Volume 12 Number 3
November 1998

Virginia Carlson, Department of Urban Planning

This research was funded by a grant from the Center for Urban Initiatives and Research
at the University of Wisconsin-Milwaukee.


Since the late 1970s, high unemployment rates and low labor force participation rates in the inner-city have reflected the difficulty that urban residents face in securing jobs in the metropolitan labor market. Although the 1996 average unemployment r ate for the Milwaukee metropolitan area was about 3.0%, within the African-American population (which is concentrated in the inner-city), the unemployment rate was estimated to be about six times as high.1

Four related theories have been advanced to address the failure of inner-city2 residents to secure and maintain employment: 1) the existence of a spatial-skills mismatch, ("job mismatch"); 2) labor market discrimination, ("jo b discrimination"); 3) the loss of entry-level manufacturing jobs, ("job loss"); and 4) firm labor-sourcing strategies that have increasingly made use of part-time and temporary labor ("job restructuring"). Proponents of the job mismatch theory contend that racial minorities achieved a plurality of central-city populations at a time when economic restructuring eliminated many blue-collar industrial jobs from cities, and employment decentralized to the suburban periphery (Wilson 1987). In some cases this employment decentralization can be attributed to company strategies to locate near specific labor pools (Nelson 1984). Inner-city residents could not follow jobs to the periphery because housing discrimination prevents inner-ci ty residents from following employment to the suburbs (Kain 1968).

The conclusions of mismatch theorists have been disputed by others who claim that job discrimination is a more forceful cause of inner-city unemployment, outweighing any effects due to geographic isolation (Price and Mills 1985). Others claim that jo b loss, an overall loss of manufacturing jobs which historically provided good entry-level opportunities, is responsible for unemployment in the inner-city (Bluestone and Harrison 1982). Finally, some have concluded that "job restructuring"--th e increased use of part-time and temporary workers (flexible labor3)-- has caused some workers to fall into poverty because temporary jobs are more likely than permanent jobs to be low-paid, intermittent, and offer few employer benefits and car eer ladder opportunities (Applebaum 1992).

The roles of job mismatch, job discrimination, and job loss in explaining inner-city employment problems have been extensively researched. Using data on industrial location and on individual labor market outcomes, researchers have documented the drift of jobs from central cities to suburbs (Wilson 1996; White, Thomas and Thompson 1995; Persky 1990); racial disparities in firm location (Carlson and Theodore 1997, McMahon 1990) and hiring practices (Kirschenman and Neckerman 1991); and the loss of manuf acturing jobs overall (Knudsen 1994, Markusen and Carlson 1989). But the role of job restructuring, which includes an increased use of flexible labor arrangements, has been examined less thoroughly.

The shortage of research on the subject of flexible labor is due in part to the fact that it is a relatively recent trend. More importantly, however, the phenomenon is difficult to study with existing secondary data. Aggregate data, which describe th e location of jobs by industrial sector and location, do not include information on individual job characteristics--such as occupation, wages and benefits, period of employment, and hours and weeks worked--which are necessary to estimate the prevalence an d growth of part-time and temporary work. For example, state Employment Security databases; County Business Patterns data from the U.S. Bureau of the Census; and Employment, Hours and Earnings data from the Bureau of Labor Statistics report the aggregate numbers of employment, hours and wages, but since the numbers are totals for industries, it is impossible to determine the structure of individual jobs. For example, it is impossible to determine whether five people employed for 100 hours breaks down in to five 20-hour jobs, or one 40-hour job and four 15-hour jobs.

Census data, which do include occupational information for individuals, are inadequate for several reasons. First, they are available only on a 10-year basis and so do not allow researchers to examine trends easily. Second, because they report on job holder rather than job characteristics, the reasons for permanent, full-time or flexible status of the jobholder are unknown. For example, was the job permanent or full-time, and then became flexible, or did the person change from one type of job to the other for some reason? The most revealing research on the use of part-time and temporary labor has utilized firm interviews and case studies (Kanter 1995, Theodore 1993). Such firm interviews allow the researcher to examine inter-firm tre nds, and to collect the detailed information necessary to understand the nature of jobs within individual firms.

This study uses a firm-interview methodology to examine the extent to which firms in the Milwaukee metropolitan area have changed their hiring strategies since the early 1980s, focusing especially on the use of part-time and temporary labor. First, a review of the literature on the increasing role of flexible labor arrangements provides a context for the study. The methodology is then discussed. Finally, the interviews are summarized along with findings and conclusions.

The Growth in Flexible Labor


Prior to the 1970s, most large-scale employers hired and trained workers for full-time employment and supplied them with employer-based benefits such as health insurance, vacation and sick leave, and pension plans. The majority of these firms utilized human resource departments to hire workers for open positions with the intention of employing these persons for a significant part of their working careers. Workers expected to have the option of staying with the same firm until being offered a career op portunity in another firm or until retirement. However, increasingly this situation is no longer the case. Firms now exhibit a trend toward hiring individuals on a more flexible (part-time or temporary) basis which results in a more conditional employme nt status for the employee, including less job security, lower wages and benefits, and fewer career-ladder opportunities (Freedman 1985, 1996). Some research suggests that this trend reflects an overall strategy on the part of employers to shift the bur den of market fluctuations from themselves to those who supply inputs (Golden and Applebaum 1992, Segal 1996). Trends in flexible labor arrangements and "just in time" manufacturing shift the risk of market fluctuations from the employer to the suppliers of inputs, including households which supply labor.

Although the increased use of both temporary and part-time employment may result from employers’ attempts to create a more flexible labor force, it is important to differentiate between the two types of employment and the different employer-employee re lationships. Part-time work may include a reciprocal notion of permanence and loyalty, along with some employer-paid benefits, while temporary employment comes with no guarantees, and shifts more of the risk of hiring from the firm to the temporary emplo yment agency and the worker. These two sectors have also exhibited vastly different growth trends in the last two decades. Research indicates that the growth in temporary employment has far outpaced increases in part-time employment, although part-time employment is a much larger share of the total workforce (Segal 1996). Temporary employment grew roughly at 11% per year over the past two decades, while part-time employment kept pace with the growth in total employment, about 2.1% annually. This study focuses mainly on the status of temporary (not part-time) work in Milwaukee.

Polivka (1996a) profiles contingent (temporary) workers as identified in the February 1995 supplement to the Current Population Survey (CPS). She measures contingent workers with three different definitions. The first consists of wage and salary work ers who expect their jobs to last less than one year for economic reasons. The second definition adds the self-employed and independent contractors to the first definition, while the third removes the one year constraint and includes all workers who expe ct their jobs to end in the future, no matter how far in the future. Unfortunately, the CPS supplement for the Milwaukee Metropolitan Statistical Area (MSA) does not contain enough observations to allow direct comparison with Polivka=s national figures. However, taking a suggestion from Segal and Sullivan (1997), we have estimated temporary workers in the Milwaukee MSA by counting those with part-year employment as reported on the 1990 Public Use Microdata Sampl e (PUMS) 5% sample. We define temporary employment in this way for three areas in the Milwaukee MSA: the inner-city of Milwaukee, the remainder of the city (the city less the inner-city), and the MSA less the city (the suburbs). From these estimates we have calculated a location quotient, whereby the race and gender percentages for each kind of workforce (temporary and non-temporary) are standardized by that race’s or gender’s representation in the total workforce (Table 1). For example, a value of 1.0 3 in Table 1 indicates that male workers in the inner-city are only slightly more represented in the non-temporary labor force than they are in the total labor force residing in the inner-city (the underlying data indicate that they comprise 52.3% of the total inner-city labor force and 53.8% of the non-temporary labor force).


Employment Concentration Ratios for Temporary and Non-temporary Workers

Milwaukee Metropolitan Statistical Area, 1990

By Race, Sex, and Area


Milwaukee Metropolitan Statistical Area




of City


of MSA












































Source: Milwaukee MSA data are author’s calculations using the 5% PUMS sample, 1990. See text for interpretation of ratios.

These local estimates match Polivka’s national data, in that, overall, minority workers are more heavily represented in the contingent workforce than they are in the noncontingent workforce (minorities comprise 13.3% of the contingent workforce but onl y 10.5% of the noncontingent, for Polivka).4 Captured by these estimates is the striking difference by race for workers residing in Milwaukee’s inner-city (a disparity not revealed by Polivka’s national summary). African-American workers are much less likely to be part of the permanent labor force than they are to be temporary. The data show clearly that inner-city white workers are much less likely to be employed in temporary jobs than are their African-American neighbors. The same situati on holds true in the suburbs. In fact, the data indicate that the temporary labor force in the suburbs is disproportionately female and minority (as compared with their representation in the total labor force).5 The employment concentration r atios by race and sex appear to be most balanced in the area described earlier as Milwaukee city less the inner-city.

Polivka also measures contingency rates within occupations (the percentage of all workers in an occupation who are temporary workers). In Table 2 we compare Milwaukee area estimates of temporary work with Polivka’s national estimates. Our method of e stimation gives higher contingency rates across all occupations, although both studies reflect similar trends (e.g., the executive, administrative and managerial support occupation has one of the lowest contingency rates for most areas, as it does for Pol ivka). The higher estimates given here can most likely be attributed to three factors. First, the estimates given here are based on Apart-year@ work [working less than 4 8 weeks per year (36 for teachers)]. Yet seasonal employment, or jobs with several starts and stops, can be part-year but permanent. In addition, the CPS supplement used by Polivka is based on expectations of a job’s continuation; reality may be differe nt from expectations and our higher numbers reflect actual, not expected, job status. Finally, and perhaps most importantly, Polivka was able to adjust for workers whose job situations were temporary by choice. The estimates here include all those who f ound themselves working part-year, for both personal and economic reasons, thereby including workers who left jobs although the job itself was not temporary.

In Milwaukee, inner-city workers are more likely to be part-year workers across all occupations than are their city and suburban counterparts (Table 2). Administrative support (including clerical) and service occupations have among the highest conting ency rates for all Milwaukee areas6, with these proportions reaching one-third of all workers in the inner-city. One notable difference between Milwaukee figures and the national estimates can be observed for precision production, craft, and r epair workers in Milwaukee. While on a national level such workers had about the average contingency rate, in Milwaukee they experienced below-average rates of temporary employment. In fact, in the inner-city, the contingency rate for precision producti on, craft and repair workers was lower even than the contingency rate for executive, administrative and managerial employees. This most likely results from Milwaukee’s historically high manufacturing employment rate, combined with high unionization rates .


Contingency Rates for Occupations


Milwaukee Metropolitan Statistical Area



Remainder of City

Remainder of MSA


Executive, administrative and managerial





Professional specialty





Technicians and related support





Sales occupations





Administrative support, including clerical





Service occupations





Precision production, craft and repair





Operators, fabricators and laborers





Farming, forestry and fishing





Total, all occupations







The preceding section considered the extent to which workers hold temporary and part-time jobs, but the issue of contingent labor can also be examined by looking at the growth of firms that supply such labor. Such firms are known collectively as t he "help supply industry" (in the Office of Management and Budget’s Standard Industrial Classification Code scheme, this industry is SIC Code 7363). The growth in the help supply industry in the Milwaukee MSA has mirrored its growth in the nati on (figure 1). Although this industry includes employee leasing firms and employment agencies, which are different from temporary employment firms, national data suggest that the bulk of SIC 7363 is made up of temporary help agencies (Segal 1997). Emplo yment in this sector in Milwaukee grew from 4,361 in 1982 to 20,224 in 1996, mirroring the national annual growth rate of about 11% per year. Help Supply Services in the nation grew from about 510,000 in 1982 to about 2,500,000 in 1995.

Figure 1 here

It is possible to look at the geographic breakdown of such employment within the Milwaukee MSA since 1991.7 The Milwaukee MSA is subdivided into four areas: the city, older inner-ring suburbs (all located in Milwaukee county), newer outer-r ing suburbs (formed after 1950, primarily outside Milwaukee county), and exurbs (freestanding communities that were incorporated before 1950, have historically had an independent economic base, and have only recently been more fully integrated into the ec onomic life of the metropolitan area). Figure 2 indicates that employment in Help Supply Services is concentrated in the city and in the old suburbs--essentially Milwaukee county. However, between 1991 and 1996, the share of employment in this sector gr ew in the city, while the suburbs surrounding the city saw their share erode in favor of the newer suburbs outside Milwaukee county. Help Supply Services employment in the city grew from 4,689 in 1991 to 8,499 in 1996 (81% growth) and from 1,005

Figure 2 here

Spatial Distribution of Help Supply Services Employment

Milwaukee MSA

to 2,674 (166% growth) in the new suburbs. No doubt the growth in the city’s share is due to the increased use of help services in the central business district. Older suburbs are losing share (Help Supply Services grew by only 40% here) as em ployment decentralizes to the periphery -- a process all older "rustbelt" cities have experienced since the post-war years.

These figures in the last two paragraphs reveal that minority inner-city workers experience higher contingency rates overall and are much more likely than their white counterparts to experience part-year employment. In addition, the temporary help ind ustry is gaining employment in the city.


Methodology: Firm Interviews

One of the most important questions regarding the growth in temporary help agencies concerns whether this growth is due to worker preferences or to changes in firm hiring practices. Some studies suggest that the growth is related to changes in the way firms organize their workforce--they now hire more temporary employees (Lapidus 1989, Carre 1992, Golden and Applebaum 1992). Others have proposed that the growth may have come about, at least in part, because firms now use outside agencies to do their hiring in much the same way they now subcontract for building maintenance and other formerly in-house services (Polivka 1996b). In other words, temporary help agencies are now performing tasks once handled by in-house human resource departments: hiring both temporary and permanent workers as needed. Here we attempt to answer this question --are jobs more temporary or is it just that temporary firms do the hiring?--by interviewing firms regarding changes in their hiring practices since the early 1980s. The methodology consists of firm interviews with fifteen human resource directors in selected Milwaukee metropolitan employers.

Such case study research is needed in order to complement the investigation of aggregate trends which we and others cited here have conducted (Polivka 1996a, 1996b; Segal and Sullivan 1997, etc.) The complementary nature of research that identifies ge neral trends and that which studies specific cases has been addressed by Massey and Meegan with respect to work done by Sayer and Morgan (1985). They point out that specific case studies can highlight general trends identified by research which looks at aggregate data. Such case study research can reveal the need for a variety of public policy responses, the variety of which may have been concealed by analysis of aggregate trend data.

In addition, firm interview methodology has long been in use by researchers probing questions concerning firm location decisions, and their links to the direction of the overall economy. The focus of these previous works was on factors leading to firm decisions to locate or expand in particular states or sites. Researchers hoped that this line of inquiry would reveal a set of general factors that could account for overall regional or local development. More recently, Markusen (1994) proposed that th e corporate interview be used to answer larger area development issues by examining specific firm attributes more closely, such as methods of decision-making, product development and production processes. Markusen proposed that, with a careful research d esign, hypotheses regarding the causes of regional growth and decline can be tested using key informant interviews. As part of her design, she conducts aggregate-trend research to identify important local industries, from which key informants are selecte d for interview (case studies). She argues that these case studies are necessary, because the decisions of major employers are often the key to local economic conditions.

Findings from Milwaukee

Personal interviews were conducted with fifteen Human Resource Managers at large local area companies regarding their hiring strategies and how their use of temporary labor has changed over the past fifteen years. Companies represented the general mak e-up of the Milwaukee economy, with an emphasis on large firms (all firms had at least 500 employees). Although promises of confidentiality prohibit us from revealing the names of these firms, we can report that all firms had employment levels (total of all branches) of at least 500 employees and represented the following sectors: three from Finance, Insurance and Real Estate; two from Construction; six from Manufacturing; one from Distribution/Wholesale; one from Retail/Entertainment; one from Utiliti es; one from Other Retail (Grocery). The following observations are based on interviews with all but the Utilities and Distribution/Wholesale firms (we wanted to limit our observations to only those sectors in which we were able to interview at least two firms).

Financial Services

First, two of the three financial services firms reported that they have shifted their recruitment and hiring of temporary labor from in-house personnel departments to temporary agencies. Neither said that their hiring of temporary workers had increas ed; instead, it was a matter of downsizing their own human resources staff and shifting the burden of recruiting and screening to a temporary agency.8 For one of these firms, the temporary firm actually had a sat ellite office located in the financial services firm itself. Each estimated that at any one time about 10% of their employees are temporary, and that these workers are filling in for permanent personnel who are home with family responsibilities, or are f illing in while permanent employees are being recruited. The remaining financial services firm has an explicit labor strategy which involves a permanent, mostly full-time staff.

Both of the financial services firms which have increased their use of temporary labor stated that, historically, they have always hired some temporary labor because of the nature of their primary labor pool, which consists chiefly of women who are &qu ot;secondary" earners within the family unit. These women fill the multitude of clerical and administrative support positions common to financial services companies. These human resource personnel maintained that they have always relied on temporar y labor (more so than other kinds of firms) because their primary labor pool is subject to more reasons for short-term leaves (pregnancy, caring for sick family members, etc.) and higher turnover rates which leave unfilled permanent positions. In these c ases, therefore, temporary workers are used both to fill short-term openings, where the jobholder is expected to return; and as a pool of "probationary" labor, which may be eventually asked to permanently fill an open position.

The third financial firm reported that they have never relied heavily on temporary labor. They have instead followed a strategy where workers are permanent employees, whether full-time or part-time, because they believe this engenders loyalty and trus t on the part of employees, and results in higher productivity in the long run. Although they also stated that their primary labor pool is second-earner women, they have followed a strategy of hiring a pool of permanent part-time employees whose schedule s contract or expand depending on the fluctuation in leaves and departures among the full-time staff. For example, if a full-time administrative assistant were to take maternity leave, two part-time employees would be asked to fill her position until she returned. Part-timers are also moved into full-time positions as they become open. In order to demand this flexibility on the part of part-time employees, the firm must be committed to a full range of training for its part-time employees, a commitment the interviewee affirmed. The firm seldom relies on temporary agencies to fill vacant positions, even for the part-time openings, because they prefer the control over recruiting that in-house screening affords.


The employment position for the construction firms is different from that of the financial services firms in that the work schedule and need for employees fluctuates depending on the season and the public funding for infrastructure and large projects. Hiring strategies for construction firms therefore must be viewed in this context. In essence, except for a core cadre of long-term employees, the industry has always relied on a pool of "temporary" labor, and has found these workers primarily through the trade unions. However, the human resources interviewees at these firms did indicate that their pool of contingent labor has declined in recent years, as public funding for construction of major public works has been relatively generous, and the construction workforce has been aging and retiring. The ability to offer longer-term employment in light of more public spending, and the need to replace an aging workforce has allowed these construction firms to move to a more permanent workforce. At the same time, union membership has fallen off somewhat, and unions themselves have found a need to do more recruitment.

The construction firms have each begun to supplement their union relationships with contacts at community agencies, including the YWCA and other local training and placement agencies. These agencies do much of their worker screening. As their pool of labor dwindled, rather than add staff to their own human resources department to do recruitment and screening, they have instead developed relationships with local job developers to serve as their labor-market intermediaries. The construction firms inte rviewed here said that help supply services firms probably would not have the kind of labor for which they are looking, because their impression is that construction workers do not use help supply services firms.


The characterization of temporary help agencies chiefly as suppliers of clerical and office work held also for the manufacturing firms. In general, these firms were clear about turning to temporary agencies for clerical and administrative aide help, b ut knew that these agencies in general could not supply the production-line workers they might need during temporary upturns. In addition, manufacturing firms find themselves in a different economic environment from that of financial services and constru ction businesses. While financial services employment has been expanding in the nation and in Milwaukee, and construction employment varies seasonally, manufacturing employment plummeted during the 1980s and continued to stagnate in the 1990s. Manufactu ring firms, therefore, find themselves with a ready supply of "temporary" labor in the form of employees laid-off from their own or other manufacturing plants, and so do not need to find other sources of contingent workers. This explains why th ey make less use of temporary agencies even though their employment strategies may include more temporary and part-time work. In their use of temporary agencies to fill clerical and administrative help positions, these firms reported that about half the time the employees were truly "temporary" while the other half were being screened for "temp to perm" positions (temporary positions that would turn into permanent employment).



Finally, the retail firms reported that they have historically always relied on part-time workers in the form of high school and college students, housewives, etc. They reported no increase or decrease in the percentage of such part-time work, but bo th firms have seen overall growth, so that the sheer number of part-time workers has increased. Neither reported using temporary agencies to find potential applicants.


These interviews serve to raise questions about the nature of the growth in temporary employment. The findings suggest, as Polivka (1996b) proposed, that to some extent the growth in help supply services employment represents an sub-contracting of fun ctions that were once performed in-house. The growth in employment in temporary agencies may not entirely represent an increase in the amount of temporary work so much as it represents an outsourcing of the formerly in-house function of hiring both temporary and potentially permanent employees. This is not to suggest that all of the increase in help supply services employment is due to this sub-contracting; rather, some of what we have seen may be due to the shifting o f the hiring of labor from in-house human resource departments to temporary help supply firms, using the "temp to perm" process.

And this shift has occurred primarily among financial services firms and in female-dominated occupations such as clerical and administrative help. Both the manufacturing firms and financial services firms reported they turn to temporary agencies to su pply this help. New workers needed for production work (in manufacturing firms) can be drawn from a separate pool of workers-employees on layoff. The growth in temporary help can also be seen as a "gendered" phenomenon, primarily affecting wom en and the occupations they dominate.

Secondly, these interviews suggest that when we use the term "temporary labor," we must make a distinction between a position that is truly temporary and one that is probationary. As we saw for the financial services firms discussed above, some temporary help is used to fill short-term vacancies, when the jobholder is on leave, while in other situations a temporary employee is really being "tested" to fill an open permanent position. Whether hired by a temporary help firm or by a n in-house personnel department, employees in these two categories of temporary workers have different prospects for long-term permanent employment at the firm. However, shifting the hiring of the "probationary" temporary labor from the firm to a temporary agency heightens the possibility, for the worker, that the position will not become permanent. Because the temporary agency has economies of scale for screening and recruiting employees that the firm does not, the firm may elect to cycle thr ough several probationary employees before deciding on one.

Finally, although none of our firms stated they increased their use of temporary workers, merely shifting the screening and recruiting process to temporary agencies may have other effects that our research so far cannot ascertain. Two questions come t o mind. First, what is the real effect on the employer/employee relationship in the long run? Fndings from interviews with temporary help agencies will shed light on this question. In addition, what has been the effect on worker job search methods? In formal interviews with persons who use temporary agencies has suggested that newspaper ads and "cold calling" are no longer as fruitful as they once were. The word "on the street" is that to find a job you must use a temporary help ag ency.


Applebaum, E. (1992). Structural Change and the Growth of Part-Time and Temporary Employment. In V. L. DuRivage (Ed.), New Policies for the Part-Time and Contingent Workforce. Economic Policy Institute Series. Armonk: M.E. Sharpe.

Bluestone, B. & Harrison, B. (1982). The Deindustrialization of America. New York: Basic Books.

Carlson, V. L. & Theodore, N. C. (1997, March). Employment Availability for Entry Level Workers: An Examination of the Spatial Mismatch Hypothesis in Chicago. Urban Geography.

Carre, F. J. (1992). Temporary Employment in the Eighties. In V. L. DuRivage (Ed.), New Policies for the Part-Time and Contingent Workforce. Economic Policy Institute Series. Armonk: M.E. Sharpe.

Freedman, A. (1996). Contingent Work and the Role of Labor Market Intermediaries. In G. Mangum & S. Mangum (Eds.), Of Heart and Mind: Social Policy Essays in Honor of Sar A. Levitan. Kalamazoo, Michigan: W.E. UpJohn Institute for Employme nt Research.

Golden, L. & Applebaum, E. (1992). What Was Driving the 1982-88 Boom in Temporary Employment?: Preference of Workers or Decisions and Power of Employers. American Journal of Economics and Sociology, 51(4), 475-493.

Hanson, S. & Pratt, G. (1995). Gender, Work, and Space. New York: Routledge.

Kain, J. F. (1968). Housing Segregation, Negro Unemployment, and Metropolitan Segregation. Quarterly Journal of Economics, 82, 175-192.

Kanter, R. M. (1995). Nice Work if You Can Get It. American Prospect, 23, 52-59.

Kirschenman, J. & Neckerman, K. M. (1991). We=d Love to Hire Them, But...: The Meaning of Race for Employers. In C. Jencks & P. E. Peterson (Eds.), The Urban Underclass (pp. 203-34). Washingto n, D.C.: Brookings Institution.

Knudsen, D. C. (1994). Deindustrialization of the U.S. Midwest, 1965-1985. In C. F. Bonser (Ed.), Dimensions of Change: Decline and Restructuring in the American Midwest. Indiana University: Institute for Development Strategies.

Lapidus, J. (1989). The Temporary Help Industry and the Operation of the Labor Market. Unpublished doctoral dissertation, University of Massachusetts-Amherst.

McMahon, W. F. (1990). An Analysis of Intrametropolitan Occupational Change: Milwaukee, Wisconsin 1979 to 1987. Unpublished report, Center for Urban Initiatives and Research, University of Wisconsin-Milwaukee.

Markusen, A. R. (1994). Studying Regions by Studying Firms. Professional Geographer, 46(4), 477-490.

Markusen, A. R. & Carlson, V. (1989). Deindustrialization in the American Midwest: Causes and Responses. In L. Rodwin & H. Sazanami (Eds.), Deindustrialization and Regional Economic Transformation: The Experience of the United States. Boston: Unwin Hyman.

Mishel, L. & Berenstein, J. (1995). The State of Working America, 1994-95. Economic Policy Institute.

Nelson, K. (1984). Labor Demand, Labor Supply and the Suburbanization of Low-Wage Office Work. In A. J. Scott & M. Storper (Eds.), Production, Work, Territory: The Geographical Anatomy of Industrial Capitalism (pp. 149-169). Boston: Allen Unwin.

Persky, J. (1990). The Economic Interdependence of City and Suburb. In L. B. Joseph (Ed.), Creating Jobs, Creating Workers: Economic Development and Employment in Metropolitan Chicago. Chicago: University of Chicago Press.

Polivka, A. E. (1996a). A Profile of Contingent Workers. Monthly Labor Review 119(10), 10-21.

Polivka, A. E. (1996b). Into Contingent and Alternative Employment: By Choice? Monthly Labor Review 119(10), 55-74.

Price, R. & Mills, E. (1985). Race and Residence in Earnings Determination. Journal of Urban Economics, 17, 1-18.

Sayer, A. & Morgan, K. (1985). A Modern Industry in a Declining Region: Links Between Method, Theory and Policy. In D. Massey & R. Meegan (Eds.), Politics and Method: Contrasting Studies in Industrial Geography. London: Methuen and Co .

Segal, L. M. (1996). Flexible Employment: Composition and Trends. Journal of Labor Research, 17(4), 543-561.

Segal, L. M. & Sullivan, D.G. (1997). The Growth of Temporary Services Work. Journal of Economic Perspectives, 11(2).

Theodore, N. C. (1993). The Employment Potential of Inner-City Enterprise: Results from a Survey of Grand Boulevard Area Businesses. Chicago: Chicago Urban League.

Wilson, W. J. (1987). The Truly Disadvantaged: The Inner City, the Underclass, and Public Policy. Chicago: University of Chicago Press.

Wilson, W. J. (1996). When Work Disappears: The World of the New Urban Poor. New York: Knopf.

White, S. B., Thomas, M. M. & Thompson, N. A. (1995). Changing Spatial Patterns of Employment Location: Milwaukee, Wisconsin, 1979-1994. Unpublished report, Center for Urban Initiatives and Research, University of Wisconsin-Milwaukee.


  1. Unemployment estimates are from the Wisconsin Department of Workforce Development; inner-city rates are from independent survey data acquired by the University of Wisconsin-Milwaukee Center for Urban Initiatives and Research.
  2. In this paper, the term "inner-city" is used to denote the core areas of major cities, usually characterized by poverty and minority populations, while "central city" is used to denote the major city of a metropolitan area. That i s, Milwaukee is the central city of the Milwaukee metropolitan area (which includes Washington, Ozaukee, Waukesha and Milwaukee counties), while the inner-city includes neighborhoods west of the river and south of downtown, such as Harambee, North Divisio n, and the Menomonee River Valley.
  3. Although the literature sometimes refers to part-time and temporary work arrangements as "contingent," here the term "flexible" is used and the use of "contingent" is reserved specifically for temporary arrangements, foll owing Polivka (1996a).
  4. The analysis performed in Table two cannot be directly compared to Polivka (1996a) because she does not report race or gender percent of total employment.
  5. Although, because of extreme housing segregation in Milwaukee, the African-American portion of the suburban labor force is only .5%.
  6. We suspect that for Polivka, professional specialty occupations have a higher contingency rate because of the different way she counted teachers.
  7. Because of a change in the way firms reported locations beginning in 1991, small, sub-city geographical comparisons cannot be made to pre-1991.
  8. The two largest temporary agencies in the Milwaukee MSA are Manpower, Inc., and Olsten Staffing Services.