Imagine you are given three choices —A, B, or C. In the ranking, A is much preferred to B and B is exceedingly preferable to C. Which do you choose? Obviously, all else being equal, you’d choose A.

Now let’s add the following restrictions to your choice:

• You, your family, and your friends will all get A. But you must make the choice of A, B, or C, for other people who you will likely never meet.

• If you choose A, no one gets B and some (perhaps many) other people will be stuck with choice C.

• If you choose B, few people will get A but even fewer will get stuck with C.

Which do you choose now?

Before you know what the choices entail, you’d likely select B as the least bad option for the people you are choosing for. It’s not as good as the choice you yourself got but it’s still better than C.

But what if I told you A is a ban on child labor in Bangladesh and B is allowing children to work in a garment factory earning 53 cents per day. Does that change your decision?

Now what if I told you choice C was the children e prostitutes or go hungry. Does that shift your choice back to B?

Because C is so horrible, you’ll likely choose B even though you’d prefer the children get A. If so, you made the correct moral choice. You weighed the available options and chose the minmax, the choice in decision theory that minimizes the possible loss for a worst case (maximum loss) scenario.

Had I simply asked you whether child labor in Bangladesh should be prohibited, though, you’d have said it should. And it should be prohibited — unless doing so would lead to an even worse e for the children. Unfortunately, most people never bother to ask, “Would my preference lead to a worse e?”

Even more unfortunate is the fact that some people would still choose to ban child labor even if they knew it would lead to a worse effect. They could claim that they had “good intentions” and therefore weren’t responsible for the horrible e. The alternative is for them to align with those horrid people who seem to willing to allow the continuation of an evil like child labor.

But in some cases, the best option is to embrace realism and choose the least bad choice so that we don’t harm others.

As C.S. Lewis once wrote, “Of all tyrannies, a tyranny sincerely exercised for the good of its victims may be the most oppressive.” Thinking we have solved the problem of child labor may help us sleep better at night, but only at the expense of children who will suffer more oppressive forms ofabuse.

So does that mean that we should just throw up our hands in despair and do nothing? Not at all. Economist Ben Powell explains that while banning child labor doesn’t help, there are other options for indirectly eliminating child labor.