the economics of being right

i have a fairly strong opinion about something.

which is that we’ve become far too attached to our opinions.

i’m aware of the irony.

and i might be wrong.

but i think one of the more interesting problems in politics isn’t that people disagree. disagreement is what you would expect in a society made up of millions of people with different experiences, priorities, and values.

it’s that we’ve created environments in which admitting you might be wrong can feel more costly than actually being wrong.

and if you think about that for a moment, it’s a strange incentive system.

1a disagreement between friends

imagine two friends, momo and lolo.

they’re both in their early thirties, both interested in politics, and both want their city to become a place where ordinary people can afford to live.

one evening, they’re talking about rent.

momo thinks the most important thing is to build substantially more housing. she’s frustrated by the number of projects that get delayed, reduced, or rejected. to her, the arithmetic seems straightforward: if more people want to live somewhere than there are homes available, making it difficult to build more homes is unlikely to help.

lolo agrees that more housing is necessary. but she worries that new development can displace existing communities and that simply relying on the market won’t necessarily make homes affordable to the people who need them most.

they have a real disagreement. about the relative importance of different problems, about what works, and about the time horizon over which to judge success.

so far, so good.

now imagine that the conversation moves into a group chat.

momo shares a short article. lolo replies with a criticism. someone else joins in and calls momo a neoliberal. another person accuses lolo of being against progress. within a few messages, neither is really talking about housing anymore.

momo feels an urge to defend herself.

lolo does too.

and something changes.

the objective quietly shifts from understanding the issue to not losing the argument.

i observe this in myself, which is probably why it bothers me.

2the difference between a belief and an identity

i like the idea of treating beliefs as working hypotheses.

not because i think everything is equally uncertain. it isn’t.

some claims are extremely well supported. others are not. and there are plenty of moral commitments that don’t disappear just because somebody produces a clever counterargument.

but many political questions involve incomplete information, complex causal relationships, and competing values.

we have to form opinions anyway.

the problem begins when we confuse having an opinion with being that opinion.

if i say a policy is likely to work, i’m making a claim about the world.

if i begin to see that policy as evidence of what kind of person i am, new information becomes much more threatening.

because changing my mind no longer means updating an assessment.

it can feel like betraying myself.

or my friends.

or the people who already applauded me for taking the position in the first place.

3the mathematics of being willing to learn

there’s a concept from statistics that i think should be taught alongside political education: Bayesian updating.

the idea is simple.

start with what you currently believe. encounter evidence. ask how likely that evidence would be if your belief were true, and how likely it would be if your belief were false. then update.

in notation:

here, H is a hypothesis and E is new evidence.

let’s make it concrete.

suppose momo is 60% confident that a particular housing reform will meaningfully increase construction. then she encounters a piece of evidence that would be more likely to appear if the reform works than if it doesn’t.

imagine, purely for illustration, that:

• her prior confidence is 60%;
• the evidence has a 70% probability if the hypothesis is true;
• the same evidence has a 30% probability if it is false.

her updated confidence becomes:

about 78%.

she hasn’t discovered the truth. she has simply become more confident for a reason.

and if the next piece of credible evidence points in the other direction, she should be willing to update again.

the important thing isn’t the exact percentage. in real political debates we rarely have likelihoods this clean.

it’s the habit.

what do i believe? why do i believe it? and what would make me change my mind?

that’s a much more interesting set of questions than simply asking whether an opinion belongs to the right political camp.

4why certainty can be a rational strategy

(big) but, even if updating beliefs is intellectually desirable, it isn’t necessarily rewarded.

imagine that expressing a political position produces several kinds of utility.

there’s the satisfaction of saying something you think is true.

there’s approval from people you care about.

there’s status, attention, or a sense of belonging.

and there’s the effort and discomfort involved in publicly acknowledging uncertainty.

we could write a deliberately simplified model:

where T is perceived truthfulness, B is belonging, S is social status, and C is the personal cost of expressing the position.

the weights differ across people and situations.

in a private conversation with a trusted friend, truthfulness might dominate.

in a highly partisan online audience, the rewards for loyalty and certainty might be much larger.

the same person can therefore behave differently in different environments.

not because their intelligence changes.

because the incentives do.

this doesn’t mean everybody is consciously calculating social rewards before posting an opinion.

most of us aren’t.

incentives can shape behavior without ever becoming explicit.

and this isn’t exclusively a problem of one party, one ideology, or one political generation.

any community that makes belonging conditional on holding the correct opinions can create it.

5but isn’t certainty sometimes justified?

yes.

and i think any essay about political humility needs to be careful here.

some things really are known with a high degree of confidence.

some claims have been tested repeatedly and don’t deserve to be treated as equally plausible alternatives.

some political disagreements are fundamentally about values rather than empirical uncertainty.

and there are circumstances in which saying “perhaps both sides have a point” isn’t wisdom. it’s avoidance.

a doctor doesn’t need to be uncertain about everything to be intellectually responsible.

neither does a politician.

the goal isn’t permanent hesitation.

it’s calibration: matching the strength of a claim to the strength of the evidence.

i can be 99% confident about one proposition, 65% confident about another, and completely undecided about a third.

those are different states of knowledge.

they shouldn’t all have to be expressed with the same rhetorical volume.

there’s also a distinction between being uncertain about the best means and being committed to a moral end.

i can be strongly committed to reducing poverty while being genuinely uncertain about the best combination of taxes, transfers, housing policy, and education reform.

it might be the beginning of serious policy thinking.

not a contradiction.

6the vanity of being nuanced

there’s another possibility.

people who enjoy describing themselves as nuanced can become just as attached to that identity as people who enjoy describing themselves as certain.

i very much fit in that category.

it’s easy to feel intellectually superior because you can see complications that somebody else is ignoring.

but “it’s complicated” can become a convenient way of never committing to anything.

sometimes there really is enough evidence to choose.

sometimes waiting for perfect information means accepting a bad status quo.

and sometimes insisting that every argument has an equally compelling counterargument is simply another way of avoiding judgment.

so perhaps the ideal isn’t being the most nuanced person in the room.

perhaps it’s being the person who can say:

here’s what i believe. here’s why. here’s how confident i am. here’s what would change my mind.

and then actually mean it.

7what would it take to make honesty less expensive?

back to momo and lolo.

imagine their group chat had responded differently.

someone asks momo which evidence she finds most convincing. someone asks lolo what outcome would change her view. one of them acknowledges a good point from the other without being accused of surrendering.

they still disagree.

maybe strongly.

but the conversation produces something.

a better understanding of the problem. a clearer view of the trade-offs. perhaps even a small change in one of their positions.

that’s what i want more of.

not agreement for the sake of agreement.

and not an endless performance of intellectual modesty.

just a political culture in which the objective isn’t to emerge from every conversation with your original position completely intact.

because if every disagreement ends with everybody believing exactly what they believed before, i have to wonder what the discussion was actually for.

what if the problem isn’t that we have strong opinions.

but that we’ve built incentives to defend them even when we’re no longer entirely sure they’re right.

and once a political opinion becomes something we need to defend rather than something we’re willing to examine, it starts becoming something else.

an identity.

which raises the next question.

why do so many of those identities come with an entire package of opinions we’re apparently expected to hold?