This paper raises concerns about the econometric approach used in the literature to estimate credit unions’ production technologies. We show that the existing studies did not recognize heterogeneity amongst credit unions’ technologies as captured by (endogenously selected) differing output mixes. Failure to account for the above leads to biased, inconsistent estimates and potentially misleading results. The estimates are also likely to be biased due to unobserved credit union specific effects that the literature broadly ignores. To address these concerns, we develop a generalized model of endogenous switching with polychotomous choice that is able to account for fixed effects in both the technology selection and the outcome equations. We use this model to estimate returns to scale for the U.S. retail credit unions from 1994 to 2011. Unlike recent studies, we find that not all credit unions enjoy increasing returns to scale. A nonnegligible number of large institutions operate at decreasing returns to scale, indicating that they should either cut back in size or switch to a more efficient technology by re-optimizing the output mix.