Recent Developments in the World Cotton Market and the Future Outlook
Author:
LEYLA U. ECEVIT
Search for other papers by LEYLA U. ECEVIT in
Current site
Google Scholar Close
Publication Date:
01 Jan 1977
eISBN:
9781451947557
Language:
English
Keywords:
SP; income; price combination; polyester price; cotton consumption; polyester staple price; price variable; Consumption; Agricultural commodities; Exports; Agricultural prices; Stocks; price consideration
Download PDF (1.5 MB)

Cotton is still the most widely used textile fiber, accounting for more than one half of world fiber consumption after decades of severe competition from man-made substitutes.1 During recent years, remarkable changes have taken place in the world cotton market. Largely as a result of a deliberate change in government policy, in the late 1960s the United States began to maintain smaller carry-over stocks, and cotton prices became more sensitive to changes in output and demand. Consequently, international cotton prices,2 which were quite stable on both a month-to-month and a year-to-year basis during the 1960s (moving within the range of $0.27 to $0.33 a pound with an arithmetic average of $0.29 a pound), began an upward trend as early as 1971. During 1973, prices increased sharply and reached an all-time peak of $0.89 a pound in January 1974. Then the cycle was completed in January 1975 when prices fell to $0.47 a pound. The phase of slowly increasing prices that followed continued throughout 1975, and during December of that year prices averaged $0.59 a pound. Sharp increases occurred during the first seven months of 1976, and by July prices reached a new peak ($0.88 a pound). Cotton prices have stayed close to this peak since then (averaging $0.85 a pound in the second half of 1976 and $0.83 a pound in the first quarter of 1977) because of decreased availability. This occurred in spite of severe competition from the low-priced man-made substitutes.

Abstract

Cotton is still the most widely used textile fiber, accounting for more than one half of world fiber consumption after decades of severe competition from man-made substitutes.1 During recent years, remarkable changes have taken place in the world cotton market. Largely as a result of a deliberate change in government policy, in the late 1960s the United States began to maintain smaller carry-over stocks, and cotton prices became more sensitive to changes in output and demand. Consequently, international cotton prices,2 which were quite stable on both a month-to-month and a year-to-year basis during the 1960s (moving within the range of $0.27 to $0.33 a pound with an arithmetic average of $0.29 a pound), began an upward trend as early as 1971. During 1973, prices increased sharply and reached an all-time peak of $0.89 a pound in January 1974. Then the cycle was completed in January 1975 when prices fell to $0.47 a pound. The phase of slowly increasing prices that followed continued throughout 1975, and during December of that year prices averaged $0.59 a pound. Sharp increases occurred during the first seven months of 1976, and by July prices reached a new peak ($0.88 a pound). Cotton prices have stayed close to this peak since then (averaging $0.85 a pound in the second half of 1976 and $0.83 a pound in the first quarter of 1977) because of decreased availability. This occurred in spite of severe competition from the low-priced man-made substitutes.

Cotton is still the most widely used textile fiber, accounting for more than one half of world fiber consumption after decades of severe competition from man-made substitutes.1 During recent years, remarkable changes have taken place in the world cotton market. Largely as a result of a deliberate change in government policy, in the late 1960s the United States began to maintain smaller carry-over stocks, and cotton prices became more sensitive to changes in output and demand. Consequently, international cotton prices,2 which were quite stable on both a month-to-month and a year-to-year basis during the 1960s (moving within the range of $0.27 to $0.33 a pound with an arithmetic average of $0.29 a pound), began an upward trend as early as 1971. During 1973, prices increased sharply and reached an all-time peak of $0.89 a pound in January 1974. Then the cycle was completed in January 1975 when prices fell to $0.47 a pound. The phase of slowly increasing prices that followed continued throughout 1975, and during December of that year prices averaged $0.59 a pound. Sharp increases occurred during the first seven months of 1976, and by July prices reached a new peak ($0.88 a pound). Cotton prices have stayed close to this peak since then (averaging $0.85 a pound in the second half of 1976 and $0.83 a pound in the first quarter of 1977) because of decreased availability. This occurred in spite of severe competition from the low-priced man-made substitutes.

Another remarkable feature of the world cotton market during the calendar years 1974 and 1975 was the apparent halt in the long-term declining trend of cotton’s relative share in total fiber consumption. This occurred as a result of absolute declines in man-made fiber consumption during those years, whereas cotton use, as measured by the consumption of cotton by textile mills, stabilized around its 1973 level. Preliminary indications on world consumption of all textile fibers during 1976 point to substantial increases over the 1975 figures. The available information is not sufficient to determine the change, if any, in cotton’s relative share. The statistical analysis, on a semiannual basis, indicates that consumption of cotton is negatively related to cotton prices and positively related to man-made fiber prices with lags of one year in each case; furthermore, it has a positive relationship to cotton consumed one year earlier and a positive trend over time. The fitted regression equation explains quite well the fluctuations in cotton consumption that occurred during the period 1958-75. Projections based on the regression equation indicate an increase of 7.6 per cent in cotton consumption during 1976 over 1975, and a further increase of 3 per cent for 1977.

I. Changes in Supply and Demand

consumption

The share of cotton in total fiber consumption, which had a long-term declining tendency, decreased further during the period 1960-73, from 65 per cent to 50 per cent, although cotton consumption increased at an average annual rate of 2 per cent. During the years 1973-75, cotton usage stabilized at about 13.4 million metric tons. Consumption of other fibers showed a tendency to decrease after 1973; thus, the share of cotton increased in 1974 and 1975 by 0.7 percentage point and 1.2 percentage points, respectively, after reaching its lowest point of about 50 per cent in 1973 (Table 1).

Table 1.

Certain Textile Fibers: World Consumption, 1960-751

article image
Sources: International Cotton Advisory Committee; Commonwealth Secretariat; United Nations Food and Agriculture Organization; Textile Economics Bureau, Inc., Textile Organon (various issues); and author’s estimates.

For natural fibers, data basically refer to consumption by mills. For man-made fibers, production is taken as a reasonable proxy for consumption.

Cellulosic (artificial) fibers include rayon and acetate.

Noncellulosic (synthetic) fibers include acrylic, modacrylic, nylon, and polyester.

This turnaround in cotton’s relative situation was the outcome of a number of factors. One was a more favorable behavior of cotton prices vis-à-vis those of man-made fibers (Chart 1).3 Cotton prices declined throughout 1974 (albeit, after sharp increases in 1973), followed by only moderate increases in 1975. Man-made fiber prices, on the other hand, which had demonstrated a declining long-term trend, increased rather sharply during 1974 owing to rising costs. Moreover, the supply of man-made fibers, which are manufactured from petrochemicals, became tight partly as a result of the oil embargo and oil shortages. Prices declined somewhat in the first half of 1975 in response to continuing weak demand, but they increased later in the year to near-1974 levels.

Chart 1.
Chart 1.

Monthly Cotton Prices (Liverpool Index), Man-Made Fiber Prices,1 and Semiannual World Consumption of Cotton, 1970-76

Citation: IMF Staff Papers 1977, 002; 10.5089/9781451947557.024.A005

1 As a proxy for man-made fiber prices, the market price series for polyester, dacron, type 54, 1.5 denier in the United States has been used.

In addition, consumer preference for the fashionable “natural look” and the popularity of cotton denims and corduroys have been effective in protecting the demand for cotton from the adverse effects of the economic slowdown of recent years. At the same time, the 1974-75 recession in the industrialized countries decreased the use of man-made fibers. Since the share of man-made fibers in total consumption is the highest in these countries, weak demand conditions, which resulted in reduced mill activity, meant lower usage of man-made fibers. However, textile production in centrally planned countries and in those developing countries with domestically oriented textile industries was not generally affected by the recession and continued to grow during 1974-75.

OUTPUT AND PRICES

The traditional market situation for cotton until the beginning of the 1970s was one of general oversupply, with carry-over stocks at the beginning of the season averaging more than six months" consumption. U. S. Government policies initiated in the early 1930s, and aimed at bolstering farm income, had gradually turned into a very complicated system of production controls, including official determinations of grades and staples to be planted, price supports, loan rate schemes, and subsidies on both export and domestic prices. As a result of this program, the Government accumulated large surpluses of raw cotton. The disposal of these surpluses from time to time in the domestic and international markets, or the ever-existing possibility of such sales, have played a significantly depressing role on the world cotton market. Consequently, cotton prices moved little until the early 1970s, fluctuations being on the order of a few cents on both a month-to-month basis and from year to year (Appendix, Table 10). In 1971, as a result of a strong increase in demand following several years of slow growth, cotton prices increased sharply.4 Then in 1972, prices averaged about 7 per cent higher than those in 1971. Yet the big leap came in 1973, when prices more than doubled during the year and reached unprecedented levels as demand continued strong, and there was a tight supply situation for some desired staples and grades. Fears of a reduced output because of the smaller acreage in the crop year 1973/74, and current and projected shortages in made-made fibers as a result of the energy crisis, further strengthened cotton prices. Rising inflation all over the world naturally added to the pressures. The peak was reached in January 1974 and was succeeded by a steady decline that continued into the first quarter of 1975. The decreasing trend was caused by slackening demand in the textile industry as inventories were reduced in response to the deep recession into which industrialized countries quickly fell in the first half of 1974. By the time the planting decisions for 1974/75 were made, however, cotton prices were still about 65 per cent above their levels of a year earlier. World acreage increased to 33.3 million hectares, second only to the record in 1972/73 (Table 2). This produced a depressing effect on prices, especially since demand was sluggish. As yields turned out to be very high, the 1974/75 crop became a record at 14.0 million tons, which pushed prices down further, since once again there was an oversupply of cotton (Table 3). Demand started to pick up slowly after the first few months of 1975, carrying prices with it. Yet, the average price for 1975 was still about one fifth below the 1974 average.

Table 2.

Cotton: World Production, Acreage, Yield, and Mill Consumption, 1961-75

article image
Sources: Derived from International Cotton Advisory Committee, Cotton—World Statistics (various issues).

Beginning on August 1.

Includes the Union of Soviet Socialist Republics, the People’s Republic of China, and Eastern European countries.

Preliminary estimates.

Table 3.

Cotton: World Supply and Disposition, 1972/73-1975/761

(in millions of metric tons)

article image
Sources: International Cotton Advisory Committee and U.S. Department of Agriculture estimates.

Crop years.

Preliminary estimates.

Includes the Union of Soviet Socialist Republics, the People’s Republic of China, and Eastern European countries.

When planting decisions for the 1975/76 season were made, cotton prices were low compared with both costs of production and prices of alternative crops. As a result, there was a sizable shift of acreage to food and oilseed crops, which, combined with a decline in yields, produced a 16 per cent drop in output.5 This large decline brought about a decrease (of about 5 per cent) in the total supply of cotton in 1975/76 in spite of the record stocks at the beginning of the season (Table 3). Since consumption during the 1975/76 season was up by 7 per cent from the previous year (according to estimates by the International Cotton Advisory Committee), cotton prices started an upward trend as demand began to pick up in 1975. Prospects of only a moderate increase in the next year’s production (1976/77) and, therefore, a continuing tight supply situation added to the pressure, and the consequent increase in international cotton prices between December 1975 and July 1976 was about 50 per cent. Latest estimates of the 1976/77 crop indicate an increase of about 6 per cent from the previous season, which, together with the relatively low level of beginning stocks, puts the total supply of cotton for 1976/77 at its lowest level since 1971/72. As a result, world cotton prices have eased only marginally since the peak in July 1976, in spite of the strong competition from man-made fibers. During the first quarter of 1977, there was a slight recovery in prices from the decline in January.

To sum up, the cotton market in recent years has become much more volatile than it used to be, shifting from a chronic situation of over-supply to one of fluctuations between oversupply and undersupply, with relatively lower carry-over stocks than before. At the same time, cotton prices have shown a general upward trend, although with sharp fluctuations. In this respect, it seems ironic that during recent years cotton has been able to improve its relative position in the fiber market, whereas previously it was continuously losing ground to man-made fibers when its price was relatively low and stable.

Another important change in recent years relates specifically to the geographic distribution of cotton production and consumption. Between the first half of the 1960s and the first half of the 1970s, the share of the United States (the world’s largest exporter) in the production of raw cotton declined from about 30 per cent to 18 per cent, whereas that of centrally planned countries6 increased from about 27 per cent to 37 per cent because of increases in both acreage and yields. The relative position of the rest of the world has not changed to any considerable extent7(Table 2). The relative distribution of mill consumption of cotton fibers among countries has also shifted (Table 4). The shares of the United States and Western Europe have declined over the last decade (from 18 per cent and 14 per cent in 1965 to 10 per cent and 9 per cent in 1975, respectively). Japan’s share was stable, at about 6 per cent, until 1974, but declined to 5 per cent in 1974 and to 4.4 per cent in 1975. The share of the centrally planned countries has increased steadily, going from 34 per cent in 1965 to 40 per cent in 1975. The increase for the rest of the countries, however, had been somewhat slower until the last two years; during 1974 and 1975, mill consumption grew steadily in spite of the recession in the industrialized countries.

Table 4.

Cotton and Man-Made Fibers: Changes in Distribution of Mill Consumption Among Countries, 1965-75

(In per cent)

article image
Source: International Cotton Advisory Committee, Cotton—World Statistics (October 1976).

Includes the Union of Soviet Socialist Republics, the People’s Republic of China, and Eastern European countries.

world trade

Cotton comprises about 9 per cent of the export earnings of the developing countries from agricultural products. Table 5 presents the export earnings of several cotton-exporting countries during the last five years. The value of exports has fluctuated because of changes in both price and volume. Although export values increased sharply in 1973 and 1974, the 1973 increase was due to increases in both price and volume, whereas the 1974 increase occurred despite a decline in volume. In 1975, the decline in volume continued and average prices were lower, resulting in a substantial decrease in export earnings.

Table 5.

Cotton: Value of Exports by Major Producers, 1971-751

(In millions of U.S. dollars)

article image
Sources: United Nations Food and Agriculture Organization and author’s estimates.

Calendar years.

Includes the Union of Soviet Socialist Republics, the People’s Republic of China, and Eastern European countries.

In volume terms, the share of the United States in world exports has averaged 25 per cent during the last four seasons, while that of centrally planned countries averaged less than 20 per cent (Table 3). World exports have been fluctuating between 3.6 and 3.9 million tons since the beginning of the 1960s. In 1972/73, world trade in raw cotton reached 4.6 million tons, a record volume, because of rising consumption in many importing countries, the desire to rebuild stocks, and higher availability of cotton. The largest growth in exports during 1972/73 was registered in the United States. World trade declined by 9 per cent in 1973/74, although the increase in U.S. exports continued, partly as a result of devaluation of the dollar, revaluation of currencies of certain major importing countries, and continuing high import demand by the People’s Republic of China. The lower volume of world trade in 1973/74, however, can be attributed to a number of factors, including the high level of accumulated stocks in importing countries, record prices for cotton, and balance of payments difficulties of major cotton importers owing to soaring oil prices. Furthermore, the severe shortage of ocean transportation has caused long delays in delivery. In many developing countries, such as Brazil, Egypt, Greece, Mexico, Pakistan, and Turkey, the effects on exports of reduced cotton output were compounded by increased demand from growing domestic textile industries. Consequently, tight export controls were imposed, which placed a restraint on world exportable supplies of cotton.

About the same factors that slowed down international trade in 1973/74 (except high cotton prices) continued to operate even more forcefully in 1974/75, reducing cotton exports by about 12 per cent below the previous year. Inflation in prices of food, raw materials, and energy aggravated the balance of payments problems of many importing countries, and the trough of the textile cycle seemed to be both wide and deep. The most significant reduction in trade occurred in the United States, where exports declined by half a million tons. Depressed demand in the Far East was largely responsible for the low export figure.

In 1975/76, despite the major reduction in world production, exports managed to record a slight increase over those in the previous year. To a large extent, this was made possible by the existence of high carryover stocks in many producing countries, such as Argentina, Brazil, the Sudan, and Turkey. Exports from the United States declined further, however, as U. S. prices were above world parities for several months; although the price situation changed later, actual 1975/76 export figures remained low, since substantial commitments were not shipped before the end of the season. As demand picked up with the upturn of the textile cycle, exports from other countries paralleled availabilities. The sharpest increase in import demand came from the Far East, with the exception of Japan—where the textile recovery has been slower. In Western Europe, a moderate increase in imports was recorded, mostly as a result of some stock rebuilding, as economic recovery in these countries has been slow. In addition to sluggish demand, competition from imported textiles has been strong in Western Europe, and profitability of mills has been adversely affected by higher cotton prices, which the textile industry has been unable to pass on to the consumer.

man-made fibers

The growth in man-made fiber production has been remarkably fast and lasting. It increased tenfold between the early 1930s and the mid-1950s, which corresponds roughly to an annual growth rate of 10 per cent. The rate of growth slowed down somewhat during the late 1950s, to about 6 per cent, but picked up later to reach the average level of 10 per cent during the 1960s and early 1970s. Production declined for two consecutive years, by 2 per cent in 1974 and 6 per cent in 1975, then recovered in 1976 with an increase of 13 per cent.

Although all man-made fibers have been effective in replacing natural fibers, various components of the man-made fiber industry have behaved in quite different ways. Noncellulosic fibers that came on the commercial market in the 1940s have grown faster than the cellulosics, whose growth came to a standstill in the 1960s. The basic factors underlying the poor performance of cellulosics are the weakening demand owing to the superiority of noncellulosics for many products, changes in fashion favoring the latter, and the high cost of pollution control and energy in rayon and acetate operations.

After surpassing the production of rayon and acetate for the first time in 1968, the production of noncellulosics doubled within a period of five years to reach 7.6 million tons in 1973. Yet, there have also been significant changes in the fiber “mix” among the noncellulosics; for example, in 1966, nylon accounted for one half of the total production, whereas, although nylon has continued to expand rapidly, polyester now surpasses nylon by a fair margin and capacity data show that these trends will continue (Table 6). Production of polyester has, in fact, been faster than that of all the other fibers, increasing more than five times during a span of ten years. Also, polyester production continued to grow during 1974 and 1975, albeit at a minimal rate, while that of man-made fibers as a whole was declining.

Table 6.

Man-Made Fibers: World Production, 1973-76, and Producing Capacity, 1976-77

(in millions of metric tons)

article image
Sources: Textile Economics Bureau, Inc., Textile Organon (June 1976 and February 1977).

The producing capacity in operable condition is shown on an annual basis as of March 1976 for the calendar year 1976 and as of December 1977 for calendar year 1977.

Includes the Union of Soviet Socialist Republics, the People’s Republic of China, and Eastern European countries.

Includes other noncellulosic fibers.

Among the producing countries, there has been a considerable variation in performance. Since the mid-1960s, some growth in rayon and acetate production has occurred in the centrally planned countries, whereas world production remained static until 1975, when it declined sharply (Tables 1 and 6). The share of these countries in the world total has increased from less than one fourth in 1966 to about two fifths in 1975 and 1976. The developing countries’ share also increased during this period, from about 8 per cent to 12 per cent. Consequently, the established producers of rayon and acetate (that is, Japan, the United States, and Western Europe) now produce a little less than half of the world output.

As for the noncellulosic (synthetic) fibers, the share of centrally planned countries has risen from less than 8 per cent in 1966 to close to 11 per cent in 1976; that of developing countries has risen from 2 per cent to about 16 per cent during the same period. Industrialized countries control about three fourths of the synthetic fiber production. Taking all man-made fibers together, the share of industrialized countries in 1976 production was about 67 per cent. In terms of capacity, however, their share was 70 per cent. The rate of capacity utilization in these countries was lower than in the rest of the world.

Prices of both polyester and rayon have shown a long-term tendency to decline since the latter part of the 1950s, with the rate of decline being much faster for polyester. Rayon prices have turned upward beginning in 1971, however, and in 1974 they averaged twice their 1970 level. The increase during 1974 was continuous through the year, and starting from 35.5 cents a pound in January, prices reached 57 cents a pound in December for regular rayon staple in the United States. (See the Appendix, Table 11.) Then they eased somewhat early in 1975 (to 50 cents a pound), but they rose again in the last quarter to 54 cents and remained stable until the last quarter of 1976, when they went up by 4—5 cents a pound. Polyester staple prices, on the other hand, started increasing during 1973, reached a peak in the second half of 1974 (at 51 cents a pound), and declined through the first half of 1975 (to 45 cents a pound by June). Since then there have been four increases, one in September 1975 (to 50 cents a pound), one in December 1975 (to 53 cents), another one in February 1977 (to 55 cents), and the last one in April 1977 (to 60 cents.8 (See Table 12, in the Appendix.)

II. Determinants of Cotton Consumption

As was pointed out earlier, the crucial and immediate question for cotton is the effect on consumption of recent changes in cotton prices vis-à-vis those of man-made fibers. To assess the effects of these prices on cotton consumption, a linear regression equation has been fitted to semiannual data over the period 1958-75. Because cotton consumption had an increasing trend and polyester prices had a decreasing trend over the long run, determining the theoretically correct signs for the price variables proved to be a challenge, but the following equation satisfied expectations in that regard:9

  • WC = f(COT(−2), POL(−2), WC(−2), T)

where

  • WC = semiannual world consumption

  • COT(−2) = deflated10 cotton price, lagged by one year

  • POL(−2) = deflated10 polyester price, lagged by one year

  • WC(−2) = WC, lagged by one year

  • T = time

Actual and predicted quantities of semiannual world consumption of cotton using the regression estimation are presented in Chart 2. Although it provides a good fit, the regression equation seems, of course, to smooth out the fluctuations. Moreover, even with the predicted values, it is possible to observe a seasonality in cotton consumption during most of the period under observation, which manifests itself in generally higher figures for the first half of a calendar year than in the second half. Use of seasonally adjusted data, or inclusion of seasonal dummies, did not improve the fit of the equation. It seems that the existence of the lagged dependent variable on the right-hand side of the fitted equation has produced a satisfactory representation of the seasonality of cotton consumption.

Chart 2.
Chart 2.

Cotton: World Consumption—Actual and Predicted Totals, First Quarter 1958-Second Quarter 1975

Citation: IMF Staff Papers 1977, 002; 10.5089/9781451947557.024.A005

The estimated equation indicates that world consumption of cotton is negatively related to cotton prices, positively related to polyester prices, has a positive time trend, and also has a positive association to consumption lagged by one year (Table 7, footnote 1). The latter two variables presumably represent the usual set of factors, such as increases in population and income, changes in consumer preferences and fashions, and technological advances.11 The cotton-price series used is the Liverpool Index, which may be considered a good indicator of world cotton prices. Still, domestic prices in any country at a point in time may diverge from international prices for a variety of reasons, exerting a somewhat different influence on consumption in that country than would be exerted if there was no divergence in prices.

Table 7.

Cotton: Extrapolated Quantities1 of World Consumption, Using Alternative Price Assumptions for Cotton and for Polyester, 1976-78

article image

Estimated regression equation used in the extrapolations was fitted on semiannual data for the period 1958-75:

WC=5,747−7,633.2(3.7)*⁢  COT(−2)⁡+3,88.4(3.0)*⁢  POL(−2)+0.481(2.8)*⁢   WC(−2)+377.2(4.6)*⁢  T

R2 = 0.95 D-W = 1.7

WC = semiannual world consumption in thousands of bales

COT(- 2) = deflated cotton price, lagged by one year, in cents a pound

POL(- 2 = deflated polyester price, lagged by one year, in cents a pound

T = time

t-ratios in parentheses

* indicates significance at the 1 per cent level

Price combinations used are as follows (in cents a pound):

Calendar year.

Growth rates for the lowest and highest projections.

Second half of 1975 through the first half of 1976.

Second half of 1976 through the first half of 1977.

Second half of 1977 through the first half of 1978.

d87944862e6961

Usage of polyester staple prices in the United States as a proxy for world man-made fiber prices has its shortcomings. One obvious reason is that polyester is not the only man-made fiber, although it became the single largest such fiber in world trade in 1975, accounting for one third of the total. Another reason is that polyester prices in the United States may not be representative of world prices. In fact, during most of 1976, polyester staple prices in Western Europe were about 20 cents a pound above the U. S. prices. Raw material costs and plant size diffferentials are the reported causes. However, owing to the unavailability of satisfactory data, the assumption had to be made that U. S. prices move in reasonably close alliance with prices elsewhere.

In addition to these considerations, one would be inclined to expect a positive relationship between the business cycle and the mill consumption of cotton, since the textile industry constitutes a considerable portion of total manufacturing, and particularly because of the coincidence of the recent recessions in general economic activity and in textile trade. The existence of such a relationship between the level of economic activity in the industrialized countries and world cotton consumption, however, could not be ascertained in the present study, largely because cotton consumption in the rest of the world had a long-term increasing trend. Using the per capita consumption of cotton as the dependent variable and including the production of cotton in the right-hand side of the equations were also among the alternative formulations that did not produce meaningful results.

However, as can be seen in Chart 3, world consumption and consumption by the industrialized countries have shown different trends in the past. Hence, a somewhat different regression equation seemed to have the best fit to the consumption of industrialized countries over the period 1958-75.

Chart 3.
Chart 3.

Cotton: World Consumption and Consumption in the Industrialized Countries, 1957-751

Citation: IMF Staff Papers 1977, 002; 10.5089/9781451947557.024.A005

1 Industrialized countries, for our purposes, include the following countries for which consistent semiannual mill consumption data were available: Austria, Belgium, Canada, Denmark, France, Finland, the Federal Republic of Germany, Italy, Japan, the Netherlands, Norway, the United Kingdom, and the United States.

IWC = f[COT (-1)/POL(-1), BCC, T]

where

IWC = semiannual mill consumption of cotton by 13 industrialized countries12

COT(- 1)/POL(- 1) = ratio of cotton price lagged by one-half year to polyester price lagged by one-half year

BCC = weighted index of the business cycle in industrialized countries13

The estimated regression shows that mill consumption in these countries has a positive association with the business cycle, a negative relation to the ratio of the price of cotton over that of polyester, and a negative time trend (Table 8). The price variables in industrialized countries seem to have a shorter time lag in their effect on consumption than is true for the world as a whole.

Table 8.

Cotton: Extrapolated1 Quantities of Industrialized2 World Consumption, Using Alternative Price Assumptions for Cotton and for Polyester, First Half 1976-First Half 1978

article image

Estimated regression equation used in the extrapolations was fitted on semiannual data for the period 1958-75:

IWC=1,235−614.9(2.6)*⁢  COT(−1)⁡/⁢  POL(−1)+85.5(6.9)**⁢   BC−30.6(3.0)**⁢  T

R2 = 0.85 D-W = 1.99

IWC = Semiannual consumption by industrialized countries (as specified in footnote 2, below) in thousands of bales

COT(- 1)/POL(- 1) = ratio of cotton price, lagged by one-half year, over polyester price, lagged by one-half year

BC = index of business cycle indicators in industrialized countries as defined in footnote 13, in the text

T = time

t-ratios in parentheses

* indicates significance at the 5 per cent level

** indicates significance at the 1 per cent level

The countries for which semiannual data were available were Austria, Belgium, Canada, Denmark, Finland, France, the Federal Republic of Germany, Italy, Japan, the Netherlands, the United Kingdom, and the United States.

Price combinations are as in Table 7.

Calendar year.

Second half 1975-first half 1976.

Second half 1976-first half 1977.

Second half 1977-first half 1978.

Growth rates for the lowest and highest projections.

III. Consumption Prospects and Future Outlook for Cotton

The regression equations reported in the previous section can be used in projecting world cotton consumption (until the end of 1978) and consumption by industrialized countries as a group (until mid-1978). As shown in Table 7, two assumptions have been made in these projections for cotton and polyester prices during 1977: high (H) and low (L). Projections based on alternative prices produce similar results, however, because of the existence of lagged variables in the equations and the availability of actual price data through April 1977.

In general, projections indicate an increasing trend of world cotton consumption (Table 7) and a leveling off of consumption by the industrialized countries (Table 8).14 The projected quantity of consumption for the period extending from the second half of 1975 to the first half of 1976 (14.0 million metric tons) implies a growth rate of 9.6 per cent since the period extending from the second half of 1974 through the first half of 1975, and it is only 2 per cent higher than the present estimate by the International Cotton Advisory Committee (ICAC) for the 1975/76 season.15 Further growth, although at declining rates, is projected for the period extending from the second half of 1976 through the first half of 1977 and for the period extending from the second half of 1977 through the first half of 1978. On a calendar year basis, the rate of growth is projected to be higher in 1978 than in 1977. For the industrialized countries as a group, zero growth is projected for 1977, following a projected growth rate of almost 4 per cent for 1976. When the 12-month period covering the second half of 1975 and the first half of 1976 is compared with the two succeeding 12-month periods, however, there is a decline in mill consumption by industrialized countries in the first year, and then a recovery in the second year. Since the regression equation for world consumption incorporates a one-year lag and the equation for industrialized countries incorporates a half-year lag for the effect of price changes to be felt on consumption, these projections may be qualified to the extent that price considerations might influence the decisions by the textile mills with a shorter time lag in the future. The upward movement projected for world cotton consumption still holds even after the impact of high cotton prices and low polyester prices in the second half of 1976 is taken into account (that is, on consumption for the second half of 1977).

The projections are subject to another more important qualification. Extrapolations based on a single regression equation have no built-in mechanism for adjusting to changes in the supply of cotton. Yet, as is especially true for many agricultural products, cotton prices and cotton use are subject to the influence of expectations of future, as well as present, availabilities. To the extent that such expectations are not reflected in the explanatory variables of the regression equation, they do not have a bearing on the projections. Furthermore, supply and use of cotton in individual countries are determined not only by the size of the cotton crop (for producing countries) but also by government policies that affect the amount of the fiber to be imported or exported. At any time, there may be physical shortages (or unnecessary stocks) of cotton in some countries that are due to the existence of less than perfect competition in the world cotton market. If such countries are relatively large suppliers or consumers of cotton, then their policy decisions will have a substantial impact on the world cotton balance. Yet, these factors would not be reflected in extrapolations from a regression fit. Keeping these considerations in mind, one can review the consumption projections and the most recent developments in the cotton market, and comment on the future outlook for cotton.

The limited availability of cotton during the 1975/76 season (when consumption exceeded production by 1.8 million metric tons, thereby causing a major decline in end-season stocks) contributed considerably to the sharp increases in its price during the first half of 1976. In spite of the continued high level of demand for cotton during the 1976/77 season16 and increasing scarcity of cotton (the total supply during 1976/ 77 was 1.3 million metric tons less than the previous season and the lowest level of supply since 1971/72), cotton prices became rather stable after July 1976. This was attributable largely to the severe competition from man-made fibers, the prices of which either remained constant or increased only minimally during 1976, causing a threat of market loss for cotton. Man-made fiber demand improved late in 1976, and prices were raised recently (Tables 11 and 12, in the Appendix). It is not yet possible to determine whether cotton lost any market share during the second half of 1976. The recent increase in man-made fiber prices helps to improve cotton’s relative price position during the first half of 1977. In addition, there are indications that cotton textile imports by several European countries from the developing country exporters are likely to slow down during 1977. It is possible, therefore, that cotton consumption in the industrialized countries during the 1976/77 season may match the level in 1975/ 76. A small increase would be in line with the long-run trend in the centrally planned economies. In the rest of the world, on the other hand, 1976/77 consumption in the export-oriented cotton industries of Asian countries, such as the Republic of China, Hong Kong, Singapore, and Korea, probably suffered somewhat from sluggish business activity in some of their export markets and also from the availability of cheap man-made fiber substitutes. In addition, cotton consumption in two large producer countries—India and Pakistan—was adversely affected by supply shortages, largely as a result of bad crops. The extent of the effect of these factors is not yet known, and present estimates of world cotton consumption for the 1976/77 season are only preliminary.

Table 9 presents alternative sets of figures representing two different possible market balances for cotton. Alternative 1 is based on consumption and production estimates by ICAC for the 1975/76 and 1976/77 seasons and somewhat conservative figures for 1977/78, when consumption is expected to match that of 1975/76 and production is expected to reach the level of 1972/73. According to this alternative, which predicts a decline of 3 per cent in consumption during 1976/77, end-season stocks on August 1, 1977 will be only 4 million metric tons (equivalent to 3.6 months’ consumption at the 1976/77 level). To maintain this already low stock level to the next season, an increase of 10 per cent in the 1977/78 crop would be necessary, and cotton would continue to be in tight supply during both 1977/78 and 1978/79.

Table 9.

Cotton: Supply and Demand Balances, 1975/76-1978/79 1

(In millions of metric tons)

article image

Crop years.

Using cotton consumption estimates of the International Cotton Advisory Committee for 1975/76 and 1976/77, and assuming a 2 per cent growth rate in 1977/78. Production for 1977/78 is assumed to increase by about 9 per cent from the 1976/77 level, whereas consumption is assumed to be at about the 1975/76 level.

Equivalent to 3.6 months’ consumption at the 1977/78 level.

Using consumption projections obtained by extrapolation from the regression equation. These figures for consumption actually refer to two half-years as in Table 7, not to August-July years. Production for 1977/78 is assumed to set a new record.

Equivalent to 1.8 months’ consumption at the 1977/78 levels.

Lowest and highest values of projections using different price assumptions. (See also Table 7.)

In Alternative 2, figures for cotton consumption are those obtained by extrapolation from the regression equation. Estimated cotton production for 1976/77 is marginally above that of Alternative 1, and the 1977/78 cotton crop is assumed to be at a record high (3 per cent above the previous record in 1974/75). With these figures, carry-over stocks at the beginning of the 1977/78 season are equivalent to only 2.2 months’ consumption (at the 1976/77 level), and they are even lower at the end of the season. The scarcity of cotton under this alternative is much more intense.

The reality will probably lie between these two alternatives, A tight supply situation is almost certain to continue. Even if a record cotton crop were obtained in the 1977/78 season and if consumption matched that in 1975/76, carry-over stocks would be equivalent to less than 4.5 months’ consumption, which is considered to be inadequate. On the other hand, if production does not increase substantially during the 1977/78 season, the possibility of a deterioration in cotton’s relative position vis-à-vis other fibers is likely to occur, since there is a limit to running down the stocks and maybe even to price increases (especially when substitutes are available). The recent recovery in cotton’s share in total fiber consumption may become a short-lived phenomenon if sufficient availability of cotton cannot be secured, that is, if cotton production does not rise to the level demanded by the growing world economy.

APPENDIX

Table 10.

Cotton: Liverpool Price Index, 1960-76

(In U.S. cents a pound)

article image
Source: Liverpool Cotton Services Limited, Cotton Outlook (various issues).
Table 11.

United States: Prices of Rayon 1.5 and 3.0 Denier, Regular Rayon Staple, 1955-76

(In U.S. cents a pound)

article image
Source: National Cotton Council.
Table 12.

United States: Prices of Polyester, Dacron, Type 54, 1.5 Denier, 1955-76

(In U.S. cents a pound)

article image
Source: National Cotton Council.
*

Mrs. Ecevit, economist in the Commodities Division of the Research Department, is a graduate of the Middle East Technical University in Ankara. She received her doctorate in economics from the University of Pittsburgh.

In addition to colleagues in the Fund, the author is indebted to Maurice E. Thigpen and Enzo R. Grilli for useful suggestions.

1

Textile fibers are defined here to include cotton, wool, silk, flax, and cellulosic (rayon and acetate) and noncellulosic (nylon, polyester, acrylic, and modacrylic) man-made fibers.

2

For strict Middling 1 1/16” quality (c.i.f. Liverpool), as measured by the index published by Liverpool Cotton Services Limited in Cotton Outlook.

3

See also Tables 10, 11, and 12, in the Appendix, for the long-term price series for cotton, rayon, and polyester. For man-made fibers, the U.S. prices are generally lower than the prices in Western Europe and Japan, but the direction and timing of price changes are similar.

4

Prices shown by the Cotton Outlook index (mentioned in footnote 2) for strict Middling 1 1/16” quality (c.i.f. Liverpool) increased from 31 cents a pound to 38 cents during the course of the year.

5

Significant reductions in plantings were observed in the United States, Latin America, and the Middle East. High prices of competing crops and government policies to stimulate food production were the main causes of these shifts to alternative products. Bad weather and insect damage, coupled with rising costs of fertilizers and pesticides, have contributed to the decline in yields in many countries, notably Brazil, India, Pakistan, and Turkey. Plantings in centrally planned countries have remained the same, but the Soviet output was reduced by 5-6 per cent, mainly because of a shortage of rainfall.

6

Includes the Union of Soviet Socialist Republics (the world’s largest producer and the second largest exporter), the People’s Republic of China (the world’s third largest producer but a net importer), and Eastern Europe.

7

Brazil, Egypt, India, Mexico, Pakistan, and Turkey comprise another 28-29 per cent of the total, and the remainder is distributed among a large number of smaller producers.

8

Because discounting from announced prices is widely practiced, these prices are estimates of producers’ actual selling prices.

9

For the estimated equation, see Table 7.

10

Deflator used was obtained from the consumer price indices of ten industrialized countries (Belgium, Canada, France, the Federal Republic of Germany, Italy, Japan, the Netherlands, Sweden, the United Kingdom, and the United States) weighted by their cotton consumption shares during the period 1970/71-1974/75.

11

The inclusion of a dummy variable to assess the effects, if any, of a change in consumer tastes during 1974-75 produced an insignificant f-ratio. We are inclined to think that the increased consumer preference for cotton was perhaps helpful during the second half of 1974 and the first half of 1975 in preventing sharper declines from the long-term trend values. To put it differently, on a calendar-year basis, cotton consumption stabilized during 1974 and 1975 because of a favorable change in tastes at a time when other fibers were declining.

12

The countries for which semiannual mill consumption data were available are Austria, Belgium, Canada, Denmark, France, Finland, the Federal Republic of Germany, Italy, Japan, the Netherlands, Norway, the United Kingdom, and the United States.

13

The business cycle indicators measure the ratio between actual and potential activities in the manufacturing sector in the ten industrialized countries as mentioned in footnote 10. These indicators were weighted by their average cotton consumption shares during the period 1970/71-1974/75 to get the weighted business cycle index.

14

Note that this refers to consumption of cotton by the local mills only.

15

The regression equation was fitted for semiannual data. The projections, therefore, are made on a semiannual basis that does not exactly correspond to the usual cotton season of August 1 through July 31.

16

Even according to the ICAC estimates, the figure for cotton consumption in 1976/77 is only marginally below the peak in 1975/76.

Same Series

Other IMF Content

Other Publishers

  • Collapse
  • Expand
Cover IMF Staff Papers
IMF Staff papers: Volume 24 No. 2
Author:
International Monetary Fund. Research Dept.
Volume/Issue:
Volume 1977: Issue 002
Publisher:
International Monetary Fund
ISBN:
9781451947557
ISSN:
1020-7635
Pages:
260
DOI:
https://doi.org/10.5089/9781451947557.024
  • Chart 1.
    View in gallery
    Chart 1.

    Monthly Cotton Prices (Liverpool Index), Man-Made Fiber Prices,1 and Semiannual World Consumption of Cotton, 1970-76

  • Chart 2.
    View in gallery
    Chart 2.

    Cotton: World Consumption—Actual and Predicted Totals, First Quarter 1958-Second Quarter 1975

  • View in gallery
  • Chart 3.
    View in gallery
    Chart 3.

    Cotton: World Consumption and Consumption in the Industrialized Countries, 1957-751

Chart 1.

Monthly Cotton Prices (Liverpool Index), Man-Made Fiber Prices,1 and Semiannual World Consumption of Cotton, 1970-76
Chart 2.

Cotton: World Consumption—Actual and Predicted Totals, First Quarter 1958-Second Quarter 1975
Chart 3.

Cotton: World Consumption and Consumption in the Industrialized Countries, 1957-751