INTERACTION EFFECTS OF PROMOTION, RESEARCH, AND PRICE SUPPORT PROGRAMS FOR U.S. COTTON



The real exchange rate, promotion, and nonagricultural research were initially included as well, but
inclusion of these variables did not lead to significant improvement in the model. It appears that
WPCOTTONt is a much better indicator of export demand for cotton than a more general exchange rate.

The domestic supply of raw U.S. cotton was modeled at the annual level because the planting
decision is made on an annual basis. Although there may be some response of production to changes in
price after planting (e.g., higher abandonment at low price), it is likely to be relatively small. The supply
function is based on a simplified version of (17). It is modeled as a function of expected cotton price, an
index of farm input prices lagged one year, and a trend variable (to capture technical change). A linear
model was estimated incorporating the following variables (the corresponding variable name from (17) is
included in parentheses):

PRODt (Qfsd)          = annual U.S. cotton production (thousands of bales)

FPCOTTONt (EPcd)   = real cotton futures price averaged over planting months (cents/lb)

PINPUTt-1 (Wf)       = index of real prices paid by farmers for inputs

TRENDt (Ra)         = trend variable that indexes years, increasing from 1 in 1975 to 26

in 200022

More complex specifications were considered,23 but the data currently available are inadequate to
allow much complexity beyond the current model. For instance, the data series is not long enough to
permit inclusion of additional variables that may influence supply (e.g., agricultural research, labor
productivity, returns to alternative crops) while maintaining sufficient degrees of freedom, especially
given the long lags expected on agricultural research.

22A trend variable was included as a proxy for the effects of agricultural research due to the difficulty in estimating a
robust specification as a function of agricultural research directly. The trend variable will provide an upper bound
on the supply shift caused by agricultural research because it also captures other factors that may increase supply
over time.

23For example, we estimated models that included CI agricultural research expenditures; total cotton agricultural
expenditures (including USDA, the State Agricultural Experiment Stations [SAES], and other nongovernmental
organizations such as CI); cumulative research at a variety of depreciation rates; and other variants on research
expenditures. In addition, we attempted to estimate production functions and to relate yield and production costs to
cotton agricultural research independently.

19



More intriguing information

1. Tourism in Rural Areas and Regional Development Planning
2. NATURAL RESOURCE SUPPLY CONSTRAINTS AND REGIONAL ECONOMIC ANALYSIS: A COMPUTABLE GENERAL EQUILIBRIUM APPROACH
3. The name is absent
4. Growth and Technological Leadership in US Industries: A Spatial Econometric Analysis at the State Level, 1963-1997
5. The name is absent
6. American trade policy towards Sub Saharan Africa –- a meta analysis of AGOA
7. The Effects of Reforming the Chinese Dual-Track Price System
8. The name is absent
9. A model-free approach to delta hedging
10. Declining Discount Rates: Evidence from the UK
11. WP 1 - The first part-time economy in the world. Does it work?
12. Fiscal Sustainability Across Government Tiers
13. The name is absent
14. DISCUSSION: ASSESSING STRUCTURAL CHANGE IN THE DEMAND FOR FOOD COMMODITIES
15. The name is absent
16. The magnitude and Cyclical Behavior of Financial Market Frictions
17. The name is absent
18. The name is absent
19. On the job rotation problem
20. Optimal Private and Public Harvesting under Spatial and Temporal Interdependence