Eva C. Galambos, Ph.D.
For twenty years, Atlanta and Sandy Springs have been at odds about the incorporation of Sandy Springs. Atlanta would like to expand its boundaries north into Sandy Springs, while the residents of Sandy Springs would like to establish their own city. Fulton County, which provides local services in Sandy Springs, has endorsed by resolution the right of the people of Sandy Springs to decide their own future.
The expansion of Atlanta to encompass Sandy Springs could occur in two ways: 1) by assent of those to be annexed in a referendum, or 2) by consolidation of Atlanta and Fulton County, also requiring a referendum. A rational approach in this debate should center on which of the following alternatives is more likely to result in efficiency in local government: Centralization into a larger city, or a move toward decentralization?
In the past century, voters throughout the U.S. have tended to say “no” in referendums on consolidation. From 1921 through 1996, only 23 of 134 referendums on consolidation were approved.1 Since 1976, the handful of “yes” votes for consolidation has occurred in smaller jurisdictions with a combined population below 250,000. Atlanta’s current population is 426,600, and Sandy Springs’ is 79,600. In addition, voters have not exhibited an enthusiasm for annexations. During the 1980s, the average annual population annexed into cities was 25 percent lower than in the previous decade.2 Why is the trend away from centralization for local government?
Although on paper the organizational charts of proposed consolidated governments seem orderly, voters have distrusted recommendations of planners and political scientists who traditionally have postulated that “fragmentation” of local government is not rational. The voters have instinctively sided with smaller, more accountable, decentralized governments instead of metropolitan government or major consolidations.
In recent years, economists have endorsed the fragmented, decentralized model. They have concluded that competition between numerous adjoining jurisdictions keeps total expenditures down in metropolitan areas.3 The multiplicity of local governments allows for contracting among jurisdictions for services when local elected officials determine buying from a neighbor is less costly than producing the service in-house. The end result is a “local public economy” — a quiltwork of service deliveries determined by decentralized elected officials who are responsive to their voters. As summarized by the U.S. Advisory Commission on Intergovernmental Affairs, “…a diversity of local governments can promote key values of democratic government — namely, efficiency, equity, responsiveness, accountability, and self-governance.”4
What reasons and data do economists offer to justify smaller governments (or “fragmentation”) instead of moving toward centralized local government? The optimum allocation of resources, or economic efficiency, is the underlying principle. The true level and mixture of local services taxpayers wish to purchase (police and fire protection, parks, sanitation services, and sidewalks, for example) is more likely to be reflected when different packages of goods are offered by competing local jurisdictions. As people vote with their feet, they move to the jurisdiction that offers the combination and level of local services at a price (or level of taxes) that best suits them. Some may deem backyard garbage collection so important that they wish to pay for it. Others may opt for a less costly method, but highly value police cruising through their neighborhoods. When only one level of local service is available, as offered by one centralized government, people have no choice. Thus there is no competition, and taxes are likely to be higher.
Service by the author on a task force that grappled with consolidation of services vividly illustrates this “choice” principle.5 The appointed group considered possible consolidation of some services in Fulton County. An academic consultant presented to the citizen task force the option of consolidating the municipal and county fire departments. The supposed benefit of this action was to raise the nine municipalities, and unincorporated Fulton County, up to the highest (and most expensive) level of fire protection offered by the City of Atlanta. However, it obviously never occurred to this “ivory tower” planner that there are areas that are perfectly happy with their current level of service. Roswell, for example, operates partly on a volunteer fire department. Given a choice, some citizens opt for what may seem a less sophisticated service in return for lower taxes.
Simply stated: expenditures for fire service are lower with several smaller governments offering various levels of service than when all are merged to the highest level. Consolidation and centralization lead to uniformity at the most expensive level, thereby negating promises of savings presented in justification of consolidation. Similarly, pay and benefit packages vary between jurisdictions. Thus when consolidation occurs, invariably all employees are moved to the highest pay scale. This entails not lower, but higher expenditures. In fact, the consolidation of Athens and Clarke County resulted in an immediate 6 percent increase of the total payroll.6
Unlike the merger of private firms where restructuring results in layoffs, in the public sector the political power of current employees translates into guaranteed continued employment. This cancels promises of savings to the taxpayer.
Economies of Scale
The U-shaped cost curve familiar to all students of Economics 101 illustrates that the per capita expenditures of local governments are high for the smallest jurisdictions, decline for middle-sized areas, and rise again for larger jurisdictions. This is easy to understand. For a very small city, overhead costs and the expenses for capital equipment cannot be spread over as many citizens, and thus unit costs are high. As a city’s population grows, the fixed costs may be spread more widely to reduce unit expenditures. However, as cities grow larger, the layers of bureaucracy rapidly grow and the economies of scale for capital expenditures are soon overwhelmed by the diseconomies of a growing bureaucracy. This effect accounts for the finding by the Carl Vinson Institute of Government that governmental consolidations for jurisdictions over 250,000 in population are unlikely to result in cost savings.7
There is real justification for merging capital-intensive services, such as water production and sewage treatment. Plants may be expanded to handle a greater capacity without incurring major labor costs, and the capital cost becomes less as it is spread over more users. Thus the Fulton County task force recommended consolidation (and privatization) for the treatment of water and sewage for Atlanta and Fulton County.8
In practice, “fragmented” local governments long ago worked out economies of scale by contracting with each other when it makes sense to buy from another provider. Contracting for services permits judicious local government managers to shop in the market for public goods, just as private managers do. What is most economical for a small city? To contract from a larger one for a service that enjoys economies of scale or to produce the service internally?
In the Atlanta metro area, as everywhere in the U.S., local officials apply common sense to stretch tax dollars. For example, in South Fulton, several municipalities jointly built and operated a new jail, which they are now privatizing. This illustrates that local governments produce services jointly when it is economical to do so on a larger scale, while preserving their local autonomy.
The Lakewood Plan in California is the most widely recognized model for contracting. It began in 1954 with the City of Lakewood contracting for many services with Los Angeles County and other cities.9 Only services benefiting from close supervision by, and responsiveness to, the local people were actually produced individually by Lakewood and municipalities that followed this model.
Studies of police services in metropolitan areas have demonstrated that smaller governments handle police protection effectively (shorter response time) when each produces the patrol function separately. Contracting with larger units, or with one another, produces more effective criminal investigation, training and other specialized functions. The most decentralized service is the one closest to the constituents — patrolling of streets and neighborhoods.10
“Duplication” Is an Erroneous Term
In 1988, the 64 full-time police departments in the St. Louis area did not produce “duplication,” but rather “specialization.” “Duplication” is a term that is often applied erroneously in the Atlanta area to refer to several jurisdictions that each supply the same service to their own constituencies. If two departments were both serving one constituency, then obviously they would be duplicating a service, but that is not what happens.
In the St. Louis area, the police departments do not overlap in their patrol functions and they coordinate their more specialized services.11 In the Atlanta area, this common-sense approach is also occurring as local officials cooperate to make tax dollars go further. They contract with each other, and the State, for police training in centralized “academies,” and some have joint radio communication systems. Within Fulton County, only the City of Atlanta refused to join the countywide radio system.
A review of 25 research studies conducted over the past two decades on “fragmentation” versus centralization in U.S. local governments suggests “…local government systems which are fragmented and deconcentrated are generally associated with lower spending and greater efficiency.”12
The power of bureaucracies grows the larger the centralized government becomes. This is evident in the difficulty locally elected officials have in privatizing municipal services in large cities. They are hemmed in by empire-building bureaucracies and government employee unions, which are stronger in the larger “fragmented” municipalities.
Consolidation or centralization is sometimes promoted as an economic development tool. The assumption is that new industry is more likely to be attracted to a centralized local government area than to one with several jurisdictions. However, a recent analysis of the nine largest consolidations, such as Jacksonville, Florida, and Nashville, Tennessee, concludes that the number of manufacturing and non-manufacturing business establishments showed no significant difference in growth in the consolidated areas as compared to similar “fragmented” areas.13
Sharing the Burdens of Social Services
The strongest argument central cities make for their desire to annex is not that citizens will gain more efficient government, but that the social service needs of the central city residents cannot be met without the suburban tax base. This argument makes no sense in the debate as to whether Atlanta should annex Sandy Springs. The responsibility for social and health services rests with Fulton County — not Atlanta. It is the Fulton County General Fund that supports Grady Memorial Hospital, health clinics, social services, senior services, and a host of special grants for indigent services. The suburban residents of Fulton County provide their share of this cost in proportion to the property tax base located in the suburbs. It has been estimated that more than 40 percent of the property taxes collected by Fulton County are derived from North Fulton County. This would not change with the formation of the City of Sandy Springs.
Regionalism and the City of Sandy Springs
A rational review of the Atlanta/Sandy Springs disagreement does not produce reasons to oppose formation of the City of Sandy Springs. Nor does the current push toward “regionalism” preclude formation of another jurisdiction in the metro Atlanta area. The priorities for regional consideration are transportation and accompanying major policies regarding land use. Regional action on these matters is compatible with a multiplicity of local jurisdictions in Metro Atlanta that provide local services to their constituents. The dual track of regional decisions on matters that transcend local boundaries (traffic, air pollution, and water pollution, for example) and city governments that are accountable and responsive to their citizens for everyday local matters may be the ideal solution.
Annexation of Sandy Springs by Atlanta would not deal with the truly regional issues that affect our air and water. Nor would it improve the efficiency of local service delivery. Incorporation of Sandy Springs, on the other hand, would have no deleterious effect on moving toward regional decisions where they are needed. It would afford 80,000 residents an opportunity for a responsive and accountable local government to meet their purely local services.
3. Carl Vinson Institute of Government, University of Georgia, The Impact of City-County Government Consolidation: A Review of Research Findings (Athens, Ga.), 1989, pp. 9-12, and Platon N. Rigos, “Metro Governance Policy: Forty-Five Years of Experimentation, Research and Ideological Conflict,” Department of Government and International Affairs, University of South Florida (Tampa, FL), 1994, unpublished
Eva C. Galambos is an economic consultant with wide experience in local government fiscal affairs. She also serves as the president of the Committee for Sandy Springs, Inc. The Georgia Public Policy Foundation is a nonpartisan, member-supported research and education organization based in Atlanta, Georgia, that promotes free markets, limited government and individual responsibility. Nothing written here is to be construed as necessarily reflecting the views of the Georgia Public Policy Foundation or as an attempt to aid or hinder the passage of any bill before the U.S. Congress or the Georgia Legislature.
© Georgia Public Policy Foundation (November 3, 1999) Permission is hereby given to reprint this article, with appropriate credit given.
The best way to make a lasting impact on public policy is to change public opinion. When you change the beliefs of the people; the politicians and political parties change with them.