Paying Employers to Hire Local Workers in Distressed Places

Abstract

Many tax incentive zones encourage employers to hire workers in distressed places. A well-developed literature on hiring low-skilled workers has much to say about why such zones have resulted in little employment for nearby residents of distressed areas but has not informed evaluation or policy for zones. Using the Detroit Empowerment Zone experience as a case, this article finds that referrals, usually from employees, determined hiring, resulting in industrial districts with few local workers and retail districts with many more, with workplaces segregated by race and ethnicity. Although skill requirements and screening approaches should not have excluded many qualified local workers, employers in industrial areas had negative stereotypes of workers from nearby neighborhoods. Trusted intermediaries such as community-based organizations may enable tax incentive zones to produce jobs for local workers by breaking down stereotypes and inserting a new “mouth” in the word-of-mouth hiring practices.

Keywords

tax incentives empowerment zones entry-level hiring local jobs inner-city jobs

The idea of providing tax breaks for businesses to locate and expand in depressed areas and hire local workers has had remarkable staying power. As of 2008, 40 states and Washington, D.C. had programs, often called enterprise zones, with this aim (Ham, Swenson, İmrohoroğlu, & Song, 2011; Peters & Fisher, 2002; Wilder & Rubin, 1996). In 1993, the Clinton Administration’s Empowerment Zone (EZ) program offered tax breaks to businesses in specific economically distressed districts in chosen cities and rural areas to encourage economic development and the employment of residents of these zones. The Bush Administration continued this program, although without additional grant funding. The empowerment zones (EZs) also provided many more programs to address poverty and unemployment in neighborhoods.

Research has shown that effects of such incentives on business decisions rarely result in the employment of low-income workers from distressed neighborhoods. Almost entirely missing from this body of research, however, is investigation of why these programs have few effects. This article addresses the question of whether and why incentives affect employment of workers from distressed areas through examination of the literature on hiring of entry-level workers and through a look at an example of such a program, the business tax incentives aspects of the Detroit EZ. The literature that evaluates business incentives rarely reflects the research that explains how employers hire workers for entry-level positions that require few skills and little education, the kind of jobs many residents of distressed areas would seek. Bringing these areas of research together helps explain incentives’ effects and suggests a few interventions that could make such programs more successful. Assuring such programs do lead to the hiring of workers from nearby distressed areas matters, for despite loss of a large percentage of central-city jobs since the 1940s, many tax incentive zones are located in central cities. No spatial mismatch interferes with local workers’ gaining jobs in the zones, unlike the more numerous jobs in the suburbs.

The Effects of Enterprise and Empowerment Zone Tax Incentives on Economic Opportunity

Do tax incentives succeed in creating economic opportunity for low-income people in distressed areas? Creating economic opportunity is defined here as generating more employment in the targeted area and connecting more residents of distressed areas to jobs than would have occurred otherwise. This requires changes in businesses’ behavior that lead to hiring workers they would not otherwise have hired.

By the early 2000s, thousands of state enterprise zones existed. Evaluations of the effects of enterprise zones on economic activity in those locations are extensive and generally show small or no impact on employment and investment with the use of stronger research design (e.g., Boarnet & Bogart, 1996; Bondonio & Engberg, 2000). The findings vary considerably with the methods of evaluation and the details of the incentive programs, which are more generous in some places than others. In an overview of studies, Peters and Fisher (2004) concluded that enterprise zones generally failed to cause economic growth. In an earlier review, however, Wilder and Rubin (1996) found that many studies showed enterprise zones created some new jobs, albeit often with flaws in research design, as Boarnet (2001) pointed out. (For additional overviews, see Bartik, 2004; Papke, 1993; Peters & Fisher, 2002.)

In contrast, little research exists on whether residents of enterprise zones receive jobs. Many states did not tie tax incentives strictly to employment of zone residents, although zones were designated in areas of high poverty and unemployment; zone residents did not necessarily benefit from jobs. Peters and Fisher (2002), in a wide-ranging assessment of a large number of enterprise zones, concluded that no evidence existed that enterprise zones employed more disadvantaged workers or more workers from neighborhoods with little access to suburban jobs. In summarizing their own and others’ research, they concluded that a small fraction of “induced jobs (if there are any) will actually go to residents of that area; and some of those newly-employed residents may actually be the poor or unemployed” the programs aimed to help (Peters & Fisher, 2004, p. 34). Papke (1993, p. 62) concluded that state enterprise zones had not improved the economic status of zone residents. Other studies have shown no improvements in income or employment outcomes for residents of zones (Elvery, 2009; Greenbaum & Engberg, 2000). Wilder and Rubin’s (1996) review concluded that enterprise zones had an uneven record of employing residents of the zones with 2% to 90% of new jobs going to residents. Because of research design problems in the studies they reviewed, one cannot determine whether these jobs would have existed otherwise. Bostic and Prohofsky (2006), in a rare assessment of effects on workers, found that employees hired under California’s enterprise zone program, especially those with very low income initially, realized higher wages and higher adjusted gross income than control groups, but the authors did not consider whether the new employees were residents of the enterprise zone or of nearby neighborhoods.

Far fewer evaluations exist of the effects of the federal EZs and enterprise communities (ECs), and only some of these assess effects on economic opportunity. The most extensive study of the Clinton Administration’s EZs designated in 1994 found that EZs’ employment grew 9% more than comparison areas between 1995 and 2000, but this varied greatly among the different zones. For instance, in Detroit the EZ businesses’ employment grew 27% more than those in comparison areas, 11% more than those in contiguous areas. The number of residents employed in EZ businesses doubled from 1997 to 2000, but their share of the jobs in these businesses stayed the same as the economy grew. The general economic upturn made linking growth to the EZ incentives and programs difficult. The growth of employment did not clearly connect to EZ activities and could have been due to other factors (Hebert, Vidal, Mills, James, & Gruenstein, 2001).

Several studies looked at employers’ use of the incentives the EZs offered, a precondition for the credits having an effect on the businesses. According to Hebert et al. (2001), 11% of businesses in Detroit reported using the EZ employment credits; 4% used the deduction for the cost of depreciable property. More than half of the businesses that used the tax credits said they were not important in their hiring or investment decisions. Two thirds of EZ businesses reported no benefits from their location in the EZ. A General Accountability Office (GAO) survey of urban EZ businesses showed that 42% of businesses with 50 or more employees and 6% of smaller businesses had used the employment credit. The majority reported that the credit was “somewhat important” in their decisions about hiring EZ residents. About 9% of large businesses used the increased deduction of depreciable expenses; about 4% of small businesses did so. Very few businesses used the tax-exempt bonds or the other credits available to businesses in distressed communities (U.S. GAO, 1999). Although the study reported the use of incentives, it did not demonstrate that the incentives caused new economic opportunity for EZ residents because businesses may have been doing what they would have done without the subsidies. In contrast to the GAO study, an analysis of the use of the EZ/EC tax incentives in 1996 through tax returns data, showed that the loss of federal revenue due to use of tax incentives was much less than expected, only 7% of the forecast amount (Brashares, 2000). A later GAO report (2006) concluded that data had not been collected to enable attribution of changes in the EZs to the program activities.

Research has also shown the variation in results among different zones on different measures of economic improvement in EZ neighborhoods. The studies do not specifically look at the effects of tax incentives. Each city with an EZ planned its particular approach to addressing economic opportunity and other goals. Numerous other programs—training, social service, and transportation programs, for instance—could have improved the economic status of EZ residents and neighborhoods. Rich and Stoker (2010), examining the differences for each EZ city through matched pairs of census tracts in and outside the EZs, concluded that results were uneven. For instance, EZ tracts in Baltimore and Philadelphia showed improvement compared with comparison tracts in jobs, poverty, unemployment, housing investment, and business investment. Detroit EZ tracts, in contrast, showed improvements in jobs, poverty, and business investment but worsened conditions in housing investment and unemployment. Unlike Rich and Stoker, Oakley and Tsao (2006), found little improvement in economic conditions compared with similar nearby census tracts in four of the EZs, including Detroit (see also Oakley & Tsao, 2007).

Evaluations of the overall effects of the EZs, not distinguishing among the different cities, on economic conditions in EZ neighborhoods also vary in the effects they show. All grapple with the difficult research design issues in separating the effects of the EZs from other influences, and they come to different conclusions in using different kinds of comparison groups and estimating different equations. As Rich and Stoker (2010) point out, these national studies hide differences in the effects of the varied approaches that different cities implemented, and these differences are likely important in understanding what works and in suggesting stronger policies. A. Hanson (2009) compared EZ tracts with applicant tracts not selected and also accounted for the fact that the Department of Housing and Urban Development selected EZs most likely to have results; he concluded that the EZs had small if any effects on residents’ employment and poverty. Busso, Gregory, and Kline (2010), also using applicant tracts in cities not designated as EZs as their comparisons, concluded in contrast that EZ neighborhoods experienced substantial increases in total employment and that the hourly wages paid to EZ residents working in the EZ also increased considerably. Ham et al. (2011), looking at all EZ census tracts compared with nearby non-EZ tracts, found improvements in the labor market in lower unemployment, lower poverty, higher average wage and salary income, and more employment.

Studies of both enterprise zones and EZs have theorized the reasons for expected effects of incentives on businesses’ employment, economic growth, and local employment and income, and some have suggested approaches that might make them more effective (for instance, Kolko & Neumark, 2010; Wilder & Rubin, 1996). Missing from most, however, is follow-up to statistical findings to understand the reasons for results. Missing from all is consideration of how systems of hiring contribute to explaining the effects on employment of zone residents, although research on hiring of low-skilled workers is a well-developed field. A few scholars have suggested why such zones might not result in hiring of zone residents based on understanding of low-income, inner-city labor markets, but they have not linked their hypotheses to research on tax incentive zones (Ladd, 1994; Rosenberg & Kasinitz, 1993; Vidal, 1995).

This article uses evidence from a study of the hiring practices of employers in the Detroit EZ and draws on research on the hiring of entry-level workers to explain the results of tax incentives for hiring workers in distressed areas, and thus aims to bring these different bodies of research into conversation with each other. The next section of this article establishes, first, that the Detroit EZ had only weak effects on business decisions to expand and locate in the EZ and to hire EZ residents in five case study areas. Unlike many studies of the effects of tax incentive zones, this study does not look at changes in the economic conditions in the EZ neighborhoods but instead looks at changes in business behavior that must occur for the EZ incentives to have eventual effects on residents and on neighborhoods.

Detroit Empowerment Zone

The EZs that were designated in 1994 offered three principal tax incentives through the first years of their operation. These sought to link assistance to businesses to local workers’ employment more tightly than most state enterprise zones did. The EZ Employment Credit allowed an employer to claim a tax credit of 20% of the first $15,000 of a “qualified” employee’s earnings. A qualified employee was one who performed substantially all his work in the EZ in the employer’s trade or business and had his or her main residence in the EZ. A “qualified” EZ business could deduct up to $37,500 of the cost of depreciable property rather than using the maximum standard deduction and rather than recovering the cost through depreciation deductions; this constituted an increase of $20,000 in the amount of the allowed deduction. States and municipalities could issue tax-exempt bonds on behalf of qualified businesses up to $3,000,000 in any one EZ. To qualify for the deduction of the cost of depreciable property and the bond financing, a business had to meet restrictive criteria. At least 35% of the business’s employees had to live in the EZ, at least 80% of total gross income had to come from activity in the EZ, “substantially” all of the use of tangible property had to be in the EZ, and “substantially” all services had to be performed in the EZ—thus aiming to encourage hiring of EZ residents through a capital subsidy. These restrictions were loosened with the tax year beginning after August 1997 so that more employers would qualify, and the meaning of the requirements was clarified (Budget Reconciliation Act, 1993; Taxpayer Relief Act, 1997; U.S. Department of the Treasury, 1993, 1999).

To evaluate the Detroit EZ tax incentives’ effects on employers’ decisions about hiring, investment, and location, this study looked at five case-study locations. The two industrial districts and three retail areas were chosen because they straddled the EZ boundary so the areas outside the EZ could serve as comparison areas where all other location conditions were the same. Furthermore, results were more likely in these districts than in many parts of the EZ because they already had concentrations of businesses, and nonprofit, community-based organizations were working to strengthen the areas and to publicize the EZ incentives. Parts of both industrial districts were designated as “Model Urban Industrial Areas,” a Detroit EZ program aimed at making existing industrial areas more attractive business locations (Detroit Planning and Development Department, 1994). The districts were adjacent to residential areas of the EZ and so had access to many EZ residents as potential employees. On the other hand, unlike several other areas, no major business expansions were already underway as the EZ began, a situation that accounted for many of the overall EZ job gains.

In face-to-face interviews in the fourth year of the EZ’s implementation, employers were asked about characteristics of their operations and about the reasons for their location, expansion, investment, and hiring decisions—the decisions that the EZ incentives and programs sought to influence (see the appendix).¹ Although the interviews occurred 14 years ago, changes since then may principally have put more barriers in the way of entry-level job seekers, those with little education and few skills. Employers now frequently advertise jobs on the Internet and expect electronic submission of cover letters and resumes, for instance. Entry-level applicants from low-income neighborhoods may have less familiarity with and access to computers, and their smartphones may not enable them to upload cover letters and resumes. The severe recession that began in 2007 reduced the number of jobs and increased the number of unemployed, increasing competition for all jobs.

Effects of Incentives

How did the tax incentives affect employers in these areas of Detroit? For the incentives to have an effect, businesses needed to use them. In the industrial districts, nearly 11% of EZ employers used the employment credit in the first 3 years of the EZ; none used the increased deduction of the cost of depreciable property or the bond financing. In the retail areas in the EZ, one employer used the increased deduction for the cost of depreciable property and the One-Stop Capital Shop (an EZ program to assist small businesses). No one used the employment credit. These results are comparable with those of Hebert et al. (2001), discussed above, for the first 5 years of EZ implementation.

When asked about location, expansion, and investment decisions, none of the employers in the EZ industrial districts mentioned that the EZ affected their decisions. None volunteered that the EZ had an effect on hiring decisions. When asked directly if the EZ had affected hiring, expansion, investment, land purchase, or location decisions, every employer said it had not.

No manager or owner in the retail districts cited the EZ as a reason for decisions about new investments, purchase of land or buildings, or hiring. Only one employer (an organization that was itself one of the EZ programs for strengthening neighborhoods) stated that the organization sought a location in the EZ.

Table 1 compares the behavior of employers in the EZ and outside it during the first 3½ years of the EZ’s implementation. If the EZ had a strong effect on employers’ behavior, the indicators should be higher for employers in the EZ rather than outside it, except that the share of employers who decreased employment should be lower in the EZ. Instead the numbers show employers outside the EZ often exhibited more of the behaviors the tax incentives were supposed to influence. The most notable difference in industrial areas was that employers in the EZ had increased employment by 18%, whereas employers outside the EZ had lost a small percentage of employment. Retail district employers, on the other hand, had increased employment much more outside the EZ than within it. These numbers present a more positive picture than the actual because they do not include employers that closed or moved between 1995 and 1998.

Table 1.

Comparison of Employers’ Activities Inside and Outside the Empowerment Zone (EZ), 1995-1998.

	Inside EZ		Outside EZ
	Number of employers	Percentage of employers	Number of employers	Percentage of employers
Industrial districts
Located in the area	2	8.3	3	8.6
Made new investments	17	60.7	24	72.7
Acquired land/buildings	7	24.1	7	20.0
Increased employment	9	33.3	14	40.0
Decreased employment	5	18.5	8	22.9
Retail districts
Located in the area	15	35.7	24	35.3
Made new investments	25	55.6	44	63.8
Acquired land/buildings	10	22.2	10	14.5
Increased employment	17	37.8	32	46.4
Decreased employment	7	15.6	7	10.1

	Number of Jobs	Percentage of 1995 Jobs	Number of Jobs	Percentage of 1995 Jobs

Industrial districts
Change in employment	144	18.3	-31	−1.5
Retail districts
Change in employment	48.5	13.4	139	31.5

Source. Interviews with employers.

Some reasons for the failure to use tax incentives were apparent. Interviews in the industrial districts showed considerable misunderstanding about the programs and lack of information about eligibility. Furthermore, half of the employers in industrial areas said that the incentives had no value to them, that the specifics of the incentives did not help them, or that they did not need them. At most, 10% or 11% of employers qualified for incentives for depreciable property or bond financing based on their employment of EZ residents and for-profit status. Employers in the EZ were usually eligible for the employment credit, but less than one third of these stated that they employed any EZ residents. The work-opportunity tax credit that encouraged employment of certain types of workers who were often hard to employ could have been attractive to some employers instead of the EZ employment credit because the work-opportunity credit was not lost when a worker moved out of the EZ (U.S. Department of the Treasury,1999). No employers in the EZ, however, had used the work-opportunity tax credit.

In retail areas employers also lacked information or misunderstood the EZ benefits. Many could have used the benefits but had not. Others, however, may not have been eligible. Some of the businesses operated with very low profits or were losing money and had no taxable income. For many businesses in the retail districts, the tax-exempt bond financing was not useful because the interest rate on bond financing reflects the riskiness of the business that backs the issue and bonds for small, independent retail stores such as many in the retail districts would be high risk. Furthermore, small business owners may have been reluctant to take advantage of tax breaks that could invite the Internal Revenue Service to investigate their books or that could attract the attention of immigration authorities. Interviewers noticed that many sales did not go through a cash register. Several employers stated that they did not hire all employees “on the books”; others asserted that some of the people working in their establishments were not employees. Numerous small businesses employed only family members.

Employers in the industrial districts also expressed reasons they would not expand or invest in their present locations, so had reason not to seek to use incentives. The employers often operated in the industrial districts because of decisions made many years ago and many were not committed to staying. Eighteen percent of employers had located in the industrial districts before 1950, and more than 60% had located at their present site before 1980 when the auto industry’s problems became severe. One third of the employers were directly connected to the auto industry as suppliers or purchasers; of these, 70% had located at their present site before 1980. About 12% of industrial-district employers said the establishment had located in the area because the owner or manager lived not far away or family had connections with the area, but these reasons no longer existed. Thirty percent of the employers in the industrial districts said that they were considering moving or selling their businesses. Eighty percent of those considering a move said they would leave the industrial district.

The employers in the retail districts contrasted sharply with those in the industrial districts in their commitment to their location, so could have been more likely to expand and invest in their establishments. Only 10% had located in the retail districts before 1950 and only a little more than one fourth had located in the area before 1980. Nearly 40% of employers said that they had located in the district because the manager or owner lived in the area or because family members were there. Nearly one quarter of employers were considering moving or selling, but of those thinking about moving, two thirds said that they wanted to stay in the same area.

Economic development planners and program managers normally seek to reach businesses to acquaint them with benefits or to find ways they can use incentives. Even if businesses did know about the incentives, however, the use of these and hiring might not have been different. In part this would be due to off-the-books business practices and to the lack of commitment to the location. Interviews with the Detroit employers as well as research on the ways that employers hire low-skilled workers suggest more explanations for the lack of incentives’ effect on hiring. The incentives assume local labor markets exist or can exist and therefore subsidies can cause marginal change in hiring. For these jobs, no spatial mismatch exists between the residences of workers and location of jobs; numerous potential employees live within walking distance of the businesses. Research on hiring suggests, however, that systems need to change for employers to hire more local entry-level workers.

Barriers to Hiring Nearby Entry-Level Workers

The following sections look at the connections between employers and nearby workers, first in where workers live and then in the ways employers hire entry-level workers, drawing on others’ research and on findings from the Detroit EZ. These relationships show why employers hired so few EZ or local workers and why local hiring varied by district. They also show that EZ and enterprise zone incentives would have to overcome major barriers to cause hiring that would not have occurred otherwise, much more than the programs implicitly assume.

Employment of Nearby Residents

The assumption behind many tax incentive zones is that one can expect neighborhood residents to work for nearby employers. If this assumption is false, zone programs attempt to counter established spatial patterns that are difficult to change. Prior research says, yes, some residents do work for nearby employers, but the numbers vary widely. In Chicago, neighborhood residents were more likely to work within 2 miles of home if more jobs existed within that area with occupations that matched residents’ skills. But if a higher proportion of neighborhood residents were Black, a smaller percentage of those residents worked within 2 miles. The share of residents of small areas of the city who worked 2 miles ranged from 1% to almost two thirds (Immergluck, 1998a, 1998b). Connections between residents and nearby jobs also vary by industry; a study of inner-city employment and residence in Phoenix showed that 15% of the employees of inner-city businesses lived in nearby neighborhoods, a percentage the authors termed low but this share varied greatly among employers and from industry to industry. Government offices employed few nearby residents whereas the employers with the largest percentage of nearby workers were in personal services, agriculture, lodging, and retail (Burns & Gober, 1998). In a neighborhood-oriented retail area of Chicago, half of the businesses employed 75% or more of their employees from the neighborhood. Twenty percent had 50% to 75% neighborhood resident workers (Theodore, 1998). S. Hanson and Pratt (1992) found that the location decisions and hiring patterns of employers with lower skilled workers resulted in localized labor markets so that most workers came from circumscribed areas, often nearby. Women were more likely than men to work close to home (S. Hanson & Pratt, 1991).²

Interviews with employers in the Detroit EZ showed differences in connections to nearby residents across the districts. Table 2 shows the employment of city residents, of workers who lived within 2 miles of the district (a distance considered “local working”; Immergluck, 1998b) and of EZ residents for producer-oriented and consumer-oriented employers in both types of districts. Consumer-oriented employers were in retail, personal services, banks, and other services for households or individuals (such as schools and churches). Such businesses may have more interest in hiring nearby residents because their customers come from the same area. Producer-oriented employers included those in construction; manufacturing; wholesaling; transportation, communication, and public utilities; and services for businesses. These employers might have less connection to nearby residents because they have few customers from those neighborhoods. Employers in retail districts, whether or not in sectors that serve local customers, employed a much larger share of workers who lived within 2 miles and many more who lived in the EZ than did employers in the industrial districts, regardless of industry.

Table 2.

Residence of Employees by Industry Within Districts.

	Percentage of employees by residence
Employers’ location and sector	In city	Within 2 miles	In Empowerment Zone
Industrial districts
Construction, manufacturing, transportation/communication/ public utilities, wholesaling, producer services	40.6	13.7	7.3
Number of employers^a	51	46	38
Retail, consumer services, banks, other services for residents, excluding police precinct^b	38.7	27.5	8.4
Number of employers	6	6	5
Retail districts
Construction, manufacturing, transportation/communication/ public utilities, wholesaling, producer services	76.0	57.3	33.3
Number of employers	10	10	9
Retail, consumer services, banks, other services for residents, excluding police precinct	72.5	54.9	42.9
Number of employers	102	100	86

Source. Interviews with employers.

Some employers in each category did not provide information.

City government employees were required by law to live in the city.

Could difference in the percentage of all jobs that are entry level and therefore available to the largest proportion of nearby workers help account for the differences in the retail districts and the industrial districts? If this were the case, even if employers hired neighborhood residents for entry-level positions, the share of all employees who live nearby would remain lower in industrial districts than in retail districts. In the industrial districts, employers reported that 39% of their employees were in managerial, professional, technical, or office support positions, not entry-level jobs except for some clerical jobs. In the retail districts, employers said that about 41% of employees filled these types of positions. A slightly higher percentage of industrial-district employers than retail-district employers had at least one entry-level position. Furthermore, only 18 employers in one industrial district filled as many entry-level positions in the previous year (192) as 35 employers in two retail districts (189 entry-level positions).

The findings are consistent with those of Immergluck (1998b) and Theodore (1998) and, in the case of retail districts, suggest the existence of the “dynamic dependencies” that S. Hanson and Pratt (1992) showed can exist between employers and a local workforce. Furthermore, the Detroit EZ findings show that different types of businesses in a district dominated by consumer-oriented or producer-oriented employers resemble each other in the geographic range from which they draw employees. The consumer-oriented employers in the industrial districts resembled other businesses in the industrial district in their employment patterns more than they did the consumer-oriented businesses in the retail districts. The producer-oriented businesses in the retail districts resembled other businesses in the retail districts more than producer-oriented businesses in industrial districts. Employment of nearby residents was a district pattern more than an industry pattern in areas of the Detroit EZ that abutted residential neighborhoods.

Another indicator of employers’ connection to local workers is the racial and ethnic mix of employees compared with that of the residential neighborhoods surrounding the districts. If employers hire predominately from nearby neighborhoods, the racial and ethnic mix of the workforce and nearby residents should be similar, even if employees later move away. By this measure, also, the retail districts were more connected to nearby residential areas than were the industrial districts. In the industrial districts, located near neighborhoods where more than 90% of the residents were African American (U.S. Bureau of the Census, 2000), two businesses employed the majority of the African American employees. More than one quarter of the employers in the industrial districts reported that they had no African American workers. The retail areas’ employment reflected the surrounding neighborhoods’ racial and ethnic mix more closely than did the industrial districts’, although employee and resident mixes were not the same. In two of the retail districts, Whites and Latinos had less than their share of employment based on proportions of nearby population, and African Americans had more. In the third district, Whites had more than their share of jobs compared with Blacks (U.S. Bureau of the Census, 2000).

Hiring Entry-Level Workers

The EZ incentives needed to alter hiring decisions to have their intended effects. The modest subsidies would have to change institutionalized approaches to hiring that made different hiring decisions quite difficult.

To learn about hiring in the five case study areas, interviewers asked employers about the entry-level position filled most often in the business or organization. An entry-level job was defined as one that someone could fill with very little work experience or specific training. These jobs were of interest because many residents in low-income neighborhoods, such as those in the EZ, might be able to fill such positions. In the industrial districts more than 70% of employers said they had at least one entry-level position. In the previous year, nearly two thirds of these employers had hired for these positions. In the retail districts, 62% of the employers said they had at least one entry-level position not reserved for family members; 70% of these employers had hired workers for those positions in the previous year. Interviewers asked about the characteristics of the entry-level jobs, the way employers sought to fill these positions, the qualifications they required and desired, how workers moved on from those positions, and the employers’ views about barriers to hiring workers who lived within 2 miles of the district.

The entry-level positions in the industrial areas often offered pay above the minimum wage, benefits, and promotion possibilities. The average starting wage for the most commonly filled positions was close to $8.25 per hour. Nearly all the positions were full-time and permanent, and 85% of the positions offered benefits. Over one fourth of the employers who hired entry-level workers said that workers were more likely to leave the positions through promotion than through quitting or being fired.

In the retail districts, the average starting wage for the most commonly filled entry-level positions was about $6.00 because more jobs paid the minimum wage ($5.15 per hour at that time) and because many positions were in restaurants where workers earned tips. Eighty-four percent of the most commonly filled, entry-level jobs were permanent, but only two fifths were full-time. Only slightly more than one third of the positions offered benefits. One quarter of the employers with entry-level positions stated that workers were promoted more often than they quit or were fired. On every measure, the averages in the retail areas described jobs that were not as good as the ones in the industrial districts.

Recruiting Entry-Level Workers

Prior research has shown that employers often hire entry-level workers through word of mouth. Social networks of current employees and customers, Granovetter’s (2005) “weak” ties, provide the contacts that help employers to hire efficiently. Granovetter (2005) wrote, “Prospective employers and employees prefer to learn about one another from personal sources whose information they trust” (p. 36). This may especially be the case with hiring for low-skilled, entry-level positions because other indicators of future work performance are often not available—the record of reliability in previous positions, for instance, or of educational achievement. This implies that if employers’ connections with nearby low-income neighborhoods are weak, they will not hire from those neighborhoods.

Others have also shown the predominance of word-of-mouth referral hiring. Moss and Tilly (2001) found that 53% of interviewed employers in Detroit and Los Angeles reported using employee referrals extensively to recruit entry-level workers. Holzer (1996) found far fewer in asking a different question—26% of employers in four central cities said that a referral from an employee led to the most recent hire for a noncollege job. Ninety-four percent of the employers that S. Hanson and Pratt (1992) interviewed used word of mouth to recruit new employees, not all entry level. Almost as many used newspaper ads, but this varied with the type of job to be filled; many businesses placed newspaper ads only for professional positions. Kirschenman and Neckerman (1991) reported that Chicago employers relied heavily on referrals from their employees to hire new workers. Kasinitz and Rosenberg (1996) found that in Brooklyn employers used referrals heavily, often from current workers who did not live in the area, to hire new employees; employers contended this was the way to find a reliable worker for an unskilled position. Residents of the Brooklyn neighborhood “simply do not hear about openings, which are rarely advertised,” Kasinitz and Rosenberg wrote (p. 189). Indeed, in hiring for entry-level positions, Detroit EZ employers followed practices that made hiring local workers likely only if they already employed local workers and had local customers.³ Employers relied most heavily on word of mouth to find workers (see Table 3). Current employees were the most important source of referrals. Owners’ and managers’ contacts and customers’ referrals were also important. In the retail districts, one fourth of the employers relied on word of mouth alone to recruit new employees. Because few industrial-district workers lived nearby and they were predominately White, nearby residents were not part of the word-of-mouth network that could alert them to job openings. In the retail districts, the many workers from nearby neighborhoods could recruit other local residents to the job openings. Customers who came largely from the neighborhoods were also a source of job referrals. Retail-district owners and managers often lived in the nearby neighborhoods or had family there. Therefore, owners’ connections also reinforced the neighborhood hiring in the retail districts.

Table 3.

Methods Employers Used to Recruit Entry-Level Employees.

	Industrial districts		Retail districts
Methods of recruiting	Number of employers	Percentage of employers	Number of employers	Percentage of employers
Word of mouth	36	83.7	63	91.3
Employees’ friends, family	29	69.4	37	59.3
Customers’ friends, family	7	16.7	20	32.0
Owners’ friends, family	8	19.1	16	25.6
Residents of the area	0	0.0	2	3.2
Vendors and suppliers	3	7.2	0	0.0
People in similar businesses	1	2.4	1	1.6
Walk-in applications	14	31.8	38	55.9
Newspaper ads	19	43.2	14	20.6
Internal posting, human resources office	12	27.3	5	7.4
Temp agency	10	22.7	1	1.5
Public employment agency	9	20.5	2	2.9
Contact local organizations: churches, schools, etc.	3	6.8	11	16.2
Union procedures	4	9.1	0	0.0
Internet ads	2	4.5	0	0.0
Radio ads	0	0.0	1	1.5
Mailing of sales bulletin	0	0.0	1	1.5
Total number of employers	44		69

Source. Interviews with employers.

Retail-district employers were more likely to recruit workers in other ways that connected with local residents. A larger percentage of retail-district employers recruited workers through walk-in applications and help-wanted signs in their windows. Walk-in applications, often in response to signs posted in windows, in neighborhood-oriented retail areas further reinforce a neighborhood labor market. In the industrial districts, employers less often recruited walk-in applications, and signs were rarely posted. Making a walk-in application was also more difficult for a prospective employee because industrial-district establishments had more evident security and building entrances were often difficult to find. Retail-district businesses’ recruitment of workers through local churches, schools, and other organizations also helped them hire local workers, as S. Hanson and Pratt (1992) also showed.

Other recruitment methods used by larger shares of industrial-district employers had little promise of attracting nearby workers. Postings in the company, use of corporate human resource departments (in several cases located in the suburbs), and union-determined hiring could exclude people not already connected with these types of businesses. Temporary help agencies that recruited workers for nearly one fourth of industrial-district employers had no particular association with nearby residents.

The use of newspaper ads, a public employment agency, and radio ads, in contrast, publicized job openings more widely and could allow residents of any neighborhood to learn about openings. A substantial share of industrial-district employers, close to 45%, advertised entry-level job openings in newspapers. One fifth of employers in the retail districts also used newspaper ads. One fifth of industrial-district employers used a public employment agency, whereas almost no one did so in the retail districts.

The social network referral system of finding new employees, reinforced by other approaches to recruiting employees, is a major explanation for the stronger local labor market in the retail districts and the weaker one in the industrial areas. The heavy reliance on word-of-mouth hiring meant that employers drew on existing social networks. In the industrial areas, these networks did not include nearby residents; in the retail districts they did.

Reinforcement of Ethnic and Racial Separation

Word-of-mouth hiring and other forms of recruiting that target or exclude certain groups segment a labor force not only spatially but also racially and ethnically. Research has shown the concentration of White employees in White-owned businesses and Blacks in Black-owned businesses. Looking at the percentage of minority employees in small businesses in 28 large metropolitan areas, Bates (1994) found that Black firms employed a much higher percentage of Black workers than White firms whether or not they were located in minority neighborhoods. In a second study (Bates, 2006) he found that more than two thirds of small Black-owned firms employed at least 90% minority workers. Seventy-seven percent of small nonminority-owned firms had 25% or less minority workers. The implication is that unless many of the businesses in the EZ were minority owned or had African American managers, African Americans were less likely to be hired although they lived nearby. Such segregation suggests discrimination, but as Moss and Tilly (2001) point out, racial prejudice and rational economic decisions are difficult to disentangle. Kirschenman and Neckerman (1991) argued convincingly that employers’ failure to hire Blacks went beyond the Blacks’ lack of access to job networks and reflected bias; employers they interviewed expressed racial preferences. Consistent with Moss and Tilly (2001), they found that employers often perceived Black workers to lack work ethic, attitudes, and skills. Employers used recruiting approaches for unskilled jobs that excluded Blacks from applicant pools (Neckerman & Kirschenman, 1991).

Research on immigrant and ethnic employment in the United States shows that many immigrant ethnic groups concentrate disproportionately in some types of businesses and occupations. Employers may prefer employees of the same ethnic group because of common language and because the business serves many customers of the same ethnic group, as well as because of greater trust in referrals from the social network (Portes & Rumbaut, 2006; Waldinger, 1996). As Waldinger (1996) wrote, “Employers who hire co-ethnics gain a reliable work force with an interest in skill acquisition—attributes that diminish the total labor bill and make for greater flexibility” (p. 24).

The social networks that lead to word-of-mouth hiring among ethnic groups also exclude others who are not part of that social network. Waldinger has argued that once an ethnic niche (areas of the labor market where particular groups specialize) exists, employers prefer to hire friends and relatives of their key workers (coethnics) because this offers greater confidence and trust in the behavior of new hires. As Parks (2004) wrote, “Jobs that are located down the street from Blacks but are held primarily by Mexican immigrants may prove as inaccessible as spatially distant jobs” (p. 142).

Consistent with this analysis of labor markets, substantial racial and ethnic separation existed in the workplaces in the Detroit EZ. Table 4 shows that in the industrial districts African American managers supervised a workforce that was nearly two thirds African American. White managers oversaw a workforce that was 60% White. In the retail districts, the ethnic and racial separation was even more marked. A worker was most likely to be employed by an owner or manager of the same race or ethnicity. This relationship held whether the employer was White, African American, Latino, Arab, or a member of a specific immigrant group. Two thirds of the employees in the retail districts worked with managers or owners of the same ethnic or racial group. The racial segregation in the EZ employers’ workforces suggests discrimination in hiring. Employers in the Detroit EZ did not articulate racial bias despite considerable probing by interviewers about barriers to hiring.⁴

Table 4.

Percentage of Employees in Racial and Ethnic Groups by Race and Ethnicity of the Owner or Manager.

Race/ethnicity of owner/manager^a	Percentage of employees who are
Race/ethnicity of owner/manager^a	White	African American	Latino	Arab	Asian	Other^b
Industrial districts
White	59.5	35.3	0.8	0.3	0.8	0.8
African American	32.1	65.3	2.3	0.0	0.3	0.0
Total number of employers providing information = 59 for 2,618 employees
Retail districts
White	62.8	11.3	21.1	3.8	2.5	3.5
African American	8.0	80.7	9.3	1.3	0.0	0.0
Latino	14.5	6.1	63.9	1.7	0.0	0.0
Arab	9.5	13.1	15.5	54.8	1.2	1.2
Other	0.0	0.0	25.0	0.0	0.0	75.0
Total number of employers providing information = 111 for 978 employees

Source. Interviews with employers.

Categories with zero or one employer are not shown.

“Other” included anyone employers placed outside the major categories; these were usually identified as immigrants from specific countries in Europe, Asia, and the Middle East.

Ethnic employees, many of them immigrants, concentrated in businesses where the owner or manager came from the same ethnic group. Ethnic connections led to job opportunities for Latino and Arab immigrants living near two of the retail districts; however, these businesses were much less likely to hire either Blacks or Whites who lived in the same neighborhoods.

Qualifications for Entry-Level Jobs

Referral hiring networks and ethnic and racial connections tended to determine employers’ hiring for entry-level positions; however, the qualifications that employers required for such positions could further influence the extent of local hiring. The “hard” skill (cognitive and technical) requirements of jobs nationally have increased, one cause of the high unemployment of low-skilled individuals (Bartik, 2001; Holzer, 1996; Moss & Tilly, 2001). Research on central-city joblessness has long debated the role of a skills mismatch. Increases in skills requirements for many jobs threaten to exclude more city residents with low levels of skills (Moss & Tilly, 2001).

Holzer (1996), Farley, Danziger, and Holzer (2000), and Moss and Tilly (2001), relying on the same telephone survey of employers, reported that a very large share of employers had stringent requirements for the last hired worker in a job that did not require college. In four central cities (one of them Detroit), 76% of employers required a high school diploma for that position. The findings of the same survey for Detroit alone showed that 74% of employers imposed that requirement. Sixty-nine percent of Detroit employers and two thirds of the four central cities’ employers required experience in similar work. Seventy-two percent of employers in the four cities and 76% of those in Detroit required work experience of some kind (Moss & Tilly, 2001). A very large share of employers required literacy and numeracy. These requirements suggest substantial barriers for central-city residents in getting noncollege jobs.

In contrast, employers in the Detroit EZ communicated fewer requirements for entry-level workers than those reported above, and numerous nearby residents had the required skills. Table 5 shows the requirements of applicants for entry-level positions in the industrial and retail districts. A considerable share of employers in both types of districts required a high school diploma, even of entry-level workers. Employers in the industrial districts were more likely to require learned or hard skills, or an indication in past experience that an applicant had these—literacy, numeracy, and work experience of any type as well as experience in similar businesses and positions. The industrial-district jobs were more likely to require physical strength or manual dexterity. Overall, the learned skill requirements were greater for a job in the industrial districts than in the retail districts. When asked what qualifications they hoped for but did not require in entry-level employees, employers in all the Detroit EZ districts preferred workers with experience in similar work.⁵

Table 5.

Required Qualifications in Entry-Level Hires.

	Industrial districts		Retail districts
Qualifications	Number of employers	Percentage of employers	Number of employers	Percentage of employers
Credentials
High school diploma	11	25.6	21	30.0
Learned skills
Literacy	21	52.5	10	14.3
Numeracy	15	37.5	8	11.4
English proficiency^a			8	11.4
Spanish proficiency^a			13	18.6
Specific work experience	11	27.5	12	17.1
General work experience	3	7.5	4	5.7
Commercial or other driver’s license	9	22.5	6	8.6
Computer familiarity	2	5.0	5	7.1
Physical ability
Physical strength	4	10.0
Manual dexterity	2	5.0
Other physical characteristics	3	7.5
Life skills
Work ethic	11	26.8	12	17.1
Good attendance, punctuality	12	29.3	8	11.4
Communication skills	7	17.1	13	18.6
Honesty	4	9.8	10	14.3
Good personality	7	17.1	8	11.4
Willing to learn			3	4.3
Positive attitude	2	4.9	7	10.0
Polite			4	5.7
Able to follow instructions	2	4.9	2	2.9
Sobriety	5	12.2	1	1.4
Appropriate appearance, cleanliness			11	15.7
No required qualifications	1	2.3	6	8.5
Total employers	44		70

Note. Qualifications mentioned by only one employer are not included. Percentages adjusted to account for missing information.

Source. Interviews with employers.

Mentioned only where many residents were immigrants.

Overall, required learned-skills qualifications in the Detroit EZ allowed unskilled, inexperienced workers near retail districts to get entry-level jobs more than they enabled such workers near industrial districts to get work in those places. Nevertheless, many workers in the areas around the industrial districts appeared to meet required qualifications for starting jobs there. In one of the industrial districts, for instance, nearly one fourth of the entry-level jobs were clerical, and employers sought many workers in manual support for industrial processes, including helpers for machine operators. In 1990, the largest share of workers who lived in the surrounding area worked in administrative support occupations, and a large share worked as machine operators, assemblers, and inspectors. Residents identified themselves in these two occupational categories in larger proportions than did the residents of the metropolitan region as a whole. Nearly half of those older than 18 years had a high school diploma (U.S. Bureau of the Census, 1990). These characteristics of the population do not mean that applicants for positions would have met the requirements, but a pool of nearby workers did exist from whom to draw. Most residents of neighborhoods near industrial districts would not have had an opportunity to demonstrate their qualifications, however, because they would not have heard of openings.

The differences in findings between the four-city telephone survey and this study of Detroit EZ employers may be because of real differences in the requirements of employers citywide and those of employers adjacent to EZ neighborhoods. Or the difference could be due to variations in the form of the questions (predetermined prompts or open ended with probes), or to differences in the job (last hired worker for a noncollege job or most common entry-level job).

Table 5 shows the importance of “soft” skills or life skills as well—work ethic, good attendance, and a positive attitude, for instance. An employer has difficulty judging such skills in an applicant for an entry-level position, although prior work experience can serve as an indicator, so the employer’s attitudes inevitably contribute to a hiring decision (Moss & Tilly, 2001). Hiring based on trusted referrals helps increase the likelihood that applicants have these skills. In the Detroit EZ, industrial-district employers emphasized requirements for work ethic and good attendance and punctuality more than retail-district employers did. Retail-district employers emphasized required characteristics that affected the way workers interacted with customers—communication skills, politeness, honesty, and an appropriate appearance and cleanliness.

Table 6 shows employers’ screening techniques, tests or checks that an employee had to pass to be considered for a job. Nearly two thirds of the employers in the industrial districts used drug testing as a screen for entry-level employees. Nearly two fifths required a medical examination. In contrast, close to 20% of retail-district employers required drug testing, and almost none required a medical examination. About 40% of employers in both kinds of districts checked references. About one quarter of the retail-district employers checked prospective employees’ criminal records, whereas only about half that proportion checked for criminal records among the industrial-district employers. A large percentage of employers in both districts, although a considerably larger percentage in the retail districts, did no screening of prospective employees. Overall, the screening techniques excluded applicants with substance abuse and other severe health problems more often in industrial districts, and more employers in retail districts used a criminal record to exclude applicants. As in the case of skills requirements, the four-city telephone survey showed stricter requirements; Holzer (1996), looking at all types of industries, reported a much more restrictive standard; and only 33% of employers in central cities said that they would hire someone with a criminal record for a noncollege job.

Table 6.

Screening of Prospective Entry-Level Employees.

	Industrial districts		Retail districts
Screening techniques	Number of employers	Percentage of employers	Number of employers	Percentage of employers
Drug testing	26	63.4	12	18.7
Medical exam	16	39.0	2	3.1
Company/civil service exam	1	2.4	6	9.3
References checked	16	39.0	28	43.6
Driver’s license required	2	4.9	4	6.2
Criminal record checked	5	12.2	16	24.9
No screening	9	22.0	26	37.7
Total employers	41		69

Note. Percentages adjusted for missing information.

Source. Interviews with employers.

Views About Neighborhood Workers

Although requirements for hard skills and screening would not have prevented local hiring, employers’ opinions about the employment qualifications of the local workforce (those living within 2 miles of the establishment) reflected and likely reinforced the hiring patterns in the Detroit EZ. Other research has shown that stigma attached to residents of specific places is a factor in willingness to consider an applicant for a job—“location-based discrimination,” in the words of Kasinitz and Rosenberg (1996). Much of this discrimination revolved around perceptions of crime in the Brooklyn study. Kirschenman and Neckerman’s (1991) interviews with Chicago employers showed that they used location of school attended, residence in a public housing project, and address as ways to screen out potential employees from the central city. Moss and Tilly (2001) argued that employers’ negative views of the inner city pushed residents lower in the hiring queue. In in-depth interviews, 64% of central-city businesses had at least one respondent who generalized that the inner-city workforce was worse than the suburban one. In a telephone survey, 15% of Detroit central-city employers said that inner-city residents were weaker job applicants or employees. Moss and Tilly (2001, pp. 172-173) stated that such statements reflected a mix of stereotypes, cultural distance, and reality that was difficult to disentangle.

In the Detroit EZ, using a scale of 1 to 5 with 1 the least qualified and 5 the most qualified, industrial-district employers with entry-level positions gave local workers an average rating of 2.6. About 13% of industrial-district employers rated neighborhood workers highly, 4 or 5. In the retail districts, employers hiring at the entry level gave neighborhood workers an average rating of 3.4; more than 40% of employers rated neighborhood workers 4 or 5.

In industrial districts, one third of the employers with entry-level positions said that they made special efforts to hire workers who lived nearby, whereas in retail districts twice that share did so (see Table 7). Nevertheless, in one industrial district, for instance, half of the employers who filled entry-level positions in the previous year did not consider any nearby residents for these openings.

Table 7.

Employers’ Views of Nearby Residents as Employees.

	Employers with entry-level jobs
Opinions about residents as employees	Number	Percentage
Industrial districts
Make special efforts to recruit local workers	15	33.3
Reasons for hiring problems
Lack of skills, education, work experience	12	29.3
Risk of crime	2	4.9
Drug/substance abuse	6	14.6
Poor work ethic or poor attendance	10	24.4
Lack of support systems for transportation, child care, phone communication	3	7.3
No driver’s license	2	4.9
Nothing interferes	9	21.4
Number of employers	42
Retail districts
Make special efforts to recruit local workers	46	65.7
Reasons for hiring problems
Lack of skills, education, work experience	15	23.1
Fail company exam	2	3.1
Dishonesty, risk of crime	3	4.6
Drug/substance abuse	4	6.2
Poor work ethic or poor attendance	8	12.3
Lack of support systems for transportation, child care, phone communication	2	3.1
Unable to meet expectations for behavior on the job^a	8	12.3
Nothing interferes	30	46.9
Number of employers	65

Source. Interviews with employers.

For instance, employers mentioned inappropriate appearance, impoliteness, difficult personality, inability to follow instructions.

Employers’ general impressions, not tied to specific hiring or a particular position’s requirements, tended to reflect discrimination that could influence their hiring decisions. In both the industrial and the retail districts, employers most often generalized about applicants’ lack of skills, education, and work experience as interfering with hiring workers from the immediate area; nearly as often they cited poor work ethic. In industrial districts the workers’ substance abuse was a third important factor employers said interfered with local employment. Lack of transportation, child care, and other support systems for work were not mentioned often in either kind of district, but these problems may have caused the difficulties with attendance that employers cited.

Despite many employers’ negative overall views of nearby workers, one fifth of employers in the industrial districts and close to half of those in retail districts said that nothing interfered with their hiring people who lived nearby for entry-level positions.

Implications for Paying Employers to Hire Workers in Distressed Places

For tax incentives in the Detroit EZ to achieve the goal of improving economic opportunity for residents of nearby distressed neighborhoods, a chain of events needed to occur. The incentives needed to influence employers’ decisions about hiring; if subsidies affected location, expansion, or investment decisions, these needed to lead to different hiring choices as well. Additional hiring in businesses in the EZ needed to lead to jobs for EZ residents. Because the incentives did not affect employers’ behavior and because employers hired nearby residents primarily in retail districts (where jobs offered low pay, infrequent benefits, part-time work, and fewer promotion opportunities), the EZ tax incentives had little effect on economic opportunity.

Evaluation of why employers said that they did not use the incentives suggests that if economic development officials had publicized the incentives more or worked with employers to make more use of these, some employers would have used the funds more extensively. Increasing the amount of the subsidy for hiring an EZ resident could have made the use of the incentives more attractive as well. However, the findings of this study on how employers hire entry-level workers suggests that the greater use of subsidies would principally have gone for activities the employers were already undertaking, not for hiring that would not have occurred otherwise, because systems of hiring reinforced the status quo.

Nevertheless, potential existed for stronger connections between residents with little education and work experience and jobs because employers in both types of districts filled substantial numbers of entry-level jobs in a year. Furthermore, many nearby residents had the kinds of skills and work experience employers sought. The major way of recruiting new workers for job openings, however, was through referrals from social networks, especially among current workers, and resulted in few hires of nearby residents in industrial districts and segregation by race and ethnicity. Workers’ social networks reinforce spatial relationships between neighborhoods and employers (S. Hanson & Pratt, 1992). The creation of new connections in neighborhoods where current workers do not live is unusual. The stereotypes held by industrial district employers about workers from nearby neighborhoods, likely partially valid, also made such connections difficult.

Enabling residents to obtain nearby jobs depends on creating new connections for neighborhood residents with jobs in industrial districts. The implication of these systems of hiring for economic development programs is that emphasis needs to be placed on creating a trusted alternative “mouth” in the word-of-mouth hiring system in industrial districts, the type of district that most tax incentive zones encompass. A community-based organization that earns the confidence of employers could screen and recommend neighborhood residents for job openings to establish a social network that would connect to local workers. In this personalized system, unlike a usual employment service, the employer and the leaders of the community-based organization know and trust each other. The first few hires also need to demonstrate that this alternative system does as well in identifying good workers as relying on already established contacts. Kasinitz and Rosenberg (1996) saw a local development corporation in Brooklyn succeed in beginning a job referral service to connect local employers and local workers. A construction contractor credited the organization’s screening with the success of the first few laborers hired from the nearby neighborhood. Bartik (2001) described a business where the CEO cultivated trusted sources of referrals, such as churches, to get promising applicants from “high-risk” groups (p. 217).

Where job referral networks existed in the retail districts, many jobs did not offer possibilities of promotion within the same business. As Waldinger (1996) has argued, however, ethnic connections can serve to help workers receive better wages and move to better jobs as coethnics have access to these; the concentration of ethnics in certain businesses was a marked characteristic in two of the retail districts in the Detroit EZ. What about creating possibilities for other workers to move to better jobs? Newman (1995) recommended an “employer consortium” to aid fast-food workers to move beyond low-wage work. Fast-food employers would select “reliable, punctual, hard-working, and motivated” (Newman, 1995, p. 26) workers who had at least 1 year of work for consideration by other employers who offered jobs with better pay, benefits, and promotion possibilities. The success of such an effort would depend on the trust other employers developed in the recommendations of the consortium.

Although or perhaps because they had less frequent experience with nearby residents as employees, industrial-district employers had a more negative view in general about workers who lived nearby. More information about local workers’ qualifications might help correct employers’ impressions over time. Intermediaries such as the community-based referral organization or the employer consortium can help provide evidence to counter prejudices.

Racial discrimination figures in the segregation of Blacks and Whites in workplaces, although employers’ negative overall impressions of residents of African American neighborhoods and stereotypes about lack of skills veil racist views. Impressions about lack of skills are no doubt also correct in many cases. Because of the connection to race, however, those who administer tax incentive zones should work closely with employers to implement employer-connected training programs to transition workers who would not normally be hired into positions (Bartik, 2001). Although many businesses in such zones are too small for a tailored training program, such a program could focus on occupations that many businesses employ in a district. In the Detroit EZ’s industrial districts, for instance, this training could have focused on clerical positions with strong emphasis on soft skills.

Employers who leave industrial and retail districts will be lost to efforts to connect them with workers from the nearby low-income neighborhoods. For many industrial-district employers, the current location was no longer as appropriate as others because the conditions facing the business had changed over the many years since the owners chose its site. Therefore, programs to retain employers are important. The employers’ views about problems suggested some of the ways to do this: improving the delivery of public services and creating programs to deliver the services that the city government does not provide reliably, reducing crime and increasing security without barricading businesses from surrounding residential areas, facilitating employers’ expansion in the same areas by making the sale of city-owned land and buildings easier, and providing services that encourage succession of ownership rather than shutdowns when owners retire.

Tax incentive zones assume that marginal changes in nearby employment can occur with subsidies, but institutional change has to happen first. Implicitly, tax incentive programs assume that neighborhood labor markets exist and that incentives to employers can encourage and reinforce these local labor markets. Because such labor markets do not necessarily exist, tax incentive zone efforts need to focus more on creating connections between residents and nearby employers. Programs to address this issue directly hold promise of encouraging the employment of nearby residents of distressed areas.

Footnotes

Appendix

Total Number of Employers, Percentage Interviewed, and Percentage in the Empowerment Zone (EZ) by Type of District and Industry.

Industry	Number of employers	Percentage interviewed	Percentage of total in EZ	Percentage of EZ interviewed
Industrial districts
Construction	8	75.0	62.5	60.0
Manufacturing	56	53.6	50.0	57.1
TCPU	8	62.5	14.3	100.0
Wholesale	16	62.5	25.0	50.0
Retail	5	40.0	40.0	25.0
FIRE	1	100.0	100.0	100.0
Services	12	66.7	38.5	80.0
Public administration	2	100.0	0.0	0.0
Total	108	59.3	44.4	58.3
Retail districts
Construction	3	66.7	33.3	0.0
Manufacturing	6	100.0	50.0	100.0
TCPU	1	100.0	0.0	0.0
Retail	95	72.6	46.3	72.7
FIRE	6	66.7	33.3	50.0
Services	60	50.0	33.3	50.0
Public administration	3	100.0	0.0	0.0
Total	174	66.1	40.2	65.7

Note. TCPU = transportation, communication, and public utilities; FIRE = finance, insurance, and real estate.

Source. Censuses constructed through field research and through community economic development organizations’ directories.

Acknowledgements

Kathy Wendler (Southwest Detroit Business Association), Laurie Garvey and Jeff Burdick (Islandview Village Development Corporation), Ron Stewart (U-SNAP-BAC), Linda Stingl (Eastside Industrial Council), and Pat Bosch (Nortown CDC) provided valuable advice. Amy Bullock, Mary Beth Bullock, Anne Dibble, Emily Gosack, Arlova Jackson, Kelly Janiga, Heidi Lubin, Gandy Madrigal, Manuel Magaña, Alejandra Montes, Amy Rogers, Del Sanders, Erica Sims, Lisa Solomon, Mellie Torres, Deborah VanHoewyk, and Catalina Velasco provided research assistance.

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research for this article received funding from the National Center for the Revitalization of Central Cities, University of New Orleans; the Bonner Foundation (through the Ginsberg Center, University of Michigan); the Undergraduate Research Opportunities Program, the Provost’s Career Development Fund, and the Coca Cola Foundation Faculty Recognition Award at the University of Michigan.

Notes

Author Biography

Margaret Dewar is a professor of urban and regional planning at the University of Michigan. Her research concerns economic development, housing, and land use in cities experiencing extensive population and job loss and property disinvestment. She is the coeditor of The City After Abandonment (Philadelphia: University of Pennsylvania Press, 2012)

References

Bartik

(2001). Jobs for the poor: Can labor demand policies help? New York, NY: Russell Sage Foundation.

Bartik

(2004). Evaluating the impacts of local economic development policies on local economic outcomes: What has been done and what is doable. In Nolan

Wong

(Eds.), Evaluating local economic and employment development: How to assess what works among programmes and policies (pp. 113-141). Paris, France: Organisation for Economic Co-operation and Development.

Bates

(1994). Utilization of minority employees in small business: A comparison of nonminority and black-owned urban enterprises. Review of Black Political Economy, 23, 113-121.

Bates

(2006). The urban development potential of black-owned businesses. Journal of the American Planning Association, 79, 227-237.

Boarnet

M. G.

(2001). Enterprise zones and job creation: Linking evaluation and practice. Economic Development Quarterly, 15, 242-254.

Boarnet

M. G.

Bogart

W. T.

(1996). Enterprise zones and employment: Evidence from New Jersey. Journal of Urban Economics, 40, 198-215.

Bondonio

Engberg

(2000). Enterprise zones and local employment: Evidence from the states’ programs. Regional Science & Urban Economics, 30, 519-549.

Bostic

R. W.

Prohofsky

A. C.

(2006). Enterprise zones and individual welfare: A case study of California. Journal of Regional Science, 46, 175-203.

Brashares

(2000). Empowerment Zone tax incentive use: What the 1996 data indicate. Statistics of Income, 20(1), 236-252.

10.

Budget Reconciliation Act of 1993, Empowerment Zones, Enterprise Communities, and Rural Development Investment Areas, Pub. L. No. 103-66.

11.

Burns

E. K.

Gober

(1998). Job linkages in inner-city Phoenix. Urban Geography, 19, 12-23.

12.

Busso

Gregory

Kline

P. M.

(2010, June). Assessing the incidence and efficiency of a prominent place based policy (Working paper No. 16096). Cambridge, MA: National Bureau of Economic Research.

13.

Detroit Planning and Development Department. (1994). Jumpstarting the motor city: Detroit Empowerment Zone strategic plan. Detroit, MI: Author.

14.

Elvery

J. A.

(2009). The impact of enterprise zones on resident employment: An evaluation of the enterprise zone programs of California and Florida. Economic Development Quarterly, 23, 44-59.

15.

Farley

Danziger

Holzer

(2000). Detroit divided. New York, NY: Russell Sage Foundation.

16.

Gottlieb

(1997, September). Commuting patterns in Cleveland’s Empowerment Zone (Working paper). Cleveland, OH: Center for Regional Economic Issues, Case Western Reserve University.

17.

Granovetter

(2005). The impact of social structure on economic outcomes. Journal of Economic Perspectives, 19(1), 33-50.

18.

Greenbaum

Engberg

(2000). An evaluation of state enterprise zone policies. Review of Policy Research, 17(2/3), 30-46.

19.

Ham

J. C.

Swenson

İmrohoroğlu

Song

(2011). Government programs can improve local labor markets: Evidence from state enterprise zones, federal Empowerment Zones and federal Enterprise Community. Journal of Public Economics, 95, 779-797.

20.

Hanson

(2009). Local employment, poverty, and property value effects of geographically-targeted tax incentives: An instrumental variables approach. Regional Science & Urban Economics, 39, 721-731.

21.

Hanson

Pratt

(1991). Job search and the occupational segregation of women. Annals of the Association of American Geographers, 81, 229-253.

22.

Hanson

Pratt

(1992). Dynamic dependencies: A geographic investigation of local labor markets. Economic Geography, 68, 373-405.

23.

Hanson

Pratt

(1995). Gender, work, and space. New York, NY: Routledge.

24.

Hebert

Vidal

Mills

James

Gruenstein

(2001, November). Interim assessment of the Empowerment Zones and Enterprise Communities (EZ/EC) program: A progress report. Washington, DC: U.S. Department of Housing and Urban Development. Retrieved from http://www.huduser.org/Publications/pdf/ezec_report.pdf

25.

Holzer

(1996). What do employers want? New York, NY: Russell Sage Foundation.

26.

Immergluck

(1998a). Job proximity and the urban employment problem: Do suitable nearby jobs improve neighborhood employment rates? Urban Studies, 35, 7-23.

27.

Immergluck

(1998b). Neighborhood economic development and local working: The effect of nearby jobs on where residents work. Economic Geography, 74, 170-187.

28.

Immergluck

Hilton

(1996). Breaking down barriers: Prospects and policies for linking jobs and residents in the Chicago Empowerment Zone. Chicago, IL: Woodstock Institute.

29.

Kasinitz

Rosenberg

(1996). Missing the connection: Social isolation and employment on the Brooklyn waterfront. Social Problems, 43, 180-196.

30.

Kirschenman

Neckerman

K. M.

(1991). “We’d love to hire them, but . . .”: The meaning of race for employers. In Jencks

Peterson

P. E.

(Eds.), The urban underclass (pp. 203-232). Washington, DC: Brookings Institution.

31.

Kolko

Neumark

(2010). Do some enterprise zones create jobs? Journal of Policy Analysis and Management, 29, 5-38.

32.

Ladd

H. F.

(1994). Spatially targeted economic development strategies: Do they work? Cityscape, 1(1), 193-218.

33.

Moss

Tilly

(2001). Stories employers tell: Race, skill, and hiring in America. New York, NY: Russell Sage Foundation.

34.

Neckerman

K. M.

Kirschenman

(1991). Hiring strategies, racial bias, and inner-city workers. Social Problems, 38, 433-447.

35.

Newman

K. S.

(1995). Dead-end jobs: A way out. Brookings Review, 13(4), 24-27.

36.

Oakley

Tsao

(2006). A new way of revitalizing distressed urban communities? Assessing the impact of the federal Empowerment Zone program. Journal of Urban Affairs, 28, 443-471.

37.

Oakley

Tsao

(2007). Socioeconomic gains and spillover effects of geographically targeted initiatives to combat economic distress: An examination of Chicago’s Empowerment Zone. Cities, 24, 43-59.

38.

Papke

(1993). What do we know about Enterprise Zones? In Poterba

(Ed.), Tax policy and the economy (Vol. 7, pp. 37-72). Cambridge: MIT Press.

39.

Parks

(2004). Access to work: The effects of spatial and social accessibility on unemployment for native-born black and immigrant women in Los Angeles. Economic Geography, 80, 141-172.

40.

Peters

Fisher

(2004). The failures of economic development incentives. Journal of the American Planning Association, 70, 27-37.

41.

Peters

A. H.

Fisher

P. S.

(2002). State enterprise zone programs: Have they worked? Kalamazoo, MI: Upjohn Institute.

42.

Portes

Rumbaut

R. G.

(2006). Immigrant America: A portrait. Berkeley: University of California Press.

43.

Rich

M. J.

Stoker

R. P.

(2010). Rethinking empowerment: Evidence from local Empowerment Zone programs. Urban Affairs Review, 45, 775-796.

44.

Rosenberg

Kasinitz

(1993, Autumn). Why enterprise zones will not work. City Journal, 3(4). Retrieved from http://www.city-journal.org/article01.php?aid=1454

45.

Taxpayer Relief Act of 1997, Pub. L. No. 105-34.

46.

Theodore

(1998). Jobs and poverty alleviation in Chicago’s Empowerment Zone. In Herring

Bennett

Gills

Jenkins

N. T.

(Eds.), Empowerment in Chicago: Grassroots participation in economic development and poverty alleviation (pp. 183-206). Chicago: Great Cities Institute, University of Illinois.

47.

U.S. Bureau of the Census. (1990). Census of population. Washington, DC: Author. Retrieved from http://factfinder2.census.gov/faces/nav/jsf/pages/index.xhtml

48.

U.S. Bureau of the Census. (2000). Census of population. Washington, DC: Author. Retrieved from http://factfinder2.census.gov/faces/nav/jsf/pages/index.xhtml

49.

U.S. Department of the Treasury. (1993, December). Tax incentives for Empowerment Zones and Enterprise Communities. Washington, DC: Internal Revenue Service.

50.

U.S. Department of the Treasury. (1999, February). Tax incentives for Empowerment Zones and other distressed communities (Pub. 954). Washington, DC: Internal Revenue Service.

51.

U.S. General Accountability Office. (1999, September). Businesses’ use of Empowerment Zone tax incentives (Report to Congress, No. GAO/RCED-99-253). Washington, DC: Author.

52.

U.S. General Accountability Office. (2006, September). Empowerment Zone and Enterprise Community Program (Report to Congressional Committees, No. GAO-06-727). Washington, DC: Author.

53.

Vidal

A. C.

(1995). Reintegrating disadvantaged communities into the fabric of urban life: The role of community development. Housing Policy Debate, 6, 169-230.

54.

Waldinger

(1996). Still the promised city? African-Americans and new immigrants in postindustrial New York. Cambridge, MA: Harvard University Press.

55.

Wilder

M. G.

Rubin

B. M.

(1996). Rhetoric versus reality: A review of studies on state enterprise zone programs. Journal of the American Planning Association, 62, 473-491.