Transportation and Land Use

The complexity that characterizes the interaction of transportation and land use in urban areas is matched by the variety of the disciplines called on to address these issues, including economics, urban planning and civil engineering. In recent decades, communication among scholars in these disciplines has improved and the acceptance of a common base of theory and method, based on economics, is increasing. The Taxation, Resources and Economic Development (TRED) conference on “Transportation and Land Use” held at the Lincoln Institute in October 1996 focused on these issues. Ten papers presented at that conference are now published in a special issue of the journal Urban Studies. The papers are organized into four groups as summarized below.

Trends in Urban Development

Gregory Ingram’s paper on “Metropolitan Development: What Have We Learned?” documents the worldwide prevalence of several trends that characterize modern urbanization. Employment decentralization and the emergence of multiple employment centers in large metropolitan areas are observed worldwide in both developing and developed countries. Although employment continues to be more centralized than population, the typical Central Business District does not contain more than about 20 percent of jobs, and much smaller percentages are common in the U.S. Manufacturing employment has become more decentralized than service employment. Decentralization has reduced traffic congestion and travel distances and has contributed to a weakening of transit systems. The increased affordability of motorized transportation worldwide has led to more trip-making, with work trips typically being less than a third of all trips in urbanized areas.

Peter Gordon, Harry Richardson and Gang Yu find evidence that the suburbanization and exurbanization of employment in the U.S. has picked up its pace since 1988. In their paper, “Metropolitan and Non-Metropolitan Employment Trends in the U.S.: Recent Evidence and Implications,” they argue that the ability of manufacturing and even of services to locate in exurban and rural areas, shunning inner-suburban and central city locations, is a consequence of the continued weakening of the agglomeration economies that shaped the now outdated downtown-oriented city.

Robert Cervero and Kang-Li Wu examine the relationship between average commuting distance and employment subcentering in their paper, “Subcentering and Commuting: Evidence from the San Francisco Bay Area, 1980-1990.” They are concerned with changes in employment densities in 22 employment subcenters and with the commuting distances and travel times of those employed in these subcenters. The authors find that employment densities have increased more in the outlying suburban centers and that commuting to these centers has experienced modal shifts away from transit and in favor of the automobile. According to their data, while jobs in these centers grew by 18 percent during the decade, average one-way commuting distances to these 22 subcenters increased by 12 percent, and average one-way travel times rose by only 5 percent.

These findings are consistent with theory: with the number of subcenters fixed and the degree of spatial mismatch between jobs and housing invariant with job growth, an increase in the number of jobs in each subcenter should result in longer commutes on average. If new subcenters are spawned in between existing ones or new ones develop in outlying areas-something that does not appear to have occurred in the Bay Area-then average commutes should decrease. The 22 subcenters account for less than half of total employment in the Bay Area, the rest of the jobs being broadly dispersed throughout. Because such dispersed employment is not included in their study, we do not know about the total effect of job decentralization on average commute distances and times.

Genevieve Giuliano’s paper, “Information Technology, Work Patterns and Intrametropolitan Location: A Case Study,” examines the impact of information technology, including the advent of fax machines, computers, modems and the internet. One of her central observations is that while the U.S. labor force increased by 14 percent from 1980 to 1990, the “contingent workforce,” a diverse group of temporary workers, part-time workers, the self-employed and business service workers, increased much faster, from about 25 to 33 percent.

This trend implies that the information revolution is causing structural shifts in the labor force as more and more workers offer temporary services to a variety of employers and, as a result, do not have a long-term attachment to any one employer. Theory suggests that such workers should locate in a way that is sensitive to their expected accessibility to jobs. Also, the advent of information technology should facilitate “telecommuting,” thus reducing the need for physical proximity to jobs.

Giuliano uses the 1990 U.S. Census Public Use Microsample for the Los Angeles region to compare the residential location and commuting patterns of contingent and non-contingent workers. The socioeconomic complexity of contingent workers makes it difficult to draw clear conclusions, but Guiliano does find that those contingent workers who live in suburban areas are likely to live in high amenity areas. Controlling for socioeconomic factors, commuting distances are shorter for part-time workers than they are for full-time workers, and among full-time workers the self-employed have the shortest commutes.

Agglomeration Economies

The next two papers offer empirical contributions on intra-urban employment agglomeration. “Spatial Variation in Office Rents within the Atlanta Region,” by Christopher Bollinger, Keith Ihlanfeldt and David Bowes, is a hedonic rent study for office buildings in the Atlanta area from 1990 to 1996. The authors find that part of the rent differences among office buildings is due to differences in wage rates, transportation rates and proximity to concentrations of office workers. More importantly, the convenience of face-to-face meetings facilitated by office agglomerations is also reflected in office rents, providing evidence that agglomerative tendencies continue to be important in explaining office concentrations, despite the ability of information technology to reduce the need for some such contacts. In their paper, “Population Density in Suburban Chicago: A Bid-Rent Approach,” Daniel McMillen and John McDonald show that population density patterns in the Chicago MSA are strongly influenced by proximity to subcenters, which include the Central Business District, O’Hare Airport and 16 other centers. Site-specific variables such as access to commuter rail stations or highway interchanges have smaller influences on population densities.

Travel Behavior and Residential Choice

Among the challenges posed by the evolving trends in transportation and land use is a better explanation of the role of non-work travel in residential location decisionmaking. Motorized mobility has greatly increased non-work travel, thus weakening the relevance of the now classical commuting-based theory of residential location. While information technology may result in more telecommuting, the importance of non-work travel relative to work travel may grow even more in the future.

Two papers attempt to develop new techniques that can be used to explain the influence of non-work travel behavior on residential location and land use patterns, and vice versa. Central to this research is the notion that when a household makes a residential choice decision it will consider the pattern of non-work trips its members are likely to make. Accessibility to non-work opportunities is likely to be important and, for many households, perhaps more important than accessibility to jobs.

Moshe Ben-Akiva and John Bowman model the probability of choosing a residential location by treating the non-work trip patterns and activity schedules of the household’s members as explanatory variables. Their model allows the treatment of trips as tours with stops at multiple destinations. In their paper, “Integration of an Activity-Based Model System and a Residential Location Model,” the authors report that their model does not fit the data as well as a work-trip-based comparison model. But, the non-work accessibility measures are more appealing conceptually and allow a richer set of predictions and simulations to be made.

Until recently, economists have suppressed the importance of non-work trips in their theories of land use. Planners have viewed land use planning as a tool that can affect behavior and travel demand. But what is the evidence that travel patterns can be influenced meaningfully by manipulating land use at the neighborhood level or in a larger area?

Marlon Boarnet and Sharon Sarmiento tackle this question by means of a travel diary survey of Southern California residents. Their paper is titled “Can Land Use Policy Really Affect Travel Behavior? A Study of the Link Between Non-work Travel and Land Use Characteristics.” The number of work trips made by residents is explained by sociodemographic variables describing the residents and by land use characteristics describing their place of residence. Generally, the land use variables describing the neighborhood are not statistically significant, but future studies could follow this approach by trying more complex specifications and using better data.

Jobs-Housing Mismatch

As first stated by John Kain in 1968, the “spatial mismatch hypothesis” claimed that black central city residents are increasingly at a disadvantage economically as jobs disperse to the suburbs. Many suburban governments limit the quantity of high-density/low-income housing, forcing workers to make long, expensive commutes. Although there is a wealth of empirical work on the mismatch hypothesis, Richard Arnott’s paper, “Economic Theory and the Mismatch Hypothesis,” is one of the first attempts to formulate a microeconomic theory of the mismatch problem. In Arnott’s model, jobs flee to the suburbs because of the advent of international trade (relaxation of global trade barriers) and the emergence of suburban-based inter-city truck transport after World War II. At the same time, large-lot zoning and discrimination in suburban housing markets force minorities to reside in central cities. An increase in the cost of commuting effectively lowers the wage paid to low-skilled labor from the city.

In “Where Youth Live: Economic Effects of Urban Space on Employment Prospects,” John Quigley and Katherine O’Regan investigate how neighborhood of residence and access to jobs affect the employment prospects of minority youth. Black youth unemployment rates are higher in metropolitan areas where blacks are more isolated geographically. Controlling for socioeconomic characteristics, minority youth who have less residential exposure to whites are more likely to be unemployed. Finally, controlling for socioeconomic characteristics as well as residential exposure to whites, minority youth living in neighborhoods that are less accessible to jobs are more likely to be unemployed. While these findings support the mismatch hypothesis, they also suggest the importance of social networks and spatial search as important mechanisms in the intra-urban labor market.

Alex Anas, professor of economics at the State University of New York at Buffalo, was the editor of the special issue of Urban Studies (Vol. 35, No. 7, June 1998). The article and figures used in Land Lines are adapted with permission.

Note: Ben Chinitz, former director of research at the Institute, helped organize the 1996 TRED conference and the following colleagues served as discussants of the papers: James Follain, Vernon Henderson, Douglass Lee, Therese McGuire, Peter Mieszkowski, Edwin Mills, Sam Myers, Dick Netzer, Stephen Ross, Anita Summers, William Wheaton, Michelle White and John Yinger. The conference participants were saddened when news arrived that William Vickrey, who had been named a Nobel laureate in economics only a few days before, had passed away while traveling to the conference. Professor Vickrey had been a leading thinker on issues of transportation and land use and a regular attendee of previous TRED conferences. The special issue of Urban Studies based on the 1996 conference serves as a tribute to his memory.