137
shrimp as shrimp is a substitute for relatively higher priced substi
tutes. This estimate, however, may simply be capturing the positive
correlation over time between the prices of raw-headless shrimp and
competing meat products. The estimated coefficient on the income
parameter RDYt has the wrong sign and is, thus, insignificant. In
addition, the estimated coefficient on the supply variable TCFFt was
expected to have a negative sign, Indicating that as the quantity
supplied changes, retail price should change in the opposite
direction. The estimated coefficient, however, has a positive sign.
This same finding of insignificance was found for the monthly retail
model.
Wholesale structural estimates
The empirical estimates of the parameters and the corresponding
standard errors for the wholesale price model are given below as
W = .247 .051 R + 1.110 P + .041 R .0011 BSFF
(.173) (.137) C (.093) C (.093) t-1 (.0015)
+ .0001 01 .0036 131 + .00033 TMCI
(.0016) C (.0129) C (.00081)
where the values in parentheses are the standard errors. Seven of the
eight estimated coefficients possess the anticipated sign.
Current wholesale price was shown to be very sensitive to changes
in current ex-vessel price (Pfc). The estimated coefficient for ex
vessel price shows that as exvessel price increases by one cent, whole
sale price increases by 1.1 cents. This translates into an elasticity
of price transmission at the means between current ex-vessel and whole
sale price of .932. An elasticity of price transmission less than one
in describing the response of wholesale price to an ex-vessel price