Title: Identification and (Fast) Estimation of Nonlinear Panel Models with Additively Separable Two-Way Fixed Effects, with Martin Mugnier (CREST)

Abstract: In this paper, we study the identification and fast estimation of a class of nonlinear panel models with additively separable two-way fixed effects widely used in empirical research. We propose a novel identification strategy and show that all structural parameters of the model (heterogeneous slopes, individual/time fixed effects, and link function) can be nonparametrically identified when T is large. We propose a novel iterative procedure to implement the routinely used MLE. This procedure fully parallelizes the updates of the estimates of individual/time fixed effects. This feature largely alleviates computational burdens when the dataset is large and delivers precise estimates using only fractional running time in the presence of a large number of fixed effects. We revisit Aghion et al. (2013) and investigate the causal effect of institutional ownership on firm innovation in the US. Allowing firms to react differently to institutional ownership changes, the estimates obtained by using our method suggest non-negligible heterogeneity in the causal effect, partly explained by the “Superstar Firm Hypothesis”.