Property Values
Police Services Resident Participation Monopoly
Government, Fragmentation and CompetitionPublic Employees Social Programs Income RedistributionPublished in the The Milken Institute Review Jobs & Capital Volume VII, Number 2 / Spring 1998
Click here to see it in original format. Index added here
Property Values
Police Services Resident Participation Monopoly
Government, Fragmentation and CompetitionPublic Employees Social Programs Income Redistribution
By Shirley Svorny
At issue in large cities today is whether residents are well served by their local governments. Can a city such as Los Angeles, with a population of 3.6 million, efficiently provide local government services? Intuitively, voters in smaller cities nationwide seem aware of the negative consequences of size, for they have systematically rejected efforts toward consolidation. And now, in Los Angeles, groups of residents from the San Fernando Valley to San Pedro are excited about a new state law ending the Los Angeles City Council’s veto over detachment; these groups are seriously considering the benefits of secession.
The most sought-after knowledge is what new, smaller cities would be like. People want to know if it makes any sense to have smaller jurisdictions. Do any benefits follow from jurisdictional fragmentation (having lots of cities in an area)? Does competition among jurisdictions make government more efficient? The answer from a survey of the existing literature on city size is a resounding yes. Residents’ concerns over the chaos that might result if many cities were to operate where Los Angeles is today are unfounded.
The most common perception is that it would be expensive to run so many governments. But the bulk of the evidence suggests just the opposite: The largest desirable city size is somewhere near 200,000 residents. That would allow more than 18 cities in Los Angeles.
One benefit from smaller governments would be the change in incentives facing residents, bureaucrats, and politicians. Behavior and outcomes are a function of institutional arrangements. This costly lesson from the world’s experience with socialism can be applied to the debate over city size. It is incentives that will change with institutional structure–incentives for citizen participation and monitoring, and incentives for government agencies to be more responsive to citizens’ needs.
A large city is much like a monopolistic producer in the private economy. Large cities can take advantage of individuals who live within their boundaries, misspending tax dollars and offering residents relatively shoddy public services. Furthermore, large city size facilitates the organization and domination of public policy by special interests. Writing nearly three decades ago, economist Werner Hirsch argued that, even if there were cost savings from greater city size, they could be "outweighed by inefficiencies resulting from top-heavy administration and the ills of political patronage in very large governments." He also noted that large organizations find it hard both to anticipate the need for change and to make necessary changes. Hirsch argued that small local government is the only level at which people in government "can effectively meet and engage in democratic dialogue." Politics change with city size, as residents rationally step back from large city government.
There is no lack of empirical tests for economies or diseconomies of city size. For four decades researchers have been churning out studies of the connection between city size and the production of various public services. These researchers nearly always found either constant or decreasing returns to city size for cities with more than 50,000 to 100,000 residents. It is almost as if the idea of the efficiency of consolidation, so widely accepted in lay discussions of public provision, had been set up as a straw man, to be knocked down by researchers doing empirical studies of city size.
Recent studies criticize earlier work for confounding, or failing to separate, the effects of population size from the effects of two other factors that influence efficiency in cities: production technology and the external conditions under which the services are provided. Production technology refers to the ability to turn inputs, such as firefighters and their equipment, into intermediate outputs (fire companies); it may exhibit constant, increasing, or decreasing returns to scale.
External conditions may influence both production technology and the ability of fire companies to turn intermediate outputs, such as the number of fire companies, into a final product (fire protection and fire prevention). For example, shake roofs allow fires to spread quickly, requiring more physical inputs (fire companies) to produce the same amount of protection. With respect to police protection, it may be more costly to provide services in large, densely populated cities with relatively great resident mobility.
This is because density and mobility may work together to reduce the chances an assailant will be recognized, and so cause arrests to decline. Interestingly, newer studies, which try to control for environmental factors in public service provision or to isolate the economies of scale associated with population from other economies, still find no evidence to favor larger city sizes. William Duncombe and John Yinger, researchers at the Syracuse University Metropolitan Studies Program, apply newer methods to examine the economies of fire protection, and conclude that "consolidating small fire departments will not result in significant cost savings." With respect to policing, two studies of police provision (in Florida and Los Angeles) that controlled for characteristics thought to affect the cost of policing in large urban areas, such as poverty and home ownership, found decreasing returns to scale in police provision.
Some have argued that city size is important to the production of goods that have some aspect of "publicness" to their consumption–goods that are jointly consumed. But looking at police and fire protection, sanitation, public health, and recreation, all arguably public goods, researchers have been unable to find that small communities are disadvantaged in providing these services.
We are faced with a literature that has found no evidence of economies of size to support arguments for consolidation. The evidence runs counter to claims that breaking up cities the size of Los Angeles would raise the costs of service provision. Intuition tells us this is what we might have expected. Most of the services that cities provide cannot be produced in large-scale facilities. Each neighborhood requires its own cleaning crew, its own set of police, its own firehouse, its own school, and its own park. This means that the optimal crew size in each of these activities will be small. Jurisdiction size can be expected to work against performance as communication and coordination costs increase.
In private industry, large scale tends to be associated with capital-intensive production, where labor costs are relatively minor. But because the services cities provide are labor-intensive, requiring work more like what can be found in a restaurant than in an automobile manufacturing facility, it makes sense that larger size would be unnecessary and a detriment to the efficient provision of services.
For the very few services where size is an advantage–such as environmental protection–separate regional jurisdictions make sense. Over the years, for many services, technological innovation has reduced the smallest efficient scale. This has occurred with respect to the provision of electric power and may be in the works for wastewater treatment, eliminating the need for relatively large wastewater and power jurisdictions. back to index
Variations in property values provide a further test of the relative efficiency of public service provision in cities of different size. The logic is that, where services are provided efficiently, residents will be better served and property values will rise. One study compared property values in small, independent cities in the Los Angeles metropolitan area to those in cities that purchased services from the County of Los Angeles.
Controlling for other factors, property values were lower in the contract cities–served by the larger provider–than in cities served by smaller, independent service providers.back to index
Numerous studies of the effect of city size have focused on police services because spending on police is a significant share of total spending for most cities. In Los Angeles, for example, police service expenditures represent 30 percent of the total city budget. Economist Elinor Ostrom has questioned the conventional wisdom that big is better in the provision of police services. The unique contribution of Ostrom’s work and that of her colleagues is that they used a "most similar systems" research design. This method pairs neighborhoods on the basis of race, income, density, age, and other characteristics, and examines the consequences of varying institutional arrangements for police service provision. Three studies compared similar neighborhoods, some served by small independent police departments and others by large regional service providers, in Indianapolis, Grand Rapids, and Nashville—Davidson County. No evidence was found of increased satisfaction or lower cost in areas served by larger providers.
To further examine the relationship between police jurisdiction size and satisfaction, Ostrom surveyed residents in the St. Louis metropolitan area, some of whom were served by the relatively large St. Louis Police Department (2,200 officers), some by the St. Louis County Police Department (436 officers), and others by 63 independent police departments. The study focused on patrol services, immediate response to reported crimes, criminal investigation, emergency services, and other neighborhood level police services. Using a wide range of measures to evaluate service–from actual victimization to feelings about police service–researchers could not find a single case to support the premise that larger departments are better able to serve neighborhoods with similar characteristics.
In yet another study of the consequences of size, measures of police manpower, property crime, and violent crime for contract cities in Los Angeles County were compared to those for cities in the County served by smaller, independent police departments. By all measures, the contract cities appeared to be less well served. One expressed concern of Los Angeles residents is that jurisdictional borders will limit the effectiveness of police pursuit of criminals. It turns out, however, that state laws regarding powers of arrest and fresh pursuit across police jurisdiction boundaries give police extended powers in the case of fresh pursuit. Furthermore, in one study of medium-sized police departments, researchers’ discussions with police about arrangements for mutual assistance found interagency cooperation to be common.
One consideration related to city size about which there is no disagreement is that small governments are better suited to match the levels of public goods that are provided with the preferences of their constituents. Where divergent groups are lumped together in one political jurisdiction, as they are in large cities, majority voting leads to a less than efficient result, favoring some individuals’ tastes over others. Not everyone can be happy. This loss has been labeled a "political externality" of city size: negative effects imposed upon the minority by the majority (or majority coalition). It refers specifically to the difficulty in providing custom-tailored services to residents of large cities.back to index
Large size is also a disadvantage with respect to resident participation in government decision-making. In big cities, it is difficult for one individual or even a group of individuals to have an impact. Monitoring government policy and expenditures is nearly impossible, unless it is one’s full-time job. The resulting lack of interest in politics and policy in large political jurisdictions has been termed "rational ignorance." A study of cities with well-established neighborhood councils found them surprisingly irrelevant to levels of resident participation; only city size appeared important.
High costs of participation have a predictable consequence: Monitoring of public actions by residents is reduced and power is shifted from the populace to organized groups, such as public employees, developers, the economic development bureaucracy, and others. In a large city, coalitions of residents and lobbyists make proposals to spend other people’s money. In a smaller town, those other people might show up to complain, but as city size increases, it becomes increasingly costly to participate. Groups that have strong financial interests in manipulating city spending benefit from the lack of organized public opposition to their demands. The result is a distortion of spending and public policy away from what residents would want and toward projects and policies sought by special interests.back to index
Some researchers hold that large cities exhibit certain characteristics of private monopoly providers of services. It is argued that monopoly governments are able to extract higher benefits for bureaucrats and provide lower-quality or unwanted services to their residents. The implication is that many small cities are preferable to one large city, as jurisdictional competition will provide incentives for better public services.
The monopoly power of large cities is thought to derive from rational ignorance and the increased cost of voter mobility. Community ties make moving costly. When changing political jurisdictions requires a substantial move, say, 30 miles rather than 5 miles, individuals are more likely to choose to put up with poor government than to sever personal and professional relationships.
It has been suggested that jurisdictional fragmentation, and the resulting competition, will constrain politicians from engaging in bureau-building activities and other spending that benefits bureaucrats at the expense of taxpayers, and so lead to lower levels of government spending. However, empirical analyses have not produced consistent results. Some researchers have found evidence of a negative relationship between fragmentation at the metropolitan level and local government size, while others have not.
The inconsistent findings may reflect the fact that competition does not necessarily reduce spending. Where cities are smaller, with more homogeneous populations, it may be easier to agree on a set of desirable services. If so, the level of city service provision might actually increase. Spending does not necessarily have to fall as a result of the constraints imposed on politicians by competition. back to index
The "factor-supplier pressure group hypothesis" (so-named by economist Gordon Tullock) suggests that public employees will work to increase the demand for their own services through the political system. City size promotes this in several ways. First, there are economies of scale in union organization. Second, the monopoly power of large cities gives politicians leeway in meeting public employee demands, because rational ignorance limits the opposition from residents. Just as automobile workers were better able to extract above-market wages in less competitive markets, so can public employees. Politicians who vote to increase employment benefits for public employees can expect their organized support in future elections. (There is evidence that public employees vote more than do other registered voters.)
Economist Bernard Lentz wrote, "In a democratic government, workers become more than mere inputs of production: Their votes and political loyalty may be directly purchased through their terms and conditions of employment." Researchers have found that city size increases the likelihood of union representation and that union representation, in turn, increases compensation and benefit packages well beyond what workers could earn in the private sector. back to index
Advocates of large city governments argue that small, local jurisdictions will find it difficult to address social issues such as homelessness. If, for example, a small city were to institute a program aimed at improving the lives of homeless people, it might attract every homeless person from miles around. But this is not an intractable problem if cities are small. In the absence of a large metropolitan jurisdiction, problems such as homelessness need to be addressed by regional coalitions. These coalitions do not necessarily have to be public; regional interfaith councils or other community groups may take on these tasks. back to index
Secession is seen by some as an effort to remove a community from the income redistribution efforts of large urban cities, to limit the exposure of relatively wealthy residents to heavy responsibility for services in less wealthy communities. If there were any evidence of a transfer of benefits to poor neighborhoods from such redistribution schemes, this might be a better argument. But most funds taken from the wealthy in large cities, ostensibly to help the poor, go to bureaucrats and their supporters, who then go on to congratulate themselves on the success of their programs to help the poor. Meanwhile, economists have found these programs, involving job training and economic development, to be of little value to poor communities. Despite the trillions of dollars that have been spent, little has changed in poor neighborhoods.
This is to be expected, as the poor have the least power to shape city spending policies; their power over resources is certainly nowhere near that of public employees or nonprofit agencies, whose support can carry politicians through an election. If wealth were really being redistributed to the poor in large cities, we would see less graffiti, cleaner public facilities, fewer potholes, and more trees in the inner city.
Thirty years ago, political scientist Robert Dahl wrote, "Any unit you choose smaller than the globe itself....can be shown to be smaller than the boundaries of an urgent problem generated by...people who are outside the particular unit and hence beyond its authority. Yet the larger the unit, the greater the costs of uniform rules, the larger the minorities who cannot prevail, and the more watered down is the control of the individual citizen."
Clearly, in choosing city size, regional issues must be balanced against local needs. But the dominance in big-city budgets of services that show no advantages of size; the gains in terms of citizen participation, monitoring, and control; and the ability to tailor services to neighborhood needs strongly support the premise that residents of large cities would be better served by smaller local governments. back to index
Shirley Svorny is a professor of economics at California State University, Northridge.