A Minor Act of Pedantry

The timing of this post is completely random, but I happened to see the phrase "economic efficiency" in a reading for class and, as it often does, it annoyed me. That word, "efficient," is effectively the economist's term of art for "good." All things good are efficient; all things efficient are good. The actual meaning of the word "efficient" in ordinary English speech is lost. That definition is specifically about the ratio of inputs to outputs: something is efficient which gets a lot for a little. A little work, perhaps; one works efficiently if you can do in only an hour what someone else might take a day to do, especially if they could be getting the work done a lot faster if only they went about it more, well, efficiently. (It is not inefficient, in other words, to just be a very slow reader, and therefore to take longer to read, but it is inefficient to be a fast reader who nonetheless takes forever to get some reading done because you keep checking Facebook or writing blog posts, i.e. what I'm doing right now.) Efficiency is in general a good thing, as long as the thing we're producing efficiently is a good thing and/or its costs are a bad thing. But it's not the only good thing. There are times when it matters not so much to be efficient as to be successful, even if it requires spending resources "wastefully." For instance, sabermetrically inclined baseball analysts focus a lot on "efficiency," the idea that one shouldn't commit too much payroll space to an unproductive player. But if you're a team that has a lot of money, and you have a lot of good, cheap players but a couple of holes on the roster, well, it might make sense to overspend, perhaps even quite a lot. Pocketing the money doesn't give you a better team. Efficiency maximizes the return for your resources, but if you have ample resources, sometimes you might as well just throw everything you have at the problem, even if that efficiency ratio suffers.

All of which is a prelude to me complaining that the way the word "efficiency" is used in economic and/or pseudo-economic political discourse conflates two very different things, one of which is efficiency and one of which isn't. The thing that actually is efficiency is the problem with which classical economics generally concerned itself: assuming a relatively fixed pool of resources, how can society get the most value from them? How, that is, can each resource be put to its respective best use? So, for instance, one might argue, as many modern liberals do, that because money, income, and wealth have relatively sharply declining marginal utilities, it increases aggregate welfare to transfer a dollar from a rich person to a poor person. That's an efficiency argument, really, though right-leaning economists are loath to admit as much: it is the argument that the same resource will be more productive, of human happiness at least, if put to one use instead of another.

The other thing is a very different thing, and that's how to increase the overall stock of resources. That is essentially the problem of growth, and in principle it's quite different from the problem of efficiency. Now, the two can be connected. For instance, a conservative answer to the above liberal argument about redistribution would be that rich people invest their money, which increases overall productivity going forward and therefore produces more total long-term benefits than just letting the poor person consume that dollar right now, even though that poor person would get more out of that dollar right now than the rich person would. That's essentially saying that one use of a certain resource is more productive than another use of that same resource because it will generate still more resources.

But, importantly, there are significant areas of non-overlap between growth and efficiency. Many efficiency questions just don't involve growth questions one way or another; for instance, magically transporting an umbrella from the umbrella stand of someone in a city experiencing sunny weather to the hand of someone in a rainy town who for some reason doesn't have an umbrella would put that umbrella to much better use. It would not increase society's stock of resources. Conversely, growth is often not about efficiency. People talk about the incentive structure of the tax code as being relatively more or less "efficient," but that doesn't seem like the right way to look at it. Those kinds of incentives can certainly influence how much work people decide to do, which affects the overall level of GDP. But they don't improve the ratio between the inputs and the outputs. They increase the inputs, and thus the outputs. (The question of whether work or leisure is actually a better use of that person's time, all things considered, is more like an efficiency question, but it usually gets ignored in these discussions; maximizing GDP is assumed to be the goal.) And when an economy is operating below potential, as ours currently is and has been for a while, we're sort of in the situation of the rich team with lots to spend and a few big holes that need filling. We've got resources that are being left idle, so even a not-very-efficient use of them will be better for growth than what we've got now.

In other words, there are a lot of things that increase total production. Some of those things are about efficiency, about getting more out of our inputs. Some of them are about growth, about getting more outputs by putting in more inputs. Some of them are about a mixture of the two. People usually use the word "efficient" to describe any policy which has a net-positive economic outcome, but that's just a pure linguistic contortion.