The Devil You Know is Better Than the Angel You Don't

An analysis of the extreme importance of predictability in social dealings

There is a famous saying that goes “The devil you know is better than the devil you don’t”. The meaning of the saying is that if you’re going to have to deal with an enemy or an untrustworthy individual it is better that it at least be someone of whom you have some knowledge and whose actions you can predict so that you can predict when they will screw you over so that you can better prepare for it and protect yourself. The advice of this quote is eminently reasonable and in fact I would even go so far as to say that it is too conservative. Not only is the devil you know better than the devil you don’t he is oftentimes also better than the angel you don’t know. Which is to say that working with a known quantity is often easier than working with an unknown quantity *even if the unknown quantity is significantly more competent/trustworthy/righteous in general*

Let’s look at a quantitative example. Suppose you are a businessman who has to award certain contracts to your business associates, if they come through for you you’ll make $1000 but if they don’t come through for you you’ll lose $10000. One of your business associates is Mr Devil. Mr Devil is generally not very reliable, in fact for 90% of the contracts you have he will not come through for you. However, the one good thing about Mr Devil is that you know him very well so you are able to perfectly predict which contracts he will come through on and which ones he won’t. The end result is that you don’t award many contracts to Mr Devil but when one of those contracts pops up that’s in the 10% of contracts that you know Mr Devil will succeed on you’re happy to award it to him.

You also have a new business associate Mr Angel. You don’t know Mr Angel at all but from his references you can tell that he’s a good businessman and that he will likely be able to come through on about 90% of your contracts, but you don’t have any idea which ones. So how do you proceed with Mr Angel? Well if you just award him a random contract 90% of the time you’ll make $1000 and 10% of the time you’ll lose $10000 which means on average you’ll make .9*$1000+.1*(-$10000)= -$100, Uh oh, that’s negative. So awarding him a random contract is not a profitable strategy. But since you don’t know anything about Mr Angel any contract you award him might as well be a random contract. This means that until you learn more about him you can never profitably work with Mr Angel.

So the dynamic we see playing out here is that even though Mr Angel is, in an abstract sense, a much stronger business partner than Mr Devil you cannot work with him profitably while you can work with Mr Devil profitably. The Devil you know is better than the Angel you don’t.

This isn’t just some made up scenario, this exact thing plays out in business all the time. Businesses consistently prefer to work with those they have a history or personal connection with because they are easier to predict, even if they are, in a general sense, inferior.

If you ever find yourself in Mr Angel’s shoes this dynamic might feel very unfair. You know you’re a better businessman than Mr Devil so why doesn’t anyone want to work with you? But as we can see here the dynamic is in fact perfectly rational and the solution is building a relationship and working your way up.

From a societal level we can observe that this implies that social and cultural structures which allow people to more accurately predict the actions of others can enrich society by creating more situations where people can work together profitably. The interesting part is that this is still the case even if the social structures themselves are arbitrary and senseless.

Let’s consider an example. Suppose you, Mr Devil and Mr Angel all belonged to a society that had a rule which said “You don’t mess up on a Tuesday” and everybody in that society took the rule very seriously and worked extra hard to make sure they never messed up on a Tuesday and if ever somebody did mess up on a Tuesday it was treated as extra shameful. This rule is clearly arbitrary and baseless, there is nothing inherently special about Tuesday. However if you knew that Mr Angel was from the society that took this rule seriously and if it was Tuesday, then your probability that Mr Angel would come through for you would raise to 95% instead of 90% and then it would be profitable for you to work with him. Potentially, working with him on this contract could lead to you learning more about him and lead to many more successful contracts in the future.

So if you ever find yourself wanting to criticize a social custom for being arbitrary or baseless just remember that even arbitrary and baseless social customs can be very useful if we all believe in them.