Some Of Our Books


« Trolling for Intuitions on Punishment | Main | Experiment Month Studies Go Live! »

April 08, 2011


Feed You can follow this conversation by subscribing to the comment feed for this post.

Jussi: Could you say more about the first caveat at the end? I was thinking throughout that the objections to the indirect model all applied in principle to the direct model, i.e., there didn't seem to be anything about the directness, per se, of the model that rendered it vulnerable or not to the objections stated.

David, that's a good question which I've been thinking about - and one with which Goodin seems to struggle too. I can give it a go (sorry about the length).

First, if the state provides welfare services directly, it could legislate strict rules for on what basis those services are to be provided. It could lay out rules that leave as little room as possible for the use discretion in application. These would be of the type - if subject A satisfies condition B (where this is as general and non-moralised as possible), then she is entitled to service C. Given that these rules are written down, any welfare recipient who feels unfairly treated could appeal on an independent court of justice. Also, the state could use inspectors to check that the rules are followed. This would provide the required checks and balances.

[In principle, the best way to get rid of discretion would be to set no conditions of the welfare benefits and services whatsoever. This would get us to Universal Basic Income which is problematic in many other ways.]

Now, this raises two questions. First, you might think that the state could legislate similar strict rules for the private companies, charities, voluntary organisations, and local communities. I think the main problem here is that the result would be against the very idea of the Big Society. Given that the non-state organisations would have no discretion at their use, it is not clear in which sense they would have been empowered to be responsible for what happens to them.

The second objection is more philosophical. This is the famous rule-following idea that no principle interprets itself. This would entail that we cannot avoid that the agents within the state institutions use discretion in interpreting the centrally provided strict guide-lines. So, for this reason, Goodin thinks that the four problems cannot be solved by the strictness of the rules but rather by their content and the back-ground conditions. I'm not quite certain about this.

In any case, it does seem to me that this left-over discretion in rule-following is a smaller amount of discretion in a multi-level state institutions than the one which the non-state institutions have available for them in the visioned Big Society.

Goodin's own solution to this problem is generosity. According to him, what pushes agents in government organisations to use discretion in rule-following in the harmful way is that they are required to help everyone who needs/deserves assistance, and *only* those. He suggests that they should instead have enough resources to be able to support everyone who needs help which would allow them to forget about whether other people too will enjoy from their services. As a result, the people who need help could not come to bear the burdens of the discretion.

Yet, even this kind of generosity might not work in the voluntary organisations and local communities. If you give more than enough resources, they might still not distribute them to those who do not satisfy their moralised conditions.

In your first paragraph you refer to the "good philosophical research" that has been done regarding the failure of "indirect" governmental intervention. I don't know your background, so perhaps you are not as knowledgable about the U.S. experience on this subject as those who have spent their entire adult lives here. But we have substantial empirical evidence going back some 50 years that indicates that direct state programs have failed to achieve their stated objectives, and are even counterproductive.

We spend more money on K-12 education per student than any nation in the world but Norway, and get terribly mediocre results. We directly fund these failing schools. Poor children suffer the most in this system, as the "rich" can afford to send their children to private schools. Teachers unions have severely restricted efforts to introduce competition into this system.

We had the welfare state for many decades providing direct monetary assistance to families without imposing conditions like actively seeking employment, and built high-rise housing projects that resembled Stalin-era Soviet apartment buildings. These were breeding grounds of crime and social pathology and most of these were demolished in relatively short order after costing taxpayers billions. That noted reactionary, President Clinton, pronounced the verdict held by the vast majority of Americans when we passed welfare reform in 1996: "the current welfare system undermines the basic value of work, responsibility, and family, trapping generation after generation in dependency and hurting the very people it was designed to help."

I won't go into detail here, but the Medicare system of "direct" healthcare for the poor is widely considered to provide substandard care and is driving the states into bankruptcy. It is in a fiscally unsustainable situation, and will be reformed one way or the other.

What weight should your "good philosophical research" have relative to the facts? Second, I don't understand how your argument works against vouchers. Why can't we give poor parents vouchers equal in value to what the taxpayers spend now per student on public education, let the parents use the vouchers, and then measure the academic performance of their children relative to the pre-existing baseline? Same principle for housing vouchers, vouchers for medical insurance, etc?

Finally, your second paragraph seems to assume that governmental intervention is required if laissez faire capitalism produces hardships for some members of society. This only follows if you assume that the intervention alleviates this hardship without causing unintended consequences that are as bad or worse than the original problem. Maybe your all-knowing and supremely wise U.K. social planners have eliminated this prospect, but I think history says otherwise.

Sorry, Medicaid not Medicare provides healthcare for the poor.

Thanks Mark.

Well, I grew in Finland in the 80s which pretty much was the golden period of the welfare state. We had public schools, health care, libraries, and universities and all of these were in good working order (the schools and healthcare are still getting graded amongst the best in the world). We also had fairly good unemployment benefits and yet low unemployment. I cannot say many bad things about it, and neither can many other Scandinavians. I'm not sure why the welfare state in the US has never quite worked out. Part of me things that it has never even properly tried.

For instance, part of me wants to claim that the reason why you got so much crime was the lack of welfare systems rather than too much of it. If there are no reasonable alternatives to sustain oneself, crime is pretty much what you turn to. And, again, Scandinavian countries compared favourably to US in the crime rates, so the causal chain cannot be from more welfare to more crime. Also, I don't think you can blame the highrises on the welfare state. They were just poorly built and poorly designed housing some of which existed in Finland too. But then we got nice government supported housing too like the houses where I grew up.

The response to the Clinton comment would take more time. Goodin in fact dedicates a whole chapter in the book to that question so maybe I do another post on the topic some time.

In any case, I don't think the facts are against the welfare state. It has worked in some areas in some times which seems to suggest to me that it can be made to work elsewhere too.

I do like the voucher ideas for many things. However, it will discriminate against minorities. If I move to an area where some other group is a complete majority, they can use their vouchers to support a business that caters them services that fit their ideology. If I am in a small minority, my vouchers won't be enough to support similar services for me. This leaves me with two choices: either to adapt to the majority or to move. These are not always reasonable alternatives. And, segregation is something we should try to resist for many reasons.

If you read the second paragraph carefully, I said that that is something that is assumed by both sides of the debate here. The debate about whether that assumption is tenable is for another day. I agree your with your second to last sentence though. But when I look back at my life in Finland, it is hard to see what the evil consequences were that would outweigh the good of providing a concrete, fairly well-functioning safety net for even the worst off of the society. True we didn't have hedge-fund managers or other billionaires at the time.

Just a quick question for Mark,

In your view, the empirical research supports the hypothesis that when government outsources or privatizes we find that things improve?

I take it that the comparative judgment is the judgment to focus on. I don't think funneling government money into for-profit colleges, for example, is going to improve the state of higher education in this country.

Thanks for your thoughtful response. First, I want to clarify that as a natural rights libertarian I believe that rights generally trump consequentialist considerations, although I acknowledge that rights must yield when the stakes get high enough, i.e. moderate deontology. Thus, I am arguing here on your turf.

Nevertheless, I think that you have implicitly agreed that even with a consequentialist approach what drives the direct/indirect decision is the empirical evidence and the inferences we can reasonably draw, not abstract philosophising. You and I can agree to disagree about what the evidence shows. But, I cannot resist pointing out that your happy experience with state intervention appears to come at a steep price. According to the "CIA World Factbook," Finland's 2010 GDP per capita is $35,300, while the figure for the U.S. is $47,400--a very large percentage difference. This disparity exists despite the fact that we spend a significant percentage of our output on military defense, which is a drag on the economy. It might also be argued that Finland and other nations that are unable to defend themselves free ride on these expenditures.

You no doubt believe that the more egalitarian distribution of wealth in Finland has greater moral weight than your inferior GDP per capita. But this depends on the value we assign to equal outcomes, a debate we can leave to another occasion. Perhaps more directly on point, which experience with direct "social justice" programs is more relevant to the situation in the U.K., Finland's or that of the United States? Since Finland is a nation of 5.2 million with a homogeneous population, I think I know the answer.

I just don't see your point about minority culture and vouchers. Unless you propose that the state build separate schools, stores, hospitals, etc. for every small minority in a given geographic area, which is completely impractical, vouchers do no worse than direct governmental programs. In fact, since private schools (and private enterprise generally) operate with less overhead (more efficiently) than their public counterparts, a voucher system is more likely to provide goods and services that are responsive to minority needs.

Hi Mark. Well, I don't think empirical evidence trumps abstract philosophising. If we know that certain structure of a society creates for instance exploitative power structures, then that is a strike against the system.

I don't really know what the steep price is. GDP isn't a very good measure of wealth of a population in any case. And, I don't think I need to rely on egalitarianism. I'm more inclined towards prioritarianism. The idea is not that equal outcomes have value per se but rather that the wellbeing of the worst off matters more than the wellbeing higher up. If you are a consequentialist, diminishing marginal utilities will get you the same result.

So, in any case, on pretty much all other measures like infant mortality, life expectancy, school life expectancy, unemployment, public dept, prison population, human poverty index, crime, and population below poverty line Finland is either equal to US or slightly better. Also, given that you don't have to pay for healthcare insurance, education, and so on the lower income can get you an equal quality of life.

I also don't think that Finland is free riding on US military in any way. Even during the cold war we remained as a neutral third party. About the experiences of UK, well many things work fine here too on the public sector. The experiences of privatisation have not been exactly encouraging here so far.

I was thinking that the state can provide fairly equal services that are culturally neutral which all groups can use (instead of segregated ones). That was one of the good features of the Finnish system. Being able to make the overheads small requires efficiency which requires large numbers. In competitive markets that requires being attractive to most people in the community. Smaller minority units from whom the minorities need to buy their services will be more expensive.

There's a good example about this in swimming pools. In Finland, some public swimming pools offer a separate times for Muslim women who do not want to swim at the same time as men or even with the majority population women. This is a very small group. Assume that all Finns would be given a swimming voucher and the building and running of the pools would be left for private companies. At any one time, the pool company would get more vouchers in by being open to all groups at the same time. And, the few vouchers that Muslim women have would not be sufficient to support any company that would build a pool for them.

In any case, I'm fairly convinced by Elizabeth Anderson's view that the main point of social justice is preventing oppression and creating capabilities to equal political participation. And, comparing the state-system and the Big Society, if this is our goal then the former system is better than the latter. Of course, you might think that there is a third system - the libertarian one that is superior to both. I'm not arguing against you here.

Well, on the subject of education, our best, most exclusive and prestigious colleges and universities are private: Harvard, Yale, Princeton, Cal Tech, MIT, etc. The question, then, is do we need state colleges and universities? Does the success of private institutions count as evidence to the contrary? I think it does.

As mentioned earlier, the U.S. spends more on K-12 per student than almost any other developed country. Yet, on standardized tests we rank far down the list. To me, this is evidence that the current state-run system is not functioning well. I think there is some evidence that school vouchers for students from poor households improve educational outcomes, although as with all social science evidence, it is hard to draw firm conclusions. But, since vouchers do not cost taxpayers more than the existing system, it seems to me that they are well worth trying.

On the main point, it is clear that you are convinced that on the standard of justice advanced by Anderson the direct state solution to social problems is superior to the indirect. I will just say that I don't understand the basis for this confidence. If vouchers provide a superior solution for housing, education and medical care for the poor relative to the direct state supply of such goods, then they will also tend (it seems to me) to reduce oppression, equalize political participation, etc. Of course, it remains to be seen whether vouchers are superior, but I don't see that you have a good philsophical argument against them. Your Muslim pool example doesn't count as one, at least for me.

With regard to empirical evidence generally, I'd just like to mention a book, one of the co-authors of which is Goodin, that examines the three principal types of welfare state regimes in the post-industrialized world of Europe and North America: liberal (e.g., the U.S.), social-democratic (e.g., the Netherlands), and corporatist (e.g., Germany), assessing them according to their respective abilities to meet avowed "moral" goals of such regimes: promoting economic efficiency; reducing poverty; promoting social equality; promoting social integration and avoiding social exclusion; promoting social stability; and promoting autonomy. Based on the aforementioned criteria of assessment, the social democratic regime fares best (and the liberal regime the worst) among the three types.

The comparative drawbacks and disappointments of the liberal welfare state regime vis-a-vis its own goals and the aforementioned criteria clearly demonstrates the problem with relying "principally on schemes of tightly targeted and only just adequate transfer payments," other regimes being conspicuously generous in both respects (perhaps this speaks in further favor of 'basic income' schemes). Indeed, the U.S. welfare state "rather overachieves in its quest for a 'mean and lean' welfare state. It certainly does not give people more than they need to escape from poverty; instead, it errs in the opposite direction, standardly giving poor people less than they need. Nor does the liberal welfare state generally give benefits to many who do not need them: again, it errs instead in the opposite direction, not paying benefits to substantial proportions of those who are in need." As our authors conclude, "contrary to liberal hopes, high incomes on average do not translate into adequate incomes for the poor. The market and private giving leave substantial proportions of people in poverty, and welfare programmes that are overly-tightly targeted mean that a substantial proportion of them remains so. The liberal welfare regime succeeds in keeping costs down, but at the cost of allowing poverty to remain comparatively high."

Please see Robert E. Goodin, Bruce Headey, Ruud Muffels, and Henk-Jan Dirven, The Real Worlds of Welfare Capitalism (Cambridge, UK: CUP: 1999).

Thanks Patrick - I'll have a look at that.


I'm puzzled by how you've framed the problem. Does Goodin really think that direct intervention in behalf of the poor is always preferable to indirect intervention? And do his critics really think that indirect intervention is always preferable? I would think that a mixed view would be more natural.

Regarding discretion, it's not fair for Goodin to note only its disadvantages. I'm sure I needn't explain how an utter lack of discretion can be deeply problematic and lead to all sorts of injustice.

Finally, I wasn't convinced by the manipulation/exploitation worries. These could be far more salient in the case of direct intervention since, in the case of indirect intervention, the poor would likely have a more of a choice regarding from whom they'll receive assistance and on what terms. In that case, they probably wouldn't have to deal with an entity that had monopoly power.


thanks for the comment and sorry for not being clear enough. Much of the framing of the problem is mine and Goodin should not be blamed for it. His argument is mainly against the idea that we should leave the intervention in behalf of the poor to charities. I applied his argument against this position to the indirect role of the government. I'm not completely against a mixed view. I think the idea is that government should directly guarantee certain minimum satisfaction of basic needs to avoid worst kinds of exploitation and other problems of discretion. Maybe a plausible mixed view could say that beyond this the role of the state could be indirect.

To some degree, I agree about the second point. I think that in the realm of the satisfaction of the most basic needs, the bad consequences of discretion in the business of satisfying them are so serious that they do outweigh the good consequences of having discretion in this domain. But, as I just noted, it might be that above the realm of basic needs the problematic consequences are not so serious and the benefits greater. It might be that giving discretion would for instance motivate the volunteers and give them a sense of responsibility and significance. This fits my second caveat according to which there is nothing bad about voluntary organisations and local communities just as long as they are not relied upon in the very fundamental functions of the welfare state.

I'm not sure though that the exploitation goes away by having many intermediate organisations offering help. Having many churches offering food for salvation does little to diminish the theistic conditions they set. Of course governments can set moralised conditions for their help too and this is none the better or less exploitative. However, it does seem to me that there is a greater possibility to come to fairly general, non-exploitative conditions for welfare support through the political process that requires consensus than there is of having voluntary organisations that will not want anything in response.

I think David Shoemaker's original objection stands, namely that "directness" per se is no solution to the various problems in aid. Sure, the state could write strict rules for its aid--but so could the intermediate agencies. The state could distribute aid justly, or the state could distribute aid with inadequate attention to local particular situations.

I think it boils down to a matter of whether one trusts the state or local institutions more. Jussi's bugbear is the church who imposes moral or theological requirements on its aid, unjustly. He is not worried about a state like, say, the Soviet Union, which supplied plenty of direct aid but not with a great deal of justice.

The usual defense of using local institutions as intermediates rests on two points: (1) that one is better placed to make decisions when one is nearer the scene; faraway central governments are less sensitive than local governments or other institutions; and (2) centralizing power is dangerous and a temptation to tyranny, while distributed power provides checks and balances. If you believe, as Jussi seems to, that central governments can plan well for localities, and that tyranny by the state is not a serious danger, then the direct solution is attractive. The fundamental questions here seem to be to be empirical.


I have to say I'm not quite convinced by the objection.

First of all, Soviet Union is completely irrelevant for this discussion. We are considering the question, if the free markets leave the fundamental needs of some groups unaddressed and we think that justice requires that the state do something about this, just how exactly should the state attempt to help to satisfy those needs? Given that the Soviet Union had no free markets, this question could not have been asked in that system. Soviet Union had no way of addressing the problems of the individuals who do not fair well enough on the markets unsupported. It did not even have free markets. And, it was not very good in addressing the fundamental needs of its citizens either.

Of course, I agree that a welfare system that serves the fundamental needs of the poorly off directly may fail to carry out its purpose. Its agents too can have too much discretion which brings about all the unwanted consequences.

However, the question is whether the evils of discretion can be better addressed in a direct system or an indirect one. I don't think this question is altogether empirical. Having discretion seems to me to built into the very nature of voluntary organisations, charity groups, and local communities. Without discretion, I very much doubt whether these groups would be motivated to act in the first place. In contrast, it seems to me that at least some level of the evils of discretion are eliminable in the direct system of a welfare state. But I won't repeat myself here.

Here's a question Heath. You note that an advantage of local communities being responsible for securing that the fundamental needs of everyone are satisfied is that these communities can pay adequate attention to local particular situations, i.e., they can be more sensitive. What does this exactly mean? The worry is that the local communities will be more sensitive to what most people or the most powerful group in the area prefers whilst not being as sensitive to each individuals fundamental needs.


Fine, we can leave the Soviet Union out of it. My point was that it’s perfectly possible for the aid-distributing organization to be either (i) corrupt or (ii) incompetent, or some combination, whether that organization is a government bureaucracy at whatever level, or whether it is a private organization.

Here is an example of corruption in the American case. It’s well-documented that public housing in large cities like Chicago in the 1980s and 90s often did not get the new furniture or repairs they were supposed to get, and which was paid for by city, state, or federal organizations. The civil servants in charge of the buildings just never did their work and often pocketed the money. One solution pursued now is to just give poor people housing vouchers, eliminating the housing bureaucracy, and turning direct into indirect aid. That has its own problems but, apparently, they tend not to be as bad.

As far as the “sensitivity” question, it’s pretty common here for the federal government to give money to state or city governments, which they then spend more or less as they see fit, within some federally-established limits. This for the very good reason that officials in, say, Denver are better placed to figure out which highways need fixing, and who to hire to do that, than officials in Washington. Some education and some medical dollars are distributed the same way. So the general principle of “make decisions at the local level” is often pursued within the public/governmental apparatus, and so local/national is not the same debate as direct/indirect. But I don’t see why local governments would be any less subject to powerful local interests than local private institutions. Indeed, local governments being in the thrall of real estate developers or large employers is very common.

Just to add a little to Heath's point on housing vouchers. I believe they go some way to preserving the dignity of the recipient, since they put the holder in the role of a consumer, who must choose how to best use the voucher, rather than simply a passive recipient of aid. Also, since the holder can use the voucher to pay for housing of whatever type and in whatever location he/she desires, it is way more responsive to their individual needs than simply building a big housing project for everyone, which may be far away from a person's family, their best employment opportunity, etc.


few points. Firstly, that sort of voucher scheme still seems like a fairly direct means of satisfying the housing needs of the badly off. It is the government after all that hands out the vouchers and makes up the conditions for the companies that can cash those vouchers. In the UK, the defenders of the Big Society believe that this is too direct. They are reducing the housing benefits which means that hundreds of thousands of people will have to move away from the areas where they now live. Now, some voluntary organisations might step in but it is unlikely that they will have the means or that they help is unconditional.

I also agree that the local/national vs. direct/indirect distinction is important. For the defenders of the Big Society, local public institutions like councils are still too direct involvement of the government. I think there are several reasons why these might less subject to powerful local interests. At least they still work on the one man, one vote basis, and they are responsible for the higher-level government organisations.


if the level of the vouchers would be as high as you seem to think, then I agree that the scheme sounds appealing. I'm not sure, however, why the state should build big housing projects anyway.


A thought struck me over the weekend about how one might put your argument; I wonder what you think of it.

You’ve mentioned the objectionable “moralized requirements” of private organizations several times. But it’s not as though the state is not moralizing about its requirements: anti-discrimination rules, or income limits, are moral constraints on how aid is distributed. At some level, the entire welfare state project is a bunch of moralized requirements. Whatever these publicly-decided-on moral views are, call them P.

When we involve private organizations in the aid-distributing process, we give them some discretion, which amounts to deviations from P. They may add objectionable requirements, like requiring recipients to listen to sermons. Or they may subtract requirements the public endorses, e.g. be racially biased. In either case, the situation is objectionable and unjust. I think your reasoning here is some combination of (1) P is established through some procedurally just mechanism and deviations are, therefore, ipso facto, unjust; (2) You just take it to be empirically more likely that deviations from P are non-procedurally, i.e. substantively, unjust.

If the aid goes through private organizations with the proviso that distribution must be governed by exactly P, I don’t see you objecting. That is just private contracting, using vouchers or the like, and you don’t seem to have a problem with that. The state need not be in the business of, say, building housing. It is deviations from P you object to.

And I think we can take this one step further. It is not deviations from P driven by the private beliefs of individuals, but deviations driven by the openly declared purposes of the non-governmental organizations. I say this because civil servants are not immune to having private agendas; for example, a lot of governmental aid to American farmers used to be distributed in a very racist fashion. The problem, rather, is that the government is then in the position of furthering agendas which the public has endorsed.

Am I in the right neighborhood?

The comments to this entry are closed.

Ethics at PEA Soup

PPE at PEA Soup

Like PEA Soup

Search PEA Soup


  • Unless otherwise indicated, the views expressed in any given post reflect the opinion of only that individual who posted the particular entry or comment.