PLANNING AND FINANCE SERIES: If high density reduces costs, why are public services in New York City so costly?

by Patrick L. Dugan, Everett, Washington

Recently, I gave a presentation to the Manitoba Planning Conference on the costs of sprawl. The presentation made the usual arguments that the costs of supporting new development are less when the development is more compact and denser. A member of the audience challenged that argument with the question, “If dense development is less expensive to serve, why are the costs of providing governmental services in New York so high?” Unfortunately, I did not answer the question very well in the session and this article will attempt a more thoughtful response.

This question has haunted me for some time since I have long noted that larger governments tend to impose higher tax rates and have more costly services than smaller governments. The City of Seattle imposes much higher taxes and provides more expensive services (on a per capita basis), by far, than almost all other cities in the state.

While there are certainly a number of potential explanations, the space provided for this column does not allow for as full an exploration of this issue as I might like to offer. Therefore, I will only focus on one of the more significant aspects of the question. The reason services are so costly in New York City, in spite of its density, is because the diseconomies of scale associated with its size outweigh the economies of density derived from its high density of development. These two different economic terms, “economies of scale” and “economies of density,” are often confused with each other. 

Economies of scale

Economies of scale occur when operational efficiencies resulting from increased production create a reduction in cost per unit. Economies of scale often arise by spreading fixed costs (such as capital investments) over more users. As such, economies of scale occur as the size of an enterprise increases. Generally, it is commonly believed that as local governments grow, the growth should produce economies of scale which allows cost savings to be achieved – average costs are reduced when spread out over a wider set of users. 

However, the potential benefits of increasing economies of scale must be evaluated against potential diseconomies of scale. Small organizations can often avoid costly and complex management and personnel systems that will be needed in large organizations. As the size of the jurisdiction also grows, so will communication and management costs; as anybody in a large organization knows, more people means more meetings. Increased complexity of issues and public opinion can also add to costs. 

While most believe that economies of scale continue to increase as cities grow, numerous studies1 have actually found that diseconomies of scale tend to take hold in larger governments — that there is a “U” shaped cost curve when per capita average costs (Y axis) are graphed against size of government (X axis). Average costs first decline as a small city grows into a mid-size city but then average costs gradually increase as the city continues to grow. These studies have noted that this “U” shaped curve varies for each type of service, with the optimum size being quite low for services provided by people (as low as 25,000 population for police services), but much higher for capital intensive services such as utilities (in the hundreds of thousands of people). It is not surprising to note then that the average costs of services in large cities such as New York are very high.

Economies of density

The term “economies of density” refers to the costs of providing services geographically, over distances. It cost more to run streets, water, and sewer services over two miles to serve 100 households, than it would cost to run those services over one mile to serve 100 households. While these costs are most prevalent in capital intensive services, they can also increase costs for services provided by people (it costs more to patrol two square miles at a given level of service than one square mile). These economies are achieved by reducing urban sprawl. Numerous studies have found significant economies of density.2  

While economies of scale and economies of density may be interrelated, they are driven by entirely different factors, and should not be confused even though they often are. Although there are studies of economies of scale and separate studies of economies of density, studies that look at both in one study are not very common. One exception is that there have been several studies that examined the interaction of economies of scale with economies of density in large water utilities in Europe. These studies generally tend to note that while diseconomies of scale exist in larger utilities, these diseconomies can be reduced with economies of density where there are high densities. One such study, Urban infrastructure: Density matters, not just size,3 concluded that: 

“Although there is good evidence for economies of scale in water treatment, this can be more than offset by diseconomies in distribution as the scale of settlement increases; on the other hand, higher density tends to reduce unit distribution costs whatever the size of settlement.”

Size drives expense

Any given city is a composite of services that are affected by both types of economies — scale and density. As a city grows in population, it experiences economies or diseconomies of scale. In theory, if that city grows in a compact pattern, it should have less costs for any given population size than if that city is more spread out.

So, New York City and other very large, dense cities are expensive due to their size, not necessarily because of their density. It probably would not be practical to conjecture whether costs would be greater if it were somehow less dense. However, in the smaller jurisdictions that many planners deal with, diseconomies of scale may well be increasing costs as the jurisdictions grow, even though more compact development could be reducing costs at the same time. For any given size of jurisdiction, a more dense pattern of development should be less costly to serve with needed services than a less dense pattern.

Patrick L. Dugan has been a city planning director and a city finance director. During the last 30 years, he has held various financial and planning positions in cities, counties, and regional agencies in three states. He has shared his views on public finance and planning with articles in The Western Planner since 1995. Now a private consultant in Washington, he can be reached at


  1. See Marc Holzer, Ph.D., et al., Literature Review and Analysis Related to Optimal Municipal Size and Efficiency (Newark, New Jersey: Rutgers, 2009) for a review of the literature on this subject. Also see this column, “What Size is Best?” Western Planner, January/February, 1997.
  2. See previous Western Planner articles “The Costs of Services Debate” (October/November, 2004) and “Theory vs. Reality in the Cost of Sprawl” (October/November, 1998) for a discussion of these studies.
  3. Hugh Wenban-Smith, Urban infrastructure: Density matters, not just size, London School of Economics, 2006, page 2.
Print Friendly and PDF Email this page