• Why, you just want to redistribute income!

    From time to time I run into a charge that I or someone supporting some seemingly non-market policy or other “just want to redistribute income (or wealth).” Something like that is often hurled as if “redistribution” were a glob of tar that soils the opponent’s entire position. Once the accusation has been leveled, the leveler then rhetorically struts around like an alpha rooster, crowing as if the argument has been won.

    Income or wealth redistribution should not end any argument. It should, in fact, be an integral part of the discussion among those who base their preference for markets on the fundamental theorems of welfare economics, which I will attempt to explain intuitively (though I am not confident I will succeed). Lest you think this is theoretical nonsense of no importance, crack any basic (or even advanced) text on economics and you’ll find these theorems. They are also behind the scenes in many claims of market efficiency. Many people probably believe in the efficiency of the market from this perspective, even if they do not recognize it.

    Let’s start with the notion of a perfect market, one that exhibits perfect competition. In such a setting, there are an infinite number of consumers and sellers, all with perfect information and no transaction costs, among other requirements (check the prior links). Also, all transactions are voluntary and there are no externalities (meaning that everyone not involved in any transaction is not affected by it). Since nobody has to transact, it follows that every exchange makes both parties better off. Since there are no externalities, nobody outside the transaction cares. It’s pretty clear from this that an economy consisting of nothing but perfect markets in everything will arrive at a point where there is no way to make someone better off without making someone else worse off (aka, Pareto optimality). Only a party to a transaction can be harmed and nobody makes a transaction that doesn’t want to. (I’m assuming people use their resources in a way they find most pleasing (utility maximization) and don’t do things to harm themselves.) There are many such market outcomes. Pareto optimality is not uniquely satisfied.

    What I’ve just described is basically the First Fundamental of Theorem of Welfare Economics. There’s a second one as well that says that any particular market outcome of the type just described (nobody can be made better off without making someone worse off) can be achieved via perfect markets given some initial distribution of wealth. If you don’t like the outcome, you can change it to another (Pareto optimal) one by redistributing wealth. I am not sure how to sense intuitively that all Pareto optimal outcomes can be achieved this way, but I think it is fairly clear that changing the wealth distribution will change market outcomes.

    Consequently, if you are among those that justify markets based on these theorems — even implicitly because you buy the concept of the efficient market even if you don’t know where it comes from — and believe that actual markets resemble perfect markets sufficiently well, the only way you can and should condone intervention in order to change the market outcome is by changing the distribution of wealth, i.e. redistributing it. That is, redistribution should not be inherently abhorred. Among market purists of this type, it is the only policy lever that remains. (There are other reasons one might favor markets. One need not buy the justification of the fundamental theorems. There are reasons why one might find markets repugnant too, at least some of them. See Deborah Satz’s Why Some Things Should Not Be for Sale.)

    Hence, what should the distribution be? is a very natural and wise question to ponder, at least among people of this type. It is highly appropriate for the question to arise. The charge that “You just want to redistribute wealth!” could be met with, “You got a better idea to address concerns about the market outcome? Or, do you just like the outcome we have, in which case, why don’t you say so?” Without redistributing income, how would a theorem-based market purist propose we address inequality of outcomes we observe? There are good answers to this question* but one of them is not, “You just want to redistribute income.”

    Amartya Sen published a paper in 1985, The Moral Standing of the Market, that makes something like these points, among many other excellent ones. I highly recommend you get your hands on this paper and read it.

    [I]f there is an absence of – or reluctance to use – a political mechanism that would actually redistribute resource-ownership and endowments appropriately, then the practical relevance of the converse theorem [the Second Fundamental Theorem of Welfare Economics] is severely limited. [...]

    But the issue of inequality does have to be addressed, whether inequality is seen in terms of utilities, well-being, incomes, resources, or freedoms (including the real “freedom to choose”). The practice of avoiding this question through evasion or silence, on the one hand, or through peculiar definitions of “optimality,” on the other, seems hard to defend.

    We must discuss and address market failures and income distribution. To avoid them is to absolve oneself of the moral consequences of (usually imperfect) market outcomes. That is very hard to defend indeed.

    * One good answer would include proposals to improve the functioning of markets that deviate from perfection, if possible. Practically speaking, this is not always possible, certainly not always quickly or in a politically viable way.

    Acknowledgments: Many thanks to Paul Kelleher and Julian Jamison for their comments on earlier versions of this post.


    Comments closed
    • This is an amazing post — not for what is said but for what is not said. “Redistribution of income” is simply another term for “theft.” Was Moses strutting around like an alpha rooster when he brought us the commandment not to steal? Redistribution through the political process is nothing more than legalized theft. Does it become more moral if the thieves cast a vote before they loot their victims?

      Also, as Aaron Director pointed out years ago, redistribution rarely helps the poor. Generally, they are one of the groups who most frequently gets plundered.

      • I just checked all the economics textbooks I own and none of them point out that the fundamental theorems of welfare economics justify theft. Is it a vast conspiracy? I encourage you to start a petition or letter-writing campaign to publishers to correct this remarkable oversight.

      • John Goodman, two points. First, all property is determined by the law, in other words the state, and at some point the state has adjudicated between competing claims. So what you are saying is that theft is what the state hasn’t sanctified, which given the tone of your remarks is, I’m guessing, something you’re not that comfortable with.
        Second, to claim that redistribution plunders the poor flies in the face of all the evidence. So, have you got any? Evidence, I mean.

        • Of course he does. He has evidence in the form of being President of a right-wing think tank. His yes-men reassure him that his ideas are unimpeachable. Case closed. Why should we need anything beyond his self-evident proclamations on the nature of property?

          I know it’s shocking, but even PhDs in economics will troll websites sometimes. Best to ignore him, lest you get sucked into his morality-laden pablum at the expense of empirical policy analysis.

      • ““Redistribution of income” is simply another term for “theft.””

        Right – like when the damned socialist Lincoln started to redistribute the ownership of some of the means of production from the capitalists to the workers (formerly known as slaves).

    • To complete the discussion, and to engage in an effective debate, one needs to know other options that may “improve the functioning of markets.” You allude to it with your asterisk.

      To those who abhor redistribution, this seems like their chosen path.

      Is empiric literature indicative (read: better) of one vs the other, ie, redistribution over improving market functions?

      • Great question. I’d frame it as “How does one optimally address market failure?” I don’t have a general answer and I doubt an empirical one exists in general. However, for specific cases, there are answers. I’m aware of a forthcoming paper by Jon Kolstad and Amanda Kowalski that investigates the welfare economics of different coverage expansion pathways that vary in degree of redistribution. I’ll point it out in a Reading List post when it is out.

    • Taxation and regulation are not theft. Property–how it is acquired, how one can use it, how one can pass it to others–is a legal and administrative construct shaped by legitimate public institutions over many generations. Many democratic societies address these issues differently, depending on theor own circumstances and values. See e.g. Nagel’s famous essay here http://www.jstor.org/pss/795521 critiquing Nozick.

    • The Ten Commandments are part of a religious heritage that commands things that support the poor and helpless through what I guess John Goodman would consider “theft” because if we do so as a community some degree of redistribution is necessary.

    • Probably useful to distinguish between “market failures” and “things I don’t like” as well.

    • It’s one thing to note that market outcomes are imperfect by some arbitrary standard, however defined (this includes abstracting social phenomena into a limited subset of variables with numerical values, plugging them into a formal welfare econ model and calling whatever optima emerge out of the back end an ideal distribution). It’s quite another to demonstrate that whatever corrective intervention is introduced to correct this failure won’t be just as arbitrary as the market outcome. Ditto for such interventions actually bringing about their proposed ends in practice without generating unintended consequences that not only fail to improve the welfare of the intended beneficiaries, but actually make everyone else worse off in the process.

      Let’s address the technical and practical realities that stand between defining ideal distributions of income or some other social good in a particular moral or theoretical framework and actually realizing those ends in practice before we conclude that whatever emerges from that process will leave everyone better off than they’d be otherwise.

      Even if two people share an ideological commitment to a particular redistributive end, they can have meaningful disagreements about the best way to achieve it based upon theoretical and practical grounds. I like the idea of improving earnings of the the least skilled and educated persons in society, but have technical reasons for concluding that a negative income tax is a vastly better means of doing so than a minimum wage.

      The same is often true in discussions regarding redistributive ends in health care. When encountering hostility to a scheme designed to improve health care by means of redistributing resources, it’s worth considering the extent to which the opposition is grounded in a blanket hostility to all redistributive ends, or upon technical/practical objections to a particular plan.

    • When some one says “you just want to redistribute income”, the point should be, so you should support a simple direct redistribution like a wage subsidy rather than this, often, monstrosity. When proposing a policy we should always ask is there a more efficient way to accomplish the goal and if the goal is income redistribution most of the ways the USA Governments do this are very inefficient.

    • There is no such thing as a free market. Never has been. Never will be. Society is promulgated on the idea that government creates the infrastructure upon which all businesses and all people rely. Our biggest infrastructure, if you will, is the idea of money. without money, we would be still stuck on the barter system. And who backs money? The government of course. The money we use is the property of the government. It is not ours: we are allowed to use it. The government could take it all back tomorrow: tax you at 100 percent if it so desired. Not only that, you are prohibited from creating your own money.

      Furthermore, there is no private interstate highway system; nor is there a national network of private local roads. And we won’t even talk about tax codes and how they redistribute income and determine what businesses shall benefit. The question is not whither government involvement in redistributing income but how MUCH government intervention do we desire?

      A free market is like the mythical Yeti: much discussed with no real proof.

    • “Consequently, if you are among those that justify markets based on these theorems — even implicitly because you buy the concept of the efficient market even if you don’t know where it comes from — and believe that actual markets resemble perfect markets sufficiently well, the only way you can and should condone intervention in order to change the market outcome is by changing the distribution of wealth, i.e. redistributing it.”

      It’s also worth asking how many people actually justify their beliefs (even implicitly) in terms of these abstract neoclassical formalisms (including neoclassical economists). Sufficient grounds for justification? Perhaps. Necessary? No – thank goodness.

      • Good subject for some qualitative analysis. Ask people if they are pro-market. Ask them why. What do they say? Likely, “more efficient” comes up. What do they mean by that? Probably they heard it somewhere. Trace it back …

    • I think one of the basic differences between those on the left and the right is their approach to market failures. The right tends to deny they exist, and if they do, any government intervention will make things worse. The end result is wealth concentration. The left sees more market failures and believes they can be fixed. Sometimes it leads to success, like the EPa cleaning up our country since the 70s, or it can lead to failure when the fix is actually written by the affected industry to exclude competition. I tend to think things can be fixed.


      • They key is that if one denies market failures and one justifies market on grounds of “efficiency” then one is implicitly following the theorems. The theorems say that the only way to change market outcomes (and retain all that wonderful efficiency) is to redistribute wealth. There’s nothing left.

        • True, but it appears that those who do not see market failures are also happy with the outcomes, or at least resigned to them. Still not sure about libertarians. I think they just assume all imperfect outcomes are due to government interference.


        • There’s an important caveat here, which is that the theorems only apply to the tautological equilibrium constructs that explain how entirely imaginary markets work when inhabited by imaginary beings. The map isn’t the territory. In the case of DSGE it’s the rough equivalent of a 16th century map of the globe – there are a few areas where it’s useful but there are large swaths of territory where it’s either not particularly helpful or downright dangerous to rely upon it.

          (One example here: http://blogs.ft.com/maverecon/2009/03/the-unfortunate-uselessness-of-most-state-of-the-art-academic-monetary-economics/#axzz1njWRa0yP)

          IMO all anyone needs to do to rebut claims based upon the neoclassical netherworld is observe the fact that society is more than a series of transactions, people are nothing like rational utility maximizing agents, and that there are virtually no transactions that ever come anywhere close to satisfying any of the formal conditions required for the welfare theorems to hold.

          It’s possible to go further and remind everyone of the fact that even if it were possible to construct redistributive policies that maximize welfare in a way that’s consistent with the precepts of the theorems, they’d have to be filtered through a very imperfect Congress and implemented by very imperfect humans. E.g the same folks who brought us pareto minimizing monstrosities like Corn Ethanol.

          -Useful and interesting review of Welfare Econ below:

    • Well, I do not think “redistribution” is theft. Representative government under the will of the people can institute social programs and social insurance programs.
      And I believe that less inequality is a good thing, and is something that flows NATURALLY from an economy. I do not believe that the people who run Goldman Sachs, AIG, Lehman, et al are “master of the universe” and will naturally and meritocrally rise to the top – if you gave me trillions at zero percent (after I had lost trillons based on my own decisions), and you do all the enforcement to prevent counterfeiting, etcetera, I know I could make money.

      It appears to me that “re-distributionalists” have imposed a unintended consequence. Take for instance “saving” the financial “system” – the full force of the government supporting people that capitalism deems a failure in companies that capitalism deems failures – - bankers who don’t know who to make loans to, financiers who don’t know how to price risk, bond raters who would rate Moms Mabley a great beauty, etcetera. I would also give as an example of “flat tax” which is opposed by progressives (or liberals – its just words)- – which has done more to assure that the well off pay less as a PERCENTAGE of their income than practically anything else.

      Laws to be fairly and rigorously enforced have to be simple and clear enough for the average person (who will sit on the jury) to understand them. In my view, there is nothing that supports the corrupt subsidy of the current system than all the bad laws recently enacted.

      Before Sarbanes Oxley or Dodd Frank, we had the S&L crisis. More than 1000 high level prosecutions took place – it is not a matter of more laws and regulations – it is a matter of more will and honor.

      I doubt that most people substantially disagree about fairness, but people look at how to acheive it differently. I happen to think that if the incompetent 0.01% had lost their ill gained fortunes, that would do more for the cause of correcting inequality than the vast majority of government policies.

    • As far as I know, an efficient market outcome requires perfect insurance markets.
      If there are perfect insurance markets, and people are risk averse – the income distribution will be pretty even (add a theoretical possibility to barter before you are even born, i.e. ”behind the veil of ignorance” , and the outcomes will be perfectly equal).

      If there are risk loving people, every casino would be full of people putting all of their money on red, until they either lose everything or have won all of the casinos money (i.e. putting everything on the table at once – if they slowly play away their money they might as well do it because they enjoy the game)

      Empirics tell you that very few, if any, risk loving people exist.
      Empirics tell you that there are big differences in incomes.

      Thus, redistribution will generally get you closer to the efficient outcome.

    • There is no such thing as a free market in a modern society. Government must be there to ensure provision of system-level infrastructure and to set the rules of the market.

      Sometimes government systems and rules have the effect of redistributing wealth upward (e.g.: protection from foreign competition for Doctors and Lawyers, but not for factory workers).

      Sometimes government systems and rules have the effect of redistributing wealth downward (e.g.: pension systems, public schools, public health insurance).

      But redistribution of wealth is not the goal, it is a side-effect of the fundamental need for government to provide systems and rules for markets to function in the modern world.

    • Well, if the distribution of income revenue from product and services sales was fair and equitable at the company level, there would be no call for any “re-distribution. Now would there?

    • @ Goodman (and others) – “Redistribution through the political process is nothing more than legalized theft. ”

      With all do respect, saying it doesn’t make it so.

      Successful businessmen and entrepreneurs certainly have a legal claim to manage the capital in their possession, but it doesn’t follow that they have either a moral or a political claim to the whole of its usufruct. The view that they are its stewards only, and that they hold it in trust to the community as a whole, is perfectly valid from society’s point of view, –especially when you consider the long history of capital accumulation. For what is capital if not the accumulated crime and sacrifice of centuries, plus interest? That is a tautology by the way.

    • Fernholz and Fernholz 2012 looks relevant here: “Wealth distribution without redistribution”, http://voxeu.org/index.php?q=node/7667 ; paper is at http://www.ricardofernholz.com/Redistribution.pdf .

      Even with perfectly voluntary exchange absent suasion, perfect (market) information, perfect competition, the absence of frictions and externalities, etc. – in a perfect market, if there are uninsurable risks to wealth, wealth tends to concentrate to a single point.

      Is this Pareto optimal? If so, then so much for Pareto optimality as an analytical tool.

      @ Goodman: Insurance is redistribution of (negative) income and wealth. Enough said on that.

    • @Greg:

      Is that a conclusion regarding insurance something that follows from a computational exercise conducted within the boundaries required for the welfare theorems to hold, or a personal opinion? It’s an interesting conclusion in either case, though I’d be interested in where you are coming from on that one.

      In classical econ, I believe the conclusion was that insurance was welfare enhancing in that it reduced the amount of material wealth that people had to hoard in order to gird themselves against various kinds of hardship and disaster, and being able to buy an insurance contract instead of hoarding wealth increased the aggregate level of savings, investment, and production and left everyone else better off than they would have been otherwise.

      If insurance is always and everywhere a negative redistribution of income and wealth, the eagerness with which people of various ilks spontaneously created and perpetuated mutual aid societies is a curious facet of economic history, as is the spontanous emergence and persistence of insurance within all market economies.

      Can’t help but ask – do you have any insurance policies that you aren’t legally obliged to hold?

      Anyhow – thanks for the interesting links and comments.

    • Pareto efficiency is a very low bar morally speaking since it does not question the historical distribution of resources. E.g. trade with China was Pareto efficient in the sense that it would have been possible to tax the winners and compensate the losers, leaving everyone better off than before. But since we didn’t do that a the time it is now Pareto optimal not to compensate the losers. And so on through history.

    • Let’s say you’re 7 years old and you go on a hike with your grandmother. During the hike you keep finding neat rocks and your grandmother keeps finding tasty mushrooms. Which is great…until neither of you have any space left in your pockets! Alas! So each time you find a new rock…and your grandmother finds a new mushroom…each of you has to make some hard decisions. You’re forced to prioritize your rocks and your grandmother is forced to prioritize her mushrooms. These tough decisions help ensure that scarce resources (your pockets) are efficiently used.

      The thing is…you can’t make these decisions for your grandmother…given that you don’t even like mushrooms and you have no idea which ones are edible. Likewise, your grandmother can’t make these decisions for you either…given that she sees absolutely no value in the rocks that you have been collecting.

      The moral of the story is that you didn’t link to the right Wikipedia articles…

      1. We all have some information but nobody has all the information… http://en.wikipedia.org/wiki/Dispersed_knowledge
      2. We all have different values… http://en.wikipedia.org/wiki/Opportunity_cost

      Public healthcare, public education, public welfare, infrastructure, national defense, environmental protection, etc. are all extremely valuable to many people. The question is…how should limited public funds be distributed among all the different possible government organizations? The answer is simple…we should allow each and every taxpayer to choose which government organizations receive his/her individual taxes.


      Now, 99% of you are going to say, “Oh no! but important government organizations would be underfunded!” http://pragmatarianism.blogspot.com/2012/01/unglamorous-but-important-things.html

      The thing is, your collective concerns clearly indicate that there is indeed a demand for public goods. And obviously we already have government organizations that can supply public goods. Therefore, why wouldn’t we want the demand for public goods to determine the supply of public goods? Why wouldn’t we want limited public funds to be efficiently allocated? If we want limited public funds to be efficiently allocated then we need each and every taxpayer to consider the opportunity costs of their tax allocation decisions. It’s a good thing for taxpayers to prioritize all the different public goods that they want.