internal rate of return on field drainage investment in lowland meadows with good 

 clay soils). From a policy perspective, perhaps the two most important reasons 

 for this high rate of return are the ongoing government investments in public 

 drainage and water projects that often greatly increase the physical effective- 

 ness of each private dollar spent on drainage activity. Also, there are still 

 certain U.K. government programs that directly and indirectly subsidize 

 agricultural drainage of wetlands, including U.K. tariffs and quotas on 

 agricultural products. Some of these indirect subsidies also affect drainage 

 investment in the U.S., though the impact is difficult to measure. 



79. Farber, S., and R. Costanza. 1987. The economic value of wetlands 

 systems. Journal of Environmental Management 24:41-51. 



Costanza and Farber apply both the conventional methodology of marginal 

 economic analysis and an energy theory of value approach (see [8] and [9] for 

 discussion of the energy theory of value or ecosystem life support function 

 approach to wetland valuation) to assess the amenity values provided by a wetland 

 system in Terrebone Parish, Louisiana. They attempt to make a summary estimate 

 using the conventional approach by employing the reasonable assumption that the 

 outputs they consider have a total value that can be derived by adding the values 

 of the individual goods and services. Costanza and Farber estimate a marginal 

 value product for wetlands in commercial harvesting, an aggregate willingness- 

 to-pay or consumer surplus figure for the social benefits conferred by 

 recreational activities pursued on these wetlands, and a wind damage protection 

 amenity value conferred by the wetland. 



Costanza and Farber estimate a marginal value product for wetlands in the 

 commercial harvesting of shrimp, blue crab, oyster, menhaden, and commercial 

 trapping (primarily for nutria and muskrat). To estimate the marginal product 

 of wetland acreage in shrimp production, the authors first tried to estimate an 

 equation introduced by Lynne, Conroy, and Prochaska (see [51]) to calculate the 

 marginal product of wetlands in the harvesting of blue crabs: 



Q = B„ + (B^ E) In W.j + (B^ E^) In W.^ + B3 Q.; + a. 



In the above equation, Q is the annual harvest, E is a scaler measure of 

 human effort (quantified as the number of man-hours for the shrimp harvesting 

 equation), W is a scaler variable that quantifies the habitat (environmental) 

 input, and a is a random error term. The subscripts indicate lagged variables; 

 thus habitat of the previous year determines the size of the current harvest. 

 However, when this equation was regressed on annual data by Farber and Costanza, 

 the estimated coefficient for the quadratic term in E had the wrong sign. Note 

 that the marginal value product of habitat (partial derivative of Q with respect 

 to W) is 



MP = (Bj E + B2 E^) W"^ . 



Farber and Costanza estimated a static version of the harvest equation used 

 by Lynn, Conroy, and Prochaska ([51]) in which only the current value of the 

 habitat variable is used to explain the harvest level. The estimated annual 

 marginal products were 1.60 pounds per acre for brown shrimp, 1.44 pounds per 

 acre for white inshore shrimp, and 0.90 and 1.23 pounds per acre for white and 



57 



