On Monday, Senator Clinton unveiled one of the first big domestic proposals of the 2008 campaign: a $10 billion, federally funded “universal prekindergarten” scheme. It fits right in with what is clearly a plan by the powerful national teachers’ unions to capture the massive public money becoming available to serve the under-five year olds set.

Just as bad, it implicitly assumes that government money can do far more than it can to improve people’s lives.

Under Mrs. Clinton’s plan, the federal government would match states’ funding for voluntary pre-K for four-year-olds, with a $10 billion annual cap on the federal dollars by the end of the first five years.

To be eligible for the federal match, states would be required to hire teachers with bachelor’s degrees in early-childhood development training, and they’d be required to stick to low student-teacher ratios as well as use some standard curricula.

The program resembles Medicaid, enacted four decades ago as a federal-state partnership to provide health care for the poor. Just as with Medicaid, individual states would decide how to structure their early-childhood programs within those few basic rules, and would be responsible for a big part of the bill.

Unlike Medicaid, states wouldn’t have to participate, but it would be hard for them to refrain. What governor can be against more education for cute kids, especially when a state’s governor and its legislature will know they’ll get “credit” for every dollar of such spending while having to come up with only fifty cents?

But the federal requirements will prove to be expensive for the states, despite the matching funds. For one thing, a mandated low student-teacher ratio means hiring more teachers. And in states like New York, New Jersey, and California, the states that would embrace the program early, the proposal likely will create a huge new demand for expensive teachers from the ranks of the politically powerful unions.

Mrs. Clinton’s plan doesn’t require that states hire unionized teachers, and the nation’s fledgling charter schools, which are usually non-union, could add pre-kindergarten classes to their existing elementary schools with the federal matching funds.

But charter schools comprise a tiny minority of public education. Unless they want to build free-standing schools for four-year-olds, most state and local governments would find themselves adding the vast majority of their pre-K classes to unionized elementary schools, adding hundreds of thousands of highly paid union jobs to state budgets.

And don’t think that the teachers’ unions want to stop at four-year-olds. In New York earlier this month, after heavy lobbying by the local United Federation of Teachers, Governor Spitzer signed an executive order that will allow 50,000 daycare workers who care for toddlers in their own homes to unionize and negotiate for higher pay and benefits.

It’s a slippery slope from encouraging bachelor’s degrees and federally approved curricula to teach four-year-olds to requiring bachelor’s degrees and federally approved curricula to watch two-year-olds. Mrs. Clinton has already started down that slope; her proposal notes that “states [could] serve younger children [with federal money] once they have provided pre-K to all four year olds who need it.”

Supporters of universal pre-K and other early childhood programs often point to the growing body of knowledge that young children develop their cognitive skills well before they’re old enough for school.

Indeed, study after study has shown that by the time they get to kindergarten, children from families that don’t or can’t value education at home just can’t catch up with peers whose parents, say, read a book to them every day since the day they were born.

One of the most comprehensive studies done to date by Georgia State University found that a sample of below-average preschoolers enrolled in Georgia’s universal pre-K made up their deficits and were average or above average on most measurements two years later, by the end of kindergarten.

But the racial gap between white and black students actually became more pronounced after pre-K and kindergarten: whether a student “lived with both parents continuously since birth” made a huge difference in achievement.

It’s no guarantee that a year of pre-school can make up for the next 12 years of poor education and poor family support. A few longer-term studies exist, but they’re often too small to be useful, or suffer from methodological problems.

Worse, for the government to follow the science of cognitive development to its logical conclusion, the feds would have to mandate that local schools force single, poor mothers to enroll their kids in government-funded, expensive full-day education programs staffed by highly trained professionals at birth, so that the kids are away from their dysfunctional families and neighborhoods.

Thankfully, this idea sounds absurd to most people — although maybe less absurd every year. And there’s no guarantee that it would work anyway if the hundreds of billions of dollars spent on anti-poverty programs over the past 40 years are any indication.

But Mrs. Clinton’s plan is an equally absurd half-measure that assumes that even more billions of dollars in state and federal taxpayer money, much of it funneled right through the teachers’ unions into schools that already do a bad job of educating disadvantaged kids aged five through 18, can bridge immense familial and cultural chasms if they just start at age four instead.

Ms. Gelinas is a contributing editor to City Journal, on whose Web site this article will later appear.