The data on costs in the paper by Leitch and Kerestes were categorized by 

 region and drainage type. Random ditch and random tile drainage costs were 

 reported for west-central Minnesota; for south-central Minnesota, general field 

 drainage costs were listed. The authors used a 12% and 8% discount rate to 

 calculate present values; the crop yield on a representative acre of composite 

 cropland in each region was used to calculate gross returns to investment. Extra 

 labor costs were computed for all cropping on all drained wetland acreage, but 

 marginal machinery costs for the drained acreage were assumed to be zero. A 

 before and after tax net return was calculated. Returns were calculated using 

 a 15-year and 25-year life of investment, and under varying assumptions about 

 maintenance expenditures. Thus the net present value of returns for drainage 

 investment reported in ([48]) ($141 per acre for random ditch drainage, $83 per 

 acre for random tile drainage, and $630 per acre for general field drainage) 

 are before tax returns on 15-year investments that were calculated using a 12% 

 discount rate. 



51. Lynne, G.D., P. Conroy, and F.J. Prochaska. 1981. Economic valuation of 

 marsh areas for marine production processes. Journal of Environmental 

 Economics and Management 8:175-185. 



Lynne, Conroy, and Prochaska calculate some of the wildlife habitat 

 preservation benefits of coastal wetlands by estimating a two-factor production 

 function for the Florida blue crab harvest. The habitat, or environmental 

 amenity input, is quantified by the areal extent of wetlands. The time series 

 data for the areal extent of wetlands were established by aerial photographs. 

 The other factor input for this production function was human effort. The effort 

 variable was quantified by the average annual number of traps used in harvesting 

 Florida blue crabs. The dynamic econometric model is based on the Verhulst 

 (logistic) dynamic population growth equation. 



The model was estimated by ordinary least squares (OLS) regression 

 techniques for a dynamic reduced form equation: 



^t ~ ^o ■*■ ^1 '^i " ^2 X2 + C3 C^.j + e^^. 



The use of the lagged value of the dependent variable "dynamizes" the equation; 

 C^, the harvest in year t, is the dependent variable, while the harvest lagged 

 one period (C^.j) is an independent variable. The other two independent 

 variables represent interactive terms that are the product of the natural 

 logarithm of marsh acreage (In M^), and Ej., the effort variable, in the case of 

 X^, and the product of (In MJ and (E^ji in the case of X2. The randomly 

 distributed error term is e^. 



The equation was estimated from data covering the 1952-1974 period. Photos 

 for the various marsh areas were available for 3-6 years of the 23-year period; 

 linear interpolation was used to establish the values of marsh acreage for other 

 years. To establish the value estimates, a marsh was defined technically as a 

 small bay or estuarine area less than or equal to 1.5 miles in width where the 

 saltwater table is at or above the land surface and river inlets up to the tree 

 lines. The effort variable was quantified by the average number of blue crab 

 traps used during the year. The overall R^ of the estimated equation was 0.78; 

 the Durbin-Watson statistic was 2.05, indicating the absence of autocorrelation 



41 



