If you are wondering whether a food needs refrigerating, you look at the label until you either see an instruction to refrigerate it, or have looked at the label enough to exclude the possibility of it containing that instruction somewhere. This seems clearly worse worse than the label just always saying whether or not the food needs refrigerating, so the procedure terminates as soon as you find that sentence.

It’s only a tiny bit worse—only a few seconds are at stake each time—but it is so clearly worse, it is interesting to me that it remains in the worse state.

But how would I expect it to get better? Here are some partial mechanisms by which I might expect things to get better usually, which this is evidence about the strength of:

  1. The market: people would be marginally more likely to buy a product with slightly more helpful labeling, e.g. they might pay an extra cent for saving several seconds of their time. I don’t feel surprised that this doesn’t work. For one thing, it would take the customer a large fraction of the time saved for each item to check whether it has the more useful label. Is there some theoretical reason that this should turn out ok in a homo economicus utopia?

  2. Brand affect and loyalty: people won’t check ahead of time whether a product will save them a second later, but after finding a product slightly more annoying to deal with, they will like the brand slightly less, and be less likely to buy it in future, even though the object they receive is identical. I don’t intuitively expect it to work on its own, I think in part because I expect people to interpret very small things in light of the larger vibe they are getting from a product. For instance, if it was helpful in several ways, they might interpret a helpful label as more of that, whereas if it was in other ways minimalist, this might detract from that vibe. I don’t have an argument against it working in theory, but at a very small scale, since it is a very small problem for a customer.

  3. People being thoughtful and nice: maybe even without incentives, it basically doesn’t cost more to have more helpful writing on your label, so why not do it? My guess is that people writing labels don’t think of it. I’m not sure how reasonable that is, because I haven’t emotionally comprehended how many people are thinking about what to write on labels.

  4. Some sort of feedback from users of labels: if the problem with 3) was lack of knowledge, then the customer seems like the most likely person to bridge that gap. Probably if you have users test a simmer sauce or something, you don’t go far enough down the details of their response to catch things like this? And nobody writes to the company to complain about something this small. Virtually all people have better things to think about than even noticing this cost.

Others? (I expect I could think of some, but it is my bed time.) Happy Thanksgiving!