To take an example: this is from Yvain's critique of libertarianism as a general approach towards the question of how much government there should be. (His position, as I understand it, is that each program should be evaluated on its own merits and according to its results; I disagree, considering morality to lie in procedure rather than outcome, but his is an understandable and eminently reasonable position).
Upon reading this, my thought process was along the lines of: "OK, this seems like a reasonable model of the situation. Let's suppose I'm in that situation and there is no state. How would I go about solving it?"
2.3: How do coordination problems justify regulation of ethical business practices?The normal libertarian belief is that it is unnecessary for government to regulate ethical business practices. After all, if people object to something a business is doing, they will boycott that business, either incentivizing the business to change its ways, or driving them into well-deserved bankruptcy. And if people don't object, then there's no problem and the government shouldn't intervene.A close consideration of coordination problems demolishes this argument. Let's say Wanda's Widgets has one million customers. Each customer pays it $100 per year, for a total income of $100 million. Each customer prefers Wanda to her competitor Wayland, who charges $150 for widgets of equal quality. Now let's say Wanda's Widgets does some unspeakably horrible act which makes it $10 million per year, but offends every one of its million customers.There is no incentive for a single customer to boycott Wanda's Widgets. After all, that customer's boycott will cost the customer $50 (she will have to switch to Wayland) and make an insignificant difference to Wanda (who is still earning $99,999,900 of her original hundred million). The customer takes significant inconvenience, and Wanda neither cares nor stops doing her unspeakably horrible act (after all, it's giving her $10 million per year, and only losing her $100).The only reason it would be in a customer's interests to boycott is if she believed over a hundred thousand other customers would join her. In that case, the boycott would be costing Wanda more than the $10 million she gains from her unspeakably horrible act, and it's now in her self-interest to stop committing the act. However, unless each boycotter believes 99,999 others will join her, she is inconveniencing herself for no benefit.Furthermore, if a customer offended by Wanda's actions believes 100,000 others will boycott Wanda, then it's in the customer's self-interest to “defect” from the boycott and buy Wanda's products. After all, the customer will lose money if she buys Wayland's more expensive widgets, and this is unnecessary – the 100,000 other boycotters will change Wanda's mind with or without her participation.This suggests a “market failure” of boycotts, which seems confirmed by experience. We know that, despite many companies doing very controversial things, there have been very few successful boycotts. Indeed, few boycotts, successful or otherwise, ever make the news, and the number of successful boycotts seems much less than the amount of outrage expressed at companies' actions.
Ultimately, the problem is lack of commitment from individuals. Talk is cheap, and the incentives and options available lead them to behaviour which fails to solve the problem. So clearly we want to change either the incentives or the options.
The first idea which came to me is that, since so many transactions are carried out by credit card, could one have one's credit card blocked from being used at certain stores? Then you would sign up to a campaign page which would work in a similar way to Kickstarter: you, and everyone else signing up, would enter your details and agree that upon the campaign reaching a certain, specified number of participants, some switch would be activated and none of you could use your credit cards at that particular store.
I won't claim this as a perfect solution: there would need to be more thought put into issues such as preventing the use of cash or alternate credit cards, and persuading people to trust the website with their details (though it should be noted that plenty of websites have managed to overcome this difficulty without any trouble). Perhaps these would turn out to be insurmountable and my idea wouldn't work. Fine then. Perhaps, with more and more websites moving online, there might be some kind of software that kept tabs on where you were shopping and would trumpet loudly that you had bought from someone you had pledged not to. The idea I am getting at is not that I have solutions - for all I know, these are both terrible ideas which could never possibly work - but that there do exist ways we can solve these problems without resorting to government.
Market failure exists, at least primarily, when individual rationality on the part of all actors leads to group irrationality. That is to say, any attempt by an individual to increase social welfare reduces their individual welfare. This creates a deadweight loss - it would be possible to make some (maybe even all) people better off without making anyone worse off. The classic solution is for the state to somehow mandate the behavioural changes which realise this increase in social welfare, but it seems to me that this model underestimates the ability of entrepreneurs to come up with new solutions.
What's more, technology is making these entrepreneurial solutions ever easier to realise. Perhaps 50 years ago, a stateless society would have had problems funding a road system. I don't know how it would have worked, but three ideas present themselves:
1) Roads are operated as a loss-leader by those selling cars and petrol. This seems unlikely, particularly in a competitive market system, partly because it is unclear that it would be worth the cost and partly because only a small portion of the increased revenue caused by a firm investing in roads would accrue to the firm making the actual investment.
2) Road users are charged a flat rate regardless of how much they use the roads. They would have to pay a fee for usage of a firm's roads, and would receive a windscreen sticker indicating a right to use that firm's roads; if caught by an agent of the firm using the roads without a sticker, they could be prosecuted. This is fairly similar if not identical to the statist system. and would have a number of problems - large economies of scale (in catching unauthorised users, logistics of road repair) leading to an uncompetetive market structure, poor incentives for road users, vast expenses for families driving on holiday.
3) Toll booths. These would also have problems - slowing down traffic, large economies of scale (because who really wants to pay twenty fares on the way to work) leading to an uncompetetive market structure.
There may well be other ways I have overlooked; I would guess that some combination of 2 and 3 would be most likely. (There might also have been a massive move towards public transport). However, the problem of funding would be ridiculously easy to solve nowadays. Something like the London congestion charge could operate, with cameras recording if you had been on a road, and if so then how often, and at the end of the month you would receive a bill. Alternatively, your car might automatically record where you went and send it to some agency.
This "failure of imagination", if you will, is common to many areas covered by the state. To take a quote from Clement Attlee (for non-UK readers, the UK Prime Minister 1945-51 and the chief founder of the modern British welfare state):
In a civilised community, although it may be composed of self-reliant individuals, there will be some persons who will be unable at some period of their lives to look after themselves, and the question of what is to happen to them may be solved in three ways – they may be neglected, they may be cared for by the organised community as of right, or they may be left to the goodwill of individuals in the community. The first way is intolerable, and as for the third: Charity is only possible without loss of dignity between equals. A right established by law, such as that to an old age pension, is less galling than an allowance made by a rich man to a poor one, dependent on his view of the recipient’s character, and terminable at his caprice.Ultimately, the problem of supporting the unemployed falls into two categories: the short-term unemployed, and the long-term unemployed. It seems fair to assume here that Attlee refers purely to the short-term problem (which is just as well, for otherwise I should be lambasting him for his - ahem - uncharitable approach towards the issue of charity). For this problem, he completely overlooks the possibility of private insurance, most likely through a Friendly society although I see no reason in principle why it should not be done for profit. Indeed, given that at this time National Insurance was a genuine insurance scheme for workers, rather than the income tax by a different name which it has become, it seems odd that he failed to think of this possibility.
Ultimately, the best way to demonstrate that something is not a market failure is to find a market solution. Perhaps these market solutions do not always exist. But it does seem rare to find evidence that state advocates have tried to find those solutions before decreeing intervention.