O God, Make Me Humble

Caesar: What's under the sheet?

Marcus Vindictus: Sheet? Oh! Oh, the sheet. Yes. To begin with, number one, a beautiful hand-carved alabaster bathing vessel!

Caesar: Nice. Nice. Not thrilling...but nice.

History of the World, Part I (1981)

There is only one prayer I know which God always answers: ‘O God, make me humble.’

When I was growing up in Minooka, Illinois, I was a nice kid. I was also an insufferable know-it-all. To be fair, I wasn’t the smartest in my school. That honor belonged to Andy Kimble, a math genius who went to USC to study film. He’s now an editor for several television shows. Still, I had a reputation for knowing way more than I ought to have about far too many topics.

There was more than a little bit of small pond effect to this. I have already confessed, after all, to my status as Medium Talent. It’s a big world, and the sooner you realize how many millions of people can think circles around you, the sooner you find your place in it. Ben and I both participate in a fairly competitive online trivia league. Within the hierarchy of the league, I am placed in what is called the “B-rundle”, which is exactly what it sounds like. Nice. Not thrilling…but nice.

My one great recurring memory of being young and in school, however, was the joy with which people responded to discovering that I didn’t know something. My fifth grade teacher. That time I spelled ‘handkerchief’ wrong in a spelling bee in 7^th grade. The first time I said the word ‘banal’ out loud. The time I pronounced ‘Mussorgsky’ as ‘Musso-gorsky’ in an orchestra rehearsal, and embarrassingly insisted that it was an accepted alternate way to say it. The scarier thing, of course, was how I responded to all these mostly harmless jokes at my expense. How desperately and successfully I hid the fact that there were a LOT of things I wasn’t understanding, and topics I didn’t completely follow.

At a young age, I found that I enjoyed the idea of being thought of as knowledgeable and intelligent, maybe as much as I enjoyed actually being those things. I also realized with some surprise that the payoffs of the two were usually pretty similar – the tangible payoffs that were immediately evident, anyway. And pursuing the former was a hell of a lot less work.

No self-flagellation here. As it turns out, there’s a growing body of research into how the real and perceived value of ability-signaling is effectively stunting true pursuit of education across the board. Bryan Caplan writes about it a great deal in his recent, provocative book The Case Against Education. Much more recently (i.e. this month), two Harvard researchers and one from Stanford released an NBER working paper on the topic called Signaling, Shame and Silence in Social Learning (h/t @RobinHanson). It’s an academic paper, but it is an interesting read. It will also be useful to anyone more generally interested in the intersection of narrative and metagame playing. But at its core, it seems to confirm that the negative ability-signaling associated with implying that we don’t know something keeps us from asking questions even when we should. This is a major problem of social organization. These people are not acting irrationally. In just about every major social sphere, we have created a system in which it is absolutely in our socioeconomic best interest to maximize our ability-signaling, even at significant cost to the thing we are supposedly signaling.

I have never seen an investment firm that has really solved this problem.

Bridgewater has famously, valiantly tried. They’ve done better than most. ‘Radical transparency’ goes a long way toward addressing the costs and benefits of expertise vs. perceived expertise. So does the embrace of mistakes and the attractive concept of a meritocracy of ideas. But even the Bridgewater code systematizes ability-signaling through ‘believability-weighting’. If you don’t think managing your reputation – even at the cost of asking questions when you don’t know the answer – is as important on Glendinning Place as it is on Wall Street, you’re kidding yourself.

But at least they’ve tried.

And so must we all. It is critically important. To the extent we incentivize ability-signaling, we impair humility. To the extent we impair humility, we degrade every decision-making process we may have. The problem I described in the Cornelius Effect was one in which, beyond a certain point, increases in talent and expertise of advisers and investment professionals tend not to manifest in better outcomes. This problem is partially external. With a world full of brilliant people, believing we can find the person who can tell us the answer through singularly brilliant insights is a fool’s errand. It is also partially internal. Humility is a necessary precondition for a talented person to make himself or herself part of a process.

But the prayer required of the organization that would actively stamp out the trappings of ability-signaling – O God, make me humble – is perilous. When you tell everyone about your vision, they will say that it is such a good idea! So long overdue! We love how transparent you are! Break the old patterns of our industry! Twelve months later, when they’re flipping through your deck, it will be different. They’ll read the bios. Hmm…what school is that? I like the strategy, but the PM’s background looks spotty…I haven’t even heard of these firms she worked at. The analyst they introduced us to seemed to be really well connected to the PM’s thinking, but remember that analyst at the other firm? He knew everything about that company! Now that was impressive!

I’m making this up, but we both know that I’m not really making this up. This isn’t an indictment of allocators or decision makers or fund managers or…anyone, really. It’s an indictment of the fact that we built an industry on professionals exploiting the knowledge gaps of their clients. It’s an indictment of the fact that the solutions we created to prevent charlatans and criminals from pursuing that exploitation were prudent man and other standards designed to minimize the appearance of risk instead of minimizing undesirable/uncompensated risk.

It’s time to revisit these standards.

This won’t be the last time we write about this.