KIRKLEY AND SQUIRES: CAPITAL STOCK AND INVESTMENT IN A FISHERY 



measurement of capital. In simple terms, invest- 

 ment is the exchange of money for some form of 

 property. Although there are several types of in- 

 vestment (Branson 1972), attention is restricted 

 in this study to an examination of net investment 

 (i„) which is defined as the difference between the 

 stock of capital in two periods of time (Baumol 

 1977). This is the same concept of investment ex- 

 amined by Tettey et al. (1986). 



Replacement investment is another type of in- 

 vestment. The sum of replacement investment 

 and net investment equals total investment. Re- 

 placement investment is measured by multiply- 

 ing the rate of depreciation times the level of the 

 capital stock. Replacement investment, however, 

 is not estimated in this study since the method 

 and rate of depreciation must be arbitrarily as- 

 sumed. 



METHOD OF ANALYSIS 



The Hedonic Approach 



While the definition and measurement of real 

 capital stock and investment are conceptually 

 straightforward, limited data and heterogeneous 

 capital inputs complicate the measurement of 

 capital stock and investment. In particular, data 

 are usually only available for a small number of 

 vessels which are often similar in their character- 

 istics; these data are inadequate for calculating 

 actual or observed capital stock and investment 

 in a fishery. However, data are available on the 

 acquisition price and characteristics of vessels 

 which permits estimation of the stock of capital 

 by an hedonic approach. In turn, estimates of cap- 

 ital stock from the hedonic approach can be used 

 to estimate net investment in a fishery. 



The hedonic approach hypothesizes that the 

 price of a commodity is influenced by its charac- 

 teristics (Rosen 1974). The hedonic price or char- 

 acteristics function in market equilibrium re- 

 flects both the distribution of marginal rates of 

 substitution over households and the distribution 

 of marginal rates of transformation over firms. In 

 effect, the hedonic hypothesis states that goods 

 may be valued for their attributes and that im- 

 plicit or hedonic prices exist as a function of the 

 attributes. 



Griliches (1971) estimated hedonic cost func- 

 tions of U.S. automobiles using advertised or list 

 prices and actual transaction prices. Griliches, 

 however, was concerned with explaining quality 

 differentials. Ladd and Martin (1976) examined 



prices and demands for input characteristics 

 using a Neoclassical input characteristics model. 

 Triplett (1986) estimated quality adjusted price 

 indexes for computers. Thus, there is a history of 

 deriving prices of commodities as a function of 

 their characteristics. 



The hedonic approach offers several attractive 

 properties for estimating capital stock in fishing 

 industries. First, the hedonic method incorpo- 

 rates changes in the quality of capital over time 

 (Triplett 1986). This is because changes in quality 

 should be reflected in vessel acquisition prices. 

 The hedonic approach, thus, permits a quality ad- 

 justed measure of capital stock to be obtained. In 

 contrast, vessel count ignores changes in the 

 quality of capital. Second, the hedonic approach 

 allows easy aggregation of the heterogeneous cap- 

 ital inputs frequently observed in a fishing fleet 

 because the heterogeneous capital is measured by 

 value rather than physical measures. 



In the hedonic approach, cost (C) or input price 

 may be expressed as a function of the associated 

 characteristics of the commodity or input, 



C = f(CHi, CH2, ...,CH„) , 



(1) 



where C is cost, CH, is the tth characteristic 

 (Braeutigam et al. 1982). The characteristic cost 

 equation can be obtained either from a dual speci- 

 fication or by determining the reduced form equi- 

 librium equation. 



The Dual and Reduced Form 



The dual specifies cost as a function of input 

 prices and their characteristics. The partial 

 derivative of the dual cost function with respect to 

 input prices yields the demand for the inputs as a 

 function of prices and input characteristics. The 

 reduced form is obtained by solving structural 

 equations, equations which explain the behavior 

 and interrelations of endogeneous variables, of 

 demand and supply for the endogeneous variable, 

 price, in terms of exogeneous variables and dis- 

 turbance terms. ^ 



In this study, the reduced form is directly speci- 

 fied and estimated for two reasons: First, all of the 

 input price data necessary for estimating a dual 

 are not available. Second, estimation of the struc- 

 tural equations is complicated by the need to use 

 a limited dependent variable method of estima- 



^See Rosen (1974) for additional information on obtaining 

 reduced form equations for hedonic prices. 



341 



