Inequality — How much is too much?

What shape is the income distribution of Andrew Leigh’s dreams? Even he doesn’t know. "I don’t have a strong sense of what the right level of inequality is", he writes. "Indeed, I’m not even sure I have the right intellectual framework for answering the question."

The question is Andrew Norton’s. In the comments threat of recent post on ‘progressive fusionism’ he writes:

…‘progressives’ tend to think that there is a correct distribution of resources that can be decided in advance. However, in practice they tend to be very vague about what this correct distribution would actually look like. Andrew Leigh, for example, has written much about inequality of income, always with the assumption that less inequality is the correct outcome, but never saying what level of inequality would satisfy him.

So how should a ‘progressive fusionist‘ answer the question? The Cato Institute’s Will Wilkinson suggests that a new alliance of progressives and classical liberals might combine John Rawls’ ideas about justice with Friedrich Hayek’s ideas about markets. From this perspective, it’s not possible to decide on a correct distribution in advance. That’s because the question isn’t a purely philosophical one. On its own, Rawls’ theory doesn’t tell you what shape the income distribution should be.

Progressive fusionism?

‘Progressive fusionism’ is Brink Lindsey’s name for an alliance of American liberals and libertarians. He argues that both groups want to give individuals more control over their own lives:

The central challenge in cementing a new fusionist alliance–and, make no mistake, it is a daunting one–is to elaborate a vision of economic policy, and policy reform, that both liberals and libertarians can support. Here, again, both sides seek to promote individual autonomy; but their conceptions differ as to the chief threats to that autonomy. Libertarians worry primarily about constraints imposed by government, while liberals worry most about constraints imposed by birth and the play of economic forces.

The basic outlines of a viable compromise are clear enough. On the one hand, restrictions on competition and burdens on private initiative would be lifted to encourage vigorous economic growth and development. At the same time, some of the resulting wealth-creation would be used to improve safety-net policies that help those at the bottom and ameliorate the hardships inflicted by economic change.

Philosophically, Cato’s Will Wilkinson suggests, fortifying "Rawls’ theory of justice with a Hayekian grasp of the coordinating function of prices, and the dynamics of spontaneous order" or fortifying "Hayek with Rawls’ rather more intelligible normative framework".

Using Rawls’ theory of justice as an intellectual framework

Rawls was an egalitarian, but this doesn’t mean that he thought that income and wealth should necessarily be distributed equally. An important element of his theory of justice is the difference principle — a principle which tells us when we should favour a system that leads to greater inequality over one which leads to less. In a 1967 essay titled ‘Distributive Justice’, Rawls explained it this way:

The basic structure [of a social system] is just throughout when the advantages of the more fortunate promote the well-being of the least fortunate, that is, when a decrease in their advantages would make the least fortunate even worse off than they are. The basic structure is perfectly just when the prospects of the least fortunate are as great as they can be (p 138).

On its own, Rawls’ theory says almost nothing about what shape the income distribution should be. To answer this question, we need to combine Rawls’ theory of justice with a theory which explains how alternative social and economic systems perform.

What if Hayek was right?

Egalitarians have been wary of markets for two reasons. First, markets inevitably create inequalities in income and wealth. Even if everyone starts out with equal quantities of human, social and physical capital, trade in the marketplace means that they will eventually end up with unequal shares of wealth and income (Robert Nozick explains why). Second, a pure market society leaves many people with little or no income. Those without assets who are unable to work, are forced to depend on begging, charity or foraging in garbage heaps.

As a result, egalitarians have often been attracted to socialism. Since the market is the cause of inequality and poverty, the less influence the market has on a person’s level of material well-being the better — or so socialists assumed. But Hayek challenged this assumption when he argued that a planned economy can never be as efficient as a market economy. This is because it is unable to use dispersed information as effectively. He argued, in effect, that attempts to engineer a patterned distribution of income will end up reducing the well-being of the least advantaged.

Rawls’ difference principle says that we shouldn’t trade off the well-being of the least advantaged in order to achieve greater equality in the distribution of resources. Hayek’s analysis says that the well-being of the least advantaged depends on a system that produces a significant degree of inequality. So if Hayek is right, Rawlsian egalitarians ought to choose a social system that includes market institutions and, as a result, sacrifices a certain amount of equality.

From a progressive fusionist perspective, the best systems would probably include a significant amount of income redistribution and government funding for services. Hayek supported both measures. In The Constitution of Liberty he wrote:

All modern governments have made provision for the indigent, unfortunate and disabled and have concerned themselves with questions of health and the dissemination of knowledge. There is no reason why the volume of these pure service activities should not increase with general growth of wealth. There are common needs that can be satisfied only by collective action and can be thus provided for without restricting individual liberty. It can hardly be denied that, as we grow richer, that minimum of sustenance which the community has always provided for those not able to look after themselves, and which can be provided outside the market, will gradually rise, or that government may, usefully and without doing any harm, assist or even lead in such endeavors (p 257-258).

How much inequality?

It’s difficult to know what set of rules and institutions would best satisfy Rawls’ difference principle. And unless you know what these rules and institutions are you can’t begin to model the resulting level of inequality. Everything from rules about corporate governance, intellectual property rights and the design of the income support system will influence the result. And on top of that the issue is further complicated by the fact that well-being is a broader concept than income or wealth (not to mention the problem posed by foreigners).

Personally, I suspect that the difference principle requires a combination of big government and free markets. But libertarians like Andrew Perraut think that this is only true in the short term:

… if markets are as massively productive as we libertarians believe and compounding returns to growth in the long term are taken into account, you could probably justify no more than very basic safety nets, for fear of distorting the economy and dramatically lowering everyone’s goods in the future.

Perhaps one way to have a more productive debate about these kinds of issues would be to use computer simulations — something like an economically sophisticated version of the computer game SimCity. Users could select their preferred rules and institutions and have the software simulate the way their hypothetical society might develop. Since the simulation would apply economic theory to generate the outcomes it would be more interesting if users had a choice of theories to play with. Users could experiment with different tax and welfare regimes and watch as GDP grew or contracted, poverty rates rose or fell, and citizens chose to stay or leave.

What’s the point?

While philosophy students might enjoy debating the similarities between Hayek and Rawls, is this debate likely to influence real world politics? There are two reasons for being sceptical. The first is that few people either know or care about Hayek and Rawls. And this includes most politically active classical liberals and left-leaning liberals. Even at the great purple heart of the left wing Australian blogosphere, Rawls has few supporters. Similarly, most libertarians and classical liberals have only a vague idea of what their Austrian hero actually wrote.

The second problem is that political movements form around policies rather than principles. As the philosopher Richard Rorty argued, "Principles are useful for summing up projects, abbreviating decisions already taken and attitudes already assumed." What they are not useful for is creating new movements. People who change the world often form alliances with others who agree with them on what to do but disagree about why.

Rorty argues for something that sounds a lot like Andrew Leigh’s experimental approach to reform. As usual Rorty credits the idea to another philosopher:

John Dewey hoped that democratic politics would cease to be a matter of batting plausible but contradictory principles back and forth. He hoped that it would become a matter of discussing the results, real or imagined, of lots of different social experiments. The invidious comparisons I am suggesting amount to saying: Look, a lot of good experiments have been run, and some of them have been pretty successful. Let’s give them a try.

But it’s worth remembering that the link between practice and principles is two way. People decide what policies are likely to produce good results based on their ideas about how the world works. And, in turn, the results they get shape their ideas. For example, think about how the experience of stagflation in the 1970s changed people’s minds about Keynesianism.

Part of what keeps a new progressive alliance from forming is that people mistake differences in ideas about how the world works for differences in moral principles. Left-leaning liberals look at the policies classical liberals support and assume that the motivation is to redistribute income from the poor to the rich. And classical liberal look at left liberals and assume that they are motivated by an envious desire to punish the rich even if it means making everyone worse off.

Starting a conversation about principles may be the best way to challenge these assumptions and encourage the two groups to start working together on new policy ideas. Maybe we should try it and see what happens.

This entry was posted in Libertarian Musings, Political theory, Uncategorised. Bookmark the permalink.

121 Responses to Inequality — How much is too much?

  1. James Farrell says:

    Thanks for making the effort to formulate something on this topic, Don. There’s two issues that shouldn’t be confused, however. It shouldn’t be that hard to come up with a degree of income inequality that reconciles fairness with the need for material incentives. Orwell in The Lion and the Unicorn suggested ten to one; I imagine that five to one would be enough to extract the maximum effort from anyone who needed that motivation, though of course it wouldn’t work in just one country. (As an aside, I don’t think the burden of proof has to be on supporters of inequality, to show that it’s desirable. To me it’s self evident.)

    The much harder problem is how to design a set of rules that would preserve such a distribution. Even with very high marginal tax rates on individuals and inheritance taxes, you would still end up with a tail of very rich people, due to windfalls, inventions, and clever dealings. The most we can practically achieve without creating too much bureaucracy and surveillance, and without breeding too much resentment is a decent living wage at the bottom end, with a Scandinavian welfare state and minimum wages (via government wage subsidies if necessary.)

  2. Thx Don,

    I enjoyed reading the post.

  3. NPOV says:

    I’d suggest taking the average GINI co-efficient of the 10 countries in the world with the highest HDI as a start. Which would make Australia’s pretty good (and the U.S.A’s pretty bad).
    The correlation between GINI co-efficient and HDI to me is pretty good evidence of the advantage of aiming for a reasonable GINI. Has anyone got a good scatter-plot showing this? I’m sure I saw one not long ago.

  4. I’ve written a paper on this topic for jurisprudence; unfortunately it’s due on the 18th of April, so I’m not in a position to make it public until it’s been handed in. Someone give me a nudge if I forget and I’ll put it up over at the Cat.

  5. conrad says:

    I’m not at all convinced about GINI coefficients. First of all, I think the main problem is to do with the extreme left of the distribution, rather than the entire distribution. Second, there are government policies that make huge differences to the real poor which have nothing to do with the coefficient. Places like HK have a much higher coefficient than Australia, but half population lives in public housing. Given this, I’d rather be poor in HK than richer and homeless in Australia — and the coefficient says nothing about these sorts of things.

  6. NPOV says:

    That’s definitely true, but it’s still about the best measure out there. And one or two exceptional cases (and there is a fair bit that is exceptional about Hong Kong) don’t invalidate a general guideline.

    Also, I’m not entirely sure the wide availability of public housing in HK does make the situation better for the poor there. Surely it would be better if they had enough money to afford housing of their own choice? Of course land is such a premium in Hong Kong that no other solution may be viable.

    Ultimately it has to be up to each country AND its citizens to decide what they want – ideally through democracy. Which is of course one area where Hong Kong fails rather miserably – so we have no real way of knowing whether most citizens of Hong Kong would prefer a government that aimed for reducing inequality.

  7. Don Arthur says:

    James – That’s an interesting point.

    I think the biggest confusion is to think we need inequality primarily to extract the maximum amount of effort from workers.

    It seems to me that major problem isn’t to stop workers from shirking but to make sure that resources go to where they are most valued. Milton Friedman put it this way:

    A system based on private property and free markets is a sophisticated means of enabling people to cooperate in their economic activities without compulsion; it enables separated knowledge to assure that each resource is used for its most valued use, and is combined with other resources in the most efficient way.

    The examples I always think of are intermediate goods — goods that are used to produce other goods. How would government planners know how to allocate these goods in the absence of price signals?

    As you say, once the system is in motion it starts to generate a long tail on the right — for all kinds of reasons.

    Personally, I think the shirking problem is a distraction. It’s more of a within-firm problem than a system design problem.

  8. hc says:

    Personal expressions of the need for redistribution suffer from biases. Public goods biases – the cost is spread around but individuals think mainly of the possible costs of their own contributions. Also hypocrisy biases – in the main its a hypothetical exercise in which it is easy to exaggerate your willingness-to-pay. In fact people can redistribute whatever fraction of their income they choose now without any public intervention by means of private charity. That they don’t do so with generosity suggests hypocrisy and public goods biases.

    Rawls is ludicrous in spite of the ‘difference principle’. Consider two societies. Society A has 20,000,000 people with income $100,000 and 1 person with income $5,000. Society B has 20,000,000 people with income $2,000 and 1 person with income $5,001. I think Rawls says society B is better off which no-one with common sense would agree with.

  9. Patrick says:

    A good post which I enjoyed reading.

    Perhaps one way to have a more productive debate about these kinds of issues would be to use computer simulations something like an economically sophisticated version of the computer game SimCity. Users could select their preferred rules and institutions and have the software simulate the way their hypothetical society might develop. Since the simulation would apply economic theory to generate the outcomes it would be more interesting if users had a choice of theories to play with. Users could experiment with different tax and welfare regimes and watch as GDP grew or contracted, poverty rates rose or fell, and citizens chose to stay or leave.

    We will get these kinds of experiments once we start settling in space – but I fear we might have to wait until then! I can’t help but think that any computer modelling of such sophistication would only recreate certain assumptions which would skew the result. But computers might easily develop enough to encompass that.

    And back on the progressive fusionism:

    Left-leaning liberals look at the policies classical liberals support and assume that the motivation is to redistribute income from the poor to the rich.

    This is often enough expressly stated by left-leaning liberals so I am willing to believe it is true.

    And classical liberal look at left liberals and assume that they are motivated by an envious desire to punish the rich even if it means making everyone worse off.

    This may be true. But I don’t think it is ‘serious’. I think the real issue is that classical liberals look at left liberals and think that they must be ideologically blinded to have so much faith in untested (or indeed tested and failed) ideas and so little regard for evidence.

    There are several exceptions to that and Andrew Norton lists most of them in his post. But on the whole, support for trade barriers and ‘fair’ trade, high marginal tax rates, anti-vilification laws and other such furphies are a prima facie proof to many liberatian-minded liberals (the ones in which you place so much hope) of either mental infirmity or ideologically blindness, basically.

    The much harder problem is how to design a set of rules that would preserve such a distribution.

    I agree, as I expect does everyone! Personally, I would opt for a strong bias in favour of lesser barriers to change and bad choices but with targeted compensation for the ‘losers’ (or in some cases, merely ‘bad choosers’, which doesn’t mean we should let them die). In general, I think that leads one perhaps somewhere inbetween Australia and Danish ‘flexicurity’.

  10. Don Arthur says:

    I think Rawls says society B is better off

    Harry, I’m intrigued. Why do you think that?

  11. James Farrell says:

    I suspect Harry meant to add one more zero to the $2,000.

  12. NPOV says:

    hc, that people don’t voluntarily sacrifice a certain percentage of their income to aid in redistribution isn’t surprising. For a start:

    a) Much of the time, it’s a hassle. I just want it done automatically. After a while anything that’s a hassle you tend to attempt to avoid.
    b) There’s very little point in me voluntarily giving my income to others if I’m on the only one to do so. I need to be sure that everyone else who can afford to will participate in the scheme.
    c) I don’t have enough time, expertise or information to determine who will best benefit from my voluntarily sacrificed income. Most likely I will give it to some cause that happens to appeal to me on a fairly irrational basis. Now of course I accept that governments don’t always do a great job of spending public money, but that is their job, and we elect them to do just that. So it’s surely logical to give my money to the government, and hold them accountable (via the various democratic institutions) to doing that job well.

    On that basis, I would be reasonably happy to volunteer to pay a little more tax than I do. Unfortunately, there’s not very many ways I can actually do that (whereas there’s literally dozens of ways I can legally reduce my tax burden considerably). Further, if I was going to volunteer to pay more tax, I would want a little more say over how it was spent (or at least, to be guaranteed it wasn’t spent on projects that I strongly disagree with).

  13. hc says:

    Don, Because the worsrt off sections of society are better off under B

  14. Patrick says:

    a) Much of the time, its a hassle. I just want it done automatically. After a while anything thats a hassle you tend to attempt to avoid.

    Fair enough. Nick Gruen loves this topic, and indeed there are a few books on it.

    b) Theres very little point in me voluntarily giving my income to others if Im on the only one to do so. I need to be sure that everyone else who can afford to will participate in the scheme.

    This is ridiculous. I said above that classical liberals don’t really believe that left liberals believe in class war but maybe I was wrong. Isn’t the point that if you believe we should all give more to the poorest people the you should do so anyway? Are you driven in fact by concern for your relative position?

    c) I dont have enough time, expertise or information to determine who will best benefit from my voluntarily sacrificed income. Most likely I will give it to some cause that happens to appeal to me on a fairly irrational basis. Now of course I accept that governments dont always do a great job of spending public money, but that is their job, and we elect them to do just that. So its surely logical to give my money to the government, and hold them accountable (via the various democratic institutions) to doing that job well.

    This is even more ridiculous. Can’t you hold yourself accountable? Can’t you focus your investments in a particular charity and hold them accountable? And on the theme of your last paragraph, wouldn’t giving to government come with a guarantee that much of the money would be spent on things you would not have personally chosen to spend it on?

  15. Don Arthur says:

    Society A has 20,000,000 people with income $100,000 and 1 person with income $5,000. Society B has 20,000,000 people with income $2[0],000 and 1 person with income $5,001.

    Even some people who agree with Rawls’ general approach (choosing principles from behind a veil of ignorance) find the difference principle intuitively implausible.

    It’s worth noting that idea of Pareto optimality runs into the same objection. In a 1968 essay of distributive justice Rawls wrote:

    … the difference principle satisfies the principle of efficiency. For when it is fully met, it is impossible to make any representative man better off without making some other man worse off, namely, the least favored representative man. Thus it defines a conception of distributive justice that is compatible with Pareto optimality.

    I don’t think Rawls would talk in terms of one society being ‘better off’ than another. What he was interested in was whether a society’s basic structure was just.

    Rather than asking readers to choose between end-states, Rawls asked them to choose between the sets of principles that generate them. He argued that if the principles are just then nobody can complain about the outcomes.

    But I take your point. Many people don’t find the difference principle intuitively appealing.

  16. swio says:

    Perhaps one way to have a more productive debate about these kinds of issues would be to use computer simulations something like an economically sophisticated version of the computer game SimCity. Users could select their preferred rules and institutions and have the software simulate the way their hypothetical society might develop. Since the simulation would apply economic theory to generate the outcomes it would be more interesting if users had a choice of theories to play with. Users could experiment with different tax and welfare regimes and watch as GDP grew or contracted, poverty rates rose or fell, and citizens chose to stay or leave.

    This already happens all the time. That’s what economic modelling is. It has two major problems.
    1) Only an economist has enough expertise to run these computer simulations. Would you trust a computer model of climate change run by a philosopher?
    2) The bigger problem is these models are completely dependent on the underlying economic theories. Again comparing to climate change, a model of the Earth’s climate would not work if you did not have Boyle’s Law, and many other carefully worked out laws that have been proven in the laboratory over literally centuries.

    This problem actually goes to the heart of almost all debates about social/economic issues today. Alot of the theory in economics is too vague or too uncertain to be useful in the real world. The world’s most emminent economics professors disagree, often quite fundamentally, about even basic questions like the effect of tax rates on growth. If the experts don’t have a consensus that produces reliable and proven laws for predicting economic variables then the rest of us can do no more than cherry pick opinions from particular economists that suit our own biases. This problem is central to the debate about libertarian ideas on society as liberatarians believe that lack of government is, in the long run, vastly more efficient and hence will produce so much more wealth that questions of distribution don’t become important. But is that a testable hypothesis? Is that a proven statement? No, its just an opinion supported by view of some economists. Its like philosophers arguing about the age of the universe 50 years ago when half the world’s astrophyscists believed in steady-state thory and the other half believed in the big bang.

    Having said all that, there is a way you can conduct computer models that might answer these questions. Massive multiplayer online games like World of Warcraft, Second Life or Eve-Online are creating enormous virtual economies with millions of real participants where things like tax rates could be played with without causing disruption to the real world. The results would be alot more reliable because they would come from the actions of real people making real decisions about resource allocation rather than theories that are still subject to debate.

  17. NPOV says:

    Yes, we should give more to the poorest (specifically, give them enough that they have the same sort of opportunities that we have), but it’s human nature to avoid schemes where free-riding and cheating is too easy. I’m not saying it’s a good reason for me to personally avoid giving more to charity, but it’s just one of the reasons that people are likely to be wary of voluntarily charity – they ask themselves (consciously or otherwise) “why should I personally give up my money for the betterment of society when nobody else is”?

    As for point c), as I said, it is quite specifically the government’s job to be responsible for spending public money. It’s a significant amount of hassle for me to “focus [my] investments in a particular charity and hold them accountable”. Seeing as I don’t personally suffer from my voluntarily sacrificed income being spent poorly, there’s little motivation for me to make an effort to monitor it constantly. Whereas governments have a significant motivation for spending public money wisely: i.e. the ballot box. So on that basis, it’s hard to see how wealth redistribution would work better if it was entirely voluntary, regardless of how much more fair or noble it might seem.

    Regarding having more control over how my taxes are spent – yes, I would like to see this option. At least one person has seriously floated the idea of voters having a say on the budget: i.e. how much they think should be spent on various parts of it. Obviously it’s impractical (and foolish) to hand this entirely over to voters, but I can certainly see the merits in such a scheme.
    For instance, if I could specify that my taxes were to be mostly spent in the general areas of education, arts, public transport, health etc. and not to be spent on defence or certain industry subsidies, then I’d happily volunteer to pay more of them.

  18. swio says:

    From a progressive fusionist perspective, the best systems would probably include a significant amount of income redistribution and government funding for services. Hayek supported both measures.

    It must be remembered that Hayek was writing in a time when socialism was widespread and it was not clear that communism doesn’t work. Alot of what he was arguing against no longer even exists today. It is my reading that the anti-government position of libertarianim today is probably well beyond what Hayek himself envisioned. In other words, finding that Hayek might support alot of progressive policies is, perhaps, not nearly the same thing as saying that a Libertarian would.

  19. Partrick,

    I think you’re throwing round the word ‘ridiculous’ too much. For me anyway, poverty alleviation is a public good.

    Even that doesn’t quite capture the way people think and are motivated on redistributive issues. People’s other regarding nature is so strong that if engaged it’s worth a lot more than tithing. Witness parents working their guts out and going without themselves to provide for their families – this is quite normal.

    Why do they do it? Are they that altruistic? They do it because of a powerful mix of culture and biology. It makes this behaviour normal, expected and so on.

    At the societal level people feel exactly the same. This is true in spades of course in wars. Social solidarity is a powerful glue – so powerful that Hayek saw people’s natural sense of community solidarity as the enemy of the market. And as Bowles and Gintis report that there’s no great resentment towards taxes when it’s felt that they’re supporting the ‘deserving poor’. People are happy sacrificing to lift the ‘deserving poor’ in the community up. But most won’t be happy doing it if they think others are free riding on their efforts.

    They will feel the same as a sibling who’s left to look after some poor member of their family – say an old sick father – when other siblings don’t contribute.

    It’s of the essence of this situation that these obligations are felt socially – not just individually.

  20. Jc says:

    Why should anyone waste their time worrying inequality if the economic pie is growing.

  21. Patrick says:

    I think youre throwing round the word ridiculous too much. For me anyway, poverty alleviation is a public good.

    The former might be true. But I happen to agree with the latter as well?

    Theres very little point in me voluntarily giving my income to others if Im on the only one to do so.

    From your comments, Nick, I understand that NPOV wasn’t talking about a ‘point’ after all. Mea culpa, and your points are quite valid. But what NPOV actually wrote does seem ridiculous to me.

    As for his comments about government not being a necessarily good allocator of money but then again neither is he, really, so government should spend it because then he can vote for or against them, whereas he (presumably) can’t actually keep himself or some smallish charity accountable??? It still seems ridiculous to me!

    PS: you say:

    People are happy sacrificing to lift the deserving poor in the community up. But most wont be happy doing it if they think others are free riding on their efforts.

    but I thought the ‘free riding’ in question was that of welfare receipients not taxpayers.

    Very happy to be corrected on that!!

  22. Jc says:

    Lewt me ask anyone here if there is an upper limit in terms of what the government’s take ought to be? We’re currently at around 33%. Should it be more. How much more?

    I would guess that a lot of people would disagree with Milt’s suggestion that the total government take ought to be no higher than 10% or his comment:

    A society that puts equality ahead of freedom will end up with neither equality nor freedom.

  23. NPOV says:

    Jc, what’s the use of an economic pie that’s growing if there a significant number of people that are not benefitting particularly from it?

    At any rate, inequality in wealth leads to inequality in power, and consequently to exploitation, corruption, and social unrest. Human nature’s a bitch.

    As far as the % of GDP that government’s should be responsible for taxing and spending, there seem to be plenty of European nations whose citizens seem perfectly comfortable with it being close to 50%. Who’s to say they’re wrong?

  24. hc says:

    NPOV, You can easily have regular deductions made from your salary by the bank and paid to the Salvos or any other reputable group.

    You are wrong to say there is no point in you making an individual contribution. There might be greater aggregate effects if everyone pays but, if that does not happen, your individual contribution is worth more than it would be if everyone contributed since it can be directed to the very needy – those at the top of the queue in terms of need.

    The ‘I won’t give unless others do’ is a pretty safe way of limiting your payments since they won’t do so voluntarily and will only do so if subjected to a tax – then you will have to give anyway. It costs you nothing to advance this claim.

    I have no response to point c. It sounds self-contradictory to me. You don’t trust goverrnment but you will trust them on this one. In a privatised set-up many of the publicly-funded benefits paid to the most needy are paid via private charities anyway.

    Are you sure you are not a ‘welfare hypocrite’ who gets a warm inner glow from telling the world how much you care for the poor and needy but need to invent implausible excuses for not doing so? Join the club – it has a vast membership – most of the human race.

  25. NPOV says:

    hc, agreed, I overstated in implying that there was no point in me making a contribution if nobody else does. I’m more interested in looking at the reasons people (consciously or otherwise) might use for not donating as much as they can afford.

    I didn’t say I don’t trust government, just that they “dont always do a great job of spending public money”. I wouldn’t even say I trusted them any less or any more than private charities, but at least I know that the government is constantly under pressure to spend money sensibly, whereas I’m not so sure that charities always are.

    I don’t think I particular make a point of telling the world how much I care for the poor and needy: I would just like to know that I live in a world where the least fortunate of us (which may well include me at some point in the future) still have access to opportunities to improve their lives, and aren’t readily exploited by those with the wealth and power to do so. My motives are as selfish as anybody else’s.

  26. melaleuca says:

    HC says to NPOV:

    “Are you sure you are not a welfare hypocrite who gets a warm inner glow from telling the world how much you care for the poor and needy but need to invent implausible excuses for not doing so? Join the club – it has a vast membership – most of the human race.”

    True. And that is why I prefer Government to coerce money out of my pocket to provide for social welfare. That way I’m not tempted to find excuses for not giving money to private charity. While I do give to private charity, I could afford to give much more but don’t do so because I readily succumb to personal greed.

    As to Joe’s question re tax, when an individual is earning more than ten times the minimum wage I’d like to see them taxed at about 70%. I’d increase that to 95% at twenty times the minimum wage.

  27. Jc says:

    Ok, so that’s a starting point, N. You think the government take ought to be closer to 50%.

    You would achieve that by raising even more income taxes on those people that are currently paying the bulk of income taxes?

    You understand that you would be raising the governments take by 50% The effect of this is that you would be making a transfer towards increased consumption at the expense of investment: the very thing that would eventually lower living standards.

  28. Jc says:

    Credit where credit’s due. At least Mel and N are being honest and forthright in terms of where they want to see things.

  29. NPOV says:

    I assume melaleuca you mean that every dollar earned over 20 times the minimum wage (~$250/hr, or nearly $500,000/yr) should be taxed at 95%? Tax rates that high did use to exist some time ago, but I’m not sure there’s a lot of evidence that anyone ever paid them.
    I suspect once marginal tax rates get over 50% the motivation to avoid them become too high for them to be terribly useful. There are surely better ways to redistribute wealth.

  30. Jc says:

    Ummmm should read

    You understand that you would be raising the governments take by 100%

  31. Patrick,

    People don’t like free riding. It may well be this intense dislike of free riding leading to what’s called in the jargon ‘strong reciprocity’ that’s an integral part of what got us going as a viable species.

    We will punish those who we feel are ripping us off, even if it harms us – wars are one of the best examples of strong reciprocity.

    People intensely dislike free riding whether of the undeserving poor or the undeserving (ie non-contributing) non-poor.

  32. NPOV says:

    Actually Jc, no I don’t particularly believe that Australia would benefit by raising the government tax take that high. In fact, I don’t think there’s a huge need to increase it all – just to spend it better.

  33. Andrew Leigh says:

    Don, a classically great post. Though I’m still not sure if Rawls, Hayek, Lindsey and Norton can tell us whether Australia’s gini of 0.31 (or if you prefer, our 90/10 of 4.2 or our Atkinson One index of 0.17) is too high or too low. The great inequality philosophers frame our thinking. But when it comes down to the ‘does Australia have too much or too little?’ question, I’m still left puzzling.

  34. NPOV says:

    Also, Jc, would would a higher tax take imply “increased consumption at the expense of investment”? Australians are pretty lousy investors. OTOH, most areas of public spending I would think qualify as investments.

    I also accept that a much higher government tax take would put something of a brake on the rate of economic growth. But if the choice is between a medium sized pie divided up reasonably equally, and a huge pie where a few get most, and the rest fight for the scraps, I’ll go for the medium sized pie any day. Even if it potentially means I personally end up with a smaller piece.

  35. NPOV says:

    Bleh, make that “why would”, not “would would”.

    Andrew, is “Australia’s gini” supposed to be a link?

  36. melaleuca says:

    NPOV says:

    “I suspect once marginal tax rates get over 50% the motivation to avoid them become too high for them to be terribly useful. There are surely better ways to redistribute wealth.”

    That’s true. An alternative is to target the spending and assets of the rich and whack them with hefty death duties. I’m not fussed as to which method is employed.

  37. NPOV says:

    Relevant article by Ross Gittins Lloking at some studies that examined how much people would voluntarily put towards a common pool of funds that benefitted others more than themselves:

  38. Russ says:

    JC, my first thoughts on that question was that it depends on how much inequality you have. In a society where everyone had the same income then the optimum tax take for reducing inequality is probably zero (there will, of course, still be taxes for public goods).

    As soon as you move away from an equal society, and an economic model of an equal society then it is reasonably easy to see situations where some segments are out-bid for important goods by richer segments satisfying their desire, rather than a need. So your question is essentially (and obviously) political: what basket of goods and services should everyone have access to regardless of wealth. It is also going to be strongly orientated towards the tail of the distribution. I don’t have much else to offer on the political front. Inequality matters, but I can’t say how much.

    Two other things come to mind though:
    1) Assuming there is an efficient political equilibrium of welfare taxing, any society that increases equality (regardless of cause) should reduce their welfare tax rate. And the corollary, any equalizing society that doesn’t reduce their tax will probably distribute the benefits of welfare more broadly. Somewhere in that there is an empirical test.

    2) In relation to NPOV’s point on the economic pie. Changes to the welfare tax rate in the long term will depend on expectations and are therefore political. In most cases expectations will rise, and so will the rate. But in the short term they will depend on the inter-relation of growth, inequality and inflation. That is, assuming you have economic growth, there are nine possible combinations of increasing/decreasing equality, and inflation/deflation (over/under changes in the wage-rate) that will affect whether people still have access to the politically acceptable basket of goods:

    1-4. Rising/stable equality, zero or negative inflation

    When economic growth decreases (or maintains), not increases inequality. Any time it does this the tax rate can be reduced, because situations are improving.

    5/6. Increasing inequality/stable equality, inflation

    Conversely, increases in equality and inflation are reducing the relative position of people through no particular fault of their own, and the tax rate needs to rise to compensate for those losses.

    7-8. Increasing inequality, zero or negative inflation

    The situation where the rich are enjoying the benefits of growth, but where the welfare of others hasn’t decreased, and may even be improving. This is the case that seems to cause divisions in left/liberal positions, but it is arguably not a stable state; as the benefits trickle down, either/or both inflation will rise and/or equality will rise again.

    9. Rising equality, inflation

    In this scenario things might be improving or regressing, so there is no clear answer. However, it is also a situation likely to lead into slower growth as wage rates and inflation rise together, so other factors may be more important.

    Arguably the Howard government had that combination of growth, stable equality and zero-negative inflation that allowed them to reduce the tax rate. Arguably too, they didn’t reduce it enough when they had the chance, resulting in wide-spread middle class welfare. Certainly the Rudd government appears to be facing a different situation.

  39. Patrick says:


    An alternative is to target the spending and assets of the rich

    Also known as a GST – easily the most progressive tax available.

    and whack them with hefty death duties.

    also easily avoided.

    But such punitive taxation as you impose would not work, whatever the format. No-one taxable would live in Australia in such an environment, basically, and in short enough order nor would anyone else interested in more than subsistence.

    Thanks. I can understand that, although I still think that the free riders Bowles and Gintis were talking about was welfare recipients and not welfare providers.

  40. Don Arthur says:

    Andrew Leigh (at 33) – I’m curious, why do you think it’s worth constructing and using measures like the gini index?

    Is it because they measure something inherently important, or is it because you can use them in theories that explain changes in other variables?

  41. NPOV says:

    Patrick, explain how the GST can possibly be the most “progressive tax available”?
    I’d like to see some numbers to back such a claim up – you seem to be implying that the rich spend a far higher percentage of their income on goods and services than the poor.
    (FWIW, I think there are many good arguments in favour of a GST, but progressivity doesn’t strike me as one of them).

  42. NPOV says:

    FWIW, mentions a study that ranked the GST as -0.017 on the progressivity index, making it definitely regressive. However, the GST package as a whole was no less progressive than what it replaced.

  43. “Though Im still not sure if Rawls, Hayek, Lindsey and Norton can tell us whether Australias gini of 0.31 (or if you prefer, our 90/10 of 4.2 or our Atkinson One index of 0.17) is too high or too low.”

    I’m sure they can’t, because as Don’s point implies it does not of itself matter what any of the indexes are. It’s up to the people calculating these numbers to explain why the results are important.

  44. Andrew Leigh says:

    NPOV, my apologies. It was a link to the comparative inequality estimates compiled by the Luxembourg Income Study folks. See: If we’re talking cross-national inequality, it’s the best source.

    Don, I like the Gini because it has an intuitive explanation – it’s half the average income gap, expressed in terms of mean income. So a gini of 0.31 implies that the after-tax household income of any two random Australians differs on average by 62% of mean income. Mean household income in a recent ABS survey was $644 per week, so the mean income difference between two random people on the street would be $399.

    But I think your question was really about whether that number has special meaning to me, and the honest answer is probably that it only matters in the context of comparing it to Australia in 1981 (G=0.28), Sweden today (G=0.25) or the US today (G=0.37). Thinking more about this, I guess it implies that if measured inequality was the same in all places and times, I’d be stumped to know what to do when you told me the level of inequality.

  45. Patrick says:

    NPOV, sorry, my line didn’t quite make sense. A GST is a part of the most progressive solution available – that is a GST with a low-income credit.

    Poor people spend a far greater percentage of their income on goods and services, but it is so little money in real terms that they are cheaply and easily compensated through the transfer system. No matter how rich rich people are, however, they basically can’t avoid GST. Cutely, if they even try, they incur GST on their efforts :)

    So a ‘perfect’ quick ‘n easy tax reform might be to lower the top marginal rate to 30 per cent, increase the GST to 20 per cent (common in Europe anyway) and introduce a GST rebate for people earning up to $35,000 and progressively decreasing from there.

    That would be progressive in practice, even if on tax:reported income ratios it might look less progressive.

  46. NPOV says:

    Patrick, that sounds reasonable enough to me. In fact, any simplification of the tax system that reduces the tax avoidance (and the cost of tax compliance) and doesn’t make anyone earning less than the median income worse off works for me.
    Oh, I don’t think there’s a good justification for reducing the total tax take at this point in time.

  47. Don Arthur says:

    Its up to the people calculating these numbers to explain why the results are important.

    I think Andrew L is invoking the Banana grower’s response:

    ME: You sure grow a lot of these Cavendish bananas. How come?

    BANANA GROWER: I like them because they’re resistant to Panama Wilt disease. You can identify them by the size of the pseudostem …

    ME: Yes yes. But what’s good about bananas?

    BG: You can sell them.

    ME: Right. But why would I want to buy a banana?

    BG: I’m a banana grower not a mind reader.

  48. NPOV says:

    Andrew L, thanks for that. I was using, which seems a bit more comprehensive.

    What I actually find surprising is that there are no genuinely third-world countries with very low GINI co-efficients. Using the UN measure, the lowest recorded is that of Denmark, followed by Japan and Sweden. The only truly poor country with a low GINI is Ethiopia (30%). OTOH, the only country that isn’t a genuinely third-world country with a high GINI is South Africa (57.8%).
    The correlation between high levels of inequality and high levels of social unrest seems to jump out too, but accepted we can’t automatically conclude that it’s a causal relationship.

  49. Backroom Girl says:

    it only matters in the context of comparing it to Australia in 1981 (G=0.28), Sweden today (G=0.25) or the US today (G=0.37).

    My problem with measures of income distribution is that I think they conceal as much as they reveal. You see, I don’t at all know whether today’s Gini coefficient of 0.31 is ‘worse’ than 1981’s 0.28.

    This is because I think that there is so much more diversity in the ways that households put together their income now that you are more likely to be comparing apples with oranges than apples with apples. And having meaningful comparisons of disposable income depends on being able to compare apples with apples.

    One of the main problems as I see it is the ‘equivalence scales’ that are most commonly used to adjust household income to take account of household size and composition. As time has gone on, these have become simpler – possibly to make it easier to equivalise income and to do all of the poverty research that some of us are so addicted to. So most people now use the ‘modified OECD’ equivalence scale, which allocates a value of 1 to the first person in a household aged 15 and over, 0.5 to each other person aged 15 and over and 0.3 for each person aged under 15. This gives you a household equivalence factor that you can use to adjust your data on earnings, total gross income or disposable income. Other equivalence scales involve using the square root (ie a household of four requires twice as much income as a household of one) or even the cube root of household size (a household of eight requires twice as much income as a household of one).

    So that’s all simple then. The only problem is that there is no allowance in these simple formulae for the kinds of things that actually make important differences to standard of living – for example, home ownership or the number of people in the household who work. So the ‘poverty line’ for a household with no workers and living in their fully-owned home (eg most aged pensioners) is exactly the same as a same-sized household that pays private rent and has two people going out to work.

    All other things being equal, a society where household employment conforms closely to a single norm (eg 1950s Australia with one full-time adult earner per household or Scandinavian countries with two full-time earners per household) will have a more equal income distribution than a country like Australia now where there is great diversity of household employment patterns -all the way from jobless households on the one hand to households with 2+ full-time earners on the other.

    Now, I grant you that having too many jobless households is not a good look (though many people in the welfare industry nevertheless staunchly defend the right of single parents, for example, to be jobless), but for most employed couples in Australia the intensity of their employment participation is largely a matter of choice. And generally we pride ourselves on the fact that parents of children in particular are able to choose when and how much they want to work. So how can we get upset if a development such as that has been accompanied by more income inequality?

  50. Backroom Girl says:

    And don’t get me started on the quality of income data in Australia – something to bear in mind when comparing Australia with other countries, where the income data is more likely to be sourced from administrative collections (tax, employment insurance records etc). Organisations such as ACOSS who think that we have a poverty rate of somewhere between 10 and 20% would do well to remember that even the ABS has no faith in the quality of data in the lowest decile – they use the 2nd and 3rd deciles as the low-income comparison group in their own analysis.

  51. vee says:

    Can I get progressive defined? I can see classical liberalism, social/modern liberalism and social democracy all for the most part being progressive. I think the term progressive and the use of progressivism are poor choices of words.

    And I still prefer the terms neofusionism and neofusionist.

  52. Peter Whiteford says:

    HC said: “Consider two societies. Society A has 20,000,000 people with income $100,000 and 1 person with income $5,000. Society B has 20,000,000 people with income $2,000 and 1 person with income $5,001. I think Rawls says society B is better off which no-one with common sense would agree with.”

    If Rawls does say this or anything which could logically lead to this conclusion then it is ridiculous – but I’d like to see the evidence that this is what he says.

    Apart from anything else like the fact that this assumed distribution is completely unlike anything that happens in the real world, this example involves a comparison of two countries with very different levels of aggregate income. You are not comparing like with like.

    Now in the standard economic way of comparing well being using the Lorenz curve from which the Gini coefficient is derived it would be unambiguously clear that Society A has higher wellbeing than Society B because everybody in Society A would have higher real incomes than the corresponding person/household in Society B, even though Society A is more unequal (although it is not clear to me that this is necessarily true and one would need to do a calculation to confirm this).

    The Generalised Lorenz curves would not intersect see

    Now if the Generalised Lorenz curves for every country in the world did not intersect then we would be able to rank people in the world correctly but the fact is that in lots of high income countries and probably elsewhere, the Generalise Lorenz curves do intersect. This means that we have a problem of interpretation the average income of country X may be higher than the average income of country Z but it may be that the poorest 10, 20 or 30% of people in country X have lower incomes than the corresponding people in Country Z. If average incomes are quite close but the level of inequality is quite different then it is possible that more people in country Z are better off than most of the people in country X.

    The reason why people are interested in inequality (i.e. the shape of the distribution) is that averages can mislead.

  53. Peter Whiteford says:

    And to expand slightly, none of this tells us what the ” right” Gini coefficient is, but measures of inequality provide useful additional information for interpreting trends in average incomes. Wellbeing is a characteristic of individual people not averages.

  54. jc says:

    A couple of economists at the Dallas Fed took a different approach in determining inequality. They took the approach of measuring consumption patterns rather than income and found the US is far less unequal than what the income side of these studies suggest.


    ts true that the share of national income going to the richest 20 percent of households rose from 43.6 percent in 1975 to 49.6 percent in 2006, the most recent year for which the Bureau of Labor Statistics has complete data. Meanwhile, families in the lowest fifth saw their piece of the pie fall from 4.3 percent to 3.3 percent.
    Income statistics, however, dont tell the whole story of Americans living standards. Looking at a far more direct measure of American families economic status household consumption indicates that the gap between rich and poor is far less than most assume, and that the abstract, income-based way in which we measure the so-called poverty rate no longer applies to our society.
    The top fifth of American households earned an average of $149,963 a year in 2006. As shown in the first accompanying chart, they spent $69,863 on food, clothing, shelter, utilities, transportation, health care and other categories of consumption. The rest of their income went largely to taxes and savings.

    So, bearing this in mind, if we compare the incomes of the top and bottom fifths, we see a ratio of 15 to 1. If we turn to consumption, the gap declines to around 4 to 1. A similar narrowing takes place throughout all levels of income distribution. The middle 20 percent of families had incomes more than four times the bottom fifth. Yet their edge in consumption fell to about 2 to 1.

    The biggest concern about the bottom rung is to determine if their level consumption is impaired compared to the upper levels of the economic ladder. If that’s the case consumption figures are showing they aren’t as badly off as the income stats present.

    So the real issue should not be income inequality. These studies ought to be premised on consumption. It’s not how much you have, it’s how much are you comsuming compared to others that really counts…. and if tht level of consumption is an accpetable standard.

  55. NPOV says:

    Jc, interesting…but you can only consume in accordance with what you earn.
    If consumption has been rising considerably faster than income, this can only be due to increasing levels of debt. But all debt needs to be repaid eventually, so clearly this isn’t a sustainable pattern. Although it would be interesting to measure the degree to which the lifestyles of less well off are funded by financial institutions writing off bad debts – i.e., the rich subsidising the poor.

  56. jc says:

    Not necessarily, N. Consumer goods are on the whole much cheaper than what they ever were for instance. Income can also be supplemented with government support transfers.

    The point is that the income gap may not be telling the story if ones prime concern is what well being a giving level of consumption will bring to individuals. I presume that ought to be the main objective of transfer payments (and not the envy issue).

    The Dallas fed’s study is significant for Australia as we are supposed to be far less unequal than the US.

  57. NPOV says:

    Well the GINI co-efficient is supposed to be measured based on post-tax income – but I don’t know if it includes government benefits too. I’d argue that it definitely should to be meaningful.

    As far your suggestion that Australia is somehow only “supposed” to be far less unequal than the US – I don’t think it’s particularly controversial to note that their poorest are far poorer than almost anyone in Australia and their richer are far far richer than anyone in Australia. But my personal experience spending time in the U.S. certainly backs it up – it’s not at all uncommon to be one minute driving through some very impressive up-scale neighbourhoods that make Toorak look shabby, but the next be driving through slums that wouldn’t look out of place in any number of third world countries. I don’t think I’ve ever had that experience in Australia, and certainly not in Melbourne.

  58. jc says:

    I dont think its particularly controversial to note that their poorest are far poorer than almost anyone in Australia and their richer are far far richer than anyone in Australia.

    Yes , but are consuming more than the income gaps suggest? And are they consuming less than an acceptable level?

  59. Patrick,

    Have a read of the article I linked to – it’s terrific – perhaps you already have. Anyway their point is that people hate free riders on the public teat of welfare. But their larger point is that people strongly support assisting the deserving poor – that there’s no evidence of a ‘tax revolt’ on that score. By implication they resent those who don’t bear that burden with them – though I think I’d concede that they resent the free riders of the undeserving poor more.

  60. Patrick says:

    Thanks Nick – you are right, I did. I’m comfortable with your summary at 59. But as I pointed out, unless NPOV meant something other than what s/he said, then as HC has also pointed out, it didn’t make any sense.

  61. hc says:

    Peter Whiteford, My understanding is that Rawls pursues maximin objectives – maximising the welfare of the worst off group.

    My example stretches that idea to the limit to show that maximin is a silly social objective.

    There are tradeoffs – you might accept a small reduction in the welfare of the poor if the welfare of the rest were massively increased.

  62. Peter Whiteford says:


    My understanding is that Rawls was in favour of complete equality, except where inequality meant that the most disadvantaged was made better-off. It is interesting that in the index to A Theory of Justice, there is no reference to income, wealth, welfare or redistribution, although there is discussion of fair wages.

    In fact, on page 157 of my edition there is specific reference to the type of argument tht you mention, but Rawls say that the maximin requires maximising ” the long-term prospects of the least advantaged”. He specifically rejects the argument that the gain of one penny would offset the loss of billions for the rest of society, saying that ” for as we raise the expectations of the more advantaged, the situation of the worst off is continuously improved … ” p.158.

  63. Yobbo says:

    Don, how can you argue that lefties like Mel are NOT motivated by envy when they support measures like a 95% tax?

    They certainly can’t be motivated by trying to increase well-being, because as patrick rightly noted, all such a measure would do is cause all the rich people to emigrate, which would leave everyone worse off.

    So, really, people who support such a measure are either

    1: Primarily motivated by a hatred of rich people


    2: Completely unable to see the obvious effect that their policies would have. In other words, borderline retarded.

    Why would libertarians want to join forces with such people?

    Not having a go at Mel personally here, but his view is a fairly common one shared by those on the “progressive” left in parties like the greens, which is who Don is suggesting Libertarians should team up with.

    The same people who propose wage ceilings on CEOs and things like that.

  64. hc says:

    Peter, I understand that Rawlsian argument but it is a cop-out. In the case you mention inequality provides external benefits to the poor since their expectations improve – it is covered by the difference principle.

    Ignoring such externalities (they could be used to justify virtually any distribution) you are back to maximin and it does not work.

  65. Andrew Leigh says:

    Don, your banana line (47) is beautifully cruel. Put another way, changes and differences in inequality are the main reason we study it. I suspect the same would be true of many other economic or social indicators too. Trust and GDP are also more meaningful in changes/differences than in levels.

    Patrick, I reckon the best response to the envy argument comes from a paper by Branco Milanovic.

    The use of value-laden terms like envy is supposed to shut us up by basically telling us that only green-eyed envious monsters are likely to covet other peoples money. Let us grant them this point: envy is not nice. But if most, many, or a significant percentage of people do feel envious of other peoples money, this is the only thing with which economists need to concern themselves. (And recall, that envy simply means that other peoples income enters our utility function). Envy, whether economists approve of it in private or not, must enter into their analysis. Perhaps, ethicists and religious ministers would disprove of such practices, and we leave the field open to them to improve the human race. When the ministers have done so, economists should go back to their assumptions, and wipe out income of others from their welfare
    functions. But not before we are informed that envy has been rooted out.

  66. Pingback: Andrew Leigh » Blog Archive » Dreaming of inequality

  67. Peter Whiteford says:


    What I am saying is that Rawls explicitly rejects the interpretation of the maximin principle that you accuse him of – i.e. you and he are in agreement. I don’t undertand how you can say this is a ” cop out”.

  68. NPOV says:

    AIUI, wage ceilings on CEO’s – specially as a ratio between employee and employer salaries, have been tried in Norway. Whether they’ve had much success in reducing inequality without compromising overall prosperity I’m not sure.

    And Yobbo, I don’t believe 95% tax rates are necessarily motivated by envy. For a start, I would perfectly happy with an adjustment to the tax system that meant I was paying more tax than I am now – provided that the corresponding level of state-provided services improved (and not necessarily services that I have any immediate need of). I agree that a 95% rate would be pointless and probably counterproductive, but I wouldn’t be so sure that it would be completely ruinous, depending on how it was implemented. It may, for instance, encourage business owners to plough profits back into growing their businesses rather than taking home huge pay cheques to spend on imported luxury goods etc.

    At any rate, yes there is a subset of the Left that is obviously anti-capitalist, even anti-economic-growth (which surely then disqualifies one from the “progressive” tag). OTOH, there’s a subset of the libertarian right that believes that any possible attempt by the government to improve the livelihood of the least fortunate among us is a Bad Thing (TM) or that we’d be better off with no government at all, even individually protecting our own lives and bit of turf by whatever means necessary. The fact that those two subsets will never see eye to eye surely doesn’t invalidate the possibility of better collaboration between moderate-social-liberals and moderate-classical-liberals.

  69. Joshua Gans says:

    Wow, these arguments are funny and philosophical. I thought this issue was: is current inequality too high or too low?

    In which case, if you can answer the following question you will know the answer. Is there a person with an above median income you can identify (say, James Packer, for example) and someone with an below medium income you can identify for which you would, taking any criteria you want, want to ‘Robin Hood style’ give $1 from the relatively rich person to the relatively poor. If so, there is too much inequality. If not, there is too little inequality.


  70. NPOV says:

    Joshua, that doesn’t really answer the question though, because my concern is more “how did we end up with a situation where James Packer has more money than he could ever possibly spend, while there are simultaneously thousands of people with no money and no place to call home through little or no fault of their own?” Wanting to taking a dollar from him and giving it to some random homeless person hardly seems like a useful way to tackle the problem.

  71. Patrick says:

    Joshua, that doesn’t even begin to answer the question, because what if you do so only to discover that the consequence is the relatively well-off person dropping dead? Or even the relatively poor person?

    In fact, there seems to be a consensus from everyone except the whacky Mel that a certain degree of greater inequality than you posit as acceptable is in fact acceptable and desirable. So you probably need some serious philosophical justifications of your own before putting that forwards as an answer.

    And as NPOV points out, you can’t actually address the question without trying to address what is possible and what are the consequences of both a) state A, b) state B, and c) trying to move from state A to state B.

  72. NPOV says:

    Having said that, the obvious conclusion to me is that yes, Australia still has considerably more inequality than I personally feel comfortable with. However if someone could find a solution that hugely improves the lot of our least fortunate while quintupling the wealth of our richest citizens, I wouldn’t necessary have a problem with it. But I want to see a solution implemented in the next 3 or 4 years, not the next 30 or 40.

  73. Backroom Girl says:

    Perhaps I’m just an old fogey, but as I get older I’m less and less convinced that inequality (especially when it is as imperfectly measured as income inequality) is in itself a bad thing.

    I’m interested in the fact that so many commenters on this blog and others focus on CEO salaries and the like, rather than the more important changes that have been occuring amongst the population as a whole. From my point of view, increased inequality of market incomes is the flipside of many societal changes that people on both sides of the political fence would agree are positive:

    * increased education, leading to growth in higher skilled, higher paid jobs
    * increased employment among women, especially mothers
    * alongside that, a social security system that allows single mothers to raise their own children and a significant degree of choice for low-income partnered parents as to how much labour they supply
    * easier divorce
    * no doubt other things that I can’t think of at the moment

    Given that all of those changes lead us inevitably in the direction of greater market income inequality, I think the truly remarkable thing is that the Gini coefficient of equivalised disposable income has only increased from 0.28 to 0.31 in the last 30 years of so.

    My final point is that, even if we were able to somehow agree on an ‘optimal’ level of inequality, I don’t think I would be in favour of many of the policy solutions that might achieve such optimality. For example, if you wish to significantly change the degree of inequality of market income, you would need to change not only the distribution of wages but also the distribution of hours worked within households. So you could either make it harder for mothers to work at all (back to the good old days of the 50s) or you could make it much harder for mothers to work part-time. You would also want as many single parents to work full-time as possible and make it much harder to get disability pensions to increase the financial incentive for those people to join the job market. Better ideas, anyone?

  74. Don Arthur says:

    What I am saying is that Rawls explicitly rejects the interpretation of the maximin principle that you accuse him of – i.e. you and he are in agreement. I dont undertand how you can say this is a cop out.

    Peter – thanks for references to Rawls.

    While it’s true that Rawls was aware of the objection Harry is making not everyone agrees that he dealt with it successfully. For example, Harry Brighouse argued that a society that applied the difference principle “might have to redistribute vast resources to the disabled, without the prospect of any benefit to the economy as a whole.”

    Brighouse suggests modifying Rawls’ theory to prevent the problem arising.

    It’s worth remembering that Rawls’ theory was about deciding between the principles that determine distributions, not distributions themselves.

  75. I suspect what’s important is not that ‘everyone should have the same/income levels/wealth should not be so unequal’ but that each should have enough. Let’s call it a ‘doctrine of sufficiency’.

    Most disputes about inequality are misconceived: they are not really about inequality per se. No one cares much about the difference in wealth between lawyers and bond traders, say, even though while both have expensive educations and considerable technical skill bond traders often earn incomes orders of magnitude larger than lawyers. What they care about is that some people live in demonstrably bad conditions due to poverty, or that their poverty prevents them from functioning.

    I do not pretend working this though is simple. Calculating the size of an equal share or an ideal income distribution is plainly much easier than determining how much a person needs in order to have enough. I think Backroom Girl is onto something in her comment, though. Working out ‘enough’ may well be easier than working out ‘optimal levels of inequality’. I also think ‘enough’ is closer to Hayek’s ‘assurance of a given minimum of sustenance for all’ than anything in Rawls.

  76. Backroom Girl says:

    I’d agree with that as an appropriate objective skepticlawyer – though we could have another very long discussion about how much would be enough.

    One of my problems with the idea that we should somehow maintain, if not improve, the income of the people at the bottom of income distribution with those in the middle, let alone the top, is that I don’t see how you could do that in practical terms. To get back to the issue I raised earlier, in the olden days while the households at the bottom had no jobs, the households in the middle had one full-time income earner. Now, the households at the bottom still have no jobs but the households in the middle have around one and a half income earners. So how could you really expect them to keep up? And how would you go about maintaining the relative position of the households at the bottom – index their pensions and benefits by one and a half times the growth in average earnings or something?

  77. NPOV says:

    Everyone having “enough” is certainly the more important part of the equation, but ensuring that wealth is not too densely concentrated is also important.
    I don’t think the latter problem is significant in Australia, or indeed in many developed countries, but it certainly is in many third-world countries.

    Further, the fact that so many people in Australia live in fairly desperate conditions is only shameful because we know we have the resources to enable them to live far better. If Australia was just a generally poor country, then it would be far less of an issue.

  78. I don’t think you can do it, BG – for exactly the reasons you enumerate. I do think some of Amartya Sen’s ideas are interesting for filling out the concept of ‘enough’, although inevitably it leads to measuring poverty in absolute, rather than relative terms. His recent attempts to extend the argument to the developed world really don’t work very well, and start becoming very statist indeed – something I think he realises.

    I discuss this in the paper I’ve written, which – irritatingly – I can’t put on line until it’s handed in. I’ll make sure it goes up on Catallaxy after I’ve dropped it off at the Examination Schools.

  79. James Farrell says:

    Helen’s point about ‘enough’ is exactly how Harry Frankfurt (the Bullshit guy) puts it, right down to the italics.

    But as soon as you concede that ‘enough’ can’t be defined absolutely, it’s almost a circular argument. A household’s conception of enough is subject to rising expectations, and is moreover in large part arrived at by comparisons with other households. Any attempt to deny this, or dismiss it as unhealthy ‘envy’, flies in the face of human nature.

    Quite apart from these considerations, inequality and poverty are correlated in practice. Show me two countries with the same per capita GDP, and I can guarantee that the one with the worse Gini coefficient is the one where more kids go without shoes. What makes it tricky is that inequality often rises with rapid econonic growth, which obviously also works against poverty. Often doesn’t mean always, though.

  80. Was waiting for someone to notice the Frankfurt reference ;)

    I do think you can give quite a bit of content to ‘enough’, though – especially when you consider Sen’s stuff on capacity and ‘functionings’. There are actually more complex empirical issues in the way of determining an appropriate level of inequality than there is in using, say the, HDI to work out what ‘enough’ actually may be.

    It would be easier if inequality was only a function of incentives – all we’d have to do is work out how much inequality we need in order to give people appropriate incentives (although this Hayekian thinks that would actually be pretty hard, and enforcing it would be even harder). People would simply leave, as Yobbo and Harry point out.

    But it’s not just incentives – there are a billion other things in the mix, as BG points out. Getting beyond ‘enough’ is going to be very, very difficult.

  81. Better HDI figures also correlate pretty tightly with economic and civil liberty, though – for a while there the EFW people used to correlate their data with the HDI. I suspect that’s more useful than GINI, which is not very revealing about developing countries where nearly everyone is pretty poor.

  82. NPOV says:

    I would think GINI co-efficients are very relevant for poorer countries, because they so precisely demonstrate that the poverty that most of the citizens there experience need not be nearly so bad, if the wealth was spread around a little better. I’d also argue that poor countries with high GINI co-efficients would stand a far better chance of climbing out of poverty if the wealth they had wasn’t so concentrated in the hands of a few that make little effort to do fairly little productive with it (“money is like manure” and all that). However I don’t claim that the solution for poor countries with high GINI co-efficient is a comprehensive tax and redistribute scheme, as those high GINI co-efficients are fairly likely to be the result of government practises that deliberately favour the few. The evidence that what poor countries need most is economic liberalisation (i.e. less government control) is just about as solid as you could ask for.

  83. Don Arthur says:

    skeptic – I agree, I think Amartya Sen’s capability approach is useful. And Martha Nussbaum’s work is worth reading too.

    BTW: Why does Sen write like that? Such wonderful ideas buried in such horrible language. I’m sure your explanation of his ideas would be easier to understand than his own.

  84. True, NPOV – although trying to convince some of my (since we’re not being snarky here) ‘progressive’ friends that free trade and less government control is a good thing in the developing world is, ahem, frustrating. Let’s just say I’ve made Tim Harford a lot of money through either recommending or giving away copies of The Undercover Economist.

    The reson I like to see the EFW indices correlated with Sen’s HDI is because it forces us (ie people in the first world) to think about what poverty actually means in practice, and what has logical priority when it comes to getting rid of that poverty (and the rotten things that go with it, like not having clean drinking water).

    BTW does NPOV that stand for ‘Neutral Point of View’? If it doesn’t, tell me now or I’m going to get into the habit of thinking of you in that way forever!

  85. Read The Argumentative Indian Don – it’s beautifully written. I was furious when I read it, after all the shiteful stuff of his I’ve had to trawl through over the years. He can write when he wants to. Some of his stuff on democracy and famines is pretty readable, too. In fact, his case for democracy based on absence of famine is actually the strongest argument for democracy I’ve ever seen.

    Nussbaum I find a bit problematic, and I hope she doesn’t suck Sen in too much. Her list of ‘capabilities’ is incredibly close to John Finnis’ account of ‘primary goods’, and her rationale for them is strongly perfectionist (of the ‘we ought to do x for y moral reasons) variety.

  86. Pingback: Will Wilkinson / The Fly Bottle » Blog Archive » Epiphenomenal Inequality

  87. Ken Parish says:


    I’m just wondering why you’re so negative about Finnis? His ideas about homosexuality, abortion, euthanasia etc are certainly uncongenial, but seem to me to me at least more a result of his conservative Christian viewpoint than a logical consequence of his natural law conception of primary goods (friendship, religion, knowledge and aesthetic appreciation, bodily life, skilful performance, marriage, and practical reasonableness). If you strike out “religion” as a mere artefact of his own religious beliefs, and rewrite “marriage” as “family” including single parent and gay and lesbian families, it strikes me as an eminently workable latter day elaboration of the Kantian Categorical Imperative, and arguably a more tenable and satisfying construct than the utilitarian focus on pursuit of happiness as the sole primary good.

  88. NPOV says:

    Skeptic – the “POV” bit is right…I think initially I intended the N to stand for “New”, but who knows…no point of view? Nobody’s point of view? Nonsensical point of view?

    And for what’s it worth, while I accept economic liberalisation is critical for poor countries I don’t buy that this implies lack of any sort of regulation over labour markets or environmental/health controls etc. Nor do I accept that further economic liberalisation is necessarily the best way to alleviate poverty in already wealthy countries (though certainly there are particular areas where this is probably the case). I’d suggest this is not too far removed from the position of most of the progressive intellectual Left these days – your friends that you call “progressive” sound anything but (more “Old Left”).

  89. I’m reading Finnis’ material on immigration for the paper on perfectionism I’m writing now, and calling it unpleasant is an understatement. I’ll give you a little taster:

    Confronted by the grave warnings thus issuing from courts of great pan-European authority, citizens of countries whose Muslim population is increasing very rapidly by immigration and a relatively high birthrate may ask themselves whether it is prudent, or just to the children and grandchildren of everyone in their country, to permit any further migratory increase in that population, or even to accept the presence of immigrant non-citizen Muslims without deliberating seriously about a possible reversal – humane and financially compensated for an incentivised – of the inflow.

    This is from a paper forthcoming in Hare & Weinstein, Extreme Speech, CUP 2008.

    I must confess that I took a screen capture of the relevant page from the BNP website, printed off a few copies, and shared them around the rest of the Oxford jurisprudence seminar on the day we discussed that paper. The more I see of perfectionism, whether of the liberal or the Kantian variety, the more I loathe it. That doesn’t mean I won’t recruit its arguments – I’m currently giving Raz’s service conception of authority substantive content using Hayek’s consideration of co-ordination and planning problems in Law, Legislation and Liberty (which is why I’m up at an ungodly hour now). Raz’s account of authority deals beautifully with the consent issue, and also provides a sound normative basis for a small state. Hayek fills the empirical gaps very nicely.

    No, I think Finnis’ religiosity is embedded so deeply in his account of primary goods that disentangling them is well nigh impossible. Apart from that, he’s a strong perfectionst, which – for the reasons I’ve outlined in various places over at Catallaxy – I’m not.

  90. Ken Parish says:


    Finnis’s real world application of his primary goods list is certainly obnoxious, but that doesn’t of itself invalidate the concept itself if rewritten as I suggested above. As for perfectionism, well again it’s possible to adopt some version of Finnis’s conception of primary goods without accepting the illiberal deployment of them to justify coercive imposition of any particular notion of “the good” by the state. Of course, that largely strips Finnis’s theory of its intended purpose of providing a “natural law” justification for state action/law and confines it to the personal realm of moral philosophy, but I don’t have a problem with that even though Finnis certainly would. Anyway, I’ll be interested in reading how you graft Hayek onto Raz.

  91. Don Arthur says:

    skeptic – Thanks for the tip. It sounds like just the book I’ve been looking for.

  92. Fred Argy says:

    Don, I have been away for a few weeks and am just catching up with my reading. I want to say how much I enjoyed reading your last two contributions, especially your insights into Hayek. It’s good to see you are back in action.

    As a welfare economist, I am mainly concerned with the change in distribution stemming from an economic reform rather than with the rights or wrongs of any given level of inequality. But, being an egalitarian, the moral issues surrounding inequality intrigue me and I am looking forward to reading the great debate you have triggered on inequality.

    At this stage, at the risk of repeating what others have said or questioned, let me say that I do not accept that labour markets distribute rewards fairly by paying people what they are worth. There is always some monopoly power, people have unequal access to information, training and networking, there is much ignorance and uncertainty, exorbitant CEO salaries (500 times the average wage of his own workers) are the product of corrupt insider markets, and so on.
    Moreover, people bring with them to the labour market not only their own innate qualities but their starting positions – the varying circumstances of birth and upbringing – which are radically unequal. In other words if one looks the whole life cycle of a person it is untrue to argue to that people have the same opportunities to develop their full education potential and abilities.

    That is why, as an economist and egalitarian, I am a strong believer in a big government role in levelling opportunities (but not in passive redistribution other than through a safety net).

  93. NPOV says:

    Fred, would you then argue that if the government could successfully ensure that all markets were competitive, and that everyone had near equal access to “information, training and networking”, that the labour market would then fairly pay people what they are worth? Can we ever get past the fact that some people are just better at convincing others to hand over their money, regardless of the genuine worth of the products or services they are offering?
    If not, then why not support passive redistribution?

  94. jc says:


    Isn’t your argument the traditonal union one that suggests wages are basically indetermiant, meaning people don’t know whow much they should earn.

    exorbitant CEO salaries (500 times the average wage of his own workers) are the product of corrupt insider markets, and so on.

    Three things:

    1. Can you offer up one single example of that happening at any public listed firm?

    2. Would you prefer to the bad old days when the school tie and family background meant more than the return a CEO provided his shareholders?

    3. Were the old days better when 50 or so families ruled the economic roost like they do in Sth America?

  95. Backroom Girl says:


    Welcome to the conversation. Like you, I think that my approach to achieving greater equality of outcomes would be to do whatever I can to improve equality of opportunity. Supplement this with an income support safety net that is designed as well as possible to encourage and help people to support themselves, while ensuring an acceptable minimum standard of living for those who cannot. As well as that, allow people as much choice as possible in how they live their lives, including how much they want to work for pay and how much time they want to spend on other things that they value.

    I guess the theme that I have been trying to develop in my contributions to this discussion is that too much focus on the distribution of disposable income (ie post taxes and transfers, but before spending) can actually lead you astray in lots of ways.

    Let me give you a simple example from the most recent figures on poverty that came out from ACOSS. This purported to find that around 10% of Australians were below the 50% median poverty line. When you look at the numbers more closely you discover that age pensioners who either own their own homes or are paying low rent for public housing are in poverty, while age pensioners who rent their homes in the private market are not in poverty. This is because the latter group have a higher measured income due to the fact that they attract rent assistance on top of their pension. And here was me thinking that home ownership in old age was a good way to protect your standard of living – apparently I was wrong.

  96. melaleuca says:

    Patrick says:

    “But such punitive taxation as you impose would not work, whatever the format. No-one taxable would live in Australia in such an environment, basically, and in short enough order nor would anyone else interested in more than subsistence.”

    That is an observably false statement. If it were true no able bodied person would live in the high-taxing European countries. Moreover if I remember some of the recent Pew study findings correctly the overwhelming majority of Westerners do not want to live in relatively low taxing America.

  97. Thinking in old ways says:

    This has been an interesting thread. Two contributions from me the first on the idea of equality the second on some data issues I know that was not the purpose of your post but since things have gone down that path I might follow.

    Don, while I recognise that Sen is usually hard to get on top of he does make some interesting and quite clear observations in Inequality Re-examined which get to the point of some of the issues you raise. Centrally he argues that the philosophical debate is not so much about inequality but rather a struggle about which equality is most important. A common characteristic of virtually all the approaches to the ethics of social arrangements that have stood the test of time is to want equality of something (preface pg ix) .

    This is I think a useful starting point for your discussion about principles what is the equality that people are interested in (and to what degree might achieving it require some other equalities to be curtailed)? Is it as suggested by BG that the focus should be on equality of opportunities but what then of the individual endowments people have does equality of opportunity mean compensating for these. Can we have equality of opportunities for children when parents have different levels of resources or when different parents choose to utilise these resources in ways which benefit (or otherwise) their kids. Todays SMH has the classic first generation story

    What happens if we take up Jamess point and settle on a ratio of 5 times and the person who gets 5 times the income only spends half of that for a few years and invests the rest and starts getting an income from that as well. Or the person is a consultant or lawyer who is willing to put in an 80 hour week can we reject the idea that they can then get ten times the income rather than five. What if they partner with someone who also earns five times the amount of the minimum and their household gets now 15 times the income and that is even before the returns on the money they are saving/investing.

    What do we do in a taxation system when two people get the same hourly rate but one works 30 hours a week and the other 60 hours do we progressively tax the person who makes a greater effort?

    Which of these bits do people consider fair or unfair or anti-egalitarian?

    Going back to the question of how much (income?) inequality is too much it is always worthwhile looking at some of the data.

    According to the ATO the top 5 percent of personal income tax payers in 2005-06 have taxable incomes above $102,200 lets say about 4 times the minimum wage the top 1 per cent have taxable incomes above $210,000 say 8 times the minimum wage.

    So whose incomes are people concerned about? Is it unrealistic to consider that the value of some peoples work is 4 or 8 times the minimum wage?

    Or are people just concerned with the 1 per cent at the top the 94,000 people with an average taxable income of $442,000 (and total income of $468,000) who get around 9.4% of all taxable income and pay 16.5% of personal income tax. (The top 5% of all taxpayers pay 32.4% of all personal income tax.)

    Certainly the top 1 per cent seems to be where much of the action is in terms of changes in the income distribution using Andrew Leighs data they seem to have increased their share of income from 4.7% in 1982-83 to 7.2% in 1995-96 and 9.4% today (in contrast the 96th to 99th percentile share edged up a bit in the first period from 10.8% to 11.5% and then down a little to 11.4 per cent.

  98. NPOV says:

    TIOW – as I said before, my take on it is not so much that we should be concerned that there are people earning 10 or more times the minimum wage, but rather that we simultaneously have people earning that sort of money while we still have a large number of people that are earning next to nothing, homeless, destitute etc. etc. If every single person in Australia was in a household earning enough to pay for accommodation, food, clothes, education, healthcare, transport and a bit left over for entertainment purposes, then it wouldn’t bother me if there were people earning 1000 times the minimum wage (providing of course they were earning it through legal means, and that there was a range of people earning 900x, 800x, 700x etc. etc. Having most of the population earning between 1x and 10x the minimum wage, but a tiny fraction earning 1000x times, with nothing in between, would strike me as highly unstable).

  99. Don Arthur says:

    TIOW – I like your question about natural endowments. Does equality of opportunity mean compensating for these?

    One of my favourite answers to that question is Elizabeth Anderson’s. In What’s the Point of Equality? she writes:

    Recent egalitarian writing has come to be dominated by the view that the fundamental aim of equality is to compensate people for undeserved bad luck–being born with poor native endowments, bad parents, and disagreeable personalities, suffering from accidents and illness, and so forth. I shall argue that in focusing on correcting a supposed cosmic injustice, recent egalitarian writing has lost sight of the distinctively political aims of egalitarianism. The proper negative aim of egalitarian justice is not to eliminate the impact of brute luck from human affairs, but to end oppression, which by definition is socially imposed. Its proper positive aim is not to ensure that everyone gets what they morally deserve, but to create a community in which people stand in relations of equality to others.

  100. NPOV says:

    Ending oppression is certainly part of it, but it’s not the whole story: there’s also Nelson’s “a nation should be judged not by how it treats its highest citizens, but its lowest ones”. As an example, the way that the U.S. treated the victims of hurricane Katrina should be enough to make any American ashamed of their country. The condition of most of New Orleans even years after the event was (and in parts still is) pretty shocking, and strikes me as a symptom of a nation that never seemed to worry too much about inequality of wealth (despite it being the source of many civil rights movements that led to much of the political equality that various groups take for granted today).

  101. Backroom Girl says:

    “If every single person in Australia was in a household earning enough to pay for accommodation, food, clothes, education, healthcare, transport and a bit left over for entertainment purposes”

    NPOV – So you seem to be subscribing to skepticlawyer’s ‘doctrine of sufficiency’, is that right? But I’m interested in your use of the word ‘earning’ – where do people who can’t earn or simply aren’t interested in earning fit into the picture? Are you in favour of Government creating a job for everyone and making them do it?

    And given that many people would argue that jobs at current minimum wages aren’t sufficient to provide people with the kind of comfortable lifestyle that you envisage, you seem to be suggesting that both wages and presumably social security payments should be significantly higher than they are now. Since I don’t believe that you can do this simply by taxing the highest income earners at 95%, presumably most taxpayers would have to pay considerably more tax in order to pay for those increases.

  102. NPOV says:

    I would think the main thing preventing someone living a moderately comfortable lifestlye on a minimum wage would be rental rates, so on that basis there is probably an argument for more public housing. I’m not sure that most taxpayers would have to pay considerably more in order to support this.
    Further, there’s a difference between temporary hard times (say, a year or so) and being stuck in a poverty trap. In fact in an ideal world everyone would start out at a minimum wage and then progress up as far as their ambitions/talents/desires took them.

    As for people that “simply aren’t interested in earning” – I’ve never been convinced this would be a problem you could solve by making it harder for such people to draw on the public purse. It also struck me that there were far more people with genuine need for assistance being hurt by lack of government support than there were dole bludgers intent on living their ideal lifestyle without having to bother doing useful work.

  103. Verdurous says:

    Backroom Girl,

    Perhaps Im just an old fogey, but as I get older Im less and less convinced that inequality (especially when it is as imperfectly measured as income inequality) is in itself a bad thing.

    What a remarkably brave and revealing comment. Inequality (whether it relate to income or individual/household wealth) means a power imbalance. Powerful people do not always act charitably towards the vulnerable (slight understatement there). To accept a high degree of inequality in exchange for some mindless goal of exponential increase in consumption seems utterly mad. It has already been demonstrated here that most enviable nations appear to have low gini co-efficients. In econo-speak there are sure to be large externalities that favour more equal societal arrangements and lower rates of civil strife.

    Some here have suggested that we might be able to identify an “optimal” level of in/equality. This is in itself an absurdity. There is no optimal level of inequality any more than there is an optimal level of shutter speed on your DSLR. It comes down to values, preferences, and deeply held philosophical beliefs. Drop the search for optimality.

  104. melaleuca says:

    One point that I don’t think anyone has raised above (sorry if I’ve something) is the extent to which a massive concentration of wealth corrupts democracy and turns it into something closer to a plutocracy. An individual or organisation with infinitely deep pockets has access and influence not available folk of modest means. This manifests itself in both crude and subtle ways:

    * a crude example is the way cashed up developers have managed to capture local councils- see this for example

    * a more subtle example is the $5,000 a seat meet-the-premier fundraisers held by our political parties- see here for example

    I’m not arguing that no-one should be be rich, but I do question stratospheric discrepancies in wealth that can’t possibly reflect the differential value of individual contributions to the community, let alone be socially just.

  105. Backroom Girl says:


    Thanks so much for the backhanded complement. I don’t think I have anywhere advocated trying for a high degree of inequality – I was really agreeing that it is almost impossible to work out the ‘best’ level of equality/inequality and also arguing that perhaps in the end it is possible to put too much weight on inequality of (imperfectly measured) income.

    Since you suggest that we drop the search for optimality, I guess that means that for you anything short of a Gini coefficient of 0 would have to be seen as failure. For me, I’m happy to stay living in the real world.

  106. NPOV says:

    BG, it’s hardly controversial to suggest that it’s “possible to put too much weight” on just about anything. I’d be more interesting in knowing whether you thought Australia’s various political parties put too much or too little weight on addressing inequality in their policy platforms. And of course, whether you thought the policies were actually likely to help, or make matters worse (e.g. by putting our overall prosperity at risk).

  107. Verdurous says:

    I was really agreeing that it is almost impossible to work out the best level of equality/inequality

    Perhaps then we share a similar position on that front. I would probably suggest that being at or near either extreme of the Gini (1 or 0) is likely to be associated with severely dysfunctional societies. Where exactly we should sit comes back to fairness, justice, values and so on rather than any mathematical calculation that might seek to find an optimal position. You may have not suggested the latter, but economics as a discipline often heads off along similar dead ends.

  108. Patrick says:

    Fair enough, Verdurous, but surely there is some valid economic input that no-one interested in ‘fairness, justice, values and so on’ would want to go without. An example would be if there was a given level of inequality below which net welfare was likely to decrease, or a certain level above which net welfare was likely to increase by a certain factor, or indeed whether particular steps which were purportedly aimed at ‘fairness, justice, values and so on’ actually caused perverse effects…

  109. NPOV says:

    I’d like to see the evidence that a GINI of 0 is likely to be associated with a severly dysfunctional societies. What is true is that various attempts to achieve “perfect equality” have backfired horribly, but I’m less convinced that the end result, were to it be even possible, would necessarily be “dysfunctional”.
    Of course, I can think of a lot of reasons to be wary of such a society. It would surely be rather boring, for a start.

  110. FDB says:

    NPOV – that depends. If you don’t factor the Morlocks into your calculations, no problem!

  111. Yobbo says:

    That is an observably false statement. If it were true no able bodied person would live in the high-taxing European countries. Moreover if I remember some of the recent Pew study findings correctly the overwhelming majority of Westerners do not want to live in relatively low taxing America.

    Mel, a lot of those people aren’t earning big dollars. Not many people with multi-million dollar salaries choose to live in European countries with high tax. Some might physically live there, but incorporate their companies elsewhere to evade tax, and even take up residency in places like Monaco for certain portions of the year.

    Either way the effect is the same, their home country is not benefitting from their wealth because it is moved offshore.

    And America isn’t even close to having the lowest tax of developed countries. There is no taxation pressure to move there, that’s for sure. In fact many Americans also emigrate to tax havens like the Bahamas and Monaco for the same reason.

  112. jc says:


    You keep hearing about how expensive London real estate is right? It’s just not Arabs buying there or the result of the banking belt. Lots of Europeans etc. take up residence in london to escape tax. The UK had or still has a deal going whereby you can earn income tax free provided it’s held offshore and you only pay tax on domestic UK income. Don’t know if that’s still going. Monaco is even more liberal and lets not even mention Switzerland or Luxembourg. Tax evasion is huge in Europe. Swiss banks apparently hold US$30 trillion dollars in tax avoided money which is not all saddam’s money..

  113. Backroom Girl says:

    NPOV, Verdurous, et al

    At the beginning of all this discussion, Andrew Leigh admitted that he didn’t really know what the ‘right’ (presumably optimal) level of inequality was. I don’t either. While, like Patrick, I can accept the idea that there is a point somewhere at which movement in either direction is likely to produce more harm than good, I have no idea how you would go about discovering it. But I also doubt whether it is a single immovable number, which can always be meaningfully compared across countries or across time.

    I remain sceptical that Australia was a better, fairer, more equal country back in 1981 than it is now, even if the Gini was lower then. I suspect that if we had the data we might well find that the 1950s Gini was even lower than in 1981 and I’m even more sure that Australia was not a better place back in the 1950s.

    So should I feel guilty as a white middle-class well-educated working woman for helping to create the more unequal society that we have now? From my point of view, Australia would not be improved if women like me were put back into the kitchen, but I have no doubt whatsoever that the Gini coefficient would go down. That is why I ask people what they would do in order to achieve the outcome that they view as better (though I notice I don’t often get too many concrete answers to those questions).

    What I also know from many years of thinking about social policy is that there are no simple answers to any of this stuff. And that most policies that will do some good also have undesired and undesirable effects. I think Fred Argy’s vision of a society that puts quite a lot of effort into trying to redress inequality of opportunities has a lot going for it, but in the end the point of trying to empower people and increase their capabilities is so that they can take more responsibility for themselves. At some point you do have to leave it up to them to make what they can of their lives.

  114. NPOV says:

    BG, fair enough, but how do you determine whether somebody lives in poverty simply because they’ve decided that’s how they want to “make what they can of their lives”? What is it that prevents, say, homeless beggars from making an effort to gain new skills and becoming self-sufficient? Why are their no homeless beggars living in small towns? I’m particularly curious to what extent the reason that there are still so many people living in quite serious poverty in wealthy countries may be because we are maladapted to the size of the communities (huge cities) we inhabit, and our normal instinct to help out others in need doesn’t function properly because 99% of the time we don’t even see the conditions that others live in.

  115. Backroom Girl says:

    Ah but NPOV – I wasn’t talking about the P word, I thought this whole conversation had been about the I word. I don’t have any problem with having policies to alleviate poverty (better still, help people escape it either permanently or at least intermittently), but then we would have to agree on exactly what poverty is, wouldn’t we? And there is a further serious question about whether the best way to address poverty is to simply redistribute more money to the people defined as poor.

    In the end, there are many people in Australia (mostly age pensioners, granted) who manage to live quite comfortable lives on the age pension, so presumably the current level of that payment is ‘sufficient’ for them. I simply don’t see how you could raise the level of income support payments to the level where no-one on them suffered any degree of financial hardship, because quite simply for a small proportion of people no amount would ever be enough. And don’t forget that we are talking pretty big numbers to increase payments by substantial amounts.

    Someone very close to me would point out that one thing that makes social policy so difficult is the problem of ‘asymmetric information’. That is, only the beggar on the street really knows if he or she is there because of the misfortunes of life or because he or she chooses to live that way. So perhaps part of the reason that there are no beggars in small towns is that it is actually a benefit to beggars if the people they beg from don’t know them.

  116. NPOV says:

    But poverty is an inequality issue as much as anything. The way to deal with poverty in a country where everyone is a subsistence farmer is surely very different to the way to deal with it in a country where the bulk of the population are well-fed, can afford houses, cars and any number of luxury goods and services.

  117. Backroom Girl says:

    Yes, but how far do you want to go? Do you really believe that people on income support should have as comfortable a life as people in (low) paid work, for example? I can tell you for sure that many people in low paid work do resent what they see as income support recipients with a standard of living very little different from their own.

    I can tell you from experience it is pretty damn difficult, if not impossible, to design a system that:

    * provides more than a subsistence level of income to people with none of their own;
    * provides sufficient reward for work (work incentive) that those who are able to get a job and support themselves are willing to do so;
    * adequately tops-up the income of people with insufficient money of their own, while maintaining incentives for people to increase their level of earnings over time;
    * minimises the extent of free-riding to keep the average taxpayer happy; and
    * doesn’t cost the earth.

    No system is perfect, but I happen to think that overall we do an OK job in Australia and that the level of income inequality we have is not too bad. We’re not Scandinavia and I don’t think we ever will be.

  118. NPOV says:

    Well ideally, the minimum wage should be seen as not just a (small) step-up from living entirely on government benefits, but as a stepping stone towards jobs that pay better wages in the future. But I don’t believe that living entirely off government benefits should be a constant struggle either: it strikes me as a fairly poor method to motivate people to find paid work instead, and unduly punishes those who genuinely are not capable of paid work. That those in low-paid work resent those living entirely of government benefits is unfortunate, but not, I think, a good basis for determining policy. Especially because there really is very little reason for anyone to stay in minimum wage type jobs for any extended amount of time – unless of course that’s their genuine preference.

    I generally agree we do an “OK” job in Australia (mind you, that’s easy to say from a comfortable middle-class background), but I don’t think I can see I’m proud of the way Australia looks after its least fortunate.

  119. Pingback: skepticlawyer » So, what does ‘progressive fusionism’ look like?

  120. Pingback: skepticlawyer » Can religious discrimination be justified?

  121. Pingback: Cambridge University Conservative Association » Blog Archive » Good and irrelevant discrimination

Leave a Reply

Your email address will not be published. Required fields are marked *

Notify me of followup comments via e-mail. You can also subscribe without commenting.