Puzzles In Choice Judgmentalism: The "Consistency" Model Versus The "Budget" Model

There was a commentroversy in the NYT recently over whether it matters that decaf coffee may have small amounts of carcinogens. The article explained that some chemicals used to make decaf coffee may be dangerous in large quantities, but that the small amounts left in decaf coffee are believed to be safe. But some commenters were angry. They said the article was ridiculous: we're constantly ingesting toxic chemicals, so what kind of dope is worried about this trivial level of exposure?  

For me, it was a new manifestation of a cultural trope that since 2020 I have come to think of in terms of the "consistency" versus "budget" problem. In some mid-range phases of COVID, advice included strategies for doing some things you want to do while still decreasing the risk of getting sick: wear your mask on transit; socialize but avoid larger gatherings; go to the gym but try to stay further apart, etc. etc.

Some of this advice pissed people off. Obviously there were several dimensions, but the one I'm interested in here is the idea that somehow combining high-risk things and low-risk things is irrational and inconsistent. "Why wear your mask on transit if you're going to socialize with a bunch of people anyway?" people asked. There was special rage about the idea someone might wear their mask to go out to a restaurant, and then take it off to eat. What, did they think COVID would magically stay away while they were eating? How stupid.

At some point, I read something illuminating about the concept of a risk-budget and how the framing of "budget" versus "consistency" made sense of these strategies. With something like risk, consistency is irrelevant. Instead, what you want to do is think about apportioning your total exposure. So you might go out for drinks, but stay away from the gym, or you might go to the gym, but avoid going out for drinks -- both good ways of constraining your total risk while doing something you want to do.

The budget model illuminates wearing a mask to go to a restaurant and taking if off to eat. Wearing your mask on the subway? Easy and requires low sacrifice of something you wanted to do. Wearing your mask while eating? Impossible. So -- you can reduce your risk by wearing your mask while you're not actually eating, and still get to go out for dinner.

For some reason I don't understand, even though the budget model obviously makes sense, the consistency model is hard to let go of. I fall into it myself, despite consciously trying to avoid it. Even these days, if I'm weighing wearing my mask on transit, I find myself thinking: but that would be pointless and stupid, given the number of other high-exposure activities I am doing. The budget model shows that's not right, but it always comes into my mind.

Evidently I am not the only one, and there's a surprising kind of negative judgement of personal choices people perceive to be inconsistent -- even when those inconsistencies affect only the person in question. It just makes people mad that someone is worried about chemicals in their decaf, if they're also eating foods like Dortitos.

Why this angry judgment? I don't know. One guess I have is that it's misplaced moralizing: we know moral consistency is important, and we don't like hypocrisy; when we implicitly moralize the choices in question, we trip the "consistency" model wire and our brain goes into hyperdrive along that route. Another guess I have is that we don't like the way others are budgeting, and we wish they had other priorities, but it's easier to accuse someone of "inconsistency." But the matter is mostly opaque to me.

Since this is a full-service blog, I will end by telling you that if you want to avoid carcinogens in decaf coffee, you can buy coffee that is decaffeinated using a "Swiss Water Process" (SWP) instead of other methods that use methylene chloride or ethyl acetate. As other commenters pointed out, the article was all about safety for consumers and did not touch on the question of safety for workers in the coffee production industry, who are presumably in contact with larger amounts of toxins -- a good reminder that sometimes we don't need a consistency model or a budget model to figure out a better thing to do.