The following is an excerpt from the book Free to Choose by Milton & Rose Friedman, pp. 115-19:

The Fallacy of the Welfare Sate

Why have all these programs been so disappointing? Their objectives were surely humanitarian and noble. Why have they not been achieved? At the dawn of the new era all seemed well. The people to be benefited were few; the taxpayers available to finance them, many – so each was paying a small sum that provided significant benefits to a few in need. As welfare programs expanded, the numbers changed. Today all of us are paying out of one pocket to put money – or something money could buy – in the other.

A simple classification of spending shows why that process leads to undesirable results. When you spend, you may spend your won money or someone else’s; and you may spend for the benefit of yourself or someone else. Combining these two pairs of alternatives give four possibilities summarized in the following simple table:

Category I in the table refers to your spending your own money on yourself. You shop in a supermarket, for example. You clearly have a strong incentive both to economize and to get as much value as you can for each dollar you do spend.

Category II refers to your spending your own money on someone else. You shop for Christmas or birthday presents. You have the same incentive to economize as in Category I but not the same incentive to get full value for your money, at least as judged by the tastes of the recipient. You will, of course, want to get something the recipient will like – provided that it also makes the right impression and does not take too much time and effort. (If, indeed, your main objective were to enable the recipient to get as much value as possible per dollar, you would give him cash, converting your Category II spending to Category I spending by him.)

Category III refers to your spending someone else’s money on yourself – lunching on an expense account, for instance. You have no strong incentive to keep down the cost of the lunch, but you do have a strong incentive to get your money’s worth.

Category IV refers to your spending someone else’s money on still another person. You are paying for someone else’s lunch out of an expense account. You have little incentive either to economize or to try to get your guest the lunch that he will value most highly. However, if you are having lunch with him, so that the lunch is a mixture of Category III and Category IV, you do have a strong incentive to satisfy your own tastes at the sacrifice of his, if necessary.

All welfare programs fall into either Category III – for example, Social Security which involves cash payments that the recipient is free to spend as he may wish; or Category IV – for example, public housing; except that even Category IV programs share on feature of Category III, namely, that the bureaucrats administering the program partake of the lunch; and all Category III programs have bureaucrats among their recipients.

In our opinion these characteristics of welfare spending are the main source of their defects.

Legislators vote to spend someone else’s money. The voters who elect the legislators are in one sense voting to spend their own money on themselves, but not in the direct sense of Category I spending. The connection between the taxes and individual pays and the spending he votes for is exceedingly loose. In practice, voters, like legislators, are inclined to regard someone else as paying for the programs the legislator votes for directly and the voter votes for indirectly. Bureaucrats who administer the programs are also spending someone else’s money. Little wonder that the amount spent explodes. The bureaucrats spend someone else’s money on someone else. Only human kindness, not the much stronger and more dependable spur of self-interest, assures that they will spend the money in the way most beneficial to the recipients. Hence the wastefulness and ineffectiveness of the spending.

But that is not all. The lure of getting someone else’s money is strong. Many, including the bureaucrats administering the programs, will try to get if for themselves rather than have it go to someone else. The temptation to engage in corruption, to cheat is strong and will not always be resisted or frustrated. People who resist the temptation to cheat will use legitimate means to direct the money to themselves. The will lobby for legislation favorable to themselves, for rules from which they can benefit. The bureaucrats administering the programs will press for better pay and prerequisites for themselves – an outcome that larger programs will facilitate.

The attempt by people to divert government expenditures to themselves has two consequences that may not be obvious. First, it explains why so many programs tend to benefit middle-and upper-income groups rather than the poor for whom they are supposedly intended. The poor tend to lack not only the skills valued in the market, but also the skills required to be successful in the political scramble for funds. Indeed, their disadvantage in the political market is likely to be greater than in the economic. Once well-meaning reformers who may have helped to get a welfare measure enacted have gone on to their next reform, the poor are left to fend for themselves and they will almost always be overpowered by the groups that have already demonstrated a greater capacity to take advantage of available opportunities.

The second consequence is that the net gain to the recipients of the transfer will be less than the total amount transferred. If $100 of somebody else’s money is up for grabs, it pays to spend up to $100 of your won money to get it. The costs incurred to lobby legislators and regulatory authorities, for contributions to political campaigns, and for myriad other items are a pure waste – harming the taxpayer who pays and benefiting no one. They must be subtracted from the gross transfer to get the net gain – and may, of course, at times exceed the gross transfer, leaving a net loss, not gain.

These consequences of subsidy seeking also help to explain the pressure for more and more spending, more and more programs. The initial measures fail to achieve the objectives of the well-meaning reformers who sponsored them. They conclude that not enough has been done and seek additional programs. They gain as allies both people who envision careers as bureaucrats administering the programs and people who believe that they can tap the money to be spent.

Category IV spending tends also to corrupt the people involved. All such programs put some people in a position to decide what is good for other people. The effect is to instill in the one group a feeling of almost God-like power; in the other, a feeling of childlike dependence. The capacity of the beneficiaries for independence, for making their own decisions, atrophies through disuse. In addition to the waste of money, in addition to the failure to achieve the intended objectives, the end result is to rot the moral fabric that holds a decent society together.

Another by-product of Category III or IV spending has the same effect. Voluntary gifts aside, you can spend someone else’s money only by taking it away as government does. The use of force is therefore at the very heart of the welfare state – a bad means that tends to corrupt the good ends. That is also the reason why the welfare state threatens our freedom so seriously.