Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

Human driving forces: commodity specialization in cotton

The contemporary landscape of the Llano Estacado is one of cotton, grazing cattle, and grain. Cotton dominates the agricultural scene, both economically and in land area. As early as 1932, a number of counties had devoted to cotton more than 50 per cent of the total cultivated land; in the cotton belt, the area planted ranged from 50 to 70 per cent (Gibson 1932, 253). By 1958, approximately 17 per cent of the cotton grown in the United States was produced in the area (Gordon 1961). This had increased to 30 per cent by 1980 (Bednarz and Ethridge 1990, 390). Most of the crop (60-70 per cent) is exported for use in making low-count yarns, thus making its production important not only to the region but to the US economy as well.

Economics drove the transition from ranching to cotton during the early decades of the twentieth century. A bale produced on 0.4 or 0.8 ha could bring US$50, the price of four or five yearlings, which might require 8-16 ha for grazing (Gordon 1961). Cotton also brought cash, a welcome addition to the largely barter economy of the early twentieth century. It was unclear to many growers in those days whether cotton would be as drought-resistant as the grain sorghum that was grown for feed. According to Gordon (1961), the Texas Agricultural Experiment Station at Lubbock was largely responsible for conducting studies that proved the comparable status of cotton vis--vis the time-tested sorghums. An abundant supply of pre-planting soil moisture was necessary to produce either a good cotton or grain sorghum crop; in-season rainfall was less important. For cotton, this meant leaving the soil bare from harvest to planting - a condition that exacerbated soil erosion. In both these respects, the climate is unforgiving October through April are the driest months, and wind speeds are highest in the spring, prior to planting. The Experiment Station also conducted experiments in the improvement of cotton culture that solved the problems peculiar to the region and allowed for the rapid development of the cotton industry (Gordon 1961).

Don L. Jones, noted agronomist and superintendent of the Experiment Station at Lubbock from 1917 to 1957, suggested five factors to explain the success of cotton in the study area: good soil; introduction of the row-crop tractor; development of a cotton plant that could do well with a relatively short growing season, scant rainfall, and high winds; the presence of groundwater for irrigation; and the self-reliance and adaptability of the people who settled the area (Jones 1959, 34). Other factors that must be added to the list are the availability of credit to finance irrigation, land, and machinery; yearly inputs; and the government's special commodity-farm programmes. Without these last two, it would have been impossible to maintain cotton production in the Southern High Plains.

Prior to its introduction to the study region and the far western United States, cotton traditionally was grown under conditions of higher humidity, longer growing season, and lower altitude than those on the Llano. The plant had to be bred to withstand drought and still make reasonable yields. Other important considerations were the percentage of lint (the fibre around the seed of unginned cotton), length of lint, length of lint vs. yield, quality vs. price, and storm resistance (Gordon 1961, 114). With the exception of yield, length of lint was perhaps the most important factor, because, below a minimum length, the cotton was not marketable. Length of staple (degree of quality of fibre) is highly weather dependent; if the rainfall is insufficient during boll (the pod of the plant) development in September, the length will be inferior. Nearly all of the cotton varieties grown in the area, however, produce adequate length.

Breeding for storm resistance was one of the most difficult problems. The high winds of the Plains created conditions not experienced in more traditional cotton areas. Low humidity, high altitude, and wind tended to cause the burrs to shrivel and let the fibre hang out onto the ground. Although the cotton could be salvaged, its dirtiness lowered the quality. If a rainstorm beat the lint into the ground, it would be totally lost. What was needed was a plant that would hold on to the fibre in the face of windstorms, but would yield the fibre to the mechanical fingers of the harvesting machines. Gordon (1961) wrote that this problem consumed the energies of the State Department of Agriculture's Lubbock Experiment Station for over two decades.

By 1934, a cotton-production culture unique to the area had evolved through the ingenuity and experimentation of farmers, ginners (owners of de-seeding operations), and scientists. Farms were much larger than those of the older cotton belt (in Texas, Louisiana, Mississippi, and other areas of the South) and the large, level fields required much less labour in production. A US Department of Agriculture study conducted between 1924 and 1927 found that it took about 96 hours of labour to produce 1 ha of cotton in Lubbock, compared with 380 hours in North Carolina (Browne 1937). The net production costs were comparably lower as well. A Lubbock Chamber of Commerce survey of the area's cotton industry in 1924 revealed that 60 per cent of the improved land was in cotton and that a total of 121 gins were operating. Land was still available and the opinion was that a farmer could pay for his farm with one cotton crop (Lubbock Morning Avalanche, 21 September 1924, p. 1, as cited in Gordon 1961).

This extreme optimism was not without its alarmists, and in the end they were proved right. The economy and the weather coincidentally went "bad." High production had suppressed prices at a time when demand had declined following World War I. In 1926, the price of cotton dropped drastically. An adverse season resulted in a crop loss in excess of 60 per cent in some communities (Gordon 1961, 151). Thus, the cotton farmer joined the appeal for relief assistance from the federal government. These years brought to an end the initial phase of cotton culture development on the Southern High Plains - a phase in which the individual was pitted against the uncertainties of climate and competitive market (Gordon 1961, 151).

Passage of the Agricultural Adjustment Act (AAA) in 1933 began the long-term direct intervention of the federal government in the cotton-growing industry. The AAA authorized the federal government to enter into agreements with farmers to control production by reducing areas devoted to basic crops, to store crops on the farm and make payment advances on them, to enter into marketing agreements with producers and handlers in order to stabilize product prices, and to levy processing taxes as a means of financing the crop-reduction programme. A large number of growers voluntarily curtailed their planting. The drought of 1934 reduced output even further. When the dryness continued into the early part of 1935, it was clear that it was not just another "dry spell."

Many regarded irrigation as the solution to these recurrent droughts, but capital was exceedingly scarce for farmers who had barely survived the depression and the weather. Irrigation was not a new concept to the Southern High Plains; wells coupled with windmills had served the cattle ranchers since the 1880s. By 1900 the existence of subsurface water in the area was widely known and publicized by the US Geological Survey. The search for water was largely a search for the economically feasible technology to bring the water to the surface in quantities sufficient for irrigation purposes (Green 1973). The first significant irrigation wells, installed in 1910 and 1911 in the vicinities of Hereford and Plainview (in Texas), were powered by vertical centrifugal pumps powered by gasoline engines and oil engines (Green 1973). Nevertheless, the real boom in irrigation did not occur until the 1930s and 1940s. The 1935 Census of Agriculture reports 2,916 irrigated farms in the Llano (see table 6.2). These were largely within the areas of shallow water. In New Mexico, truck farmers were irrigating. In Texas, cotton was the most important irrigated crop because high profits could be realized in spite of low prices.

None of this would have been possible without the extensive credit that was made available for the installation of the pumps. Local bankers, pump companies, and other concerns aided by federal government loan agencies organized to install and finance irrigation plants, forging another government/capital link in the growing cotton industry.

Table 6.2 Farm size and irrigation profile

  Study area Irrigated land in study area
Year Number of farms Average size(acres) Number of farms Acreage % of farms % cropland
1900 2,132 4,603 (N) 269 21,977 -a -a
    782.4 (T)        
1910 12,345 1,336 103,205a -b -a  
1920 12,357 817.9a -a -a -a -a
1925 19,510 1,144.1 -a -a -a -a
1930 33,802 1,157.4 -a -a -a -a
1935 35,880 1,119.3 2,916 116,872 8.13 1.22
1940 30,558 1,502.8 -a -a -a -a
1945 29,526 1,665.6 -a -a -a -a
1950 28,066 1,586.0 10,940 1,839,935 38.98 21.62
1954 26,290 1,745.7 14,536 3,096,291 55.29 34.99
1959 22,797 2,002.1 15,855 3,938,856 69.55 43.82
1964 20,018 2,207.5 14,648 4,099,247 72.27 44.53
1969 21,080 2,108.16 -a -a -a -a
1974 16,843 2,471.08 10,721 3,956,035 63.65 43.26
1979 17,026 2,507.45 9,793 3,973,184 57.50 44.35
1984 15,741 2,422.86 8,134 3,090,880 51.67 35.96

Source: US Bureau of the Census (1900-1984).
a. Incomplete data.
b. No data.
N = New Mexico
T= Texas

During and after World War II, an increase in domestic and foreign demand for most commodities, including cotton, encouraged an increase in irrigation and cotton production. The new pumping technology made the deep-water reserves of the Ogallala accessible, opening what appeared at the time to be an endless source. With ample groundwater and a plentiful supply of natural gas, the number of irrigation wells increased at the rate of approximately 2,000 per year for the Southern Plains area (Gordon 1961, 169). But during the years immediately prior to World War II, irrigation expansion had to compete with other technological innovations for capital. The row-crop tractor, first produced by the McCormick Company in 1925, was ideally suited to the light, rock-free soil and flat terrain of the plains (Jones 1959). Efficient utilization of the new tractors required the lister-planter and cultivating equipment. Because the introduction of the tractor coincided with the depression, conversion to tractors did not occur until the latter part of the 1930s. One study, cited in Gordon (1961), showed that 44 per cent of all tractors engaged in cotton production were located in the state of Texas, with the highest concentration found in the Plains area. Mule teams in the fields were rare by 1945.

Another important technological change in cotton culture was the invention of mechanical harvesters. Mechanical harvesters were popular even before the labour shortage created by World War II. By 1951, approximately 21,000 were in operation, a sufficient number to harvest all the cotton grown (Jones 1954). In addition, improvements in gin machinery made machine-harvested cotton more practical. Mechanical stripping created problems of trash in the seed cotton as burrs, leaves, stem, and other material were gathered by the machine. Storage of cotton on the field prior to ginning exacerbated the problem. A local person, Ennis E. Moss of Lubbock, invented the lint cleaner in 1950 and began (in 1958) manufacturing machines to clean the lint after ginning.

The Korean War again created an escalated demand for American cotton, and existing surpluses were quickly exhausted. High market prices encouraged planting and irrigation expansion, so, by 1953, there were again surpluses and the Secretary of Agriculture announced marketing quotas for the 1954 crop. Marketing quotas have been in effect ever since. Planting allotments accompanied the marketing quotas and were particularly severe where the most recent development had occurred. The response was to increase fertilizer and other inputs, thus increasing yields on the allowed area. As a result, total yields were barely decreased (see table 6.3). Parity prices virtually guaranteed commercial farmers on the Llano favourable commodity prices for their crops and a substantial profit. Thus, a most productive and profitable agricultural industry developed. Lubbock became a major cotton market and a centre for the processing of cotton and cotton seed (Hill 1986). Cattle-feeding and meat-packing were booming industries, as were agricultural implement and equipment suppliers and manufacturers. The manufacture and installation of irrigation equipment also became big business in the area. The 1950s and 1960s, in short, were the heyday for agricultural producers.

The 1965 Food and Agriculture Act, which lowered levels of price support to world market levels, combined with increasing costs of production to begin a turnaround in the agricultural economy of the region. Cotton provisions of the 1973 Agriculture and Consumer Protection Act had little or no effect on the level and stability of farm prices, consumer prices and supplies, or world markets. Cotton disaster payments were about the only provision of the Act that added to net farm income in 1974 and 1975. At the same time, farming costs and irrigation costs in particular were increasing. As groundwater levels fell, more energy was required to lift water to the surface, and specific yields declined. Energy costs increased by an order of magnitude between the late 1960s and the early 1980s. In several cases, irrigation costs constituted more than half of the variable costs of crop production on the South Plains of Texas. According to Hill (1986), the agricultural economy of the area continued to be sustained by the general inflation experienced in the United States from 1973 to 1981. Despite the temporarily inflated prices, inevitably the boom years were over - the limits of the water supply were slowly realized, credit was increasingly costly, government programmes were less supportive - and the slow erosion of prosperity on the Southern High Plains had begun.

Table 6.3 History of cotton production

Year Study area
Number of farms Acreage Bales
1925 12,776 907,697 325,362
1932 20,006 1,564,922 -
1935 16,152 777,264 181,311
1940 16,904 1,112,154 413,577
1945 12,024 1,057,395 584,426
1950 15,970 1,889,614 1,694,926
1955 16,521 2,193,384 1,483,058
1959 16,136 2,023,588 1,831,150
1964 14,306 1,998,591 1,875,863
1974 9,328 2,009,774 1,195,779
1978 9,225 3,416,940 1,910,728
1982 6,381 2,169,293 1,214,099

Source: US Bureau of the Census (1925-1982).

Contents - Previous - Next