President Bush is taking a step backward with a drive to promote $138 million in new spending for abstinence-until-marriage programs in his most recent round of welfare reauthorization proposals.

In fact, $50 million a year is already being spent by the federal government on abstinence programs, with precious little return on the dollar.

For more than three decades, federal, state, and local governments have addressed the teenage pregnancy issue through family planning, youth development initiatives, abstinence, and contraceptive-education programs — with negligible results and huge expense. Analysts have wrestled with ambiguous program-evaluation results; indeed, many analysts have provided solid evidence and persuasive arguments that most programs do not work well.

To place these results in context, recent research from the U. S. Department of Health and Human Services indicates that teenage birth rates are down 22 percent since 1991, with the rate for 2000 the lowest rate ever recorded. Moreover, birth rates for black teenagers declined more rapidly in the 1990s than for any other population group, a reduction of 31 percent. The birth rate for Hispanic teens dropped by 13 percent from 1994 to 1999 — even if it did rise by one percent in 2000.

Indeed, some might reasonably assert that having eight percent of the teenage female population — aged 15 to 19 — pregnant, and close to five percent of this population segment giving birth, is not a stand-alone public policy problem. We are encouraged to ask: Is teenage pregnancy really the legitimate domain of public policy and government largess?

While the president and congressional committees have agreed to remove the so-called "illegitimacy bonus" from this year's round of welfare reauthorization legislation, most Americans will not be aware that the U.S. government already spends $571 million annually on pregnancy prevention, and another $988 million on the medical costs of teen births, for a total tab of $1.56 billion per year.

In addition, another $1.15 billion is spent by the federal government providing welfare cash assistance and other welfare services to teen mothers. Thus, the grand total is about $2.7 billion a year. On a "unit cost" basis, the federal government currently finances teenage pregnancy to the tune of $1,867.00 per teenager per year. It is likely an abuse of the public trust to spend huge sums of public money for five to eight percent of any population subset, particularly since the source of this problem involves voluntary social acts.

And are we getting much for our money? Not much. The Alan Guttmacher Institute and the Henry J. Kaiser Family Foundation report that between 60 and 80 percent of unintended pregnancies among teens occur among teen contraceptive users. (Apparently, promoting contraception increases pregnancy rates.)

In addition, comprehensive youth development programs have shown little or no effect, according to a Washington State study. Another Alan Guttmacher study demonstrates that a slightly greater number of teens are choosing to delay sexual intercourse. But there is scant evidence that comprehensive sex education programs provide the rationale for this more conservative behavior. And on the abstinence-only strategy, too few teens are captured by the wisdom of abstaining from sex to possibly justify all the time, effort, and money expended by the federal government.

Finally, national teenage pregnancy and birth rate figures have often been intertwined with the rising rates of out-of-wedlock births in the general population. For teenage girls, however, the percentage of out-of-wedlock births as a proportion of the general population has remained stable for more than four decades at about 14 percent. For those who say the problem of teenage pregnancy is getting worse, they are often mixing up their public policy problems; that is, teenage pregnancy and illegitimacy.

In the usual fashion, these kinds of social problems often correct themselves when the culture and civil society begin to work out such central issues. Indeed, state guidance and spending largesse are not often required to find a resolution. As Isabel Sawhill and Sara Brown, leaders of the National Campaign to Prevent Teen Pregnancy, have noted, declining teenage pregnancy is largely the result of a teenage population adopting more conservative attitudes toward sex. That we have been able to observe these declines is miraculous when we consider the heavy promotion of easy sexual morays by the media and Hollywood.

But, obviously, individual, rational behavior by young women is having beneficial results. If we can just keep politicians from their perilous task of "doing good," declining teen pregnancy rates are likely to continue.

Kimble Ainslie is an entitlements policy analyst at the Cato Institute .