Bryan Caplan  

The W.D. Ross Club?

PRINT
Richman Link... Opting for Enslavement...
Mankiw's latest paper takes the ethical significance of merit (or, as he calls it, "Just Deserts") seriously.  I'm tempted to propose that the two of us become the founding members of a W.D. Ross Club.  Its purpose: to promote the "distribution of happiness in proportion to merit." (The Right and the Good)


Comments and Sharing





COMMENTS (13 to date)
Kurbla writes:

You came relatively close to the legendary Marx's first phase of communist society from Critique of Gotha Program, called socialism by Lenin. There are two important differences: (1) surely you consider investment as merit, and (2) you wrote about "promoting" and not "organizing" or "enforcing" such merit based distribution.

Still, although you do not have to "abolish" inheritance and wealth earned through investment of such inheritance, you should at least "criticize" or "discourage" it, as some kind of, say, bad habit.

bdm writes:

Mankiw writes on page 19 of the paper:
"What about transfer payments to the poor? These can be justified along similar lines. As long as people care about others to some degree, then antipoverty programs are a type of public good...we would all like to alleviate poverty. But because we would prefer to have someone else pick up the tab, private charity can’t do the job. Government-run antipoverty programs solve the free-rider problem among the altruistic well-to-do."

Mankiw's views seem inconsistent with his caveat. If redistribution is a public good and provision of public goods can justify taxes, then Mankiw really has no rejoinder to liberal arguments for redistributive justice. All the redistributionist has to do to justify his position is prove that the public wants more redistribution.

Unless Mankiw is willing to agree to Caplan-esque welfare-state-abolitionism, I don't see his theory as providing any defense against arguments for more redistribution.

bbb writes:

@ "to promote the 'distribution of happiness in proportion to merit.'"

To try to promote the distribution of happiness in proportion to an externally defined value-standard is the central characteristic of a centrally planned and organized society.

Your preferred standard for the organized distribution of happiness in society carries just as much legitimacy as any other imaginable one.

Doc Merlin writes:

How do you seriously compare people's utility functions?
Utility being ordinal and not cardinal (even if in a few ways we can treat it as cardinal), and subjective means that we can't compare and see who gets more utility from something. We can only see for an individual, what their utility is from various sources, but cannot compare utilities between individuals!

Tom West writes:

but cannot compare utilities between individuals!

I find that phrase, while true in the specific, is practically false.

More to the point, the vast majority of human beings believe you compare utilities between individuals, act as if you compare utilities between individuals, and *do* compare utilities between individuals.

Oddly enough, in doing so, they produce a nicely functioning society.

In other words, comparing utility functions between individuals is "good enough".

Doc Merlin writes:

@Tom West

But we avoid comparing them between individuals as much as possible! This is why for example, one can't legally hit anyone without their permission (and in many cases even with their permission). Someone's enjoyment of hitting others may be many times the other's disutility, but we can't really say so we ban it.

Also, you can arbitrarily add constants to utility functions without changing their affect at all (because marginal utility is what matters not total utility), so again the idea of merit becomes very thorny.

Tom West writes:

But we avoid comparing them between individuals as much as possible! This is why for example, one can't legally hit anyone without their permission (and in many cases even with their permission). Someone's enjoyment of hitting others may be many times the other's disutility, but we can't really say so we ban it.

Wait, I think that's an argument for *my* side :-).

We assume everybody's (or almost everybody's) utility function is enough alike (i.e. comparable) that we feel comfortable in banning the practice altogether.

Hmm. This may be more complicated than I thought.

eccdogg writes:

"Mankiw's views seem inconsistent with his caveat. If redistribution is a public good and provision of public goods can justify taxes, then Mankiw really has no rejoinder to liberal arguments for redistributive justice. All the redistributionist has to do to justify his position is prove that the public wants more redistribution."

I think Mankiw is adressing the reasons for the redistribution not the specific policy.

You often here the word "Justice" used in these discussions. That seems to indicate that it is the right of the poor to take from the rich and that the rich owe it to the poor.

What I believe Mankiw is saying is that the poor do not deserve redistribution, but the rich/middle class may want to give them charity because they care for them and that due to free rider issues state sponsored charity may be a public good.

I think Mankiw's distintion may produce identical policy outcomes to one of redistributive justice, however I believe how you make the decision would be different. In a redistributive justice scenario if 60% of the poor vote to take from the 40% richer than them that is just fine. But in Mankiw's just deserts theory you would need to get at least a majority (and probably a large majorty) of the people losing money for the redistribution to be just. Because the argument is that the rich as a group want to give to the poor, but to stop free riders it must be forced from everyone.

Ryan writes:

Doc Merlin,

You can definitely compare utilities between groups of individuals, assuming effectively random group assignment and large enough sample size. In my field of health economics we do this all the time in cost utility analysis.

http://en.wikipedia.org/wiki/Cost-utility_analysis

Steve Roth writes:

I don't have time to address this in detail, so just two points on Mankiw's paper:

1. While he does not play the popular shell game of just analyzying personal income taxes for progressivity (he uses CBO figures that include and allocate payroll taxes, corporate taxes, etc.), he does ignore state and local taxes. I find this to be a wildly under-examined area, but the one big-picture analysis I've found says that (because state and local taxes are in total quite regressive) above about 60K a year, total taxes are essentially non-progressive.

2. He says that perhaps all economists can do is "try to estimate the cost of redistribution—that is, the negative impact on efficiency that comes with attempts to achieve more equality."

This reveals a profound a priori bias: that the effects of redistribution and greater equality on efficiency can only be negative. That presumption is wildly, simplistically, and demonstrably untrue. There are effects in both directions--countervailing, interacting, mutually reinforcing, and etc.

His presumption that policy choices are just about "picking the best point on the tradeoff" reveals a childishly simplistic, one-dimensional view of how economies work.

Since he obviously "knows better" (on some level of consciousness), I have to make a presumption as well: that he intentionally (on some level of consciousness) chooses to ignore that knowledge.

ryan yin writes:

Steve Roth,
Why do you say in #2 that it's an a priori assumption (or that it's not true)? It's not so much an assumption as a theorem. It's possible you don't like the theorem, but it's demonstrably untrue that all those economists (right-wing, left-wing or other) just assumed it.

I'm curious about what you have in mind. Suppose there exists some situation where the government could increase both equality & efficiency simultaneously through redistribution. Ok, why doesn't the gov't keep redistributing until there's a tradeoff? (If you assume the gov't does, then it follows you think that right now there's a tradeoff -- but you just said you don't think that.) Is it because (as Mankiw might be suggesting) that these two aren't the gov't's ? That doesn't seem to be what you mean.

Steve Roth writes:

ryan yin:

>Why do you say in #2 that it's an a priori assumption

Because 1. That's how he states it, and 2. I've read (large chunks of) his textbook and many of his other writings.

>or that it's not true

Because there are obviously redistributive policies that have positive effects. Almost every economist agrees that 20th century public education policies had profoundly positive effects. That's just one example out of dozens or hundreds.

>Suppose there exists some situation where the government could increase both equality & efficiency simultaneously through redistribution. Ok, why doesn't the gov't keep redistributing until there's a tradeoff? (

1. Maybe because people like Mankiw argue that redistribution can have only negative effects on efficiency. 2. Because economics isn't a simplistic continuum with one or two on/off switches. 3. Because it often takes decades to change policy directions.

ryan yin writes:

Steve Roth,
If that's what you took away from any of Mankiw's writings, or the writings of any economist, you might want to reread. (Just because someone does not at all points explain why they think a proposition is true does not mean that it's just an a priori assumption.)

Almost every economist agrees that 20th century public education policies had profoundly positive effects.

Setting aside the hyperbole, I'm hard pressed to think of one who thinks that's true for purely redistributive reasons (i.e., that there was no market failure besides the distribution of wealth but that free public education made the economy more efficient).

Comments for this entry have been closed
Return to top