First-Mover and Late-Developer Advantages Ch 2

Chapter 2 “Created” Externalities and the Regulation of Cut-throat Competition in the Union Era The ILGWU pledged to improve the economic conditions of the garment crafts by raising wages, reducing work weeks, making employment more secure, and ending substandard workplace conditions and other sweatshop practices. In trying to deliver on this promise, however, New York’s local ILGWU unions had to address organized employer opposition to collective bargaining, a diverse membership scattered among many small and medium-sized firms that posed difficulties for organizing and bargaining, and the practices of cut-throat competition that threatened to limit economic gains for union members (Levine, 1969, pp. 110, 114-116). Overcoming Employer Opposition While New York’s share of national women’s wear markets continued to grow in the early 1900s, it faced growing external competition from other districts with lower factor costs and fewer strikes. Employers feared that bargaining with a locals affiliated with a national union based in New York would lead to even more strikes and costly increases in wages The union strategy for overcoming for overcoming employer opposition was to rely on district-wide strikes to force employers to the bargaining table and the New York ILGWU unions were reasonably successful in achieving that goal. The first steps toward district-wide unionization were taken in 1909 and 1910 when ILGWU locals successfully launched large and long “general” strikes (the 1909 “Uprising of the 20,000” in support of shirtwaist and dress workers and the 1910 “Great Revolt” involving about 60,000 strikers in support of cloak and suit workers). These strikes, along with pressure from civic leaders and government officials, forced employers to recognize the ILGWU unions and to turn to their industry associations to represent them in negotiating collective bargaining agreements. The ILGWU and the Regulation of Cut-throat Competition Controlling cut-throat competition proved to be a longer-term challenge involving the development and gradual refinement of a bargaining agenda for stabilizing competition among employers in the district and for limiting market entry by non-union and other low-cost apparel suppliers from outside the district. External competition in particular was explicitly recognized as a major constraint on raising wages as early as 1910 during the ILGWU’s discussions of bargaining strategy in preparation for the 1910 cloak and coat negotiations (Carpenter, 1972, pp. 48-49) and the arbitrators of labor disputes under that agreement later cautioned the union and the employers’ association that they needed to extend bargaining coverage to firms outside of New York “for their own good” (Carpenter, 1972, p. 560). The early labor contracts contained traditional labor benefits of wage increases, improved working conditions, and greater employment security for union members. However, the 1910 cloak and coat agreement (known as the “Protocol of Peace”) also included provisions that affected market structure and district efficiency. 5 This contract provided for arbitration procedures to prevent the disruption of production by strikes over both worker grievances and future contract negotiations; it stabilized employment among manufacturers by limiting the amount of contracting they were allowed; and it established a systematic structure of minimum wages that recognized the different market conditions in various product sectors (Levine, 1969, pp. 194, 196-98; Carpenter, 1972, Ch. 2). This combination of economic benefits for workers, structural change affecting market competition, and the efficiencies of reducing strikes and stabilizing employment was subsequently replicated in agreements covering New York City’s other major product specializations in 1913. These contracts, however, failed to adequately control cut-throat competition in the district. Restrictions on contracting were hard to enforce and the small size and frequent turnover among firms made it difficult for unions to maintain sufficient organizational strength to limit non-union competition within the district. The threat of competition from other districts proved even harder to contain and a common employer complaint in later years was that “out-oftown reproductions” of their garments could be purchased far below New York prices (Carpenter, 1972, p. 61). Avoiding work stoppages also proved problematic as arbitration procedures became congested from the large numbers of shop-floor disputes and spontaneous strikes and employer lockouts became more frequent alternatives to arbitration. Finally, between changing labor market conditions and union wage bargaining, New York was losing its wage advantage over other districts. For example, the average annual earnings of apparel workers in New York City were 58% above those in Los Angeles by 1921 (Levine, 1969, Table VII). The inability of unions to block non-union competition and control disruptive strikes, along with the recession of 1920-21and the revival of a nation-wide “open shop” movement among employers (Jacoby, 1985, Ch. 6), led New York garment employers to abandon their 5 Because skilled workers often became supervisors in apparel factories and were also a source of entrepreneurial talent, a “management” perspective on competition and efficiency was present in the early craft-based locals in New York and the decision by the New York City locals to establish a national union affiliation with the “businessoriented” AFL also reflected an understanding of the importance of profits as a source of revenue for improving wages and working conditions. (See Taft, 1957 and Galenson, 1960). collective bargaining agreements in the early 1920s. Ending collective bargaining, however, still left the New York apparel industry with the problems of spontaneous strikes, inefficient competition from marginal firms within the district, and growing competition from suppliers outside of New York. Recognizing that an enforceable district-wide solution to these problems was needed “to control unfair and destructive competition between jobbers and submanufacturers [contractors]” (Carpenter, 1972, p. 73), and that the New York industry lacked the means to develop and legally enforce its own mechanisms for market stabilization and restructuring, the major garment employers soon authorized their associations to re-open negotiations with the unions and to develop a more effective system of market regulation through collective bargaining (Carpenter, 1972, p. 564). 6 By this time, the New York industry had largely completed the transition from artisan and craft production to an industrial contracting model consisting of a network of manufacturers, jobbers, and contractors and the ILGWU devised a comprehensive strategy for economic regulation that reflected these changes. The union insisted that the 1924 cloak and suit negotiations adopt coordinated bargaining with the three employer associations representing jobbers, manufacturers, and contractors, and jobbers were made the focal point for regulating the industry in recognition of the central position they occupied in the industry. The union proposals also envisioned a major restructuring of the industry to reduce excess capacity, stabilize employment, and “remove” wages and working conditions from competition (Levine, 1969, pp. 422-424). The employers’ associations in the cloak and suit sector initially rejected union’s plan, but subsequently accepted a more limited proposal recommended by a mediation panel appointed by the Governor of New York (Levine, 1969, pp. 425-27; Carpenter, 1972, pp. 115-138, 326366). The 1924 agreement included several of the key elements of the union’s bargaining agenda: (1) the obligation of jobbers and manufacturers to deal exclusively with contractors and 6 Contractual provisions for fixing prices, limiting entry, and requiring exclusive dealing are the kinds of monopolistic restraints on trade that would have been illegal if conducted by employers. However, such practices were exempt from anti-trust prosecution under Section 8 of the 1914 Clayton Act, provided they were reached through collective bargaining and were for the benefit of employees. De jure, the regulation of cut-throat competition to protect ILGWU employment and economic benefits met this test for over four decades of collective bargaining in New York City. other suppliers that had active contracts with the union (2) the establishment of a minimum employment size for manufacturers and contractors, (3) the recognition that jobbers had contractual responsibilities for the terms and conditions of employment offered by their contractors covered by the contract, and (4) the requirement that jobbers (as well as manufacturers and contractors) participate in district-wide procedures for improving shop-floor working conditions and resolving industrial disputes. The ILGWU was reported to represent over 90% of women’s wear employment in the district during this period (Levine, 1969, p. 424), giving it effective control over almost all women’s wear employment, and many of the provisions of the cloak and suit agreement were subsequently extended to contracts covering other product sectors. Further contract enhancements for regulating competition were negotiated through the late 1920s (Teper, 1937, pp. 19-20, 28; Carpenter, 1972, pp. 115-138, 326-366, 488-489). Limits were placed on the number of contractors each jobber could employ in order to control excess capacity; contracting relationships were made more permanent by designating the specific contractors assigned to each jobber; contract prices were required to cover fixed as well as variable costs in order to reduce turnover among contractors; jobbers were obliged to share work equitably among their contractors; and the peaceful settlement of labor disputes was strengthened by establishing full-time paid positions for the chairmen of arbitration procedures. To reduce competition from outside the district, equal prices were to be paid for comparable contracted work regardless of whether it was performed by New York contractors or elsewhere, penalties were levied against runaway shops leaving the district, and the union had a contractual obligation to organize competing employers outside the New York City district. Unions were also the major enforcement arm for these agreements through strikes and boycotts against firms that violated their contractual obligations, and union sanctions were backed as needed by judicial remedies. By the end of the 1920s, a transition from cut-throat competition to market governance through collective bargaining was well underway. District-wide Efficiencies “Created” by Union Regulation While monopolistic restraints on trade are normally considered to be inefficient, many of these practices had positive efficiency consequences for the district when compared with the inefficiencies of cut-throat competition. Compulsory arbitration, for example, helped to reduce the disruptions of spontaneous and unauthorized strikes, controlling excess capacity improved resource allocation, price-fixing reduced uncertainty, and minimum size-of-firm requirements helped to eliminate marginal firms and ensure that surviving firms could achieve minimum efficient scales of production. In addition, the ILGWU directly contributed to district efficiency by encouraging mechanization and efficient work practices by providing training and industrial engineering services to help unionized firms improve their production systems and work organization (Disher, 1947). Efficiency effects also flowed from the pairing of jobbers with specific contractors, and from the preferences granted to these dedicated contractors in the awarding of contracts and in the equitable sharing of work at uniform contract prices. Continuing and exclusive relationships between jobbers and contractors provided new opportunities for investments in relationshipspecific knowledge that improved both the productivity of contractors and the efficiency of contracting relationships. Jobbers could learn the strengths and weaknesses of their dedicated contractors and train them to be more efficient, contractors could learn about their jobbers’ products and expected quality standards, and stronger channels of communication between jobbers and contractors could help jobbers to secure orders and improve profits. These stable contracting relationships applied primarily to the “core” group of experienced contractors in New York City who worked mostly for one jobber and had a regular flow of orders, in contrast to a remaining group of “peripheral” contractors who had highly-competitive, arms-length market relationships with multiple jobbers. 7 Regulation of cut-throat competition through collective bargaining reportedly caused some employers to leave New York (Levine, 1969, Ch. XXXII; Carpenter, 1972, pp. 547-551) and prompted resistance to unions in other districts. However, the employers who remained in New York had come to realize by the end of the 1920s that the regulation of product competition through collective bargaining generated both monopoly rents and efficiency gains that could be profitably shared with their unions while also allowing the district’s market share to continue to rise. 7 Stable jobber-contractor relationships and the distinction between the core and periphery sectors of the district have persisted in a modified form long after collective bargaining ceased to be an effective regulator of competition in the New York garment district (Uzzi, 1997, Palpacuer, 1997, Doeringer and Crean, 2006). The Great Depression of the 1930s, however, challenged this union-regulated system of monopoly protections and regulatory efficiencies as New York City employers sought to cut costs and prices in the face of sharply reduced demand for clothing and as union strength was eroded by membership losses. Nevertheless, regulation of the industry through collective bargaining continued in New York City, aided in part by the market stabilization and supportive union policies of the 1933 National Industrial Recovery Act (NRA), which translated many of the district’s collective bargaining regulations into the NRA industry codes (Carpenter, 1972, Chs. 16-18). As a result, New York City’s market share in women’s wear remained close to its peak of about 75% through at least 1935 when the NRA was declared unconstitutional. Estimates of Efficiency Benefits The magnitude of the district efficiencies created in New York City from the beginning of collective bargaining in 1909 through the early 1930s are revealed in studies and hearings conducted by the NRA in conjunction with its examination of wage differentials among different garment regions. Testimony by the head of an employer association based in Chicago, for example, provides a qualitative perspective on these efficiencies from a unionized district that competed with New York City: “New York is a natural market, it gets volume orders, its methods of production are superior, it has a much greater percentage of men workers and these are men who are anxious to work hard, its selling costs are about 5% less than Chicago, it has less inventory loss because it can buy from hand to mouth, its houses need not spend time or money to learn what is new in material and style, it is not required to expend sizable sums annually for express charges, and its product is generally made by machine whereas about 75% of the operations here are by hand” (Women’s Wear Daily, Sept. 19, 1933, cited in Carpenter, 1972, p. 616). Quantitative analyses of unit production costs in different regions conducted by the NRA further confirm that New York’s efficiency advantages were substantial. For example, wages for major occupations such as cutters and sewing machine operators in other regions would need to be anywhere from 25% to 40% below those in New York in order to offset New York’s unit cost advantages (Carpenter, 1972, p. 614). In the two decades of more or less continuing collective bargaining between 1910 and 1930, New York City also continued to see its market share grow. With the demise of the NRA in 1935, the apparel industry in New York largely returned to earlier system of regulation through collective bargaining under the new union and bargaining protections of the National Labor Relations Act in1935, and later under the auspices of the dispute settlement and wage stabilization policies of the National War Labor Board during World War II. However, wartime mobilization also strengthened garment districts outside of New York City and by the end of World War II New York had a smaller, but nonetheless commanding market share of 54%. Late Development and Market Competition in Los Angeles Los Angeles was still part of Mexico and a largely agricultural and ranching community in the 1840s when New York was becoming a fashion center. After Los Angeles was annexed to the United States at the end of the Mexican-American War in 1848, it began to develop good port facilities and railroad networks and, by 1900, it had a garment industry specializing in cloaks and coats that served a small regional market. While it subsequently added specializations in work clothes and sportswear, its apparel market remained largely regional through the 1920s, 1930s, and early 1940s. Employers in the Los Angeles area were aggressively anti-union in the late 19th century. Major employers formed the Merchants and Manufacturers Association (MMA) in 1896 to promote an :open-shop: industrial climate and the MMA advised apparel industry employers on how to implement anti-union strategies -- discharging and blacklisting union members, replacing strikers, and lobbying for pro-business labor policies (Milkman, 2006, p. 35; Laslett and Tyler, 1989, p.15). This anti-union climate was further reinforced by an active Ku Klux Klan movement (Lassiter, 1953, p. 171), and by the influential Los Angeles Times newspaper that had long opposed radical movements and was a strong advocate for low labor costs and freedom from unions as the basis for the development of the Los Angeles economy (Gottlieb and Wolf, 1977). The apparel industry was no exception to employer opposition to unions. There were sporadic attempts by the ILGWU to organize workers in the fledgling apparel industry prior to the late 1920s, but these had only minor success and were often short-lived. For example, an ILGWU local was established in 1907, but survived for only 3 months (Laslett and Tyler, 1989, p. 15) and a women’s cloak and suit local of skilled tailors working for small contractors was established in 1910, but this local had negotiated only 17 contracts with individual employers by 1919 (Laslett and Tyler,1989, pp. 15, 18-19). The ILGWU never represented more that 4% of employment in women’s wear prior to 1930 (Laslett and Tyler, 1989, Appendix 4, p. 119); there were no unions in the dress sector until 1923 or in sportswear until 1937 (Laslett and Tyler, 1989, p. 20); and the only garment union to gain a significant foothold in Los Angeles before 1930 was a tailors’ local that succeeded in negotiating contracts covering the vast majority of skilled male workers in the women’s cloak and suit sector in 1924 (Laslett and Tyler, 1989, pp. 19-20). As Los Angeles was poised to enter national markets for sportswear in the late 1920s and early 1930s, it had a full complement of agglomeration economies needed to support women’s sportswear production – a pool of labor of immigrant labor, a regional base of suppliers and buyers, and a rapidly-developing base of knowledge about the design and manufacturing of sportswear. It also had advantages of lower labor costs and more abundant land than New York City, and it remained free of unions as the coalition of merchants, manufacturers, and the media, successfully mobilized public opinion and civic power to suppress union organizing in women’s wear. Nevertheless, the Los Angeles district needed to consider whether any changes were needed in market structures and business practices in order to match the efficiencies of the industry in New York City. Coincidentally, New York’s employers and unions went through a self-assessment covering many of these same issues at the beginning of the 1920s when cut-throat competition returned to the district when employers abandoned collective bargaining after a decade of attempts to stabilize competition. The leading academic expert on the New York City garment district during this period reported an industry consensus that larger firms and greater market concentration were needed to gain scale economies and to help control cut-throat competition, and that scope economies needed to be expanded through greater product diversification. The industry then debated three options for achieving these goals: (1) the continuation of the jobbercontractor system, which was already moving towards larger firms; (2) the replacement of the jobber-contractor system with the earlier model of inside manufacturing with minimal outside contracting; or (3) the restoration of regulation through collective bargaining under a tripartite arrangement whereby manufacturers and jobbers would continue to grow in size and scope to gain further efficiencies in marketing, contractors would become larger and better able to manage garment production efficiently, and unions would regulate cut-throat competition and eliminate marginal contractors (Levine, 1969, pp. 417-417). While the employers in the New York garment district chose the option of continuing the jobber-contractor system and returning to collective bargaining to regulate the industry in the 1920s, Los Angeles’ employers chose to continue the district’s industrial structure that was organized primarily around manufacturers with inside shops with relatively little contracting, and to keep in place its opposition to unions and collective bargaining. From the perspective of the Los Angeles apparel manufacturers, including some “runaway” manufacturers from New York City, the lessons from New York, and from the closer experience with unions in San Francisco (Lassiter and Tyler, 1989, pp. 170-7), were clear. The jobber-contractor system offered no particular advantages over the manufacturer-dominated system in Los Angeles. Manufacturers were good at design, marketing, the organization of economies of scale and scope, and at drawing upon contractors as needed, while unions and collective bargaining led to major inefficiencies from disruptive strikes, costly wage increases, and interference with management prerogatives. As a result, Los Angeles opted for unregulated markets and relatively low factor costs as the efficient way for competing in national markets. Cut-throat competition among firms remained prevalent through the 1920s, although it was somewhat muted by the presence of larger manufacturers and the lesser reliance on small contractors than in New York. The wage cuts and collapsing employment at the start of the Depression triggered a wave of garment union organizing efforts and strikes in Los Angeles between 1930 and 1933. However, these were initially unsuccessful because the Los Angeles police arrested picketers and protected strikebreakers (Laslett and Tyler, 1989, ch. 5). Employer opposition to unions remained intense after the passage of the NRA and the NLRA in the 1930s (Laslett and Tyler, 1989, chs. 2-7), but union organizing and bargaining were somewhat more successful towards the end of the NRA and under the NLRA. About 25% of the women’s wear workers in Los Angeles were reported to be union members by 1936 (Laslett and Tyler, 1989, pp. 41-42). However, impasses during negotiations to extend union organization and improve contracts in 1936 resulted in a number of strikes that prompted strong employer retaliation. Strikers were arrested for disturbing the peace, employers subsidized police details to prevent pickets from blocking access to their factories, and a second employer association (the Southern California Manufacturers Association) was formed in 1937 to promote the open shop movement in the region (Laslett and Tyler, pp. 37-41). Together the MMA and the Southern California Manufacturers Association advised employers on how to establish company unions that would pass muster under the NLRA and they lobbied successfully for local legislation that limited picketing. Nevertheless, garment union membership overall had climbed to about one-third of the industry during the Depression (Laslett and Tyler, 1989, Appendix 4, p. 119) and ILGWU locals had almost 16,000 members in over 600 factories by 1939 (Laslett and Tyler, 1989, p. 42). Apparel production grew in Los Angeles during wartime mobilization and unions made further organizing gains under wartime industrial relations policies. It was estimated that unions represented over 50% of Los Angeles’ apparel workers during World War II (Laslett and Tyler, 1989, Appendix 4, p. 119) and the ILGWU also succeeded in securing some of the limitations on contracting found in New York City’s collective agreements by using militant bargaining tactics. For example, a major contract was negotiated with several sportswear and dress employer associations in Los Angeles that contained a number of restraint of trade practices (Carpenter, pp. 841-47; Federal Trade Commission, In the Matter of the California Sportswear and Dress Association et al., 1955), but only after an unusual wartime strike in 1942. Developments after World War II New York City was still the largest and most unionized U.S, garment district at the end of World War II and much of its system of market regulation through collective bargaining remained intact (Carpenter, 1972, pp. 853-859). Its overall market share had fallen, but New York City still accounted for more than half of all value-added in women’s wear and almost three-fourths of the blouse market (U.S. v. Greater Blouse, Skirt, and Neckwear Contractors’ Association, et al., U.S. District Court, Southern District of New York, CR 153/181, 1959, cited in Carpenter, 1972, p. 852). Nevertheless, a number of factors beyond the control of district regulation were working against production in New York City. The trend nationally was for production to decentralize towards districts in the South and West where land and labor costs were lower, fabric could be sourced more cheaply from textile suppliers that had relocated to the South, and unions were weaker. Fashion preferences were also shifting away from New York City’s traditional specializations in dresses and tailored clothing in favor of more casual blouses, skirts, and trousers, and many of these products could be produced more efficiently in large factories using mass production methods than by the smaller-scale and flexible production networks prevalent in New York. Entry by low-cost non-union suppliers was reducing New York’s market share and Los Angeles, in particular, was beginning its climb to become the rival fashion district to New York. The decentralization of production and the growing number of large manufacturers also made it easier for suppliers to sell directly to retailers instead of going through jobbers and buyers in New York City. Market entry by other districts in the 1940s and 1950s was followed by entry by offshore suppliers beginning in the late 1960s, and this import penetration was subsequently reinforced by a series of trade liberalization policies that the apparel unions were unable to block -- the Caribbean Basin Initiative (1985), NAFTA (1994), and the more general reductions in apparel tariffs and quotas from 1994 to 2005 under the Uruguay Round of GATT negotiations. The New York district began its long-term decline in the 1970s as output, employment, and union membership started to fall. The ILGWU union locals represented less than half the apparel employees in the district by the 1980s and only 30-40% in the 1990s (derived from Levitan, 1998, quoted in Milkman, p.88). Apparel employment in New York City is now under 20,000 (2009), about one tenth of it peak employment and roughly the same level as in 1890, and the district lacks a sufficient union bargaining presence to regulate competition and employer conduct to any significant degree. Nevertheless, many of the efficiencies of relationships persisted in the shrinking “core” sector of the New York City garment district. Meanwhile, Los Angeles continued to prosper as a “late-developing” district. It had agglomeration economies similar to those of New York, an established reputation for sportswear design, a manufacturer-contractor system with somewhat larger firms than New York that suited its sportswear specialization, and the wartime strength of the Los Angeles garment unions proved to be temporary. The union’s remaining strength in Los Angeles was greatest among craft workers in the rapidly declining coat and suit sector while it was weakest in the post-war growth sectors of sportswear and dresses where the ILGWU represented only 10-15% of the workforce during wartime bargaining (Laslett and Tyler, 1989, pp. 45, 49, 52, 56, 62-63). An ILGWU organizing drive in sportswear was launched shortly after the end of World War II, but was met with strong employer resistance, substantial financial support from the MMA, and court injunctions against picketing (Laslett and Tyler, p. 58). The garment industry unionization rate in Los Angeles County quickly fell from its wartime high of about 50% to 24% by 1947 (derived from Lassiter, pp. 11, 66-67, 72-73; Laslett and Tyler, p. 56) and was only 9% by 1979 (Laslett and Tyler, Appendix 4, p. 119). 8 Los Angeles returned to an unregulated market as its apparel unions weakened and collective bargaining gradually disappeared from its post-war apparel industry (Laslett and Tyler, 1989, pp. 61-62, 84-87). Recent accounts confirm that arms-length competition was once again the norm for contracting in Los Angeles by the 1980s (Ross, 2004; Bonacich and Appelbaum, 2000; Milkman, 2006). In other respects, however, the New York City and Los Angeles are becoming more alike. Manufacturers in Los Angeles are concentrating more and more on design and marketing while relying heavily on contractors for garment assembly; there is considerable overlap in product specializations, and unions and collective bargaining have diminished substantially in New York. Both districts are experiencing declines in output and employment as a result of foreign competition and Los Angeles replaced New York as the largest U.S. garment district in the late 1980s and early 1990s. Epilogue: Is there a Future for Market Regulation in the U.S. Apparel Industry? The purpose of this paper is to show how economic institutions can redesign markets and augment traditional agglomeration economies, using historical evidence from the two largest garment districts in the United States. The key economic institutions in New York were the 8 ILGWU membership fell from a high of over 12,000 in 1945 to under 10,000 in 1953, under 7,000 in 1962, under 5,400 in 1972, to 3,700 by the end of 1979, and to 3,000 in 1988 (Laslett and Tyler, pp. 92, 99). Using different data on ILGWU union membership as a fraction of total apparel employment in Los Angeles, Milkman also reports sharp declines in ILGWU membership -- from about 32% of all Los Angeles apparel workers in 1950 to only 4% in 1980, and down to 2.1% by 1992 (Milkman, p. 89). Even where unions remained in place, employers sometimes failed to comply with their collective bargaining agreements (Federal Trade Commission, In the Matter of the California Sportswear and Dress Association et al., 1955). ILGWU and the industry associations with which it bargained. The performance advantages that were created are rooted in the re-design of highly-competitive, but nonetheless inefficient, supplier markets through a series of “regulations” contained in collective bargaining agreements negotiated and enforced during the era of powerful garment unions from 1909 through the 1970s and 1980s. Collective bargaining helped to unify fragmented apparel supply chains in the New York City garment district and to replace inefficient cut-throat competition with a blend of monopoly and efficiency practices that stabilized contracting relationships, improved industry performance, and created new district externalities, while also improving wages and working conditions. Except for a brief period during the late 1930s and in World War II, apparel unions in Los Angeles never came close to being strong enough to regulate competition in the district. Highly-competitive market conditions have generally prevailed throughout the history of the Los Angeles district and contributed to Los Angeles eventually overtaking New York as the largest U.S. garment district. The summary evidence of the performance advantages from regulation through collective bargaining during the decades of strong garment unions in New York City is found in the data on the district’s share of domestic output and value added (see Table 1). New York was the preeminent “first-mover” garment district in the United States. It secured a dominant share of national women’s wear markets by the 1880s and 1890s based on its abundant agglomeration economies, a labor market with increasing supplies of skilled immigrant labor, and a set of “cutthroat” competitive practices that kept wages lower and overall production costs lower than those in other garment districts. By 1899 New York accounted for 64.5% of the U.S. clothing market and its market share had increased further to 69.3% by the end of the district’s non-union period in 1909. New York City’s market share rose further during the formative period of market redesign from 1909 to 1921 as the ILGWU local unions helped to replace chaotic cut-throat competition with more orderly and efficient competition through a series of collective bargaining agreements negotiated with apparel industry employer associations. Collective bargaining helped to protect the district’s market share by erecting entry barriers to New York’s markets, Table 1 Market Shares: New York City, Los Angeles County, New York State, California (Selected Years) Los New York New York California Angeles City % County % State % % 1899a 64.5 1909 a 69.3 1921 a 74.2 75.8 b 1925 1931 b 79.0 75.1 1.5 2.5 1933 a 72.8 2.7 1935 c 73.2 2.8 1947 d 54.4 4.6 58.9 5.7 1958 d 41.3 6.3 44.9 7.5 1967 d 35.7 6.1 38.2 7.7 1977 d 26.8 11.9 29.3 14.3 1987 d 22.8 12.8 24.6 20.9 1992 d 21.9 18.9 23.1 24.5 e 20.0 22.8 21.8 28.5 1997 Sources: a = Levine, 1969: (Value of Shipments – SIC Basis); b = Census of Manufacturers: (Value of Shipments – SIC Basis); c = Teper, 1937: (Value of Shipments – SIC Basis); d = Census of Manufacturers: (Value Added – SIC 233 & 234); e = Census of Manufacturers: (Value Added – NAICS 315212 & 31523) but it also redesigned market structures so as to reduce strikes, make competition more stable, and improve the efficiency of firms and contracting relationships. Market re-design began with the “Protocol of Peace” agreement in 1910 and was substantially extended in subsequent collective bargaining agreements through the 1920s as major revisions in collective bargaining regulations were made in recognition of the growing influence of jobbers in the New York district. The district’s market share in women’s wear rose to 74.2% by 1921 and it held an even larger market share in its leading specializations --79% of women’s tailored suits, coats, and skirts and 77% of shirtwaists and dresses (derived from Levine, 1969, p. 515). The available state-level data point to further growth in New York City’s market share through at least the mid-1920s as market regulation under collective bargaining was expanded. 9 These increases in market share under union regulation came at least in part from the efficiency gains that allowed New York City to retain many of its existing markets while also drawing production from other large districts such as Philadelphia, Cleveland, and Boston (Levine, pp. 385) and much of this gain proved to be robust through at least the first half of the Depression. New York City’s dominant market position proved remarkably resistant to entry by entry by firms in districts with lower factor costs and unregulated markets. Although New York City subsequently lost market share during World War II to Los Angeles and other districts, it still accounted for over half of all domestic value-added in women’s wear in 1947 (compared to 4.6% in Los Angeles), more than 40% in the 1950s, and over one third in the 1960s. Los Angeles was the second largest U.S. garment district after the war, but it made only gradual gains in market share until the 1970s when increasing domestic and international competition began to diminish the ability of unions to regulate the New York district. However, New York City was still supplying over 25% of domestic value-added in women’s wear in 1977, more than twice that of Los Angeles, and New York continued to hold its lead into the mid-1990s. While it may be tempting to attribute New York’s dominant market shares to the trade restraints negotiated with the ILGWU or hard-to-replicate agglomeration economies associated with New York City’s larger “culture economy” (Scott, 1996; Florida, 2004) the evidence points in other directions. New York City’s markets were, in fact, successfully contested throughout the union era. Production of more standardized products with lower fashion-content was continually being decentralized from New York to districts with lower factor costs, greater scale economies, and newer plant and equipment, and suppliers in other nearby districts also competed successfully for contracts from New York-based jobbers and manufacturers in more-fashionable product lines (U.S. v. Greater Blouse, Skirt, and Neckwear Contractors’ Association, et al., U.S. District Court, Southern District of New York, CR 153/181, 1959, cited in Carpenter, 1972, p. 852). The strongest evidence of the contestability of New York City’s apparel markets is 9 Changes in state-level market shares are a reasonable proxy for those in New York City because the New York garment district accounted for 98% of all women’s wear production in New York State in 1921 and 91%-95% between 1947 and 1992. provided by Los Angeles’ ability to become a credible competitor to New York by building its own pool of fashion and design knowledge and eventually over-take New York. 10 Moreover, the economic stabilization and price fixing arrangements in New York City’s apparel industry bargaining agreements faced repeated legal challenges beginning in the late 1930s as the Federal Trade Commission and the Department of Justice sought to define the appropriate balance between legitimate trade restraints negotiated in support of workers’ interests (which had immunity under the Clayton Act) and those where collective bargaining was largely a mechanism by which employers could impose otherwise illegal restraints on trade (see Carpenter, 1972, Chs. 21 & 22 for a detailed examination of these cases). Sometimes these legal challenges ended in judicial rulings barring specific anti-competitive practices contained in collective bargaining agreements and other times they ended in lengthy stalemates, but the continuing threat of these such investigations is reported to have had a chilling effect on the use of restrictive trade practices under collective bargaining agreements in the apparel industry (Carpenter, 1972, pp. 862-869). On balance, New York’s restrictive trade practices proved to be a more effective regulator of competition from within the district while its ability to regulate external competition depended more on keeping its unit costs competitive with districts that had lower factor costs and often greater scale economies. What set New York apart from competing districts was its ability to augment its substantial traditional agglomeration economies with collective efficiencies created through collaborative collective bargaining. The magnitude of this advantage is apparent from the various studies conducted by NRA during the early 1930s and from the commanding lead New York maintained for decades over other districts in terms of its shares of domestic production and value added. Los Angeles, by comparison, has benefited from the long-term shift towards casual fashion and the scale economies that are possible in sportswear manufacturing. Manufacturing units remain larger in Los Angeles than New York (an average of 22 employees in 2008 compared to 15 in New York), there are more large-scale manufacturers and contractors, and a larger fraction of output consists of the mass production of moderately-priced products such as 10 It has also been suggested that Los Angeles growth in the 1970s and 1980s was driven by the rapid growth of contracting with unregulated sweatshops that substantially lowered the district’s production costs (Light, 2006). blouses and shirts. Unregulated competition and arms-length market relationships between manufacturers and contractors may also be a more efficient way of organizing production of medium-priced sportswear, than the more-regulated contracting relationships in New York, and there is some evidence that Los Angeles’ opposition to market regulation may have given it a further cost advantage from the adoption of illegal sweatshop practices (Bonacich and Appelbaum, 2000; Light, 2006; Weil, 2007). Viewed over the last century, late-developing Los Angeles market-based production strategy would seem to be superior to New York’s more-regulated production system. It is presently the largest U.S. district, with more than three times as many employees as New York, a larger share of the remaining domestic output of women’s wear, and a lower rate of job loss. However, New York is performing better than Los Angeles by measures such as productivity and pay and its surviving firms appear to be returning to the districts original model of craft-like production, “core” contracting relationships, and small establishments producing small batches of products with higher fashion content than Los Angeles. The apparel industry in both New York and Los Angeles has shown a remarkable ability to survive in the face of over forty years of rising import competition and it remains to be seen which market design has the better odds of survival in the future.