A Taxonomy of Commitments
What are Commitments? — Part 1
Here at CommitPool, we’re big fans of commitments. We believe that enabling individuals to make credible commitments to themselves will help people stick to their personal goals and become just a little bit healthier, happier, and wealthier as a result. In working to make that happen, we’ve been thinking a lot about commitments, what they are, and how they work. This post, the first in a series, is our attempt to share some of those thoughts.
Let’s get to it!
What is a commitment?
The common definition of a commitment — a device for constraining the one’s future available actions — is incomplete. To fully examine how commitments work, we also need to understand their purpose: why is the committer voluntarily restricting their own options? We contend that there are two major categories of reasons, creating two major categories of commitments.
Persuasive Commitments
For when you’re trying to get somebody to do something that they will only do on the condition that you take a particular action. Restricting your own options can persuade the other person to do what you want. There are two primary actors in a persuasive commitment. The committer, who in this post we’ll call Charlie; and the person Charlie is trying to persuade, who we’ll call Peggy.
Here’s a simple example: the Charlie Car company wants to build 10,000 cars of their new model, which uses a unique part made by Peggy Supplier. Other cars don’t use that part, so Peggy will only manufacture enough parts on the condition that Charlie buys them. So Charlie makes a commitment that he will buy 10,000 parts from Peggy, which persuades Peggy to make the parts.
Self-Commitments
For when you’re trying to ensure that your future self takes a particular action. Your current self may suspect that your future self’s priorities will change, so only by restricting your future self’s options to that action can you ensure it will be taken. Self-commitments only have one actor: Charlie the committer.
A simple example: Charlie has a goal to exercise regularly, but he recognizes that, when it comes time to actually do the exercise, he sometimes puts it off. To keep his future self from straying, he makes a commitment to himself to exercise.
Constraints
In addition to the commitment type/goal, commitments can have different mechanisms for constraining behavior. We contend that, again, there are two main categories of constraints.
Physical Constraints
Physical constraints are literal restrictions on action. They can be “physical” in the sense of the analog world, like removing access to unhealthy foods or locking oneself in a room. They can also be digital, such as locking funds inside a smart contract.
Everything about a physical constraint is established along with the commitment itself, before Charlie takes (or does not take) the promised action. Crucially, a physical constraint is not dependent on Charlie’s behavior, nor does it depend on anything happening following Charlie’s behavior. The only thing that affects a physical constraint’s effectiveness is it’s restrictiveness: how well it restricts Charlie from taking an action other than what he promised.
Incentive Constraints
Where physical restraints are restrictive, incentive constraints use influence to constrain Charlie’s behavior. They actually don’t restrict Charlie’s behavior at all; instead, they guide his choice of action by adjusting how attractive his various options are. Note that we can still call this a constraint because the relevant perspective is Peggy’s, not Charlie’s; as long as Peggy sees that Charlie’s realistic options are limited to what he’s promised, she’ll act as if Charlie will indeed take the promised action.
Incentive constraints can apply their influence in two directions. Whereas a physical constraint can only prevent Charlie from doing something other than what he’s promised, an incentive constraint can make an option either more attractive (i.e., with a reward) or less attractive (i.e., with a penalty).
This flexibility, however, comes with a trade-off. Like physical constraints, incentive constraints are established previous to Charlie’s actions. But what is established up-front is the mechanism for the reward or punishment, which is implemented only after Charlie takes an action (and is conditional on that action). Someone or something must execute that mechanism once Charlie takes his action. So unlike physical constraints, incentive constraints are necessarily under somebody’s control — and, as we’ll see in the next post, that introduces all sorts of challenges.
What’s Next
Due to their flexibility, incentive constraints are applicable to many more situations than physical constraints, but they’re also quite tricky to implement properly. The remainder of this series will focus primarily on incentive constraints. Part 2 will dig into persuasive commitments, and Part 3 will examine self-commitments in more detail.
Thanks to Reuben Youngblom for reviewing several early versions of this post.
The What are Commitments? series:
- Part 1: A Taxonomy of Commitments (this post)
- Part 2: Penalties and Persuasive Commitments
- Part 3: Self-Commitments (upcoming)
About CommitPool
CommitPool is a protocol and app for holding yourself accountable to your goals.
Connect with us here:
- www.commitpool.com
- @CommitPool on twitter
- CommitPool on github
- CommitPool discord community
