Beginning with President Roosevelt's New Deal, our government gave middle-income citizens a stake in policies that help the poor. You would create a broad-based program, such as social security, that would collect most of its taxes and pay most of its benefits to middle-income taxpayers, with some spillover benefits to low-income citizens. Today, the case for this type of social contract is not very compelling.

Willie Sutton

Since the 1960's, the share of income accounted for by the top fifth of households is up somewhat. More important has been the increase in all levels of income. The average real income of people in the second fifth of households today exceeds the average real income of people in the top twenty percent in the 1960's. See the following table, which comes from the census report on income distribution, in dollars of constant purchasing power.

The combination of a large rise in overall income and a slight increase in the share at the top means that households earning over $100,000 now account for something close to three-fourths of all income. If we think of "rich" in absolute terms ($100,000 per year in household income, adjusted for inflation) rather than in relative terms (the top 20 percent), the "rich" now earn enough income to fund both baseline government functions plus programs to help the poor.

We do not need the middle class to pay taxes any more. In fact, with income taxes, the middle-class taxpayer is on the road to extinction. Data from the U.S. Treasury compiled by Daniel Mitchell for the Heritage Foundation show that the bottom 50 percent of the income distribution accounts for only 4.2 percent of tax revenues, as shown in the follwing table:

As an aside, given this data, it obviously is valid to say about any income tax cut that the benefits do not flow to the lower middle class. You cannot cut someone's taxes unless they are paying taxes to begin with.

The most significant taxes on middle-income familes are social security taxes and property taxes. Both of these taxes help fund middle-class benefits: retirement income and schools.

Going forward, there is an opportunity to have the taxes paid by rich people used to fund new middle-class entitlements. These might include expanded coverage for health care, child care, elder care, pet care, minivan care, cell phone care, you name it. Fifty years ago, the money for such entitlements would have had to come from the same income group receiving the benefits. Today, it is possible to transfer income from the wealthy to the middle class. I am not saying that it is desirable to do so, but I am pointing out that it has become feasible.

Alternatively, the tax capacity of the rich could be used to try to help alleviate poverty. What is interesting is that if it no longer is necessary to tax the middle class in order to provide money for the poor, then perhaps we no longer need to funnel aid to the poor through programs that have large middle-class constituencies. While the middle class wants its tax payments to go for social security and public schools, perhaps rich people are willing to see their taxes go toward effective programs that directly help the poor.

To put this another way, one of the arguments against privatizing social security or enacting school vouchers is that if you give the middle class the opportunity to opt out of social security or public schools, then there will be no political support for subsidizing the poor. However, now that there is a significant capacity among the rich to subsidize the poor (because there are more rich people nowadays and rich people are richer), we may not need to seduce the middle class into helping poor people.

Leaky Buckets

If a poor person is an alcoholic or hears voices, then it may not be sufficient to provide money. It may be at least as important to plug the leak by curing the addiction or mental illness.

It appeared to me that one example of "leaky bucket" poverty was the case of Barbara Jennings, as described in the book A Hope in the Unseen, by Ron Suskind. Actually, the book is focused on her son, Cedric Jennings, who manages to enroll in Brown University, after overcoming the handicaps of an inner-city boyhood. It tells a fascinating story of the emotional issues that he faces, given that he is not naturally comfortable either in the ghetto or in the elite academy.

What struck me about Barbara Jennings was that she frequently was evicted for nonpayment of rent, in spite of having a steady job--at low pay, to be sure--working for the U.S. Department of Agriculture. While I can understand that a single mother with a slender salary would not be living in suburban splendor, my guess is that there are plenty of people with less income who pay their rent on schedule.

The book never explains where Jennings' money goes. At one point, however, Suskind mentions that she obtains some credit counseling. At the end of the book, she has managed to right herself financially, even paying off over $11,000 in debts. Suskind does not explain how this occurs.

It appears to me that Barbara Jennings fixed the leak in her bucket. That in turn enabled her to improve her financial position dramatically.

Middle-class programs that happen to provide benefits to the poor are programs that pour more water into the bucket. My guess is that we need to focus more on fixing the leaks.

With the general rise in incomes in the past thirty years, poverty has fallen dramatically. This point has been made by W. Michael Cox and Richard Alm in their book Myths of Rich and Poor.

Part of their argument is that measuring "poverty" as being in the lowest 20 percent of the income distribution means that you are counting that segment as poor by definition. Instead, they say that if you look at consumption behavior, most people in the bottom twenty percent of the income distribution are well-off compared with middle-class consumers in the 1960's.

Another part of their argument is that for a given individual, income fluctuates. Only a fraction of the people now in the lowest income bracket will remain in that bracket for many years.

Cox and Alm make a good case that less than five percent of the U.S. population is truly, persistently poor. My guess is that a high proportion of this hard-core poverty consists of leaky buckets. This group may not be helped much by programs that are derived from the expansion of middle-class entitlements, such as social security and free public schools.

In theory, better education is a "fix-the-bucket" strategy. However, as I have pointed out, both in terms of efficiency and equity, I consider the existing public school system itself to be a leaky bucket. I would feel much better about having my wealth flow to poor people as vouchers for education than to have it be taxed to go to public school bureaucracies.

The New Social Contract

Today, the rich are richer and more numerous while the truly poor are fewer and with more complex needs. As a result, the broad-based middle-class programs are no longer a panacea. An alternative is to use programs that are targeted to help the poor and are paid for by the rich.

It is not necessary to use government taxes and expenditures to accomplish this new social contract. It is possible that "venture philanthropy" would be more efficient. For example, wealthy individuals have funded school voucher programs in Washington, DC.

The possibility of a new social contract provides some perspective on the debate over tax cuts. President Bush's opponents are arguing that smaller government and smaller taxes necessarily will be worse for the poor. However, it is not clear that this must be the case. Taxing the rich to fund expanded middle-class programs under the old social contract may be less progressive than allowing them to use their money to engage in "venture philanthropy" under a new social contract.