Let’s Break Up The Big Cities
Why Bigger Local Government Isn’t More Efficient: The Case for Breaking up Cities
Speech given by Howard Husock, Director of Case Studies at John F. Kennedy School of Government, Harvard University, on Friday, May 18, 2001 at Montreal’s Omni Hotel.
Thank you. It’s great to be here in the city that Boston television weather forecasters never fail to cite as the source of all our cold weather, what they like to call the Montreal Express. Something tells me your terminology might just be different. That’s a lesson I learned when I visited Hong Kong recently and was told by the local health authorities that the virulent 1968 flu epidemic we called the Hong Kong flu was known in Hong Kong as the Sydney Flu.
There’s no doubt, however, that there really is a Montreal Express – and a Toronto Express and a Halifax express – when the subject is amalgamation of local governments, a process which has been moving forward at something like a breakneck pace. But, like the Hong Kong flu and the Sydney flu, the same thing can have different effects and be perceived differently by differing populations. When it comes to municipal amalgamation, there is an analogous movement in the United States, but it has largely not gotten off the drawing boards. In fact, the most powerful, grassroots forces in the United States are moving in just the opposite direction: toward breaking up some of our big cities into more manageable units of government, as well as establishing small, new municipalities in areas which had been governed by large, regional county governments.
The movement to decentralize municipal governance reflects a commonsense point of view, richly supported by empirical evidence, that makes clear that the claims made for amalgamated so-called mega-cities are more myth than fact.
In the U.S., the amalgamation idea – most famously championed by the former mayor of Albuquerque, New Mexico, David Rusk, in his book Cities Without Suburbs – is seen as a means to redirect some of the wealth of current suburban jurisdictions to our deteriorated inner city neighborhoods. At the same time, it’s claimed that bigger cities will be better able to compete economically, presumably by recruiting big employers.
Crucial to this point of view is a crucial mistake, the mistaken idea that bigger government can be more efficient and effective government. Proponents of regionalism, both here in Quebec and in the U.S., both look to such alleged efficiency as a way to generate cost-savings that will help fund the causes they seek to support.
Yet study after study has shown that the efficiency gains of bigger government do not materialize. This was the conclusion of the study by Andrew Sanction of the University of Western Ontario with which I know many of you are familiar. A similar study of the Miami area by faculty at Florida International University found that for the most part, economies of scale in municipal government are illusory. Professor Milan Dluhy found that such economies existed there in only two areas: fire protection and library services. Localities, he found, can provide other services-police, recreation, public works, waste management – at equal or less cost than an amalgamated, or, in the U.S. context, county jurisdiction. The recent budget problems of an amalgamated Toronto would seem, too, to undercut the idea of realizing easy efficiency gains this way. Somehow, however, the myth of efficiency through amalgamation lives on, so that it is worth explaining exactly why bigger government won’t save money. And it is worth considering the possibility that the greatest efficiency gains may actually be realized by moving in the opposite direction: breaking cities up into their component neighborhoods.
Why exactly isn’t big government more efficient government? Forty years ago, in a short but classic essay, an economist named Charles Tiebout provided the key insight. Local governments in metropolitan areas, he wrote, do much more than merely co-exist side by side. They compete with one another. Not all of us want the same things from local governments. Those with small children may care most about education. Joggers may want to spend more on parks. Others may want their streets cleaned three times a week. Cultural difference, too, is part of the mix. Separate municipalities offer the chance for residents to emphasize different values with a minimum of political friction. The patriotic may wish to spend local funds to celebrate Dominion Day. Those with other political leanings may wish to support a sister city program linking themselves with, say, a town in Cuba. The overall point is this: small communities can offer differing packages of services and amenities and we can vote with our feet as to which ones we prefer. Moreover, even when they offer the same sorts of services, they compete as to which can deliver them more efficiently. The town which offers the package of services most like that which you want and delivers at the lowest tax rate will get your vote, in effect. You’ll move in. If things change, you may well move out. We know that competition disciplines the private marketplaces; so, too, does it discipline the public one.
The late Daniel Elazar, who, for many years was director Temple University’s Center for the Study of Federalism, pointed out that some of the best managed cities in the United States are those in which metropolitan areas are divided amongst a number of jurisdictions: the Bay Area, where San Francisco, Oakland and San Jose complement one another; or the prosperous and efficient Twin Cities, Minneapolis and St. Paul. It’s worth noting that the Bay area, particularly, in which the three major jurisdictions are complemented by numerous smaller municipalities such as Sunnyvale and Redwood City, has emerged as the most globally competitive metropolitan area in the entire United States. This, after all, is Silicon Valley.
Some of the most telling evidence of the potential efficiency of a competitive municipal marketplace comes from a brilliant young economist named Caroline Hoxby of the National Bureau of Economic Research in Cambridge, Massachusetts. Hoxby compared the effectiveness and cost of public education in metropolitan areas with a multitude of small school districts, such as the Boston area, versus those with large, amalgamated school districts, such as New York and Los Angeles. She found that, after adjusting for factors such as income and race, as the number of districts increased, student achievement went up, though modestly – reading scores rose 2 percent, on average – while cost went down significantly – 17 percent on average.
All this is well and good but I’ve still not fully answered the question of what exactly happens when local government gets larger, or why larger government is less likely to be efficient. The answer involves, first, what the voters may want; second what they may not want. First, as amalgamation occurs, all the various services and amenities packages of the many individual jurisdictions get put together. No one wants services to be reduced – but, as a matter of political reality, no municipality can provide certain services only to one area. All the jogging enthusiasts who’d previously been outnumbered in Jurisdiction A, suddenly can turn to the new Mega-City and demand that, because there are great jogging trails in what used to be Jurisdiction B, they, too, deserve such amenities. In other words, rather than being reduced, service provision inevitably rises to meet the many tastes that had previously been separate. Cost increases as facilities as basketball courts are built in areas which previously had made them a lower priority. Public employment must necessarily increase, not decrease.
At the same, time, government, in an amalgamated city, inevitably becomes more distant from the individual voter. It is harder for any one voter, or group of voters, to influence policy. This situation works to the advantage of well-organized interest groups, with the resources to employ staff to influence policy on their behalf. Even the most zealous unpaid neighborhood activist is little match for the full-time paid staffs of public sector labor unions, for instance, who know local officials, help elect them and understand how the system works. Inevitably, unions will, in representing the interests of their members, resist cutbacks in municipal employment. They will insist that the efficiency gains of smaller municipalities be eliminated. Thus, for instance, if Jurisdiction A formerly paid its recreation workers less than those in Jurisdiction B, where recreation was not as important to voters, we can expect that the new amalgamated city will have just one pay scale – at the higher rate.
This is true because interest groups will have more power and leverage in an amalgamated city. It is one thing for police to stage a job action that affects a small jurisdiction; quite another if it affects the entire metropolitan area.
In my opinion, then, not only should cities avoid amalgamation with other municipalities, they should consider doing just the opposite. In order to achieve the political and efficiency gains that the Tiebout theory predicts, instead of swallowing their neighbors, they should break up into their own component parts. Movements in this direction are, in fact, going on all over the United States. In the Miami area, eight new municipalities have incorporated in the last ten years, in effect seceding from the mammoth Dade County government. In Los Angeles, second-largest U.S. city, a highly-organized movement called Valley Vote seeks to split off the city’s San Fernando Valley neighborhoods, with a population over a million, from the rest of the city. A referendum is expected next year. If they win, proponents may consider further divided the Valley into 20 or more still smaller jurisdictions. In New York, the largest U.S. city, in the early 1990s, one of the five boroughs, Staten Island, voted outright to secede and was thwarted only because state law required approval from the state government, which would not agree.
In addition to alleged efficiency gains, proponents of regional municipal government claim, too, that somehow mega-cities will be better able to compete in the global marketplace. This is a misapprehension of how cities actually become and remain prosperous and what would likely occur after amalgamation. First, as Jane Jacobs has said convincingly in her classic book The Economy of Cities, healthy cities are not chasing after businesses from elsewhere. Instead, they are nurturing their own, indigenous new and better ideas. This is the story of Silicon Valley, or even Starbucks in Seattle. You can be sure that if Seattle had set its restaurant tax too high, it might have strangled Starbucks in its crib. The best way forward for cities is to keep taxes low, to allow the easy conversion of buildings to new uses, to limit the number of license and regulatory requirements for new businesses.
What’s more, when and if new or expanded businesses are seeking to build new plants or warehouses, mega-city governments can be an impediment, not a facilitator. Here’s why. One of the reasons that citizens agree to the permits or zoning changes which allow new businesses to locate in their municipalities is the belief that the benefits they will gain – in employment and municipal taxes – will offset their costs, such as greater congestion. In a mega-city, citizens in one small neighborhood will fear that they are more likely to incur costs than to see benefits – and to employ "Not in My Backyard" tactics to stop new construction. In fact, proponents of amalgamation make clear that they do, indeed, seek to redistribute wealth from more affluent neighborhoods to poorer ones. The result will be less wealth, overall – as wealthier neighborhoods realize that they have less and less reason to accept growth. Because growth, no matter where it occurs, provides jobs for residents from throughout the metropolitan area, this demagogic attack on better-off neighborhoods will hurt the poor, as well.
How can poorer neighborhoods improve their condition? They must heed the real lesson of globalization – they must compete by lowering their cost of doing business. I’m not familiar with the situation in Montreal but in most metropolitan areas in the U.S., the big cities have by far the highest cost per capita of government. I was asked, for instance, to study the city of Kalamazoo, Michigan, which was pushing for what it called a regional compact, under which its suburban areas would share tax revenue with the city. My analysis found that in every area of municipal spending – police, public works, recreation – Kalamazoo’s costs were half again as great as the average of its suburban neighborhoods. The city didn’t have a funding problem – it had a spending problem. And, as I said earlier, amalgamation will not lower costs but, inevitably, drive them up.
But now let me make a quite different point. Breaking up the cities, or preserving existing independent municipalities, does not mean there is no room for regionalism. On the contrary, independent municipalities can work together on those specific activities for which it is in their best interest to cooperate. In the U.S., we call such arrangements special purpose districts. Thus, in the Boston area, even though we have dozens of independent cities and towns, there is but one metropolitan-wide public transit district and one metropolitan-wide water and sewer district. Similarly, municipalities agree to provide what they call fire protection mutual aid: in the case of large fires, fire trucks from throughout the area provide back-up capacity – so that no one municipality has to pay for trucks that will go unused most of the time. In other words, breaking up the cities could, rather than being the opposite of regionalism, could set the stage for effective regionalism.
In Erie County, New York, for instance – that’s Buffalo, an area struggling to become economically competitive – the county executive has recently started what he calls the "Who does What Commission." Members are examining what sorts of services might be better purchased or provided by groups of municipalities, rather than each going it alone. They have begun to save money on such mundane services as property assessment – without asking voters to sacrifice the control over their local affairs that they so rightly value.
In the Canadian context, it is quite possible, for instance, that the so-called downloading of social welfare responsibilities – in the motivation of the conservative government there to create the Toronto mega-city – could have been handled by creating a special service district for such purposes only, without resorting, amalgamation.
If one looks at the sweep of history, amalgamations tend to be the dreams of planners, while the formation of individual home towns tend to be the preference of the general voter. In the U. S., even as some have called for amalgamation, the number of municipalities has continued to grow: from 16,800 in 1952, to 19,300 in 1992, for instance. The average citizen understands the virtues of smaller, local government, tempered with appropriate regional cooperation. Advocates of amalgamation ignore this wisdom at their own risk.