The Costs and Benefits of Costs and Benefits
Primer
You’re deciding whether or not to eat at McDonalds. You start off with a simple cost-benefit analysis:
Benefits: cheap, tastes good, easy.
Costs: makes me feel gross.
Long and Short Run
This initial analysis just considers the short run. If you care at all about your future self, which you almost certainly do, then the long run matters as well.
The following cost should be added:
Costs: will make me less healthy.
Discounting the Long Run
You may have a “live in the now” approach to life, meaning that you don’t care about your future self as much as you care about your current self.
If so, then you should discount the effect on your future self. For example:
Benefit: tastes good = 20 happiness points
Cost: will make me sick = -100 happiness points * 1/10 discount multiple = -10 happiness points.
In this case the fact that it’ll make you sick really sucks — it accounts to negative 100 happiness points. But because this will happen to your future self, it doesn’t matter to you as much. This future self is only one tenth as important to you as your current self, so you discount the -100 happiness points by 1/10.
Note that you shouldn’t just disregard your future self — you care less about him, not zero.
Uncertainty
Actually, you don’t know for sure whether or not it’ll make you sick in the long run. How do you deal with this uncertainty? Don’t worry — expected value has your back!
Let’s say that you think there’s a 5% chance of it getting you sick, and a 95% chance of it not getting you sick. If it gets you sick, then you lose 100 happiness points. If it doesn’t, then you don’t gain or lose any happiness points.
Here’s how you perform the calculation:
sick: 5% * -100 happiness points = -5 happiness points
not sick: 95% * 0 happiness points = 0 happiness points
sum: -5 happiness points + 0 happiness points = -5 happiness points
Note that you shouldn’t just disregard the chance of getting sick — it’s low, not zero.
Options
So, let’s say that the benefits of eating McDonalds outweigh the costs. Even after accounting for the long run, discounting the long run appropriately based on how much you value your future self, and accounting for uncertainty using expected value.
Is it time to eat McDonalds?! No, not so fast.
You have to evaluate your other options first. You’re only going to eat dinner once, so if you have McDonalds, then that means that you’re not going to have Chipotle.
The benefits of McDonalds may outweigh the costs, but what if the benefits of Chipotle also outweigh the costs of Chipotle? Now you have two options to choose from, and both of them have benefits that outweigh the costs.
The answer is obvious — choose the one that is better. Choose the one where the benefits outweigh the costs by a larger amount.
This may seem so obvious to not be worth saying, but I see people in real life who forget to do this a lot.
Person: Hey, the benefits outweigh the costs — let’s do it!
Me: … but what about the alternatives?
Does This Utilitarianism Approach Make Sense?
In short, yes.
A rational decision maker will, when presented with a choice, take the action with the greatest expected utility.
- https://wiki.lesswrong.com/wiki/Expected_utility
There are some edge cases though. Check out the Stanford Encyclopedia of Philosophy for a discussion. Also see Game Theory 101: Rationality and Expected Utility Theory — a YouTube playlist that talks about Von Neumann–Morgenstern utility theorem.
But… It Isn’t Fun Being That Analytical
Great point! That’s exactly what I want to talk about in this article.
Running Example: What To Eat For Dinner
Let’s suppose the following:
McDonalds: benefits - costs = +20 happiness points
Chipotle: benefits - costs = +25 happiness points
Suppose you were originally going to eat McDonalds, but after taking the time to analyze your decision, you realized that you’d be better off eating Chipotle. More specifically, you’d be better off by 5 happiness points.
So then, your analysis provided you with an extra 5 happiness points.
Wait, that’s not true! Didn’t the analysis itself cost you something? Yes, yes it did.
The analysis itself took time and energy. It wasn’t fun, and you were hungry! You didn’t want to be analyzing, you wanted to be eating! The analysis itself costed you 15 happiness points.
Cost-benefit of performing the cost-benefit analysis:
Benefit: +5 happiness points (making the better choice)
Cost: -15 happiness points (time, energy, hunger)
In this case, you would have been better off not performing the original cost-benefit analysis at all.
Discussion
Performing an analysis has a cost. It costs you time, energy, and patience.
Performing an analysis also has a benefit. It leads to better decisions.
How much better will the resulting decision be? And how much will the analysis itself cost you? These are the questions you need to ask yourself. These are the costs and benefits of a cost-benefit analysis.
