How intangible factors contribute to economic development: Lessons from a mature local economy

0305-750x/90 $3.00 + 0.00 Pergamon Press plc

World Development, Vol. 18, No. 9. pp. 1295-1308,199O. Printed in Great Britain.

How Intangible Factors Contribute to Economic Development: Lessons from a Mature Local Economy PETER B. DOERINGER Boston University

and DAVID G. TERKLA University of Massachusetts, Boston Summary. - Building on a case study of a mature manufacturing region in Massachusetts, this paper provides new evidence on the importance of nontraditional cost factors in determining regional growth. Specialized mature manufacturing firms that have broken the product cycle factors to promote are identified. We then discuss development strategies for using “intangible” such firms.

1. INTRODUCTION There is an extensive literature on the determinants of state and local economic development, most of which involves the econometric modeling of growth processes (i.e., Wheaton, 1979; Howland, 1979; Browne, 1982; Carlton, 1979; Kieschnick, 1981; Wasylenko and McGuire, 1985; Wheat, 1986; Bartik, 1985; and Plaut and Pluta, 1983). These studies focus principaily upon the long-term effects on growth of “visible” costs - factors such as wages, taxes, and energy costs that are readily measured. Economic theory predicts that these visible costs should have a greater effect in the long term than in the short term, and these studies generally support the proposition that visible costs influence economic growth. However, they are less clear as to how fast the existing economic structure adjusts to cost differences or whether there are other less obvious factors that may also influence growth patterns. Our research on state and local-level economic development further confirms the role of visible costs (Doeringer, Terkla, and Topakian, 1987; Terkla and Doeringer, forthcoming). It also reveals additional findings that have not figured prominently in the regional development literature.

In our analysis of variations in employment growth among states during the 1970s for example, we introduce a measure of industrial structure in addition to cost factors. We find that the industrial structure of each state in the United States is more important than the relative magnitude of its visible costs in explaining employment growth.’ Moreover, visible cost factors such as wages, tax rates, transportation, and energy could not explain any of the variation in employment growth among local areas within a state.

2. OMITTED VARIABLES AND THE NEED FOR A CASE STUDY One explanation for this surprising result is that important tangible cost factors were omitted from standard regional development models, particularly at the local level. For example, companies often mention the availability of utility connections and easy access for truck transportation as relevant considerations for the location of new plants (Doeringer, Terkla, and Topakian, 1987). Such cost factors may be highly visible (and quantifiable) features of the local economic terrain to a business considering expansion or relocation, and yet still be hard to

1295

1296

WORLD

DEVELOPMENT

identify in a systematic way for purposes of economic modeling. A second set of cost factors that may have been omitted, however, includes the intangible elements of local areas such as the business climate, community leadership, and the quality of the labor force. As Schmenner (1982) and Schmenner, Huber, and Cook (1987) suggest, these hard-to-measure interactions between the firm and the local environment may have a pervasive influence upon growth, especially at the local level2 One piece of evidence of such “invisible” factors is found in studies of national economic growth (Dennison, 1962) that have identified an important long-term growth “residual.” While some hard-to-measure factors (such as scale economies and technical progress) are formally included in some of these models, others are not. The missing variables include factors such as the quality and attitudes of the labor force and improvements in managerial practices that promote productive efficiency. Another set of examples is found in studies of business performance. These studies highlight the importance of factors such as entrepreneurial behavior (Porter, 1980; Abernathy, Clark, and Kantrow, 1983; Peters and Waterman, 1982; Lawrence and Dyer, 1983); the development of scope economies (Chandler, 1962 and 1977; Chandler and Daems, 1980; Elbaum and Lazonick, 1986); corporate strategies (Bluestone and Harrison, 1982 and Piore and Sabel, 1984); and “product cycle” phenomena (Norton, 1986). Similarly, econometric studies of regional growth have often found that the labor relations climate affects employment growth (Kieschnick, 1981; Wheaton, 1979; Terkla and Doeringer, forthcoming). Traditionally, regional growth policy has focused on readily-measured costs and on the potential for subsidizing such costs as a stimulus to growth. There is some indication, however, that such subsidies have often met with only limited success (Kieschnick, 1981; Kotlowitz and Buss, 1986). Meanwhile, the systematic exploration of how invisible cost factors might affect growth, and how these might be incorporated into development policy, has been neglected. While difficulties in directly measuring factors such as business strategies or product cycle considerations have impeded the econometric analysis of how invisible factors affect regional growth, these data limitations can be circumvented through case studies of business performance at the local and regional level, Case studies by themselves do not provide a reliable guide to how hard-to-measure cost factors affect local

econometric growth. However, they can provide insights for building an analytical framework that can be used to analyze growth processes in other local settings. This paper reports the results of a case study conducted in the Montachusett region, a relatively self-contained labor market area located about 50 miles northwest of Boston in north central Massachusetts. Montachusett was once a site of America’s industrial revolution, and has since experienced several periods of economic growth. In recent years, however, it has suffered from economic decline as its industrial base has matured. Based upon a consistent pattern of findings that cut across a wide range of industries in Montachusett, we propose a view of local growth processes that emphasizes the importance of managerial practices and the interaction between these practices and the local economy.

3. PRODUCT CYCLES AND PRODUCT SPECIALIZATION The case study findings lead us to identify four groups of firms in the local economy - innovative firms. mature mass production firms, customized firms, and “hybrid” firms - that contribute in different ways to the growth of the local economy. Both innovative firms and mass production firms exhibit typical product cycle characteristics in their evolution. These product cycle firms depend more heavily on the visible costs of the region as they mature, while drawing little from the invisible strengths of the region. As a result, these firms appear to be economically “footloose” and destined to decline or leave the region. The experience of customized and hybrid firms contrasts sharply with that of product cycle firms. Customized and hybrid firms rely on strategies of product specialization and production flexibility to remain competitive. They are highly dependent upon a variety of invisible economic factors - niche-seeking business strategies, an adaptable labor force, flexible wage setting, and constructive labor-management relations. Such firms are prosperous even though they often belong to declining industries. As the following sections will show, the dramatic difference in the performance between product cycle and specialized firms is founded in a set of invisible factors present in the local economy. More attention should therefore be paid to policies affecting invisible growth factors than has been the case in the past.

1297

INTANGIBLE FACTORS 4. PROFILE OF THE MONTACHUSET-I ECONOMY The Montachusett region is a useful location for examining the importance of invisible factors, because it has a large and diverse group of manufacturing industries. These range from mature industries such as furniture and apparel, to younger industries such as machine tools and plastics, to high technology electronics. The wide range in economic performance exhibited by these industries yields important lessons about the sources of business growth, controlling for the distinctive features of the state and local economic environment. Most of the region’s older industries have suffered from decline, but this has not been the universal pattern. Machine tools and plastic resin-making industries, for instance, actually outperformed their national and state counterparts in the postwar period, while others have had at least some periods of prosperity. Even in industries that have sustained continuing job losses, there are particular firms that remain strong performers. This variance in performance among a broad range of firms and industries provides a unique opportunity for identifying and analyzing the effect of less visible economic influences on local growth. (a) The labor market The region has a relatively stable population of slightly under 200,000 and a labor force of about 75,000. As is typical of mature regions, labor has been in abundant supply in recent years. Because of job losses in mature industries, the unemployment rate for the area was chronically high during the late 1970s and early 1980s. Median family earnings in 1979 were roughly equivalent to median family earnings nationwide, and the age and educational characteristics of the population also approximate those of the United States as a whole. However, slow growth and relatively high unemployment have tended to depress incomes in the area somewhat below those statewide, despite the importance of manufacturing in the economy. Median family income was about 7% below the statewide level and, at a more detailed occupational level, the pay gap between the region and the state ranged from over 20% for assembly jobs to 7% for clerical and sales work. Of the major manufacturing occupations, only those in the machine trades achieved pay parity with the state. report that the quality of industrial . .Employers .. skulls and craftsmanship is high among the

region’s workforce. Most important, however, employers are unusually impressed with the attitudes of workers in the region. At all levels of skill, and across all demographic groups - recent immigrants, young workers, and experienced workers - there was a consensus that worker effort and commitment to work were high. (b) Industrial structure During the 19th and early 20th centuries, the Montachusett economy approximated the economist’s model of regional cost advantage. The presence of water power encouraged the location of a variety of mill-based industries such as shoes and textiles. Water availability and ready access to high quality hardwoods and to pulpwood led to the development of lumber, paper, and furniture industries. Proximity to northeastern markets, and to the port of Boston, further added to these locational advantages.3 By the 1970s. only a few remnants remained of the region’s early industries such as paper, apparel, and furniture. Mill industries had been replaced in importance by two further waves of industrial development plastics and nonelectrical machinery. High technology industries also came to the region, but not to the degree experienced by neighboring areas in Massachusetts. Nevertheless, the region remains heavily industriai; manufacturing provides almost half of all private sector employment in the region, nearly twice the national average. Within manufacturing, over three-fourths of all jobs are accounted for by five mature sectors - plastics and resins, nonelectrical manufacturing. paper. furniture. and fabricated metals. High technology industry provides only 6% of manufacturing jobs in the region. In addition to the prevalence of manufacturing, the Montachusett economy is distinguished by the number of small and medium-size firms. Forty-five percent of all manufacturing establishments have between 20 and 250 employees, compared to 30% for the United States as a whole. Even a higher percentage of the region’s manufacturing jobs (56% compared to 39% nationally) is in these small and medium-sized firms. Many of the larger firms in the region are unionized, whereas most of the smaller firms are not. (c) Changing regional costfactors In seeking to explain Montachusett’s long-term industrial changes, the most obvious reasons

1298

WORLD

DEVELOPMENT

would seem to be related to visible costs. Changes in markets, transportation. and energy costs have all worked against the historical locational advantages that first attracted millbased industries to the region. Similarly, changes in technology in some industries have increased the demand for technical skills while diminishing the importance of the region’s skilled blue-collar labor force. As a result, many firms in the region report serious cost disadvantages in their traditional product lines. At the same time, however, new firms in industries such as plastics and metal fabrication have joined the economy when the structure of visible costs would have often pointed to alternative locations. This paradox is consistent with our econometric analyses showing that local economies such as Montachusett seem to be able to grow in ways which are independent of such cost disadvantages (Terkla and Doeringer, forthcoming).

5. THE ECONOMIC PERFORlMANCE MONTACHUSETT FIRMS

OF

In order to probe these differences in economic performance among local firms, we conducted detailed interviews with local business executives and trade union officials in 1985 and 1986. This survey covered about 50 manufacturing firms: all firms with 80 or more employees and a sample of smaller firms. The interviews were used both to collect detailed information on what costs affected competitiveness and on the relationship between management style and a firm’s economic performance, and to elicit managers’ perceptions of the region’s business environment. This interview data base does not lend itself to formal modeling. However, the interview methodology was structured so as to allow for the development of hypotheses about growth during the early interviews that could then be “tested” during subsequent interviews. While space constraints prevent a complete report on the interview findings, a sampling of the results for the key Montachusett industries will give some flavor of the major themes underlying business performance in the region.

(a) Machine tools, mold making, and metal fabricating Machine tool firms in Montachusett design and manufacture complete, and often one-of-a-kind, machine tools. The smaller shops in particular

are finding it difficult to compete with foreign suppliers in established markets and are constantly searching for new markets. Mold makers are relatively young companies that are producing high precision molds primarily for plastics fabricators, but they too face growing foreign competition particularly for the low quality end of the mold market. Metal fabricators range from firms that produce keys and cutting tools to those that produce heavy capital goods such as turbines and transmission housings. Most firms in these industries produce on a contract basis - the large firms bid on large and complex projects while the small firms rely on subcontracts, often with short delivery times. Those firms that are doing well have been able to compete by concentrating increasingly on markets where quality and speed of delivery are more important than price. The most successful firms engaged heavily in product differentiation and continually searched for new market niches. Where foreign competition has been a problem. some firms have begun to develop franchising arrangements which grant them rights to market foreign products in exchange for providing engineering expertise to improve machining and design capabilities. The quality of the labor force is the primary advantage of the area for these firms. They employ skilled labor and most believe that the skills and work attitudes of the local workforce are better than elsewhere in the country. While the largest firms (and smaller firms as well in mold making) have acquired state-of-the-art computerized technologies, most firms continue to use standard machine tools and rely heavily on skilled machinists to set up equipment and to produce to close tolerances. The firms in competitive difficulty were those that focused on standardized products and that relied on narrow, “dedicated” technologies. These firms rarely provided anv “extras” to customers in terms of speed of delivery, quality. or technical services.

(b) Paper The paper industry consists of two very different types of products - cardboard boxes and paper making. Both are mass production industries. Box making requires only low-skilled labor and remains in the area largely because land is inexpensive enough to permit storage of large inventories of boxes and because high transportation costs provide a buffer against competition from lower cost producers. Paper making draws upon some of the same transporta-

INTANGIBLE

tion cost protections, and also benefits from a heavily subsidized local waste water treatment plant. Wages and energy costs, however. are sufficiently high in the area to make competition from the southern United States a problem. As a result, the local paper making industry was in considerable trouble until recently when some paper making companies switched to types of speciality papers for which there were relatively short production runs. The move from standardized to specialized paper making required increased flexibility in job assignment and work rules. The importance of these changes is illustrated by the contrast between two plants owned by the same company, but organized by different local unions. One plant successfully negotiated the needed flexibility in work practices while the other was forced to close because it could not obtain similar concessions from its union. Flexibility and specialization notwithstanding, the industry views its future as precarious until its strategy of specialized production can be further tested.

1299

FACTORS

(d) Apparel and furniture Although the apparel and furniture industries in Montachusett consist entirely of mass production firms that have shown considerable decline in the past two decades, a handful of firms in both industries have developed survival strategies. The successful apparel firms have secured product niches, not through specialization, but from their reputation for quality and reliability and from their ability to respond quickly to odd lot fill-in orders. In the furniture industry, which appears to have the bleakest future among the mature industries in Montachusett, successful firms are specializing in market niches such as children’s furniture and custom orders for large institutions such as the military or universities. However, even some of these firms are gradually becoming marketing rather than fabricating organizations as they continue to expand in areas involving the assembly of knockdown imported furniture.

(e) High technology (c) Plastics The plastics industry consists of both resinmaking and fabricating firms. Nationally, resin making has become increasingly located near petroleum sources in the South to reduce energy and transportation costs. Fabricators have also been relocating to the South to be close to resin supplies and to take advantage of lower labor costs. Reflecting these national trends, low-value, standardized plastic production of items such as toys and household products that use commonplace resins has trouble competing because of Montachusett’s relatively high energy and laboi costs. Like standardized producers in other industries, some fabricators are now purchasing foreign products which they then market nationally. Firms specializing in high value production, such as computer casings and artificial limbs, appear much more profitable, as are the resin makers that supply exotic materials to these fabricators. These fabricators work closely with their resin makers to develop new products and applications of new materials for which there will be secure product niches. Because of the precision and complexity of many of these products, fabricators also work closely with the local moldmaking industry and often employ their own skilled machinists and metalworkers. This branch of the plastics industry is a growth sector and draws upon local agglomeration economies as well as the highly skilled, local labor force and firm-specific specialization strategies.

Montachusett has a small high technology sector. Some plants are well-established. locally owned businesses that are now moving from mechanical into electronic versions of their traditional products. Some are branch plants of companies headquartered closer to Boston; they came to the area largely by chance and have remained because of the availabilitv of land and labor. Many of the smaller firms are independent “start-up” companies that located in the area either because it was near the residence of the company founder or because of the efforts of local economic development officials. None of these high technology firms has a strong economic attachment to the area. The established plants are becoming much less reliant on the area’s skilled labor force as they switch from mechanical to electronic products; the branch plants switch products frequently based upon the ebb and flow of business elsewhere in their parent companies, rather than on changes in local economic advantages; and the start-up companies employ mostly professionals who have a residential attachment to the area.

6. INTERPRETING THE INTERVIEW DATA Despite the tangible cost disadvantages of the region facing many of its industries - transportation costs for the raw material inputs to the

WORLD DEVELOPMENT

1300

plastic resin industry, high labor costs relative to foreign sources for apparel and plastic fabrication, distance from markets for the paper industry, and the like - the interviews revealed considerable.underlying strength in both the nonelectrical machinery and fabricated metals industries. A second set of industries - apparel, plastics, and chemicals - may either prosper or decline in the next decade, depending both on developments within these firms and in their larger product markets. There is similar uncertainty about the high technology firms in Montachusett because of the inherent volatility of high technology product markets and because these firms have not developed strong attachments to the area. Only the paper and furniture industries seem to have a poor long-term prognosis, although even they have some successful product lines. These divergent patterns of business performance, both within and across industrial sectors, can be modeled in terms of the behavior of two distinct sets of firms - product cycle firms and specialized customized and hybrid firms (see Table 1). Each set of firms has its own set of business strategies; each draws in different degrees upon the economic strengths and weaknesses of the region’s economic environment; and each has distinctly different prospects for the future. Innovative firms are doing well, while mature product cycle firms are in trouble. Customized firms prosper by developing unique product and service niches that can stave off competition from mass production while hybrid firms continue to find success in mass production by incorporating features of specialization.

(a) ProdLcct cycle firms Product cycle firms are those whose pattern of growth fits the model of the industrial product cycle, particularly as it has been applied to explaining the experience of both mature and innovative firms.’ One group of these firms has reached the mature phase of the product cycle. These firms found in industries such as paper, plastic fabrication, furniture. and metal fabrication - compete in national and international markets for relatively standardized, mass produced goods. Mature product cycle firms are highly sensitive to visible costs. As a result, these are the firms that are rapidly losing markets to lower cost regions and countries and that have been the major contributors to the economic decline of mature regions. Those that are managing to survive depend upon visible cost advantages such as the availability of low wage/high effort immigrant labor. or low cost waste water treatment, or the protection from competition in regional markets that is afforded by high transportation costs facing producers outside the region. At the other end of the product cycle spectrum are innovative firms, whose prosperity depends upon bringing new products to markets where there are relatively few competitors. These firms principally produce high technology electronics, but they are found in mature industries as well. The latter include firms in the nonelectrical machinery area that manufacture time recorders and life safety systems for buildings. chemical manufacturers that sell primarily to the pharmaceutical and photographic industries. and resin

Table 1. Classification of Monrachusett industries by product straregy Specialized

Product Cycle Firms:

Innovative production

Traditional mass production

Hybrid mass production

Firms:

Customized production

High technology electronics

Paper boxes

Apparel

High value-added metal fabrication

High technology plastics and resins

Low value-added plastics fabrication

Paper

High value-added plastics fabrication

Standardized furniture

Specialized furniture

Resin compounders

Low value-added metal fabrication High technology electronics

Machine tools

INTANGIBLE

compounders that are designing new materials. For these innovative firms, visible costs are relatively unimportant. Instead they value their Montachusett location for reasons such as the personal preferences of executives or because of family ties to the area. (b) Customizedfirms

FACTORS

1301

shifts in product lines, and a product mix that requires greater labor and capital flexibility than is characteristic of traditional, large-scale mass production firms. Illustrations of this type of niche-seeking behavior include the shift among paper producers toward specialized, technical, and industrial papers and the move among apparel firms toward specialized, luxury items.

and hybridfirms

Customized and hybrid businesses represent a distinct departure from the product cycle model.’ Customized and hybrid firms belong to mature industries. They produce traditional products, often using traditional technologies, but they are buffered from the competitive problems that trouble product cycle firms in mature industries, The most important competitive buffer arises from niche-seeking behavior. Customized and hybrid firms in Montachusett’s mature industries devise strategies for developing and marketing specialized products, and specialized services attached to products, which allow them to avoid product cycle maturity. Customized firms operate in protected market niches - concentrating on relatively short runs of products that are specialized to meet the demands of particular customers. The typical customized firm is small to medium-sized (fewer than 100 employees); it utilizes highly flexible labor and capital equipment; and it produces one-of-a-kind or small batches of product that are made to order.6 These firms have the capacity to produce diverse products and they must frequently switch from one product to another. Some of these firms have been in customized production for a long time, but many have only recently abandoned mass production for customized production in order to survive. Customized firms are found in the mold-making industry (such as manufacturers of molds for custom designed artificial limbs and joints), machinery and metal fabrication industries (such as manufacturers of prototype products for the defense industry), and in parts of the plastics fabrication and furniture industries. Hybrid firms, unlike customized firms, engage in mass production. Yet they create protected market niches by providing specialized “product services” - marketing, technical assistance in product design, and reliable quality assurance tied to an otherwise undifferentiated product. On the whole, these firms are somewhat larger than customized firms, and their technologies and production processes superficially resemble traditional mass production companies in their respective industries. However, they tend to have shorter production runs, more frequent

7. SUPPORTING CUSTOMIZED

COMPETITIVENESS IN AND HYBRID FIRMS

While niche-seeking behavior is the hallmark of customized and hybrid firms, the effectiveness of niche-seeking strategies depends upon a foundation of other factors. Some of these factors are firm specific, some are area specific, and some are generic to the local labor market. (a) Firm-specific organizational factors Competitiveness in customized and hybrid firms begins with a set of organizational features designed to exploit specialized product lines and flexibility in production. Chief among these features is the ability to respond flexibly and rapidly to product demand. Product demand shifts rapidly for customized and hybrid firms. One-of-a-kind markets come and go quickly and relatively short runs of mass produced goods must often be geared to meet tight production schedules. These considerations place a premium on rapid responsiveness to market demand. Rapid responsiveness in these firms is facilitated by a combination of nonbureaucratic organizational structures, skilled labor, versatile capital, and adaptable work scheduling and assignment practices. Unlike larger producers with higher overheads, more specialized machinery. and more bureaucratized production schedules, the smaller firms of the region have unusual flexibility in production that allows them to respond expeditiously to demand variations. This rapid responsiveness is an important source of the competitive edge that these firms have been able to obtain over other producers in the market. Rapid responsiveness has been particularly important in apparel, where local suppliers of mass produced goods can quickly replace exhausted retail supplies; in machining and metal fabricating, where urgent repair and replacement orders can be filled and small orders of particular parts can be supplied; and in the furniture

WORLD

1302

DEVELOPMENT

industry. where almost no inventories are carried. Such responsiveness has given these firms preferred access to national, and even international, markets.

(b) Area-wide agglomeration and collaboration factors Agglomeration factors are derived from the mix interaction among local firms and (Richardson, 1969, Ch. 2). The most common example of visible agglomeration factors is transportation costs, which are minimized by proximity among firms that buy and sell from one another. In the Montachusett region there are a few instances of such input-output relationships among local firms. Plastic fabricators purchase resins from local suppliers; a local box firm supplies packaging for a variety of local producers; machine shops make and repair molds for plastic fabricators (and occasionally do fabricating), manufacture parts for the paper industry, and handle subcontracts from local nonelectrical machinery manufacturers. These contracting relationships, however, are based far less on typical transportation cost advantages than on “collaboration economics” the ability to participate in, and respond rapidly to, changing design and manufacturing practices as a result of physical proximity and close working relationships among firms that buy and sell from one another. Where quality or speed are major factors, proximity allows firms to consult with one another quickly to resolve problems over product specifications and schedules. A typical example is in custom molding, where there is a need for considerable interplay among the resin firms that are producing specialized materials, the mold makers that design and repair molds, and the precision fabricators that produce the final product. For this cluster of firms, proximity is an important factor in both flexibility of production and cost competitiveness.

(c) Invisible laborfactors The third set of invisible factors exists in the labor market. These labor factors are associated with the area, rather than with particular firms. (i) Skills A major factor contributing

to the success of a

number of industries - nonelectrical machinery, metal fabricating, and mold making - is the skill of the labor force. These industries have depended upon a labor force in the region that has the ability to set up and operate precision machinery. Beyond these formal skills, these industries also require labor that has the mechanical aptitude to adapt existing machinery and operating procedures for the manufacture of ever-evolving, customized products. This local skilled labor pool has been developed over the years principally through informal on-the-job training in local companies. On-thejob training has been reinforced by strong traditions of father-son transfers of machining and woodworking skills. The strength of these informal sources of training has meant that vocational and trade schools have played a relatively minor role in the supply of skills. These traditional sources of skill, however, have been weakening in recent years. Demographic changes and emigration from the region have contracted the supply of young workers, and large numbers of skilled workers are reaching retirement age. This has already led to some spot labor shortages, such as for highly skilled, experienced machinists. (ii) Worker attitudes While the skilled labor market has long contributed to the strength of mature industries in the Montachusett economy, worker attitudes are becoming important as the economy moves toward specialized, custom production. The quality of skill in the local workforce receives high marks from employers, but skill advantages are outweighed by the more subtle qualities of these workers in terms of their attitudes and work ethic. There was high praise from local employers about the work commitment of the local labor force, especially from those managers who were able to make comparisons with worker attitudes in similar facilities at other locations in the country. Attitude and commitment were reported to be particularly important to maintaining a rapid response production capability and in sustaining product quality.’ (iii) Wage setting Both the interviews and related statistical analyses failed to reveal any growth benefits from the marginally lower wage rates in the region. However, this slight wage edge is indicative of the pay flexibility in the local labor market. Most of the region’s firms are wage “followers.” reacting to the local market, rather than being or following national wage wage ‘*leaders,”

INTANGIBLE

bargains, as is the case with large-scale industries such as steel and automobiles (Doeringer and Piore, 1985, Ch. 4; Mitchell, 1980, Ch. 5; and Flanagan, 1976). As a result, there are few institutional pressures to push wage rates out of line with competitive forces in the region. The region’s firms also seem to be better able to differentiate among occupational groups in setting their pay than would be customary in areas where unions and large-scale, bureaucratic firms are an important force in regulating the labor market. In large firms, the relative wages of skilled and unskilled workers are typically linked together within a rigid internal wage structure. In Montachusett, however. employers are able to pursue more differentiated wage strategies. Firms can target wage setting on specific occupational pools in order to compete selectively for labor. This gives them flexibility in wage setting and reduces labor costs since wage pressures from one occupation are less likely to spill over into other occupations. (iv) Labor-management relations In addition to the skills and attitudes of individual workers, the relationships between unions and management were a significant factor in business success. Indeed. econometric studies have shown that employment growth is inversely related to union militancy (Terkla and Doeringer, forthcoming; Wheat. 1985; Bartik, 1985; Plaut and Pluta, 1983). Montachusett has a strong tradition of unionization in both traditional mill industries and in machinery firms. While some companies had previously experienced difficult labor-management relations, the more successful, unionized firms were characterized by an improved labormanagement environment. This improved environment, however, was evidenced less by fewer strikes and reduced pressure on wage rates than by an accommodating approach to day-to-day issues of work rules, safety. and flexibility in the utilization of labor.” Conversely, the less successful unionized firms were facing unwillingness on the part of local unions to remedy inefficient work practices. Again, this judgment was reinforced by managers who were able to compare labor relations in Montachusett with those at company facilities in other regions of the country. Flexible labor relations have been easiest to achieve in small and medium-sized firms. The freedom from bureaucratized organization rules and the personalized and paternalistic labor relations practices that characterize such firms have helped them to realize the potential flexibility in the labor force and the capital stock.

1303

FACTORS

8. CONCLUSIONS The presence of customized and hybrid firms in the Montachusett economy illustrates the importance of product specialization as an alternative to subsidies and protection for rescuing mature industries. It further suggests that innovation and high technology are not the only avenues for establishing a competitive manufacturing sector. Those mature firms concentrating on more specialized products find their market niches defensible within surprisingly broad cost limits. Although there is always the danger that specialized markets will dry up and need to be replaced, or that other countries will eventually be able to compete in certain customized product areas, the ability to provide special services. such as product design, product maintenance and repair. and prompt delivery schedules safeguards against competition. Product specialization may thus be a successful strategy for long-term competitiveness. Other firms in the same industries that have failed to specialize have either closed or have become marginal in the region. Those that have been able to survive do so by relying on automating labor-intensive immigrant labor, production lines, developing more aggressive marketing programs, or using foreign product sources. These strategies have bought temporary relief from competitive pressures, but they have not fundamentally improved the underlying competitiveness of the firms. As a result. these firms remain vulnerable to their visible cost environment. Over time, business strategies emphasizing product specialization have dramatically changed the character of most mature industries in Montachusett. Some have made the transition from mass production to specialized and customized product lines. Others have become more heavily specialized in the distribution and marketing of goods that they no longer produce, while still others have shifted toward providing a variety of business and technical assistance services to accompany the products they manufacture. The corollary of specialization in production is that labor and capital must be relatively unspecialized and flexible. The capital equipment in the region, such as the machine and woodworking tools or plastics fabrication machinery, has always been characterized by general purpose applications. The labor force in the region has a broad skill base critical both for the operation of unspecialized machinery and for accommodating the product shifts associated with strategies of innovation and specialization.

1304

WORLD

DEVELOPMENT

(a) Implications for growth theory These findings suggest the need to recast our understanding of the factors that determine business performance and of the options that firms face in developing technology and product strategies. Theories of competitive advantage based solely on visible costs are not a sufficient explanation for regional growth. The discovery of an Important subset of customized and hybrid firms in the mature industry of the Montachusett region also introduces an important extension of the conventional product cycle model. In conventional product cycle analysis, mature firms are in trouble because they operate in saturated product markets where they are no longer cost competitive (Norton, 1986; Van Duijn. 1983; Wasson, 1978; Dhalla and Yuspeh, 1976). Similarly, mature regions are thought to be in trouble because of their industrial mix (Van Duijn, 1983). Too many firms operate in markets where regional factor costs put them at a competitive disadvantage. The importance of industrial mix has been borne out time and again by empirical studies of regional growth; the role of local factor costs, however, has been more ambiguous. The usual remedies suggested by the product cycle model are either to offset cost disadvantages through subsidies or to start the product cycle over again by attracting new and innovative industries that can draw upon existing cost advantages. Our findings partly confirm the basic thesis of the product cycle model - that many firms, industries, and regions evolve from innovation to maturity. Parts of the furniture, plastics, and metal fabricating industries are involved with product lines, introduced years ago, that are still being mass produced. Some of the product lines remain protected from competition by transportation cost barriers, but others are showing declining sales and employment. Similarly, there are machinery manufacturers and metal fabricators that are continuously developing new products that they hope to place into mass production. At the other end of the spectrum are innovative firms that are introducing new products and new methods. But the study also demonstrates that there exist various types of specialized products both custom-made products and mass produced products accompanied by specialized services that deviate from this pattern. Markets for some of these specialized products may never grow sufficiently large to warrant mass production. Common examples include the many one-of-akind products demanded by defense contracts

and the unique markets for many machine tool and fabricated metal products such as cutlery molds and paper-making screens. While sales may be recurrent, they are at a level that only justifies customized production. For others, such as fill-in apparel orders or dormitory furniture, demand allows for the scale economies of mass production, but is not sufficient to justify scale economies of marketing and distribution. For these products, services such as rapid delivery schedules, technical consulting assistance, and product troubleshooting create the same kinds of protected market niches that are found in customized production. By concentrating on such specialized product lines, firms in industries that have matured can survive, and even prosper, over the long term. In the Montachusett economy, the foundation of specialized production has largely been in mature products using mature technologies. The capability to achieve specialization in this way is the legacy of its mass production industries that have now substantially disappeared, and of the job shops that serviced these industries. Moreover, the same business strategies of specialization and the same qualities of flexibility of labor and capital that have sened these “postmaturity” firms so well have also allowed some of these firms (and new entrants as well) to expand into similar types of technologically more advanced markets. Specialized production. whether of mature or new products. seems to have achieved a permanent place in the IMontachusett economy. Postmaturity specialization thus provides an alternative to innovation as a source of economic revival - an alternative that need not require either sophisticated research and development (R&D) capabilities or advanced technologies. (b) Policies for reviving mature industries and economies

The ingredients for reviving mature industries and mature economies - encouraging specialization, the expansion of business and labor flexibility, and the identification and further development of highly localized, invisible cost advantages - argue strongly for rethinking local economic development policies. Currently, disproportionate attention is paid to policies that affect visible cost factors, such as taxes. energy costs, and wage rates - factors that in this study have proved much less important in influencing local business expansion and location decisions than the invisible factors analyzed - and to policies for attracting high technology firms.

INTANGIBLE

To the extent that visible factors were found to be important in this study, it was to those mature firms engaged in mass production for highly competitive markets. This suggests that subsidies of visible costs may be most attractive to firms whose long-term competitive prospects are grim, or to those mobile firms who are indifferent to the invisible strengths of the local economy. Instead, much more emphasis should be placed on developing the logic of invisible growth factors. This means directing policy attention toward identifying and strengthening invisible factors associated with specific firms and specific communities - the organizational and strategic characteristics of business firms, agglomeration economies involving the more qualitative product linkages among firms, and the extent to which different firms can draw upon invisible factors present in local economies. While our analysis has little to offer in terms of direct advice for businesses on how to improve their strategic behavior or how to create strong linkages among firms, it does carry implications for the targeting of economic development efforts and for improving the generic strengths of the labor market.

(c) Targeting growth policy One clear finding of the study is that the revival of mature industry depends substantially upon transforming mass production firms that are dependent upon visible cost advantages into firms that engage in customized or hybrid production. Rather than relying upon the “vintage” of a technology or a product as a guide for targeting economic development efforts, as is typical with programs to attract high technology industry, a region’s growth objectives would be better served by concentrating upon firms and product lines that draw strategically upon local invisible strengths. In the Montachusett region, this translates into policies that focus on strengthening established firms that have become more specialized, encouraging further specialization, and developing recruitment programs tailored to specific types of firms and production lines that depend more heavily upon these invisible factors.

FACTORS

1305

economically troubled. however, the skill pool is being disrupted. The younger, more mobile skilled workers have begun to leave the area while the aging of the labor force in mature industries is leading to a massive wave of retirements of the area’s most skilled and experienced workers. This combination of emigration and retirement not only generates a potential crisis of skill replacement over the next few years. but it also threatens traditional training arrangements whereby one generation of skilled workers prepared the next through on-the-job training. Traditionally, the schools were not expected to play a major role in maintaining the skill base of the local economy. However, the high skill levels, prevalence of small and medium-sized firms, and shifting employment patterns in the Montachusett economy, suggest that it may be time to revive apprenticeships to support the growth of customized firms.’ Apprenticeship prepares workers with sufficient skills to allow them to produce a wide variety of products. sometimes in different firms and often in a customized production setting. By training broadly, workers acquire more flexible skills and the ability to be productive without a long period of learning routine jobs through repetition. The resulting skills of the journeyman are the very kinds required by an economy engaged in specialized and customized production and where employment levels in particular firms are often in flux.

(e) The labor-management environment In Montachusett, there were numerous examples of negotiation over more flexible job assignments and work rules needed to produce more specialized product lines efficiently and to meet rapid response production schedules. There are also benefits to be had from extending these workplace approaches to the wider community, as has occasionally been tried in other parts of the country (Siegel and Wineberg, 1982, Chs. 3 and 4). Communities that develop a reputation for constructive and progressive labor relations should have an advantage in attracting industry as well as in expanding the existing employment base.

(d) Skill development

(f) Can this experience be generalized?

One of the strongest features of the Montachusett region has been a labor force of skilled craftsmen with a strong mechanical orientation. As the area’s mature industries have become

Specialized production is, of course, not a new phenomenon. Machine manufacturers have long catered to the unique equipment needs of their customers and product specialization has been

1306

WORLD

DEVELOPMENT

a long-standing survival strategy for the oldest mature industries such as textiles and paper (Wagoner, 1968; Estall, 1966; Commonwealth of Massachusetts, 1984, pp. 69 and 136). Moreover, the prospects for specialized production have been noted in several Western European countries (Piore and Sabel, 1983). What is important about the Montachusett experience is the potential it reveals for regaining competitiveness through business strategies that promote product specialization. The range of industries present in the Montachusett economy that have faced the choice between mass production and specialization is quite large and the patterns of success and failure across these industries suggest that the potential for economic revival through specialization strategies is not limited to a handful of industries having unique characteristics. While the range of mature industries examined and the underlying logic of the transition from mass production to specialized production argue for the general validity of the propositions about invisible factors that we have advanced, specialization will not work for all mature industries nor for all mature areas. The types of industries

for which the findings are least likely to apply are those so dominated by mass consumption markets and large-scale mass production technologies (e.g. autos, electronics assembly, and other standard products) that there is little scope for specialized production. lo Nor are these findings likely to apply to mature economies, such as that of Detroit, that are dominated by such largescale firms and that contain a workforce accustomed to rigidly defined industrial practices and adversarial labor-management relations (Flynn, 1984). There are, however. many local areas throughout the northern United States that resemble Montachusett in terms of the concentration of manufacturing jobs and the prevalence of small and medium-sized firms. These include LewistoniAuburn (Maine), Elizabeth (New Jersey), Warren (Ohio), Gettysburg (Pennsylvania) and Sheboygan (Wisconsin), to name a few.” Such areas account for over one-third of the manufacturing jobs in states such as Pennsylvania and New Jersey. It is in communities such as these that the lessons of Montachusett for economic development and labor market policies are most likely to be applicable.

NOTES

1. Studies which examine cyclical growth or decline have also generally concluded that the industrial composition of an area. and how the area’s industries are faring nationally, are more important than visible costs in determining regional employment change in the short term. See Howland (1979) and Browne (1982). 2. For example, Schmenner et al. (1987) demonstrate that firms make location decisions in two stages - selecting a region on the basis of highly visible cost considerations and then choosing a site within a region based on more qualitative considerations. 3. See Doeringer, Terkla, and Topakian (1987, ch. 3) for a brief economic history of the Montachusett region. Stone (1930) also provides useful background on the types of industries that developed in the region. 4. For the development of the product cycle see Kuznets (1930). Burns (1934). Alderfer and Michl (1942). Van Duiin (1983) and Norton (1986). For its application to iegibnal ‘growth, see ‘Krumme and Havter (1975). Norton and’ Rees (1979). Rees and Sta’fford ‘( 1984). US Congress, Office oi Technology Assessment (1984). Hekman (1980) provides a useful industry study of the regional product cycle. 5. Customized product firms are similar fo the “specialized” firms discussed by Piore and Sabel (1984). The term customized. however, is used to

connote firms that are one-of-a-kind producers.

or few-of-a-kind

6. For a further discussion of flexibility tion, see Piore and Sabel (1984).

in produc-

7. For a discussion of the economics of attitudes at the workplace see Akerlof (1982) and Doeringer (1984). 8. Piore and Sabel (1984) propose a similar explanation for the growth of non-union firms. See also Katz (1985, Ch. 7). 9. In the late 19th and early 20th centuries. many skilled workers in metalworking and woodworking factories received broad skill training through formal apprenticeship programs, often in conjunction with craft unions. During the early 20th century, Montachusert was home fo a well-known skill training scheme, the “Fitchburg Plan,” that involved part-time schooling and part-time work in the machine trades. See Wagoner (1968). however, that the example of the 10. We recognize, mini-mill in the steel industry has many of the characteristics of specialized production that we encountered in Montachusett. See Piore and Sabel (1984). For a complete list of such areas, see Doeringer, Il. Terkla, and Topakian (1987). pp, 102-104.

INTANGIBLE

1307

FACTORS

REFERENCES Abernathy, W. J., K. B. Clark, and A. Kantrow, Industrial Renaissance (New York: Basic Books Inc., 1983). Akerlof, G. A.. “Labor contracts as a partial gift exchange,” Quarterly Journal of Economics, Vol. 97. No. 4 ~November i982), pp.-54>570. Alderfer, E., and H. E. Michl. Economics ofAmerican Industry (New York: McGraw-Hill Book Co., 1942). Bartik, T., “Business location decisions in the U.S. Estimates of the effect of unionization. taxes, and other characteristics.” Journal of Business and Economic Statistics, Vol. 3, No. 1 (January 1985), pp. 14-22. Bluestone, B., and B. Harrison, The Deindustrialization of America (New York: Basic Books Inc. 1982). Browne, L. E.. “Two years of stagnation: A regional perspective,” New England Economic Review, (September/October 1982), pp. 35-44. Burns, A., Production Trends in the United States Since 1870 (New York: National Bureau of Economic Research, 193-t). Carlton, D. W., “Why new firms locate where they do: An econometric model,” in W. C. Wheaton (Ed.), Interregional Movements and Regional Growth (Wash&ton, DC: The Urban Institute, 1979). Chandler, A. D. Jr., The Visible Hand: The Managerial Revolution in American Business (Cambridge,“MA: MIT Press, 1977). Chandler, A. D. Jr., Strategy and Structure: Chapters in the History of the Industrial Enterprise (Cambridge, MA: MIT Press, 1962). Chandler. A. D. Jr., and H. Daems. Managerial Hierarchies (Cambridge, MA: Harvard University Press, 1980). Commonwealth of Massachusetts, Governor’s Commission on the Future of Mature Industries, Final Report and Appendices (Boston, MA: Commonwealth of Massachusetts, ‘June 1984). Dennison. E. F.. The Sources of Economic Growth in the United States (New York: Committee For Economic Development. 1962). Dhalla, N. K., and S. Yuspeh, “Forget the product life cycle concept,” Harvard Business Review, Vol. 54 (January-February 1976). pp. 102-112. Doeringer, P. B., “Internal labor markets and oaternalism in rural areas,” in P. Osterman (Ed.), Iniernal Labor Markets (Cambridge, MA: MIT Press, 1984). Doeringer, P. B., and M. J. Piore, Internal Labor Markets and Manpower Analysis (Armonk, NY: M. E. Sharpe Inc., 1985). Doeringer, P. B., D. G. Terkla, and G. T. Topakian. Invisible Factors In Local Economic Development (New York: Oxford University Press, 1987). Elbaum, B. L., and W. Lazonick, (Eds.) The Decline of the British Economy (Oxford: Oxford University Press, 1986). Estall, R. C., New England: A Study in Industrial Adjustment (New York: Praeger Publishers. 1966). Flanagan, R. J:, “Wage interdependence in unionized labor markets,” Brookings Papers on Economic Activity, No. 3 (1976). pp. 635-673. Flynn, P. M., “Lowell: A high tech success story,”

New England Economic Review (September/October 1984). pp. 39-49. Hekman, J. S., “The product cycle and New England textiles,” Quarterly Journal of Economics, Vol. 94. No. 4 (June 1980). pp. 697-717. cycle and long-run Howland, M., “The ‘business regional growth.” in W. C. Wheaton (Ed.). Inter. regional Movements and Regional Growth (Washington, DC: The Urban Institute. 1979). Katz, H. C., Shifting Gears (Cambridge, MA: MIT Press, 1985). Kieschnick, M., Taxes and Growth: Business Incendves and Economic Development (Washington, DC: Council of State Planning Agencies, 1981). Kotlowitz. A., and D. D. Buss, “Costly bait: Localities giveaways to large corporations cause growmg outcry,” Wall Street Journal (September 24. 1986), pp. ‘1, 27. Krumme, G., and R. Hayter. “Implications of Corporate strategies and product cycles adjustments for regional employment changes.” in L. Callings and D. F. Walker (Eds.), Locational Dynamics of Manufacturing Activities (New York: John Wiley and Sons, 1975). Kuznets. S.. Secular Movement in Production and Prices (Boston. MA: Houghton Mifflin, 1930). Lawrence; P. R., and D. Dier, Renewing American Industry (New York: The Free Press. 1983). Mitchell, D. J. B., Unions, Wages, and Inflation (Washington. DC: The Brookings Institution, 1980). policy and American reNorton, R. D.. “Industrial newal.” Journal of Economic Literature. Vol. 24. No. 1 (March 1986), pp. l-10. Norton, R. D.. and J. Rees. ‘.The oroduct cvcle and the spatial decentralization of American manufacturing.” Regional Studies, Vol. 13. No. 2 (August 1979). pp. 141-151. Peters, T. J.. and R. H. Waterman. Jr... In Search of Excellence (New York: Harper and Row, 1982). Piore. M. J., and C. Sabel, The Second Industrial Divide, Possibilities For Prosperity (New York: Basic Books, Inc., 1984). Plaut. T. R., and J. E. Pluta, “Business climate. taxes and expenditures, and state industrial growth in the United States,” Southern Economic Journal, Vol. 50 (1983) pp. 99-119. Porter, M. E., Competitive Advantage (New York: The Free Press, 1980). Rees, J., and H. Stafford, “High technology location and regional development: The theoretical base,” Appendix A, in US Congress, Office of Technology Assessment, Technology, Innovation and Regional Economic Development (Washington, DC: Government Printing Office, July 1984). Richardson, H. W., Regional Economics (New York: Praeger Publishers. 1969). Schmenner, R., Making Business Location Decisions (Enalewood Cliffs: NJ: Prentice-Hall. 1982). Schme&er, R.. J. Huber, and R. Cook. **Geographic differences and the location of new manufacturing facilities,” Journal of Urban Economics, Vol. 21 (1987). pp. 83-104. I

1308

WORLD

DEVELOPMENT

Siegel, I. H.. and E. Weinberg. Labor Managemenr Cooperarion (Kalamazoo, MC W. E. Upjohn Institute for Emplovment Research. 1982). Stone. 0. L., fiistky of Massachusetts Industries. Their Inception. Growth and Success, Vol. 2 (Boston, MA: The S. J. Clarke Publishing Co., 1930). Terkla, D. G., and P. B. Doeringer, “Explaining variations in employment growth: Structural and cvclical changes among states and local areas.” Jburnal of Uyban Ecoiomics (forthcoming). US Coneress. Office of Technoloev Assessment. Technolog;, Innovation, and Regioyal Economic Developmenr (Washington, DC: Office of Technology Assessment, July 19&t), Van Duijn, J. J.. The Long Wave in Economic Life (London: George Allen I% Unwin. 1983). Wagoner, H. D., The U.S. Machine Tool Industry From 1900 to 19.50 (Cambridge, MA: MIT Press, 1968).

Wasson. C. R.. Dynamic Competitive Strategy and Product Life Cycles (3rd edition (Austin. TX: Austin Press. 1978). Wasylenko. M., and T. McGuire, ‘-Jobs & taxes: The effect of business climate on states‘ employment growth rates.” National Tax Journal. Vol. 38, No. 4 (December 1985). pp. 197-511. Wheat. L.. “The determinants of 1963-77 regional v manufacturing growth: Why the South and West grow.” Journal of Regional Science. Vol. 26, No. 1 (1986). pp. 635-659. Wheaton. W. C., “Metropolitan growth. unemployment. and interregional factor mobility.” in W. C. Wheaton (Ed.), Interregional Movements and Regional Growth (Washington, DC: The Urban Institute. 1979).