Printer Friendly

Requiem or rebirth? Internal labour markets and labour market restructuring in the Kitchener and Sault Ste. Marie regions.


Since the 1980s, social scientists and economic geographers have examined how 'standard' forms of employment have been displaced by non-standard arrangements such as part-time and temporary employees (e.g., Atkinson 1987; Noyelle 1987; Betcherman 1995; Allen and Henry 1997; Carnoy et al. 1997; Harrison 1997; Peck and Theodore 1998; Christopherson 2002; Theodore and Peck 2002). Underlying much of this literature is the argument that these changes represent a secular, if not epochal, shift towards a post-Fordist regime with declining employment tenures and the dissolution of internal labour markets (ILMs) in favour of contingent workers recruited on the external labour market (ELM) (e.g., Allen and Henry 1997; Christopherson 2002).

However, other observers are critical of the assumptions of a universal shift in firm employment strategies and stress continuing significant national differences in employment tenures and standard/non-standard employment arrangements (Bowers and Martin 2000). As importantly, narratives of a 'golden age' of secure employment often ignore the majority of women and people of colour who laboured under highly insecure conditions (Smith 1999; McDowell 2002). Other researchers argue that in the United States, there may be structural and spatial limits to the extension of temporary and contingent work; they maintain that although ILMs have been significantly restructured, they have by no means been replaced by contingent work arrangements but remain an important, if not central, part of many firm employment strategies (Moss et al. 2000; Theodore and Peck 2002).

In this paper, I focus on the contemporary role of ILMs. The paper is based on a postal survey and interviews conducted with ninety firms and institutions in Kitchener-Waterloo and Sault Ste. Marie, Ontario, in 1995-1996 that assessed changes to firm employment practices during the first half of the 1990s. This period was characterized by severe economic recession in the Canadian and especially Ontario economy followed by a slow recovery in output and employment (see Rutherford 1996; Gertler 1999). Although the survey reveals important changes in employment practices towards nonstandard arrangements, it also shows that employers are clearly aware of the need to maintain if not develop ILM structures.

I also argue that geographers need to reconceptualize the relationship between ILMs and ELMs. Geographers have often focused on contingency and ELMs as phenomena analytically separated from the ILM and firm work organization. Employment contingency is driven primarily by lower labour costs and more permissive regulatory environments (see Allen and Henry 1997; Harrison 1997). There has been less attention to the role ILMs play in the organizational integration required for innovation and other firm goals. Following Grimshaw and Rubery (1998), my central argument is that firms do not use external/ contingent labour markets and ILMs in exclusion to each other but rather as part of a continuum of strategies which is leading to increased blurring of the ILMs and ELMs and a renewed emphasis on social and cultural skills. Second, rather than a simple ILM/ELM dichotomy, restructuring is leading to a number of different ILMs, each with its own relationship with ELMs. Finally, although less directly exposed to local labour market conditions, ILMs are not 'space-less' and through what Manwaring (1984) terms the extended internal labour market (EILM) are often strongly linked to the local labour market. Indeed, ELMs continue to be a determinant of the substitutability of labour within ILMs in ways that cannot be completely controlled by firms. Thus in both Kitchener and Sault Ste. Marie, geography continues to be an important determinant in ILM-ELM change, even though restructuring may have an eroded, some of the more distinctive heartland-hinterland employment contrasts which characterized their labour markets before 1980.

Flexibility, Risk and the Demise of the Internal Labour Market?

Throughout the 1970s and early 1980s, ILMs were the object of intense theorizing and empirical research (Doeringer and Piore 1971; Osterman 1984; Marsden 1986). Classic dualist theories of labour market segmentation viewed the ILM as linked to differences in economic structure between large oligopolistic primary firms and smaller firms subject to more intensely competitive market conditions. Primary sector firms and the public sector sought to lower turnover and develop needed skills through strong ILMs for both white-collar professional employees and blue-collar workers in unionized firms. Seniority was the main criterion for employee advancement up vertical job ladders, and firms only hired externally if they failed to fill vacancies internally. This practice had significant gender, racial and class implications, because ILMs were dominated by white male workers, whereas women and workers of colour were either occupationally segregated or overwhelmingly employed by secondary sector firms (Milkman 1983). Thus radical or second-generation labour market theorists in the United States viewed ILMs as an institution of labour control designed to segment an otherwise increasingly homogenous industrial workforce (Peck 1996).

ILMs also posed a significant challenge to neoclassical economic theory that claimed that wages were narrowly determined by accumulated human capital and supply and demand market conditions. ILMs' bureaucratic procedures and use of seniority and custom and practice in determining wages, conditions and advancement, diluted the effect of human capital and provided alternatives to the wage-clearing mechanism for firms (Peck 1996, 55). Moreover, this impact was not simply confined to firms utilizing ILMs but influenced medium term wage determination and overall levels of employment in ELMs. Thus broader institutional processes were seen as more determinant of labour market outcomes than markets.

However, recently, with some exceptions (e.g., McDowell 1997; Grimshaw et al. 2001) social scientists and economic geographers have become less interested in the ILM as an object of study. Instead, attention has focused on flexible employment strategies and the increased outsourcing of formerly secure employment to contingent labour markets. The decline of ILMs is viewed, as both supply and demand led. Noyelle (1987) links this trend to the availability of a more educated workforce such that skilled workers can be obtained on ELM, whereas Harrison (1997) sees contingent employment resulting from firms taking advantage of increased unemployment, the weakening of unions and employment de-regulation to lower labour costs. For Christopherson (2002, 10), the development of contingent work arrangements is due largely to capital markets which emphasize short-run profit maximization and rapid flows of capital. This regime is facilitated by relatively high quit rates by employees, the use of labour market intermediaries (see Benner 2003) and the development of project-based production systems which do not require longer-term employment strategies or a developed ILM. (1)

Employment outsourcing also means that segmentation is no longer based primarily on sector (i.e., primary/secondary) but between core and periphery workers (Atkinson 1987; Noyelle 1987; Christopherson and Noyelle 1992). Thus Harrison (1997) argues that employment regimes in the United States are characterized by a 'new dualism [which] differs from the original version in at least one important way: even high level jobs are no longer secure. In the age of flexibility, even the most profitable big firms are inclined to shed even white collar employees. The shrinkage in the number of safe, stable, secure occupations seems to be keeping pace with the big firms' downsizing' (Harrison 1997, 199). This new dualism increasingly characterises also significant segments of public sector employment (see Reimer 1999) and a growing share of US manufacturing jobs (Peck and Theodore 1998; Theodore and Peck 2002).

It is not surprising, given these trends, that Cappelli et al. (1997) find that the breakdown of 'traditional methods of managing employees and developing skilled workers inside companies ...' This trend, they argue, reflects 'pressures from product and labor markets [which] are brought inside the organization ... [establishing] market-mediated employment relations' (Cappelli et al. 1997, 4). Such developments also challenge theory for 'the current fragmentation and weakening of traditional socio-political practices in the labour market appears in conflict with economic theories which stress the role of institutional structures in determining patterns of pay and employment' (Grimshaw and Rubery 1998, 199).

However, recently, there has been a growing critique of this focus on contingent employment. For Smith (1999), this focus over-estimates both Fordist employment security and the extent to which workers are now less secure. As importantly, Moss et al. (2000) note that some aggregate indicators of ILMs such as job tenure and estimates of the firm-specific component of wages in the United States have not changed significantly over the last twenty years. They argue that such findings are not necessarily at odds with increased contingent employment but are 'consistent with a world in which firms alternately tear down and rebuild internal labor markets, averaging out to little change' (Moss et al. 2000, 96), and ILMs are still required to meet both employees' needs for job security and employers' needs for predictable supplies of adequately skilled workers. Furthermore, the employment of contingent workers does not necessarily lead to a reduction in labour costs nor, as Theodore and Peck (2002) emphasize, are ILMs necessarily less flexible than contingent ones. Finally, contingent employment arrangements can be at odds with the needs of firms for organizational integration of the workforce for learning/innovation (Lazonick 1991; Lowe 1999). Even high-tech firms in Silicon Valley have found that turnover rates of highly skilled employees are above optimal levels and have developed more internal training and compenzation packages to keep key employees (Carnoy et al. 1997, 8; Hudson 2001).

Although still relevant to firms, ILMs have been significantly restructured. In the United Kingdom, Grimshaw et al. (2001) see traditional ILM structures replaced by a greater reliance on market solutions. However, Salzman et al. (1998) and Moss et al. (2000) argue that after downsizing, many firms begin to organize for growth and innovation and often need to rebuild ILMs. Reconstituted ILMs differ from earlier ILMs, because jobs have often been redesigned into broader functional categories that eliminate highly graded hierarchies (Lovering 1990). Job movement is more likely to be horizontal than vertical, because 'skill barriers to entry are greater and the gaps between job functions are greater, but skill development and responsibilities as well as wage progression are also greater within each broad functional area' (Salzman et al. 1998, 17-18). Similarly, although ILMs have long been segmented with restricted access, current restructuring is also exclusionary and polarizing (see Yates 1999 on the automobile industry). Thus lower-skilled workers are often given little access to training or advancement, or their jobs may be outsourced or eliminated altogether (Salzman et al. 1998; see also Carnoy et al. 1997; Lundvali 2001).

Yet, the most important development is that although ILMs have always had some social or cultural skill component (Blackburn and Mann 1979), they have become increasingly important, organizationally driven and based on firm-specific knowledge (Thrift 1997; Salzman et al. 1998; McAdam and McCreedy 2000). In addition, 'knowing how to get on' is critical (Hudson 2001, 111; McDowell 1997). This imperative reflects factors both within and outside the firm. The first is that leading edge firms, especially in the growing service sector, are increasingly attempting to develop individual and especially group, continuous learning environments, which demand 'soft' communication skills (Lundvall 2001; Tomlinson 2002). Second, many technical skills are often available on the ELM because of overall rising educational standards and downsized skilled workers and thus perceived socio-cultural abilities become more decisive to firms when recruiting (Saizman et al. 1998).

The continued viability of ILMs and the enhanced role of socio-cultural skills indicate a need to re-theorize classic dualistic theory and contingent employment research which often assumes a static division between ILMs and ELMs. Instead, Grimshaw and Rubery (1998, 200) argue that ILMs and ELMs should be viewed as integrated entities, because 'internal and external competitive pressures mutually interact to shape employer strategy and the labour market positions of employees'. Drawing on Osterman (1994), they argue that labour market change is based on the interaction between technology and competitive strategy; customs, norms and politics which are historically specific to particular firms and ELM institutions such as minimum wage legislation and social security systems. Although related, these processes do not necessarily move in synchronization. Thus internal norms and customs of the firm are often highly path dependent and structured in part by employee perceptions of fairness (Grimshaw and Rubery 1998, 215).

Although insightful, Grimshaw and Rubery pay relatively little attention to the relationship between ILMs and local labour markets. Although some economic geographers have implied that ILMs are intrinsically less geographic than other labour market phenomena (see Clark 1981; Peck 1996), ILMs have always had a link to external, local labour markets (see Manwaring 1984; Maguire 1988; Hanson and Pratt 1995). Invariably, labour must be recruited from somewhere, and labour markets are both spatially and occupationally segmented (Martin 2001; Castree et al. 2004). Local labour markets are themselves legacies of earlier rounds of investment (see Massey 1984) that means that even firms with well-developed ILMs may use local institutions and encounter local socio-cultural norms when recruiting. Stereotypes of race, gender and skill existing in local labour markets can be the basis of segmentation within the firm (Milkman 1983; Hudson 2001). Thus Maguire and Warde and Hanson and Pratt examine how social relations such as gender, skill and sectarian identities of the local labour market and ILM are mutually constitutive. For example, firms often utilize what Manwaring (1984) terms the EILMs for word of mouth recruitment of shop floor employees via existing workers, family and ethnic networks as a method of workforce control. Scale and gender also play a role, as skilled and male employees may be recruited over longer distances than female workers in semi-skilled jobs, in part because of the latter's domestic commitments (Hanson and Pratt 1995, 224). This extends to how concepts of masculinity and femininity are developed and reconstituted both within the workplace and labour market (see McDowell 1997).

Certainly, current workplace and ILM restructurings are changing linkages with local labour markets. As stressed above, firms systematically assess behavioural and social skills (e.g., Sayer and Walker 1992; Salzman et aL 1998; Hudson 2001) for internal training and are often driven by organizational integration. Technical skill training of course is not irrelevant and links with colleges and institutions of higher education may become more important as firms need colleges to provide remedial training/filtering services and technical specialties and spot training (Salzman et al. 1998, 24). Yet increased educational attainment by job applicants may be less an indicator of higher skill requirements than a signal to employers that applicants have both attained appropriate social skills and 'learned how to learn' (Crouch et al. 1999, 6). There thus exists a blurring between the ILM and the ELM, which is in part linked to a blurring of social and technical skills. Local labour markets matter in this process in that identities which are in part constructed outside of the firm, shape the way labour is used and can limit its substitutability within the firm. Equally, firms do not encounter the labour market passively, and new forms of work organization, training and occupational change can create the basis of new solidarities and divisions within a locality.

ILMs and Restructuring in Canada

Like other advanced economies, there has been significant restructuring of Canadian labour markets (Rutherford 1996). The pressure for restructuring was such that between 1990 and 1995, establishments (especially larger ones) representing an estimated 54 percent of employment in Canada downsized, whereas 66.6 percent re-engineered their operations (Statistics Canada 1998, 16). Deregulation has also made the use of contingent workers more attractive (Vosko 1998), and establishments representing 26.8 and 30.0 percent of employment, respectively, increased their use of temporary and part-time workers. As in the United States, the post-1980 period saw the erosion of ILMs and employment security for many groups of workers in both the public and the private sector (Lowe 1999; Gunderson 2002). Although average job tenures for both males and females rose during the 1990s, this trend largely reflected a markedly poorer labour market performance and reduced voluntary employee turnover (Yourck 2002). During the post-war boom, many segments of the Canadian labour force were highly insecure and contingent (Smith 1998), and the current Canadian labour market is even more flexible than the United States, with Canadian employers responding faster to shifts in demand with changes to their labour force size (Stanford 1995).

These trends suggest a weak role for ILMs with a majority of Canadian firms having a low commitment between workers and employers and to investment in training (Betcherman 1995). Canadian employers tend strongly to hire from outside of the firm for necessary skills with 74 percent preferring this method (Garlaneau et al. 2001, 27). Although this practice is partly driven by the poor record of employers' investments in training (see The Premier's Council 1990), lower training rates are more evident in establishments with no employee turnover than those with higher turnover (Leckie et al. 2001, 21).

However, aggregate surveys may not capture the true status of ILMs in Canada. Thus Betcherman (1995) finds a substantial number of medium and larger firms in business services, and some manufacturing had developed internal job ladders and above-average training investments (Betcherman 1995, 83). The recent Evolving Workplace Survey indicated that, between 1990 and 1995, many firms relied on overtime and increasing employee functionality (establishments representing 29.5 and 57.4 percent of employment, respectively), meaning that ILM strategies may have remained relevant to firms even as they downsized and outsourced employment (Statistics Canada 1998).

Finally, while in the post-1980 era, federal and provincial governments allowed the greater use of contingent workers; the period also witnessed the adoption of the Charter of Rights and Freedoms and Employment Equity legislation which gave a high profile to non-discrimination in firm hiring and promotion practices (see Bilson 1995; Smith 1998). Furthermore, the higher percentage of workers in unionized establishments in Canadian manufacturing such as auto assembly and parts compared with the United States has meant that Canadian unions have been better able to maintain Fordist ILMs (see Kumar and Holmes 1997). Unionization may not prevent the use of contingent workers, but because it often requires negotiation, firms may choose to intensify their use of full-time employees such as through overtime (Lowe 1999; Doubleday 2000).

A Tale of Two Cities: The Restructuring of ILMs in Kitchener and Sault Ste. Marie

The debate on the role of ILMs within advanced economies, including Canada, indicates the need to gain a greater understanding of firm and establishment practices. To assess this question, a survey using postal questionnaires and semi-structured interviews was conducted in 19951996 with employers, government training officials, educators and unions in two Ontario cities Kitchener-Waterloo and Sault Ste. Marie. In total, ninety questionnaire responses were received from employers in the two regions (a 20 percent response rate); the respondent firms covered approximately 11 and 26 percent of total employment, respectively, in the Kitchener and Sault Ste. Marie areas. (2) Fifty follow-up interviews were carried out with both respondents and other key actors in the local labour market.

With its diversified economic base, Kitchener is typical of 'heartland' Southern Ontario urban areas, whereas Sault Ste. Marie is more representative of single and resource-dominated communities of 'hinterland' Canada. Furthermore, in both Kitchener and especially the steel-dominated Sault Ste. Marie, ILMs and EILMs were important features of the local labour market (see Heath 1978; English 1996). During the early twentieth century in Kitchener manufacturing, labour relations were principally based on ethnic and paternalistic German networks. Unionization formalized ILM processes, and the arrival of new groups of immigrants such as the Portuguese in the post-World War II era also led to a reinforcement of the EILM. In the Sault, the EILM often drew upon the Italian community and formalized seniority structures arrived with unionization in 1946. Although recent research has challenged some of the stereotypes of hinterland and resource communities stressing their diversity, different patterns of female participation in the labour market and in and out migration (Hayter and Barnes 1992; Randall and Ironside 1996; Halseth 1999). The experience of the two regions in the first half of the 1990s may provide insights into other Canadian localities.

During the first half of the 1990s, the recession which hit the Ontario economy was clearly felt in both Kitchener and the Sault, although its impact was markedly different. Despite experiencing significant layoffs and plant closings in manufacturing, the Kitchener region long considered one of the most successful and diversified industrial localities in Canada (Walker 1987) witnessed continued growth. Between 1989 and 1994, employment grew by over 3 percent compared with a decline in Ontario of nearly 2 percent (Table 1), and Kitchener had one of the lowest unemployment rates in Canada. However, even after an overall employment recovery, the employment share for older males and women in manufacturing and women clerical occupations declined during the first half of the decade. Unemployment rates are several points higher in the older industrial cities of Kitchener and Cambridge than they are in Waterloo.

Sault Ste. Marie offers an important contrast to that of the Kitchener region. Unlike the latter's diversity of employment and like many hinterland communities, the Sault has long been highly dependent on Algoma Steel. Indeed, in 1991, it was estimated that up to 63 percent of total employment (37,810 jobs) were directly or indirectly dependent on Algoma Steel. Like many other steel companies, Algoma encountered increasing difficulties from the early 1980s onwards and was purchased by Dofasco in the late 1980s. Faced with bankruptcy in 1991/1992, an Ontario government and union backed employee buyout was negotiated. Although it stabilized the firm's financial situation, it did not prevent the continued downsizing of the firm.

The downsizing of Algoma from over 12,000 in the early 1980s to just over 5,000 by 1994 meant that total employment in the Sault fell 17 percent between 1989 and 1994 (see Table 1). The restructuring and downsizing of Algoma coincided with the arrival of large government agencies such as The Ontario Lottery Corporation. Thus there were significant increases in public administration employment--up 30.3 percent (compared with 1.6 percent in Ontario) and finance insurance and real estate employment that increased by 40 percent (compared with only 0.6 percent in Ontario). As will be explored later, these trends had significant implications for the gendering of the local labour market. Although Algoma and the resource sectors had been dominated by male employment, by 1994, there were actually more women employed in the Sault than men (17,371 vs. 16,608) (Statistics Canada Labour Force Annual Averages, 1989-1994, Cat. No. 71-529) (Table 2).

Restructuring, work reorganization and skills In the first half of the 1990s, surveyed firms and institutions came under intensified competitive pressures. During the early 1990s recession, manufacturers had restructured due to the free trade agreement with the United States, and then, NAFTA and formerly sheltered public and private services had also significantly reorganized. Thus publicly owned hospitals and other institutions encountered severe budget constraints, as the Ontario government coped with recession and deficit reduction. Financial services such as insurance were confronted with increased competition from banks and other financial institutions, as these services were de-regulated, whereas domestic food retailers were restructuring in the face of competition from large U.S. retailers such as Wal-Mart.

This intensified competition contributed to a high level of adoption of new technology and work organization in both localities. In Kitchener and the Sault, 92 percent of surveyed establishments had adopted at least one new technology (such as personal computers, CAD/CAM systems) in the last five years, whereas 82 percent of Kitchener and 64 percent of Sault workplaces had adopted at least one new work practice such as Total Quality Management, multiskilling or team working (Table 3). These innovations had an important impact on employer perceptions of technical and social skill change. Nearly 75 percent of Kitchener respondents reported a minor to major increase in social skills (for example, problem solving, communication, etc.), whereas approximately 60 percent reported minor to major increases in the Sault. In contrast, nearly 35 percent of Sault respondents reported a major increase in social skill requirements compared with less than 20 percent in Kitchener, a difference that partly reflected the major efforts to develop the cultural skills to support self-directed work teams at major private sector employers such as Algoma and St. Mary's Paper.

Technical skills (numeracy, programing ability, etc.) were perceived to be increasing. Nearly 84 percent of Kitchener respondents reported a minor to major increase, whereas in the Sault, approximately 73 percent responded in the same categories (Table 3). However, in some higher technology firms, the traditional divisions among engineering, manufacturing, marketing and design were blurring, as production lead times collapsed and the line between technical and social skills became less distinct:
   Because we are a full service company, we conceptualize
   what the problem is and we solve it. We design
   the solution, we make the solution, we install the
   solution [we need] multi-skill, multi-talent sort of
   people. One person might do three quarters of the
   whole project by being part of the design, manufacturing
   and installation team. and he's not a sales rep
   but he's also dealing with the customer ... (Kitchener
   manufacturing, 1995).

This process was also evident in more traditional manufacturing and service firms, where downsizing and re-engineering had a significant impact on needed skills. In financial service and communication firms for example, there was a greater emphasis on increasing those in direct customer contact or sales. One service company respondent described the impact of this change:
   In our field when I was hired, sales was not a big
   prerequisite, it was more clerical and accounting
   and figuring out bills for people and polite manners.
   Now sales are a very big part if we are to stay viable.
   Sales are so high on their list. Many people don't fit
   into that category now.., you have to be able to
   accept change fast and furious and be pleased with
   it (Sault communications services, 1995).

Although these data seemingly confirm arguments that contemporary restructuring is leading to increasing skill levels (see Block 1990; The Premier's Council 1990), other research sees that this is a form of task enlargement or work intensification. The Evolving Workplace Survey (1998) indicated a relatively low rate of adoption of 'high performance' work practices and gave strong evidence that workplace restructuring in the 1990s was primarily directed at work intensification through job enlargement rather than upskilling. Thus the adoption of such innovations as employee suggestion programs, flexible job design, quality circles and self-directed work teams occurred at establishments representing less than 12 percent of employment (Statistics Canada 1998, 19). The view that work re-organization was leading to work enlargement rather than upskilling was stressed by one food retailing union representative:
   With the recession, companies have rationalized
   their operations. This means that they cut off as
   many employees as they could in the hope of
   increasing production. The widening of the responsibility
   of the worker rather than the upgrading of the
   worker is what is happening. The worker has had to
   assume a lot more responsibility at the same skill
   level (Food retailing union representative, Kitchener
   region, 1995).

As stressed above, to Salzman et al. (1998), work enlargement and intensification is the result of restructuring which is largely organizationally as opposed to technologically driven. Technological change is primarily supportive of the former and is associated mostly with basic computer literacy rather than higher-level computer skills. Organizationally driven change puts a high priority on cultural or soft skill training. This suggests a need to be careful when arguing that current restructuring is leading to increasing technical skills. Although task enlargement can be associated with skill increases (Wood 1989), the most important change would seem to be a shifting of the boundaries between technical and social skills, in which due to restructuring pressures, organizational integration is driving the re-composition of skill. This view was underlined by a Kitchener area bank human resource manager:
   We are most interested in soft attributes and people
   skills [because] I don't think we can produce a product
   that somebody else couldn't have on the market
   within a month. There is no sustainable advantage
   in the technology [so] the only thing we can do is
   have better people and have better competencies to
   deliver to our customers (Kitchener region financial
   services, 1996).

Furthermore, a number of employers felt that increasing skill levels were partly an outcome of increased unemployment raising the overall available skills in the local labour market (Sault financial services; Kitchener region food retailing). As Salzman et al. also stress, rising skill levels are in part compositional, as lower-skilled workers are laid off or jobs outsourced, and thus remaining workers have overall higher skills which are augmented by the need to be trained in the jobs of those who have been downsized. This perspective was confirmed by two Sault Ste. Marie respondents:
   A lot of firms here were in the situation that they had
   do more with fewer people. That meant that the
   remaining people had to be trained formally or informally
   in doing other work (Sault Ste. Marie, provincial
   government training official, August 1995).

   When any company cuts, the people left are diversified
   and you have to improve your skills, you are
   required to do a lot more work in a lot less hours
   (Sault Ste. Marie communications services firm, 27
   August 1995).

Restructuring the internal labour market Work intensification and the blurring of social and technical skills were also evident in the restructuring of labour market strategies. Overall, evidence suggested a strong use of the ELM with 50 percent of employers in both localities reporting they would prioritize the hiring of needed skills from outside of the firm rather than developing them from within (Author's survey). However, evidence that the ILM was still relevant, and differences in local labour market conditions were revealed when establishments were asked how they would react to an increase in demand. In Kitchener, the most selected strategy was to hire full-time employees with 38 percent of respondents compared with less than 25 percent in the Sault. In contrast, 45 percent of Sault respondents stated they would be likely to hire part-time compared with 20 percent in Kitchener. Similarly, 30 percent of Kitchener respondents would increase overtime compared with 14 percent in Sault Ste. Marie (Table 4). As in many other hinterland communities (see Randall and Ironside 1996; Halseth 1999), restructuring has increased employment volatility more than in heartland economies, and this difference in employment strategies reflects the weaker employment picture in the Sault, which heightened employer caution about hiring full-time and the greater use of part-time work (Provincial government official, 1995).

Similarly, although a significant proportion of employers in Kitchener and Sault Ste. Marie stated they would either hire unskilled or skilled employees in response to skill shortages (46 and 42 percent, respectively), retraining current employees from the ILM remained a significant option (24 and 35 percent, respectively) (Table 5).

Aggregate data suggests a mixed role for the ILM, and interviews indicated a complex situation. In Sault Ste. Marie, some firms/institutions downsizing or restricting employment growth intensified their use of the ILM. In one hospital in which nearly 50 percent of employees were part-time, the impact of cuts to funding from the provincial government meant that the institution was less likely to seek needed skills from the outside and had developed an inventory of skills amongst existing full- and part-time employees and was stepping up training. The Human Resource manager of the hospital commented:
   Historically, hospitals had enough money to go out
   and hire the best so you'd see an awful lot of external
   posting. You'd have your postings to satisfy your
   collective agreements internally and you'd go and
   say "now where can I find the best physio in the
   province?" How can I get that person here? [Now]
   we've developed a human resource plan that we
   may not be able to afford the best so we've committed
   a lot of dollars to training [so] we can retrain
   internally (Sault health care sector, 1995).

Furthermore, within the manufacturing sector, there were very few firms which employed contingent labour for their shop floor operations. This observation contrasts with the findings of Peck and Theodore (1998) who found a high level of contingent worker present in Chicago manufacturing. Although increasing contingent workers in manufacturing production operations in Southern Ontario is occurring, for example, in food processing (see Denomme 1998), only one Kitchener region automotive component operation reported the use of contingent shop floor employees. A predominantly female workforce included approximately 10 percent part-time workers who were employed as a union-negotiated 'buffer workforce' in case of shifts in demand. Although most surveyed manufacturing establishments were unionized, it is difficult to ascertain its impact on contingent employment. In the strongly unionized Algoma Steel, even the limited use of temporary workers (often older former employees) was viewed as problematic by managers. However, non-unionized high-tech manufacturing operations did not include contingent workers, as this was felt to compromise the need for continual innovation (Kitchener manufacturer, 1995). Even in lower technology operations, employers' preferences were to use overtime rather than contingent workers to cope with demand increases. Thus:
   Part-time workers might be needed in the future, but
   at this plant the problem is that no one specializes in
   a specific job ... The operator needs to know about
   specific products and a part-timer might not be
   aware of changes to the customers specifications
   which might change daily (Sault auto parts manufacturer,

   There are very, very limited places out there where a
   part-time person is going to help you ... because ...
   [if] you bring in temporaries you're going to be
   much more inefficient, particularly at the skilled
   aspect of what we do (Kitchener footwear manufacturer,

There was also evidence that even firms highly dependent upon contingent employment were reconsidering some elements of this strategy. A leading southwestern Ontario grocery chain based in the Kitchener region had over 7,000 employees, of which 75 percent were part-time. However, the firm employment regime had to become more sophisticated in recruitment, as it began to hire professionals such as pharmacists. Furthermore, the manager considered that the ILM was too lean, and demographic changes in its workforce were necessitating the development of a stronger ILM. He stated bluntly that the firm had
   reached the limits of hiring part-time employees
   since we have a high level of turnover at the entry
   level and thus we need to create more full-time positions.
   We have an increasingly well educated workforce
   and not enough internal positions for them.
   Traditionally people worked with us in school and
   then went on to a career, but now continue on with
   us as there is less alternative employment available
  (Kitchener region grocery retailer, 1995).

Even where contingent employment was increasing, the evidence suggested that this represented more of a blurring of the ILM and ELM rather than the erosion of the former. This is best illustrated by the restructuring of the financial services sector. One major Canadian bank in the Kitchener region had approximately 360 employees, of which 100 were part-time. The bank had obtained needed skills from within and, during its restructuring in the early 1990s, developed an employment continuity program, so that employees could be re-trained and re-located within the organization if their position was eliminated. However, although less than 10 percent of its 'hiring' was done externally, the bank was beginning to do this for certain sales positions, because with financial service de-regulation sales of financial products had assumed a greater role in the banks revenues. This placed greater training and time commitment demands on existing staff which some were not prepared to accept. Thus increasing external hiring coincided with intensified internal training. Furthermore, in the bank's view, the advantages of using part-time rather than full-time workers was becoming less clear as internal flexibility increased.
   Historically, part-time has allowed us more flexibility;
   however, our full-time staff is realizing that they
   need to be more flexible also so into the future I'm
   not sure which of those two groups will offer or distinguish
   itself by being more flexible (Kitchener
   region financial services, 1996).

Similar examples of blurring came from two Kitchener region insurance firms. One had undergone a major downsizing in the early 1990, but more significantly, both had moved away from employing company-based sales agents towards the use of 'independent' agents who effectively operated as a small business with their own training budgets. However, one firm adopted this structure in part, because it wanted to retain the competencies of independent agents, as they had lower turnover than company-based agents. Because of its good reputation, agent retention was an issue as major banks were increasingly seeking these skills. To reduce this risk, the firm had also developed work teams and given more intensive training in soft interpersonal skills to employees. Furthermore, it was prioritizing the use of the ILM for needed skills utilising overtime by existing employees to cope with short-term increases in demand and hiring on contract only if the former strategy was inadequate. 'Since it is difficult to find people who can be effective in sales, particularly in life insurance' (Kitchener region financial services, 1995).

The other insurance firm was providing retention incentives to its agents, by having a policy of purchasing their businesses when they retired. Historically, the firm was highly dependent on its ILM for needed skills, and this remained significant even as the ILM was restructured. Thus the firm had adopted team training, widened the responsibilities of clerical employees and flattened its employment pyramid such that employee movement was more likely to be lateral than upward. Promotion was now based on the ability to be trained, because 'historically speaking putting in your time and seniority was the method of promotion, but this barely counts anymore' (Waterloo financial services, 1995). Accordingly, the compenzation policy had become a merit-based one linked to promotion. Finally, temporary workers were hired on their long-term potential to the firm, and it only hired skills externally if these were at such a high level that they needed to go outside of the insurance industry itself to recruit.

It was apparent that advancement within the ILM was increasingly based on the ability to be trained. Moreover, although such training contained a significant technical component, there was evidence of an increased blurring of technical and social skills. As one Kitchener training official stressed:
   Until a few years ago most firms thought training was
   simply computer training only and it was very individualized.
   Now we find that training has become
   broader with a greater emphasis on broader communication
   skills and programs have shifted from technical
   only to organizational cultural skills ... Firms
   find that their sales people have communication
   skills but not technical ones; whereas technical people
   don't have the communication skills, thus there
   has been a greater need for giving communication
   skills to their technical employees (Kitchener, provincial
   government training official, 1995).

Furthermore, not only was cultural training becoming more important, but in the case of one Kitchener region auto parts manufacturer, the emphasis was clearly on control and educating the workforce about managerial prerogatives:
   When we take them into the classroom, we are not
   teaching them about their work, we are teaching
   them about the company, business, running this
   plant and economics ... We try and put them in our
   shoes to have them know what it is like, and show
   them that we are all looked upon here by higher
   authorities that have the right to shut us down
   (Guelph auto parts manufacturer, 1995).

Finally, the ability to be trained became the basis for new forms of segmentation. This was in part because training was focused on managerial and technical/scientific staff. Thus in Kitchener, even though they accounted for only about one-third of the workforce in surveyed establishments, managerial, technical, scientific and sales employees received approximately nearly three quarters of all training (author's survey). Although this skewed distribution of training by occupation is consistent with other studies (see Betcherman 1995; Lundvall 2001), there was evidence of segmentation by age and gender within different occupations. Within financial services, older male employees were considered less able to cope with increasing training demands, as the onus for completing courses shifted onto the individual employee and work demands heightened necessitating what one employer termed 'home study' (Kitchener region financial services, 1996). Similarly, a Kitchener region auto components firm found that older supervisory and managerial employees had difficulty accepting many aspects of team working.

Impacts on the gendering of the ILM also differed. As McDowell (1997) found in her research on financial services in the City of London, a shift towards a more entrepreneurial culture in insurance eroded traditional notions of masculinity defined by an earlier, more bureaucratic regime, and as noted above, both within services and manufacturing, it was often middle-aged male managers who had the greatest difficulty coping with flatter, more network type organizations or demanding more entrepreneurial skills. Although historically women's advancement within ILMs was limited by their restriction to particular occupations, this was changing. Although in the public sector, clerical occupations could still be described by one Human Resource manager as a 'ghetto' and heavier manufacturing remained overwhelmingly male, women had made gains in middle and some higher administrative positions. However, such a shift did not uniformly favour women employees, and restructuring could also be associated with de-feminization. Thus in both financial and telecommunication services, clerical occupations were increasingly replaced by associate management positions. In both the cases, wider job responsibilities and the need for increased qualifications led to a decline in the female employment share, because for many new male employees, it was considered to be the first step within the firm's ILM.

Recruitment regulation, training and the local labour market

The early 1990s witnessed significant changes in the environment in which firms related to the local labour market. The recession had significantly increased unemployment especially in Sault Ste. Marie, and although unemployment rates declined in the Kitchener region after 1992, employment rates were slower to recover. Furthermore, although the most significant cuts to unemployment (later employment) insurance and social assistance occurred after 1995, by the early 1990s, these programs were already considerably less generous than they were in the 1970s (McIntosh and Boychuck 2000). Thus firms had little difficulty recruiting the most needed skills and became much more selective. At the same time, the adoption of human resource management (HRM) philosophies within larger establishments and the introduction of Employment Equity legislation by the Ontario New Democratic Party government made recruitment especially via the EILM a more scrutinized and formal procedure. Employment equity mandated that all firms and institutions employing over 50 employees had to compare their current employment patterns against the composition of the local labour market. If designated equity groups--such as women, the disabled, people of colour and others--were under-represented, firms had to file plans with the government as to how this difference would be overcome.

Employment Equity legislation was highly controversial, and it coincided with government programs designed to assist equity groups such women gain access to non-traditional trades (3) and larger firms who were adopting HRM strategies. Indeed, several interviewed managers considered that the law mandated practices they already had in place. However, it was more difficult to discern the direct impacts of legislation and HRM philosophy in the more diverse Kitchener region than in the smaller and more specialized Sault Ste. Marie. The former had a long history of women being employed in manufacturing, and female labour market participation was higher than the Canadian and Ontario average (English 1996), although some respondents noted continued resistance from employers and workforces against women entering non-traditional occupations. Yet in the Kitchener region, there was evidence of a continued use of racial and ethnic stereotypes.

Like many larger Southern Ontario localities, the Kitchener region had received new immigrants from the former Yugoslavia, Central America and Africa. Many of these immigrants took English as Second Language (ESL) training, and ESL Centres were used by local employers for recruiting workers--especially for lower-skilled occupations (Stroud 1995). These Centres reported that employers could demand workers on the basis of racial stereotypes such as when urban educated E1 Salvadoran immigrants were recruited by local agricultural establishments for their perceived rural work ethic and skills. For new immigrants, such stereotypes combined with a lack of recognition of their education and professional credentials contributed to concerns amongst some immigrant group representatives that this could contribute to exclusion from higher-skilled and higher-paid occupations (Kitchener region immigrant representative, 1999).

In Sault Ste. Marie, new immigrants constituted a much smaller proportion of new immigrants, but here, the most significant change was related to gender. Similar to many other older manufacturing and single industry resource communities in Canada, during the 1980s and 1990s, there were declines in male employment as the female proportion of the workforce (Hayter and Barnes 1992; Randall and Ironside 1996; McKenna and Roberge 2001). Yet the legacy of a male-dominated labour market remained. A number of Sault Ste. Marie respondents considered that Algoma Steel had provided a hostile environment for women. One government official noted that 'when young women applied for jobs at Algoma, they were told there were no locks on the washroom--they were not prepared to layout $1.50 for a lock' (Ontario provincial government training official, 1995). Moreover, this negative influence extended outside of Algoma Steel, discouraging women from applying for non-traditional employment, and in the 1970s, a woman applying to work with the city police was forced to undergo a five-year court case before she was hired. However, the downsizing of Algoma and the development of service employment had begun to shift attitudes, as women now outnumbered men in the local labour market. Thus the arrival of the Ontario Lottery Corporation in the early 1990s was noted by one respondent as changing the dynamics of the local labour market, by 'bringing 300 good jobs with an employer who hires and promotes equitably' (Women's group representative Sault Ste. Marie 1995). The same respondent also reported some evidence of a resultant re-negotiation of domestic divisions of labour, as female labour market participation increased and male participation decreased (see also McKenna and Roberge 2001).

Equity legislation and the adoption of HRM in both Kitchener and Sault Ste. Marie were associated with a restructuring of recruitment practices--especially the EILM. Recruitment practices in both the Kitchener region and Sault Ste. Marie were equally divided between informal and formal methods with informal methods such as word of mouth and direct applications representing over 40 percent of recruitment (Author's survey). Use of the EILM was connected to distinct geographic patterns, as employers sought to access workers with desired skills. Thus several high-tech employers in the Kitchener region recruited from Mennonite and rural backgrounds especially at the shop floor level to access the reputed strong work ethic and machinery experience of this community. Similarly, another large financial services firm recruited much of its staff from smaller communities just outside of the Kitchener CMA.

While suggesting the continued importance of the EILM, significant changes were occurring. Thus word of mouth was either being avoided, or if utilized, it was done in conjunction with interviews or tests to obtain the most qualified candidate. In both the localities, this shift away from word of mouth was concentrated in the public sector and larger firms partly to conform to Employment Equity. As one Sault Ste. Marie education sector respondent stated, 'word of mouth is a no-no because we are an equity employer and so we don't hire anybody by word of mouth' (1995). However, underlying this was increased attention to the social or cultural skills of employees. Although Manwaring (1984) and Hudson (2001) view the EILM utilizing family and friend networks as a form of social control of employees, it was clear that larger firms were no longer taking this method as sufficient for achieving these ends. In Kitchener, one food-processing firm found that the use of the EILM led to many production areas with a predominance of Portuguese workers. The poor English skills of many older workers, however, made adopting teamwork based on intergroup communication more difficult. Thus coupled with an increase in firm labour market power due to higher unemployment, firms were effectively 'taking control' of recruitment. As Sayer and Walker (1992) and Rutherford (1994, 1995) stress, the adoption of new forms of work organization necessitating increased reliance on employee discretion, judgement and commitment to the firm have been coupled with significantly more stringent recruitment standards to better ensure that these qualities are present in the workforce. More generally, employers saw changes to the EILM as associated with social control issues.
   Recruitment since 1990 has been through newspapers
   and trade magazines for skilled and semiskilled
   workers. We do not use the family and friends
   approach anymore. In the past it caused discontent
   when the manager's son was promoted before someone
   else. Today, the best people get the job ... (Sault
   auto parts manufacturer, 1995).

Furthermore, in the Kitchener region, the use of word of mouth recruitment could be an indirect result of a reliance on internal promotion. As one firm owner stressed: 'We post everything internally for employees and that's basically where the word of mouth comes in' (Kitchener manufacturer, 15 April 1995). Another manager argued:
   The majority of staff are coming up through the
   ranks ... Word of mouth is used in the sense that
   people come in and ask about jobs that they may
   have been told about by friends and relatives
  (Waterloo financial services, 1995).

However, there were indications of a stronger reliance on the ILM in Sault Ste. Marie. The primary method of obtaining skills for firms in Sault Ste. Marie was internal postings (nearly 20 percent) and promotions in the Sault Ste. Marie compared with 11 percent in Kitchener (Author's survey). Furthermore, Sault Ste. Marie employers were more likely to retrain existing employees to overcome skill shortages than those in Kitchener and were also more likely to use formal in-house training as the most important training method. This reflected the higher proportion of government employment found in the Sault Ste. Marie and larger private sector employers such as Algoma Steel, and indeed, several Sault-based respondents remarked on the existence of a seniority and job-specific skill culture pervading in the community.

There were also differences between the two localities in how recruitment channels and standards were changing. Nearly 42 percent of Sault respondent stated that there had been an increase in minimum education requirements between 1990 and 1995, whereas over 25 percent reported an increased emphasis in cognitive and problem-solving abilities (compared with 21 and 22 percent of Kitchener replies, respectively). In part, this difference reflected higher levels of unemployment in the Sault which allowed employers to raise the educational bar for recruitment. Even in the more buoyant Kitchener region, although increasing standards could be driven by changes in work organization, they could also be the result of increased unemployment allowing firms to be more selective of potential employees.
   Word of mouth was important when we needed 50
   employees by Monday, but now we don't do this or
   walk-ins. In terms of education it is now very simple
   to recruit people with at least high school and many
   of our latest recruits have at least high school if
   not university. In terms of assessing people a gut
   reaction used to be important but we are now looking
   at a more specific training program which is more
   scientific (auto parts manufacturer, Kitchener region,

Although considerable attention has been paid to the role of learning and formal education as key factors in contemporary restructuring (see Florida 1995; Lundvall 2001), the latter was not necessarily viewed as the most important attribute of job applicants. Indeed, the relative availability of educated and skilled workers meant that employers favoured previous work experience over education as a guide to a prospective employee potential--especially as it pertained to being trained within the firm's ILM. As one manager stated: 'If you really want to know if someone can do the job you look at what they've done. That equates to work experience. References are worse than ads--they're meaningless' (Sault Ste. Marie public sector, 1995). Another argued:
   Previous work experience is the most important
   qualification. We are looking for people who have
   the ability to be retrained, and adapt ... You have to
   be retrained as soon as step in the door to a certain
   degree. Experience tells you more than educational
   qualifications (Kitchener financial services, 1996).

The ambivalent role of formal education extended to internal and external firm training. Thus 59 percent of Kitchener replies indicated some use of external training facilities (such as community colleges or private trainers) compared with 52 percent in Sault Ste. Marie (Author's survey). In both the localities, firms had increased their use of external trainers in the previous five years, with 41 and 40 percent doing this in Sault Ste. Marie and Kitchener, respectively. This reflected contradictory trends with the firms' use of the ILM. Firms were increasingly expecting that their employees develop skills to maintain employment with the firm, but although firms were increasing investment in training, in general, training itself remained informal and often reactive, with relatively few establishments and institutions having a formal training plan. More Kitchener respondents had a formal training department (28 percent) than in the Sault (24 percent), but government training officials noted that many firms had cut training budgets and/or eliminated separate human resource departments which meant that employees requiring specific skills were trained externally. Many manufacturing firms had also cut formal apprenticeship programs and instead were offering 'mini-apprenticeships' which significantly reduced the transferability of skills outside of the firm.

Firms and institutions had increased their use of local community colleges and private trainers, and although one Kitchener region public sector respondent regarded increased use of consultant-based training as weakening the coherence of institutional training cultures, the externalization of training was being driven by the above forces. This meant significant changes in the role of local community colleges. Thus when colleges were established in the 1960s, over 80 percent of entrants had come directly from high school, by the mid-1990s, 60 percent came after completing both secondary education and some employment experience. However, colleges faced increased competition from private trainers who offered more specialized, less-expensive training. Although several employers commented on their ability to influence college course development, firms sometimes had difficulty communicating their long-term training needs to colleges and had to accommodate the latter's mandate to provide general education programs. Some specialized high-technology firms considered that their technical requirements had outpaced what community colleges could provide. There was thus emerging a dual external training system in which generic skill training was provided by colleges, whereas more specialized programs were the focus of private trainers. Significantly, firms were putting pressure on colleges to increase the behavioural skills, work discipline and 'job readiness' of college graduates. In both the localities, college respondents complained of this contradictory stance on the part of employers. As one argued:
   We are trying to do more on the general education
   and generic skills. The problem is that same companies
   that criticize us will complain more the moment
   we cut back on technical training. It is not easy to
   demand more of our students than we already do as
   they spend a lot of hours in school and many of them
   have families and jobs as well (Community college
   respondent, Kitchener region, 1995).

Discussion and Conclusions

The evidence presented in this paper reveals that even during a period characterized by restructuring and increased use of contingent labour, ILMs remained important for many Kitchener and Sault Ste. Marie region firms and institutions. It also stresses how ILM restructuring is linked to changes in the labour process, the local ELM and the wider regulatory and institutional context. Thus I have argued that where there is a surplus of needed technical skills and work place change is driven primarily by organizational rather than technological imperatives, then the ILM is an important institution for the development of cultural or social skills. This process is not 'space-less' but is intrinsically related to the restructuring of firm linkages with the local labour market.

The evidence in this paper suggests that the use of the ILM is generally an outcome of work place change, work intensification and a blurring between technical and social skills. Although social and cultural skills have always been important elements of the ILM, they are assuming a heightened importance, as workers are given more discretion and a critical strategy for some firms is the ability to channel employee's tacit knowledge towards the goals of the firm. Thus a key component of training becomes an attempt to systematically create employee identification with the firm. This behaviour was evident even in older, downsized manufacturing operations where as one manager argued 'you can only rule by fear so long' (Sault Ste. Marie Resource firm, 1995).

Increased blurring between technical and social skills was also linked to a blurring between the ILM and the ELM. Indeed, an important finding of this study is that there is not a singular model of ILM-ELM relations emerging but several. There remain downsized but still significant 'Fordist' ILMs which are often unionized in mature manufacturing and in the public sector where seniority and vertical job ladders are still important. There is also an emerging core of new ILMs in financial services and newer high-tech manufacturing in which job structures are more horizontal than vertical, and there are attempts to enhance both internal and external labour flexibility. Finally, there exists in perhaps the majority of establishments in smaller manufacturing and services, truncated or traditional ILMs which feature extensive use of contingent labour. In mature ILMs, the downsizing of ILMs could lead to their more intensive use rather than hiring more part-time or temporary workers. In new and some Fordist ILMs, recruitment standards were being risen in part due to higher unemployment and when firms increased their share of contingent employees, the latter could be included in ILM training and opportunities for advancement. Thus in financial services, firms were creating a cadre of independent/entrepreneurial agents; this replicated some aspects of the ILM via training, the buy-back of agencies and indeed these firms valued these agents precisely for reasons associated with ILMs such as lower turnover and higher levels of commitment to the firm.

As noted earlier, in both the localities, there was a re-defining of desirable skills which had significant gender implications which in particular adversely impacted older male, middle managers and production workers in both manufacturing and services, as they coped with flatter more networked organizations and continuous training work regimes. Similarly although women had made overall gains in employment including middle and higher administrative positions, older females with less formal skills could lose out as clerical and lighter manufacturing positions declined. Such shifts indicate that although gender remains important to employment outcomes, it may be intersecting with class and age in new ways (see McDowell 2002). These changes were in part linked to evidence that ILMs are more exposed to market forces/criteria (see Hudson 2001) with formerly clerical and other position are much more sales oriented than previously, whereas wage/salary determination is merit rather than seniority based. However, merit is still bureaucratically constructed and in part based on the ability to pass training courses having substantial social/cultural content.

Thus although there is evidence confirming Cappelli et al.'s (1997) view that ILMs are less 'sheltered' from market logic, this is not a linear process, for as the above-suggests firms continue to require a stability in skills and employment that cannot be captured in market-based transactions alone. Furthermore, although contingent labour markets may resemble a commodity like 'spot market' as Peck and Theodore (1998) argue, many temporary agencies attempt to develop a regular and dependable supply of labour for firms. More generally, the ELM is also institutionally bounded, and simple dichotomies between bureaucratic-governed ILMs and market-ruled ELMs are misleading (Block 1990; Lazonick 1991). The question then is less whether ILMs continue to function but in whose interests. Evidence in this paper indicates that downsizing, outsourcing and higher unemployment has tipped the balance of ILM operations towards employers and the development of new forms of segmentation which are partly age, gender and skill based.

Importantly, the restructuring of ILMs raise the issue of the role played space and scale. However, there is some evidence that geography matters more within than between the two localities. In large part differences between Kitchener and Sault Ste. Marie are due simply to industrial composition--especially in the case of the latter with the predominance of Algoma Steel. Employers may seek increased control of applicant qualities, yet the degree to which ILMs could be re-fashioned also reflected the legacy of earlier local labour market practices such as the prevalence of seniority and job-specific skill culture in Sault Ste. Marie. Within the two localities, employers often had distinct 'maps' of the local labour when accessing different skills--especially perceived socio-cultural attributes such as work ethic. Furthermore, such maps were not simply based on individual characteristics but could involve the use of perceived gender and racial stereotypes which limited the substitutability of labour within the firm. What constituted the 'local' labour market could vary considerably with the catchment area for higher end skills being geographically wider (national and international in some cases) than for less-skilled occupations.

The study also casts some light on the different trajectories of hinterland and heartland labour markets Randall and Ironside's (1996) finding of significant diversity between hinterland communities suggests that caution is required before making generalizations. Certainly Sault Ste. Marie's size (over 80,000) makes it significantly larger than most hinterland resource or single-industry communities; however, it does possess some similarities. Like many single industry and resource 'hinterland' communities, the Sault had undergone significant change since the 1970s--especially with the downsizing of its major employer (of males) Algoma Steel and increasing diversification and female employment (Hayter and Barnes 1992; Randall and Ironside 1996; Halseth 1999; McKenna and Roberge 2001). it has attracted public sector employment and some other advanced services which have partly offset employment declines in steel and forest products and provided some of the first sizable employment opportunities to women with higher education and formal skills. In contrast, change in Kitchener has been significant but less dramatic (especially in terms of gender), and the decline of older manufacturing has been more than offset by newer high-tech employment and diversified services. The region has attracted a large number of new immigrants which may be contributing to new forms of segmentation in the local labour market. In this way, it confronts similar issues to larger Canadian CMAs, and indeed during the 1990s, it became increasingly integrated with the Greater Toronto Area commuter shed. Finally, the evidence from Kitchener and Sault Ste. Marie is suggestive of broader trends. Thus Coffey (2000, 135) found that there was an overall decline in urban employment specialization between Canadian cities between 1971 and 1991, and Randall and Ironside (1996) found that many hinterland communities had increased employment diversity, as primary industry declined and services grew. In some important ways, restructuring has promoted a convergence between formally more distinctive 'hinterland' and 'heartland' employment structures, although as in many other hinterland communities, it is evident that public sector-related employment in Sault Ste. Marie plays a greater role in service employment growth than in more heartland labour markets.

More generally, the raising of recruitment standards and development of ILM strategies were due to higher levels of unemployment and the presence of a more educated local labour force and changing regulations such as Employment Equity. This was particularly evident with firm use of the EILM. Although the EILM was clearly still important especially in manufacturing firms who sought to 'import' desired social skills and work ethic from the surrounding community, it was being restructured in line with changes in the organizational requirements of firms combined with changing regulations. New requirements were being grafted onto existing EILMs with assessment of applicants becoming more systematic than previously. Furthermore, as the contradictory pressures on colleges to simultaneously increase social and technical skills reveal, ILM development was also associated with an increased reliance on external training which has heightened the importance of local training institutions.

Yet there is a need to be aware that the study covers a specific period (1990-1995), and as such, care must be taken in extrapolating results. The latter 1990s were generally a period of employment growth and shortages of some key technical skills, which may have led firms to adopt different strategies. Finally, research often tends to change biased, and the survey interviews revealed that in a substantial minority, if not the majority of workplaces there was relatively little change and what Betcherman (1995) terms traditional ILMs characterised by limited tenures and low levels of commitment and training remain significant. Indeed, as Hudson (1999) and Lundvall (2001) argue, many firms compete on the basis of low labour costs and not on knowledge or the organizational integration of their workforce and thus exhibit 'slash and burn' labour strategies. As Christopherson (2002) stresses, path dependencies towards this model are in part created by the pressures of short-term profit maximizing financial regime. However, there is also evidence that ILMs remain at least partially autonomous and reflect longstanding corporate cultures that mediates these restructuring pressures.


The author gratefully acknowledges the financial support of the Social Sciences and Humanities Research Council of Canada, the insightful comments of Lawrence Berg, Roger Hayter, David Doloreux and three anonymous reviewers. The usual disclaimer applies.


ALLEN, J., and HENRY, N. 1997 'Ulrich Beck's risk society at work: labour and employment in the contract service industries' Transactions of the Institute of British Geographers 22, 180-196

ATKINSON, J. 1987 'Flexibility or fragmentation? The United Kingdom labour market in the eighties' Labour and Society 12, 87-105

BENNER, C. 2003 'Labour flexibility and regional development: the role of labour market intermediaries' Regional Studies 37, 621-633

BETCHERMAN, C. 1995 'Inside the black box: human resource management and the labor market' in Good Jobs, Bad Jobs, No Jobs: Tough Choices for Canadian Labor Law, ed R. Adams, G. Betcherman and B. Bilson (Toronto: C.D. Howe Institute) 79-97

BILSON, B. 1995 'Workplace equity: a seat at the policy picnic?' in Good Jobs, Bad Jobs, No Jobs: Tough Choices for Canadian Labor Law, ed R. Adams, G. Betcherman and B. Bilson (Toronto: C.D. Howe Institute) 70-97

BLACKBURN, R, and MANN, M. 1979 The Working Class in the Labour Market (London: Macmillan)

BLOCK, F. 1990 Postindustrial Possibilities: A Critique of Economic Discourse (Berkeley: University of California Press)

SOWERS, N., and MARTIN, j. 2000 'Going mobile? Jobs in the new economy' Organization for Economic Cooperation and Development Observer 221/222, 113-117

CAPPELLI, P., BASSAW1, L., KATZ, H., et al. 1997 Change at Work (New York: Oxford University Press)

CARNOY, M., CASTELLS, M. and BENNER, C. 1997 'Labour markets and employment practices in the age of flexibility: a case study of Silicon Valley' International Labour Review 136, 27-48

CASTREE, N., COL, H., WARD, K., et al. 2004 Spaces of Work: Global Capitalism and Geographies of Labour (London: Sage)

CHRISTOPHERSON, S. 2002 'Why do labor market practices continue to diverge in a global economy? The "missing link" of investment rules' Economic Geography 78(1), 1-20

CHRISTOPHERSON, S., and NOYELLE, T. 1992 'The US path toward flexibility and productivity: the re-making of the US labour market in the 1980s' in Regional Development and Contemporary Industrial Response, ed H. Ernste and V. Meier (London: Belhaven) 163-178

CLARK, C. 1981 'The employment relation and the spatial division of labour' Annals of the Association of American Geographers 71(3), 412-424

COFFEY, W. 2000 'Canadian cities and shifting fortunes of economic development' in Canadian Cities in Transition: The Twenty-First Century, ed T. Bunting and P. Filion (New York: Oxford University Press) 121-150

CROUCH, C., FINEGOLD, D. and SAKO, M. 1999 Are Skills the Answer?: The Political Economy of Skill Creation in Advanced Industrial Countries (Oxford: Oxford University Press)

DENOMME, M. 1998 'Continuity and change: production restructuring and labour market flexibility in the food processing industry in Canada's technology triangle and the City of Toronto' Unpublished MA Thesis, Department of Geography, University of Waterloo

DOERINGER, P., and PIORE, M. 1971 Internal Labour Markets and Manpower Analysis (Lexington, MA: D C Heath)

DOUBLEDAY, S. 2000 'Channeling demand: the role of employment agencies in the temporary help industry of the Kitchener CMA and the Hamilton CMA' Unpublished MA Thesis, Department of Geography, University of Waterloo

ENGLISH, 3., and MCLACHLAN, K. 1996 Kitchener: An Illustrated History (Kitchener: Robin Brass)

FLORIDA, R. 1995 'Toward the learning region' Futures 27(5), 527-536

GARLANEAU, D., KREBS, H., MORRISSETTE, R., et al. 2001 The Quest for Workers: A New Portrait of Job Vacancies in Canada (Ottawa: Statistics Canada, Canada Human Resources Development)

GERTLER, M. 1999 'Negotiated path or business as usual? Ontario's transition to a continental production regime' Space and Polity 3, 171-197

GRIMSHAW, D., and RUBERY, 3. 1998 'Integrating the internal and external labour markets' Cambridge Journal of Economics 22, 199-220

GRIMSHAW, D. WARD, K., RUBERY, J., et al. 2001 'Organizations and the transformation of the internal labour market' Work, Employment and Society 15, 25-54

GUNDERSON, M. 2002 Rethinking Productivity from a Workplace Perspective. Canadian Policy Research Networks Discussion Paper no. 17 (Ontario: Canadian Policy Networks Ottawa)

HALSETH, C. 1999 "We came for the work:" situating employment migration in B.C.'s small resource-based, communities' Canadian Geographer/Le Geographe canadien 43(4), 363-381

HANSON, S., and PRATT, C. 1995 Gender, Work and Space (London: Routledge)

HARRISON, B. 1997 Lean and Mean: Why Large Corporations Will Continue to Dominate the Global Economy (New York: Guilford)

HAYTER, R., and BARNES, T. 1992 'Labour market segmentation, flexibility and recession: a British Columbian case study' Environment and Planning C 10, 333-353

HEATH, F. 1978 'The post war drive for benefits' in 50 Years of Labour at Algoma: Essays on Aspects of Algoma's Working-Class History (Sault Ste. Marie: Algoma University College)

HUDSON, R. 1999 'The learning economy, the learning firm and the learning region: a sympathetic critique of the limits to learning' European Urban and Regional. Studies 6(1), 59-72

--. 2001 Producing Places (New York: The Guilford Press)

KUMAR, P., and HOLMES, J. 1997 'Canada: continuity and change' in After Lean Production: Evolving Employment Practices in the World Auto Industry, ed T. Kochan, R. Lansbury and P. Macduffie (Ithaca, NY: ILR Press) 85-108

LAZONICK, W. 1991 Business Organization and the Myth of the Market (Cambridge: Cambridge University Press)

LECKIE, N., LEONARD, A., TURCOTTE, J., et al. 2001 The Evolving Workplace Series: Employer and Employee Perspectives on Human Resources Practices (Ottawa: Statistics Canada, Human Resource Development Canada)

LOVERING, J. 1990 'A perfunctory sort of post-fordism: economic restructuring and labour market segmentation in Britain in the 1980s' Work, Employment and Society 4(May), 9-28

LOWE, G. 1999 'Contingent work: trends benefits and costs and challenges presented at pathways to the future'. Paper Presented at the 37th Annual BC Human Resources Management Association Spring Conference, Whistler BC, 6 May

LUNDVALL, B.-A. 2001 'Innovation and learning in a national system--in industries and firms, Paper Presented at the Eindhoven Centre for Innovation Studies, Eindhoven University of Technology, the Netherlands, 20-23 September

MAGUIRE, M. 1988 'Work, locality and social control' Work Employment and Society 2(1), 71-87

MANWARING, T. 1984 'The extended internal labour market' Cambridge Journal of Economics 8, 161-187

MARSDEN, D. 1986 The End of Economic Man? (London: Wheatsheaf)

MARTIN, R. 2001 'Local labour markets' in The Oxford Handbook of Economic Geography, ed G. Clark, M. Feldman and M. Gertler (Oxford: Oxford University Press) 455-476

MASSEY, D. 1984 Spatial Divisions of Labour: Social Structures and the Geography of Production (London: Macmillan)

MCADAM, R., and MCCREEDY, S. 2000 'A critique of knowledge management: using a social constructionist model' New Technology Work and Employment 15(2), 155-168

MCDOWELL, L. 1997 'A Tale of Two Cities? Embedded Organizations and Embodied Workers in the City of London' in Geographies of Economies, ed R. Lee and J. Wills (London: Arnold) 118-129

--. 2002 'Father and Ford revisited: gender, class and employment change in the new millennium' Transactions of the Institute of British Geographers 26, 448-464

MCINTOSH, T., and BOYCHUCK, G. 2000 'Discovered: El, social assistance and the growing gap in income support for unemployed Canadians' in Federalism, Democracy and Labour Market Policy in Canada, ed T. Mcintosh (Montreal: School of Policy Studies, Queen's University, McGill-Queen's Press) 65-178

MCKENNA, M., and ROBERGE, R. 2001 'Restructuring, gender and employment in flux: a geography of regional change in Cornwall, Ontario' Canadian Geographer/Le Geographe canadien 45(2), 223-236

MILKMAN, R. 1983 'Female factory labor and industrial structure: control and conflict over "women's place" in auto and electrical manufacturing' Politics and Society 12, 159-203

MOSS, P., SALZMAN, H. and TILLY, C. 2000 'Limits to market mediated employment: from deconstruction to reconstruction of internal labor markets' in Nonstandard Work: The Nature and Challenges of Changing Employment Arrangements, ed F. Carre, M. Ferber, L. Golden and S. Herzenberg (Champaign, IL: Industrial Relations Research Association) 95-121

NOYELLE, T. 1987 Beyond Industrial Dualism (Boulder, CO: Westview Press)

OSTERMAN, P. 1984 Internal Labor Markets (Cambridge, MA: MIT Press)

--. 1994 'Internal labour markets: theory and change' in Labour Economics and Industrial Relations: Markets and Institutions, ed C. Kerr and P.D. Staudohor (Cambridge, MA: Harvard University Press) 303-339

PECK, J. 1996 Workplace: The Social Regulation of Labor Markets (New York: The Guilford Press)

PECK, J., and THEODORE, N. 1998 'The business of contingent work: growth and restructuring in Chicago's temporary employment industry' Work Employment and Society 12(4), 655-674

PREMIER'S COUNCIL 1990 People and Skills in the New Global Economy (Toronto: Government of Ontario)

RANDALL, J., and IRONSIDE, R.G. 1996 'Communities on the edge: an economic geography of resource-dependent communities in Canada' Canadian Geographer/Le Geographe canadien 40(1), 17-35

REIMER, S. 1999 'Contract service firms in local authorities: evolving geographies of activities' Regional Studies 33(2), 121-130

RUTHERFORD, T. 1994 'From sitting by Nellie to the classroom factory? The restructuring of skills, recruitment and training in a South Wales motor components plant' International Journal of Urban and Regional Research 18(4), 470-490

--. 1995 'Control the one's you can restructuring, recruitment and training in Kitchener-Waterloo Manufacturing, 1987-1992' Canadian Geographer/Le Geographe canadien 39(1), 30-45

--. 1996 'The socio-spatial restructuring of Canadian labour markets' in Canada and the Global Economy: The Geography of Structural and Technological Change Space Economy, ed J.N.H. Britton (Montreal: McGill-Queens Press) 407-432

SALZMAN, H., MOSS, P. and TILLY, C. 1998 'The new corporate landscape and workforce skills: what firms want; how they get it; and the role of education, training and community colleges' National Center for Postsecondary Improvement 508 Ceras School of Education Stanford University, Stanford, CA (Available at

SAYER, A., and WALKER, R. 1992 The New Social Economy: Reworking the Division of Labour (Oxford: Basil Blackwell)

SMITH, U. 1999 'Insecurity in the labour market: the case of Canada since the Second World War' Canadian Journal of Sociology and Anthropology 24(2), 193-224

STANFORD, J. 1995 'Bending over backwards: Is Canada's labour market really inflexible?' Canadian Business Economics 4(1), 70-87

STATISTICS CANADA 1998 The Evolving Workplace: Findings from the Pilot Workplace and Employee Survey (Ottawa: Statistics Canada, Human Resource Development Canada)

STROUD, T. 1995 'English as second language centres and the Kitchener labour market' Senior Honours Essay, Department of Geography, University of Waterloo

THEODORE, N., and PECK, J. 2002 'The temporary staffing industry: growth imperatives and limits to contingency' Economic Geography 78(4), 463-495

THRIFT, N. 1997 'The rise of soft capitalism' Cultural Values 1(1), 29-57

TOMLINSON, M. 2002 Measuring Competence and Knowledge Using Employee Surveys: Evidence Using the British Skills Survey of 1997 CRIC Discussion Paper 50, ESRC Centre for Research on Innovation and Competition, University of Manchester

VOSKO, L. 1998 'Regulating precariousness? the temporary employment relationship under the NAFTA and the EC Treaty' Industrial Relations 53(1), 123-153

WALKER, D. 1987 Manufacturing in Kitchener-Waterloo: A Long-Term Perspective Publication Series No. 26 Department of Geography, University of Waterloo

WOOD, S., ed. 1989 The Transformation of Work? Skill Flexibility and the Labour Process (London: Unwin-Hyman)

YATES, C. 1999 'Defining the fault lines: new divisions in the working class' Capital and Class 66, 119-147

YOURCK, S. 2002 'Canadian job tenures rose in the 1990s' Globe and Mail 26 October


Department of Geography, Syracuse University New York 13244-1020, USA (e-mail:

(1) By 1994, the percentage of employees in Canada in 'non-standard' employment arrangements consisting of part-time, and temporary employment multiple job holders and own account self-employed had risen from 28 percent in 1989 to over 30 percent of the Canadian workforce (Lowe, 1999). By the mid-1990s in the United States, non-standard employment had also risen to over 30 percent of the workforce up from 26 percent in 1980 (Carnoy et al. 1997).

(2) The survey sample included a census of all manufacturing establishments employing over 200 people and all service establishments employing over 100 people. In addition a 10 percent sample of all manufacturing establishments employing less than 200 people and all service establishments employing less than 100 people were also included. In all, 450 questionnaires were mailed with ninety firms (a 20 percent response rate) and institutions responding--sixty-five in the Kitchener region and twenty-five in the Sault. The response was skewed, especially in the Kitchener region, towards manufacturing and larger establishments. However, although only 29 percent of respondent establishments were in service-related sectors in Kitchener, they did represent 63 percent of the total employment of all surveyed establishments. In the Sault, the situation was reversed with 72 percent of respondents being service-related establishments but representing less than 42 percent of the total employees of surveyed firms and institutions. The survey's limitations then are that it is based primarily on evidence from larger establishments and institutions, it is not a longitudinal study and thus captures only a limited amount of change over time. Finally, it focuses primarily on demand as opposed to supply side labour market change.

(3) Employment Equity legislation in Ontario was abolished after the defeat of the New Democratic Party by the neo-liberal Progressive Conservatives in 1995.
Table 1
Employment change in Kitchener, Sault Ste. Marie and Ontario by
industry 1989-1994


                                            % change
                         1989      1994     1989-1994

All industries          197,000   203,000       3.10
Manufacturing            68,000    53,000     -21.9
Construction             14,000    10,000     -27.7
TCOU                     10,000    10,000       0.0
Trade                    37,000    34,000      -7.8
Service                  49,000    69,000      39.4
FIRE                     11,000    16,000      50.6
Public Administration     5,000     8,000      66.4

                               Sault Ste. Marie             Ontario

                                            % change       % change
                         1989      1994     1989-1994       1989-94

All industries           41,107    33,980     -17.3           -1.6
Manufacturing            10,605     5,047     -52.4           -7.3
Construction              2,948     1,789     -39.3           18.9
TCOU                      2,634     2,141     -18.7           -7.2
Trade                     6,147     5,419     -11.8            1.4
Service                  14,616    14,162      -3.1           12.1
FIRE                      1,170     1,639     +40.1            0.6
Public Administration     2,565     3,341     +30.3            1.6

FIRE, finance insurance and real estate employment; TCOU,
transportation, communication and equipment operators.

SOURCE: Statistics Canada (1995): Labour Force Annual Averages
1989-1995 Ottawa: Statistics Canada.

Table 2
Employment change by occupation and gender in Kitchener and Sault
Ste. Marie 1989-1994


                                        Male (%)   Female (%)   Total

All occupations                             4.6        1.3        3.1
Managerial                                 56.4       61.5       58.7
Clerical                                    0.7       -8.8       -7.0
Sales                                       8.9      -16.6       -5.3
Service                                     1.4       12.4       -5.4
Processing, machining and fabrication     -24.5      -32.7      -26.9
Construction                              -36.4      -68.9      -37.2
TCOU                                       31.2      -41.3       21.3
Material handling and other crafts         14.1        6.4       12.1

                                                   Sault Ste. Marie

                                        Male (%)   Female (%)   Total

All occupations                           -29.4       -1.2      -17.3
Managerial                                 -2.7       24.7       12.4
Clerical                                  -23.9       -8.6      -11.4
Sales                                     -11.9        4.7       -1.9
Service                                    -7.4      -22.6      -16.9
Processing, machining and fabrication     -55.8       38.3      -55.5
Construction                              -50.3        0.0      -50.3
TCOU                                      -21.3      -32.1      -22.8
Material handling and other crafts         34.1      -50.8       11.4

TCOU, transportation, communication and equipment operators

SOURCE: Statistics Canada: Labour Force Annual Averages 1989-1995
Cat. no. 71-52

Table 3
Work place restructuring in Kitchener and Sault Ste. Marie 1990-1995

                          Introduce new         Introduce new work
                         technology (%)            practice (%)

Kitchener                      92                       82
Sault Ste. Marie               92                       64

                      Minor-major increase     Minor-major increase
                     in technical skills (%)    social skills (%)

Kitchener                      84                       75
Sault Ste. Marie               73                       58

Kitchener n = 61; Sault Ste. Marie n = 21.

SOURCE: Author's survey.

Table 4
Employer response to increasing demand

                        Hire full-time
                         employees (%)        Use overtime (%)

Kitchener                     37                     30
Sault Ste. Marie              24                     14

                        Hire part-time
                    temporary employees (%)      Other (%)

Kitchener                     20                     13
Sault Ste. Marie              48                     14

Kitchener n = 103. Respondents could choose more than one category;
Sault Ste. Marie n = 21.

SOURCE: Author's survey.

Table 5
Employer response to skill shortages

                       Change          Hire and          Retrain
                     recuitment     train unskilled      current
                    channels (%)     employees (%)    employees (%)

Kitchener                18               16               24
Sault Ste. Marie         12               19               35

                    Hire skilled    Change pay and        Other
                    employees (%)   conditions (%)         (%)

Kitchener                27               12                3
Sault Ste. Marie         27                6                1

Kitchener n = 397; Sault Ste. Marie n = 108.

N = total number of respondents. Respondent could indicate more than
one category for each of five different skilled/occupational

SOURCE: Author's survey.
COPYRIGHT 2006 Canadian Association of Geographers
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2006 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Rutherford, Tod D.
Publication:The Canadian Geographer
Geographic Code:1CANA
Date:Jun 22, 2006
Previous Article:Frontier engineering: from the globe to the body in the Cold War Arctic.
Next Article:Vulnerability to climate change hazards and risks: crop and flood insurance.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters