Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Collective models identifying resource shares are promising tools to analyze intra-household welfare and poverty. However, their empirical application has proven difficult in practice as authors contend with large standard errors and unstable estimates. This paper uses a prominent framework to show how a common feature of the structure of these models makes the task so difficult and proposes an empirical strategy to stabilize the estimates.