Journal of Planning Literature http://jpl.sagepub.com
Beyond the Spatial Mismatch: Welfare Recipients and Transportation Policy Evelyn Blumenberg and Michael Manville Journal of Planning Literature 2004; 19; 182 DOI: 10.1177/0885412204269103 The online version of this article can be found at: http://jpl.sagepub.com/cgi/content/abstract/19/2/182
Published by: http://www.sagepublications.com
Additional services and information for Journal of Planning Literature can be found at: Email Alerts: http://jpl.sagepub.com/cgi/alerts Subscriptions: http://jpl.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations (this article cites 69 articles hosted on the SAGE Journals Online and HighWire Press platforms): http://jpl.sagepub.com/cgi/content/refs/19/2/182
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
10.1177/0885412204269103 ARTICLE Journal of Planning Literature
Beyond the Spatial Mismatch: Welfare Recipients and Transportation Policy Evelyn Blumenberg Michael Manville Beneath the broad umbrella of agreement about transportation’s relationship to poverty is considerable discord about the specific nature of the problem and about where and how transportation solutions should be applied. Much of the existing scholarship on this topic focuses on the spatial mismatch hypothesis, the geographic separation between employment and housing. Although this concept has merit, to meet the transportation needs of welfare recipients, policy makers must move beyond conventional notions of the spatial mismatch hypothesis. This article draws from theoretical and empirical scholarship on travel behavior, transportation infrastructure, poverty, gender studies, and residential segregation and recommends transportation policies to better connect welfare recipients to employment.
Keywords: spatial mismatch; poverty; transportation; welfare reform
The passage of welfare reform legislation in 1996 dra-
matically altered the way the United States dispenses social services. In the aftermath of the Personal Responsibility and Work Opportunity Reconciliation Act (U.S. Congress 1996), welfare is no longer a program designed to protect vulnerable populations from the uncertainties of the labor market. Rather it is a vehicle for putting them to work and inserting them, within
five years at most (and often much faster), into the very realm from which they had once been defended. Thus, welfare reform set out to do what the original welfare program had been designed to prevent. Welfare was established to buffer “suitable” women from the labor market, enabling them to remain at home and care for their children (Gordon 1994); in contrast, welfare reform was initiated, as former President Clinton put it, to “end welfare” by mandating employment. A host of obstacles stands in the way of welfare reform’s success, not least of which is transportation. The central challenge of welfare reform lies in connecting its recipients to the labor market. Because this was never a central concern of traditional welfare programs, few mechanisms existed to facilitate the transition to employment, and the reform law has, as a result, been accompanied by other programs and services designed to prepare welfare clients for entry into the workforce. Some of these programs focus on providing transportation to welfare recipients on the assumption that lack of Beyond the Spatial Mismatch
EVELYN BLUMENBERG is an associate professor of urban planning in the School of Public Policy and Social Research at the University of California, Los Angeles. Her research focuses on low-wage labor markets, urban poverty, transportation and the poor, and women’s economic status. MICHAEL MANVILLE is a Ph.D. student in the Department of Urban Planning at the University of California, Los Angeles. His interests include urban governance, transportation in low-income communities, megaprojects, and city marketing. Journal of Planning Literature, Vol. 19, No. 2 (November 2004). DOI: 10.1177/0885412204269103 Copyright © 2004 by Sage Publications
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
Beyond the Spatial Mismatch mobility is a significant barrier to employment. This assumption is not unreasonable. There is a substantial empirical literature on transportation’s relationship to both poverty and employment and a more recent and growing literature about its relationship to welfare use. Most studies consistently find that transportation is a barrier to employment for the poor in general and for welfare recipients in particular. For example, in a study conducted in Illinois (Julnes and Halter 2000), more than 25 percent of former Temporary Assistance to Needy Families (TANF)1 clients interviewed reported having transportation problems. Among unemployed former recipients, this figure was 41 percent; among the employed, it was nine percent. In a survey of randomly selected public housing residents, Henle and Kinsella (1996) find that transportation is second only to child care as an obstacle to employment. And researchers studying the relationship between transportation and employment find that reliable transportation leads to increased access to job opportunity, higher earnings, and increased employment stability (Blumenberg 2002; Cervero, Sandoval, and Landis 2002; Danziger et al. 2000; Ong 1996, 2002; Ong and Blumenberg 1998; Raphael and Stoll 2000). Beneath the broad umbrella of agreement about transportation’s importance to employment, however, is considerable discord about the specific nature of the problem and about where and how transportation solutions should be applied. Much of the existing literature focuses on the “spatial mismatch” hypothesis, an idea first put forward in 1968. Although constantly evolving, the spatial mismatch hypothesis attributes growing inner-city poverty to, among other factors, the separation of employment and housing and the “surplus of workers relative to the number of available jobs in those areas where the black population is concentrated” (Ihlanfeldt and Young 1994, 221). As the United States economy has restructured itself in the postwar years, manufacturing, service, and retail employment has shifted away from central cities, oftentimes locating in the suburbs and sometimes moving overseas (Bluestone and Harrison 1982; Jackson 1985; Norton and Rees 1979; Piore and Sabel 1984; Sugrue 1996). Advocates of the spatial mismatch hypothesis argue that low-income residents have been left behind in urban areas, distant from suburban employment opportunities and without the resources to overcome this difficulty (Cutler and Glaeser 1997; Ihlanfeldt and Sjoquist 1998; Kain 1968, 1992; Ross 1998). Over the years, however, the spatial mismatch hypothesis has been put to test by a host of skeptical scholars. Our purpose in this review is not to revisit yet again the voluminous mismatch literature; four reviews of the spatial mismatch research have been
183
recently published and a number of recent studies have highlighted the thorny methodological issues associated with this body of scholarship (Cooke and Ross 1999; Deka 2002; DeRango 2001; Perle, Bauder, and Beckett 2002). While the literature reviews have been written by scholars sympathetic to the hypothesis— and anyone reading them should bear in mind the potential biases of their authors—they nevertheless summarize and analyze the mismatch debate and leave no reason for us to do the same (Holzer 1991; Ihlanfeldt and Sjoquist 1998; Kain 1992; Preston and McLafferty 1999). Instead, our purpose is to focus on the broader relationship between transportation and employment opportunities among the poor, particularly welfare recipients, and show that welfare recipients face spatial barriers to employment that extend beyond the spatial mismatch. Employment opportunities are affected not solely by distance but also by travel time and accessibility to different travel modes. In many instances, for example, those with access to automobiles can navigate the route from home to work with far greater ease than bus riders, even when drivers live farther from their places of employment. But automobile ownership among the poor is far from universal, and access to reliable vehicles is even less commonplace. In the latter portion of this article, we ask whether public or private transportation is the best way to overcome the transportation barriers that welfare recipients face. We argue that too great a focus on the spatial mismatch hypothesis obscures important aspects of the transportation-welfare problem. Ideally, policy discussions of welfare and transportation should move beyond a conventional notion of the spatial mismatch hypothesis and incorporate a much broader conception of spatial barriers to employment. Policies to facilitate welfare recipients’ travel to jobs should draw from numerous bodies of research, including those on travel behavior, transportation infrastructure, poverty, gender differences, and, not least, residential segregation. ECONOMIC OPPORTUNITY AND THE SPATIAL MISMATCH HYPOTHESIS
The spatial mismatch hypothesis is predicated on the notion that the poor living in the inner city do not often live near available employment opportunities and that the distance between their residences and places of employment is frequently amplified by a lack of adequate transportation. This is not a new idea. As early as 1965, in the aftermath of the Watts riots, the McCone Commission identified poor transportation and spatial isolation as key contributors to the hopelessness and rage that eventually resulted in rioting and urban
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
184
Journal of Planning Literature
unrest in South Central Los Angeles. The Commission concluded, Our investigation has brought into clear focus the fact that the inadequate and costly public transportation currently existing throughout the Los Angeles area seriously restricts the residents of disadvantaged areas, such as South Central Los Angeles. This lack of adequate transportation handicaps them in seeking and holding jobs, attending schools, shopping, and in fulfilling other needs. It has had a major influence in creating a sense of isolation, with its resultant frustrations among the residents of South Central, . . . particularly the Watts area. (Governor’s Commission on the Los Angeles Riots 1966, 35)
To support this statement, the Commission rattled off a list of unsettling facts. Heavily populated East Los Angeles was only six miles from job-plentiful City of Commerce, but a paucity of transit options made the City of Commerce “almost inaccessible” to residents of East Los Angeles. Hundreds of women from South Central Los Angeles traveled each day to jobs in the affluent neighborhoods of Brentwood, Beverly Hills, and Pacific Palisades, but their commutes averaged over two hours and required four buses. A resident of Watts who wanted to travel to the nearest Sears had to take two buses and spend one and a half hours en route. Low-income residents were not located near low-wage jobs, and there seemed no easy way to connect the two. At the time, this phenomenon was a problem without a name. That changed three years later, when the economist John Kain (1968) published an analysis of the effect of housing discrimination on African American employment. Kain concluded that the spatial separation of black housing and employment exacerbated the poverty of inner-city African Americans. He cited suburbanization and the attendant movement of lowwage jobs away from the inner cities as major factors in this separation. Although he never used the phrase himself, Kain’s idea came to be dubbed the “spatial mismatch hypothesis,” and his article, although not the first to note the effect of the suburbs on low-wage labor, became a standard in the field. Three months after his article was published, Kain’s theory was bolstered when the National Commission on Civil Disorders (better known as the Kerner Commission) released its report on the urban strife that had swept across America the previous summer, igniting riots in Detroit, Newark, Atlanta, and elsewhere. Like the McCone Commission before it, the Kerner Commission viewed the geographic disparity in job growth as one of the primary causes of the previous year’s turbulence. “Most new employment opportunities,” the
Kerner report noted, “do not occur in central cities, near all-Negro neighborhoods. They are being created in suburbs and outlying areas—and this trend is likely to conti nu e i ndef i ni tel y ” ( N a ti ona l A dvi s or y Commission on Civil Disorders 1968, 392). Kain’s original work is worth discussing a bit more because, having given birth to a widely debated and somewhat amorphous concept, it is often the source of some confusion. As Ihlanfeldt (1994) points out, in 1968 Kain actually advanced not one but three hypotheses. The first said simply that residential segregation helped shape the geographic distribution of African American employment. The second said that segregation also increased black unemployment, and the final argument was that both of these conditions were exacerbated by the decentralization of jobs. It is this final idea—that surplus labor in minority inner-city areas leads to either increased unemployment or the need to travel farther (which results in lower net earnings because of travel costs)—that has grown into the modern idea of spatial mismatch (Ihlanfeldt 1994). Over the years the idea has also expanded and contracted to reflect the changing concerns of those who study poverty; at the time of Kain’s initial investigation, the focus was on African American men, but scholars now have applied it to a wide range of the urban poor, and in recent years the emphasis has shifted to low-income women as welfare reform has seized the policy agenda (Ihlanfeldt and Sjoquist 1998; O’Regan and Quigley 1998b). As we will discuss shortly, the conventional idea of spatial mismatch has undergone a number of evolutions and refinements in recent years, which have both strengthened our understanding of the role distancebased isolation plays in the lives of the poor and clouded any attempts at developing a single, widely accepted definition of spatial mismatch. For now, however, it is enough to assert that many scholars have demonstrated that spatial access to employment— defined very broadly—can be pivotal for impoverished job seekers, including low-income women and welfare recipients. Twenty-five years after his original paper was published, Kain (1992) reviewed six comprehensive studies of the spatial mismatch hypothesis and found that all but one supported it to some degree. The lone dissident was a review by Jencks and Mayer (1990), which found little evidence that the spatial mismatch exists. Evidence for it, they included, “is so mixed that no prudent policy analyst should rely on it” (Jencks and Mayer 1990, 219). In a subsequent review, Ihlanfeldt and Sjoquist (1998) found the spatial mismatch valid and relevant to welfare reform. In addition, a number of studies show that job accessibility is an important factor in the economic well-being of welfare recipients. Using
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
Beyond the Spatial Mismatch data for the Detroit Metropolitan area, Allard and Danziger (2003) and Allard (2002) find that proximity to employment opportunities is associated with a higher probability of working and leaving welfare, while Blumenberg and Ong’s (1998) study of Los Angeles shows that neighborhoods with entry-level work nearby have lower welfare usage rates than comparable job-poor neighborhoods. Other statistics support the application of the spatial mismatch to welfare recipients. Over time, the welfare program has assumed an increasingly urban character. While urban residents constitute approximately 35 percent of the nation’s population, welfare recipients in the eighty-nine largest urban counties account for 60 percent of the national welfare caseload (Leonard and Kennedy 2001), and many of them reside in inner-city neighborhoods. The U.S. General Accounting Office (1998) reports that in the 1990s, almost three-quarters of all welfare recipients lived in central cities or rural areas, while in 1997, approximately three-quarters of all employment in 114 select metropolitan areas was located in the suburbs (U.S. Department of Housing and Urban Development 2000).2 Even the sustained period of economic expansion in the late 1990s did little to close this central city-suburb employment gap. While central-city employment grew by 8.5 percent between 1992 and 1997, suburban employment grew at more than twice that rate (U.S. Department of Housing and Urban Development 2000). And although some cities—such as Atlanta, Las Vegas, and Boston—experienced strong central-city job growth, other central cities—such as Baltimore, Cleveland, Philadelphia, and Washington, D.C.—suffered from continued decline (U.S. Department of Housing and Urban Development 2000). For example, from 1991 to 1996, the number of jobs in Cleveland fell 1.6 percent, while suburban job growth in the Cleveland area grew 121 percent. Seventy-eight percent of the Cleveland-area welfare recipients live in the central city, while only 34 percent of its overall population does (Bania, Coulton, and Leete 1999). THE SPATIAL MISMATCH HYPOTHESIS QUALIFIED
Despite the considerable support for it that exists in some quarters of academia, the spatial mismatch has never lacked for detractors, and critics have assailed it from almost the moment it was introduced. Even among those who adhere to the idea of spatial mismatch, there have been significant disagreements over the best way to measure it. As Ihlanfeldt (1999, 342) notes, the spatial mismatch hypothesis is “a straightforward concept that has been very hard to prove.” In response to these criticisms, mismatch research has
185
grown from the rather crude method of simply measuring jobs per area (which fails to account for the number of prospective applicants for those jobs), to measuring jobs per worker, to evaluations of commute time, and finally to gravity-based models that try to combine all these factors. Mismatch scholars have also responded with increasing methodological sophistication to concerns regarding sample selection bias—the fact that samples typically exclude unemployed adults whose lack of a job may be, in part, because of their spatial isolation (Cooke and Ross 1999; Deka 2002; Shen 1998). Finally, researchers have attempted to address the endogeneity of residential location choices. A positive association between spatial access to employment and economic outcomes can be because of the effects of employment proximity but can also be because of the residential relocation of the employed to job-rich, suburban neighborhoods. Has a person become employed because this person lives in a job-rich neighborhood or has employment allowed this person to relocate closer to the job? To account for these relationships, some scholars have examined employment outcomes of youth, whose housing location decisions are considered exogenous since they are typically determined by their parents (Holzer, Ihlanfeldt, and Sjoquist 1994; Raphael 1998; Stoll 1999). Substantive criticisms of the spatial mismatch hypothesis are twofold. The first asserts that racial discrimination rather than distance-based isolation is primarily responsible for inner-city unemployment and poverty. In the 1960s, scholars such as Fogelson (1967) and Scobles (1968) attacked the McCone Commission for its emphasis on ghetto transportation and charged that the allure of the transportation explanation was based less on the weight of empirical evidence and more on its convenience. Transportation, they argued, was an inherently solvable problem and allowed policymakers to evade the more immutable dilemmas of racism and police brutality. “The South Central ghetto,” Fogelson (1967, 132) noted, “is indeed isolated, but not for reasons as simple and reassuring as dreadful bus service.” This argument was reinforced by the work of Floyd (1968), who examined job programs in South Central Los Angeles and suggested that some employers used the idea of transportation barriers as a “convenient excuse” for not hiring minorities (Floyd 1968, 10). Harrison (1974) similarly argued that discrimination played a more influential role than spatial location in determining the employment outcomes of inner-city residents. The best-known skeptic, however, is Ellwood (1986), who in an influential study of inner-city Chicago wrote that “race, not place” was the source of the inner-city poverty problem. His thesis is bolstered by such studies
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
186
Journal of Planning Literature
as Kasinitz and Rosenberg’s (1996), who examined the Red Hook neighborhood of Brooklyn where African Americans lived in close proximity to appropriate jobs and yet still were not hired. Interviews of employers confirmed that discrimination locked out many minority applicants. Likewise, Carlson and Theodore (1997), in research on employment outcomes in Chicago, determined that race was a more significant determinant for African Americans seeking jobs than was space. Briggs (1997) also finds proximity to be a less significant factor than racial discrimination and ethnic segmentation of the labor market. The importance of these studies should not be understated, but other researchers have pointed out that the race versus space argument can be taken too far, as neither is wholly independent from the other. Preston and McLafferty (1999), in their review of the mismatch literature, point out that “to a large degree race is space in the United States. Cities bear the imprint of racialized processes, . . . and space in turn affects these processes” (p. 390). This line of thinking conforms to the findings of Kirschenman and Neckerman (1991), whose interviews with Chicago-area employers found that employers thought of African American men as dishonest and uncooperative but also found that much of this belief was grounded in the fact that the men lived in the inner city. “Race is an important factor in hiring decisions,” the authors concluded, “but it is not race alone. Rather it is race in a complex interaction with employers’ perceptions of class and space, or inner-city residence” (p. 204). Race is a sign of space and vice-versa. The second substantive criticism of the spatial mismatch hypothesis is built less around the distance to employment and more around the mode used to traverse that distance. A number of scholars have argued that the conventional spatial mismatch story is less relevant in newer urban areas where low-income households tend to be more dispersed and where employment opportunities, although not growing as fast as in the suburbs, are numerous. In these cities, low-income inner-city residents suffer from a modal mismatch, a drastic divergence in the relative advantage between those who have access to automobiles and those who do not (Blumenberg and Hess 2003; Blumenberg and Ong 2001; Kawabata 2003; Shen 1998; Taylor and Ong 1995; Wyly 1998). The idea of a modal mismatch is relatively new; as recently as 1992, Kain (1992, 392) noted, “None of the spatial mismatch studies, including my original 1968 study, does a good job of dealing with mode choice.” More attention has been given to it in the years since as the merits of public and private mobility have been increasingly debated. The logic behind modal mismatch is derived from evidence suggesting that the spatial mismatch is far
from uniform in the extent and manner to which it affects cities. There is a host of reasons for this, but many of them can be attributed to cities’ age and size. Mismatch in the conventional sense—where geographic distance between home and available employment opportunities is the primary employment barrier—is far more common in the older metropolitan areas of the Northeast and the Midwest. These are cities that developed before the age of the automobile, whose compact urban form included concentrated low-income residential neighborhoods, industrial employment in the neighborhoods surrounding the central business district, and well-developed urban transit systems to facilitate intracity travel. In these cities, the spatial mismatch was a product of the “hole in the doughnut” syndrome—the formation of a hollowed-out urban core in which “hypersegregated” minorities remain trapped when the industrial areas declined. When employment shifted outward, these cities and their residents were hardest hit by the spatial barriers described by Kain. These conditions characterize the plight of Philadelphia (Private Industry Council of Philadelphia 1998), Milwaukee (Center for Economic Development 1998), and Detroit (Sugrue 1996) among others. In each of these cases, employment opportunities disappeared; those who had the means moved away, and those who could not spiraled into decline. Such is the conventional mismatch story, and in it we find many of the characteristics of Rust Belt disinvestment. But for the most part, the problems that characterize these areas bear only parting relevance to those that characterize new urban centers like Phoenix and Denver. These cities—and those like them (e.g., Houston, Las Vegas, Los Angeles)—came of age when American land use policies had been subordinated to the needs of the automobile, and the distribution of their residents and employment reflects the greater mobility that cars confer. Data from the 2000 U.S. Census show that almost one-half (47 percent) of the metropolitan population lives outside of the central city. But the spatial dispersion of the poor is more extensive in the Sunbelt: in the South and the West, 53 and 49 percent of the poor, respectively, live outside of the central city, compared to 38 and 39 percent in metropolitan areas in the Northeast and the Midwest. Sunbelt cities, of course, tend to be larger and less compact. Further, many jobs remain in the central city. The rapid increase in suburban employment does not mean that entry-level jobs no longer exist in central cities. Care must be taken to distinguish between net new job growth and turnover in existing jobs, and there is plentiful evidence suggesting that urban job turnover rates actually exceed the rate of suburban job growth. In other words, at any given time, there may be more
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
Beyond the Spatial Mismatch entry-level job vacancies in a central city than there are in its suburbs (Shen 1998, 2001), and the short commute distances among low-income, central-city residents suggest that many central-city residents find employment in these jobs.3 Wang (2003) finds that low-wage workers tend to have better job proximity than highwage workers. However, as other scholars note, the more relevant issues may be the relative ratio between job openings and potential applicants (Ihlanfeldt and Young 1994; Kawabata 2003; Leonard and Kennedy 2001; Ong 1996; Turner 1997) as well as the match between the skills required in central city occupations and those available in the local workforce (Browne 2000; Kasarda 1983, 1985, 1988, 1989; Simpson 2000). The archetype of this newer urban model is Los Angeles, which has been called the “first American city” for its low-rise, postpedestrian development (Weinstein 1996) and for its eventual organization around the private automobile.4 Los Angeles is a useful example not only because it is the progenitor of this new breed of city but also because, as the oldest and largest, it is also heavily studied. Unlike Detroit’s hollow core, Los Angeles has a checkerboard pattern of growth and lacks—with the possible exception of Watts—a concentrated area of isolated minority poverty (Blumenberg and Ong 1998). Data from the 2000 U.S. Census shows that 46 percent of Los Angeles’ poor live outside of the central city, while numerous employment opportunities remain concentrated in and around the central business district. Having been built around the car, most of Los Angeles’ residents have access to one. According to a recent welfare-to-work transportation needs assessment in Los Angeles, 70 percent of its welfare recipients have access to automobiles, and 60 percent typically commute by private vehicle; in contrast, less than 27 percent use transit to travel to work (County of Los Angeles 2000). Average commute distance among welfare recipients is relatively short, approximately seven miles (County of Los Angeles 2000). These figures suggest that in Los Angeles and other cities with similar development patterns, spatial access to suburban employment opportunities may not be the central obstacle in moving welfare recipients to work. Unquestionably, Los Angeles has many transportation problems. But by conventional measures, it is not mismatched, and neither are most cities like it. WELFARE RECIPIENTS AND SPATIAL BARRIERS TO EMPLOYMENT
Newer cities suffer less of a spatial mismatch largely because many of their residents have access to automobiles, not because their land use patterns are amenable
187
to human mobility. For those without cars, life is extraordinarily difficult, probably more difficult than it is for transit-dependent residents of older, East Coast cities where transit networks are more extensive. Transit-dependent women are perhaps the most disadvantaged since they must juggle household and workrelated responsibilities and the complicated travel patterns these engender. But even low-income adults with automobiles face spatial barriers because they disproportionately are saddled with old and unreliable vehicles. For policy purposes, therefore, it is important to move beyond the “spatial mismatch” and to focus more broadly on spatial barriers to employment that include, but are not exclusive to, “distance.” Lengthy Commute Times On average, low-income commuters travel shorter distances than those with higher incomes (Loveless 1999; Murakami and Young 1997; Rosenbloom 1994). But a short commute distance should not be confused with a short commute, nor should it suggest that those who live and work in the central city lack transportation problems. There is a difference between commute distance and commute time, and in the central city, even over short distances, travel times may be quite long, especially for people who rely on public transportation. Long travel times can be problematic since they are difficult to sustain and may, as a result, lead to higher job turnover and lower overall earnings. Overall commute time tends to vary positively with income; consequently, low-income workers, on average, commute relatively short distances. Rosenbloom (1994) finds that the lowest income workers—those who make under $5,000—travel the shortest distance from home, and Murakami and Young (1997) find that low-income single mothers make significantly more trips within a three-mile radius of their homes than do non-low-income adults. Loveless (1999) shows that low-income households take 20 percent fewer trips and travel 40 percent fewer miles than middle- and upperincome households.5 But these short trips can take a long time when traveling by public transit. McLafferty and Preston (1997) find that commute times vary with both income and the use of public transportation. Even in cities with good transit service, transit travel times are, on average, much longer than automobile travel times—a result of walking to and from stops, waits at stops and for transfers, and frequent vehicle stops along the way. For all workers, commute distance varies only slightly by mode. In 1995, the average commute distance for workers using private transportation was twelve miles, compared to thirteen miles for those reliant on public transportation (Hu and Young 1999). Commute times,
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
188
Journal of Planning Literature
however, were more than twice as long on public transit than for those who used private vehicles—twenty minutes compared to forty-two minutes (Hu and Young 1999). Although low-income commuters travel shorter distances than higher-income commuters, those who are transit dependent spend, on average, thirty-six minutes in travel (Murakami and Young 1997). As a consequence, welfare recipients who rely on public transit can reach far fewer jobs than those who travel by automobile (Blumenberg and Hess 2003; Blumenberg and Ong 2001; Kawabata 2003; Shen 1998). When Bania, Coulton, and Leete (1999) examined the number of entry-level jobs accessible by public transportation in Cleveland, they found that the transit dependent suffered heavily in comparison to those who had automobiles. A twenty-minute car commute, for instance, gave a welfare recipient access to 12.8 percent of the area’s entry-level jobs, while a twenty-minute transit commute gave them access to less than 2 percent. The difference between car and transit commutes became even more striking when the commute time was increased to thirty minutes. Auto users could get to more than 40 percent of the area’s opportunities, while transit riders were left at just above 7 percent. “Those who depend on the transit system,” the authors concluded, “pay a big price in terms of increased commute times or decreased access to job opportunities” (Bania, Coulton, and Leete 1999, 15). A study of welfare recipients in three California counties (Alameda, Fresno, and Los Angeles) shows that welfare recipients who use automobiles can reach most jobs within these counties in a thirty-minute commute. Access to jobs by transit, however, is a more questionable proposition. In some areas, it is reasonably good; in others, (for instance, the Los Angeles neighborhoods of Watts and Monterey Park), it is strikingly inferior to the automobile. In Watts, the ratio between the number of jobs accessible by car to those accessible by public transit within a thirty-minute commute is fiftynine to one, while in Monterey Park it is seventy to one (Blumenberg and Hess 2003). Shen (1998) and Kawabata (2003) report similar findings for Boston, Los Angeles, and San Francisco. Commute Distance and Earnings For low-income commuters who travel longer distances, employment may be difficult to sustain. Wages tend to rise as one travels farther from the city center (Viscusi 1992) and also to rise with commute distance in general. This relationship is consistent with findings for the labor market at large and is not unique to lowincome workers—across all incomes, commute distance correlates positively with earnings (Taylor and Ong 1995)—but the explanation for this relationship
tends to differ between the poor and the population at large. For middle- and upper-income adults, high income in many instances precedes longer commutes. Although better paying jobs tend to be dispersed throughout metropolitan areas (Simpson 1992) and job seekers can sometimes earn higher incomes by working in the suburbs, conventional location theory holds that well-paid city workers will ultimately choose to live in the suburbs even if they are employed in the central city. This choice reflects the desire for more housing and land, the relative costs of which are lower on the fringes of urban areas (Muth 1969; Simpson 1992). The preference for more space is frequently accompanied by a desire for better residential amenities—more parks, quality schools, and lower crime rates—and these too are more prevalent in the suburbs than they are in inner cities. Lastly, of course, higher incomes generally allow ownership of automobiles, and this purchase reduces much of the inconvenience and opportunity costs inherent in longer commutes (Taylor and Ong 1995). In combination, these reasons justify and precipitate moves to the suburbs. The picture for lower income people is somewhat less clear, but conventional notions of spatial mismatch would hold that the suburbs (whether they are commuted to or relocated to) would be the source of higher earnings for the poor rather than a destination to move to once higher earnings are secured (Downs 1968, 1973; Kain 1968; Kain and Persky 1969; Rosenbaum 1995). For the poor, the costs of residential migration to the suburbs are hotly debated (Briggs 1997), but there is no argument that such migration occurs less often for those of lower income than for the population at large. This is partly a result of residential discrimination, which greatly constrains the location choices of minorities (Massey and Denton 1993), but there are nonracial factors that explain it as well. People with lower incomes, according to Muth’s (1969) location theory, need less space, and so choose instead to live in urban areas, where costs per foot of real estate tend to be higher. The poor are also tethered to the city, according to some interpretations, by the relative largesse of city social service programs (Muth 1969; Peterson 1981). Theoretically, low-income households are compensated for their central-city locations by their proximity to employment and the resultant savings in commute time and expense (Alonso 1964; Muth 1969). But if proximate jobs are less available, the poor become saddled not only with more expensive real estate but also with the need—like the nonpoor—to travel farther to find work. These longer distance trips, even within the central city, may be difficult to sustain and result in lower earnings. Ong and Blumenberg (1998) find that welfare recipients in Los Angeles do not see an appreciable
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
Beyond the Spatial Mismatch increase in earnings as commute distances rise and, in many instances, do not see a rise in earnings at all. Even if incomes rise with commute distance, they point out, the difficulty of making long daily trips (particularly if trips are by transit) can lead to frustration and increased job turnover. If welfare recipients cannot maintain jobs, their employment will be erratic and their earnings will inevitably fall.6 Similarly, Wachs and Taylor (1998) point out that the distance/income correlation will eventually reach a point of diminishing returns. The costs of travel are implicitly deducted from wages, so extensive travel time may reduce real earnings to the level they would be if the jobs were located much closer to home. Women’s Labor and Travel The McCone and Kerner Commissions viewed poverty as an African American problem, but African Americans today hold no monopoly on the American underclass (Preston, McLafferty, and Liu 1998). Although African Americans are disproportionately represented on the welfare rolls and their plight should not be understated, they do not constitute the majority of welfare recipients. Nationally, African Americans account for 39 percent of the welfare caseload; by way of comparison, 25 percent of welfare recipients are Hispanic, and 31 percent are white (U.S. Department of Health and Human Services 2002). The Kerner Commission was also concerned with the fate of young males since young males were the demographic group implicated in the urban riots under investigation. But as the policy focus has shifted to welfare recipients, it has become necessary also to examine the transportation needs and spatial barriers faced by women, who today constitute 90 percent of adult welfare clients (U.S. Department of Health and Human Services 2002). Although women, on average, have shorter commutes than men, their travel schedules remain both complex and demanding. Urban women travel an average of eight miles compared to eleven miles for males (Federal Highway Administration 1994).7 Scholars posit a number of explanations for this sex difference in commute length and continue to debate whether women’s shorter work trips are the consequence of rational decision making or the result of constraints on their mobility. As MacDonald (1999) argues, women’s shorter work trips may be a rational response to lower wages that do not justify the costs of longer trips, a function of the more even spatial distribution of feminized occupations, or a survival strategy used to more easily balance paid employment and domestic responsibilities. But women’s shorter work trips may also indicate a “spatial entrapment” caused by the spatial mismatch between residential location and employment opportu-
189
nities, by highly localized labor markets for female labor, and by differential access to transportation modes, particularly automobiles (Wachs 1992). Whether women’s shorter commutes represent rational “choices” or “spatial entrapment” is a debate that will no doubt continue. Regardless, within both explanations lie some of the spatial barriers facing women that extend beyond the conventional mismatch story. The foremost issue may be women’s need to make multiple household-supporting trips over and above the work commute; this is particularly relevant for single mothers who typically have sole responsibility for the household. The data show that single-parent households tend to make more trips per day than other households, and studies show that this holds true for lowincome single mothers as well (Hu and Young 1999; Rosenbloom 1994). The reasons are not hard to discern. Compounding the task of traveling to work is the need to perform domestic errands, to drop off and pick up children at either school or child care, and to react to emergencies and other unplanned family events (a sick child, for instance, or a problem with a residence). Although there is conflicting evidence regarding the effects of domestic responsibilities on travel distance, it is likely that shorter commute distances allow women to more easily balance home and work responsibilities (Gordon, Kumar, and Richardson 1989; Hanson and Pratt 1995; Madden 1981; McLafferty and Preston 1997; Singell and Lilleydahl 1986; Turner and Niemeier 1997). Another consequence of women’s multiple roles is their disproportionate reliance on driving relative to comparable male parents (Rosenbloom and Burns 1994), a finding that also extends to low-income adults. Rosenbloom (1998) finds that women with incomes under $15,000 are more likely to drive to work alone than comparable low-income men. The reason for this, she speculates, is that feminized employment tends to be spatially dispersed, a conclusion also drawn by others (Singell and Lilleydahl 1986; Wyly 1996), and that cars enable women to more easily manage commuting and child-serving travel. But as we discuss in the subsequent section, not all women have equal access to automobiles, and therefore, depending on both the availability of local employment and public transit service, those welfare recipients without access to automobiles may be the most spatially isolated. Reliable Access to Dependable Personal Vehicles Most low-income households own at least one automobile, but auto ownership is still not universal. According to data from the 2001 National Household Travel Survey, 92 percent of all households have at least one vehicle, but only 73.5 percent of households with incomes less than $20,000 do (Pucher and Renne 2003).
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
190
Journal of Planning Literature
Households without an automobile are most likely to be in the lowest income brackets: households with incomes under $25,000 constitute 65 percent of households without vehicles (Federal Highway Administration 1995). Further, African American households are especially likely to lack vehicles, accounting for 12 percent of all households but 35 percent of those without cars (Federal Highway Administration 1995). But even the presence of a vehicle in the household is not necessarily the best measure of access to automobiles. Many low-income adults must share the use of a car with other household members. The ratio between household adults and vehicles is significantly lower in low-income households (.7 to 1) than in non-lowincome households (greater than 1:1) (Murakami and Young 1997). There is also a large difference between an automobile and a reliable automobile, and most welfare recipients cannot afford the latter. The cars driven by the poor tend to be old (Blumenberg and Haas 2002; County of Los Angeles 2000; Murakami and Young 1997; Ong and Houston 2002), often pollute heavily, and are usually less than a year away from extensive and costly repairs (Cervero, Sandoval, and Landis 2002). THE PUBLIC TRANSIT POLICY RESPONSE
In 1968, when the Kerner Commission first confronted the dilemma of spatial mismatch, it offered three possible solutions. “Providing employment for the swelling Negro ghetto population,” it said, “will require society to link these workers more closely with job location. Matching black workers to jobs can be done in three ways: by developing incentives to industry to create new employment centers near Negro residential areas; by opening suburban residential areas to Negroes and encouraging them to move there; or by creating better transportation between ghetto neighborhoods and new job locations” (National Advisory Commission on Civil Disorders 1968, 392). The Commission went on to note that while “all three will require large public outlays,” the transit option “has received little attention from city planners and municipal officials,” compared with the others (p. 392). Today, however, transportation arguably is the option receiving the most attention from planners and policy makers (Cervero, Sandoval, and Landis 2002). This is partly the result of the speed necessary to comply with welfare reform. Welfare reform requires that its participants become self-sufficient in five years, and five years is short time indeed to dismantle an urban structure that took many years to create. Transportation may address only the symptoms of spatial-related employment barriers rather than its causes, but in a rel-
atively short time, its proponents argue, transportation improvements can have a visible effect; they can get people into the labor force (Hughes 1995; Kain 1968; Meyer, Kain, and Wohl 1965). In the same amount of time, by contrast, a comprehensive program of urban reinvestment will likely have barely begun. Calls to reorganize transportation policy and have it address the needs of the urban poor began shortly after the Kerner Report was published. In 1973’s Social Justice and the City, David Harvey lamented the absence of a transportation policy to overcome the spatial mismatch, which he viewed as one of the city’s great engines of inequality. Citing the work of John Meyer (1968), Harvey (1973, 62) advocated what he termed an “inside-out system” for “conveying people in from central city areas to job opportunities in the suburbs.” Today, this is the most commonly prescribed remedy for the spatial mismatch, although it goes by the name “reverse commuting.” Like the inside-out strategy, the reverse commute is precisely what it sounds like—a trip to work that cuts against the traditional grain of suburb to inner city, and instead moves city residents out to the suburbs. The reasoning behind it is equally simple. If entry-level jobs have shifted outward and left low-income city workers behind, then the easiest solution may be to move workers to suburban jobs. Reverse commuting received perhaps its biggest boost when Congress passed the Job Access and Reverse Commute (JARC) program as part of the Transportation Equity Act of the 21st Century (U.S. Congress 1998; Marbach and Leckrone 2002). The JARC component of the bill specifically cites the uneven growth of jobs and allocates federal funds for projects that connect inner-city residents to suburban employment centers (49 USC § 5309). “The twofold purpose of this program,” according to California’s Department of Social Services (1999) “is to: (1) Develop transportation services designed to transport welfare recipients and lowincome individuals to and from jobs; (2) Develop transportation services for residents of urban centers and rural and suburban areas to suburban employment opportunities. Emphasis is placed on projects that use mass transportation services” (p. 3) That final sentence is important. For understandable reasons, mass transit is an appealing method of reverse commuting, particularly pleasing to transit advocates searching for ways to boost ridership on underused public transportation systems. Added to this appeal is the practical fact that many low-income adults are without cars. Although only 4 percent of higher income Americans lacks automobiles, 36 percent of the poor do not own a car (Murakami and Young 1997).8 The largest, and perhaps most obvious, reason for this disparity is that automobiles are expensive and low-income work-
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
Beyond the Spatial Mismatch ers often lack the necessary resources to purchase them. And the initial cost is by no means the only or even the largest expense. The American Automobile Association (American Automobile Association and Runzheimer International 2002) estimates that the average annual cost of owning a new car is $7,533 and $2,500 for a tenyear-old car. Such obstacles to car ownership only add resonance to the argument that public transit must be a vital component of any welfare-to-work transportation strategy. Implicit in this idea is the promise of two problems solving each other; if public transportation can help low-income adults overcome spatial barriers to employment, it validates both itself and welfare reform efforts. History has shown, however, that public transit’s record as an employment mechanism for the poor is somewhat lamentable. Public transportation has often stumbled in its attempts to connect the poor with jobs, even during times when funding was plentiful and the importance of transportation to the poor has been widely acknowledged. In the wake of the McCone Commission’s report, and subsequent assertions by John McCone himself that “the availability of public transportation directly affects, if it does not control, the employability of persons living in poverty areas,” the State of California funded the Transportation-Employment Project, which ran a new bus line through underserved parts of Los Angeles (Governor’s Commission on the Los Angeles Riots 1967, 9). The TransportationEmployment Project was intended to gauge the effect of better transit on low-income workers, and its results were sadly ambivalent. Although ridership on the new bus line was consistently strong, and the new routes made some existing commutes far less circuitous, there was little evidence that the bus contributed meaningfully to any decline in unemployment (California, Transportation-Employment Project 1970). Similar experiments run in St. Louis, Boston, and New York also yielded disappointing results (Falococchio and Cantelli 1974; Floyd 1968). “The overwhelming consensus,” O’Regan and Quigley (1999, 33) note, “is that thes e projects of the 1 9 6 0 s a nd ea rl y 1 9 7 0 s demonstrated only meager success.” More recently, studies have shown a weak statistical relationship between the availability and access to public transportation and low-income employment. In a study of low-income employment in Portland and Atlanta, Sanchez (Sanchez 1999) finds that proximity to transit might have a small positive effect on employment, but he also cautioned that his results were far from conclusive. Likewise, Thompson (1997), in an investigation of Dade County, Florida, finds only a slender connection between transit and employment. Finally, Ong and Houston (2002) find that the level of
191
transit service near welfare recipients’ homes has a small effect on increasing the probability of employment and transit use among single women in Los Angeles without access to private vehicles. One of the reasons transit has difficulty overcoming spatial barriers to employment is that spatial barriers to employment are a mass transit problem almost as much as they are a poverty problem. The same forces that have dispersed employment—suburbanization and deindustrialization—have also created serious challenges for transit agencies, which have had to contend with expanded service areas, decreasing ridership (Pucher and Renne 2003), and an emerging work/residence pattern in which the dominant commute is now from suburb to suburb (Pisarski 1996). Mismatch also confounds transit because most public transportation systems have been designed for middle-class suburban riders heading inbound to downtown areas and not for those traveling within the suburbs or heading outbound from the central city. This last point, in particular, is central. Transit works best in the inner city, where there are dense clusters of jobs and residents and where employment is frequently located adjacent to transit stops (Levinson 1992). These conditions are increasingly rare for entry-level work, and they occur with particular infrequency for reverse commutes. Suburban homes may be clustered around transit stops, but suburban employment generally is not (Holzer and Ihlanfeldt 1998; Lacombe 1998; Orski 1998; Regenstein, Meyer, and Hicks 1998). A middleclass commuter who emerges from a city rail station often finds herself in the middle of a pedestrianfriendly central business district. Low-income workers arriving at a suburban bus depot, by contrast, more likely find themselves miles away from the corporate office park or highway-side building that is their place of employment. According to Orski (1998) and the Urban Institute (Regenstein, Meyer, and Hicks 1998), 40 percent of entry-level suburban jobs are not accessible by transit routes. In their own study, Holzer and Ihlanfeldt (1998) place that figure at closer to 50 percent. Likewise, in a 1998 report on welfare recipients in Boston, the Bureau of Transportation Statistics found that that city’s comprehensive transit system was of almost no use to job-seeking welfare participants. Although 98 percent of surveyed recipients lived within a quarter mile of a transit stop, the transit system provided excellent service only within the immediate metropolitan area. Over 70 percent of the city’s entrylevel jobs, meanwhile, were in the suburbs (Lacombe 1998). Transit also tends to run in sync with the typical nineto-five business day, but low-income workers often hold second-shift jobs whose hours extend past its
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
192
Journal of Planning Literature
prime operating times (Garnett 2001). Most welfare recipients are women, and while 62 percent of all women work standard schedules, only 57 percent of low-income women do the same (Presser and Cox 1997). The disparity in work schedules becomes even greater among households with children; low-income women with children are nine to ten times more likely to work nonstandard hours than women without children (Presser and Cox 1997). The reasons for this vary. According to Presser and Cox (1997), 31 percent of lowincome married women they surveyed claimed they found “better child care” if they worked nonstandard hours, while more than half reported that the hours were requirements of the job. Presser and Cox (1997) also note that the jobs these women held were some of the country’s fastest-growing, meaning that the trend toward nonstandard hours will likely increase. Similarly, Garnett (2001) predicts that jobs with nontraditional hours will soon account for 30 percent of all new job growth. And since off-peak commutes on average are 35 percent longer than those taken during standard hours (O’Regan and Quigley 1999), nonstandard work schedules only complicate the task of getting welfare recipients to work. Further, mass transit cannot easily accommodate women’s complex and demanding travel schedules. A mother who takes two buses and more than an hour to get to a suburban job is in no position to quickly depart in the event of a family crisis. Nor is transit often amenable to “trip-chaining”—the multiple stops that many commuters, particularly single mothers, must make en route to the job. Family life, in other words, is rarely conducted on a fixed schedule, but transit is. A certain amount of tension is thus inevitable. Finally, the attention given to reverse commuting may be out of proportion to the extent to which it actually occurs. Although the working poor reverse commute more than the average worker (O’Regan and Quigley 1998a; Pisarski 1996), overall the practice is still far from common. In 2002, researchers at the University of California, Berkeley, released a comprehensive study of reverse commuting in California in the 1990s. One of its most significant findings was that over the last decade, reverse-commutes accounted for only 7 to 11 percent of all work trips. And among the relatively small percentage of reverse-commute trips, the overwhelming majority (nineteen of every twenty) was taken by automobile (Cervero et al. 2002). Even among the poor, the authors stated, carpooling dwarfed reverse commute trips taken by public transit. Public transit use was, admittedly, much more common among low-income commuters than it was among all other income groups, but the private automobile remained the chosen mode across all income groups of
getting to work. On the surface, these findings may appear counterintuitive, for transit seems, economically at least, to be the ideal fit for low-income commuters. But the problem lies less in how much transit costs and more with its conformity to the needs of lowincome travelers. The barrier to using public transit is typically time and convenience, not cost. WHAT ROLE FOR PUBLIC TRANSPORTATION?
The challenge for public transportation providers lies in targeting scarce funds where they are most effective and avoiding costly projects that are unlikely to generate results. Current research suggests that transit strategies might best be concentrated in two avenues— so-called demand-responsive services, which do not operate on fixed routes but instead cater to the specific needs of local riders, and fixed-route transit within urban areas (as opposed to routes running from urban areas to the periphery.) The first of these, it should be noted, has yet to undergo extensive empirical evaluation. Most demand-response systems aimed at lowincome riders are quite new, arising as they have in the wake of welfare reform, and their effects to this point are difficult to measure. Still, they merit some discussion. The second point about intraurban transit services is part of a much larger and older debate in transportation policy. Demand-Responsive Transit Service Demand-responsive service (sometimes called paratransit) is public transit service tailored to particular riders’ needs. The most common method of implementation is the vanpool, a shuttle that carries workers who live in roughly the same location to different job sites throughout the metropolitan area. The advantage of demand-responsive service is fairly obvious; it eliminates the expense and waste of running a fixed-route system through a geography that lacks clusters of destinations. Like the spatial mismatch hypothesis, the roots of demand-responsive transit service can be found in the urban unrest of the 1960s. In 1967, when the Transportation-Employment Project was completing its 24-month trial period of new mass transit in Los Angeles, the administrators of the project noted that buses may not be the best answer to the problem of getting Project area people to the scatterings of available jobs. Considering the overwhelming numbers of potential origins and destinations, and the fact that many of the passengers use the services only until they are able to use their own automobiles, it seems doubtful that many cases will be found where passenger volumes over fixed routes
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
Beyond the Spatial Mismatch will be of sufficient magnitude to justify the costly operation. (California, Transportation-Employment Project 1967, 23)
Instead, the project recommended “less organized and more ad hoc arrangements,” including vanpools and jitneys (California, Transportation-Employment Project 1967, 23). But the project never actually initiated such programs, and the idea fell by the wayside. In the early 1970s, however, the Urban Mass Transportation Administration sponsored a number of demonstration projects using demand-responsive transit service (Cervero 1997), and vanpools and similar arrangements became a very popular method for moving the disabled. Today a number of larger cities—Los Angeles included—contract vanpool service run by quasi-public agencies and private companies to comply with the Americans with Disabilities Act, a law that greatly increased the provision and use of demand-responsive service (Cervero 1997, Access Paratransit Services 2003). Data from the 2000 National Transit Database present the relative cost-effectiveness of public transit by this mode. According to the data, the operating expense per unlinked passenger trip is $16.74, approximately eight times that of a bus ($2.19). These services tend to be extremely expensive mostly because the cost of vehicles and personnel cannot be spread over a large number of passengers. The higher costs are also the result of the equipment needed to move the disabled and to the higher wages of drivers, who tend to earn more than transit or taxi drivers (Cervero 1997). Preliminary evidence from the newly implemented demand-responsive services for the poor indicates that they, too, will be costly endeavors. A number of pilot projects have been undertaken in cities throughout the country, and early reports on their efficacy have been mixed. While none of the projects have been as expensive as moving the disabled, costs nevertheless range from $10 to almost $20 per ride, which rivals the cost of giving every passenger a private trip in a taxicab (Cervero et al. 2002). Similar to fixed-route public transit, demandresponsive programs face the challenge of providing efficient service while accommodating the complex travel patterns of low-income commuters, particularly mothers. Bridges to Work, a project of Public/Private Ventures, began perhaps the largest experiment in demand-response transit in 1997 in metropolitan Chicago and several other cities. In Chicago, the program used a combination of express buses and vanpool services to move workers from impoverished residential areas (most of which were public housing) on the South and West sides of Chicago out to the fast-growing
193
industrial complex near O’Hare Airport. Over the course of two years, the Bridges to Work program had enrolled 1,960 people, 982 of whom were eligible for benefits. Of the 982, the Bridges program placed 599 in jobs, a placement rate of 61 percent. However, program evaluators remain uncertain about the role of transportation itself in achieving these results and, instead, stress the importance of workforce development. In a field report, the research team wrote, Yes, lack of transportation prevents many poor people from reaching job-rich locations. But most low-income job seekers will need a variety of supports to get and stay employed. Attending to these elements in program design and implementation is essential. (Elliot, Palubinski, and Tierney 1999, 19)
Most surprisingly, in its recommended list of “best practices,” the Bridges to Work team strongly asserted the importance of focusing exclusively on moving people to and from work and avoiding trips to other destinations such as child care centers. “Operating highquality, efficient, flexible, punctual transportation is hard enough,” the authors argued, “without imposing on the service the demands of getting people and their children to the day care center” (Elliot, Palubinski, and Tierney 1999, 15). Although this recommendation seems quite sensible, a contradiction is embedded in it. The appeal of demand-responsive transit stems, in large part, from its flexibility, and the need for that flexibility comes from the fact that many welfare recipients cannot separate their commute trips from trips to child care centers and other household-supporting destinations. A demand-responsive program that enforces such a separation does not solve this problem. A single mother using such a service would still need to find an entirely different way to get her children to day care; the fact that transportation providers do not want to add complexity to their schedules does not mean that complexity is removed from their clients’ lives. The work of Presser and Cox (1997) and Henle and Kinsella (1996) suggest that child care may soon be the single greatest concern of low-income working parents. Eliminating day care from transportation programs offers the benefit of making the programs work at the cost of rendering them ineffective. Despite these difficulties, paratransit may still offer one of the best strategies for moving low-income commuters, particularly in neighborhoods with relatively few employment opportunities. The existence in many cities of illicit jitneys and black-market paratransit services is a testament to how badly the urban poor need flexible options for mobility (Cervero 1997).
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
194
Journal of Planning Literature
Fixed-Route Mass Transit One of the most important lessons to be drawn from the body of research on transportation and poverty is the importance of commute time. A long commute is not a barrier to work if it can be accomplished quickly, whereas a short distance commute can be a severe obstacle if there is no speedy way to complete it. Understanding the importance of time helps identify the ideal uses of automobiles and public transit. We have already discussed why fixed-route transit often does not excel at moving people long distances outside of cities. Commutes are frequently long and fail to get riders very close to their destinations. As troublesome as these difficulties are on a daily basis, they can be even more so during the crucial period of time when welfare recipients are searching for jobs and have to navigate a variety of unfamiliar settings (Blumenberg and Haas 2002; County of Los Angeles 2000). Learning transit routes and schedules, making multiple transfers, and most of all, spending hours in travel time seem difficult to justify when the payoff—that is, a job—is, at best, uncertain. The difficulties associated with public transit actually decline somewhat once employment is secured. Tolerance for even fairly complicated transit routes rises once a routine is established, and familiarity can l e a d to a di mi ni s hed pe rcepti on of ba r r i e r s (Blumenberg and Haas 2002). Still, the convenience of automobiles is difficult to compete with in these circumstances, and this may explain the tendency of welfare recipients to “graduate” away from transit—to get jobs, save money, and then purchase vehicles (Kain and Meyer 1970; Ong 1996; Reidy 2000). Transit does excel in moving people from place to place within dense urban areas. Cities have high densities of jobs and residences; they also are more congested and have less parking, which detracts from the advantages of the private automobile. And it is in precisely these areas that low-income populations are often underserved by fixed-route public transit (Blumenberg and Hess 2003; Grengs 2002). Better and more frequent intracity service, therefore, could have a dramatic effect on the lives of the urban poor. The purchase of additional vehicles or the implementation of rapid bus service could improve on-time performance, thereby reducing the fears of tardiness to work and consequent welfare program noncompliance. Replacing circuitous routes with direct ones could also reduce travel times. And more direct service can positively alter transit rider’s perceptions of their transportation burdens, just as confusing or labyrinthine routes can dissuade them from even attempting to make trips. In some cases, simply providing better transportation information can
reduce barriers to transit ridership. In one of its early assessments of Los Angeles, the TransportationEmployment Project noted that the tortuous series of changes needed to move through Los Angeles had a demoralizing effect on low-income residents. “People who are badly in need of transportation are not using the transit facilities that exist,” the Project (California, Transportation-Employment Project 1967, 42) stated, “and the pseudo-immobility thus created unnecessarily compounds their feelings of isolation.” Over thirty years later, research on welfare participants in Marin County, California, reinforced this notion—welfare clients complained that confusion about the availability of transit in general was compounded by confusion over which services were available under the state’s welfare program (Metropolitan Transportation Commission 2002a). In transportation policy circles, there has also been significant discussion of expanding service hours of transit to better accommodate the nonstandard work shifts of many welfare participants. Some researchers believe that there is a considerable latent demand for such service and that the introduction of more latenight buses would generate substantial increases in ridership. Welfare-to-work research in Solano County, California, for example, identified the lack of so-called owl service as one of the region’s biggest barriers to employment, since many of the available warehousing and light manufacturing jobs require late shifts (Metropolitan Transportation Commission 2002b). Endeavors such as these must bear in mind other factors, however, such as perceptions of safety. Even if buses run well after midnight, it is questionable whether lone women will feel comfortable riding them on a regular basis (Schulz and Gilbert 1996). Fare Vouchers, Passenger Equity and Transit Priorities One of the most common services offered to lowincome riders is reduced transit fares, either through vouchers or special transit passes. The efficacy of these initiatives has long been a matter of some debate since much of the literature on fare pricing suggests that improvements to service, rather than lower fares, are of greater import to the poor (Blumenberg and Haas 2002; Cervero 1990; Domenicich, Kraft, and Valette 1968). But lower fares are not without their advantages. They certainly make life easier for those with low incomes, which is one reason why welfare programs typically subsidize the work-related travel of their participants. But since tardiness or missed work can result in sanctions for welfare participants, the quality of transit service is arguably more important than its price. If a bus takes too long to arrive at a stop, goes to the wrong places, or spends too long getting to the right ones, rela-
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
Beyond the Spatial Mismatch tively little is gained by making the ride cheaper. In the context of welfare reform, more and better service— that is, more buses and buses that run on time—are of potentially greater value. In an ideal world, of course, the poor would have both low fares and better service. In the current climate, however, a trade-off is more likely: lower fares mean less farebox revenue, which in turn means fewer service improvements. “Cheap transit for poor people,” as Thomas E. Lisco (1970, 63) once wrote, “almost invariably means poor transit for poor people.” Low fares are also a rather imprecise way to help the poor because there is no guarantee that a person taking advantage of a subsidized fare is in fact a person of low income (Cervero 1990; Wohl 1968). Some scholars suggest that demand-side vouchers—similar to the demand-side subsidies currently provided by social service agencies—are preferable to lower fares (Cervero 1990). Demand-side vouchers, in addition to being less paternalistic, offer recipients more transportation choices and may force transit agencies to improve services to compete for low-income riders, who might otherwise spend their voucher money on mileage reimbursements, taxi fares, or other types of transportation. Vouchers may also allow transit agencies to offer higher quality services at higher prices while making these improved services available to low-income riders. Almost all county welfare programs offer their clients transportation subsidies (Blumenberg et al. 2003); however, these subsidies clearly do not address the transportation costs for the much larger group of lowincome workers who are ineligible or otherwise do not participate in the welfare program. To the extent that fares might be adjusted to benefit the poor, they may need to be restructured more than lowered. Flat fares, which some transit agencies now use, can be inequitable because riders traveling shorter distances—who tend to be low-income (Garnett 2001; Murakami and Young 1997; Shen 2000)—subsidize those who travel farther (Cervero and Wachs 1982). Flat-fare pricing also discourages off-peak travel, which is, again (Garnett 2001; Presser and Cox 1997), mostly the province of the poor. Time- and distancebased fares are potential solutions for these problems. The question of equity in transit raises an issue far larger than fares, however, which is the overall allotment of transit funds. In recent years, the politics of mass transit financing have led to a steady shift away from inner-city, lower income riders and toward “choice” suburban commuters—people who have the option and inclination to drive. As public anxiety over traffic congestion and air pollution has increased, the role of transit has been modified, so that it now is concerned less with providing mobility to those who lack it
195
and more with luring those who do have mobility away from their automobiles (Garrett and Taylor 1999). Garrett and Taylor (1999) write that transit planners today face a “growing tension . . . between meeting the strong demand for transit services by predominantly low-income and minority residents on the one hand, and accommodating the political interests and desires of a more mobile, dispersed, and largely white suburban-based electorate on the other” (p. 7). This “growing tension” has typically resulted in a focus on rail, which is a form of public transit that appeals to higher income commuters. Income segregation in public transit is notable. While bus ridership declines as income rises, the use of streetcars, commuter rails, and subways increases (Pisarski 1996). The poor are eight times as likely as the affluent to ride a bus, and the affluent are three times as likely as the poor to ride suburban rail (Pucher and Renne 2003). In 1995, the median household income of an urban bus rider was below $20,000, while it was over $40,000 for a rail commuter (Garrett and Taylor 1999). Likewise, in Los Angeles in 1993, 57 percent of bus riders earned less than $15,000 a year, compared to 20 percent of the Los Angeles County population as a whole (Metropolitan Transportation Authority 1991-1993). Eighty-three percent of the city’s bus riders were nonwhite, and most were women (Metropolitan Transportation Authority 1991-1993). These characteristics contrast sharply with a typical Southern California rail commuter, who at that time was a white male who earned $65,000 per year, owned an automobile, and had access to free parking at work (Garrett and Taylor 1999; Rubin 1994). There is nothing intrinsic to rail that makes it useful to only the relatively affluent. Rail can provide excellent service to anyone who happens to live and work near its lines, but it is often targeted toward the nonpoor, and its allure is counterbalanced by its high costs. Pucher, Hendrickson, and McNeil (Pucher, Hendrickson, and McNeil 1981) show that the average passenger subsidy for rail is almost three times that of bus service, and Pickrell (1983, 1992) buttresses this finding with studies of rail transit costs in Miami, Portland, and San Jose. Large expenditures on rail necessarily detract from bus improvements, and this leads to the aforementioned dilemma of the poor subsidizing additional transit options for the better off (Wachs and Taylor 1998). In the 1970s, this issue motivated the successful opposition of Cleveland planners to commuter rail line extensions and helped them win expanded bus service instead (Krumholz and Forester 1990). Likewise, in the 1990s, the Labor/Community Strategy Center and Bus Riders Union, a grassroots coalition in Los Angeles, successfully sued the city’s Metropolitan Transportation Authority (MTA), alleging that the amount of funds
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
196
Journal of Planning Literature
expended on subways constituted spatial discrimination and neglected the majority of transit riders (Garrett and Taylor 1999; Grengs 2002). In this instance, although 94 percent of the MTA’s customers were bus patrons, the agency spent 70 percent of its budget on rail. Almost three-quarters of its budget, in other words, went to one-twentieth of its ridership (NAACP Legal Defense Fund 1996). Rail-bus battles have been hard-fought. Rail is the darling of middle-class environmentalists, who often see it as a vital strategy in their efforts to reduce smog, congestion, and other externalities of the private automobile. But while the elimination of air pollution and traffic are laudable goals, there is little evidence that expanded rail service, by itself, will accomplish them. Data from the 1995 Nationwide Personal Transportation Survey show that public transit captures only 2 percent of daily person miles traveled (Hu and Young 1999). Transit riders constitute a slightly higher percentage among metropolitan commuters; just fewer than 5 percent report transit as their usual mode of transportation to work (Rosetti and Eversole 1993). Still, given these figures, even doubling transit’s mode-share would do little to reduce auto travel, and there is good reason to believe that any road space freed up by drivers choosing to use transit would not stay vacant for long (Downs 1992). Other motorists would soon fill the gaps because traffic is itself a disincentive to drive, and any reduction in traffic also increases the rewards, however briefly, for those who choose to use cars. Thus, for every person who abandons a car to climb aboard a train, at least one person is likely to climb into a car and occupy the vacated road space. This phenomenon, called “triple convergence,” is a result of commuters switching from other times, other routes, and other modes in response to decreases in congestion (Downs 1992). THE ROLE OF AUTOMOBILES IN THE TRAVEL OF THE POOR
“The invention of the automobile,” William Ogden wrote in 1938, “has had more influence on society than the combined effects of Napoleon, Genghis Khan and Julius Caesar.” Kenneth Jackson (1985, 170), who cited Ogden’s claim in his history of suburbanization, added that nothing had so altered American urban form as the internal combustion engine. One can argue such sweeping statements, of course, but not the incredible attention the academic world has bestowed on the private car. The literature on automobiles and employment outcomes is far larger than that on transit and employment, no doubt because there are a great many more automobiles in the United States than there are
transit services or transit patrons and no doubt because the automobile, without question, dominates American transportation. The evidence that automobiles make it easier to find and keep work is diverse and persuasive.9 Low-income adults drive to their jobs and use automobiles to search for new jobs for the same reason everyone else does—it works. In both compact cities known for comprehensive transit systems and in larger cities known more for their auto dependency, research has shown that the private automobile is a pivotal factor in improving employment rates among the poor. The research is often compelling. Ong (1996) finds that access to automobiles correlates positively with employment, hours worked per month, and mean monthly earnings.10 Other studies (Blumenberg 2002; Cervero, Sandoval, and Landis 2002; Danziger et al. 2000; Ong 2002; Ong and Blumenberg 1998; Orski 1998; Taylor and Ong 1995; Waller and Hughes 1999) reach similar conclusions and find the automobile to be a far more powerful determinant in job seeking and job retention than public transit. Cervero et al. (2002b), in their exhaustive study of reverse-commuting in California, conclude that the weight of empirical evidence and case experience lends considerable credence to the argument that assisting the inner-city poor with the purchase of a car can stimulate employment. Statistically, . . . owning a car is a far more powerful predictor of whether people will find jobs and get off welfare than the availability and quality of transit services. (p. 173)
Cervero’s team cites evidence from three large metropolitan areas in California where peak period travel by bus took three times as long as travel by private vehicle. Kawabata (2003) finds that workers with autos are more likely to work at least thirty hours per week, more likely to be employed, and had higher earnings than people dependent on transit. “The likelihood that an individual is employed,” O’Regan and Quigley (1998b, 17) wrote, “varies a lot by auto access.” Working in the Chicago region, Metaxos et al. (1999) approached the problem from the other end and noticed that lack of auto access was twice as common in neighborhoods of high unemployment as it was in the region as a whole, and Raphael and Stoll (2000) find that auto ownership can all but eliminate interracial differences in employment. In the general population, African American unemployment exceeds that of whites by 11 percent. Among the population of those who own cars, however, African Americans are actually more likely than whites to have jobs. When the same experiment is conducted between whites and Hispanics, the employment gap between them is similarly reduced.
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
Beyond the Spatial Mismatch A note of caution is in order about these results. Causality is an issue in many studies that examine automobiles and employment outcomes. While access to personal vehicles can increase the probability of employment, employment also increases the probability of auto ownership. Addressing this issue, Ong (2002) finds that auto ownership remains an important predictor of employment. Likewise, Raphael and Rice (2002) controlled for this causality bias and still found auto access to be an important factor in employment. The advantages of the car are real and not difficult to discern. Every reason that transit has difficulty meeting the transportation needs of the poor is a reason that private vehicles succeed. Cars allow flexibility, accommodate schedules that may include unforeseen travel requirements (sick children, etc.), decrease travel time by avoiding the multiple stops and sometimes circuitous routes of public transit, and allow women to travel safely after dark. Indeed, in some instances, cars are more important to low-income workers than to higher income workers.11 The poor, as has been mentioned, are more likely to have inconsistent work schedules that require travel at off-peak hours, and cars eliminate the need to contend with isolated transit stops and long headways typical of off-peak travel. Moreover, welfare recipients can be sanctioned for arriving late to work; therefore, a late bus can result in lost benefits. In research conducted in Iowa on sanctioned families, more than half of all participants cited transportation as a major factor in their noncompliance (Goldberg 2001). “Reliability,” Cervero et al. (2002b) point out, “is of utmost importance to many [welfare] clients, in that if they arrive to work late more than once, they are usually let go” (p. ES-7). Owning a car can provide a measure of control over one’s fate and alleviate the need for dependency on buses or trains. Since transit commutes tend to be 63 to 94 percent longer in time than solo drives (Taylor and Ong 1995), using transit allows less time for trip chaining, can increase child care costs, and can result in additional travel time and fewer work hours. Evidence of this sort can upset advocates of public transportation, but the efficiency of the automobile should not be construed as an indictment of mass transit. Although the automobile is remarkably useful, many of its advantages are structural and related to issues of land use and governance rather than to any inherent superiority it may enjoy. The private automobile has a dramatic effect on employment for the simple reason that the United States is designed around cars, and as a result, those who have one will do better than those who do not. Both the efficacy of automobiles and the relative inefficacy of transit spring from the same source, which is the vast amount of public policy that
197
over the past fifty years has shaped urban development in America and turned it into an autocentric nation (Duany, Plater-Zybek, and Speck 2000; Fogelson 2002; Gutfreund 2004; Jackson 1985; MacShane 1994; Shoup forthcoming). Zoning and other land use policies have pushed residences and workplaces farther apart; policies have subsidized the individual driver with free parking, extensive freeway systems, and fuel subsidies; and local communities have accommodated the thirst for private land by promoting low-density development (Gutfreund 2004; Jackson 1985; Wachs and Taylor 1998). In such circumstances, the private car is bound to perform better—and the lack of a car, for those unable to afford or drive one, is likely to be far more crippling (Webber 1992). But the idea of the private automobile as an antipoverty instrument is hardly devoid of problems. Cars carry with them well-known social costs, including traffic congestion, air pollution, and increasing dependence on foreign oil. These costs must be incorporated into any proposal to assist welfare recipients with auto ownership; a clumsy effort at introducing cars risks unnecessarily setting environmental and social goals at odds with each other. Yet such dilemmas, although real, need not be intractable. Raphael and Stoll (2000), for instance, point out that if cars are used primarily for reverse commuting and particularly for reverse commuting at nonstandard hours, they are unlikely to make traffic congestion significantly worse. It is also worth noting that neither environmental nor transportation policy is likely to be successful if it is confined largely to those without choices. If the landscape of the United States dictates that a car is necessary for economic success and self-sufficiency, it seems neither fair nor pragmatic to discourage auto ownership among the poor while expecting them to climb beyond poverty (Blumenberg and Waller, 2003; Wachs and Taylor 1998; Waller and Hughes 1999). Most middle- and upper-income families could not make do without automobiles (Berger 1992; Rosenbloom 1992), and lowincome families, with fewer options and resources, seem unlikely to be different. Nor, it could be argued, should policy deny a single car to a poor family, in the interest of traffic or pollution mitigation, when almost 60 percent of American households have two or more vehicles and the freedom to purchase as many more as they can afford (Pucher and Renne 2003). Traffic is not a problem caused by the poor, and barring them from cars will not free our streets and highways of congestion. Spurious criticisms aside, however, there remain significant problems with the idea of low-income automobility. Cars may help low-income job seekers get employment, but they can also become financial bur-
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
198
Journal of Planning Literature
35%
32%
30% 25% 20%
24%
23%
18%
17%
16%
15%
12% 12% 9%
10%
10% 6%
5%
5%
4%
3%
5% 5%
0% Cars and trucks, new
Cars and trucks, used
Gasoline and motor oil
Vehicle finance charges
Vehicle Purchases
Maintenance and Vehicle insurance repairs
Vehicle rental, leases, licenses, other
Public transportation
Other Vehicle Expenses Total
Bottom Income Quintile
FIGURE 1. Transportation Expenditures SOURCE: U.S. Department of Labor (2002).
dens in their own right. Cars in low-income households are nearly 40 percent older than the average car (Murakami and Young 1997), and these older vehicles are often less than a year away from extensive repairs and, as mentioned previously, can be unreliable. Moreover, with auto ownership comes added costs and responsibilities, such as fuel, maintenance and repairs, auto insurance, and monthly loan payments. Data from the 2002 Consumer Expenditure Survey show that transportation expenditures account for 17 percent of the total expenditures in low-income households (households with incomes in the bottom quintile), third in size following housing and food (Bureau of Labor Statistics 2002). Since most low-income households travel using automobiles, it is not surprising that 95 percent of these transportation expenditures are vehiclerelated. As Figure 1 shows, certain expenses constitute a higher percentage of expenditures in lower-income households than in all households; these include used cars and trucks, gasoline and motor oil, and maintenance and repairs. Although low-income households spend a smaller percentage on transportation (17%) than higher income households (18-21%), transportation costs may still be burdensome, taking necessary resources away from other important expenses such as housing and food. Even if welfare recipients could afford better cars, in many states they are not allowed to have them. A still
frequent component to state welfare programs is the vehicle asset limitation, which prohibits anyone on welfare from owning a vehicle worth more than a certain value. While the asset limitation for at least one vehicle has been lifted in approximately 50 percent of all states, the remaining states still impose some type of limit (The Urban Institute 2002).12 For example, California, the state with more welfare recipients than any other, maintains an asset limitation of $4,650. And for people living in inner cities—a category that includes the majority of welfare recipients—automobile insurance is prohibitively expensive.13 Lastly, of course, there are political barriers to assisting welfare recipients with the purchase of automobiles. Welfare reform was spawned in a climate of hostility to welfare itself and sprang in particular from a notion that too-generous government programs were compounding poverty by creating an endless cycle of complacency and entitlement (Cloward and Piven 1987; Glazer 1971; Murray 1984). The veracity of these notions may be debatable, but their prevalence is not, and programs designed to help welfare recipients own automobiles would need to contend with them. These obstacles have not stopped a number of researchers (Blumenberg and Waller 2003; Gardenhire 2000; O’Regan and Quigley 1998a, 1998b; Ong 1996, 2002; Orski 1998; Shen 2001; Taylor and Ong 1995; Waller and Hughes 1999) from recommending that
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
Beyond the Spatial Mismatch public funds be spent helping welfare recipients secure cars instead of on expanding public transit services. The laws governing the welfare program allow states to use federal block grants to provide direct purchase assistance for automobiles, to help pay for insurance, and to provide loans for would-be buyers. The last option, in particular, seems to hold promise, for it not only helps welfare participants get cars but also helps them build (or rebuild) credit while establishing a relationship between them and local banks. Ways to Work, a partnership that secures two-year loans for Temporary Assistance for Needy Families (TANF) recipients in twenty-three cities across seventeen states, reported in 2001 that over 85 percent of its loans had gone to vehicle purchases (Goldberg 2001). Moreover, early evaluations of the program show that participants average as much as a 20-percent increase in monthly income once the loan is received—testament again to the economic impact of personal vehicles (Goldberg 2001). Whether providing auto access to the poor is advisable or not, it seems inarguable that the poor themselves place a high priority on obtaining cars. According to Pucher and Renne (2003), who comprehensively review evidence from the 2001 National Household Travel Survey, a car is obviously one of the first purchases households make as soon as they can, even if it strains their already limited budgets. Indeed, it is probably unique to the United States that three-quarters of even its poorest households own a car. That reflects the extent to which the car has become a virtual necessity for even the most basic transportation needs in most American metropolitan areas. (p. 55) CONCLUSIONS
Almost forty years ago Meyer, Kain, and Wohl (1965, 2) wrote that the true advantage of transportation improvements was their ability, amid the web of urban policies, to offer “some possibility of action.” Other problems, by contrast, were “so intrinsically complex and politically explosive as to defy easy analysis or solution.” This statement is no less true today and may even be more so. It is at once a reason to embrace transportation solutions and to be skeptical of them. Transportation remedies may be inherently incomplete, but they are also speedy, and in an era of time-limited welfare, this is no small advantage. There is little question that welfare recipients face spatial barriers to employment. But while the conventional notion of the spatial mismatch may still apply in some metropolitan areas and for some low-income residents, spatial barriers to employment are numerous,
199
and conventional mismatch is only one of a host of obstacles. For the poor as well as nonpoor, work is not often in close proximity to home. The difference is that for the nonpoor, traveling between two locations is often much easier. This idea of a modal mismatch is i ncre a s i ng l y domi na ti ng the l i te r a tu re on transportation, poverty, and employment. To be effective, therefore, transportation policies aimed at the poor must also look beyond the spatial mismatch literature. The research on transportation and employment has increasingly examined the important role of mode and, in particular, the pivotal role of cars in the economic fortunes of the poor. The evidence on the relationship between employment and public transit is much weaker. These findings are related to distance-based barriers and the land-use decisions that create them. The same policies that helped create the spatial mismatch have also challenged the utility of public transportation, making it less effective even as it became more necessary. It therefore is unrealistic and perhaps unfair to think that mass transit will fill the transportation gaps faced by the poor. As Wachs and Taylor (1998) note, transit and the urban poor suffer from the same affliction, which is a national urban policy that for too many years has been harmful to urban areas. To saddle transit with the responsibility of bringing welfare recipients to work ignores this fact, constructs a scenario for transit’s failure, and thus risks the further erosion of its credibility. We must be careful, however, not to paint an overly bleak picture. Recognizing the limitations that American urban form places on transit’s role in poverty reduction also allows us to better understand those cases where transit is likely to be effective. Demand-responsive transit, for instance, shows early promise and rests on a built-in understanding of the realities behind urban and suburban growth. Although at this point it seems quite expensive per ride, it can provide door-todoor service rivaling that of the automobile, it requires less capital outlay than bus or rail, and its flexibility allows vehicles to be shifted easily from one area of a city to another.
Transit can also make a contribution by reorienting itself to better serve inner-city populations. In many urban areas, there are still high concentrations of employment located in the central city adjacent to lowincome neighborhoods, providing ample opportunity for improvement in intraurban mobility. Even if many low-income residents do not use transit as their primary commute mode, they will most certainly use it in countless other facets of their lives (Blumenberg and Haas 2002). Improved intraurban transit will also allow for better access to those jobs that remain in metropoli-
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
200
Journal of Planning Literature
tan areas and shorten commute times that are now prohibitively long even within the central city. The final lessons to be derived from the current research on welfare and transportation are that transportation is but one component in a large and vexing problem, and commuting is but one aspect of that component. While low-wage workers may easily commute to and from work, their job prospects may be greatly diminished by their inability to travel in search of employment. Similarly, low-wage mothers may find commuting on public transit relatively problem free. However, nonwork travel—trips to the child care center, schools, job training and welfare offices, and so forth—may greatly complicate commute trips and make public transit an untenable option. Seamless access to jobs and transportation will be meaningless unless welfare recipients find employment that pays livable wages. They may also be ineffective unless coupled with adequate education and training that allows low-income adults to compete s u cces s f u l l y f or empl oy me nt oppor tu ni ti e s (Blumenberg 2002; Danziger et al. 2000). And neither education nor transportation will matter if taking jobs forces young women to leave their children unsupervised. Transportation will never be a panacea, and any effort at poverty reduction must, by necessity, factor in the multiple concerns of those it seeks to aid. But transportation policy can, if crafted wisely, greatly improve the plight of welfare recipients and help them move beyond their current mobility constraints. ACKNOWLEDGMENTS
We thank the California State Department of Transportation and the Norman Y. Mineta International Institute for Surface Transportation Policy Studies for their financial support of this research. We also thank Mark Garrett, whose insights greatly strengthened the article. Although numerous individuals and institutions have encouraged, commented on, and supported this research, responsibility for all errors lies with the authors. NOTES 1. Temporary Assistance to Needy Families (TANF), the program that, under welfare reform, replaced Aid to Families with Dependent Children (AFDC). 2. Welfare Recipients tend to be more spatially concentrated than the working poor and, therefore, are likely to compete with other lowskilled workers whose residential locations are geographically much less concentrated. 3. However, it is important to acknowledge that the absolute number of jobs may be less important than the relative competition for each job vacancy. In densely populated inner cities, many job seek-
ers may vie for each opening. This competition not only reduces the chances for employment but also, by virtue of demand soaring past supply, can depress wages (Kawabata 2003; Leonard and Kennedy 2001; Ong 1996). Kirschenman and Neckerman (1991) argue that the labor/jobs ratio is exacerbated by firms that tend to avoid recruiting in low-income neighborhoods. Ironically, those employers who do tap this reservoir of labor can often get employees to work at very advantageous wage rates.
4. It would be inaccurate to state that Los Angeles was originally designed around the automobile. Contrary to popular perceptions, Los Angeles’ land use pattern began with the railroad and interurban streetcar companies and then was solidified by the automobile and the development of the freeway system. But the point remains the same—the idea of a pedestrian- or transit-oriented city had been dispensed with by the time Los Angeles began to grow (Wachs 1996). 5. An emphasis on those who are employed necessarily results in a sample selection bias, excluding those who are unemployed because of difficulties traveling to available employment. However, Deka (2002) estimates that the mean commute time of the nonworking poor is likely to be substantially lower than that of those currently employed.
6. A number of other studies on women show that longer distance commutes may not be justified given the narrow range of wages in feminized occupations (Bianchi and Spain 1996; JohnstonAnumonwo 1997; Rutherford and Wekerle 1988) 7. McLafferty and Preston (1997) show, however, that not all women commute shorter distances than men; these results can vary by race and ethnicity. 8. Statistics on auto ownership tend to vary but for the most part tell the same story. Being poor, living in the central city, and being nonwhite all contribute to the absence of autos. Citing vehicle asset data from the U.S. Department of Health and Human Services, the General Accounting Office (1998) reported that less than 6 percent of welfare recipients owned cars, but this figure is considerably lower than most other studies that show ownership figures as high as 65 percent (Blumenberg and Haas 2002; County of Los Angeles 2000; Edin and Lein 1997; Federman et al. 1996; Murakami and Young 1997). 9. The lack of reliable transportation—and, in particular, the absence of automobiles—has also been linked to nonemployment outcomes, such as unmet food and housing needs and difficulty traveling for medical care (Corcoran, Heflin, and Siefert 1999; Zogby and Malin 2001). 10. It did not, however, have any effect on wages; a car seems to allow one to work more hours but not to get paid more for each hour worked (Ong 1996). 11. The manner in which the poor spend their money tends to support this assertion. Although nonpoor households have higher auto ownership rates than poor ones, Gardenhire (2001) finds that poor households “convert additional income into autos at twice the rate of non-poor households.” 12. See Urban Institute (2002), http://www.urban.org/Content/ Research/NewFederalism/AboutANF/AboutANF.htm for a description of the data. 13. The insurance issue deserves a bit more examination as it is particularly troublesome. In 1998, the Joint Economic Committee of Congress released “Auto Choice: Impact on Cities and the Poor” (Miller 1998), which enumerates the ways in which current insurance policies exacerbate poverty and inequality. In inner cities, auto insurance is exorbitant, and costs on average $1,000 per year more than in the suburbs. In central city Los Angeles, for example, the average insurance for a 38-year old female with a clean driving record is almost $3,500 per year. This is a case, of course, of the largest expense
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
Beyond the Spatial Mismatch falling on those with the fewest resources. Families in the bottom income quintile who buy auto insurance spend seven times the percentage of their incomes on it than do families in the top quintile— those at the bottom often spend more than the face value of their car. High costs occur for two reasons. First, inner-city accidents, although far less likely to cause serious injury, are far more likely to generate insurance claims. Large numbers of claims naturally drive up premiums. The second reason is more insidious, for it feeds on itself. Families unable to afford insurance, but who nevertheless need cars, often drive uninsured. Uninsured drivers become criminalized, which further hurts their chances of climbing out of poverty, and their presence on the road causes premiums to rise still higher, which increases insurance costs for everyone else. As premiums rise, additional drivers are priced out of the insurance system, which leads to more uninsured drivers, and the cycle begins anew. A 1995 survey conducted in California revealed that 28 percent of the state’s drivers are uninsured (in Los Angeles County the figure was 37 percent) and that California drivers paid an extra $1 billion in premiums as a result. REFERENCES Access Paratransit Services. 2003 Access paratransit rider’s guide. Los Angeles, CA: Author. http://www.asila.org/accessparatransit/ guide/index.cfm (accessed June 25, 2004). Allard, Scott W. 2002. The urban geography of welfare reform: Spatial patterns of caseload dynamics in Detroit. Social Science Quarterly 83 (4): 1044-62. Allard, Scott W., and Sheldon Danziger. 2003. Proximity and opportunity: How residence and race affect the employment of welfare recipients. Housing Policy Debate 13 (4): 675-700. Alonso, William. 1964. Location and land use; Toward a general theory of land rent. Cambridge, MA: Harvard University Press. American Automobile Association and Runzheimer International. 2002. Your driving costs. Heathrow, FL: American Automobile Association. http://www.ouraaa.com/news/library/drivingcost/. Bania, Neil, Claudia Coulton, and Laura Leete. 1999. Welfare reform and access to job opportunities in the Cleveland metropolitan area. Working Paper of the Center for Urban Poverty and Social Change, Cleveland, OH: Case Western Reserve University. http:// povertycenter.cwru.edu/Fordv4.PDF. Berger, Michael. 1992. The car’s impact on the American family. In The car and the city, edited by Martin Wachs and Margaret Crawford. Ann Arbor: University of Michigan Press. Bianchi, Suzanne, and Daphne Spain. 1996. Balancing act: Motherhood, marriage and employment among American women. New York: Russell Sage. Bluestone, Barry, and Bennett Harrison. 1982. The deindustrialization of America: Plant closings, community abandonment, and the dismantling of basic industry. New York: Basic Books. Blumenberg, Evelyn. 2002. On the way to work: Welfare participants and barriers to employment. Economic Development Quarterly 16 (4): 314-25. Blumenberg, Evelyn, and Peter Haas. 2002. The travel behavior and needs of the poor: A study of welfare recipients in Fresno County. San Jose, CA: Mineta Transportation Institute, San Jose State University. Blumenberg, Evelyn, and Daniel Baldwin Hess. 2003. Measuring the role of transportation in facilitating the welfare-to-work transition: Evidence from three California counties. Journal of the Transportation Research Board 1859:93-101.
Blumenberg, Evelyn, Douglas Miller, Mark Garrett, Lisa Schweitzer, Karen Kitsis, Michael Manville, and Bravishwar Mallavarapu. 2003. California transportation needs assessment: The transportation barriers of welfare recipients and low-wage workers. Los Angeles: UCLA Lewis Center for Regional Policy Studies.
201
Blumenberg, Evelyn, and Paul Ong. 1998. Job accessibility and welfare usage: Evidence from Los Angeles. Journal of Policy Analysis and Management 17 (4): 639-57.
. 2001. Cars, buses, and jobs: Welfare recipients and employment access in Los Angeles. Journal of the Transportation Research Board 1756:22-31. Blumenberg, Evelyn, and Margy Waller. 2003. The long journey to work: A federal transportation policy for urban families.. Washington, DC: Brookings Institution. Briggs, Xavier de Souza. 1997. Moving up versus moving out: Neighborhood effects in housing mobility programs. Housing Policy Debate 8 (1): 195-233.
Browne, I. 2000. Opportunities lost? Race, industrial restructuring, and employment among young women heading households. Social Forces 78 (3): 907-29. Bureau of Labor Statistics. 2002. 2002 Consumer expenditure survey. Washington, DC: Bureau of Labor Statistics. California, Transportation-Employment Project. 1967. A research project of the State of California to determine and test the relationship between a public transportation system and other opportunities of low income groups. Sacramento, CA: Business and Transportation Agency.
. 1970. A research project of the State of California to determine and test the relationship between a public transportation system and other opportunities of low income groups. Sacramento, CA: Business and Transportation Agency. California Department of Social Services. 1999. All-county information notice 1-44-99, to all county welfare directors and welfare-to-work coordinators. Sacramento, CA: California Department of Social Services. Carlson, V., and Nic Theodore. 1997. Employment availability for entry-level workers: An examination of the spatial mismatch hypothesis in Chicago. Urban Geography 18:228-42. Center for Economic Development. 1998. The economic state of Milwaukee: The city and the region, 1998. Milwaukee: University of Wisconsin–Milwaukee. Cervero, Robert. 1990. Transit pricing research. Transportation 17 (2): 117-39. . 1997. Paratransit in America: Redefining mass transportation. Westport, CT: Praeger. Cervero, Robert, Onesimo Sandoval, and John Landis. 2002. Transportation as a stimulus of welfare-to-work—Private versus public mobility. Journal of Planning Education and Research 22 (1): 50-63. Cervero, Robert, Yu-Hsin Tai, Martin Wachs, Elizabeth Deakin, Joulia Dibb, Andrew Kluter, Cornelius Nuworsoo, Irina Petrova, and M. Reinhur Pohan. 2002. Reverse commuting and job access in California: Markets, needs and policy prospects. Berkeley: University of California, Institute for Transportation Studies. Cervero, Robert, and Martin Wachs. 1982. An answer to the transit crisis: The case for distance-based fares. Journal of Contemporary Studies 5 (2): 59-70.
Cloward, Richard, and Frances Fox Piven. 1987. Sources of the contemporary relief debate. In The mean season: The attack on the welfare state, edited by Fred L. Block, Richard Cloward, Barbara Ehrenreich, and Frances Fox Piven. New York: Pantheon.
Cooke, Thomas J., and Stephen L. Ross. 1999. Sample selection bias in models of commuting time. Urban Studies 36 (9): 1597-611. Corcoran, Mary E., Colleen M. Heflin, and Christine Siefert. 1999. Food insufficiency and material hardship in Post-TANF welfare families. Ohio State Law Journal 60 (4): 1395-422. County of Los Angeles. 2000. Assessing the transportation needs of welfare-to-work participants in Los Angeles County. Los Angeles: Urban Research Division, Chief Administrative Office. Cutler, David M., and Edward M. Glaeser. 1997. Are ghettos good or bad? Quarterly Journal of Economics 112 (3): 827-72.
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
202
Journal of Planning Literature
Danziger, Sandra, Mary Corcoran, Sheldon Danziger, Colleen M. Heflin, Ariel Kalil, Judith Levine, Daniel Rosen, Kristin S. Seefeldt, Kristine Siefert, and Richard M. Tolman. 2000. Barriers to the employment of recipients. In Prosperity for all? The economic boom and African Americans, edited by R. Cherry and W. M Rodgers III. New York: Russell Sage. Deka, Deva. 2002. Predicting commute time of non-workers in the context of welfare reform. Journal of Urban Affairs 24 (3-4): 333-52.
DeRango, Kelly. 2001. Can commutes be used to test the spatial mismatch hypothesis? Urban Studies 38 (9): 1521-529. Domenicich, T., F. Kraft, and J. Valette. 1968. Estimation of urban passenger and travel behavior: An economic demand model. Highway Research Record 238:64-78. Downs, Anthony. 1968. Alternative futures for the American ghetto. Daedalus XCVII (Fall 4): 1331-78. . 1973. Opening up the suburbs. New Haven, CT: Yale University Press. . 1992. Stuck in traffic: Coping with peak-hour traffic congestion. Washington, DC: Brookings Institution. Duany, Andres, Elizabeth Plater-Zybek, and Jeff Speck. 2000. Suburban nation: The rise of sprawl and the decline of the American dream. New York: North Point.
Edin, Kathryn, and Laura Lein. 1997. Making ends meet: How single mothers survive welfare and low-wage work. New York: Russell Sage. Elliot, Mark, Beth Palubinski, and Joseph Tierney. 1999. Overcoming roadblocks on the way to work. The Bridges to Work field report. Philadelphia: Public/Private Ventures. Ellwood, David T. 1986. The spatial mismatch hypothesis: Are there teenage jobs missing in the ghetto? In The black youth employment crisis, edited by Richard B. Freeman and Harry J. Holzer. Chicago: University of Chicago Press. Falococchio, John, and Edmund J. Cantelli. 1974. Transportation and the disadvantaged. Lexington, MA: D. C. Heath. Federal Highway Administration. 1994. NPTS urban travel patterns. 1990 Nationwide Personal Transportation Survey (FHWA-PL-94-018). Washington, DC: Office of Highway Information Management. . 1995. Our nation’s travel: 1995 early results report. Washington, DC: U.S. Department of Transportation. Federman, Maya, Thesia I. Garner, Kathleen Short, W. Bowman Cutter IV, John Kiely, David Levine, Duane McDough, and Marilyn McMillen. 1996. What does it mean to be poor in America? Monthly Labor Review 119 (5): 3-17. Floyd, Thomas H. 1968. Using mass transportation to alleviate poverty: A progress report on experiments under the Mass Transportation Act. Paper presented at the American Academy of Arts and Sciences, Conference on Transportation and Poverty. June 7, 1968, Brookline, MA. Fogelson, Robert. 1967. White on black: A critique of the McCone Commission Report. In Mass violence in America: The Los Angeles riots, edited by Robert Fogelson. Salem, NH: Ayer Company. . 2002. Downtown: Its rise and fall. New Haven, CT: Yale University Press.
Gardenhire, Alissa D. 2000. Employment effects: How do transportation barriers impact the labor force participation of the poor? In Proceedings of the 79th Annual Transportation Research Board Meeting. Washington, DC: National Academies Press. . 2001. Understanding automobile ownership behavior of lowincome households: How behavioral differences may influence transportation policy. In Personal travel: The long and the short of it, from the Transportation Research Board Conference Proceedings, June 28-July 1, 1999. Washington, DC: Transportation Research Board.
Garnett, Nicole Stell. 2001. The road from welfare to work: Informal transportation and the urban poor. Harvard Journal on Legislation 38 (1): 173-229. Garrett, Mark, and Brian Taylor. 1999. Reconsidering social equity in public transit. Berkeley Planning Journal 13:6-27. Glazer, Nathan. 1971. The limits of social policy. Commentary 52 (3): 51-8.
Goldberg, Heidi. 2001. State and county supported car ownership programs can help low-income families secure and keep jobs. Washington, DC: Center on Budget and Policy Priorities. Gordon, Linda. 1994. Pitied but not entitled. Single mothers and the history of welfare. Cambridge, MA: Harvard University Press. Gordon, Peter, Ajay Kumar, and Harry Richardson. 1989. Gender differences in metropolitan travel behavior. Regional Studies 23 (6): 499510.
Governor’s Commission on the Los Angeles Riots. 1966. Violence in the city: An end or a beginning? Final report. Los Angeles: Author. http:// www.usc.edu/isd/archives/cityinstress/mccone/. . 1967. Staff report of actions taken to implement the recommendations of the commission’s report. Los Angeles: Author. Grengs, Joe. 2002. Community-based planning as a source of political change: The transit equity movement of Los Angeles’ bus riders union. Journal of the American Planning Association 68 (2): 165-78. Gutfreund, Owen. 2004. Twentieth century sprawl: Highways and the reshaping of the American landscape. Oxford: Oxford University Press. Hanson, Susan, and Geraldine Pratt. 1995. Gender, work and space. New York: Routledge. Harrison, Bennett. 1974. Urban economic development. Washington, DC: Urban Institute. Harvey, David. 1973. Social justice and the city. Baltimore: Johns Hopkins University Press. Henle, T., and A. Kinsella. 1996. Welfare reform and child care: The welfare reform debate. Critical policy issues. Dekalb: Northern Illinois University, Office for Social Policy Research. Holzer, Harry J. 1991. The spatial mismatch hypothesis: What has the evidence shown? Urban Studies 28 (1): 105-22. Holzer, Harry J., and Keith R. Ihlanfeldt. 1998. Customer discrimination and employment outcomes for minority workers. The Quarterly Journal of Economics 113 (3): 835-67. Holzer, Harry J., Keith R. Ihlanfeldt, and David L. Sjoquist. 1994. Work, search, and travel among white and black youth. Journal of Urban Economics 35 (3): 320-45. Hu, Patricia S., and Jennifer R. Young. 1999. Summary of travel trends: 1995 Nationwide Personal Transportation Survey. Washington, DC: Federal Highway Administration, U.S. Department of Transportation. Hughes, Mark Allan. 1995. A mobility strategy for improving opportunity. Housing Policy Debate 6 (1): 271-97.
Ihlanfeldt, Keith. 1994. The spatial mismatch between jobs and residential locations within urban areas. Cityscape 1 (1): 219-44. . 1999. Is the labor market tighter outside the ghetto? Papers in Regional Science 78 (3): 341-63. Ihlanfeldt, Keith R., and David L. Sjoquist. 1998. The spatial mismatch hypothesis: A review of recent studies and their implications for welfare reform. Housing Policy Debate 9 (4): 849-92. Ihlanfeldt, Keith R., and Madelyn V. Young. 1994. Housing segregation and the wages and commutes of urban blacks: The case of Atlanta fast-food restaurant workers. Review of Economics and Statistics 76:425-33. Jackson, Kenneth T. 1985. Crabgrass frontier: The suburbanization of the United States. New York: Oxford University Press.
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
Beyond the Spatial Mismatch Jencks, Christopher, and Susan E. Mayer. 1990. Residential segregation, job proximity, and black job opportunities. In Inner-city poverty in the United States, edited by Lawrence E. Lynn Jr., and Michael G. H. McGeary. Washington, DC: National Academy Press. Johnston-Anumonwo, I. 1997. Racial differences in the commuting behavior of women in Buffalo, 1980-1990. Urban Geography 12: 542-62.
Julnes, George, and Anthony Halter. 2000. Illinois study of former TANF clients, final report. Springfield: Institute for Public Affairs, University of Illinois. Kain, John. 1968. Housing segregation, negro employment, and metropolitan decentralization. Quarterly Journal of Economics 82 (2): 175-97.
. 1992. The spatial mismatch hypothesis: Three decades later. Housing Policy Debate 3 (2): 371-460. Kain, John, and Joseph J. Persky. 1969. Alternatives to the gilded ghetto. The Public Interest 14 (Winter): 74-87. Kain, John F., and John R. Meyer. 1970. Transportation and poverty. The Public Interest 18 (Winter): 75-87.
Kasarda, John. 1983. Entry-level jobs, mobility, and urban minority unemployment. Urban Affairs Quarterly 19 (1): 21-40. . 1985. Urban change and minority opportunities. In The new urban reality, edited by Paul E. Peterson. Washington, DC: Brookings Institution. . 1988. Jobs, migrations and emerging urban mismatches. In Urban change and poverty, edited by Michael McGreary and Lawrence Lynn. Washington, DC: National Academy Press. . 1989. Urban industrial transition and the underclass. Annals of the American Academy of Political and Social Science 501:26-47. Kasinitz, Philip, and J. Rosenberg. 1996. Missing the connection: Social isolation and employment on the Brooklyn waterfront. Social Problems 43 (2): 180-96. Kawabata, Mizuki. 2003. Job accessibility and employment outcomes for low-skilled autoless workers in US metropolitan areas. Environment and Planning (A) 35 (9): 1651-68. Kirschenman, Joleen, and Kathryn M. Neckerman. 1991. “We’d love to hire them, but . . .”: The meaning of race for employers. In The urban underclass, edited by Christopher Jencks and Paul E. Peterson. Washington, DC: Brookings Institution. Krumholz, Norman, and John Forester. 1990. Making equity planning work. Philadelphia: Temple University Press. Lacombe, Annalyn. 1998. Welfare reform and access to jobs in Boston (BTS98-A-02). Cambridge, MA: Volpe National Transportation Systems Center, U.S. Department of Transportation, Bureau of Transportation Statistics.
Leonard, Paul, and Maureen Kennedy. 2001. What cities need from welfare reform reauthorization. Washington, DC: Brookings Institution Center on Metropolitan and Urban Policy. Levinson, H. S. 1992. System and service planning. In Public transportation, 2nd ed., edited by George E. Gray and Lester A. Hoel. Englewood Cliffs, NJ: Prentice Hall. Lisco, Thomas E. 1970. Mass transportation: Cinderella in our cities. The Public Interest 18:52-74.
Loveless, Shirley. 1999. Access to jobs: Intersection of transportation, social and economic development policies—Challenge for transportation planning in the 21st century. In Refocusing transportation planning for the 21st century: Proceedings of two conferences. Washington, DC: Transportation Research Board.
MacDonald, Heather I. 1999. Women’s employment and commuting: Explaining the links. Journal of Planning Literature 13 (3): 267-83. MacShane, Clay. 1994. Down the asphalt path: The automobile and the American city. New York: Columbia University Press. Madden, Janice F. 1981. Why women work closer to home. Urban Studies 18 (2): 181-94.
203
Marbach, Joseph R., and J. Wesley Leckrone. 2002. Intergovernmental lobbying for the passage of TEA-21. Publius 32 (1): 45-64. Massey, Douglas, and Nancy A. Denton. 1993. American apartheid: Segregation and the making of the underclass. Cambridge, MA: Harvard University Press. McLafferty, Sara, and Valerie Preston. 1997. Gender, race and the determinants of commuting: New York in 1990. Urban Geography 18 (3): 191-212.
Metaxos, Paul, Siam Soot, Hashish Sen, Vane Thakuriah, J. Jarzab, and J. DiJohn. 1999. A transportation planning process for linking welfare recipients to jobs. Transportation Research Record 1726:24-32. Metropolitan Transportation Authority. 1991-1993. Origin-destination passenger survey. Los Angeles: Author. Metropolitan Transportation Commission. 2002a. Marin County welfare to work transportation plan. Prepared by Nelson/Nygaard Associates, San Francisco. Oakland, CA: Author.
. 2002b. Solano County welfare to work transportation plan. Prepared by Nelson/Nygaard Associates, San Francisco. Oakland: Author. Meyer, John. 1968. Urban transportation. In The metropolitan enigma, edited by J. Q. Wilson. Cambridge, MA: Harvard University Press. Meyer, John R., John F. Kain, and Martin Wohl. 1965. The urban transportation problem. Cambridge, MA: Harvard University Press.
Miller, Dan. 1998. Auto choice: Impact on cities and the poor. Report submitted to the Joint Economic Committee of the 105th Congress. Washington, DC: Government Printing Office. Murakami, Elaine, and Jennifer Young. 1997. Daily travel by persons with low income. Paper presented at the NPTS symposium in Bethesda, Maryland, October 29-31. Washington, DC: Federal Highway Administration, U.S. Department of Transportation. http://npts.ornl.gov/npts/1995/Doc/LowInc.pdf. Murray, Charles A. 1984. Losing ground: American social policy, 19501980. New York: Basic Books. Muth, Richard F. 1969. Cities and housing. Chicago: University of Chicago Press. NAACP Legal Defense Fund. 1996. Labor/Community Strategy Center vs. Los Angeles County Metropolitan Transportation Authority, Plaintiff’s revised statement of contentions of fact and law. http:// www.ldfla.org/mta_fr.htm. National Advisory Commission on Civil Disorders. 1968. Report of the National Advisory Commission on Civil Disorders. Washington, DC: Government Printing Office. Norton, R. D., and John Rees. 1979. The product cycle and the spatial decentralization of American manufacturing. Regional Studies 13 (2): 141-51. Ong, Paul. 1996. Work and car ownership among welfare recipients. Social Work Research 2 (4): 255-62. . 2002. Car ownership and welfare-to-work. Journal of Policy Analysis and Management 21 (2): 239-52. Ong, Paul, and Evelyn Blumenberg. 1998. Job access, commute and travel burden among welfare recipients. Urban Studies 35 (1): 77-93. Ong, Paul M., and Doug Houston. 2002. Transit, employment and women on welfare. Urban Geography 23 (4): 344-64. O’Regan, Katherine, and John M. Quigley. 1998a. Cars for the poor. Access 12:20-5. . 1998b. Spatial isolation, transportation, and welfare recipients: What do we know? Shortened version of paper presented to the UCLA Conference on Transportation and Welfare Reform. Los Angeles: University of California. . 1999. Accessibility and economic opportunity. In Essays in transportation economics and policy, edited by Jose A. Gomez-Ibanez, William B. Tye, and Clifford Winston. Washington, DC: Brookings Institution. Orski, C. 1998. Welfare to work. Innovation Briefs 2 (3): 1-2.
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
204
Journal of Planning Literature
Perle, Eugene D., Harald Bauder, and Nancy Beckett. 2002. Accessibility measures in spatial mismatch models. Professional Geographer 54 (1): 106-10. Peterson, Paul. 1981. City limits. Chicago: University of Chicago Press.
Pickrell, Donald. 1983. The causes of rising transit operator deficits (No. DOT-I-83-47). Washington, DC: U.S. Department of Transportation. . 1992. A desire named streetcar: Fantasy and fact in rail transit planning. Journal of the American Planning Association 58 (2): 158-76. Piore, Michael, and Charles Sabel. 1984. The second industrial divide: Possibilities for prosperity. New York: Basic Books.
Pisarski, Alan. 1996. Commuting in America II: The second national report on commuting patterns and trends. Landstowne, VA: Eno Transportation Foundation. Presser, Harriet B., and Amy G. Cox. 1997. The work schedules of loweducated American women and welfare reform. Monthly Labor Review 120 (4): 25-34.
Preston, Valerie, Sarah McLafferty, and X. F. Liu. 1998. Geographical barriers to employment for American-born and immigrant workers. Urban Studies 35 (3): 529-45. Preston, Valerie, and Sara McLafferty. 1999. Spatial mismatch research in the 1990s: Progress and potential. Papers in Regional Science 28:387-402. Private Industry Council of Philadelphia. 1998. Greater Philadelphia works: Proposed plan for the expenditure of the federal welfare to work grant. Philadelphia: Private Industry Council of Philadelphia. Pucher, John, Chris Hendrickson, and Sue McNeil. 1981. Socioeconomic characteristics of transit riders: Some recent evidence. Traffic Quarterly 35 (3): 461-83. Pucher, John, and John L. Renne. 2003. Socioeconomics of urban travel: Evidence from the 2001 NHTS. Transportation Quarterly 57 (3): 49-77. Raphael, Steven. 1998. The spatial mismatch hypothesis and black youth joblessness: Evidence from the San Francisco Bay Area. Journal of Urban Economics 43 (1): 79-111. Raphael, Steven, and Lorien Rice. 2002. Car ownership, employment, and earnings. Journal of Urban Economics 52 (1): 109-30. Raphael, Steven, and Michael Stoll. 2000. Can boosting minority carownership rates narrow inter-racial employment gaps? Chicago: Joint Center for Poverty Research. Regenstein, Marsha, Jack A. Meyer, and Jennifer Hicks. 1998. Job prospects for welfare recipients: Employers speak out (No. A-25). Washington, DC: Urban Institute. Reidy, Matthew F. 2000. The longest commute: The geography of poverty, employment, and services. New England Journal of Public Policy 16 (1): 53-68. Rosenbaum, James. 1995. Changing the geography of opportunity by expanding residential choice: Lessons from the Gautreaux program. Housing Policy Debate 6 (1): 231-70. Rosenbloom, Sandra. 1992. Why working families need a car. In The car and the city, edited by Martin Wachs and Margaret Crawford. Ann Arbor: University of Michigan Press. . 1994. Travel by women. Washington, DC: Federal Highway Administration . 1998. Transit markets of the future. The challenge of change. Washington, DC: National Academy Press. Rosenbloom, Sandra, and Elizabeth Burns. 1994. Why working women drive alone: Implications for travel reduction programs. Transportation Research Record, 1459:39-45. Rosetti, Michael A., and Barbara S. Eversole. 1993. Journey to work trends in the United States and its major metropolitan areas, 1960-1990. Washington, DC: U.S. Department of Transportation, Federal Highway Administration.
Ross, Stephen L. 1998. Racial differences in residential and job mobility: Evidence concerning the spatial mismatch hypothesis. Journal of Urban Economics 43:112-35. Rubin, Tom. 1994. A look at the Los Angeles Metropolitan Transportation Authority. Los Angeles: Metropolitan Transportation Authority. Rutherford, Brent, and Gerda R. Wekerle. 1988. Captive rider, captive labor: Spatial constraints on women’s employment. Urban Geography 9 (2): 116-37.
Sanchez, Thomas W. 1999. The connection between public transit and employment: The cases of Portland and Atlanta. Journal of the American Planning Association 65 (3): 284-96. Schulz, Dorothy, and Susan Gilbert. 1996. Women and transit security: A new look at an old issue. In Women’s travel issues. Proceedings from the second national conference, October 1996, edited by Sandra Rosenbloom and Shirl Byron. Washington, DC: U.S. Department of Transportation, Federal Highway Administration. Scobles, Harry. 1968. The McCone Commission and social science. Phylon 29 (2): 167-81.
Shen, Qing. 1998. Location characteristics of inner-city neighborhoods and employment accessibility of low-wage workers. Environment and Planning B 25 (3): 345-65. . 2000. Spatial and social dimensions of commuting. Journal of the American Planning Association 66 (1): 68-82. . 2001. A spatial analysis of job openings and access in a U.S. metropolitan area. Journal of the American Planning Association 67 (1): 53-68. Shoup, Donald. Forthcoming. The high cost of free parking. Chicago: Planner’s Press. Simpson, Pat. 2000. Skills shifts and black male joblessness in major urban labor markets over the 1980’s. Social Science Research 29 (3): 327-55. Simpson, Wayne. 1992. Urban structure and the labour market. Worker mobility, commuting, and underemployment in cities. Oxford: Oxford University Press. Singell, Larry D., and Jane H. Lilleydahl. 1986. An empirical analysis of the commute to work patterns of males and females in twoearner households. Urban Studies 23 (2): 119-29. Stoll, Michael A. 1999. Spatial mismatch, discrimination, and male youth employment in the Washington, DC area: Implications for residential mobility policies. Journal of Policy Analysis and Management 18 (1): 77-98. Sugrue, Thomas. 1996. Origins of the urban crisis. Princeton, NJ: Princeton University Press. Taylor, Brian, and Paul Ong. 1995. Spatial mismatch or automobile mismatch? An examination of race, residence, and commuting in U.S. metropolitan areas. Urban Studies 32 (9): 1453-73. Thompson, Gregory L. 1997. Transit accessibility and labor force participation rate of at-risk groups: Dade County. Tallahassee: Florida Institute for Marketing Alternative Transportation, Florida State University. Turner, Susan. 1997. Barriers to a better break: Employer discrimination and spatial mismatch in metropolitan Detroit. Journal of Urban Affairs 19 (2): 123-41. Turner, Tracy, and Debbie Niemeier. 1997. Travel to work and household responsibility: New evidence. Transportation 24:397-419.
U.S. Congress. 1996. Personal Responsibility and Work Opportunity Reconciliation Act. Public Law 104-93. . 1998. Transportation Equity Act for the 21st Century. Public Law 105-178, 105th Congress.
U.S. Department of Health and Human Services. 2002. Characteristics and financial circumstances of TANF recipients, October 1999-September 2000, Temporary Assistance for Needy Families (TANF) Program. Washington, DC: Administration for Children and Families, Office of Planning, Research and Evaluation.
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
Beyond the Spatial Mismatch U.S. Department of Housing and Urban Development. 2000. The state of the cities 2000. Megaforces shaping the future of the nation’s cities. Washington, DC: HUD Office of Policy Development and Research. U.S. Department of Labor. 2002. Consumer expenditure survey. http:// www.bls.gov/cex/home.htm.
U.S. General Accounting Office. 1998. Welfare reform: Transportation’s role in moving from welfare to work. Final Report to the Chairman, Committee on the Budget, House of Representatives. Washington, DC: Government Printing Office. Urban Institute. 2002. Welfare rules database. Assessing the New Federalism. Washington, DC: Author. http://www.urban.org/Content/ Research/NewFederalism/AboutANF/AboutANF.htm. Viscusi, W. K. 1992. Fatal tradeoffs: Public and private responsibilities for risk. New York: Oxford University Press.
Wachs, Martin. 1992. Men, women and urban travel: The persistence of separate spheres. In The car and the city, edited by Martin Wachs and Margaret Crawford. Ann Arbor: University of Michigan Press. . 1996. The evolution of transportation policy in Los Angeles: Images of past policies and future prospects. In The city: Los Angeles and urban theory at the end of the twentieth century, edited by Allen Scott and Edward W. Soja. Berkeley: University of California Press. Wachs, Martin, and Brian Taylor. 1998. Can transportation strategies help meet the welfare challenge? Journal of the American Planning Association 64 (1): 15-9.
205
Waller, Margy, and Mark Allan Hughes. 1999. Working far from home: Transportation and welfare reform in the ten big states. Washington, DC: Progressive Policy Institute and Public/Private Ventures. Wang, F. 2003. Job proximity and accessibility for workers of various wage groups. Urban Geography 24 (3): 253-71. Webber, Melvin. 1992. The joys of automobility. In The car and the city, edited by Martin Wachs and Margaret Crawford. Ann Arbor: University of Michigan Press.
Weinstein, Richard S. 1996. The first American city. In The city: Los Angeles and urban theory at the end of the twentieth century, edited by Allen Scott and Edward W. Soja. Berkeley: University of California Press. Wohl, Martin. 1968. Users of transportation and their income circumstances. Paper presented at the American Academy of Arts and Sciences Conference on transportation and poverty, June 7, 1968, Brookline, MA. Wyly, Elvin. 1996. Race, gender, and spatial segmentation in the Twin Cities. Professional Geographer 48 (4): 431-44.
. 1998. Containment and mismatch: Gender differences in commuting in metropolitan labor markets. Urban Geography 19:395-430. Zogby, J., and P. J. Malin. 2001. Children’s health care and transportation access. New York: Children’s Health Fund.
Downloaded from http://jpl.sagepub.com at UCLA on September 15, 2008
