It sounds like heresy to say so, but maybe the consolidation of the five boroughs into the City of New York, whose 100th anniversary we celebrate this month, wasn't such a good idea. The question is not just of theoretical interest. Today prominent urbanists are urging a new wave of consolidation, exhorting cities to merge with their suburbs to form region-wide metropolitan governments. But equally energetic advocates have mounted fierce political campaigns to make exactly the opposite view prevail—to form smaller, decentralized city governments for neighborhoods that secede from a larger whole. There are good reasons to believe that the secessionists are right. Indeed, even New York, notwithstanding all the anniversary hoopla, would work much better split into pieces.

The idea of metropolitan government—a benevolent, expert central administration for urban areas—has tempted efficiency-minded urban theorists for generations. Here, they've argued, is a way of bringing order to the chaos of central cities surrounded by a crazy quilt of independent suburbs. Though in the 1920s the movement saw metropolitanism as a recipe for improved city services and non-corrupt leadership, by the 1960s advocates added two more goals: redistribution of wealth, with affluent suburbs supporting public services in poorer inner cities, and environmental protection, with enlightened planners dictating the shape of future development, preventing "urban sprawl" and keeping "ticky-tacky" suburbs from devouring farmland.

Today's foremost champion of metropolitan government, David Rusk, onetime mayor of Albuquerque and self-described former civil rights and anti-poverty worker, travels the lecture circuit spreading the gospel of his 1993 book, Cities Without Suburbs. A unified metropolitan government, Rusk proclaims, can "profoundly transform the long-term outlook for failing central cities and help re-energize American society." Cities that annex their suburbs, Rusk claims, adding his own twist to earlier arguments, not only will improve the lot of their poorest residents but also will be more prosperous overall than those that don't. As to how exactly the new metropolitanism will produce new wealth, he is disconcertingly vague.

Though big-city mayors and elite opinion makers like Rusk's views and consider them mainstream, on the political front lines these ideas don't carry much weight. Across the country local activists have been rejecting the push to create bigger jurisdictions. They want to retain—or create—smaller governments, by seceding from existing city governments, by incorporating new, smaller jurisdictions carved out of larger ones, or by resisting annexation by larger governments. New Yorkers are familiar with this move toward the local: five years ago the Staten Island secession movement blazed up fiercely, though the State Legislature eventually snuffed it out. But similar efforts are catching fire in many other locales.

Most notably, in Los Angeles, the ValleyVote movement, with Governor Pete Wilson's support, proposes to detach the entire San Fernando Valley (population: 1.2 million) from the rest of the city. Proponents say such a secession could come to a referendum vote in the year 2000. The next step could well be to split up the residential expanse of the Valley into 23 newly incorporated municipalities. In Florida four successful referenda since 1992 have given birth to four new municipalities within Dade County, the administrative district for 1.5 million residents of the suburbs surrounding central Miami. Six other new municipalities are in the works. Around Tucson, suburbs that have long resisted annexation by the central city leaped at the chance state legislation offered them last year to incorporate as independent towns. Two have already done so; six others have proposed or already scheduled votes. Suburban activists outside Houston and Atlanta have strenuously resisted the efforts of those cities to bite off pieces of their suburbs. Nor is localism limited to the U.S. In Canada, Citizens for Local Democracy (whose leaders include revered urban theorist Jane Jacobs) successfully organized resistance to the Ontario government's "megacity" plan to submerge, under the banner of cost reduction, six independent municipalities into a Rusk-like greater Toronto.

The conflict between localism and metropolitan government is a clash not over form but philosophy of government. To liberals like Rusk, localism reflects yet another example of what journalist Robert Kuttner has called the "revolt of the haves"; it is a greedy retreat from the commonweal, sacrificing the city for the short-term improvement of the suburbs. But Rusk believes the revolt of the haves is futile: it threatens the suburbs themselves, as inner-city chaos overtakes the surrounding communities. In this vein, Myron Orfield, a state legislator from Minneapolis and author of an influential text on metropolitan government, laments that "powerful resentments, based on class and race," have prevented cities from merging with suburbs to create efficiency and social justice. If the suburbs shared their often considerable riches, older, poorer neighborhoods wouldn't decline into dilapidated urban ghettos.

Orfield's view, the elite wisdom about secession and incorporation movements, is simply wrong. Localism is popular not because it promises a sweetheart deal for a few privileged suburbanites at the expense of the greater good, or because the unsophisticated fail to understand a demonstrably superior metropolitan approach. Instead, it rests on common sense—which economics and political science amply confirm. Voters' common sense tells them that the closer they are to government, the more it will respond to their demands. They will see their hard-earned tax dollars spent on the kind of projects they prefer and will have a greater assurance that interest groups—such as public employee unions—will not usurp local government for the benefit of their own members, who may not even live in the city in which they work.

In fact, there are good reasons to go one step further. To improve older neighborhoods in older cities requires not a single, bigger government but increased numbers of smaller ones. Rather than expanding cities, we should break them up into an array of independent, neighborhood-based governments that would set their own property-tax rates, elect their own officials, and give city residents the same control and sense of community that their suburban counterparts take for granted. City dwellers could direct public spending to the things they consider most important. They could ask the local public works director why their street went unplowed or unpaved, or push the local chief of police to deal with the rowdy playground gang before things get out of hand. Inevitably, such a system would favor economic growth over redistribution. Freed from centralized bureaucracies, these neighborhoods, including many of the older, poorer ones, would prosper. As for paying to maintain, or build, expensive regional infrastructure systems: for that purpose, these independent local governments could cooperate in a loose confederation, or "special purpose district."

I will admit to some personal bias in all this. As a resident and onetime minor elected official in Brookline—a Boston suburb of 56,000—I'm convinced that the town's decision 125 years ago not to merge with the central city is a key reason for our good schools, clean streets, and low crime rate, despite our closeness to the central city. My town is a living advertisement for the local control that leaders of today's secession, incorporation, and annexation campaigns seek.

These campaigns don't add up to a movement in the sense that metropolitanism is a movement, supported by foundations and promoted at conferences. The leaders of the new localism are responding to purely local conditions. Yet though they are generally unaware of their counterparts elsewhere, local activists make strikingly similar observations and complaints. Consider two of the most active and important sites of the new localism, both in areas of rapid population growth—southern California and south Florida.

In Los Angeles the head of the ValleyVote movement is a commercial real estate broker named Jeff Brain, whose office, in a converted toy store at the rear of a strip mall, overflows with clippings from the L.A. Daily News, the Valley newspaper, which has pushed hard for a secession vote. In his professional life, Brain rents out storefronts in this and other small strip malls along the length of Ventura Boulevard, one of the San Fernando Valley's main streets. Brain also helps organize a network of 30 local home owner and business organizations, with such typical LA names as the Tract 115105 Neighborhood Association. Though some affluent enclaves dot the Valley, on the whole this is Queens, not Westchester; New Jersey, not Connecticut. This is the valley of the shop-till-you-drop Valley Girl, where residents in small ranch houses and whitewashed apartments gaze up at the Santa Monica mountains, beyond which lies Beverly Hills. The Valley—white, middle-class, and heavily Jewish—is increasingly Latino and Asian. Some 27 percent of Valley residents are Spanish-speaking; 11 percent live below the poverty line. New restaurants sport signs in Farsi or Cyrillic characters; at Torah-land, a growing Orthodox Jewish community looks for bargains on yarmulkes and talises.

Jeff Brain's complaints are very specific. Most nights, only two police cars patrol his own community of Sherman Oaks, which, like other residential areas in the Valley, has its own name but not its own government. Sidewalk cleaning in the commercial areas along Ventura Boulevard is dismal, even though merchants pay extra fees to the city for a private contractor to do the work. Brain also resents the private school tuition he pays for his four children to keep them out of the giant LA Unified School District, with its unmanageable 800,000 students. Moreover, the city pays little for Valley road repair (only $4 million out of a $28 million budget), even though the Valley contains a third of the city's total area. Look at the number of residents per City Council member, Brain complains: some 250,000 people per Council district. With fewer than 100,000 residents, nearby independent cities like Burbank and San Fernando have their own mayors and city councils, along with lower business taxes and an easier permit process. Brain looks enviously at older commercial main streets in these neighboring communities, where small beautification steps have paid big dividends by helping to attract and retain shoppers and businesses, while Van Nuys Boulevard, another of the Valley's main drags, grows shabbier by the day.

For Brain, being too distant from local government to get anything accomplished is more than a metaphor. "In order to get something for our area," he observes, "we have to go downtown. That's an hour's drive to start. Then we have to convince council members from all over the city. They say, `Why should you get something if my neighborhood doesn't?' Then you reach an impasse."

Such frustration drove Brain to mount a successful campaign this past fall to persuade Governor Wilson to sign legislation allowing city neighborhoods to detach, without first getting the permission of their existing city councils. Next he'll try to get the more than 100,000 petition signatures necessary to get a regional advisory board to examine the issue, as required by law. Brain expects the petition drive—and then the secession vote—to succeed. And he believes that large parts of Los Angeles will follow suit, seeking to detach and incorporate on their own. "We've had interest from throughout the city and throughout the state," he says.

In Miami, localism has advanced even further. From his dark, wood-paneled office as chairman of one of the city's leading law firms, Eugene Stearns has already been through the fire of referenda campaigns like those Jeff Brain anticipates. A native Miamian who has watched the city change around him, Stearns is no political neophyte. He helped run the campaigns of former Florida Democratic governor Ruben Askew and served as his chief of staff. He thought he had left political life behind to build up his law firm—until 1991, when he heard that developers planned two new 800-room hotels close to his home in Key Biscayne, an enclave of 8,500 residents within so-called unincorporated Dade County. Key Biscayne had no local government: the county government, housed in a downtown Miami office tower, supplied the police force, firefighters, and garbage collectors.

"I figured I was a fairly connected guy, and I wanted to do something to get this hotel decision re-examined," he says—"not necessarily to block it entirely but to scale it down some. And I was told, `This is a done deal: forget it.'" His unexpected sense of powerlessness led him to reflect on the overall quality of services in Key Biscayne.

Well-known and affluent, with luxurious waterfront condos and pastel Mediterranean-style homes, Key Biscayne nevertheless had embarrassingly pitted streets and shabby recreation areas far worse than in nearby independent municipalities—even blue-collar ones like Miami Shores, where many cops and schoolteachers lived. As in Los Angeles's San Fernando Valley, political representation was disproportionately weak, with just 13 out of 100 county commissioners representing the 1.5 million county residents who lived outside the core city of Miami (population: 365,000). Stearns grew convinced that centralized government made it too difficult for citizens to influence decisions about their neighborhoods. Moreover, he was certain that bigger government, with its layers of management, wasted on unproductive bureaucracies funds that would otherwise flow into the capital improvements—new baseball fields, new public docks—that he and his neighbors wanted. "It all comes down to bigger isn't better," he says.

Stearns decided not to waste time fighting specific policies but to attack the underlying problem. He proposed to lead an independence movement to incorporate Key Biscayne as a village, with its own council and full-time manager. In November 1992 the incorporation referendum sailed through with a resounding 70 percent of the vote. Since then, Key Biscayne's assessed property valuation has skyrocketed by some $800 million, because local control has made the municipality a still more desirable place to live—and Eugene Stearns, from his nerve center in downtown Miami, has become the leader of an incorporation movement throughout south Florida. He has provided his law firm's services pro bono to leaders of four similar referenda, with others in the offing. Three of them succeeded, making Stearns the godfather to the new municipalities of Pine Crest, Aventura, and Sunny Isles, as well as his own Key Biscayne.

The referendum that failed is Stearns's most interesting battle. He served as advisor to Shirley Gibson, a black retired policewoman who sought to carve out an independent municipality called Destiny from the subdivisions surrounding the car dealerships and racetracks on Route 441, north of Miami. The area, 80 percent black, is a magnet for upwardly mobile African-American families from inner-city Miami neighborhoods like Liberty City and Overtown. But Gibson, an entrepreneur who had started a business that quirkily combined two of her own interests—a beauty-supply-shop-cum-private-detective-agency—believed that, for all the energetic striving of its residents, the area was suffering neglect.

"We had the worst bus service, the worst parks, the worst police protection," she recalls. "The kids couldn't drink out of the water fountains in the parks, and the bathrooms didn't work." Yet county planners told Gibson that she should be grateful for what she had—that the proposed new city was a "recipient community": it received services valued at $21 million but paid only $17 million in taxes. Gibson rejected this. She knew from experience that the high overhead costs of centralized county government gobbled up tax money before it could reach her neighborhood parks. "I'd been a policewoman; I knew that you had six layers of supervision in a big department and only two in a smaller one," she explains. Stung, too, by the condescending label "recipient community," with its insinuation of public assistance, Gibson, with Stearns's help, led the 1995 incorporation campaign.

And she would have won, had not the Miami Dolphins' Joe Robbie football stadium stood on the outskirts of the proposed new municipality. Owner Wayne Huizenga, concerned about how a new local government might tax his stadium, led an opposition that outspent the pro-Destiny forces by almost eight to one ($300,000-plus to $40,000) and emphasized the possible job losses middle-class black families, many of whom worked for government, might face. Destiny failed to gain incorporation by 700 votes of the 5,500 cast. Gibson does not rule out another referendum, although a change in the law now requires the permission of the sitting Dade County commissioners. Despite the loss, Gibson notes, services in her neighborhood have improved, a change she attributes to the mere threat of secession.

The reason to believe that these incorporation and secession leaders are right begins with history—even though the metropolitanists assert that history is on their side. But in fact the consolidation of formerly independent municipalities in New York, Philadelphia, Pittsburgh, and Boston, which metropolitan advocate Rusk cites to buttress his case, didn't arise from a Rusk-like belief that bigger was better. Rather, newly developing areas saw consolidation as the best means to plug into the services core cities offered. By the 1920s, as soon as suburbs discovered other means short of consolidation to hook up to regional infrastructure—typically special-purpose districts—they stopped joining central cities.

Even when given the option of voting for mergers and annexations, Americans historically have supported the creation of more local governments, not fewer. "The American system is one of complete decentralization, the primary and vital ideal of which is that local affairs should be managed by local authorities," wrote legal observer Thomas Cooley perceptively in 1868. Municipalities have differentiated themselves from one another for good reason. The formation of independent cities and towns fueled the explosive economic takeoff of the late 1800s; it defused tensions between immigrant and native-born; and it allowed the upwardly mobile to build communities that reflected their hard-won new social status. Industrialists, including the founders of Cudahy (incorporated so they could build a meatpacking plant just south of Milwaukee), have always sought safe haven from the hands of government regulators. A boon for owners, such municipalities also provided a living wage for employees and buttressed the property values of residential areas in nearby municipalities. Distinct ethnic and cultural groups established their own niches, as in the once "dry" towns of Pasadena and Compton, California, or Oak Bluffs, Massachusetts, where Methodists for many years used local control to keep out alcohol. The lesson is crystal clear: independent jurisdictions are a crucial means through which a nation as diverse as the U.S. can develop a modus vivendi among peoples of sharply different values and wildly various backgrounds.

Even as advocates (including the Department of Housing and Urban Development) beat the drums for metropolitan government, the number of local governments in the U.S. kept rising. From 1952 to 1992, the number of municipalities grew from 16,807 to 19,279. While a few core cities—Indianapolis, Jacksonville, and Nashville—have merged with their surrounding counties to form metro governments in recent years, citizens have overwhelmingly scorned the metro vision over the past half-century. Champaign and Urbana in Illinois twice rejected consolidation, even though they're often thought of as a single college town. Voters in the Knoxville and Richmond areas refused city-county consolidation, as did voters in David Rusk's own Albuquerque (before Rusk's tenure as mayor). Even where citizens actually have embraced metro governments, suburbanization has inexorably continued, with metro governments generally stalled at their original lines.

There's no shortage of theory to explain why this long-standing American preference for localism makes sense. The key fact: we don't all want the same things from our local jurisdictions. Those with small children may care most about education, unmarried joggers may want to spend public money on parks, and the tidy-minded may want the streets cleaned three times a week. Forty years ago, in a brief but classic essay, economist Charles Tiebout argued that local governments do more than coexist side by side. Instead, they compete with one another for residents by offering different packages of services. Of course, wealthier communities can provide more amenities than poorer ones; that's part of the free-market incentive structure. But at equivalent income levels, governments can differentiate themselves, in terms of the kinds of services they offer and also the cost-efficiency with which they provide them. If they fail to provide what people want at reasonable cost, residents can "vote with their feet," wrote Tiebout. When municipalities lose residents, property values fall, leaving remaining residents with a powerful incentive to figure out what's gone wrong.

I can attest, through personal experience, that Tiebout's argument is sound. Several years ago, I agreed to join a volunteer financial-planning advisory committee to examine my town's costs for specific services. The committee immediately gathered cost data from comparable municipalities. Where our costs differed, we asked hard questions. Why do our police officers take twice as much sick time as those in other towns? Is it a management problem, a defect in our union contract—or are our police simply older than those on other forces? How can we improve matters? We were acting as Tiebout would have us act: "Public service agencies," he has written, "may be forced to compete over the service levels offered in relation to the taxes charged."

Daniel Elazar, director of Temple University's Center for the Study of Federalism, observes that some of the nation's most smoothly functioning cities may owe part of their success to competition of this sort. Elazar notes that in the Bay Area, three flourishing midsize cities—San Francisco, Oakland, and San Jose—compete (and also cooperate) with one another and with Silicon Valley towns like Palo Alto and Sunnyvale. Prosperous and efficient Minneapolis and St. Paul, along with a gaggle of nearby cities with populations between 100,000 and 150,000, do the same.

In other words, because of this competitive, anti-monopolistic mechanism, smaller—not bigger, as the metropolitanists contend—is more efficient. New research from the Institute of Government at Florida International University, located right in the middle of Dade County's wave of incorporations, soundly debunks the big-is-efficient argument that is the linchpin of the metropolitanists' case. Public administration professor Milan Dluhy examined the costs per resident for a wide range of core municipal services in metropolitan Dade County and in 24 "fragmented municipalities" within and around the county. Dluhy found that economies of scale existed in only two areas: fire protection and library services. Localities can provide all the other services—police, recreation, public works, waste management—at equal or less cost.

Evidence from other economists strongly supports the case that smaller is cheaper and more efficient than bigger. David Sjoquist analyzed 48 metropolitan areas in the South and found that competing local governments kept costs down. As he puts it: "The level of expenditures will fall as the number of jurisdictions rises." Conversely, Richard Wagner and Warren Webber, looking at counties in 16 southern states, found that consolidation and centralization led to greater spending, not less.

Why exactly would costs increase after governments merge? Jane Jacobs tartly gave one explanation in a recent talk: "Anyone who has had to deal with a big-city bureaucracy knows that the idea that bigger is more efficient is laughable." Another explanation invokes Tiebout's point about differing "packages" of government services. A newly consolidated government, replacing a group of smaller jurisdictions that had offered a variety of service packages with differing costs, comes under pressure to provide the same package of services to everyone—and always the highest level and greatest variety of services, since no community is willing to accept a service cut.

Tiebout's smaller-is-better theory finds interesting confirmation in National Bureau of Economic Research economist Caroline Hoxby's investigation of one key government service: schools. By Tiebout's lights, many small school districts should be better and cheaper than one huge one, because they will compete with one another to minimize costs and, within the confines of state regulations, to offer somewhat different educational "packages." Hoxby looked at metropolitan areas in which lots of small municipalities run their own school districts. Exhibit A: the Boston area, with some 70 municipal school districts half an hour from downtown. Many have only a few thousand students, compared with Dade County, Florida's Amalgamated District, with 400,000-plus students, or the LA Unified School District's 800,000 students. Sure enough, Hoxby found (after controlling for a wide range of factors, from race to income, that could explain variation) that even a fairly small increase in the number of districts—from 3 to 13, say—had a big effect. Costs per child dropped 17 percent, while reading and math scores went up 2 percent—perhaps, in part, because parents attended school meetings and other school events at a rate two-thirds higher than in larger districts.

But if you look at school districts across the country, you see relatively few Tiebout-style systems like the Boston area's but rather vast, centralized amalgamated districts that look like the metropolitanists' dream come true. And their performance doesn't do much to buttress the theory that big jurisdictions are efficient and cost-effective. Between 1952 and 1992, even though consolidation bypassed America's municipalities, it did descend upon the nation's school districts, reducing their number by over 75 percent (from 67,355 to 14,422). At the same time, costs skyrocketed and quality plummeted—not coincidentally. Nor is it a coincidence that teachers' unions swelled in power: larger jurisdictions put average citizens at a disadvantage, since even the most zealous unpaid neighborhood activist is little match for the full-time paid staffs of public-sector unions, who know local officials and understand how the system works well enough to make the interest of their members prevail.

In the same way, as government jurisdictions get larger, control gradually melts away from voters; realizing the difficulty of influencing officials, and increasingly impotent against the organized electoral power of public employees, individuals give up. As Jeff Brain likes to point out, voter participation is much lower in the San Fernando Valley areas that are part of Los Angeles than in those that are independent municipalities. Growing voter apathy gives organized public employees and other special interests a clear field to advance their own agendas, while the higher campaign spending that comes with big government allows unions and government contractors to sway officials by providing campaign funds and volunteers.

Political science sheds further light on why voters tune out. Successful governmental systems, political scientists believe, have a high degree of political homogeneity—where voters generally share similar preferences—in contrast to political heterogeneity, with tightly wound tensions among many disparate voting groups. Given a large number of small jurisdictions, voters can sort themselves out according to what kind of place they want to live in, and they can pick the kind of representation they want. Homogeneity is high, and even losing voters are not violently disappointed and disaffected, since they live within the same universe of desires as the winners. There is no large group of voters whose dearest hopes have been dashed and who feel impotent and excluded. Maybe that's why polls consistently show that local government is much more popular than the federal government—which has taken on more and more tasks for itself but is not very likely to reflect a homogeneous preference. The title of a 1959 essay by James Pennock sums up this idea with still-unsurpassed brevity: "Federal and Unitary Government: Disharmony and Frustration."

Not only is metropolitanist Rusk mistaken when he asserts that bigger jurisdictions are more efficient than smaller; he is equally in error when he claims that they promote growth better. He contends that "elastic" cities—those empowered to annex new-growth suburbs—will economically outperform "inelastic" cities, frozen into historic boundaries. "Metro areas containing elastic cities," he pronounces, "have had higher growth rates than metro areas with inelastic cities." But economists John Blair, Samuel Staley, and Zhongcai Zhang have left this assertion in tatters. Rusk appears to have made the classic mistake of confusing correlation with causality. If cities that annexed their suburbs after World War II prospered, that does not prove that the annexations sparked the prosperity, and certainly Rusk offers no explanation of why they would. These economists observe that the cities Rusk singles out as elasticity success stories outperformed "for reasons unrelated to elasticity": Rusk compared newer-growth, non-manufacturing cities with older, manufacturing-based cities during a period of manufacturing decline, and he compared state capitals to non-capitals during a time of governmental growth. Finally, they note, if you compare the entire metropolitan areas of inelastic cities with the metropolitan areas of their elastic rivals, there's not much difference. Rusk himself notes that elastic metro Houston built new housing for 2.4 million new residents between 1950 and 1990, while inelastic metro Detroit built housing for 1.9 million. Not very persuasive evidence of elasticity's superiority.

The reverse of Rusk's proposition is closer to the truth: metro government is more likely to discourage than to foster growth. And further, quite possibly the older, inner-city neighborhoods Rusk is most concerned about—for him they are the most telling indictment of inelasticity—are depressed partly because of their imprisonment within the growth-stifling big-city governments we already have.

Why? Consider what I'll call the golden goose effect. Communities are willing (indeed, eager) to accept new development, the golden goose, so long as they can be sure of getting the golden eggs—strengthening their tax base and adding or improving such neighborhood amenities as schools, parks, or police protection. Metro government changes this whole calculation. Suddenly there is no guarantee that city hall will use new tax revenues the neighborhood generates to improve the neighborhood. Without the assurance that it will be able to make use of new taxes, a community's incentives change dramatically. Suddenly new developments bring a guarantee of costs but not of benefits. Areas asked to accept the new industrial park may get no improved services or new school buildings; the additional tax revenue, if not simply swallowed up in the day-to-day administration of the consolidated government, may well be spent in other parts of the city—probably those with the most political clout, which will probably not be the poorer areas.

In fact, poorer parts of cities, much as they would presumably be eager to attract new development, have reasons to be wary of it: they may get the smoke and noise, while more affluent and politically powerful neighborhoods get superior services. I recall, as a young newspaper reporter, covering the story of a home owner in Boston's Roxbury black ghetto who fought bitterly, and successfully, to close down a large neighborhood baking plant. Why should she put up with noise and trash, she reasoned—and still have to live in an area where streets were poorly swept and police response time poor? The same reasoning helps explain neighborhood opposition to new superstores in New York. Who wants the traffic without a guarantee of improved services?

Helping these poor neighborhoods is Rusk's real agenda for metro government, which at heart he believes will help them not so much by facilitating economic growth as by facilitating redistribution. Rusk discloses his true priorities in his book's conclusion: "It is not important that local residents have their garbage picked up by a metrowide garbage service or their parks managed by a metrowide parks and recreation department," he writes. "It is important that all local governments pursue common policies that will diminish racial and economic segregation. In baldest terms, sustained success requires moving poor people from bad city neighborhoods to good suburban neighborhoods and moving dollars from relatively wealthy suburban governments to poorer city governments."

Stripped of its pretenses about efficiency and economic growth, the metro movement turns out to be a campaign to support and enlarge the growing package of social services that our big cities provide—what might be called the municipal welfare state. Rusk asserts that those who have fled the crime and disorder of inner cities must be joined with those who have been "left behind." Even though these voters might prefer government that provides a menu of basic services—police and fire protection, public works, education—they should instead underwrite a social service edifice for the poor. Rusk never considers the possibility that these citizens recognize that the existing welfare state is an abject failure that has worsened urban life for everyone, taxing wage earners excessively and intensifying the dysfunction that Rusk deplores in underclass neighborhoods. He doesn't dream that expanding city borders to grab the wealth of those who have fled failed urban policies would simply extend the problem, not solve it. He imagines that middle-class habits will rub off on the poor if government drops them into middle-class neighborhoods, rather than understanding that it is the incentive of reaching a better neighborhood that encourages the habits of thrift and industry. Nor does he notice, or credit, the success of so many who by their own efforts have propelled themselves out of the inner city into the suburbs over the past two decades.

Instead of promising more of the redistributionist machinery that has failed so roundly over the last generation, breaking up the cities holds out a different, more valuable promise to poor neighborhoods: it offers them the incentive and the means to encourage economic growth. Knowing for sure, as suburbs now do, that they will benefit directly from new investment, poor municipalities would try to make their business climate accommodating—to developers, say, or street vendors or small, home-based businesses that might be zoned out elsewhere. Independent municipalities will have the option of limiting regulation and accepting employment-generating businesses, even waste-recycling centers or power plants, that middle-class areas may not want, but whose financial benefits a poor community might find well worth the costs. (State and federal health and safety standards would still apply, of course.) Neighborhoods would have the incentive to find the highest and best use of their land and buildings.

The fact that the Camdens and Newarks of the world have not prospered should not be taken as evidence that poor but independent cities can't succeed. For 30 years, such cities have focused on seeking state and federal aid for public works projects or the rebuilding of formerly middle-class housing, even after the middle class has fled, rather than on figuring out the economic advantages they might themselves offer. Poor but independent communities might, to be sure, become places where corrupt local officials take power; but they might, just as well, become places that can develop their own competent leadership, based on the exercise of real authority.

Whatever the complications very poor neighborhoods pose to creating a system of independent city neighborhoods, they should not obscure the tremendous benefits that such a system would bring the vast majority of neighborhoods. Not the least of these would be the new political cultures that will have a chance to take root, more communal and truly democratic than today's. Here's how it would work.

Neighborhoods that already have their own, informal identity, and often their own zip codes, would become formal municipalities. In New York, for instance, this would mean that the Upper West Side, Harlem, the Upper East Side, the Village, Canarsie, the Rockaways, Forest Hills—all the city's identifiable residential neighborhoods—would become independent municipalities, empowered to set their own property-tax rates, operate their own police and fire departments, make their own zoning and land-use decisions, pick up their own garbage, and clean and repair their own streets. This does not mean, however, that each municipality will in fact do all these things. Like suburbs, they will provide some services themselves and contract out others to either a private firm or another public entity, typically a county.

Just outside the Los Angeles city limits, for instance, a large number of so-called contract municipalities, members of the Los Angeles County Association of Contract Cities, contract for law enforcement services from the county sheriff's office. Cities like Bradbury and Rolling Hills have taken this as far as it can go: with but a handful of employees, they contract for all their services. Around Tucson, discussions are currently intense about what price Pima County should charge newly incorporated towns for, say, a week of a deputy sheriff's time. Over time, individual municipalities will doubtless figure out which services should remain local, which joint effort best provides, and what is the best way to pressure outside contractors to keep prices down. Key Biscayne attorney Eugene Stearns believes that ordinary police protection is best provided at the local level but that the cost of bomb squads or SWAT teams—operations that any single jurisdiction rarely uses—should be shared. Similarly, it may make sense for municipalities to operate their own school systems, as Caroline Hoxby's research suggests, but to form a buyer's pool for select educational goods or services. The costs and revenues of those services for which metropolitan economies of scale exist—as Milan Dluhy's research indicates—could be shared through special-purpose districts that would oversee airports, say, or libraries or arterial roads. Such districts have swiftly and steadily proliferated over the last two decades.

Key to any urban breakup would be figuring out which functions would be local and which would be regional. And what amount of the existing debt would each new municipality inherit? The residents of a jurisdiction containing the region's water treatment plant, clearly, would not become solely liable for the debt payments related to its construction. But what about school buildings or recreation facilities? "We expect," observes Richard Close, an attorney who has actively promoted San Fernando Valley secession, "that figuring out the terms of separation will keep a major accounting firm busy for several years."

Not every public facility or geographic area should come under the control of those who live in or near it. Voters in an entire metropolitan area should control what Eugene Stearns calls "regionally significant areas," either through regional votes or through appointed commissions chosen from the entire metropolitan area. In practice, this would mean that such entities as airport authorities, metropolitan-wide water supply and sewage districts, and port authorities would function as before.

It would also mean that a metropolitan area's central business district would be governed by a mayor and a city council whom voters from the entire metro area would elect. Who exactly would be eligible to vote? The Census Bureau already provides a framework for eligibility by delineating which municipalities—by virtue of their preponderance of economic ties—lie within a single "standard metropolitan statistical area." Those living within the New York SMSA, including those in parts of Nassau and Westchester counties—and perhaps parts of New Jersey and Connecticut as well—would have the right to vote for a metro mayor and council, who would set policy for midtown and lower Manhattan. Suburbanites would also get to vote for their own mayors, just as residents of the former City of New York would get to vote for their local neighborhood mayors, too.

Some new municipalities wouldn't have to raise all their tax revenues internally. Eugene Stearns—who envisions conjuring such new municipalities as Little Havana, Little Managua, and Coconut Grove out of Miami—proposes that well-to-do municipalities share a percentage of their local revenues with poorer communities. "I want to make clear," he says, "that this is not a rich man's attempt to avoid the cost of government." New municipal boundaries, he suggests, might divvy up poor communities among richer neighboring areas. Any municipal breakup campaign must recognize that some sort of regional revenue sharing will have to guarantee that all areas have decent schools and adequate police and fire protection. But assuring, say, a floor below which school spending must not fall is a far cry from aiming at overall tax equalization.

Stearns's proposal should serve as the start of a discussion, not the final word. What percentage of tax revenues should municipalities share? And who should pay it? All communities above a certain income level? All communities above the property tax median? How would cities like New York, which finance public assistance in part locally, finance it after breakup—and who would determine the levels of support, since cutting the size of the dysfunctional edifice of low-income assistance is a likely and desirable outcome of breaking up the cities? Only the winnowing of the political process will answer such questions.

The one thing we shouldn't do is to use the central business district's property-tax revenues to subsidize tax rates in residential neighborhoods. Such redistribution dampens business development. The better approach is to use Stearns's revenue sharing to accomplish desirable redistribution desirable and to reduce, radically, central business-district taxes. Let the office buildings and residents of the economic heart of the city pay to support the actual services they require; beyond that, let the economic value they generate be reflected in the property values—and property taxes—of the residential areas. Keeping commercial taxes low will keep businesses in the cities—and preclude politicians from cutting deals with favored firms, offering lower taxes in exchange for staying put.

This is a radical proposal, true. The neighborhoods of our big cities have long been bound together; residents think of themselves as New Yorkers or Chicagoans. But they also think of themselves as residents of Flatbush and Canarsie—and in that capacity they lack the means of exerting political control over the places they call home, in contrast to people short distances away on Long Island or in Westchester County, who have, in effect, greater rights to influence where they live. Perhaps the strongest argument in favor of the unthinkable possibility of breaking up the cities is that the movement to do so has already, spontaneously, begun. David Rusk has written that we should not consider the "political geography of mature metropolitan areas" to be "immutable." Just so—but not in the way he believes.


City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next