So says Amanda Goodall in an article in the Financial Times 9 Sept 2020. You can read the article in full here (dodging the FT paywall!).
http://www.amandagoodall.com/FTRandom2020-09-07_101201.pdf
Dr Goodall of the Cass Business School, London has produced many papers on management and HR. She has tried to promote the idea of Lotteries for Jobs with Margit Osterloh, a Swiss academic.
Amanda tells me “It is a hard one [the idea of lotteries for jobs] to get off the ground. It has been hard to publish our article.”
It is very rewarding to find others in the field working on this form of ‘Local Democracy’ as Elster calls it. There are further developments which I will post here shortly.
Filed under: Academia, Distribution by lot, Employment, Press, Proposals | Tagged: Advocacy, Distribution by lot, jobs |
Hmm. Might be better to choose the hiring panel by lottery, with half of them randomly chosen from women, and half from men, and some portion of those women and men being people of colour.
Or maybe even to choose them 75% from women and 50% from poc to help redress the existing imbalance.
The pool for this stratified-random-sample hiring panel could include managers and fairly top level employees in the company, and also some outside examiners with relevant expertise.
Ways of getting people who might be qualified to apply are also needed, the article indicates. In addition to the panel actively seeking people (as the article suggests), managers and employees in the company could be invited to nominate people they think would be well-qualified by secret ballot. Each secret ballot could be required to nominate two people. All nominees would be automatically invited to apply. Obviously there are some problems with this form of nomination as those making the nominations are obviously likely to have some conflicts of interest (for example to nominate their friends, and to not nominate someone they see as a competitor to themselves). If one of the two people they nominate is themself, that’s fine.
LikeLike
As for the Swiss Supreme Court (mentioned in the article), it should be chosen by juries for set terms of office. And so should the rest of the Swiss judiciary.
The reason for this is so that they will be chosen by informed and highly representative portions of the Swiss, with all those interested in judicial office being put on a fair and level playing, not on a playing field skewed in favour of for example fellow-partisans, ideological fellow travelers and lackeys of Swiss politicians.
The Swiss Supreme Court should not be chosen by politicians and political parties as is now the case. This is at odds with basic tenets of democracy, including the independence of the judiciary from the other branches of government (executive and legislative), rule by the people and the political equality of citizens.
The way to bring the selection of the Swiss Supreme Court and the rest of the Swiss judiciary into accord with these basic tenets of democracy is for them to be chosen by jury. A suitable method of proportional representation using ranked choice ballot can be used to choose the Swiss Supreme Court several members at a time.
Article on the the proposal to choose Swiss Supreme Court by lottery (there must be a way to do a link but I don’t see it).
https://www.reuters.com/article/us-swiss-justice-idUSKBN25G0YS
My 2016 article “Why America’s Judges Should be Chosen by Citizen Juries,” https://dissidentvoice.org/2016/08/why-americas-judges-should-be-chosen-by-citizen-juries/
Although this article is written in the context of the US, this is how judges should be chosen everywhere, in my view. (The only possibly reasonable exception to this is where the parties might agree to choose their own judge, or where for example there is a three judge panel with one judge chosen by each party, and the third judge chosen by agreement of the parties.) I am so far as I know the first to propose and argue that judges be chosen by jury, which I first did in published form in 1997.
LikeLike
Simon, your idea to replace the normal selection panel with a randomly selected jury of individuals misses the point of lotteries — that they remove human judgement. Hence Ollie Dowlen’s description that a lottery is ‘arrational’ or Peter Stone’s notion of ‘sanitizing’.
But why would you want to remove at the very least SOME forms of human judgement? I suggest Khaneman’s Thinking, fast and slow, to understand how decision-makers, even the most benevolent and highly trained cannot avoid biases.
LikeLike
Judges selected by CJs? At least you may take heart from the US where judges are randomly assigned to cases.
According to
https://www.oknd.uscourts.gov/case-assignment-and-numbering
“Case Assignment and Numbering
Civil and criminal actions are assigned by a random selection process in order to ensure that the identity of the judge is not disclosed before the case is officially filed. The Clerk’s Office uses an automated case assignment system to randomly assign cases to judges. When a new case is filed, the clerk’s office will draw a new judge from the system. A judge is assigned to each new criminal case, and a judge and magistrate is assigned to each new civil action.”
I believe this random allocation process for judges is universal throughout the USA, which would make it one of the most frequent and widespread uses of Random Selection in the world?!
LikeLike
Simon,
A minor point, but the voting method used by a nominating jury matters. the two vote system you propose has deep flaws, if we are nominating a pool that will then be judged rationally and a winner chosen by interview, etc. One flaw is referred to in voting theory as “raising turkeys.” I have an interest in advancing my favorite choice, and a second person who is so bad they can’t possibly be selected over my favorite. With this logic occurring to all nominators, half of your final pool might be turkeys.
An interesting historical note is that in the original U.S. Constitution the electors of the Electoral College cast two equal votes, with the candidate getting the second most votes becoming vice president. After the fiascos of the presidential elections of 1796 and 1800, this procedure was amended out of the constitution.
LikeLike
Terry,
> the two vote system you propose has deep flaws
This is not something I am proposing. It is only something I am or was considering re how to identify and nominate good potential managers within the company. As I said, “Obviously there are some problems with this form of nomination …”
I agree that people who may themselves be in the running for a position have the conflict of interest problem you refer to (the “raising turkeys” problem), as well as the ones I mentioned.
(Just to be clear, I was not suggesting that there be a nominating jury, but rather that all those of certain categories who work in the company be able to in effect vote for two people who they think would be good.)
If the people working in the company are likely to know who among their co-workers and supervisors would be a good choice for a management opening, then it makes sense to try and find out what their honest opinions are. This is tricky, including due to the conflict of interest problems we have mentioned.
> the original U.S. Constitution the electors of the Electoral College cast two equal votes, with the candidate getting the second most votes becoming vice president.
Definitely one of the more comedic failures of foresight in political history.
LikeLike
Conall
> Simon, your idea to replace the normal selection panel with a randomly selected jury of individuals misses the point of lotteries — that they remove human judgement.
No, the point of randomly sampled juries is to make judgements, not to “remove human judgement.”
The point of for example a trial jury is to make a judgement, not to “remove human judgement.”
Random selection removes human judgement from the selection of the jurors. But hopefully it does not remove human judgement from the decisions that juries make.
> even the most benevolent and highly trained cannot avoid biases.
I do not disagree. That does not mean for example that instead of using a trial jury to decide the result of a trial, we should instead use lottery (or a coin toss). Deciding the result by lottery would eliminate some “biases.” For example, it would mean that black and poor criminal defendants would have the same chance of being convicted as white and rich ones (which is not the case at present in for example the USA), but it would not be a good way to achieve that result (because many innocent people would be convicted).
The pre-chosen pool of candidates for Barbara’s lottery of course has to be chosen by someone, so her proposal does not entirely eliminate the role of human judgement. The question I am asking is whether it is better for achieving equity to choose a manager from this pool at random, or by using a jury designed (in the way I mention) to achieve equitable choices of managers.
LikeLiked by 1 person
Conall
> [In the US] judges are randomly assigned to cases.
Good point.
This is, in addition to trial, grand and coroner’s juries, a further precedent for the use of random selection in the legal system.
LikeLiked by 1 person