http://econweb.umd.edu/~murrell/Articles/Can%20Neoclassical%...
> One doubt about the evidence might lie in the comparability of estimates obtained from separate studies, although Table 1 addresses this issue by matching studies with similar methodologies. One study, Koopman (1989b), employs observations on both centrally planned and market economies, thus providing direct comparability. He uses observations for 1960-79 for Soviet republics and a matched sample of Canadian provinces, U.S. states, and Finland, and employs a translog functional form that allows for differences in technology between Soviet and non-Soviet regions. The average level of technical efficiency in Soviet agriculture is estimated at 93 percent, while it is 92 percent for agriculture in the market economies.
