Imagining better ways of picking people
I am in the process of giving my assets to education and I think your Trust Agent method is beautiful for many reasons. I am a retired Professor of Theoretical Chemistry and am funding projects at 5 major universities. I will soon have 3 Agents. If you have ever participated in the sheer lunacy of applying for a research grant, your method is most appealing. Thanks,
Interesting notions, if also as illustrations of how chesterton's fence came to be so.
One note on the increasing fervour with applications is that this often coincides with these fields becoming overall much larger. The circle being: larger candidate needs = requires rigorous standards = need for gatekeeping + credentialism = legible application processes = larger candidate pools (I looked at this in terms of wage effects and found HR becoming more important since there were more applicants - https://www.strangeloopcanon.com/p/understanding-wage-stagnation).
I love this. I've been making this same argument about foster families for ages. Average families are acceptable child-rearing environments. It's cheap and easy to find average families. But what do we have instead? An application process, plus a ton of standards that look complex and misguided to exactly the kinds of parents and families we're trying to find. So, we should just place kids with randomly-selected families that already have one or two kids. It would instantly solve the family supply problem and axe a ton of administrative burden. But it's not a process we're in *control* of, so it is anathema to bureaucracy.
Weird coincidences: "the application apocalypse" is basically compounded with other Substack articles like "The Occult Deficit" and other 1971 memes https://rogersbacon.substack.com/p/the-cult-deficit-analysis-and-speculation https://rogersbacon.substack.com/p/something-is-not-right https://wtfhappenedin1971.com/
Really interesting ideas! I think you might be missing a part of the hiring equation, which is that it can be hard to fire people (especially in fields with strong HR). This makes hiring more of a one-shot deal than a many-shot game and makes hirers more cautious. It also gives hirers incentive to be able to defend their hiring decision, for which application documents, references, and certifications are perceived as useful, or at least concrete.
Intern World directly gets around this in a clever way and I really like that idea. I'm not sure the others do, though.
I think we already live in a Dark Bellamyville, where instead of remuneration being tied to inverse popularity, it's tied to inverse social utility. Want a job that feels good to do - working with kids, being an artist, anything that helps people directly? Generally you're not going to be paid well, and why should you, your job is so *fulfilling*. I don't think this is healthy, but it's a rather persistent pattern.
Ok I haven't fully thought out all the details of this, so bear with me. Of these options, I like Bellamyville the best, but I also wonder if it's worth examining the premise that there are people who are the "best" choice for something, that the goal is to identify them accurately, and that the main problem with the current system is that it doesn't do that. I kind of want to substitute a paradigm of "good enough" and a plea for a deliberate move away from manufactured scarcity.
For example, in the current system, the most selective U.S. colleges are the Ivies, and going to one gives you a big advantage in the next round (grad school, job, whatever). I've heard Ivy admissions officers say that for each class they admit, they reviewed applications for an additional 2-3 classes worth of people who would have been similarly successful at the institution (I personally think that's an underestimate). McGill, on the other hand, one of the most prestigious schools in Canada, admits 13x as many undergrads as Yale or Harvard, based entirely on grades, test scores, and CV stuff (jobs, activities). No essays, no letters of recommendation, and admissions are rolling. This method obviously has its problems (grades are hackable, standardized tests are bullshit), but I like that it has transparent criteria, less work for the admissions office and the applicants, and a "good enough" approach (e.g. learned enough high school chemistry to succeed in undergrad chemistry) where above a cutoff, it is somewhat random. So, as far as academia goes, I think I would vote for small, wealthy, elite schools to restructure and greatly expand their student bodies (this has the additional benefit of diluting their prestige). This would not require these institutions to lower their standards (per their own admissions officers) but merely to cut nonessential "perks" and ostentatious displays of wealth and maybe spend slightly more of their endowment's annual return in order to provide a high-quality education to *as many students as possible.* Ok so maybe nobody gets 1:1 office hours with a Nobel laureate and nobody gets ice sculptures or caviar or a fancy new school of global affairs, but that's ok, everyone will be ok actually.
Where resources/positions are fundamentally limited, I think it might work well to have a "good enough" threshold based on certain skills/job requirements, and then random picking from the good enough candidates. But in many cases, I think we would benefit from hiring more people and having them work fewer hours (and reducing executives' pay proportionately), though this is obviously at odds with capitalism, which requires unemployment.
Some people are definitely better at certain things than others, but for something as difficult to predict as "going to be a good scientist 7-10 years after admission to this PhD program" or as low-stakes as "likely to be able to write these corporate-speak brand memos," I don't think that, beyond a certain threshold, most candidates are going to be meaningfully better than most other candidates, certainly not in a way that justifies the current application system, but also probably not in a way that justifies many of your suggested systems. Whether someone will become a good scientist is gonna be so dependent on their environment, their relationships with others, and a lot of luck that it might just be better to, wherever possible, focus on increasing the number of people who can be accommodated by our institutions to do the kind of research they are excited about.
Does the MacArthur "genius" grant function as a sort of Trust Windfall? No applications, no strings attached to the money, a few large sums of money given out to people from a bunch of different fields each year, etc.
Have you hired people? Some interesting ideas, but mostly unworkable.
Haha loved this one. I think my vote will go to Selectopia (it has the best name too).
I wrote about something similar a while ago, but in relation to testing (https://bewrong.substack.com/p/inverted-u-graphs-and-testing?s=w).