Penalties and Persuasive Commitments
What are Commitments — Part 2
In Part 1 of this series, we introduced a Taxonomy of Commitments, which lays out four categories of commitments based on commitment goals and constraint types. Commitment goals reflect what the committer is trying to accomplish, while constraint types describe how the commitment constrains the committer’s behavior.
This post focuses on persuasive commitments and incentive constraints (the bottom left quadrant above). To assist in our exploration, we’ll continue referring to our shorthand characters: Charlie, the committer, and Peggy, the person Charlie seeks to persuade. Charlie and Peggy will meet another character along the way.
Persuasive Commitments Recap
By making a persuasive commitment, Charlie seeks to persuade Peggy to behave in a certain way. In our example from Part 1, the Charlie Car company wants Peggy Supplier to manufacture a unique part for 10,000 of Charlie Car’s new model, but Peggy is hesitant to invest in the requisite manufacturing capacity unless Charlie is really going to pay for all those parts. So Charlie makes a commitment to buy 10,000 parts, which persuades Peggy to manufacture them.
What this example leaves unexplained is how exactly Peggy has been persuaded. What is it that has convinced Peggy that Charlie will indeed pay for the parts? Charlie’s commitment is designed to constrain his options to “pay for the parts”, but where is that constraint coming from?
As we saw in Part 1, there are two main ways that Charlie can constrain his options: physical constraints and incentive constraints. Physical constraints are not relevant to our car parts example — and indeed, they are appropriate in only a minority of situations — but an incentive constraint is a good fit. If Charlie can successfully demonstrate to Peggy that it’s in his best interest to pay for the parts, then it will be easy for Peggy to believe that he will do so.
That is the power of an incentive constraint. Even without physically restricting Charlie’s behavior, an incentive can influence his behavior by making a given option more (or less) attractive. Note, though, that to create a successful commitment, it is not sufficient for an incentive constraint to influence Charlie’s. It must also convince Peggy that it will do so.
For Peggy to change her behavior in response to Charlie’s incentive constraint-based commitment, she needs to believe that the incentive is both sufficiently likely to be applied and strong enough to influence Charlie’s behavior — we can describe such an incentive as “credible”. Both of these factors depend on the form the incentive takes and who is enforcing it.
Penalty Enforcers
For the sake of clarity and simplicity, our focus in this post will be on penalties. Similar concepts will apply to rewards.
For a penalty to be credible, its enforcer must:
- Have the means to impose the penalty. If they don’t have the means to impose the penalty, then of course it is highly unlikely that the penalty will be applied when necessary, impacting its credibility. Enforcers who can apply the penalty with less delay, lower cost, and with less ambiguity help create a more credible penalty.
- Be fair. They cannot or do not have an incentive to change the rules of when the penalty will or won’t be applied (including how to measure whether Charlie’s behavior is worthy of punishment), regardless of what anybody else may want them to do. If the enforcer can be bribed or otherwise enticed into penalizing Charlie when he shouldn’t be penalized — or not penalizing Charlie when he should be — then the penalty is unlikely to influence Charlie’s behavior.
There are three broad categories of penalty enforcers; Charlie himself, Peggy, or a third party. Let’s see how they fare on the above criteria.
Charlie (first-party enforcer)
Charlie certainly has the means to penalize himself, but a self-imposed penalty is rarely credible. In theory, he could send some money to a cause he despises, place himself under house arrest, or even break his own nose as penance. But can we really expect Charlie to punish himself fairly? He will always have an incentive to change the rules and let himself off easy.
The only way Charlie can be a credible enforcer of his own penalty is to relinquish control of the actual enforcement to more subconscious brain functions that are disconnected from his ability to let himself off easily. Charlie can achieve this by selecting a penalty that would be imposed on his own psychology that is driven more by deeper emotional forces and outside the reach of his conscious mind — such as a hit to his sense of identity, honor, or ego. As one can imagine, the threat of a true decline in one’s sense of self-worth can be quite a powerful motivator, and under the right circumstances can serve as a credible penalty from Peggy’s perspective.
However, those circumstances are rare. Most Charlies’ identities are not sufficiently susceptible to self-inflicted punishment for this to work. Not everybody is a pirate following a code or a samurai with iron-clad honor, and even those of us who like to think of ourselves as principled people have to admit that we can often be quite “flexible.” In most situations, there’s little reason for Peggy to believe that Charlie will not find some way to squirrel out of penalizing himself.
Do you think there might be another way for Charlie to relinquish control without relying on others to enforce the penalty on him? Stay tuned for our next post!
Peggy (second-party enforcer)
Peggy — or a group of Peggy’s — can impose a penalty on Charlie when there is a possibility of repeated interactions between Charlie and Peggy(s). If Charlie fails to fulfill his commitment in the first interaction, Peggy can penalize him by avoiding further interactions with him, and/or by broadcasting his lack of credibility so that others also avoid interacting with him. This is a reputational penalty.
Even when subsequent interactions are not expected, it’s also possible for Peggy to punish Charlie with bodily harm or blackmail of some form. Consider loan sharks, who operate outside the legal system. When borrowing cash from a loan shark Peggy, Charlie can expect to have his arm twisted (or worse) by Peggy if he doesn’t pay her back.
Note that in both repeated- and single-interaction scenarios, the extent to which Peggy will penalize Charlie fairly (i.e., not twist his arm if he pays back on time) is held in check by reputational forces on Peggy herself. Loan shark Peggy’s future clients won’t borrow from her if they can expect to have their arm twisted regardless of their repayment, and repeated-interaction Peggy’s gossip won’t be as effective next time if her friends can’t trust her story.
Trey (third-party enforcer)
If neither Charlie nor Peggy are equipped to enforce a penalty on Charlie for failing to fulfill his commitment, then they can enlist a third party — who we’ll call Trey — serve as a penalty enforcer.
Trey can be anybody or anything, as long as he has the means to enforce the penalty and Charlie and Peggy both trust him to do so fairly. Trey could be an individual person, such as a mutual friend or somebody with a well-known reputation to uphold. Trey could be a private organization (a group of people), such as a social club both Charlie and Peggy belong to or a company specializing in arbitration.
Trey could also be a government’s legal system. A large part of what legal systems provide is a mechanism to penalize Charlies for not keeping their promises to Peggy’s. Legal systems can impose penalties (such as fines or imprisonment) on residents for not abiding by local laws and regulations. Legal systems can also act as the third-party enforcer on agreements between parties like Charlie and Peggy. By entering into a legal agreement (i.e., a contract) with Peggy, Charlie is effectively telling his government’s legal system that it should penalize him for not taking the action he promises in the contract. Legal contracts are third-party commitment devices, where the third-party enforcer is the legal system.
Our Charlie Car company example is a good illustration of a legal system acting as a third-party penalty enforcer. Charlie’s commitment to Peggy to purchase 10,000 of her parts can take the form of a legal contract that outlines the penalties Charlie must pay if he does not follow through on the purchase. In that case, Peggy can bring Charlie to court, where a judge will consider the facts and enforce the penalty.
Which leads to the final type of Trey: smart contracts (or more generally, smart contract blockchain networks like Ethereum). Whereas a legal contract ultimately relies on the interpretation and order of a judge as the means to enforce penalties defined within it, a smart contract both describes the commitment and penalty — the logic is embedded into the smart contract’s code — and acts as the means to enforce it!
Here’s how it works. The commitment logic — including the conditions under which the penalty should be imposed — is embedded into the smart contract’s code. If those conditions are met, code is triggered that executes the penalty. Once the smart contract has been deployed, it cannot be stopped nor can its rules be changed. This has the triple benefit of reducing the cost, time, and uncertainty associated with enforcing the penalty, all of which bolster its credibility, creating one of the most robust commitment devices.
Astute readers may be wondering whether smart contracts don’t actually belong in the third-party enforcer section. Could they also be categorized as a second-party, or even a first-party, enforcer? It’s a valid question, and one to which we’ll return in our next post covering self-commitments (spoiler alert: we think they are).
Penalty Types
As we’ve already seen, penalties can take several forms.
Financial penalties, such as a loss of money. In the course of human history, these types of penalties are a relatively new development. Like money itself, financial penalties can apply to a wide range of situations and types of commitments, and they can function as the sole form of incentive constraint or complement another type of incentive. This flexibility, bolstered by their ability to be easily scaled up or down, makes them very effective incentives. The main catch is that they typically can only be credibly enforced by third parties (Trey). As we’ll see in the next post, that poses a challenge for self-commitments.
Social penalties, such as a loss of reputation. By definition, social penalties can be enforced only by Peggy or Trey, but can be a powerful motivator. Humans are social creatures, and the threat of lower social standing and decreased access to subsequent interactions can be an effective constraint on Charlie’s behavior.
Physical penalties, such as a loss of personal freedom (imprisonment) or bodily harm. We reference these types of penalties only for analytical completeness; CommitPool does not condone such penalties, and will never utilize (or even reference) them in our protocol or product.
Psychological penalties, such as shame or a loss of ego or sense of self-worth. With the notable exception of abuse (which, like physical penalties, we do not condone), psychological penalties can only be imposed by Charlie himself. Since they are self-imposed, they can be credible penalties when they are involuntary psychological reactions, not consciously calculated responses. Which is why psychological penalties are limited to deep identity and self-perception because they are involuntary reactions. Still, their strength varies person to person, which impacts how certain Peggy can be that a given Charlie will be constrained by one.
Wrapping Up
In this post, we reviewed the different types of penalties and penalty enforcers that can undergird a persuasive commitment. Not all penalties can be enforced by all types of enforcers, and some combinations work for more commitment scenarios than others. Financial penalties enforced by third-parties, for example, are quite common — and effective! — in modern economies, and reputational penalties enforced by Peggy’s are common everywhere, while self-enforced psychological penalties are rarely employed within persuasive commitments.
Those same patterns, however, do not hold for self-commitments. In our next post, we’ll take a closer look at those dynamics and — hint hint! — how we’re designing CommitPool to change them.
Thanks to Reuben Youngblom and Ken Beckers for reviewing several early versions of this post.
The What are Commitments? series:
- Part 1: A Taxonomy of Commitments
- Part 2: Penalties and Persuasive Commitments (this post)
- 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
