Mediocrity in Labor

Why you're a 4/10

I was reading this piece from DanT’s Technopoptisim blog, and I enjoyed the following footnotes:

[NFL Films founder] Ed [Sabol] was a graduate of Columbia, received his MFA from NYU, and interned for Warner Brothers. Wait, just kidding, this was back when America innovated, so Ed Sabol was a WWII vet and a raincoat salesman who started his own company. Although he was a [world-class] swimmer who turned down swimming in the 1936 Olympics because he wouldn’t swim in a pool built by Hitler.

[Pete Rozelle], a graduate of Yale with a JD from Harvard Law School – oh, you know where this is going. He was a WWII vet who got his first break in sports while at community college, worked his way up to NFL Commissioner, and then radically changed it into a goliath of business.

Today, it seems quaint for two major executives to lack traditional credentials. I couldn’t find the current head of NFL films, but the reigning Chief Media Officer got a Harvard MBA while the COO of NFL Media has an unspecified Princeton degree and background on Wall Street. Commissioner Roger Goodell is less credentialed, though he still earned an economics degree from a liberal arts college. Tech titans like Bill Gates and Mark Zuckerberg break this trend, and I’ll get to them later.

To start the discussion, let me list a few observations I hear across a variety of industries:

• Every corporate office runs poorly. I’ve spoken to many people about their jobs, and I’ve never heard a single person respond “yeah my company is super efficient.” People often describe bullshit jobs, where none of the work results in operational changes.

• Positions require college degrees, but one could perform the role without a university education. Many of us find memes like this relatable:

From: https://www.reddit.com/r/ProgrammerHumor/comments/mew4g6/technical_interview_vs_job_tasks/

• People have a lot of credentials. This doesn’t just include the normal BAs, MAs, and MBAs. You also see Lead Six Sigma, HR certifications, and proofs of completion of various IT/Data Science/Programming courses.

• Companies often pay people more for obtaining additional certifications (e.g. completing a part-time MBA), but these certifications don’t seem to improve performance. Oftentimes, students will joke about how silly the program is.

• Jobs that didn’t previously require degrees (e.g. executive assistants/secretaries) seem to require them now.

• The top performers never seem to boast the most impressive credentials. Sometimes, the college dropout is the best employee, or the one with only a BA performs better than the ones with MAs and MBAs. Other times, the best performer has great credentials in a different subject. An example would be a great data scientist with a Ph.D. in English.

What’s going on here? Why are we spending more of our time in classrooms, and where did we get the idea that book lernin’ beats hands-on experience? When my dishwasher broke, I wanted someone with plumbing experience. I wasn’t searching for an MA in Applied Hydrological Sciences. This article will present two theories: asymmetry and optics.

Asymmetry (Also Loss Aversion)

You’ve reached the “thought experiment” part of the Simplify article. Prepare for paragraphs that begin with the phrase “imagine that.”

Imagine that you’re captured by aliens, and they force you to play one of two games. In each game, a 50/50 coin flip decides if you win or lose. In the first game, a win gains you $10, and a loss costs you the same amount. The second game contains the same rules, with $1,000 instead of $10. Which do you prefer? Though the games offer the same expected payout, most of us would prefer the $10 game. This is what economists call loss aversion: the pain of losing $1,000 outweighs the pleasure of gaining $1,000. In more technical terminology, this occurs because consumers act according to a concave utility function. In less technical terminology, the badness of bad stuff is bigger than the goodness of good stuff.

Now, I’ll apply this to hiring. Say that I make $80,000, and I need a new data analyst on my team. If I hire the right gal, my wage could double to $160,000. If I screw up, my supervisor could sack me. In this scenario, I may select between two candidates: a risky one and a safe one. The risky one may perform well and double my salary. The safe one offers no chance of doubling my salary, but she also won’t cause my firing. Who do you choose? Most of us would probably opt for the safe one. Doubling one’s salary seems nice, but getting fired would really suck. Your concave utility function selects the safe candidate.

But, wait, that doesn’t sound anything like the decision hiring managers face. I just drew that one up to draw some symmetry with the alien scenario. However, real-life hiring decisions involve much less interesting games. We don’t see an equivalent to the “lose only $10” scenario or “gain $1,000 scenario.” Instead, the game of hiring often contains only a large loss (getting fired) and a meager gain (getting a 5% raise instead of a 3% one).

I like numbers, so I’ll present a quantitative scenario. A manager needs to hire someone for her team, and she spots two candidates: Kyle and Kilroy.

• If she hires Kyle, there’s a 50% chance he brings the company $300,000 in revenue, and 50% he brings in $100,000.

• Kilroy offers a riskier choice: he has an 80% chance of bringing in $1,000,000 and a 20% chance of producing $0. Who does our manager select?

Econ 101 says she’ll select Kilroy. His expected value is $800,000, while Kyle’s is only $200,000. Companies maximize profit, so the choice is easy. Econ 701, however, might tell a different story. If Kilroy produces no revenue in his first year, the manager risks losing her job. That’s a really bad scenario that she’d like to avoid. Sure, Kilroy has an 80% chance of making $1,000,000 for the firm, but who cares? It’s not her money. Maybe she gets that 5% raise instead of the standard 2%. After all, the big raises often only come when people leave. Even if the company does offer meaningful raises and promotions to high performers, much of the compensation process may sit outside the employee’s control. For instance, economic downturns or revenue shortfalls in one department may cause promotion freezes across the entire industry. I don’t mean to deny an internal locus of control. I have received raises and promotions at previous jobs, though even these depended on positive economic and corporate outlooks. I’m just saying that the [high performance] —> [high pay] connection is a lot messier than the line between [low performance] —> [getting fired] one. As a result, our manager probably avoids Kilroy.

Optics

In a corporate office, you’ll often hear the notion of CYA, which stands for “Cover Your Ass.” Workers don’t want to fuck up, of course, but they also want good excuses when they do. With that in mind, let me update our Kyle vs Kilroy comparison.

• Kyle has an MBA from Harvard, and he currently works for a big four consultancy. He has an 80% chance of producing $300,000 in value and a 20% chance of producing $0.

• Kilroy dropped out of college and lacks any relevant work experience. If you hire him, he has an 80% chance of producing $1,000,000 in value, and, like our friend Kyle, he also has 20% of producing diddly squat.

There’s no risk vs reward issue here. Each candidate poses a 1-in-5 risk of an awful performance. Who does our manager choose this time? Well, she might still select Kyle.

The same argument applies to cases where the employee adds revenue. If Kilroy brings home the bacon, it’s not your money blah blah blah. The difference stems from that 20% chance. They equate in monetary value, but they look much different in an annual performance review.

Imagine you hire Kyle. You can pull a sort of “how could I have known” argument if he fails. He fit the bill of a great employee, and just happened to not work out. You could probably present yourself as the victim. That bastard Kyle scammed you! He must have cheated his way through the MBA program! Maybe he misrepresented his experience at that consulting firm, and he actually performed an unrelated function there. Such a tale could help you avoid a firing.

Now, imagine that same scenario for Kilroy. You hired a shelf-stacking dropout. What did you expect to happen, moron? It’s just not a good look.

DanT noted this effect in sports, as coaches and GMs often prefer safe strategies to more effective ones. I’ve noticed this as well, but I’ll have to add the caveat that I haven’t followed any non-hockey sports in a few years. When I did watch the NFL and college football, I always noticed that post-season play was weirder. I would see more fourth-down attempts, two-point conversions, onside kicks, trick plays, and even teams kicking the ball backward for safeties. In the NHL playoffs, you’ll see coaches empty their nets much sooner than they would in November.

Optics and asymmetry explain this phenomenon. Coaches won’t earn a raise for finishing 9-8 instead of 8-9, but they could get fired if those losses occur due to wacky tactics. Better to lose conventionally than to marginally increase your odds of winning unconventionally. This optics focus seems to dissipate come playoff time, so coaches feel free to leave their punter on the bench for a 4th and 3.

Interview Advice

This is a bit of a tangent, but I think the previous paragraphs provide some implications for interviewing. I’m not going to pretend that I’m some master interviewer, but I have received offers from the largest companies in my city, a major consulting firm, and (the one I took) a tech startup in an industry I knew nothing about. So, I think I’m pretty good.

When I interview, I don’t try to look like the best candidate. Instead, I try to look like the safest bet. I want to show my future boss that I have the lowest probability of sucking, even if that means I don’t have a particularly high probability of being exceptional. When the hiring manager asks the inevitable “tell me about a time you went above and beyond” question, I recount an instance when I did a bit more of the thing I was supposed to be doing already. I don’t craft a tale where I worked on an initiative well outside my job description. That’s what a risky hire would do.

I also consider optics. Descartes argued that we can only know one fact for certain: we are a momentary blip of consciousness. I feel safe in adding a second: no one has ever read the code on my Github page. Yet, I put the page on my resume and LinkedIn, and I discuss my personal projects in the “tell me about yourself” question at every interview. I don’t expect the interviewer to check my code or even care about it. Rather, I think it makes my hiring seem like a “good look.” We hired a guy who codes in his free time! We hired a guy who has proof of his coding ability! Seems like safe bet.

Consequences

Over-credentialing leads to two issues: wasted time and sub-optimal hires.

Dolly of the 21st Century Salon discusses wasted time

Everyone at work in an administrative role just keeps acquiring degrees because they expect to get more money and more respect, even though these hapless people are getting “degrees” from diploma mills and racking up major debt.

The majority of applicants for administrative assistant roles at my workplace -- formerly known as “secretaries” -- have “master’s degrees” in things like “health care administration” and yet they can’t produce a coherent cover letter.

I don’t know what’s happening with all these meaningless degrees. I assume these poor people have been scammed.

A lot of human effort goes towards nothing. Students waste time in these classes. The credential-granting organizations, meanwhile, place thousands of people into bullshit jobs.

DanT of Technopoptisim discusses hiring incentives

I will say that in my experience there is absolutely no correlation between my best employees and education. With one exception (who actually had a Ph.D. from a top school) the best employees I've had are uniformly college dropouts. However, the actual college graduates rarely drop below a 4/10. So in systems optimized for mediocrity, hiring college grads is smart.

Thus, companies fill their slots with 4/10s instead of 8/10s because they’re worried about accidentally hiring a few 3/10s.

What We Lose

Let me return to those tech tycoons. Yes, I’ve previously argued that much of our technology sector provides dubious value, but I couldn’t say that the software industry hasn’t been successful. We’ve clearly seen at least one tech boom, and these firms fill dozens of slots in the Fortune 500. At the same time, software remains one of the few high-paying white-collar jobs that one can land without the relevant credentials. I’ve known many developers without a CS degree, and they seem to perform as well as those who spent their early 20s learning about data structures and algorithms. What if every industry looked like this? It’s difficult to picture such a counterfactual, but it sounds like one worth working towards.