Incentives Matter

A little hobby of mine, that I have managed to get my daughter hooked on to as well, is etymology.

I’ve long held that concepts become more interesting, more relatable and therefore more memorable – in the literal sense of the term – once you’re able to tell yourself a story about the underlying concept. Look up the etymology of the word “average”, for example, and it is likely to be a story you won’t forget in a hurry. By the way, here’s a fun question the daughter asked some months ago, and I’ve been kicking myself for not having thought of it first.

So what is the etymology of the word incentive?

From Medieval Latin incentīvus (“that strikes up or sets the tune”), from incinō (“to strike up”), from in- (“in, on”) + canō (“to sing”).

I like words. I like stories that can be fashioned out of, and about, words. If you click around on the search result that I have linked to, you realize that you can go down quite a rabbit hole about the history of the word incentive. Words such as kindle, singing, and incendiary crop up, and the associations these words can conjure up in one’s mind can result in a very pleasant couple of hours. But the phrase that resonated the most with me was “sets the tune”. It fits nicely with what incentives actually do in real life – they do set the tune on which we are tempted to dance.

Now who sets the tune, for whom, and with what consequences – that’s a whole other story, and practitioners of public policy can tell this tale much better than most other folks. But even outside the always-fascinating drama that is always being staged in the theater of public policy, this story is at the heart of what plays out in applied economics. Who is incentivizing whom, towards what end, and do the incentives end up producing intended or unintended consequences, and at what cost – these are fascinating questions to answer.

My favorite story about getting incentives right comes from Marginal Revolution University:

And my favorite story about getting incentives wrong comes from Calvin and Hobbes:

And that’s the tricky thing with incentives. Getting them right is a surprisingly difficult thing to do. The reason it is a surprisingly difficult thing to do is because of a variety of reasons, but it is possible to start to think about building a framework that one might use to design incentives.

Begin by asking yourself this question: Who is designing the incentive, and for whom?

Let’s begin with a simple example. Let’s say I am designing an incentive for myself. If, I say to myself, I can finish writing the blog post you’re reading right now without taking a break, I’ll reward myself by having a cup of coffee. In this case, I am designing an incentive for myself – I am setting a tune for myself to dance to.

Note two other things about this little incentive scheme:

  1. It is a positive incentive. I am not going to punish myself if I do not finish my designated task – that would be a negative incentive. I am, instead, going to reward myself if I finish my designated task. Think of the old English phrase “the carrot and the stick” to get a sense of what a positive and negative incentive mean.
  2. It is a non-monetary incentive. I am not going to reward (or punish) myself with money. There is no prize money, nor is there a fine. There is, instead, a non-monetary reward – a nice hot steaming cup of coffee. Incentives need not always be monetary!

So, a positive, non-monetary reward to finish a task. What could possibly go wrong? Consider the opening paragraph of Ch. 6 of a lovely little book called In The Service of the Republic:

In 1902 in Hanoi, under French rule, there was a rat problem. A bounty was set—one cent per rat—which could be claimed by submitting a rat’s tail to the municipal office. But for each individual who caught a rat, it was optimal to amputate the tail of a rat, and set the rat free, so as to bolster the rat population and make it easier to catch rats in the future. In addition, on the outskirts of Hanoi, farms came up, dedicated to breeding rats. In 1906, there was an outbreak of bubonic plague that killed over 250 people.

Kelkar, Vijay; Shah, Ajay. In Service of the Republic . Penguin Random House India Private Limited. Kindle Edition.

By the way, the footnote associated with this little tale contains the link to the fuller story, and is worth reading in its entirety. It would appear that the cobra story from India doesn’t have any corroborative evidence. For those who don’t know the background, there is a very similar story from India, only involving cobras isntead of rats. (If you will permit a slight digression: I was telling both of these stories to my daughter, and her only observation was to note that the cobra story was unlikely because “won’t cobras be an example of an apex predator? They can’t grow as quickly as rats, correct?”)

But what can go wrong is what the government in Hanoi discovered – that the person for whom the incentive has been designed may well end up hearing a completely different tune than the one that the designer of the incentive intended. That is, the designer would like you to do x, but you end up doing y instead.

Teachers may set up assignments to incentivize learning, but students are playing a different game. They are looking to minimize efforts in order to maximize marks. Ditto for managers and members on a team in the corporate world. Ditto, as I and my wife have been discovering to our chagrin, for parents and kids! That last bit has been a particularly aggravating discovery, since both my wife and I are economists.

But this phenomenon of incentives not working out as envisaged has an entire “law” of its own, called Goodhart’s Law. This is what it says:

“Any measure that becomes a target stops being a measure”

You’ll find different phrasings of the same idea online, but that’s the simplest way to express the idea. If the measure (to stop the culling of the rat population) is rat’s tails that have been cut off, and you make this the target – well, they stop being a measure of the culling of the rat population!

And that’s why designing incentives is so very tricky. The Indian government found this out to its cost in the aftermath of demonetisation, for example, but rather than look for examples elsewhere, I think you learn about incentives best when you try to think of examples from your own life.

But you cannot – simply cannot – be a student of economics without appreciating both what incentives are, and how difficult it is to design and implement them. The study of this facet of economics will last for your entire life, and you will always find something interesting to learn about it, every single time.

Incentives matter.

Now, if you will remember, I had promised myself a cup of coffee if I finished writing this blog post without taking a break. Goodhart’s law would imply that I would indeed finish writing this blog post without taking a break, but presumably at the cost of either its length, or its quality, or possibly both. I leave it to you to judge if that has been the case.


I’ll go brew that cuppa.

2 thoughts on “Incentives Matter

Leave a Reply