[Insert Table 3 here]Hypothesis 2 predicts that CEO-CFO LSM is positively associated with the likelihood of CFOs’ becoming board members of the focal firm. We use the event history analysis to test this hypothesis. Such analysis focuses on the length of time that passes until a CFO becomes a board member of his/her own firm. We choose Cox proportional hazard models for the following reasons. First, an event history model like the Cox model incorporates the fact that a CFO can be at “risk” of becoming his/her firm’s board member in a given year and yet has not become a board member. The hazard function estimates the likelihood of becoming a board member in a particular year conditional on the fact that the CFO has yet become a director on his/her firm’s board up to that point. Second, the model uses the time series of information of a CFO in estimating the hazard of becoming a board member. Lastly, the Cox model is semiparametric and makes no assumption about the particular shape or nature of the survival distribution (Cox, 1972).We cluster observations at the CFO level and apply the robust variance estimator option to account for potential correlation of observations.