Some ideas just won’t die. Enterprise zones, the perennial Republican urban policy proposal championed by former Housing Secretary Jack Kemp, have reemerged as a Democratic proposal spearheaded by Kemp’s successor, Henry Cisneros. The idea sounds like simple common sense: use tax breaks and other incentives to encourage private employers to locate in poor areas, bringing job opportunities to local residents. Our study of employment patterns in an impoverished Brooklyn neighborhood, however, suggests that enterprise zones are unlikely to be an effective strategy. The primary reason for ghetto unemployment is not the lack of nearby jobs but the absence of social networks that provide entry into the job market.

The idea of enterprise zones has something to appeal to nearly everyone. Conservatives like them because they promise reduced government and a private-sector answer to poverty. Liberals are attracted by the promise of “empowerment” for the local community. And enterprise zones have a strong nonideological selling point: they are cheap. An enterprise zone program entails very modest initial expenditures; the costs are paid primarily in uncollected taxes on as yet unmade profits.

A wide variety of enterprise zone programs have been implemented since the original Kemp-Garcia proposals of the mid-1970s. Since 1981, 35 states and the District of Columbia have introduced some version of an enterprise zone program, with modest results at best. All such programs offer a package of government incentives to encourage businesses to locate or expand in high-poverty areas. The primary benefits are tax breaks and abatements, although in some versions regulations are relaxed and additional tax credits are available for employee training. Early proposals had called for radical reductions in the role of government, including reduced environmental regulations and a lower minimum wage, but these provisions were usually dropped as legislation was drafted.

The Clinton administration’s program, dubbed the Economic Empowerment Act of 1993, was approved in August as part of the compromise budget plan. It will establish 9 “empowerment zones” and 95 “enterprise communities” in both urban and rural areas. In all of these locales, firms would be eligible for tax incentives: exemptions for private activity bonds and accelerated depreciation for investments within the zone. Employers in the empowerment zones would also be eligible for employment and training tax credits and some small incentives to hire local residents. And, unlike past Republican proposals, the administration’s plan provides that both types of targeted areas would be eligible for a host of proposed and already active community-based education, housing, and anticrime programs. Such programs, however, are brought together in a rather disorganized, ad hoc way.

The Mismatch Theory

The intellectual basis of the enterprise zone concept is what is known as the “spatial mismatch” theory of inner-city poverty. In 1968, economist John Kain first suggested that poor blacks lacked access to jobs, particularly manufacturing jobs, in significant part because those jobs were moving to the suburbs, far from the inner-city areas where most poor blacks lived. The idea was elaborated by some, most notably sociologist John Kasarda, and questioned by others. The most exhaustive review of the academic literature on the issue, conducted by Christopher Jencks and Susan Mayer, found the evidence inconclusive.

It is certainly true that in many cities joblessness among ghetto residents grew at the same time as the number of manufacturing jobs declined and those that remained were shifting to the suburbs. It is probably also true that some policy decisions have made it more difficult for the poor to find work. The lack of mass transit in suburban areas, for example, has often made outer-city and suburban jobs far less accessible for ghetto residents. But if an enterprise zone program succeeded in bringing jobs into inner-city neighborhoods, would local residents have access to those jobs? Our study of Red Hook, Brooklyn, a community that already has the mix of low-income residents and industrial employers that enterprise zones aim to create, suggests that the answer is no.

Red Hook is a neighborhood of approximately 13,500 residents, located on a peninsula directly across New York Harbor from the Statue of Liberty. While the name “Red Hook” once referred to a larger section of South Brooklyn, today the name is used almost exclusively to describe the area cut off from the rest of the borough by a confluence of highways and the toll plaza at the entrance to the Brooklyn Battery Tunnel. It is a mixed-use area; the Red Hook Houses, one of the city’s oldest and largest public housing projects, sit close to a variety of small industrial firms. In contrast to many Brooklyn neighborhoods with similar concentrations of poverty and public housing, the area is also quite near middle-class areas

Neighborhood residents, however, think of themselves as isolated. Highways cut the peninsula off from the adjacent neighborhoods, making passage to Red Hook difficult on foot or by car. The area is underserved by mass transportation; the most common route to the subway requires crossing a six-lane highway, passing underneath an elevated expressway, and walking three blocks through a predominantly Italian-American neighborhood that many members of Red Hook’s black and Hispanic majority view, not without reason, as hostile territory.

But Red Hook’s isolation is far more than a matter of infrastructure. Residents report a pervasive sense of having been cut off—economically, politically, and culturally—from the rest of the city. They often speak of themselves as being “the forgotten people,” at “the end of the line.” Indeed, many New Yorkers know Red Hook only as the neighborhood where elementary school principal Patrick Daly was fatally shot in December 1992.

In 1989, 46.3 percent of Red Hook residents lived below the poverty level—far more than the 20 percent necessary for a community to be designated an official enterprise zone in New York State. (Red Hook is not, however, an enterprise zone.) Three out of four local residents live in the sprawling, forty-acre Red Hook Houses. The rest live in “the back,” a cluster of private housing and small tenements occupied by the remnant of the old working-class white population, as well as by African-Americans and Puerto Ricans. The 1990 census found that 48 percent of area households received public assistance and 21.5 percent of Red Hook workers were unemployed. In 1989, 58 percent of all children born in the area, and 78 percent in the project, were born to single mothers.

Red Hook’s mix of industry and impoverishment seems to confound the spatial mismatch theory. Numerous manufacturing plants and jobs exist in a neighborhood with high levels of poverty, unemployment, and crime. Yet while poverty is worse for public housing residents than for those outside the project, and worse for blacks and Hispanics than for whites, lack of access to jobs is perceived as a problem by all groups in the neighborhood.

It could be true, as Lawrence Mead’s work would suggest, that a substantial number of local residents are simply not interested in low-wage work. On the other hand, when a discount supermarket chain announced its intention to open a store in the area in early 1991, it was besieged by more than three hundred applicants—two-thirds of them Red Hook residents—for minimum-wage positions. We surveyed these applicants and found that housing project residents, including those with considerable work experience, were more connected to service-sector jobs outside the area than to industry. By contrast, applicants from “the back” were far more acquainted with blue-collar work. But both groups of Red Hook applicants tended to have highly unstable job histories. This seemed to result from both a propensity to leave jobs (frequently over disagreements with bosses or coworkers) and the chronic instability of the low-wage job market. Many applicants had experience with temporary work, day labor, layoffs, and job loss due to company closings. We also found a surprising lack of qualifications, even for many unskilled jobs. In an area that employs many truck drivers, for example, only 9 percent of the three hundred job applicants had driver’s licenses.

In order to find out more about why Red Hook residents don’t get Red Hook jobs, we conducted a survey of local employers. Using a list compiled by the South Brooklyn Local Development Corporation (LDC), we generated a sample that included all 19 of the large employers (those with forty or more employees) headquartered in the community and 15 of the smaller ones. Thirty-two of these 34 employers were private-sector firms; 13 of those (or 41 percent) employed at least one Red Hook resident. Most of those local employees, however, were whites from outside the project. Only three of the 34 employers reported that the majority of their workers were Red Hook residents.

When we analyzed our data further, we realized that we had underrepresented two key sources of employment for people in the project. Only two not-for-profit or governmental agencies showed up in our initial sample, because others were headquartered elsewhere, so we sampled three additional work sites: two day-care agencies and a public school. All five had employed Red Hook residents in the past, and four did at the time of the study. In all four cases, the employees included project residents. Each of these agencies uses highly systematic hiring procedures: all had a formal application process and four of the five required state or other licensing procedures, educational credentials, fingerprints, and criminal background checks.

Because the LDC list had also missed most small retailers, we conducted an additional small survey of 23 retail establishments facing the project. Only 7 of them employed Red Hook residents, and only one, a locally owned barbershop, regularly employed local blacks.

Our interviews with employers revealed two general reasons why they tend not to hire locals. There is substantial “positive discrimination” in favor of some groups of nonresidents, as well as considerable “negative discrimination” against residents. Though racial and ethnic stereotypes play some part in these patterns, they are far too complex to be explained exclusively in such terms.

Positive Discrimination

The patterns of positive discrimination take a variety of forms. For one, the area’s best jobs, highly paid longshoring and shipbuilding positions, are monopolized by powerful unions to such a degree that employers actually have little choice in whom they hire. And there are many more union members than jobs; indeed, no new longshoreman has been hired in the port of New York since 1972. In such cases residential location does play a role in who gets jobs—but it is residential location in the distant past rather than the present. For example, a large contractor hires carpenters from a union hiring hall near Red Hook. Many members of this father-and-son local live in the distant suburbs. Yet they have first crack at Red Hook jobs, in part because their fathers, uncles, or grandfathers once lived near the waterfront.

Most local jobs are in lower-paying industries. Here, too, there are many examples of positive discrimination in favor of nonlocal job-seekers. The primary quality employers look for in an unskilled position is reliability, and most report that the best way to find reliable employees is by personal referral. “We hire almost entirely by word of mouth,” said the owner of a small recycling center. “I have a couple of guys who brought in almost everybody, all the unskilled anyway. They bring in their friend, their cousin, their uncle, whatever. I have had maybe five guys come in off the street, and only two or three have worked out even for a short time.”

Hiring “off the street” is rare even in the lowest-paying jobs. A commercial baker noted:

We’ll give anyone who walks in off the street an application. The likelihood of their being hired is slim, because we don’t know them.... We don’t tend to just take people off the street, because I’ve had a lot of bad experiences. That’s true of all ethnic groups, including white Americans.

The reliance on referrals often amounts to a reliance on ethnic networks. This was no surprise in small retail businesses, which are usually family-run. Korean, Chinese, Arab, and Dominican retailers all hired exclusively from their own ethnic group. But larger businesses also often hire through ethnic networks, though employers are sometimes only peripherally aware of how the process works. One large contractor reported that more than half his workers are immigrants from Spain who commute into Red Hook each day from Jersey City and Newark. “They all know each other over there, I guess, and when they hear we have an opening, somebody just shows up.”

Sometimes ethnicity is an explicit basis for positive discrimination. “There’s a general perception amongst employers that Irish, Polish, and Mexican immigrants are the latest desirable employees,” said a warehouse manager. “There’s a friend of mine [who says] that all the Mexican guys he’s come in contact with are incredibly good workers. You hear that enough times and then if a Mexican guy came here for work I’d probably hire him.”

There is considerable scholarly literature on the role of social networks, including ethnic networks, in job placement. Networks serve at least three functions in the labor market. First, they provide specific information about the availability of job openings and how to pursue them. Second, they provide role models of successful employees. Third, and perhaps most important, they can provide direct sponsorship. A current employee may directly vouch for a prospective worker; if the new employee works out, the credibility of the sponsor is enhanced.

Social networks, not residential geography, tend to determine employment decisions. Social networks may be centered on a particular neighborhood, but they are often not, particularly in the sort of high-crime ghettos that discourage the formation of strong local ties. In areas like Red Hook, residents tend not to he embedded in social networks that connect them to private-sector jobs, even jobs in their own neighborhood. There is little reason to suppose that promoting local businesses in similar areas would change this pattern.

Negative Discrimination

Our study also found considerable discrimination against local residents. Although most employers were anxious to make clear that they do not practice racial discrimination, few showed any hesitation about describing “locality” discrimination against “ghetto” people in general and residents of the Red Hook project in particular. In some cases employers preferred not to hire people who lived too close to work for fear that the complicated lives of the poor would spill out into the workplace:

We have drawn people from the Red Hook area for a number of jobs and we haven’t had particularly good experiences. It’s interesting that the thing that seems to be the problem is that people feel almost too close to home. They tend to want to go home for lunch and stay there.... They tend to have a fair number of visitors. Sometimes children, spouses, boyfriends, and girlfriends come in to visit them.... Sometimes that can be very disruptive.

More often, however, the focus of locality discrimination was the fear of crime. An employer with a large facility on the waterfront pointed with pride to his all-black team of security guards, West Indian immigrants who commute from central Brooklyn and Queens, but scoffed at the idea of hiring local African-Americans: “What, the bums hanging around outside? You want me to hire the guys who are trying to rob me?”

A recycled responded in this way to our attempt at subtly raising the question of hiring “locals”:

Look. I know what you’re driving at. Let me save you some time. Do I hire people from the projects? I have had maybe five people come in here over the years. I gave one or two a shot. And they didn’t last a day. Mostly they gave me the impression that they were just looking around so they could come back for their “real job” [burglary], at night.

Concern about crime in the area is by no means unreasonable. Most of the Red Hook employers have experienced pilferage, and many have had employees mugged. It would be unfair to dismiss these concerns as simple euphemisms for racism.

Many private-sector employers see the neighborhood as a hostile environment to be kept at bay. Industrial firms in the area frequently do not post signs, and it often takes considerable persistence to find one’s way into the secure buildings. Steel-reinforced doors and blacked-out windows are common. Negative feelings about the area are often transferred to the residents:

We see junkies, we see prostitutes standing outside, and I wouldn’t hire someone who lived in that. The projects have a very strong stigma attached. It’s not just a matter of race. I don’t discriminate. I’d be very upset if one of my people did. But naturally you distrust somebody who comes from a place like that. It’s unfortunate, because there are probably lots of good people who live there.

Even some employers who do employ ghetto residents discriminate against those from Red Hook. One moving company supervisor, who primarily employs black men as day laborers, explained that he prefers to hire men from outside the area. Those from the neighborhood, he asserted, would come under pressure from friends to help them steal, while men with similar backgrounds but from distant neighborhoods would be under less such pressure.

Who Does Hire Locals?

As we have noted, there was one significant exception to the pattern of employers favoring nonlocal workers. Four out of the five not-for-profit and public-sector employers we surveyed did hire large numbers of people from the project. In contrast to the private firms, all of these employers perceived having employees who were “part of the community” as a significant benefit. In part this is because neighborhood residents are their clients, so good community relations is important to their “business.” Yet it also seems likely that local residents, particularly those in the project, do better in public-sector employment because that is where their connections—their social networks—are. Unfortunately, public-sector opportunities for poorly educated applicants are severely limited; even the least-skilled positions in this sector, such as for day-care workers, require complicated licensing procedures.

During the time we were conducting our survey, another exception to the pattern of nonlocal hiring occurred. The South Brooklyn Local Development Corporation began a referral service aimed at placing residents in local jobs. At first, success was limited: the corporation primarily placed young people in very low-wage jobs and placed a few women in clerical positions, mostly outside the neighborhood. But it repeatedly tried to place men in industrial and construction jobs. Eventually a few construction contractors, needing minority workers to fulfill the affirmative action requirements of city contracts, agreed to take on a few laborers from the LDC pool. The contractors later reported that the workers had “basically worked out.” They did not, however, start recruiting in the Red Hook Houses. Rather, they saw this as a happy exception, and they credited the favorable experience, probably with some justification, to the considerable screening done by the LDC. The employers were willing to “take a chance” on people they would not normally have hired because they trusted the judgment of the LDC director with whom they had an ongoing relationship—another example of the importance of social networks.

Why Jobs Aren’t Enough

Our study of Red Hook demonstrates that the presence of low-skill jobs in poor neighborhoods is no guarantee that the poor will have access to those jobs. In a loose labor market it is easier and more efficient for employers to find new employees through informal referral systems, particularly when local residents are perceived as unreliable and prone to crime. These systems typically end up recreating the demographics of the current labor force.

Enterprise zone programs, by focusing on the location of jobs rather than the formation of social networks, fail to address the central problem. The concept confuses physical proximity—a matter of little importance to most employers, and in some cases even a detriment—with “social embeddedness,” a quick index of reliability. The idea of enterprise zones reflects a basic misunderstanding of the relationship between living space and working space in dense cities like New York. We do not, after all, expect middle-class New Yorkers to work within walking distance of their homes: why should we expect the poor to do so?

How, then, can we connect ghetto residents with private-sector jobs? While we have no easy answer, it is clear that policymakers should be thinking in terms of social embeddedness rather than geographic location. One promising approach—exemplified by the limited success of the South Brooklyn LDC and more systematically by America Works, the private-sector employment agency for welfare recipients—is to create proxy networks that can inform, socialize, and vouch for employees in much the way that informal social and ethnic networks do now. Such efforts could go a long way in giving poor people access to jobs—even if they have to take the subway to get to them.


City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next