April 24 (Bloomberg) -- Mitt Romney suggested last week that the Department of Housing and Urban Development “might not be around later.” Predictably, this set off a minor firestorm.

Eliminating every HUD program would be unwise and inhumane, but the best ones could be saved even if the department were shuttered. If we want to take Romney’s suggestion seriously, we should start by designing a shutdown plan, under which HUD’s most critical functions, such as housing vouchers and the Federal Housing Administration, remain operational.

In my last column, I argued that the Romney campaign could do the U.S. a signal service by presenting a serious path toward fiscal responsibility; closing departments might help achieve that goal. HUD has five major functions: tenant-based housing vouchers, mortgage assistance, support for housing supply, community-building efforts (such as community development block grants and empowerment zones) and more general programs for housing and urban affairs.

The U.S. spends about $19 billion a year on the rental assistance program known as Section 8 housing choice vouchers, which is the largest line item in HUD’s budget. Romney’s father, George, helped lay the groundwork for this initiative when he served as HUD secretary during the Nixon administration.

Housing Help

This program now helps about 2.2 million households pay rent. Vouchers are administered by local housing authorities, which operate under HUD rules. Recipients must earn less than 50 percent of the area’s median earnings and 75 percent of the aid must be distributed to families whose income is less than 30 percent of the area’s median. The local housing authority sets a payment standard, and voucher recipients are expected to spend 30 percent of adjusted gross income on rent and utilities. The difference between that 30 percent payment and the local standard is covered by the voucher.

Does this make sense? Compared with the public housing programs it replaced, the voucher program is a triumph. It enables choice, is far less wasteful and, unlike large housing projects, doesn’t automatically create clusters of poverty. One recent study argued that the program’s benefits far exceed its costs, in part because vouchers improve education outcomes.

The best evidence of vouchers’ long-term impact is provided by a randomized evaluation called Moving to Opportunity. Although that study found that vouchers don’t improve youth educational outcomes or economic results for adults, it showed that recipients live in significantly better neighborhoods, and are less depressed and less obese. Vouchers are no magic bullet, but eliminating them would cause enormous hardship to millions of poorer Americans.

Nonetheless, the vouchers program does have flaws that could be remedied through restructuring. It costs about $1.5 billion to administrate (this isn’t surprising given the role of local housing authorities, which check any proposed housing unit for health and safety concerns). With the radical improvements in the U.S. housing stock, however, this function is probably no longer necessary.

In addition, there are long waiting lists for vouchers, which makes receipt somewhat random. For every $1 of extra income that a recipient family receives, its voucher declines by 30 cents, an implicit tax that deters working by the poor.

Finally, though vouchers are supposed to be mobile, there are longstanding complaints about the difficulties of transferring them from one housing authority to another.

Under New Management

Here are two proposals for a post-HUD voucher system. The simpler method is to transfer the operation to the Treasury Department and the tax code. In this scenario, every poorer family would be eligible for rental assistance, based on income and payment standards in their area. Families would receive a check or tax credit from the Internal Revenue Service that is equal to the difference between their documented rental payments and their ability to pay, as long as their rental payments don’t exceed the local payment standard.

This mechanism could radically reduce administrative costs, enhance mobility and increase fairness. If we wanted to reduce the implicit tax on those who work, we would need to design the program to resemble the earned income tax credit, which doesn’t decrease as much as earnings rise. Voucher quantities would need to fall. As a result, we would need to implement a gradual transition to limit the harm to current recipients to keep spending constant as vouchers became more widely available.

Alternatively, vouchers could be taken over by Health and Human Services, which could roll them into the current Temporary Aid to Needy Families, or TANF, program. It makes little sense to have three separate silos of aid to the disadvantaged (food stamps, housing vouchers and TANF). Consolidation could ensure the right form of aid was targeted to the needs of particular families.

Mortgage-assistance programs should also be maintained, even if HUD were to vanish. The Federal Housing Administration, the mortgage insurer for poorer Americans, covered almost one-fifth of home-purchase loans in 2009 and 2010. My co-author, Joseph Gyourko, argues that the FHA is woefully undercapitalized relative to its level of risk, with about $30 billion of assets being held against $879 billion of insured mortgages. He may be right that the FHA will be “the next housing bailout,” but historically, the agency’s insurance fee of close to 85 to 90 basis points per year has enabled it to cover its costs.

I share Gyourko’s fears about the FHA, but I have trouble believing that we would want to eliminate the primary tool for helping poorer Americans to become homeowners. Ginnie Mae also bundles and sells mortgages insured by other government agencies, including the FHA and the Department of Veterans Affairs, and it has had few problems.

Mortgage Insurance

If HUD were to vanish, then the FHA and Ginnie Mae could become standalone entities, perhaps joined with the remaining public portions of Fannie Mae and Freddie Mac. Ideally, they would be isolated from political influence in an entity that was charged with insuring mortgages for fees that were high enough to offset any reasonable financial losses. This entity could be housed in the Treasury Department, which knows a bit about banking and the financial system.

HUD’s housing-supply-related functions are an appropriate place for pruning. The low-income-housing tax credit probably should be phased out: The program subsidizes construction everywhere, but that help isn’t needed in most parts of the U.S. The private sector is perfectly capable of delivering huge amounts of affordable housing in states such as Texas, and we don’t really need any more buildings in declining areas, such as Detroit. Indeed, one study found that public housing crowds out private housing, so that each new subsidized unit creates only one-third to one-half of an extra housing unit.

The most expensive areas, such as greater New York City and coastal California, where local governments overregulate new construction, could benefit from federal public support for new construction. When localities overrestrict, they impose costs on people elsewhere who might like to move in. But any national supply subsidy, including the low-income credit and project-based housing vouchers, ignores the profound differences across local markets and can be gradually eliminated.

There is an existing public-housing stock that needs maintenance, and HUD tends to that task. This responsibility could be given directly to local housing agencies. Ten-year phase-out grants can avoid immediate hardship, and push toward a future in which the federal government is no longer in the public-housing business.

HUD also has a sizable community-building portfolio, including community development block grants, but Congress could direct these funds either to localities or to states that would have a mandate to transfer the money to appropriate communities. The principle of these grants was to engender local autonomy, which makes HUD’s oversight relatively unnecessary.

Moral Leadership

Finally, I am loath to see HUD completely lose its moral leadership in fair housing. The federal government should have an officer who is charged with ensuring that the disadvantaged are treated fairly in housing, and that localities don’t hurt the country by engaging in excesses of NIMBYism. This function could be handled with an undersecretary-ranked official at either Justice or Commerce. In principle, further distribution of community block grants could be made contingent on good behavior, as certified by this official.

HUD’s valuable research function could also be taken over by this new housing official, and would find a natural home at the Commerce Department. The Bureau of Economic Analysis already does significant research, which could be integrated with HUD’s current housing portfolio. Other HUD functions, such as support for American Indian housing, could be rolled into other departments, such as Interior.

How much would this reorganization save? At best, a couple billion dollars a year in administrative costs and perhaps $10 billion in reduced spending on supply-related policies (which could be eliminated even if HUD survived). Any larger financial benefits of eliminating the agency would depend on more conservative practices at the FHA and fewer new projects.

I’d prefer to start by closing the Department of Agriculture rather than HUD, or at least closing them simultaneously in a grand bargain under which urban and rural America lose their representatives in the Cabinet. Of course, just as with the best parts of HUD, Agriculture’s food stamps program, which aids more than 46 million people a month, would need to be put in another department, perhaps Health and Human Services.

U.S. policy, with its huge highway subsidies and a home-mortgage-interest deduction that bribes people to move from urban renting to suburban owning, is already stacked against city life. Yet it isn’t impossible to imagine a sensible post-HUD future, as long as our commitment to the most vulnerable Americans isn’t radically diminished.

(Edward Glaeser, an economics professor at Harvard University, is a Bloomberg View columnist. He is the author of “Triumph of the City.” The opinions expressed are his own.)

Read more opinion online from Bloomberg View.

Today’s highlights: the View editors on European austerity and antibiotics in farm animals; Jeffrey Goldberg on the space shuttle; Ramesh Ponnuru on the Buffett rule; Enrique Krauze on Peru’s economic revival; and David Crane on California’s rising taxes.

To contact the writer of this article: Edward Glaeser at eglaeser@harvard.edu.

To contact the editor responsible for this article: Max Berley at mberley@bloomberg.net.