Fj = ^hj,a + p) 



(10) 



(=1 



Data from the National Marine Fisheries Ser- 

 vice were used as annual observations on the 

 nominal value of gross investment in Gulf fish- 

 ing vessels (U.S. Department of Commerce 1965- 

 77^). These values were deflated to real terms 

 using the industrial price index. The quality of 

 the time series for real net investment in fishing 

 vessels, N/, depends on how well the annual 

 values of/, reflect quality changes in vessels over 

 time. 



The annual levels of the implicit rental price of 

 vessels (c) were computed using the definition 

 outlined on the right-hand side of Equation (2). 

 Coen (1975) assumed that the real after-tax rate 

 of return desired on equity capital, p, is constant 

 over the economic life of the investment. Follow- 

 ing the lead of Coen, a value for p of 5% was 

 employed in this study. 



The real rate of interest on nonreal estate loans 

 at commercial banks, r, along with annual rate of 

 inflation equals the nominal rate of interest on 

 debt capital. Annual values for all these variables 

 were obtained from US. Department of Com- 

 merce publications. The annual values for the 

 fraction of investment expenditures that are debt 

 financed (v];) used in computing Z were found by 

 dividing the annual change in total debt in the 

 fishing industry by the annual level of gross 

 investment in durable inputs provided by the Na- 

 tional Marine Fisheries Service. The time series 

 for a, the fraction of capital expenditures fi- 

 nanced with internal equity capital, was equal to 

 one minus the percentage debt financed 

 (l-il/). 



Investment tax credit rate, i^, was equal to 7% 

 during the 1965-68 period, 0% during the 1969-70 

 period, and I09c during the 1971-77 period. The 

 maximum corporate income tax was assumed to 

 represent i^ for the Gulf shrimp fishery. The 

 double-declining balance method was assumed in 

 determining the present value of the stream of 

 annual tax depreciation allowances in A . 



The time series data on prices paid for vessels, 

 a component of the rental price, were measured 

 using cost data collected from shrimp vessel 

 builders. Griffin et al. (1978) have shown that 



vessel length, material of construction, and year 

 of purchase were the most significant factors de- 

 termining the price of a vessel. The equation esti- 

 mated in that study was used to extend available 

 vessel price information over the entire time pe- 

 riod covered by this study. 



Econometric Results 



Statistical as well as economic criteria can be 

 employed to evaluate the estimated equations for 

 the various categories of fishing vessels. The eco- 

 nomic criteria include the reasonableness of the 

 elasticities for the economic variables and as well 

 the partial production elasticities implied for the 

 production function. 



Empirical estimates of the annual real net in- 

 vestment model for steel, wooden, and fiberglass 

 vessels indicate statistically significant coeffi- 

 cients for all but one of the explanatory variables 

 at the 10% level or less (Table 1). The lone excep- 

 tion was the coefficient associated with the lagged 

 capital stock variable in wooden vessel model, 

 which was not significantly different from zero at 

 less than the 20% level. All the coefficients associ- 

 ated with the explanatory variables have the 

 signs hypothesized earlier in this paper. Finally, 

 the coefficients on the lagged dependent variable 

 satisfy the constraint of being both greater than 

 zero and less than one. 



Table 1 . — Estimated coefficients for the annual net investment 

 model for fishing vessels. Gulf Shrimp Fleet, 1965-77. 



8U.S. Department of Commerce. 1965-77. Vessel charac- 

 teristics data. National Marine Fisheries Service, NOAA, 

 Wash., D.C. 



1 Numbers in parentheses indicate absolute values of f -statistic. 



Economic criteria employed in evaluating the 

 reasonableness of the empirical results and in 

 comparing the investment behavior among vessel 

 types include the partial production elasticity of 

 fishing vessels (p). This elasticity is given by 

 -61/62 which is computed using the estimated 

 beta coefficients in Equation (7). It appears that 

 steel vessels, with a partial production elasticity 

 of 0.479, are highly productive and play an impor- 

 tant role in the supply of fish. The fishing sector, 

 unlike other sectors of the economy, depends pri- 



154 



