The differential marginal utility of money

by on May 20, 2015 at 1:08 am in Economics, Philosophy, Political Science, Uncategorized | Permalink

1. You cannot build and sustain a polity on the idea of redistributing wealth to take advantage of differences in the marginal utility of money across varying wealth classes.

2. The ideas you can sustain a polity around often contradict the notion of socially arbitraging MU differences to try to boost total utility.

3. The MU argument, in isolation, is therefore rarely compelling.  Furthermore its “naive” invocation is often a sign of underlying weakness in the policy case someone is trying to make.  The proposed policy may simply be too at odds with otherwise useful social values.

4. This is related to why parties from “the traditional Left” so often lose elections, including in a relatively statist Europe.

5. That all said, sometimes we should in fact take advantage of MU differences in marginal increments of wealth and use them to drive policy.

6. Figuring out how to deal with this tension — ignoring MU differences, or pursuing them — is a central task of political philosophy.

7. The selective invocation of the differential MU argument — or the case against it — will make it difficult to improve your arguments over time; arguably it is a sign of intellectual superficiality.

1 Adrian Ratnapala May 20, 2015 at 1:25 am

I’m confused. I translated #1 & #2 to mean “genuinely distributing wealth to those who will benefit most (as opposed to those with clout) is politically infeasible”. Yet #3 seems to say those policies are genuinely bad, not that they are unrealistic.

And separately:

4. This is related to why parties from “the traditional Left” so often lose elections, including in a relatively statist Europe.

I think parties of the traditional left lose elections about half the time. As do parties of the traditional right.

2 RoyL May 20, 2015 at 4:15 am

I think a better phrasing is.

This ecxplains why when party’s from the left win they only tinker on the margins and genuine leftists quickly see them as just as bad as the right party was. Of course they also decide the right party has moved further right because it opposes the left party’s rightist policies from the right, so they continue to support the left party while trying to drive it further left. As soon as they appear to be succeeding, the right party wins the next election.

This goes on until all the marginal leftist improvements, which are never coordinated or rationally integrated, bankrupt the state, and then their is a reaction which often, but not always, involves a dramatic policy shift to the right of all parties, which is denied by everyone.

3 magilson May 20, 2015 at 11:15 am

Maybe. But there must be a historical start to this “truth”. Because If anyone is arguing the United States or Europe are less left/statist/progressive/whatever than *they* used to be I think that’s wrong. And that’s the implication of your translation, is it not?

4 Floccina May 20, 2015 at 4:57 pm


5 Luis Pedro Coelho May 20, 2015 at 2:24 am

The classical MU argument has, in my view, been moderated by the findings of behavioral economics, namely loss-aversion. Taking from the higher-incomes to give it to the lower incomes may be negative utility as the higher incomes are valuing their loss at an exaggerated rate (it’s a loss), while the lower income recipients under value it.

Many on the Left are too quick to grab on to the findings of behavioral economics as a critique of neoclassical economics, but while they often do point away from simplistic free-market views, they do not necessarily point towards left-wing solutions. They are just as likely to point to non-market conservative views.

For example, isn’t it another consequence of the asymmetry of the utility function with respect to the status quo (loss aversion) that social mobility destroys utility? I mean, if the tide is lifting all boats, then you can argue that it’s still better for everyone (the libertarian view), but if your utility function is heavily rank-based (a standard left-wing view) and you accept loss-aversion from the behavioral literature, then social mobility is suspect from an utility point-of-view.

This sounds shockingly old-school conservative when we discuss our own societies (“why should the children of the poor compete with my kids for a place in a good university? they have lower expectations, after all, State U is a step up for them. My kids, on the other hand, would be crushed if they had to go to their safety school”), but is quite acceptable when discussing international inequalities (“it doesn’t morally matter that people in Mexico have much less material wealth, their society has lower expectations”).

6 RoyL May 20, 2015 at 4:18 am

+1 and Amen

7 AB May 20, 2015 at 6:31 am

The traditional left IS (and always has been) suspicious of the social mobility, as it is antithetical to class consciousness.

8 Pshrnk May 20, 2015 at 7:48 am

And the left always overpromises, thus raising expectations too high, leaving lowere SES folks feeling they have lost something when they may have actually gained.

9 Boonton May 20, 2015 at 9:26 am

The classical MU argument has, in my view, been moderated by the findings of behavioral economics, namely loss-aversion. Taking from the higher-incomes to give it to the lower incomes may be negative utility as the higher incomes are valuing their loss at an exaggerated rate (it’s a loss), while the lower income recipients under value it.

Consider policy A:
Tax upper income people about $5 an hour. Give the ‘working poor’ a $5 per hour raise via the Earned Income Tax Credit. Super policy from a classical right wing economics point of view. Better to ‘top off’ income at the bottom rather than try min. wages increases which just freeze the unskilled out of the job market entirelly.

Policy B:
Raise the min. wage $5 an hour. As a result corporate earnings and upper executive bonues drop by roughly the cost of the increase.

It seems like the ‘loss aversion’ would make policy B preferred. The ‘1%’ would rather have a few thousand dollars less in income but no tax increase than a few thousand more in income but a tax increase even though an economy filled with Mr. Spocks would see these as the same thing.

Perhaps this is why income support programs like the Earned Income Tax Credit seem to encounter a lot more resistance than they deserve since they supposedly work to address the concerns of both liberals and conservatives (helping the poor while preserving incentives to work and avoiding gov’t intereference in labor markets).

10 sort_of_knowledgeable May 20, 2015 at 10:50 am

The EITC is seen by my left leaning friends as corporate subsidy allowing the corporations to pays less than a living wage when they should be paying more.

11 Luis Pedro Coelho May 22, 2015 at 6:50 am

> Policy B: Raise the min. wage $5 an hour. As a result corporate
> earnings and upper executive bonues drop by roughly the cost of the
> increase.

Why would this follow from this policy? You have a very heroic set of assumptions here.

12 Slocum May 20, 2015 at 11:36 am

“Policy B:Raise the min. wage $5 an hour. As a result corporate earnings and upper executive bonues drop by roughly the cost of the increase.”

But that’s not what happens. Half of minimum-wage workers are employed by businesses with < 100 employees. These are small, low-margin businesses that have no 'upper-level executives'. And when you look a list of the the large employers who do employ low-wage workers, the list is heavily dominated by discount retail and fast-food — which are already low margin businesses. It really doesn't matter if a given McDonald's is a franchise or company-owned store, if a big minimum wage hike renders it unprofitable, it won't stay in business. And the idea that McDonald's management are paid enough that you could marginally reduce executive compensation and thereby provide a $5/hr pay increase (or any significant increase) to all employees is a silly fantasy. The only place McDonald's could get that kind of money is from customers by increasing prices. But customers may not be willing to pay. After all, nobody *has* to eat fast food — they always have the option of cooking.

13 Floccina May 20, 2015 at 5:18 pm

And once a few of the small restaurants go out of business McDonald’s can raise prices a bit and those executives can get a big bonus.
Although many people in the small immigrant run business (who are gaining at the expense of McDonald’s) already make less than minimum wage so to them minimum wage does not matter at all (though they are occasionally busted when they hire non-family. One was busted in my town a few years ago).

14 Boonton May 21, 2015 at 12:00 pm

And if instead of a $5/hr increase society opted instead for additional income subsidies then you are talking about increased taxes. If taxes go up $5 might not some people respond by hitting McDonald’s a bit less? In that case how many will close due to reduced demand?

Two can play at the ‘just so story’ game.

15 John May 21, 2015 at 2:09 pm

I’m not sure that the first paragraph is a strong point. Just because there are asymmetries in marginal valuations based on direction of flow I don’t see that it invvlaidaets the usual assumption of diminishing marginal utilty. As I understand you, your point about behaviour ecnoomics is that if we start with someone having $100 and take away $1 the MU will mu1, For the same person, can also start them with $99 and then give them another $, bringing them to the $100. In this case we would call that mu2. The claim then is that mu1 >= mu2 (which I can undstand on one level but then start sratching my head when I ask what are the total utilities assocated with the $100 for the person — or is that even a question we can ask in the behavioral approach?)

If I’ve got that right, the implication is that (assuming that at some gross level we can make the interpersonal utility comparisons) that the behavioral insight does is shift the margin for where the mus for the $ transfered hit the “equalibrium” point. Clearly that still leaves the potenial the some rich people will value those $s more than any of the recipients but not that all rich people will do so.

BTW, perhaps it’s just how the point is expressed but are you really saying that the rich over value the loss and the poor under value the gain from an internal valuation point within each of their utility “functions” or some other valuation measure? That would be something different from what I take things as above.

16 Luis Pedro Coelho May 22, 2015 at 6:38 am

Diminishing marginal utility means that if we have two people A, who has 100 and B who has 50, and we transfer 25 from A to give to B, then total utility has gone up:

util(100) + util(50) < util(75) + util(75). So, transfers increase utility. QED (assuming no dynamic effects, which is where the discussion needs to move to next).

Behavioralists point out, however, that A will feel the loss of 25 more than just "util(100) – util(75)" and B will feel the gain of 25 as less than "util(75) – util(50)". Thus, there is no single 'util' function, it depends on the status quo and even in the absence of dynamic effects, transfers are not so clearly utility increasing.

17 mulp May 20, 2015 at 2:36 am

If wealth is not redistributed by people with money paying those without money to labor to produce stuff, especially productive capital assets for the people with lots of money, then GDP will be stagnant or decline.

I am using redistribute as Obama used “spread it around” talking to Joe the Plumber about people spending money to put people to work.

If no one pays millions of people to work, those millions of people will never spend any money, and if profits are not spent 100% to pay workers, the GDP will decline because workers will have less income than the total prices of all produced, so less will be produced. You don’t produce stuff to throw it away, or bury unless you are an emperor building an army for your afterlife.

If you are a monopolist reaping huge monopoly profits, as it currently going on rampantly, the last thing you want is paying your profits to labor to buy more productive capital assets because those assets will increase the supply in the face of stagnant demand, forcing price and profit cuts. Worse is a speculator building productive assets with money borrowed from you to compete with you, taking your profits when interest rates are so low that servicing the debt is easy by accepting slightly lower than market returns on capital.

That is the problem Apple faces. Google is doing everything it can to spur investments in production capacity for advanced phones running Android using the google cloud apps. Apple is trying to maintain its monopoly by spending profits buying competitors at the margin to prevent anyone to compete. It makes sense for Apple to buy companies and lay off workers.

WalMart has for years sought to lower costs by lower wages. Now they blame the stagnant and declining incomes of its customers from lower wages and lower welfare benefits.

To your point, the MU of profits to the billionaire is zero if paying workers if profits are to be maintained so they will be kept away from paying labor to produce anything.

On the other hand, a poor person will find every dollar of income has maximum utility paying for labor, and less MU paying to burn capital assets which goes to monopoly profits at fossil fuel companies that see paying labor as low MU.

I am a big fan of tax dodges and high tax rates. I find in my own behavior that I make stupid decisions to not pay taxes. Thus, a carbon tax that was going to go up every year indefinitely would lead to tax dodging investment in labor beyond all reason. A carbon tax would in hundreds of billions of productive capital assets being idled, unwillingly by the owners, by those seeing great profit in tax dodging with wind and solar and batteries and replacing burning capital with paying labor to build electric cars and heat pumps.

A high carbon tax in the certain future would redistribute wealth from capital to labor to capital with virtually no tax revenue raised. But today, the wealth sits as capital doing nothing.

18 Ricardo May 20, 2015 at 2:45 pm

“But today, the wealth sits as capital doing nothing”

So you mean it’s held as gold coins in a vault? Or perhaps as factory equipment that isn’t being used?

19 Alan May 20, 2015 at 2:42 am

Is point 1 an axiom, an expression of a hope or a statement of an observed fact? I really don’t know. If it is a statement of fact (not of an ideology), how might it be demonstrated, other than by the usual political and economic process of claiming that almost any observation supports what you already believed anyway?

20 prior_approval May 20, 2015 at 3:23 am

You mean that 1 is just self-recommending mood affiliation when viewed through a Straussian lens?

21 Alan May 20, 2015 at 7:23 am

Exactly. I’d find it much more plausible if I heard it from a Pakistani cook at a food stall in a back alley in Somalia.

22 Topper Harley May 20, 2015 at 9:18 am

Two questions: Are there any attractive women at the food stall? And why this instead of hearing it from your Pakistani cab driver?

23 Dan May 20, 2015 at 9:03 am

I’m confused as well. I think he means that, historically, a modern industrial democracy must basically be capitalist if it is to survive, as can be seen by the end of communism. Once you’ve got a working capitalist democracy, you can do redistribution, but it’ll always be in tension with the underlying capitalist basis of the state. Speaking as a pro-redistribution guy, I don’t even disagree with his point — I just don’t think it gets you very far in the argument against redistribution.

24 derek May 20, 2015 at 10:18 am

It would be worth looking at prosperous societies that are forced to live a bit closer to the edge than the US. Canada, Sweden, the northern democracies who have stable economies and a social welfare system that works reasonably well.

Since the 80’s the challenge of government policy has been to sustain the social programs as best as possible by having economic growth. These marginal utility decisions mean whether a project or industry exists or not to be taxed. If a tax rate or fee schedule is wrong, there is no taxes or fees to be had because it doesn’t happen. If it is right then economic growth happens, taxes are collected and some people on a waiting list for surgery get it done.

Welfare rates are shockingly low. Minimum wage is higher. If doctors aren’t paid enough, or nurses, there aren’t poorer doctors or nurses, there aren’t any, or enough. Bureaucrats are cheap and expendable. Unemployment insurance is hard to get and of short term. The hard poor due to mental illness or some kind of dysfunction are targeted in ways to minimize cost.

There is a natural experiment about to happen. Alberta just voted in an NDP government. The marginal utility constraints were low for a long time due to the flood of money from the resource sector, and we will see how a socialist government and a tight commodity market get along. The opposite happened in Saskatchewan a few years ago, another natural experiment that is enlightening and rather depressing for the left.

25 Bernard Yomtov May 20, 2015 at 9:54 am

Excellent question.

26 Sam May 20, 2015 at 2:53 am

What about marginal tax reforms and other kinds of reforms predicated on stochastic dominance theorems? Higher orders of SD incorporate stronger ethical standards, with pigou-dalton and transer sensititive consistent reforms in a sense encapulating the standard ethical intuition behind differential MU of income (or consumption). These seem to be extremely useful theorems for identifiying which reforms will be unambiguously pro-poor.

That said, I agree that, at least rhetorically, framing redistribution this way makes it inherently contentious.

27 Millian May 20, 2015 at 5:50 am

“sustain a polity” is the new “mood affiliation” or “ZMP worker”. It’s an unexplained meaningless MR phrase that may well be a strawman.

28 meets May 20, 2015 at 10:04 am

It’s very real for the Labour Party.

29 derek May 20, 2015 at 10:29 am

You are wrong. A real situation developed a few years ago here. With aids there were needle exchange programs set up with funding available for groups to distribute them with the goal of minimizing the spread of disease. A worthy intention. At the same time people were facing very long wait times for medical care, having to travel distances for obstetrics, waiting times for diagnostic procedures measured in years.

If I wanted a clean needle to shot up something, a phone call would have one delivered in 15 minutes. If I had discomfort or even more serious symptoms it was four months before a colonoscopy. And of my appointment at the gp was early enough I would have to step over addict sleeping on the sidewalk.

Guess what policies changed. We now have an obstetrics service locally and the free delivery of needles isn’t available.

30 Ricardo May 20, 2015 at 2:54 pm

A sustainable polity is an enforced social ordering that is compatible with human nature.

Mood affiliation is the tendency to align yourself with arguments not for their own sake, but because you identify with people who express those arguments.

Zero marginal productivity workers are those who contribute no net value (i.e., their value added does not exceed their compensation).

31 Bernard Yomtov May 20, 2015 at 8:03 pm

A sustainable polity is an enforced social ordering that is compatible with human nature.

Of which there have been very many, of very many different types, throughout history. It may be that some policies are inconsistent with a sustainable polity, but given that history, surely the burden of proof is on the person making th claim.

Mood affiliation is the tendency to align yourself with arguments not for their own sake, but because you identify with people who express those arguments.

As used on this blog it seems more to mean simple partisanship. That is, it seems to be to be used to describe a position held because one’s political allies – not always people you want otherwise to identify with – hold them.

Zero marginal productivity workers are those who contribute no net value (i.e., their value added does not exceed their compensation).

So in a simple textbook example, marginal productivity declines as you add workers beyond a certain number. Say you get to zero. Which of the N workers is the ZMP worker? Does the concept make sense as applied to individuals?

32 rayward May 20, 2015 at 6:44 am

Inequality is an issue that must be addressed if it creates an economic problem; if not, then social norms, or pressure, should be the only mitigating force. Is inequality an economic issue? Piketty believes inequality sustains and builds on itself, but that alone doesn’t make inequality an economic issue; a social issue for sure, but not necessarily an economic issue. On the other hand, if inequality inhibits economic growth, including investment in productive capital, or if inequality contributes to financial and economic instability, then it should be addressed. Discussions about marginal utility are a diversion. There’s lots of theoretical and historical evidence that inequality, at a high level, creates an economic problem, that it does inhibit economic growth, that it does contribute to financial and economic instability. In his interview in the NYT last year about Piketty’s book, Cowen said as much when he argued that Piketty’s thesis (r>g) must be wrong because the rate of return on productive capital has been falling (for the same period as inequality has been rising). That’s my parenthetical, but if we ignore history, we are doomed to repeat it. My view, repeated often, is that markets will correct excesses, including excessive inequality. Intervention (by governments and central banks) can mitigate the negative consequences of the correction, or they can prevent markets from correcting the excesses. In the latter case, we are doomed to repeat history. I suspect that many readers of this blog are aware that economists associated with the Austrian School share my view that markets self-correct excesses, and that they would prefer for governments and central banks to allow markets to do their (efficient) work. I suspect that some (not many) readers of this blog may not appreciate the consequences of the market-driven correction. I prefer intervention, but only to mitigate the negative consequences of the correction; I oppose intervention that prevents markets from correcting excesses. Better still, I prefer a preemptive strike that mitigates the excesses before markets do it with a blunt instrument. That’s my marginal utility argument.

33 Gopchik May 20, 2015 at 11:05 am

If you stopped posting I could stop scrolling past a wall of ignored text.

34 Robert Wiblin May 20, 2015 at 6:56 am

“You cannot build and sustain a polity on the idea of redistributing wealth to take advantage of differences in the marginal utility of money across varying wealth classes.”


35 The Anti-Gnostic May 20, 2015 at 7:20 am

Because nobody likes government taking their money to spend on strangers who are encouraged to view these transfer payments as entitlements and the net payors as arbitrary beneficiaries of some unjust, unearned ‘privilege.’

36 Robert Wiblin May 20, 2015 at 8:02 am

Nobody likes this? No way – lots of people like this, not least the beneficiaries, and it’s one of the main justifications for the welfare state in rich countries.

37 The Anti-Gnostic May 20, 2015 at 8:20 am

Then the binding proposition is “having enough agreeable net payors,” not redistributive justice, but perhaps we agree.

38 Dan May 20, 2015 at 9:06 am

Well, the Indians sure didn’t like it, but the white settlers loved it!

39 Boonton May 20, 2015 at 9:32 am

That doesn’t answer the question. The issue isn’t can we craft a policy that makes some people better off while not making anyone else worse off. That would be a Pareto optimal move which no sane person would object too. The issue is differeing utilities.

If Donald Trump could be taxed $1 more and save 10,000 people, clearly the fact is Trump would be $1 worse off yet everyone would agree that trivial harm would be far outweighed by the utility of 10,000 people being saved. In other words, the combined ‘likes’ of 10,000 saved people would be greater than Trump’s dislike at losing a dollar.

So of course people won’t like the gov’t taking their money but Tyler seems to be saying it is impossible to craft a policy that exploits the fact that many people would like getting an extra dollar more than some people would dislike losing a dollar. I’m not sure how he proves that.

40 derek May 20, 2015 at 10:32 am

But what if Trump, when taxed that amount decides that it is no longer worth taking the risk on a project because of diminished returns?

The pairing of labor over the last three decades is exactly this happening.

41 Boonton May 20, 2015 at 10:40 am

That would seem to contradict the idea that Trump’s utility on that $1 is very low. If the $1 wasn’t worth that much to him, why would he be expending a great amount of effort to earn it? If he isn’t extending a great amount of effort, why would the tax cause him to dramatically alter his behavior?

42 sort_of_knowledgeable May 20, 2015 at 11:26 am

“If he isn’t extending a great amount of effort, why would the tax cause him to dramatically alter his behavior?

So it seems like we have a problem measuring utility. Utility ends up being measured by hard one fights to avoid taxes or gains benefits.

43 Boonton May 20, 2015 at 2:49 pm

Hard to measure in real life but not in theory. How much someone values an extra dollar can be seen by what they are willing to do to get one. If Trump pulled out his wallet and 4 quarters dropped out of his pocket and fell into a storm grate, would he get on his knees and try to fish them out? What if it was a $100 bill? What about a person who gets $175 a week who likewise drops a $100 bill?

44 T. Shaw May 20, 2015 at 8:15 am

“. . . improve your arguments over time . . .” Translation: acquire more bullets.

JOHN C. WRIGHT at Instapundit: “Leftism is politicized envy.”

“You didn’t build that.”

“Society’s lottery winners.”

Talleyrand, a consummate politician, said, “Governing has never been anything other than postponing by a thousand subterfuges the moment when the mob will hang you from the nearest lamp-post, and every act of government is nothing but a way of not losing control of the people.”

45 Boonton May 20, 2015 at 9:47 am

“You didn’t build that.”

“Society’s lottery winners.”

Are these really statements of envy or simply facts?

46 Lord May 20, 2015 at 11:47 am

Because one already exists with the goal to redistribute upwards.

47 Further or Alternatively May 20, 2015 at 7:18 am

“You cannot build and sustain a polity on the idea of redistributing wealth to take advantage of differences in the marginal utility of money across varying wealth classes.” I suppose this is an observation that polities have (always and everywhere?) been built and sustained on other grounds, such as ties of blood, language, geographical isolation or matters of that kind. But surely a polity can be built, at least in part, on shared ideals having an important role in a nation’s history and sustained by continuing to venerate those ideals. Is it impossible that a group of Englishmen could have started a polity (let’s call it the United States of Utopia) with a stirring 18th century declaration that began: “We hold these truths to be self-evident, that all men are created equal, that they are endowed by their Creator with certain unalienable Powers and Duties, that among these are the Periodic Redistribution of Wealth and of the Fruits of Labour, Toil and Industry among all men, so as to maintain and preserve the divinely ordained state of Equality”? And is it impossible that those Englishmen, having fought and won a war against some other Englishmen who had a King and no redistribution of wealth, could have written a constitution with similar effect to that declaration and continued to venerate and adhere to it for a few centuries afterwards? True, that is not quite the same as proposition #1 – it would be a polity built on a shared history and some important documents, rather than built on the principle of redistributing wealth to take advantage of differing marginal utilities – but it might end up looking quite similar.

48 The Anti-Gnostic May 20, 2015 at 7:34 am

And is it impossible that those Englishmen, having fought and won a war against some other Englishmen…

There’s your prior. The US is a propositional nation, but it was a specific ethno-cultural group who dreamed up the proposition and fought to defend it. Of course, the founding propositions last only as long as the men who wrote the nation’s charter are still around to explain what they meant.

Over time, the proposition can mold the people. Anglo-Canadians and Anglo-Americans differ in temperament, to my non-scientific observation. My wife, who is from Canada, agrees.

The US is a radical experiment, and seems to be maturing into an aggregation of nations.

49 Millian May 20, 2015 at 7:35 am

Sustaining a polity has nothing to do with morality, or correctness, or even long-term sustainability.

Certain European polities of the early-1930s to mid-1940s were sustained on very awful ideas.

50 RJHJR May 20, 2015 at 9:05 am

“You cannot build and sustain a polity on the idea of redistributing wealth to take advantage of differences in the marginal utility of money across varying wealth classes.”

I agree with the previous comments that question this. Building and sustaining a polity requires a mix of policies. The polity itself is created partly by the involvement of various interest groups. Interest groups and their interests change over time.

Norway has a lot of income redistribution. It also has an ethnically homogeneous population. Criticism of income distribution tends to be centered on immigration. The situation in Sweden, Denmark, and Holland is similar.

The U.S. has always been an ethnic mix. Rather than try to forge a consensus on redistribution, we’ve taken a more every-person-for-themselves approach. Perhaps it would more accurate to say that it is harder to include lots of redistribution in your policy mix if you have large groups of people who see each other as different.

51 Cole May 20, 2015 at 11:07 am

The Scandinavian countries reinforce Tyler’s argument unfortunately. All of their social programs are tied up in fairness and a common national identity. It is embarrassing to all Swedes if any Swedish child grows up without access to health or education. Therefore, they provide services to all of their citizens. They don’t have social programs that target the poor–they target everyone equally. If you speak with the people in these countries, they don’t view what they do as particularly redistributional or based on a notion that the marginal value of a dollar is higher for the poor.

52 RJHJR May 20, 2015 at 3:52 pm

“It is embarrassing to all Swedes if any Swedish child grows up without access to health or education.”

Then the Swedes are aware that they are redistributing resources to people who wouldn’t otherwise get them.

“All of their social programs are tied up in fairness and a common national identity.”

That was my point. They redistribute because they think it’s fair and because they think other Norwegians/Swedes/etc. are like themselves. (I studied Norwegian in Norway and I had a Norwegian girlfriend for many years. I’ve heard quite a bit about the superiority of the Norwegian safety net and how it prevents people from falling into poverty.)

Redistribution makes no sense unless you believe that the education or health care going to a person that otherwise wouldn’t get it is of greater value to that person than the value that someone else places on the tax they pay to cover the cost of the first person’s education or health care. I’ve never heard a Scandinavian talk about the marginal utility of income, but they clearly believe that redistribution increases welfare.

It’s possible that Cowen was referring only to arguments that explicitly depend on MU, and not to redistribution arguments that rely on it implicitly. If so, then his claim is unproven and trivial. I believe he was trying to make a more serious point about redistribution policies that rely on MU at some level, even if it’s not explicit. (See his claim about traditional parties of the left, which rarely explicitly mention MU.) In that case, the palatability of redistribution seems to depend on ethnic homogeneity and possibly tax structure. (Countries with lots of redistribution also tend to rely on regressive VAT taxes.)

53 John May 21, 2015 at 2:14 pm

But is is correct to call what the these countries are doing “redistribution” when they see the situation as all already having the access right? From that point of view, when some are doing without while other enjoy there is something of a redistribution from those who are going wihtout to those enjoying more than otherwise.

54 Ben J May 20, 2015 at 9:21 am


I love MR, but so many of your posts are pointlessly vague.

This blog post consists of you stating your position, with literally no justification for any of the many points you make. It reads like a bad elevator pitch of a very long argument against redistribution.

Who is your audience? Why are you writing this?

55 meets May 20, 2015 at 10:08 am

Speak for yourself. Some of us enjoy this as much as the other stuff.

It’s more philosophy than economics. You can’t figure everything out with just economics graphs.

56 Boonton May 20, 2015 at 10:16 am

I’m not sure it is an argument against redistribution since point #5 says

“5. That all said, sometimes we should in fact take advantage of MU differences in marginal increments of wealth and use them to drive policy.”

That sounds like he is trying to say you cannot build a policy on wealth taxes on top and redistribution on the bottom on the grounds that a rich person losing $100 experiences less emotional distress than the emotional joy ten poor people get from receiving an additional $10. He doesn’t quite explain why can’t build such a policy on this. Does he means the electorate just won’t go for it? Does he mean there is some economic mechanism that steps in and short circuits your attempt to transfer a $1 from a person who has low utility for money to one with high?

Then #5 seems to be saying you can and often should build such a policy around “MU differences in marginal increments of wealth”. That is while you can’t take from what people already have, you can and should take some of the increase in what they have….in other words instead of wealth you tax income since income (minus consumption) would be a ‘marginal increment in your wealth’.

I’m not sure then who he is arguing with since the US has never been very big on wealth taxes and I’m not sure Europe has either. The only exception might be estate taxes but they actually do seem rather sensible to me and hardly deadly from an electorial perspective.

57 Urso May 20, 2015 at 10:57 am

It’s Socratic.

58 Peter May 20, 2015 at 12:34 pm

This. I can’t believe people eat this shit up every day. This could have been an interesting post if it wasn’t, you know, unsubstantiated/unelaborated on, and generally horrible

59 Ricardo May 20, 2015 at 2:57 pm

De gustibus non est disputandum.

60 sp6r=underrated May 20, 2015 at 6:55 pm

Great point. Writers who operate in oppressive regimes for survival have to be cryptic in their writings. Tyler is a tenured professor at a public university. He has a guaranteed job for life. There is no need for him to write in this style. I think the most reasonable hypothesis for why he writes like this is that many of his points are actually highly debatable. By not being clear he can be avoid pinned down to any position. Doing so prevents falsification.

61 John Liselotte May 20, 2015 at 10:13 am

Tyler do some more bloggingheads

62 Ano May 20, 2015 at 10:24 am

I would value some links or longer blog posts in support of items 1 and 2. They are doing a lot of the work in this line of argument, and they’re not obvious to me.

63 Lawrence H. White May 20, 2015 at 10:50 am

In a nutshell: It’s smarter to be a long-run Benthamite, taking into account the negative feedback effects from redistributive policies, than a naive short-run Benthamite.

64 Lord May 20, 2015 at 11:44 am

Isn’t this exactly what the right is based on, that the wealthy have greater marginal utility for money so anything we can do to direct more money to them will increase utility, thus upward redistribution is most productive?

65 PD May 20, 2015 at 12:08 pm

Let’s suppose a hedge fund mogul installs a solid silver toilet in his mansion. A rival mogul then puts a solid gold one in his. Meanwhile 50,000 people in India have no toilet at all. The cost of the gold toilet would cover 50,000 toilets in India.

Leaving aside employment for gold miners, etc., why is this the best of all possible worlds and something it would be folly to try and change? Is this what Tyler believes?

66 Urso May 20, 2015 at 2:41 pm

It is clearly not the best of all possible worlds. However, the question is a little more complex – where’s the guarantee that the system that you design to replace it will be better? Most decisions aren’t as simple & obvious as the 1 toilet : 50,000 toilet hypothet.

67 Ricardo May 20, 2015 at 3:00 pm

Let’s suppose you buy a nice steak for $15. Meanwhile, someone in India has no food at all. You could buy a $5 burger instead, and feed the Indian for a week.

Is this what you do? In what ways do you and the hedge fund manager differ?

68 Michael Foody May 20, 2015 at 3:24 pm

Oh oh, I know this one! We differ by SEVERAL ORDERS OF MAGNITUDE! Ha ha, was that a serious question?

69 Urso May 20, 2015 at 6:19 pm

Yeah but the other way; there a lot more $15 steaks than there are solid gold toilets. The first hypothet is too ridiculous; you want to pass a law banning solid gold toilets, knock yourself out but you’re not getting anywhere. The steak/hamburger is much more realistic, which is why it’s much more difficult.

70 Michael Foody May 20, 2015 at 1:35 pm

LOL, so what if you can’t build a polity based on a particular argument for redistribution? You also can’t build and sustain a polity based on a belief in deadweight loses from taxation. It’s almost as if you don’t “build or sustain a polity” based on the particulars of arguments for or against redistribution!

So from your asserted rather than argued 1 and 2, you get to 3. Let me ask you this: what argument In Isolation is compelling and not naive?

If I identified as a libertarian I would also be dismissive of particularly powerful counter arguments against the very method of inquiry that I use for evidence of my ideology’s superiority. I might even resort of ad hominem against people who make an argument that I cannot counter.

Seriously though, this sort of ‘lack seriousness irrespective of accuracy’ argument totally applies to deadweight losses and every manner of neoliberal policy prescription! The thing is deadweight loses from taxation, prices floors and ceilings, tariffs etc are real and a big deal. Diminishing marginal utility of wealth is also real and a big deal. These things exist in tension with one another despite their inability to build and sustain a polity. 😛

71 msreekan May 20, 2015 at 6:06 pm

This is how I understood the idea — seems there is considerable difficulty in aligning political solutions with individual marginal utility. Left advocates an outright one-size fits all like approach which tends to ignore the whole concept while the other end of the spectrum specifies that it’s implausible to reconcile any end-specific policy prescription with differing marginal utility. Also any effective public policy should be able to account for plausible value enhancing individual activities like social arbitraging and prevalent norms before advocating any specific aggregate ends, otherwise there is little gain with regards to overall social utility.

72 Los Ranchos May 20, 2015 at 8:58 pm

There is a subtle difference between the utility shifts of the individuals and the utility shift of society when implementing redistribution policies. If it’s just moving consumption around it may be difficult to gain support or claim real societal advantages. If it’s investing in education or health care it may find more favor and you may be able to build a polity around that. I suspect Tyler is talking about trying to build a polity based on redistributing consumption – that’s tougher – more like supporting equality of outcomes than equality of opportunity.

Comments on this entry are closed.

Previous post:

Next post: