Randall L. Patton, Kennesaw State University
Paul Krugman (1993, p. 5) has written that “the most striking feature of the geography of economic activity…. is surely concentration” (emphasis in the original). There are few better examples of highly concentrated economic activity than the U.S. carpet industry. Today, carpet mills located within a 65-mile radius of Dalton, Georgia, produce about 85% of the carpet sold in the U.S. market. The U.S. industry accounts for about 45% of the world’s carpet production. While many segments of the textile industry have struggled in the post-World War II era, carpet makers have prospered. The U.S. carpet industry also exemplifies the southward drift of textile production within the United States during the twentieth century. Indeed, it is probably useful to conceptualize the U.S. carpet industry as two distinct industries with different trajectories. The early American carpet industry was, like other textile segments, a product of borrowed (from the United Kingdom) technology and skill that struggled throughout its existence against imports. The second American carpet industry grew from deep southern roots and utilized locally developed technology and skills. The second industry also came along at just the right time to ride the boom in consumer spending associated with the economic golden age that followed World War II.
The First U.S. Carpet Industry
The first U.S. carpet industry emerged at the end of the eighteenth century. Skilled weavers produced carpets and rugs with handloom technology. In its early years, American carpet makers encountered the same problem as other textile manufacturers – imports. Congress protected the infant U.S. industry, along with textiles generally, in 1816 and raised protective tariffs in the 1820s. In an early survey of the industry conducted in 1834, Timothy Pitkin found 20 carpet mills producing about 1 million square yards. By 1850, a government survey found 116 mills producing 8 million square yards of carpets and rugs (employing more than 6,000 workers). Twenty years later, U.S. carpet mills numbered 215, wove more than 20 million square yards, and employed 12,000 persons. In the nineteenth century Americans used carpet to cover poor quality, soft wood floors. A commentator wrote in 1872 that the “general use of carpets was a necessity some few years ago, from the fact that the floors of our houses were generally built of such poor material, and in such a shiftless manner, that the floor was too unsightly to be left exposed” (Greeley, 1872). The mid-nineteenth century saw the introduction of the varnished hardwood floor. With the hardwood floor came a declining demand for wall-to-wall carpets and an increasing demand for smaller rugs to provide stylistic accents.
Employment and production figures indicate that, although there was an incremental increase in productivity, production effectively rose in concert with the number of workers. Erastus Bigelow introduced power loom technology for various types of carpeting in the early 1840s, and others quickly followed with competing designs. Though Bigelow’s idea – the use of power looms in carpet production – would eventually result in great productivity gains, Bigelow’s own looms were not the primary source of the gains, nor did those gains materialize overnight. Handloom production outweighed power loom production as late as the 1870s in the Philadelphia area. Power looms were expensive and manufacturers had great difficulty in matching the quality of goods produced with handlooms.
Boom and Bust in the First Half of the Twentieth Century
After 1870, refinements in power loom technology allowed manufacturers to produce reasonable substitutes for higher quality handloom woven goods. This resulted in a decline in the production of the cheapest carpets as consumers moved toward higher quality goods as the price of higher quality weaves declined. Large rugs became a staple in upper-middle class American homes by the early twentieth century. Sales ballooned to more than 83 million square yards by 1923. Firms such as Bigelow-Hartford produced lavish catalogs and advertised products direct to consumers in the early twentieth century, bypassing the traditional commission agents who had dominated marketing in the nineteenth century. The industry seemed, however, to have peaked in 1923. Sales fell off even before the Great Depression, and the economic disaster of the 1930s offered no respite. Firms such as Bigelow and Mohawk struggled. Industry production hovered in the 60 million square yard range throughout the 1930s. Most mills converted to war production during the Second World War, a move that helped forestall a deeper crisis. Just after World War II, the industry experienced a brief boom, with sales jumping to nearly 90 million square yards in 1948, but the boom quickly turned bust. Even the seemingly robust sales of 1948 amounted to a scant increase over the peak of a quarter century earlier. When compared with population growth, the industry’s sales had actually declined. Worse still, sales fell through the early 1950s back into the 60 million yard range.
The Second U.S. Carpet Industry
Carpet in the United States had three salient characteristics in 1950. Carpets were (1) woven on power looms out of (2) wool in (3) mills located in the northeastern United States. In just one short decade, each of those critical elements had changed dramatically. By 1960, most carpet in the United States was made on tufting machines from synthetic fibers such as nylon in factories located in the southeastern United States – and the vast majority of these new mills were located in and around the Appalachian foothills town of Dalton, Georgia.
The U.S. economy entered a prolonged boom period after World War II that many historians have labeled the “golden age.” The release of pent-up consumer demand associated with the sacrifices of World War II, Keyenesian government policies aimed at maintaining a high level of demand, and other factors helped produce a period of unparalleled economic growth. Northeastern carpet manufacturers tried a variety of approaches during the late 1940s and early 1950s to reverse their industry’s fortunes, but had little success. Annual per household carpet consumption stood at 1.97 square yards in 1950, virtually unchanged from the beginning of the twentieth century. Industry executives expressed increasing frustration throughout the early 1950s with their inability to tap the booming housing market of the postwar period. Many northern carpet mills began to open new plants in the South. Moving south allowed older firms to escape unionized work forces, take advantage of the region’s lower labor costs and, occasionally, benefit from incentives offered by state and local governments in the region (Greenville, Mississippi, built a $4 million facility to entice the Alexander Smith Company in the early 1950s, for example). Bigelow, Mohawk, and other northeastern companies built facilities in Virginia, South Carolina, Georgia, and Mississippi during the 1950s.
With few exceptions, these facilities produced carpet using weaving technology. The shining new mills in Greenville, Mississippi and Liberty, South Carolina, used the latest and most productive looms and were constructed according to the most up-to-date standards – single-floor construction and concrete floors, for example, to make the use of lift trucks possible. Yet the industry encountered one insurmountable barrier. In spite of decades of incremental progress, woven carpets were still too expensive to penetrate the working class market. The wholesale price of woven carpets rose slightly during the 1950s. The quite modest increases were interpreted within the industry as something of a success.
The woven carpet manufacturers also tried other strategies to boost sales in the 1950s. Some manufacturers experimented with selling carpet “on time” (credit) through retailers; others emphasized style and elegance. The chief impact of the advertising campaigns seems to have been to raise awareness of and desire for carpeting in general. In 1949, this would have seemed a winning strategy.
Tufted Textiles Take the Floor
During the same decade, however, a new southern industry produced a cheaper substitute for woven goods – tufted carpets and rugs, whose sales grew from near zero in the late 1940s to more than 100 million square yards by 1958. The origins of this new carpet industry in the South can be traced to a combination of purposeful action and historical accident.
The Tufted Bedspread Industry
The historical accident, as Krugman called it, was the revival of the hand tufting tradition in northwest Georgia (and elsewhere in the region) in the early twentieth century. To create a tufted bedspread, the craftsperson inserted raised tufts of yarn into a pre-woven piece of backing material (generally cotton sheeting) to form a pattern, then boiled the sheeting to shrink it and lock in the tufts of yarn. Catherine Evans, a young woman living near Dalton, Georgia, saw an old hand tufted bedspread at a friend’s house in 1895. Evans duplicated the design and made a similar spread as a wedding gift. Evans and some of her relatives began teaching other area women the art of tufting. From these beginnings, a cottage industry developed. By the 1920s, local entrepreneurs had created numerous “spread houses.” The spread houses operated a putting out system, sending “haulers” into the countryside with sheeting and yarn. The haulers returned later to pay the farm families for their hand work and pick up tufted spreads for finishing – washing and, for some, dyeing. These spreads found a ready market, not just regionally, but in the northeast as well. (Wannamaker’s department stores stocked Georgia bedspreads in the 1930s.) This cottage industry became a source of economic growth in north Georgia even during the Great Depression.
Here the residue of purposeful action intersected with Catherine Evans’ historical accident. By the 1920s, the South had become home to the lion’s share of U.S. textile production. Some of this shift southward was due to capital movement from North to South, but most of the shift could be accounted for by new southern firms – large firms such as Georgia’s West Point Manufacturing and North Carolina’s Burlington Mills and smaller firms like Dalton, Georgia’s Crown Cotton Mill and American Hosiery Mill. After the Civil War, and especially after 1880, southern firms had borrowed northern technology, begun at the bottom of the quality chain with the coarsest fabrics, and initiated what might be called a process of regional learning. Much of this development was the result of a purposeful effort to industrialize the region. By the early twentieth century, the South still had not developed a regional textile machine-making industry, but the cotton mills, hosiery mills, and other textile firms had recruited and trained a large number of mechanics to maintain machinery purchased in the northeast. Mechanics from the Dalton area and nearby Chattanooga began adapting sewing machines for the purpose of inserting raised yarn tufts, and in the early 1930s many of the spread houses moved toward becoming spread mills, or factories. Spread mill owners employed a largely female work force to operate the sewing machines that now created the raised patterns.
From Spread Mills to Carpet Mills
By the end of the 1930s, a number of these firms had begun to experiment with multi-needle machines that could tuft wider swaths of backing material more quickly. Some firms, such as the cleverly named Cabin Crafts (to conjure the image of a cottage industry that already had ceased to exist) had begun making small rugs by covering the entire surface of a piece of backing material with tufts. Hosiery mill mechanics like Albert and Joe Cobble founded firms in the southern industrial dynamo of Chattanooga, Tennessee (less than 30 miles from Dalton) to build special machines for the tufted bedspread and small rug industry. From these technological roots, area entrepreneurs began experimenting with making large rugs and wall-to-wall carpeting with this tufting process. About 1949, the Cobble Brothers firm and an innovative Dalton spread making company, Cabin Crafts, introduced tufting machinery wide enough to produce carpeting in a single pass. Carpet makers could buy cheap pre-woven backing materials. Manufacturers tried cotton with mostly poor results. Eventually Indian jute became the primary backing material for tufted carpets through the 1960s. In the 1970s, manufacturers developed suitable synthetic substitutes for jute.
The traditional woven carpet industry primarily used wool. (One manufacturer lamented in 1950 that it was “unfortunate that the carpet industry was tied to the back of a sheep.”) Wool made an excellent material for floor coverings – it was durable and resilient. The new southern tufting mills used cotton yarn at first. Cotton did not compare with wool as a floor covering material – it crushed easily and wore more quickly. Yet already by 1955, southern carpet mills were selling more carpets than northern mills, in spite of the clearly inferior nature of the product. The key was price: the wholesale price of tufted carpet was about half that of woven products. Consumer surveys in the 1950s demonstrated that few carpet buyers could name the manufacturer of the carpets they had purchased. The same consumers were almost without exception unable to distinguish between a tufted and a woven construction with a visual inspection. The old woven firms’ ad campaigns of the 1950s probably helped move more tufted carpet than woven.
The tufted carpet industry experienced a meteoric rise in the 1950s, but many skeptics saw it as a fad that would fade. One machinery executive quipped that “every year was the last big year for tufting” in the 1950s, according to industry observers. The obvious inferiority of cotton made the argument plausible. Surely consumers, many in the old woven industry argued, would eventually tire of placing glorified bedspreads on their floors. Tufted manufacturers experimented with rayon (disastrously) and staple (chopped, spun) nylon (with some success) in the 1950s. The most significant breakthrough in terms of raw materials came in the mid-1950s from the DuPont Corporation. Woven manufacturers and others had experimented with DuPont’s nylon as a carpet fiber, but nylon lacked the bulk needed in floor coverings. DuPont helped insure that the bust never came by developing bulked continuous filament (BCF) nylon in the mid-1950s. DuPont’s initiative was clearly stimulated by the growth of carpet sales. In essence, tufted manufacturers created a market large enough to justify DuPont’s research and development costs. DuPont even helped the new industry along by launching its own ad campaign for carpets made with its trademark 501 nylon in the late 1950s and early 1960s.
BCF nylon helped insure the long-term future of the tufted carpet industry. Tufted carpets used, and still use, a variety of fibers. Staple nylon could be used in constructions and styles that were not possible with a continuous filament yarn – plush, lustrous constructions. And in recent years, the industry has made increasing use polypropylene and other continuous filament yarns. DuPont’s BCF nylon (and similar products introduced by Monsanto a bit later), however, fit perfectly with the least expensive, low pile height, loop constructions that sold best in the emerging modest income market.
By the end of the 1950s, the new tufted carpet industry had raced past the old woven industry. While the total volume of carpet sales skyrocketed, woven sales actually fell. Tufted products accounted for all the growth in the industry through the 1970s. Tufted carpet sales increased from about 6 million square yards in 1951 to nearly 400 million yards in 1968. Carpet finally became a staple of middle and working class home furnishings – indeed, it became the default floor covering over much of the nation for decades. The logjam had been broken by product substitution. Per household sales increased for the first time since the turn of the century. By 1990, Americans consumed over 12 square yards of carpet per family per year, up from 1.97 in the early 1950s. Woven sales drifted downward in the same period from 67 million yards to just over 40 million. Woven products did not disappear. High-end consumers still sought the assumed quality of woven goods, and woven products continued to dominate specialty commercial markets – hotel lobbies, casinos, etc. But tufted carpet achieved total dominance of not just the residential carpet market, but the residential flooring market in general.
Average Mill Value of Carpet Shipments, 1950-1965 (price per square yard)
|All Broadloom Carpet and Rugs||Woven||Tufted|
Carpet Industry Output, 1951-1968 (square yards)
|Tufted Carpet Shipments(square yards)||Woven Carpet Shipments(square yards)||Total IndustryShipments
The tufted carpet industry was the nation’s fourth fastest growing industry in the 1960s, trailing only aircraft, television picture tubes, and computers. Robert Shaw, CEO of Shaw Industries, for two decades the nation’s leading manufacturer of carpet, recalled the late 1950s and 1960s as the era of the “gold coast” in the Dalton area, an era in which demand constantly outstripped supply and small manufacturers and large could succeed with few controls and a “seat-of-the-pants” management style.
Carpet Capital: An Industrial District
The brief narrative sketched above outlines the emergence of an industrial district. By the 1960s, the district had developed several distinct features. The carpet complex was characterized by the rapid emergence of new firms. No single firm accounted for as much ten percent of the industry’s output. The industry had developed from the deep roots of textile manufacture and, specifically, bedspread making. Carpet making emerged out of a process of regional learning (albeit a small region, similar to Jane Jacobs’ “city regions”). Carpet manufacture was also a decentralized affair. A few large firms, such as Cabin Crafts and E.T. Barwick Mills, spun some of their own yarn and finished some of their own carpets in-house by the 1960s, but most of the hundreds of small firms relied on independent yarn spinning or production mills and independent commission finishing firms. Carpet finishing provided the industry with significant flexibility. Mills produced some carpets with pre-dyed yarns, but tufted significant yardage with undyed yarn. This allowed manufacturers to delay the critical decision on color until later, increasing the company’s flexibility. Commission finishing companies provided these services. Initially post-production dyeing was handled in dye becks, or large drums. That is, finishers dyed carpets by the piece (albeit large pieces, 900 feet or more in length). Dye becks were produced locally and regionally.
The Dalton district offered a classic example of the great Victorian economist Alfred Marshall’s industrial district based on external economies. Clearly this industry originated in northwest Georgia because of the peculiar skill set developed among managers, mechanics, and workers. The finishing companies and other suppliers clearly filled the role of Marshall’s “subsidiary trades” devoted “to one small branch of the process of production.” Innovation and ideas were “in the air,” as Marshall put it. With so many firms and workers in close proximity, improvements in technology, management practices, marketing, and other arenas were rapidly transmitted throughout the industry. Though different in many ways, Paul Krugman has observed, the relatively low-tech carpet industry of the Appalachian foothills was quite similar to the high-tech Silicon Valley in these respects.
In the 1960s, European firms introduced continuous dyeing equipment to the U.S. carpet market. Continuous dyeing equipment held out the potential for more effective use of mass production techniques – an endless stream of white carpet moving through a dye range capable of rapidly shifting colors. The continuous ranges were, however, frightfully expensive compared to dye becks. The relative expense of the equipment in this evolving industry offers a window into the strategic options available to management. A tufting machine might have sold for $10,000 in the late 1950s, with Cobble Brothers or some other firm offering in-house financing. Through the 1960s, the well-nigh indestructible tufting machines were available second-hand – a bit slower than brand new models installed by larger mills, but still effective for smaller product runs. That particular barrier to entry into this new industry was quite low. To establish a beck dyeing operation, the equipment alone would have cost more than $700,000 by the end of the 1960s. The stakes in finishing were much higher, but the risks were shared among the finisher and his many customers. Just one of the new continuous dye ranges in the early 1970s cost more than $800,000. The capital stakes rose for finishers.
The Maturing of the Industry
The carpet boom slowed in the 1970s as did the rest of the US economy. The recessions of the mid-1970s brought an end to the double-digit annual growth rates of the earlier period. In a slower growth environment, attention to cost became critical. Some firms adapted to the changing environment, but many did not. Adaptation generally involved vertical integration. Particularly during the 1980s, a few firms took the lead in bringing yarn spinning (and eventually production of extruded, continuous filament yarn) in-house, integrating backward toward raw materials. The most successful large manufacturers also integrated forward through finishing, investing in their own dyeing facilities. The recession of 1981-82 proved a pivotal moment – many smaller and mid-sized firms had continued to struggle along and occasionally prosper during the inflationary 1970s. The recession of the early 1980s claimed nearly half of the 285 mills that had been in operation in 1980; by 1992 the industry counted only about 100 mills, down dramatically from its early 1970s peak of more than 400. Shaw Industries, a revamped Mohawk Industries, and a few others bought competitors and moved the industry towards greater consolidation. Moreover, the top four firms, led by Shaw Industries, accounted for more than 80% of total production by the early 1990s.
The Industry Today
The carpet industry today is essentially the domain of a few large firms, led by Shaw Industries and Mohawk. The nation’s largest carpet making firms are headquartered in northwest Georgia. Shaw and other carpet firms have moved into the production and distribution of other flooring surfaces – tile, wood, vinyl, etc. – as carpet has slipped in market share. No longer the unchallenged leader in covering America’s floors, carpet is still the single most popular choice. Perhaps the most notable change associated with the industry today is its increasing use of workers of Hispanic descent. Since the late 1980s, Hispanic immigrants have moved in large numbers to Dalton, as they have to many new destinations throughout the nation. The region’s employers laud the immigrant workers as the saviors of the industry, a solution to the region’s recurrent labor shortages. Some community leaders and longtime residents express anxiety about the pace of cultural change in the small communities that still serve as hosts to the industry.
Cole, Arthur H., and Harold Williamson. The American Carpet Manufacture: A History and an Analysis. Cambridge, MA: Harvard University Press, 1926.
Deaton, Thomas M. From Bedspreads to Broadloom: The Story of the Tufted Carpet Industry. Acton, MA: Tapestry Press, 1993.
Ewing, John S., and Nancy Norton. Broadlooms and Businessmen: A History of the Bigelow-Sanford Company. Cambridge, MA: Harvard University Press, 1955.
Flamming, Douglas. Creating the Modern South: Millhands and Managers in Dalton, Georgia, 1884-1984. Chapel Hill: University of North Carolina Press, 1992.
Friedman, Tami J. “Communities in Competition: Capital Migration and Plant Relocation in the United States Carpet Industry, 1929-1975.” Ph.D. diss., Columbia University, 2001.
Greeley, Horace, et al. Great Industries of the United States: Being an Historical Summary of the Origin, Growth, andPerfection of the Chief Industrial Arts ofThis County. Hartford: J.B. Burr and Hyde, 1872.
Krugman, Paul. Geography and Trade. Cambridge, MA: MIT Press, 1993.
Patton, Randall L. Shaw Industries: A History. Athens, GA: University of Georgia Press, 2002.
Patton, Randall L., with David B. Parker. Carpet Capital: The Rise of a New South Industry. Athens, GA: University of Georgia Press, 1999.
Scranton, Philip. Proprietary Capitalism: The Textile Manufacture at Philadelphia, 1800-1885. Cambridge, MA: Cambridge University Press, 1985.
Scranton, Philip. Figured Tapestry: Production, Markets, and Power in Philadelphia Textiles, 1885-1941. Cambridge, MA: Cambridge University Press, 1989.
Walters, Billie J. and James O. Wheeler, “Localization Economies in the American Carpet Industry.” Geographical Review 74 (Spring 1984): 183-91.
Zuniga, Victor, and Ruben Hernandez-Leon, “Making Carpet by the Mile: The Emergence of a Mexican Immigrant Community in an Industrial Community of the U.S. Historic South.” Social Science Quarterly 81, no. 1 (2000): 49-66.
Citation: Patton, Randall. “A History of the U.S. Carpet Industry”. EH.Net Encyclopedia, edited by Robert Whaples. September 22, 2006. URL http://eh.net/encyclopedia/a-history-of-the-u-s-carpet-industry/