Protected Hours

Reading note, a bit over my skis on the banking culture here. Caveat emptor .

Researchers looked at how ‘protected hours’ (Saturdays off) worked for investment banks in New York City. Analyzing 16 million taxi rides, researchers found that bankers worked the same total number of hours, but did have Saturday’s off. There’s a few lessons in this.

1/ The feedback loop between outcomes and actions for aspiring-junior-bankers is loose. Unlike a coder, or salesperson, there’s not a clear pile of work or numbers to say ‘I did this’.

1a/ “Landers, Rebitzer, and Taylor (1996) argue that some firms may use willingness to work long hours as an indicator of some valuable, yet hard to observe, characteristics of employees.” So for starters, we might have the wrong proxies.

2/ Without outcome measures people substitute. Often this is looking the part. Punting on fourth down is optics. Suits are optics. Words rather than deeds or money, is optics.

2a/ However society is getting more numerate. That means we’ll measure outcomes more tightly: Amazon review, Uber ratings, credit scores, Zillow estimates, predictions.

3/ So we have loose feedback and have to do something. Okay, let’s use long hours. But not too long. And not on Saturdays. So we get a decree. However, we can’t do just one (eco-,social-, economic-) thing. An increase in one area cascades to others.

3a/ According to Nassim Taleb, small effects occur in Mediocristan (normal bell curves) and large effects will occur in Extremistan (fat tails, power laws, networks, complexity). There will be fallout from change and the system dictates the size. .

4/ You can’t do just one thing, part 2. In banks that didn’t adopt a protected weekends policy, there was also a decline in taxi rides to the banks on weekend. They guessed at a socio-culture effect. If they aren’t, I guess I won’t either.

5/ You can’t do just one thing, part 3. After a certain time, say 10pm, banks reimburse employees for taxi rides. Guess how late the employees decided to stay? Or maybe they just always finished after 10?

6/ Even though there are side effects, protected weekends aren’t bad. For one, it solves for ambiguity aversion.We don’t like not knowing.

7/ Mad Libs news is the effect which ambiguity aversion causes.Stonks are ________ [up/down] based on ________[news/tweets]_______ from [Elon/Mark/The President/The Speaker] about __________ [International issue, domestic issue, spiritual issue, political issue].

8/ Protected weekends solves for this. Junior bankers know more now than they did when they started: I’m probably working till at least 10 and I’m not working Saturday.

9/ The long hours also filter out people who really don’t want to do the job. Long hours in investment banking are the Peloton Christmas commercial: not for you. Unless it is. Then it’s really for you.

10/ Bottom line: protect hours are probably good, but like any intervention have unexpected consequences.