Inter-Fiber Competition: Econometric Modeling of U.S. Cotton and Polyester Fiber Demand
No Thumbnail Available
Files
Date
2004-04-29
Authors
Journal Title
Series/Report No.
Journal ISSN
Volume Title
Publisher
Abstract
The estimation of the U.S. cotton and polyester demand is essential to effectively track both the performance of the U.S. domestic textile industry as well as the U.S. economic performance as reflected in consumer demand for these two fibers. In order to track the above mentioned parameters the two most effective benchmarks would be the U.S. fiber mill demand and U.S. total or retail fiber demand. The level of fiber demanded by U.S. mills for conversion into yarn and subsequently fabric and apparel, is reflective of the competitiveness of the U.S. spinning industry. This is true because the same fiber demand could be shifted to a low cost manufacturing base like Asia, South America or Mexico given their economies of scale and/or preferential trading agreements with the U.S. At the retail level, the total fiber demand for cotton and polyester are reflective of the U.S. consumer demand for cotton and polyester products, which in turn reflects the general U.S. economic performance. The inter-fiber competition dimension captures the demand substitution effects between cotton and polyester specifically.
A review of previously published literature revealed some interesting observations about U.S. inter-fiber competition and their demand estimation. First, most of the models developed in the recent past focused on estimating U.S. inter-fiber demand between cotton and man-made fibers or between cotton and synthetic fibers. Earlier research work had clearly established that within man-made fibers, cotton shared a substitutive relationship only with synthetic fibers and not the cellulosic fibers even though data for individual fiber demand within synthetic fibers was not available. This effectively renders the clustering of all man-made fibers under one grouping ineffective in tracking the derived individual fiber demand. Further, with regards to competition to U.S. cotton, polyester is known to be its chief competitor. However, there was not a single model found in previous literature that captured inter-fiber competition between U.S. cotton and polyester. This could also be the reason for the lack of estimation models to explain inter-fiber substitution effects between cotton and polyester. Second, the previous models ignored the effect of cotton promotion. Though there are references to cotton promotion efforts as a driver of cotton consumption, no previous estimation model included that as an explanatory variable. Third, even though the last fiber demand model estimated cotton fiber demand from 1965 to 1997, it did not account for cotton promotion, whose effect on cotton consumption was evident since 1980. The choice of time span for this study was from 1990 to 2001, in order to account for more recent economic changes in the U.S. fiber demand situation. Finally, not a single model found in the previous literature estimated U.S. polyester demand at the mill and end use level. Even the clustered man-made fiber models were built on the hypothesis of being influenced by man-made fiber promotion apart from the obvious parameters of U.S. income and fiber prices. However, man-made fiber promotion, which was a major consumption driver for polyester during its introductory phase in the 1960's, has largely disappeared due to very low margins. Instead cotton promotion as described above has become the key factor since 1980. Since then, polyester has largely relied on cheaper prices and its technological development to produce new value added properties due to which it remains the most popular among synthetic fibers as well as cotton?s chief competitor. Ordinary least squares regression, which has been the most widely used estimation method for U.S. fiber demand as seen in past research, was used to develop the demand estimation models for U.S. cotton and polyester demand in this study. The explanatory variables for the regression model were based on past research as well as newer variables introduced to reflect the changes in the U.S. cotton and polyester demand scenario during the study period of 1990 to 2001 as described above. The demand models developed based on a new hypothesis accounted for nearly 95% of the variation in U.S. annual cotton demand at both mill use and end use levels. This was a significant improvement over past models. The U.S. cotton promotion was found to be the strongest influencing factor on U.S. total cotton demand. On the other hand, U.S. cotton textile deficit was the most significant factor influencing U.S. cotton mill demand. In the case of polyester, its technological development was found to be the major factor influencing end use demand whereas at the mill level, the U.S. total fiber demand was its major driver.
Description
Keywords
inter-fiber competition, cotton promotion, fiber econometric model
Citation
Degree
MS
Discipline
Textile Engineering
