There's something about the subject of public housing that saps the enthusiasm of even the most dutiful students of government. Self- described policy wonks who have little trouble discoursing on the Medicaid dual-eligible problem or the mass transit mode split start to fidget when anybody brings up Section 8 or Hope VI.

I've often wondered why that was. How to provide shelter for people may not be the juiciest of political questions, but it's ultimately just as important as health, education or transportation. And yet it's difficult to find anyone who likes talking about it--let alone someone really good at writing about it

Perhaps the topic is under-reported because, unlike health care or transportation, public housing isn't a good that most middle-class people have the need or desire to consume. Almost everyone rides on buses or trains and files a Medicare claim if he lives long enough. But hardly anybody who's reading this column lives in a public housing project or, I would guess, is close friends with anyone who does. So except for a very small number of sub-specialists, housing is a virtual black hole.

All this is a rather wordy way of saying that Howard Husock, who teaches at Harvard University and is a fellow of the Manhattan Institute, has made an important contribution to public policy and to American government in general. Over the past decade, in essays for the Institute's City Journal and for a handful of other policy magazines, Husock has been asking the difficult questions about housing that virtually no one else wants to ask.

It's not always easy to swallow his conclusions: Husock believes that most public-housing programs are not only flawed but illogical in their very conception, and that they could all be wiped off the books at a single stroke without doing society as a whole any serious harm. "It's hard to find evidence," he says, "that America ever needed its public-housing system." But even if you find him a bit extreme for your tastes--not an unreasonable reaction--you have to concede that he is a provocative thinker and a burst of fresh air in a field that badly needs one.

Seven of Husock's most interesting essays are about to be published as a book, which is called--not too surprisingly--"America's Trillion- Dollar Housing Mistake." It sounds like a screed, but it's more than that. It's a catalog of inflated hopes and insufficient realism. One by one, Husock takes aim at the failures, the reforms that were meant to correct them, and the reforms that were launched to correct the reforms.

It requires no unusual insight to discredit the vertical ghettoes of 1950s. Even the majority of Americans who rarely think about public housing are aware that the high-rise projects were a disaster--a breeding ground for overall social dysfunction that served to cordon off the underclass from productive society. On this issue, Husock merely reinforces the accurate conventional wisdom.

But we aren't building high-rise public housing anymore; we're tearing it down. In the past decade, from Baltimore to Oakland, vertical ghettoes have been demolished and replaced by modestly scaled mixed-use projects in which subsidized welfare families live side by side with working- and middle-class tenants paying market-rate rents. These new projects not only aren't hideous, some are as attractive as anything in middle-class suburbia.

This strikes nearly all casual observers as a giant step forward. It doesn't strike Husock that way. To start with, he argues, the new low- rise developments can't possibly house as many people as the old towers did. When the towers come down, the tenants have to go somewhere, and what they do is fan out to nearby working-class neighborhoods, using federal housing vouchers to pay the rent. Most of these are aging, fragile communities struggling to stave off dysfunction themselves. A large influx of welfare families brings increased crime and disorder and sometimes threatens a neighborhood's very survival.

This is not a racial issue. In most cases, the older neighborhoods are majority-black. The homeowners, Husock says, "complain that they thought they'd left the ghetto behind, only to find that the federal government is subsidizing it to follow them." It's not that Husock thinks the old projects were preferable to the new ones. What he's arguing is that it's foolish to believe tearing down a few ugly buildings can solve a complex social problem. The problem doesn't go away.

Husock applies the same contrarianism to another fashionable instrument of housing policy in the 1990s: community development corporations.

CDCs are widely admired as a grass-roots institution in which local residents build and maintain their own small-scale housing projects, bypassing the clumsy and insensitive federal housing bureaucracy. Over the past two decades, CDCs have built more than half a million low- priced housing units. In Congress, they are popular with the Republican majority as well as with Democrats.

The only problem, Husock insists, is that CDCs aren't really a grass- roots institution. Conceived by the Ford Foundation and other philanthropies in the 1960s, they continue to be funded largely with foundation dollars and through corporate contributions generated by the federal Low Income Housing Tax Credit, which offers tax breaks to companies willing to use CDC projects as an investment. Although CDC boards are required to include low-income neighborhood residents, the real power is in the hands of professional staff and outside funding sources, well-meaning but far removed from the street-level democracy that the whole movement is supposed to embody.

So it goes on the somber sidewalks of Mr. Husock's neighborhood. For every innovative housing program that is supposed to be bettering the lives of low-income people, he will give you chapter and verse on why it can't possibly succeed.

Underneath all this naysaying is a simple point: Stripped of jargon and pretense, the dilemma of the American underclass is a dilemma of human conduct. It is not one that buildings or subsidies or demographic dispersal schemes will ever solve. "Our 'housing problem,' as Husock puts it, "is really just another name for our single-parent family and illegitimacy problems."

This is not a new point. Students of social policy willing to think clearly have been making it for decades. "There is no direct, simple relationship," Jane Jacobs wrote in 1961, "between good housing and good behavior, a fact which the whole tale of the western world's history… should long since have made evident."

As obvious as this may seem, Husock would argue that virtually the entire edifice of U.S. housing law in the years since has been based on a determination to ignore it. Housing reformers, he says, have taken their cues not from the wisdom of Jane Jacobs but from the manifesto of the New York Association for Improving the Condition of the Poor, promulgated in 1854: "Physical evils produce moral evils." If only that were true, American social policy since 1960 would have recorded glittering successes. Unfortunately, the evidence is overwhelmingly on the other side.

Given this discouraging reality, and accepting the fact that American governments are not going to get out of the public-housing business anytime soon, what sort of reform might actually achieve something? Husock offers a simple, albeit controversial answer: If inner-city problems are problems of behavior, a sensible public-housing program is one that encourages recipients to behave.

Husock would change federal housing policy in much the same way welfare policy was changed in 1996. Rather than making housing a permanent entitlement based solely on need, he would make it a temporary benefit aimed at helping the beneficiaries to work their way out of it. In other words, he would impose time limits.

Some governments are actually experimenting with this. Charlotte has for several years operated a voluntary time-limit program for public- housing projects. The city doesn't have the authority just to cut people off, because that would run afoul of HUD regulations. But it can offer incentives, and that's what it does. Families that agree to a five-year time limit get the nicest, newest units. Philadelphia is launching a similar experiment with its Section 8 recipients.

I agree with most of Husock's ideas. I think we've been much too eager to provide people with shelter for life while asking very little of them in return. I'm absolutely in accord with his view that moral blight isn't the product of physical blight. It's not a symptom; it's the disease.

What seems problematic to me is Husock's insistence that, even if there were no public housing at all, the free market would provide decent shelter for virtually everyone. Maybe it would, if we started paying the working poor a respectable wage. Given the massive erosion of blue-collar wages in the past 30 years, I wonder if the idea of housing for all in a deregulated market doesn't have a bit of wishful thinking behind it.

If we did something about wages for the working poor, we could have a policy that was conservative in its work requirements but humane when it came to rewarding those who met them. This would be reform in the most logical and constructive sense. Despite the successes of the 1996 welfare law, I don't think we're nearly there yet.

But maybe that's a subject for another day.