As with any set of policies, there are winners and losers. With the welfare state, lobbyists win, politicians win, bureaucrats win, some of the poor win, some of those win who have a strong subjective preference that some of the poor win via help from the government, and most others lose. Given that there are winners and losers, it's always possible to construct utility functions that justify the welfare state or show that it's a net negative.
Let's for a moment examine one of the more well thought out analyses that's often used to justify the welfare state. A Theory of Justice by John Rawls does just that. He does this by stepping through a set of clever thought experiments.
Everything flows from the first innovative thought experiment called the original position:
The original position is a hypothetical situation developed by American philosopher John Rawls as a thought experiment to replace the imagery of a savage state of nature of prior political philosophers like Thomas Hobbes. In the original position, the parties select principles that will determine the basic structure of the society they will live in. This choice is made from behind a veil of ignorance, which would deprive participants of information about their particular characteristics: his or her ethnicity, social status, gender and, crucially, Conception of the Good (an individual's idea of how to lead a good life). This forces participants to select principles impartially and rationally. [...]
Rawls argues that the representative parties in the original position would select two principles of justice:In other words, Rawls claims that those in the original position would all adopt a maximin strategy which would maximize the prospects of the least well-off. Inequality is allowed to happen, but only to the extent that it creates incentives for the well-off to produce more such that it benefits the least well-off:
The reason that the least well off member gets benefited is that it is assumed that under the veil of ignorance, under original position, people will be risk averse. This implies that everyone is afraid of being part of the poor members of society, so the social contract is constructed to help the least well off members.
- Each citizen is guaranteed a fully adequate scheme of basic liberties, which is compatible with the same scheme of liberties for all others;
- Social and economic inequalities must satisfy two conditions:
- to the greatest benefit of the least advantaged (the difference principle);
- attached to positions and offices open to all.
The Difference Principle regulates inequalities: it only permits inequalities that work to the advantage of the worst-off.The vast majority of academic criticism of A Theory of Justice came from the Left (i.e. it's not socialist enough). However, "in Anarchy, State, and Utopia, Robert Nozick argues that ... Rawls's application of the maximin rule to the original position is risk aversion taken to its extreme."
But depending on when the maximin rule is applied, I'm not sure it is extreme in its risk avoidance. A thousand years ago, it was pretty important to avoid being among the worst-off pretty much everywhere in the world. Even now, an awful lot of the world is pretty awful and if Rawlsian justice were achievable in those awful places, it would be pretty tempting to adopt it. Unfortunately, it's precisely because no sort of justice is achievable in most of the awful places that they are awful in the first place.
To me, the primary problem with the original position, the veil of ignorance, the maximin rule, and the difference principle, is that they don't even attempt to take into account the effect of the structure of society on the future. For example, assume that human civilization will ultimately span 100,000 years (it's spanned somewhat less than 10,000 years so far). If a society constructed around completely free markets grew at a mere one-hundredth of one-percent per year faster than a society constructed around Rawls' difference principle, the median GDP per capita over the 100,000 year period for the free market society would be 148 times greater than for the difference principle society. Even if we assume that the worst off get only one-hundredth as large a piece of the pie, the better bet from the original position, if you didn't know which time frame you were going to be part of, is the free market approach for all but the most extremely risk adverse people in the original position.
One of the biggest problems I have with the welfare state is that I believe that it's stealing from me now, but even worse, it's stealing even more from my children's future, and a tremendous amount from their descendants.
To summarize, the welfare state has winners and losers, but more and more losers over time.