Occasionally, I happen upon people who claim that they base their economic justice views on the idea of Pareto Improvements. On this view, economic justice requires that all moves to make individuals better off must not make other individuals worse off. Those who promote this view often seem to have roughly laissez-faire capitalist preferences. In this post, I outline just a few of the problems with this view, especially as applied to reach laissez-faire capitalist conclusions.
1. Initial Appropriation Is Unjust
Initially, nothing is owned. All of the land, for instance, belongs to nobody. In its unowned state, by definition, anyone may access it at any time. A move that transforms an unowned parcel into an owned parcel is not a Pareto Improvement. Everyone who is not the new owner of the parcel is immediately made worse off by the appropriating move. Their previously existing access to the parcel is taken from them.
Some try to argue that initial appropriation does not violate the Pareto Improvement requirement, but they fail. Nozick made the most overt effort to do this (you can also read things into Locke to suggest he’s doing it as well, though that’s obviously anachronistic). But Nozick’s effort is a sleight of hand. He claims that, in the medium and long run, allowing appropriation will so improve things economically that those immediately dispossessed by appropriation will be better off in welfare terms.
I call this a sleight of hand because it seamlessly shifts between Pareto as applied to individual transactions and Pareto as applied to institutional design. Nozick is not really saying that, as a transactional matter, the move where you steal everyone’s access to a parcel to land passes the Pareto test. He’s saying, as an institutional matter, the move where you impose laissez-faire ownership institutions passes the Pareto test in the medium to long run against the baseline where you don’t impose those institutions.
The crucial problem here, as I’ve explained before, is that this institutionalist Pareto argument succeeds only because Nozick is comparing laissez-faire ownership in the medium and long run to a hypothetical system with no resource governance institutions. He’s essentially saying that the imposition of laissez-faire institutions outperforms the no-institutions baseline. But this is true of almost any set of institutions. The imposition of social democratic institutions of resource governance also outperforms the no-institutions baseline in the medium and long run.
So, when applied in its purely transactional sense, the Pareto Improvement requirement makes all appropriation impossible because it makes people better off (new owners) by making others worse off (those whose access is taken from them). In this Nozick institutionalist sense, the Pareto Improvement requirement does not lead to laissez-faire institutional conclusions because basically all institutional sets are Pareto Improvements over the no-institutions baseline.
2. Future Flows Belong To Nobody
Bracket for a moment the issue of how the Pareto Improvement requirement can square with the reallocation of existing wealth. I have things to say on that, but I want to go for lower-hanging fruit here by confining the discussion to what the Pareto Improvement requirement says about the proper allocation of future flows of income.
When you talk to these Pareto Improvement guys, they’ll usually object to changing the institutions that govern the way income is distributed. So, for instance, if you propose that we reform our institutions to be more social democratic, they say that this violates the Pareto Improvement requirement because it improves the plight of the bottom only by making the top worse off. But this is not true.
To see why this isn’t true, consider this question: who has received the income of 2016? That’s not a typo. Actually think about that. To whom has the income of 2016 been distributed? This is an absurd question, right? Nobody has received that future flow of income. It presently belongs to nobody. It will only belong to someone after it has been distributed next year.
Thus, it doesn’t matter (for Pareto Improvement purposes) what institutions we use to distribute 2016’s income. Nobody is made worse off because nobody is deprived of something that they used to have. Pareto can’t be used, then, to argue against using different distributive institutions for future flows of income.
Even if you supposed that somehow Pareto required that people be entitled to the same income level next year as they had the previous year, that would still leave “growth” as belonging to no one in particular. Some prior anti-poverty proposals have actually used this logic to explain how they would be funded. They would say something like: over the next 10 years, GDP will go up by 20 percentage points, and so we are only asking that we distribute 5 of those percentage points to the poor.
Per (2) above, I don’t think this particular reasoning is necessary to pass Pareto because all future flows of income presently belong to nobody. But it deserves pointing out that, even if you stretch Pareto in such a fashion, all income growth above the present surely belongs to nobody right now and can be distributed in any way without running afoul of the Pareto Improvement requirement.
4. Whose Expectations?
After you make point (2) and (3), the last ditch effort of the Pareto Improvement guys is to say that future flows of income (even the future growth of income) should be assigned as if they already belong to those who expect to get them. So, for instance, levying a tax on a rich person that only managed to capture the person’s market income growth (meaning the rich person still gets as much income this year as last year) would violate Pareto because the rich person presently expects to get that growth amount as well. Relative to that expectations baseline, then, it is said that the rich person was made worse off and that therefore the tax violated the Pareto Improvement requirement.
Perhaps it’s needless to say, but this expectations baseline business is totally unworkable. I could go on and on explaining why. Polls of young Americans show a totally impossible percentage of them expect to be rich adults. More generally, if you were to add up all of the expectations of Americans (a relatively optimistic group), you’d probably find that the future flows of income are claimed three times over. Against this background, how could any institutional future avoid violating the Pareto Improvement requirement?
Moreover, one of the core components of capitalist systems is to constantly dash expectations. This is what so-called disruptive innovation is all about. When a new company comes on the scene, it takes market share from its competitors and may drive those competitors out of business altogether. This move is not a Pareto Improvement because the expectations of the investors (and other actors) involved in the incumbent company are not satisfied. You can see this in real time by watching stock prices fall. What is the difference between that situation and the situation where tax changes also cause the company to take in income lower than their expectations? None, at least not in terms of the Pareto Improvement requirement (as modified by this expectations gloss).
In the realm of working people, the dashing of expectations is a far more common phenomenon. Consider the person who, at the age of 30 or so, had a union forklifting job and fully expected to have that job until he retired, with all of the raises and seniority that generally entails. Then, that person’s company is bought up, the union is run out, and the raises he expected never come under the new management. How can this process be justified under this Pareto Improvement requirement (as modified by this expectations gloss)?
I could go on, but it suffices to say this expectations gloss gets you nowhere if you are trying to justify laissez-faire capitalism. Satisfying the present expectations of all of the people is not only impossible (meaning violations of Pareto are inevitable) but is also antithetical to the basic mechanics of capitalist innovation.
This Pareto Improvement rationale for laissez-faire capitalism is unsuccessful, and the above is only a short sample of why.