Mark Thoma lists three different reasons for market intervention,
 Some types of government intervention – weights and measures, disclosure requirements, truth in advertising, safety requirements, etc., are intended to make markets work more efficiently by creating conditions that better approximate competitive ideals. The debate over health care can be cast in this light, i.e. as a debate about how best to solve a market failure that prevents broader coverage at lower prices.
 As second type of intervention redistributes income ex-post, i.e. after the market process has occurred, often through taxation and spending programs. In an economy with significant market failures that cause an inequitable distribution of income and wealth, ex-post redistribution may be justified to redress the imbalances caused by the market failures (and to create equal opportunity).
Thus, the first two types of intervention occur due to market failures, in the first case the intervention is to correct the failures and improve market efficiency, and in the second case the intervention redistributes income ex-post to make-up for inequities caused by existing market failures.
 There is also a third possibility, intervening when markets are working reasonably well. Here, the assumption is that even well-functioning markets can produce inequitable outcomes and hence ex-post redistribution is required. Unlike the first type of intervention which corrects market failures, this type of intervention often leaves markets functioning less efficiently. Much of the objection to government intervention is made on this basis.
He goes on to say that he thinks that reason #2 is the most compelling reason for wealth redistribution and that reason #3, redistribution for redistribution’s sake, is relatively unimportant. I disagree. I disagree partly because I do not see certain markets as failing in ways that systemically harm specific groups to a great extent, but I mostly I disagree because my values are utilitarian and Thoma’s are not (apparently). I consider inequality among outcomes to be the most compelling reason for redistribution of goods. If we can give people large benefits at low costs to other people, we should do so. It is important to remember that the outcomes of perfectly competitive market processes are not moral judgments. People who have highly valued skills and are rewarded for it by market processes are not more “deserving” than people who don’t.
I don’t mean to say that identifying market failures is not a worthwhile task. If certain market failures mean that the labor market rewards some labor skills substantially less than if the labor market did not have those failures, it suggests that correcting those failures could be a low cost way to make workers lives better. If lack of access to credit due to some market failure seriously harms the poor, working around such a market failure may be a good idea, but my motivation for doing so would not be that the poor are not fairly rewarded by the market, but simply to make people’s lives better.
I also don’t mean to say that perfectly distributed income is the ideal; it is important to consider revealed preferences. Other things equal, if one person works harder than another, it likely means that they value the rewards of working more than the other person; leisure and other things which are hard to measure are consumption goods just like material goods. When making judgments about the appropriate amount of redistribution, it is also important to pay attention to how redistribution affects economic growth, because redistribution now may come at a high cost for the future, and this is important even though the future will be significantly richer than us.
I recognize that my utilitarian view is probably a minority view. I think most people probably give much more weight to fairness concerns than to utilitarian concerns.