The authors are Post Doctoral Research Associate, Professor, and Professor, respectively, Department of Agricultural and Applied Economics, Texas Tech University. The authors acknowledge Plains Cotton Cooperative Association and DTN Cotnet for cooperation in obtaining the data.
This research is supported by Cotton Incorporated and the Texas State Support Committee.
The analysis of the West Texas and East Texas/Oklahoma spot market using the Daily Price Estimation System (DPES) indicated an overall improvement in quality in the 2005/06 marketing year relative to the 2004/05 marketing year. The results also showed an increase in price in 2005/06 compared to 2004/05, though it remained below its 2003/04 level. The combined total bales and total sales in the two regions were higher in 2005/06, with increased sales in both the West Texas and East Texas/Oklahoma regions. For the 2005/06 marketing year, the results indicated higher premiums for low leaf grade, higher staple length, and lower first digit color grade. Price discounts in 2005/06 for lower staple length and higher than base quality for first digit color grade remained unchanged. Discounts for leaf and micronaire increased compared to the 2004/05 levels, while discounts for strength, bark, and first digit of the color grade decreased.
Discounts for level 1 bark in the 2005/06 marketing year generally fluctuated between 100 and 300 points/lb. (Figure 14). Compared to the previous marketing year, discount for level 1 bark in 2005/06 was, on average, 150 points lower (Figure 15).
Discount for micronaire 3.35 fluctuated between 150 and 250 points/lb between early November
and the middle of-December then followed a slight downward trend between the middle of-December and the middle of-March (Figure 12). Compared to the 2004/05 marketing year, micronaire was more severely discounted both above and below the base level (Figure 13).
Uniformity 80 discounts in the 2005/06 marketing year exhibited two phases. From early November to the end of January, uniformity 80 was rarely discounted, but from early February through the end of March, discounts reached as high as 75 points per pound (Figure 16). Compared to the previous marketing year, uniformity level below the base value was more severely discounted in 2004/05. Uniformity level above the base value received higher premiums as well in 2004/05. Overall, uniformity had more impact on cotton price in West Texas in the 2004/05 marketing year compared to the 2005/06 marketing year (Figure 17). This observation stands when average premiums are calculated for the period between mid January and late March.
Other Extraneous Matter and Preparation
As indicated in Table 1, the percentage of level 1 and level 2 other extraneous matter and preparation remained below the 1 percent level in the 2005/06 marketing year. The low incidence of other extraneous matter and preparation makes it difficult to interpret and draw conclusions on the patterns of these attributes.
The analysis of the 2005/06 marketing year shows that cotton price partly recovered in the 2005/06 marketing year. The average spot price for the 2005/06 crop was 47.60 cents a pound, about 30% higher than its 2004/05 level. While total number of sales in 2005/06 for both regions were similar to 2004/05 marketing year, total bales for both regions increased compared to the previous year. The average price was higher in 2005/06 than in 2004/05 despite the high sales volume. This may be explained by an improvement in overall quality of the crop compared to the previous marketing year as well as a stronger general cotton market.
For the 2005/06 marketing year, the results indicated higher premiums for low leaf grade, higher staple length, and first digit color grade. However, no premiums were observed for higher level of uniformity. Premium levels for better than base quality strength appear to be very minimal to nonexistent. Price discounts were higher for leaf and micronaire and lower for second digit of the color grade, strength, and bark in 2005/06 compared to the 2004/05 marketing year.
Brown, J.E. and D.E. Ethridge. "Functional Form Model Specification: An Application to Hedonic Pricing." Agricultural and Resource Economics Review. 24(2), 1995: 166-173.
Brown, J.E., D.E. Ethridge, D. Hudson, and C. Engles. "An Automated Econometric Approach for Estimating and Reporting Daily Prices." Journal of Agricultural and Applied Economics.
27(2), 1995: 409-422.
U.S. Department of Agriculture (USDA). The Classification of Cotton. Washington, DC: USDA, Agricultural Marketing Service, Agricultural Handbook 566, September 1995.
Appendix A: The DPES Model and Yearly Parameter Estimates
The Daily Price Estimation System is a computerized econometric model based on the theory of hedonic price analysis (Brown and Ethridge, 1995). The premise of this approach is that the value of a commodity is determined by the value of the utility bearing characteristics that comprise the commodity. The implicit prices of these characteristics may be determined by disaggregating the price of the commodity into its measurable characteristic components. In the DPES, the relationship between the price of cotton and its various measurable quality attributes is estimated using a nonlinear regression model. The equation used for regression analysis is:
The variable definitions and parameter estimates are presented in Appendix Table A1. At the end of each marketing year, the data for that year are compiled and diagnostic tests are run on the model. The purpose of running diagnostics tests is to detect any systematic error that might have occurred in the DPES, but which remained undetected in the daily diagnostics. The model specification above is the result of the year-end diagnostic analysis for the 2005/06 marketing year. The procedures of Brown et al. (1995) indicated that this model specification best fits the 2005/06 marketing year data. The parameters of the model for the 2005/06 year model were computed by weighting the individual estimates for each day by the number of sales transactions during that day.
Appendix Table A1: Definition of Variables and Parameter Estimates for the 2005/06 Marketing
Year Model. Dependent Variable = Log(Price)
TEXAS-OKLAHOMA PRODUCER COTTON MARKET: 2005/06
This report summarizes the price, premium, and discount estimates for the 2005/06 marketing year (August 1, 2005 to July 31, 2006) based on spot market transactions throughout the West Texas and East Texas/Oklahoma regions. The estimates are derived on a daily basis using the Daily Price Estimation System (DPES), which is a computerized price analysis system that combines statistical estimation and spreadsheet computation to estimate the relationship between spot price in these two regions with nine cotton quality attributes as defined by the USDA1. These attributes include fiber length, fiber strength, fiber length uniformity, micronaire, leaf content, color grade, bark content, other extraneous matter content, and preparation (USDA, 1995). A detailed description of the model and the weighted averages of the parameter estimates are presented in Appendix A of this report. The purpose of the analysis is to determine the base price and calculate the premiums and/or discounts associated with each attribute. The report is organized as follows: the first section presents the 2005/06 crop statistics; the second section summarizes and discusses the average price, premiums, and discounts for the 2005/06 marketing year; and the last section discusses the movements of premiums and discounts associated with each characteristic.
2005/06 Crop Statistics
The average cotton spot price for the 2005/06 marketing year was 47.60 cents a pound, about 31 percent higher than its 2004/05 level (Table 1). The total number of bales per sale increased from 108.47 to 127.3 bales per sale, representing the fifth crop year in a row with an increase in average bales per sale. The total volume of transactions for the 2005/06 marketing year amounted to 10,516 with a total of 1,338,764 bales sold. The West Texas region accounted for 80 percent of the transactions and 85 percent of the volume of sales, roughly comparable to their 2004/05 levels. As Figure 1 indicates, most of the transactions in the 2005/06 marketing year occurred between the beginning of December and the end of January, about a month shorter than 2004.05 crop year, which totaled most of its transactions from the middle of January through end of February. The base prices in the two regions exhibited similar behavior during the 2005/06 marketing year, averaging about 41.94 cents a pound for West Texas and 41.80 cents a pound for East Texas and Oklahoma, about 3 cents a pound increase compared to the 2004/05 marketing year. As Figure 2 indicates, the movement of the base price in West Texas shows two distinct patterns in 2004/05. At the beginning of the marketing year, from November 7 to January 2, the base price fluctuated around 45 cents a pound with occasional dips to 40 cents a pound; then followed an upward trend to reach 50 cents a pound in the middle of-March.
With regard to quality attributes during the 2005/06 marketing year, the average leaf grade decreased to 2.80 compared to 3.57 for the 2004/05 marketing year. This level is comparable to the 2003/04 marketing year (Table 1). The first and second digit of the color grade also improved substantially compared to the 2004/05 marketing year. While the first digit color grade was at its lowest level in the last four seasons, the second digit of the color grade was similar to its 2003/04 level at 1.08. There was a slight increase in staple length (34.50 vs. 33.97 32nds/inch) and strength (28.82 vs. 28.31 gram/tex). Micronaire and uniformity increased slightly (3.83 vs. 3.71) and (80.18 vs. 80.10), respectively, between the two marketing years. Average level 1 bark decreased considerably from 47.01% in 2004/05 marketing year down to 8.44% in 2005/06 marketing year. Average level 1 other extraneous matter and preparation also decreased in the 2005/06 marketing year.
Average 2005/06 Prices, Premiums, and Discounts
The DPES uses a hedonic model to evaluate the relationship between spot price and a set of nine quality attributes (i.e., staple length, first digit color grade, second digit color grade, micronaire range, leaf grade, uniformity, strength, bark, and preparation) and a regional variable to separate West Texas and East Texas/Oklahoma regions. The goal of this estimation is to calculate the daily premiums and discounts associated with each characteristic. The results of this estimation are then used to calculate the weighted averages of the parameters for the entire marketing year and to calculate the yearly base prices, premiums, and discounts in the two regions. Tables 2 and 3 present the yearly averages of the base price and premiums and discounts associated with the nine quality characteristics for the West Texas and East Texas/Oklahoma regions. The average base price was 41.94 cents per pound for the West Texas region and 41.8 for the East Texas/ Oklahoma region.
Table 2. 2005/06 Weighted Average Price Estimates from DPES, West Texas
Table 3. 2005/06 Weighted Average Price Estimates from DPES, East Texas/Oklahoma
Patterns of Premiums and Discounts
This section summarizes the average premiums and discounts for each quality attribute during the 2005/06 marketing year. It also provides a comparison of the 2005/06 and 2004/05 marketing years to identify whether there were changes in premiums and discounts for each quality attribute. The premiums and discounts for quality characteristics associated with the West Texas spot market are presented.
For illustration purposes, the premiums for leaf grade 3 for West Texas are presented in Figure 3. The daily premiums for leaf grade 3 were relatively stable in the 2005/06 marketing year. lthough there were some occasional spikes up to 200 points per pound and dips to no premium at all, for the most part premiums for leaf grade 3 fluctuated between 50 and 100 points per pound. Leaf grade premiums and discounts for the West Texas market in the 2005/06 marketing year were compared with the 2003/04 marketing year (Figure 4). The results show that premiums for low leaf grade and discount for high leaf grade were slightly higher in the 2005/06 marketing year.
The discounts for color grade 42 fluctuated considerably between early November and end of February between 50 and 400 points/lb. (Figure 5). First digit color grades 1, 2, and 3 received premiums amounting to 142, 101, and 61 points/lb, respectively. In comparison to the 2004/05 marketing year, first digit color grades 1, 2, and 3 received higher premiums while first digit color grade 4, 5, and 6 were discounted similarly in 2005/06 (Figure 6). Discounts for the second digit color grade were lower in the 2005/06 marketing year compared to the 2004/05 marketing year. Thus, increased levels of yellowness were less severely discounted in the 2005/06 marketing year than in the 2003/04 marketing year (Figure 7).
The discounts for staple length 33 for the 2004/05 marketing year fluctuated, for the most part, between 100 and 350 points/lb. There were some occasional spikes as noted in Figure 8 and a noticeable upward trend throughout the marketing year. The discounts of staple length 33 remained rather stable between early November and early January with no distinct pattern. However, between the middle of December through the beginning of March, it trended up, reaching as high as 500 points per pound. In regards to premiums and discounts, lower staple length levels (below the base) were slightly more discounted in 2005/06 than in 2004/05. Higher staple length levels received considerably higher premium in the 2005/06 marketing year, contrary to 2004/05 where almost no premiums were gained for higher staple length (Figure 9).
As illustrated in Figure 10, strength 26 discounts generally fluctuated between 0 and 60 points/lb for the 2005/06 marketing year. Compared to the 2004/05 marketing year, lower strength levels (below 27-28 gram/tex.) were less discounted in 2005/06 (Figure 11). Similar to previous years, premiums for higher levels of strength continued the trend of seeing minimal premiums, which are lower than their levels in the 2004/05 marketing year.