The effects of alternative functional forms on parametric estimation o
f a multi-product cost function are examined using a sample of rural U
S banks. Functional forms considered for possible choice are three ver
sions of second-order Taylor-series expansions (or locally-flexible fu
nctional forms-translog, normalized quadratic, and generalized Leontie
f. The generalized quadratic Box-Cox transformation and the Likelihood
dominance criterion are used as vehicles for conducting nested and no
n-nested tests. Although all three forms are rejected by the nested te
st, the likelihood dominance criterion indicates that the translog is
clearly preferred over the other two as the functional form of choice.
Using either of the other functional forms would substantially alter
conclusions about both scale elasticities and inefficiencies, includin
g the impact of bank size on these measures.