Pareto improvements are changes in allocation that harm no one. State A > State B if A is a Pareto improvement on B, i.e. if no one in State A is worse off than in State B and at least one person is better off.
This is a partial ordering, which means that for some states "A > B == undefined". The example is that (2,2) < (2,3) == TRUE (and similarly, (2,2) < (3,2) == TRUE), but (2,3) < (3,2) is undefined.
Since Pareto efficiency can not evaluate all states relative to each other, economists must define a total ordering which introduces subjectivity and ethical choices.
"Astute readers will have figured out by now that literally every point falling on the entire solid curve in the graph must be “Pareto optimal” by the economist’s definition of that term..."
Isn't that wrong? I can see how it could be true for the points on the curve between Y and Z. It's obviously not true for R and U, which result in decreased happiness for someone.
What is the curve supposed to represent, anyway? Or to put it another way, why is it on the graph? What does it mean? What purpose does it serve?
Y and Z are Pareto improvements on X, and they are also Pareto optimal. Points on the line between X and Y are also Pareto improvements on X but not Pareto optimal.
R and U are not Pareto improvements on X, because as you note some people lose happiness, but they are Pareto optimal, since according to the model, no further Pareto improvements can be made on those positions.
The point of the article, as I understand it, is that everyone would prefer Pareto improvements, but these are often infeasible. Economists seem to prefer Pareto optimums over non-optimums even when they are not Pareto improvements, perhaps because the word "optimum" and "efficient" have such positive suggestions.
I think the curve is sort of relevant to making this point, but the harder I think about it, the more wrong it seems.
It's meant to be (I guess) a theoretical model for what different public policies can achieve. But we would all prefer points between Y and Z to X, so why don't we enact policies that achieve that? If those policies are impossible or undiscovered, we should instead redraw the curve of optimality to reflect that, i.e. it should connect R, X, and U.
This is a question I've been giving a lot of thought to over the last 6 months or so. My conclusion:
We should be attempting to create an ethical, moral set of rules first, and then optimizing for efficiency, rather than creating an efficient system and then moralizing it.
Not that I have any answers for how to do this yet, but that's where I've decided my priorities lie.
Fortunately, this is a topic that has been widely discussed in academic philosophical circles. Read Rawls' "Theory of Justice" - he goes into great detail on how one ought to balance the competing forces of efficiency (e.g. maximal utility) and distributive justice.
I sympathize but the problem with that conclusion is "Who sets the ethical set of rules"? There's a lot of room for bottom-up using the democracy/capitalism combination, provided you don't let those who have money at one particular point in time totally break the system through lobbying, rent-seeking and regulatory capture (fail).
> There's a lot of room for bottom-up using the democracy/capitalism combination, provided you don't let those who have money at one particular point in time totally break the system through lobbying, rent-seeking and regulatory capture (fail).
the "provided" applies to all systems and doesn't seem solvable. The only way to avoid it is to keep such things outside the political sphere.
> Does this mean you're against corporations being able to spend money on political campaigns?
Corps and unions spend money on political campaigns because of ROI. No law will stop that as long as there is ROI, as long as there is some benefit to owning politicians.
So, if you don't want them to do that, you need to take the R out of the system, you need to make it so that owning politicians isn't profitable.
Pareto improvements are changes in allocation that harm no one. State A > State B if A is a Pareto improvement on B, i.e. if no one in State A is worse off than in State B and at least one person is better off.
This is a partial ordering, which means that for some states "A > B == undefined". The example is that (2,2) < (2,3) == TRUE (and similarly, (2,2) < (3,2) == TRUE), but (2,3) < (3,2) is undefined.
Since Pareto efficiency can not evaluate all states relative to each other, economists must define a total ordering which introduces subjectivity and ethical choices.