy. The approach taken to its composition, agenda, expert advice and evidence was problematic in several respects: opaque, apparently ad hoc and with inadequate attention to the risks of bias and manipulation by elite actors. The Irish experience provides a warning about how the symbolic value of the ordinary citizen can be exploited for political purposes.

Using focal random selection to close the gender job- and pay-gaps

Those well funded Swiss researchers have just produced another Report on the benefits of using a lottery as part of the job-appointment process.

By ‘focal’ they mean a two stage process (focussed?) with all applicants undergoing an ability test and the top three being entered into a draw, so the winner is selected at random.

The alternatives were: to select entirely on ability, or else entirely at random (from a pool of well-qualified applicants).

Their conclusion

“Our findings suggest that the pool of high-performing women who apply for top jobs can be substantially enlarged by the introduction of focal random selection. Consequently, the pipeline for women to leadership positions can be made less leaky without lowering candidates’ performance. Moreover, focal random selection closes the gender pay gap among high performers. In addition, differences between men and women in entering competition caused by gender stereotypes are completely eliminated by randomness. Our findings, therefore, point to the relevance of gender stereotypes as an underlying mechanism of gender gap in competitiveness.”

Not bad! Fix the ‘glass ceiling’ and the gender pay gap with the judicious use of lotteries!

You can read the paper here (no paywall)

Random Selection for the Supreme Court

In an opinion piece in The New York Times, Yale law student Melody Wang lays out an extremely cogent argument for random selection in choosing cases. She emphasizes the power of random selection to prevent corrupt practices, and to focus advocates on directing their arguments to the general good, rather than to specific decision makers.