Promoting at random
Recently, I published a post criticizing performance a promotion processes calling it “the myth of objectivity”.
I did not expect it to be received so warmly and, for some people, controversially.
So I thought to expand that concept with a couple of ideas that might be even more controversial.
Convincing someone to give you the job is different from being able to get the job done
Malcolm Gladwell’s “Revisionist History” ran an episode in 2020 called “The Powerball Revolution”. In this episode, Gladwell tries to apply the idea of “democratic lotteries” with Adam Cronkright of Democracy In Practice.
The process is the following:
Every student who wants to be elected as a representative will just put their name forward
On election day each applicant will draw a colored ball, and people with the winning color will be elected
This seems counterintuitive right? How is this democratic if the people don’t vote?
The underlying idea is that election campaigns are biased toward people who are great speakers, have good friends, or are popular. None of it matters when it comes to defending the interests of the student body as a whole.
Moreover, they find that many introverts and/or people coming from minority backgrounds will avoid a popularity contest but are comfortable putting their name forward if it’s a random chance.
So they put this to the test in New Jersey and followed how the student representatives group became more diverse and also changed its priorities. In the end, they polled the students and found that their work was much more appreciated as a whole.
And more people in scientific research believe that this is a good idea
This is a powerful idea, and some researchers are advocating to use of the same system for research grants.
In 2022, some researchers from the University of Catania won an IgNobel for their research on the role luck plays in life:
The largely dominant meritocratic paradigm of highly competitive Western cultures is rooted on the belief that success is due mainly, if not exclusively, to personal qualities such as talent, intelligence, skills, efforts or risk taking. Sometimes, we are willing to admit that a certain degree of luck could also play a role in achieving significant material success. But, as a matter of fact, it is rather common to underestimate the importance of external forces in individual successful stories.
The researchers ran several simulations to predict the role of luck in individual success (measured by wealth), and in their conclusions, they state:
The model shows the importance, very frequently underestimated, of lucky events in determining the final level of individual success. Since rewards and resources are usually given to those that have already reached a high level of success, mistakenly considered as a measure of competence/talent, this result is even a more harmful disincentive, causing a lack of opportunities for the most talented ones.
Our results highlight the risks of the paradigm that we call ”naive meritocracy”, which fails to give honors and rewards to the most competent people, because it underestimates the role of randomness among the determinants of success.
The strategy do they suggest using? Just give funds at random, and make lots of bets.
And you know who else does that? Venture Capitals.
Wait! The math doesn’t make sense when it comes to re-elections!
An objection that I think is relevant to raise is “ok, I believe you. If I know nothing about the skills of someone, random is better. But once they have done the job for a year or two, I have now enough information to judge them fairly”.
I think this objection is valid and haven’t seen any specific research on the topic of re-elections.
If we look at the simulations ran by the team in Catania, they tried different strategies:
Random
Proportional to past successes
Mixed (i.e. 20% of funds reserved for top performers, rest at random)
They say that the simulation shows that mixed strategies work better than using only past successes but worse than random ones.
My intuition is that, while random is probably the best strategy, humans also need agency and trust in their leaders.
From this point of view, an election with 20% of seats assigned using a classic ballot, limited to outgoing members, and the rest random, would have a better overall perception.
Why don’t we do the same for promotions or hiring?
Create minimum requirements to apply, and maybe do one review to verify the requirements. And then draw at random.
I know it can’t happen
And I know that because it makes me, the proponent of this idea, uneasy. As humans, especially in the corporate world, we like to be in control and smarter than average.
So maybe we would like to use a mixed strategy: pick one person ourselves and the rest at random.
But what if random worked?