Fairnessland and Economic Growth

By Jason Brennan On March 15, 2011 · 33 Comments · In Liberalism, Social Justice

Update: Some readers are having a surprisingly difficult time understanding this post. So, here is the question I'm interested in:

Suppose it turned out, empirically, that improving the income level of the poor at any given time by equalizing incomes eventually leads to the poor in that society having less than they otherwise would have had under a less equal but faster growing scheme. If so, which is preferable, all other things equal?

A. Equalize things now so that the poor now do much better.

B. Allow for growth so that the poor in the future do much better.

It's also worth asking how much one's answer depends upon just what the differences in growth rates are stipulated to be.

Imagine there are two nearly identical societies, Fairnessland and ParetoSuperiorland. Imagine both are some form of liberal society, and both equally protect citizens' political and civil liberties.

In both societies, there are three economic classes, poor, middle, and rich. Fairnessland's basic structure allows inequality, and the inequality it allows at any given year is to the maximal benefit of the poor. Empirically, it turns out that at any given year, the distribution of wealth that maximally benefits the poor is 15-19-24. ParetoSuperiorland, on the other hand, has a basic structure that provides less well for the poor. Throughout its history, ParetoSuperiorland has a 10-20-40 distribution.

However, imagine that because ParetoSuperiorland allows more inequality, it has faster growth. (I'm not here claiming that allowing inequality really does cause growth. So no need to argue with me about that.) Fairnessland has an annual growth rate of 2%, with each class's income growing equally quickly. Suppose, however, that ParetoSuperiorland has a faster growth rate.

Here is the comparative performance of the two economies over time, starting in year 1900, if ParetoSuperiorland has a 4% growth rate.

	ParetoSuperiorland 4%			Fairnessland 2%
	Poor	Middle	Rich	Poor	Middle	Rich
1900	10	20	40	15	19	24
1901	10.4	20.8	41.6	15.3	19.4	24.5
1902	10.8	21.6	43.2	15.6	19.8	25.0
1925	26.7	53.3	106.6	24.6	31.2	39.4
1950	71.1	142.1	284.3	40.4	51.2	64.6
2000	505.1	1010.1	2020.2	108.7	137.7	173.9

Here is the comparative performance if ParetoSuperiorland has a 3% growth rate:

	ParetoSuperiorland 3%			Fairnessland 2%
	Poor	Middle	Rich	Poor	Middle	Rich
1900	10	20	40	15	19	24
1901	10.3	20.6	41.2	15.3	19.4	24.5
1902	10.6	21.2	42.4	15.6	19.8	25.0
1925	20.9	41.9	83.8	24.6	31.2	39.4
1950	43.8	87.7	175.4	40.4	51.2	64.6
2000	192.1	383.4	768.6	108.7	137.7	173.9

Feel free to imagine that in ParetoSuperiorland, the middle and rich classes have even higher growth rates than the poor.

(Note: This is a conceptual post meant to discuss values. I'm not making an empirical, economic argument here.)

Some issues:

1. Note that at any given time, the poor of ParetoSuperiorland would do better by transforming ParetoSuperiorland into Fairnessland. So, at any given time, the poor of ParetoSuperiorland would see a 50% increase in their income if their country adopted the institutions of Fairnessland. Should ParetoSuperiorland adopt the basic structure of Fairnessland?

2. However, over time, ParetoSuperiorland does better for its poor than Fairnessland. In the first example, when the ParetoSuperiorland's growth rate is 4%, the poor class starts doing better in only 25 years. Does this make ParetoSuperiorland superior to Fairnessland?

3. Suppose you could be born into one of these countries. Which country would you prefer to be born into? Why? Should everyone else have the same preference? How could a country justly decide between being like ParetoSuperiorland and being like Fairnessland?

4. Suppose ParetoSuperiorland and Fairnessland are neighbors and allow immigration. Would the poor from ParetoSuperiorland tend to immigrate to Fairnessland or vice versa?

5. Some philosophers claim that in order for a society to be just, people in that society need to affirm the right theory of justice, or something close enough to it. Suppose that everyone in Fairnessland believes in the right theory of justice, whatever that is, but everyone in ParetoSuperiorland believes in some wrong theory (not a terribly wrong one, but a wrong one nonetheless). What difference does that make in your overall assessment of these two societies?

David Sobel

From what baseline is Pareto-Superiorland pareto superior? It seems not to be pareto superior to Fairnessland as some people’s lives would be better if they lived them in Fairnessland. Consider the person born in 1900 and who dies in 1928. Perhaps the idea is that everyone’s expected well-being is higher in Pareto-Superiorland given their life expectancy and it is in that sense pareto-superior even though people who die young would have had better lives had they lived in Fairnessland? Surely some people’s life expectancy is less than 30 some years and so perhaps the argument would need to rely on average life-expectancy or life-expectancy behind the Veil?
Jeff

I think you have a flawed assumption here – that the fact that the poor in ParetoSuperiorland in your hypothetical have more income actually means that they’re “doing better” than their peers in Fairnessland.

Why might this be untrue?

1. Because the poor in ParetoSuperiorland are at a much greater disadvantage in terms of their ability to compete for goods and services against the middle and rich classes than are the poor in Fairnessland.

2. Because the narrower income gaps may indicate more class mobility in Fairnessland. In other words, it may be significantly more attainable for a member of the poor class to become middle class.

3. Because most recent research suggests that absolute income (once you’re past the point of satisfying all your basic needs) does not actually increase happiness. We’re not after income in the absolute sense, we’re after doing well compared to our peers. Hence, more societal inequality, more people in a bad position compared to their peers, less happiness.

A couple of links relating to the third point:

http://www.outsidethebeltway.com/happiness_is_being_richer_than_your_friends/

http://www.eurekalert.org/pub_releases/2010-12/uosc-olh120910.php
http://profile.typepad.com/theotherchuckd TheOtherChuckD

Welcome to the first truly weak post on BHL.

What you’ve shown us is that, over the course of a century, a country that maintains either double or 50% more annualized growth than some other country becomes much wealthier over the course of a century. This is the same magic of compounding we learned in middle school math by calculating how much money Peter Minuit would have had if he put $24 in the bank in the early 17th century instead of using it to buy Manhattan Island (the answer: lots).

Aside from the absurd notion that countries keep stable policies for a century at a time, this chart fails to answer big questions:

- Why do you assume the number of poor and the poor’s share of wealth stays the same given the amorphous policies you posit? Does this ever happen?

- Is there massive inequality within the bottom deciles, or do all poor have enough to live with a bit of dignity? That’s a central question for people debating the welfare state.

- What is it about ParetoSuperiorland’s policies that generates the inequality and the growth? Are you making assumptions about the growth effects of marginal tax breaks? Wage, hour and workplace safety laws? Tariffs? Are the assumptions tethered to any empirical evidence?

- Do the poor have political power? Will they ever demand it? If they do, what will they ask for? At what cost?

- Which society is more likely to maintain steady growth at the rates you describe over the time period?

If you make the assumption that the policy you like yields large enough benefits to “raise all boats” over the disfavored scenario, your scenario wins. All you’ve shown is that you can pick numbers to make your “model” work.
http://profile.typepad.com/andrewlevine Andrewlevine

1. Yes, the flatter income structure enhances a sense of social belonging and trust that is essential for building societies that are worthwhile to live in (in terms of non-income considerations).

2. Yes, if true, a ParetoSuperiorLand that improves over time is better than FairnessLand. We can predict the future, I’m assuming here.

3. I swear I’m not trying to be difficult for no good reason, but I find it hard to put myself in the mindset of someone who cares to get as much income as possible for its own sake. This is just so inconsistent with my values that I’d have to say “I want to be born in whatever country those people don’t pick.”

4. Fairnessland would get the majority but not all of the poor migrants. It would get a minority of rich migrants but not zero.

5. Taking consideration of your new neighbors’ values into account is critical when deciding where to live. There’s a good reason the people with the means and inclination to move absolutely anywhere they please, don’t all congregate wherever their wealth will be maximized (Monaco, the Cayman Islands), but where the people share their values (New York, London, Paris). And why so many people in general try to escape paranoid, crime-filled states in favor of more neighborly ones. So having the “right” theory of justice (if there is one) clearly matters to people in selecting one’s compatriots. To answer the question, I’d start by brushing up on my Fairnesslandese and looking into their immigration laws. (And also where I could find a link to donate to the ParetoSuperiorLand Pro-Fairness League.)
John V

Jason,

In answer to what your post was getting at:

Yes. Left liberals are more concerned with relative numbers than absolutes.
Fernando Teson

A fine post, Jason. It highlights two problems in political morality: Is income inequality per se problematic? How important is it that people have the right moral impulses, and how does one measure this against the concern for the material situation of the worst-off? I know people who would say that Fairnessland is a better society in spite of the fact that the poor are worse off. This assertion requires a more sophisticated political theory than the one that says that economic institutions should be arranged to benefit the poor.
David L

It seems to me that to answer the “ought” question, you have to assign some theoretical discount rate to human utility. Some of PSL’s poor may prefer to switch to FL’s more favorable current distribution structure right now, favoring higher current distributions over the higher future distributions they would receive if they retained PSL’s structure. Some may prefer to bear PSL’s lower current distributions in anticipation of economic growth bringing a better life in the future. You could argue that this is due to different discount rates, although of course most people don’t think of it that way.

That assumes rational choice, of course. In reality some poor may prefer to have a higher distribution as a percentage of the distribution of the rich, even if it’s a lower absolute distribution. That’s not rational, but people aren’t rational.

The bottom line is that no one person is in a position to answer the “ought” question for anyone but themselves because of varying individual preferences. So, imperfect an answer as it is, you have to rely on some implementation of democracy to determine which structure to use.
Jason Brennan

OtherChuckD:

I’m just asking questions about value. I’m not making an economic argument about how institutions work. I thought my caveats would help you understand that.
brian

1) Thanks for framing the question this way. It sheds light on how intractable the division is, while not relying on arguments about one side hating the other, not caring about the poor, etc. My preference, especially given the existence and ability to migrate, would be to keep ParetoSupiorland as is and allow the two different structures to compete.

2) I believe it does make ParetoSuperiorland superior. Money provides people with options, and the ability to take greater advantage of their rights. The problem with average happiness and income studies is it overlooks significance of outliers. Some people may be much, much happier with very large incomes even when most people don’t care that much. Restricting those who do care is a significant reduction in their happiness. Combined with the higher incomes of the poor it’s a win win for me.

3) If I had competitive, curious parents and was reasonably sure I would also be competitive and driven I would want to be born in ParetoSuperiorland where I could take more advantage of growth rates. On the other hand if I was not curious, driven, and competative I would probably choose Fairnessland. Either of these situations applies regardless of which class I was born into.

4. Suppose ParetoSuperiorland and Fairnessland are neighbors and allow immigration. Would the poor from ParetoSuperiorland tend to immigrate to Fairnessland or vice versa?

4) Holding that both societies have equal protection of negative rights, and immigration is fairly easy, I think you may see some initial poor immigration into fairness land for a quick bump in income. However, long run I think for each classes the net migration will be towards ParetoSuperiorland.

5. Some philosophers claim that in order for a society to be just, people in that society need to affirm the right theory of justice, or something close enough to it. Suppose that everyone in Fairnessland believes in the right theory of justice, whatever that is, but everyone in ParetoSuperiorland believes in some wrong theory (not a terribly wrong one, but a wrong one nonetheless). What difference does that make in your overall assessment of these two societies?

5) Assuming the wrong theory is reasonably functional and not terribly wrong I don’t think it much matters. I don’t think it would much matter to most people either, except under the specific, and I assume rare, cases where the wrong theory gives much worse outcomes. I am thinking along the lines of Newton’s theory of gravity, for almost all practical applications it works, but it’s also wrong.
http://profile.typepad.com/theotherchuckd TheOtherChuckD

Jason: I read your caveats, but it’s still a poorly formed argument. You let compounding growth do the hard work for you.

Basically, you’ve thrown enough money at ParetoSuperiorland to make the inequality problem in ParetoSuperiorland pale in comparison to the economic stagnation problem of Fairnessland. That stagnation problem occured because … well, because it was convenient to your model.

It’s like debating the merits of football’s 3-4 defense against the 4-3 defense and assuming that one defense always starts deep in its own red zone. In the end, the arguments for the disadvantaged position go down in a hail of touchdowns and field goals.

Assumptions are absolutely necessary to making theoretical arguments, but using them to stack the deck to this degree weakens the whole argument to the point that it can’t be taken seriously.
brian

It may be even more interesting to do the growth rates at 2.5, 3, and 3.5 for poor, middle, rich in ParentoSuperiorland. This should increase the inequality drastically between poor and rich after 100 years while still providing higher absolute income.
Hyena

What about Better-than-fairstein: the country relies on geographic patterns of economic concentration to leverage differing rates of inflation.

For example, the City of Los Angeles, the Bay Area or Manhattan are centers of economic productivity with high wages and high costs of living. By subsidizing the single service of remote access to urban agglomerations–through busing, trains or whatever–the poor would be able to command the highest wages possible while fully benefiting from lower costs of living beyond the metropolitan core.
Jason Brennan

ChuckD:

Yes, of course the stagnation occurred because of my model. Yes, of course compounding growth did the work for me. Yes, of course I threw enough money at ParetoSuperiorland to make the the inequality problem pale at the stagnation problem. That’s the set-up so that we can then ask questions about what it would mean to care about the poor if faced with these conflicts.

I didn’t even make an argument, so it’s not clear I have a poorly formed argument.
http://profile.typepad.com/schaut Rick Schaut

Stuff like this is what gives rise to jokes involving Economists and can-openers.

As a, relatively, simple exercise, add volatility to your model. Suppose, also, that, while ParetoSuperiorLand does have a faster rate of average growth vis-a-vis FairnessLand, it also has a higher level of volatility around that average growth. Is there any realistic notion of volatility that affects all income groups “equally”? Is there no point where one might be willing to sacrifice some level of average growth in order to achieve more stability?

To the extent that this model obscures key empirical issues, it fails to illuminate some of the more significant value differences between BHL’s and liberals.

Or, maybe not. Does a preference for overly-simplistic models reflect on the values of the average bleeding-heart libertarian?
http://profile.typepad.com/zarth Zarathustra Gazvoda

@TheOtherChuckD

I think that Jason’s point was basically to get readers to ask themselves whether they’d prefer to live in an economy that promotes overall growth over equality, or equality over overall growth. His caveats, as I understand them, were there to make the point that his post is primarily a question about values, rather than an argument that inequality really does promote overall growth (which would be difficult if not impossible to prove).
http://profile.typepad.com/theotherchuckd TheOtherChuckD

You made an argument by rigging the rules of your model. The absurd timeframe, positing stagnation and leaving out all other concerns (like income mobility), all serve to lead readers by the hand to a set of conclusions.

It’s a common libertarian strawman: would you rather be poor in a poor but equal country or a rich but unequal country? Would you rather be poor in 2011 or rich in 1911? Arguments like these rest on the built-in assumption that inequality powers growth and safety nets hamper it.

It’s worth referencing an earlier BHL post: would you still be a libertarian if the assumptions you’ve made about the positive economic effects of libertarianism were wrong?

A far more interesting question would involve taking your thumb off of the scale.

Assume ParetoSuperiorland had the same growth rate as Fairnessland, but high inequality. However, citizens had a far better chance of moving from the bottom decile to the top. Fairnessland has greater income inequality but people in the bottom decile rarely if ever become rich.

Any answer would really have to grapple with issues of the kind of society we want and the tradeoffs we make, instead of just encouraging readers to reach for the bigger pie.

I know it’s not the question you’re asking, but at least it can be asked without stacking the deck to such an extent.
http://dmorr.livejournal.com Dave Orr

Jason, I think your challenge in running a hypo like this in blogspace is that it’s going to be a rare reader who reads this and thinks about it as a hypo.

It’s a very common technique in certain blogs to write an analogy or hypothetical with the expectation that it be taken as an allegory. So I would expect most readers here to analyze this as an allegory, one which seems clearly to say that Rawlsian fairness has an overly high cost to it.

If what you’re trying to say is, suppose Rawlsian fairness has a high cost to it: is it worth it to you, and what cost is too high, this is probably not the best way to do it on a public blog.

PS: Facebook/twitter signins appear to be broken.
David Sobel

My goodness people, surely it is ok to sometimes use very simple models that ignore many of life’s complexities to focus on a particular issue.
Corey B.

Most of the commentators here have complaints of the form, “If I change your thought experiment, I get different results.” Others say, “If I assume Brennan is saying something he did not say, then he would be wrong.”
http://profile.typepad.com/zarth Zarathustra Gazvoda

Any answer would really have to grapple with issues of the kind of society we want and the tradeoffs we make, instead of just encouraging readers to reach for the bigger pie.

I know it’s not the question you’re asking, but at least it can be asked without stacking the deck to such an extent.

Now that’s a much better criticism. He definitely stacked the deck there.
http://profile.typepad.com/theotherchuckd TheOtherChuckD

@david

My goodness people, surely it is ok to sometimes use very simple models that ignore many of life’s complexities to focus on a particular issue.

It is, but the argument didn’t ignore life’s complexities, it resolved them in favor of one way of thinking to such an extreme extent that the question, as posed, boils down to “would you rather live in a country that had 2% growth or 4% growth over the past century, if both countries started the century at the same income level?”
http://notesfrombabel.com Tim Kowal

I very much like this post. It provides a nice framework for discussing and evaluating two big divides that separate conservatives/libertarians on the one hand, and liberals on the other: (1) whether to focus on long or short time horizons in deciding political and/or moral questions, and (2) whether to focus on procedural fairness versus substantive fairness. I plan to cite back to this construct in my own blogging. I also enjoy the comments, and look forward to reading more.

Well done!
Don

Let me tell you a story. As a teenager I worked a while at an apple orchard. One day I was helping in the cider press room, working with a bunch of Mexican migrant workers. They were older, knew what they were doing, and could work way harder than me. We made small talk through the one guy who was bilingual. At one point they asked me how much I was making, so I told them. I had assumed they were making much more, since even I knew they were better workers, but when my answer was translated I quickly saw in their faces that I had been mistaken. I can still remember the shame.
Don

Sorry, browser ate my first and last paragraphs. The point of that story is, yes, income inequality per se is a problem.
Dan Kervick

I’m happy to try to play these intuition-pumping games with models. However I would resist that this particular model represents anything close to the social choices we are actually faced with.

Equality in income promotes the creation of a large number of goods, not all of which are easily measurable or estimatable. The common idea that we have to trade aggregate well-being for equality can be based on erroneous methods of measuring aggregate well-being.

And income inequality is often extremely wasteful. It creates manifold, and acceleratingly bad, prisoner dilemma situations in which firms and individuals, in rationally pursuing their own interests in an environment in which thy have little influence over the choices of others, collectively generate an outcome that is much worse overall than the outcome that could have been produced with some cooperation and a commitment to more equality. Firms relate to their workers primarily as workers, or “human resources”. Left to their own devices they will compete with each other by reducing costs – i.e. by firing workers and keeping their wages down. At least, they will do this so long as they are operating in an economy which runs well short of full employment, and it certainly appears on the basis of historical experience that societies that do not make a social commitment to full employment end up generating large pools of unemployed and underemployed workers. However, those workers are also other companies’ customers. So the aggregate outcome is the depression of economic activity, and lower levels of both consumption and savings. I would have hoped that this was a lesson that was learned in the incredibly prosperous postwar period. But sadly, it was not.
http://profile.typepad.com/theotherchuckd TheOtherChuckD

Re: The update

It’s also worth asking how much one’s answer depends upon just what the differences in growth rates are stipulated to be.

Exactly. The number you pull out of a hat will make 99% of the difference, especially if you have a very long time horizon.

It’s also worth asking how the growth will be distributed and why one should assume, absent government interventions, that income of the poorest grows at the same rate as the income of the richest.

If all that extra growth Jason threw at ParetoSuperiorland went to the rich, as it is wont to do in highly unequal societies, the question gets more complicated for our theoretically poor Paretosuperiorlanders.

Since the rich can earn interest while they sleep while the poor rely on labor, assuming equal distribution of gains would probably mean assuming that there is some sort of labor shortage (which requires immigration restrictions) or safety net/redistribution in ParetoSuperiorland.

Note that Fairnessland also has growth at the same rate across all incomes, so it probably has the same interventions, or at least the same policies on inequality.
http://profile.typepad.com/andrewlevine Andrewlevine

I just read the update. The starting baseline still matters a lot to the answer. Developing countries rely on growth more than developed countries do. Once it’s feasible for every (or nearly every) citizen to get food, clean water, education, travel, and some other things that are taken for granted in developed countries, and which are essential to being free to pursue one’s destiny from youth to death, relative wealth becomes more important than further growth if that growth contributes to widening inequalities.
John V

theotherchuckD,

“It’s a common libertarian strawman: would you rather be poor in a poor but equal country or a rich but unequal country? Would you rather be poor in 2011 or rich in 1911? Arguments like these rest on the built-in assumption that inequality powers growth and safety nets hamper it.”

Your assumption of the causation is wrong. The assumption is not that inequality powers growth. The assumption is that faster growth leads to inequality. BUT, as the model shows, that inequality from faster growth is coupled with better absolute conditions for the poor.

Now, you can dispute that that is wrong or not as accurate as Jason is suggesting…if Jason is even stating it as iron-clad. But let’s keep the framework of the writer you are reading in mind and what is, ironically, a strawman of what he is really thinking.

BTW, there’s more than just safety nets involved here. There’s the level of taxation…regardless of what it is used for as well as the amount of freedom of exchange, which is mainly about rules and regulations and not taxes.
John V

theotherchuckD,

“It’s also worth asking how the growth will be distributed and why one should assume, absent government interventions, that income of the poorest grows at the same rate as the income of the richest.”

He made it clear that the growth rates between rich and poor are not equal. I think you should reconsider what Jason’s argument was because you really aren’t addressing what he said. Instead, you’re addressing this matter as you like to argue it.
Paul Gowder

Dave Orr nailed the problem here — most readers in blogland just don’t get the purpose this sort of example serves.

Having only skimmed over the comments, many of which are annoying, I’m not sure if someone else has raised this already. But just in case not:

Can “intergenerational justice” do any work here?
John V

“Once it’s feasible for every (or nearly every) citizen to get food, clean water, education, travel, and some other things that are taken for granted in developed countries, and which are essential to being free to pursue one’s destiny from youth to death, relative wealth becomes more important than further growth if that growth contributes to widening inequalities.”

Why? This is an assertion and not an explanation that can be taken for granted. Besides, why are you trying to decide at what point absolute should not be as important as relative growth. By doing so, you’re sort of admitting that slowing down the improvement rate of the absolute condition of everyone in an effort to reduce inequality is more important than allowing that absolute condition to condition to improve. Why? Why should it be up to someone’s tastes and notions of “what’s right” to slow down the engine so the rich and poor aren’t so far apart. I think YOU are making assumptions about that inequality in a wealthy society that you need to substantiate (and without pointing to problems from bad government interference in the economic sphere.).
http://profile.typepad.com/schaut Rick Schaut

Jason,

“Some readers are having a surprisingly difficult time understanding this post.”

I don’t know. I thought most of the folks objecting understood the post rather well. Are you, perhaps, not understanding the nature of the objections being raised?

Your update still asks me to assume some empirical result that fails to capture what my values are. If this is “a conceptual post meant to discuss values,” have you not failed in our aim if your conceptual construct leaves me with no way to articulate what my values are?
http://profile.typepad.com/theotherchuckd TheOtherChuckD

@johnv

He made it clear that the growth rates between rich and poor are not equal.

Actually, he doesn’t. He assumes the same growth rate for rich, middle and poor. If you look at the charts, you can see that each group’s income increases by the same percentage. Over time, this creates a large divergence in absolute terms, but the percentage increase is lockstep across income levels.
brian

“Once it’s feasible for every (or nearly every) citizen to get food, clean water, education, travel, and some other things that are taken for granted in developed countries, and which are essential to being free to pursue one’s destiny from youth to death, relative wealth becomes more important than further growth if that growth contributes to widening inequalities.”

This strikes me as a near vs. far problem, to borrow from Hanson. Relative wealth is a near concern once basic needs are meet. People in their day to day lives are much more concerned with their neighbors wealth/position/status then their absolute level once basics are taken care of. On the other hand, most people in the U.S. live better then kings a millennia ago. It is easy to loose site of the far mode, but arguable it’s more important.

Not sure if this is what the previous poster was thinking with the inter-generational justice.
John V

theotherchuckD,

Yes, but in the follow-up edit, he mentioned it and THAT’S what I am referring to.

And it wasn’t until I read his follow-up that I even noticed the growth rates were equal. I just assumed they weren’t…and that would have been fine. The point he is making doesn’t change because of it.

Personally, I would preferred he made the growth rates disproportionate in the first place so as to avoid this type of rejoinder that you gave, which, again, actually is totally beside the point he is making.
John V

My one quibble with the OP is that he makes the baseline higher in Fairnessland. I don’t see any realistic reason to assume a better starting absolute position for the poor in a more equal society since the aim at equality can be seen as antithetical to the idea improving the absolute condition of the poor…unless the number are supposed to reflect distribution + income or share of the wealth + distribution or just share of the wealth.

This little tangent I am making is something that is often overlooked since transfers in actual dollars are generally not reflected in income or wealth distributions that people often cite when pointing to inequality.
http://profile.typepad.com/theotherchuckd TheOtherChuckD

I don’t see any realistic reason to assume a better starting absolute position for the poor in a more equal society since the aim at equality can be seen as antithetical to the idea improving the absolute condition of the poor…

Assume that and you’ve assumed away every discussion in this blog.
John

One of the comments raises an interesting possibility. Some of the poor in ParetoSuperiorland move to Fairnessland and then send money back to their families in ParetoSuperiorland.

What stresses does that put on the social cohesion of Fairnessland?
John

I wonder what this variant to the game would result in. Keep everything the same with the exception that in ParetoSuperiorland there’s 5 percent chance each year of finding oneself in a different class?
John V

theotherchuckD,

“Assume that and you’ve assumed away every discussion in this blog. ”

No. I haven’t. The idea that poor are materially or financially better off in any absolute sense under a system with more equality isn’t just an assumption, it’s a leap of faith.
brian

‘No. I haven’t. The idea that poor are materially or financially better off in any absolute sense under a system with more equality isn’t just an assumption, it’s a leap of faith.’

Especially if as stated you assume the government is ‘some form of liberal society, and both equally protect citizens’ political and civil liberties.’

So the inequality is not a function of repression, lack of rights.
Dan Kervick

Well, I’ve been trying to mull over this reformulation of the hypothetical, but I realize I’m hung up on the terms “income level” and “having less”.

What goods are you assuming to be included in a person’s income during a given period of time? Money only? The value of all goods and services delivered to a parson during that time? The value of all goods acquired or achieved by that person during that time?

Anyway, I value equality both instrumentally because of the other individual and social goods it generates, and also more immediately to the extent that it is a measure of social justice. So yes, my first impressions are that the answer to this question depends on the amounts of good things produced or forsworn, and thus there is no answer to the “A or B” question that is entailed by the description of the hypothetical example alone, even holding other things equal.
http://profile.typepad.com/theotherchuckd TheOtherChuckD

The idea that poor are materially or financially better off in any absolute sense under a system with more equality isn’t just an assumption, it’s a leap of faith.

The idea that the poor are materially or financially better off in an absolute sense under a system with more equality makes the following leaps of faith:

- (This is the big one) A more equal society gets more equal at the cost of total economic growth at a 1-to-1 rate or close to it;

- There are no externalities to inequality;

- Inequality itself is not a barrier to class mobility;

- There are no economic benefits to more equality, such as a stronger internal market for products;

- Extreme inequality doesn’t create political pressures that could be far more punitive and damaging than had inequality been reduced over time;

Boring, rote libertarianism, called “vulgar libertarianism” elsewhere on this blog, makes all these assumptions. BHL doesn’t, which is why I like it.
http://profile.typepad.com/theotherchuckd TheOtherChuckD

Should read:

The idea that the poor aren’t materially or financially better off in an absolute sense under a system with more equality makes the following leaps of faith:
http://www.emergentfool.com Kevin Dick

I guess I’ll actually answer the questions you asked instead of arguing with them:

(1) No.

(2) Yes.

(3) I would prefer PS because I think wealth is the biggest contributor to my happiness. I don’t think everyone should have the same preference. I don’t know a way to justly decide between these two options. Ideally, people would have the choice and could vote with their feet.

(4) If I were one of the poor, I would migrate to wherever my income would be the highest. So if were poor, I’d choose FL the first 25 years and PS after that.

(5) I believe in revealed preference. So intentions don’t matter to me.
Dale Dorsey

Yikes. The discussion here seems to have gotten way off track from Jason’s original post, which is an interesting philosophical question.

I pick Pareto-Superiorland! (I’m assuming that $$$s are shorthand for welfare levels.)

If that commits me to rejecting the intrinsic importance of equality, I think I’m OK with that.
Dale Dorsey

Or, I should say, “if that commits me to believing that if equality has intrinsic value, it is lexically dominated by increases in welfare to the poor, then I’m OK with that.”
John V

theotherchuckD,

too many assumptions and I don’t quite understand what you are trying to say. I read a few times and and it doesn’t make any sense.

I make only one assumption in my statement.

A free and open society.
http://profile.typepad.com/theotherchuckd TheOtherChuckD

@John V:

Actually, you make a lot of implicit assumptions when you say “the aim at equality can be seen as antithetical to the idea improving the absolute condition of the poor.”

I listed all the things you had to assume for that statement to be true.
http://profile.typepad.com/chrisbenjamin Chris Benjamin

theotherchuckD I can understand why you wouldn’t like to make a choice between the two worlds, but I believe that choosing whether we’d like to be more like ParetoSuperiorland or Fairnessland is actually a choice we make through government policy. Inequality per say doesn’t cause greater economic growth, but things used to reduce inequality do reduce economic growth. Take for instance progressive taxation – the idea that the richest should face higher marginal tax rates. As people face higher marginal tax rates on the profits they get from investments, they are going to be discouraged from saving and investing, and will hence probably spend more of their income on consumption. Because more income will be spent on consumption and less saved and invested, economic growth will be reduced in the long term. Now as more is invested and more capital accumulated over time, (by capital I mean tools, machinery, knowledge, technology and things that improve the production process) through competition these benefits will be passed onto workers and consumers. Workers will receive greater wages due to their increased productivity (a worker using a digger is more productive than a worker using a spade and so has more bargaining power when it comes to negotiating wages) and benefits will be passed onto consumers in the form of cheaper prices (this due to competition between firms).
John V

theotherchuckD,

But I don’t make those assumptions. I made one and that covers it.

The problem I have with you laundry list of assumptions, if I am starting to get what you were saying a little better now, is that none of the things you listed in their opposite form prevent what I claimed….unless we assume things about government that we simply cannot.

Left-liberals need to be more self-critical about their solutions because expending such energy to make a strict list of conditions for what I said (as if that’s even a reason to not follow through with the ideas I’m putting forth) is useless unless the solution is just as thoroughly vetted and strictly assessed. Imperfect, yet free, societies are far easier to deal with than imperfect government. Society evolves to find solutions to most problems. Government is not so agile and pliable.
http://profile.typepad.com/theotherchuckd TheOtherChuckD

@Chris:

Yes, that’s doctrinaire Libertarianism 101 on marginal tax rates, taken right from a textbook.

Unfortunately, things don’t usually work out that way in the real world (see: Bush tax cuts). It turns out that individuals are idiosynchratic and changes to things like tax rates do not lead to lockstep changes in the behavior of individuals.

The very rich especially are not slaves to swings of a few percent in the top marginal tax rate, as ones’ standard of living changes very little if you take a little off the top. Even when the top marginal income tax rates were 90%, we had strong growth and lots of innovation.

You add the assumption that “Workers will receive greater wages due to their increased productivity.”

Workers’ wages have been stagnant for decades, despite leaps in productivity. It turns out that inequality has a whole string of side effects, not the least of which is the concentration of political power to break unions and cut benefits.
John V

theotherchuckd,

I think the issue of productivity gains vs. wages is somewhat affected by the skills gaps, who has them and how different skill sets relate to different jobs and thus income.

I suspect that if you break down a country as large as the US into sectors and skills and demographics, the parts that form the whole start to show their uniqueness…which gets lost in aggregates.
Kien

The question seems to be a relatively straight forward consideration of inter-generational equity. Should the poor of today give up some wealth today so that the poor of the future (who would likely be wealthier) can be wealthier than they would otherwise be? This question arises in the context of climate change. Most people would say that the poor of today should not have to forego wealth so that the future poor are better off. However, a case could be made that the rich of today ought to forego wealth so that the future poor are better off.
http://profile.typepad.com/wilkinsonwill Will Wilkinson

I think maybe Jason should be hanged by the neck until dead.
Luka Yovetich

I find this to be a thought-provoking post. Perhaps there’s something fundamentally bad about the thought experiment that I’m not seeing. But my guess is that some of those commenting here aren’t used to considering hypothetical scenarios in the way that philosophers normally do.
Mike Kelley

Christ, there are some stupid people reading this blog.
Joel O’Dorisio

Your assumptions are seriously flawed in that the wealth distribution of Pareto superior land are actually much closer to the wealth distribution in Sweden rather that in a libertarian society. the actual wealth distribution of USA is top 20% has 85% of the wealth and the bottom 40% has less than 1%. so both of your models are socialist wealth distribution schemes. Additionally both of your models assume that the distribution of wealth will be fixed 100 years after the start of your model. clearly flawed.

Good luck with your next
Mike Kelley

Joel, he’s not talking about Sweden, the United States, or libertarian societies. He’s asking you to do philosophy and you’re too stupid to do it.
John V

Joel,

It’s a hypothetical to provoke a thought experiment. It’s not some precise model. You’re so busy trying to “debate” that you missed the point of the post.

I was reminded of a term posted by Julian Sanchez that precisely describes what you are doing. It’s the opposite of an “insight”. It’s an “outsight”:

http://www.juliansanchez.com/2007/08/15/outsight/
nemi

Even if I know that I would be one of the poor, I think that I would have chosen the high growth alternative (given that I was born in 2000 – otherwise I would of course choose the more egalitarian).

But even if I was born, as a poor, in the year 2000 – I think that my choice would be irrational. Pretty much every study show that we overestimate the amount of wellbeing that we will arrive from a higher consumption.
http://www.david-welker.com David Welker

The problem with this thought experiment is that it is like one that asks us, would you rather die by being slowly burned to death, or would you rather die by slowly being cut into pieces. There are no good answers.

Basically, neither a highly unequal society nor is a society where productivity is stifled is a desirable scenario. That is why people are hesitant to go along with this thought experiment.

Also, one cannot deny that this experiment is designed (or appears to be designed) to make some sort of argument about the real world. So, I think questioning the relevance and motives behind this hypothetical is a reasonable inquiry.

That said, I will gladly step into the “trap” that is being laid. Yes, there is a limit to the amount of productivity that I am willing to sacrifice in order to increase equality, temporarily going along with the ASSUMPTION (which I believe is wrong) that even the most modest attempts to increase economic equality would have SEVERE consequences for productivity.

Part of what is wrong with this hypothetical, is that it has an assumption that I believe is wrong embedded into it. Hypothetically, I have to choose death by slowly being burned, or slowly cut into pieces. You can understand why this is not an enjoyable hypothetical.
Nathan P.

The trouble liberals have with this is identical to the trouble libertarians have with being asked “if your policies were proven to make everyone poorer, would you still believe in them?”

In fact, that’s exactly the question being asked here. Liberals are being asked how they would react if their policies were proven to make everyone poorer.

Most libertarians will sputter in response to this question, “but my policies would never make everyone poorer!”

Most liberals in this thread have sputtered, “but my policies never make everyone poorer!”

I’m actually tremendously encouraged by this post and its ensuing comments. It makes me happy to have it be so vividly reaffirmed that people are people, and encourages me to think that through understanding our similarities common ground may be possible.
http://blog.hecker.org/ Frank Hecker

I’m coming very late to this discussion, and as a naive layman I’m in great danger of generating an “outsight” (great term BTW). But I’ll venture a comment anyway:

Clearly people’s sense of what constitutes a fair situation is often based on relative and not absolute outcomes, as shown, e.g., by the “ultimatum game”. However clearly absolute outcomes matter to people too, not just in terms of basic sustenance but in terms of the rewards of living in a wealthy society even if you yourself aren’t (relatively) wealthy, e.g., lots of poor people seem to like their iPhones. So I can easily imagine different people having differing preferences as to which society they’d like to live in, based on their own values, their sense of how they themselves would fare in such a society, and so on.

Given that, my preference would be for both societies to exist in parallel, and that the barriers to migration between them be low enough that motivated people are able to live in the society that best matches their own preferences. This seems analogous to what happens in real-life to some degree, e.g., when people choose employers or places to live: They make trade-offs between absolute and relative well-being, and different people choose different trade-offs. Thus I don’t see there being one “right answer” to the thought experiment that’s potentially universally acceptable by everyone.
grberry

If I believed Rawls was right (I don’t), I would have to conclude that Fairnessland is the more just society, because the worst off in it (the worst off in the first year) are better off than the worst off in Paretosuperiorland (the worst off in the first year).

Since I don’t believe Rawls was right, I have no hesitation in choosing Paretosuperiorland. Even if I ended up in the poor of the first year, I could be confident that my grandchildren would be better off in Paretosuperiorland than in Fairnessland.

And thus:
1) No, because I have an interest in the wellbeing of my descendants, I would not support transforming Paretosuperiorland to Fairness land at any time when I believed that the transformation would adversely affect future growth rates.

2) I believe that absolute well-being is more important than relative well-being, so yes, Paretosuperiorland is superior in the long run. There will come a day when it is superior for every individual, and even sooner there was a day when it was for the median individual.

3) I would choose Paretosuperiorland, because I would expect to be better off there.

I don’t think everyone else should have the same preference. Rawls made the false assumption that under the veil of ignorance everyone is maximally risk adverse, despite the empirical evidence to the contrary. If there is in fact anyone who is maximally risk adverse, they should prefer Fairnessland. (Nobody who has ever made or accepted a bet qualifies as maximally risk adverse though.) It is a matter of preference, not justice.

Those most affected by a decision between Paretosuperiorland and Fairnessland are the young and unborn. A just decision making process would require them to have the greatest say in the decision. There is no mechanism available for giving the unborn a vote or voice. Nor for giving the very young (say below age 4) the same. The elderly should have very limited say. I don’t see any practical way to implement a just decision process.

4) This is ultimately an empirical question. Arguably the closest we have to an empirical test is the U.S. and Canada. I don’t know the historical results. Speculating, I think the poor might migrate to Fairnessland until a few years before the breakeven point, then the direction of migration by the poor would reverse. The rich and middle class would tend to start migrating to Paretosuperiorland from the very beginning. Unless the poor greatly outnumber the rich, I think the net migration would be toward Paretosuperiorland from the very beginning.

5) I would disagree with those who think the right theory is required. Such philosophers are overestimating the importance of philosophy, and thus of themselves. Hardly an unexpected error, but an error nonetheless. Their error would make no difference in my assessment of these or any other societies.