In the estimation ofour extended mo

In the estimation ofour extended model (Eq. (2)), we expressly recognise the endogeneity of some corporate policies that may affect several explanatory variables. Examples of such policies could include demanding financial debt, which implies an increase in cash as well as a variation in a firm’ capital structure, or a new issue ofcapital with analogous effects. In a Generalised Method of Moments (GMM) framework, we use lags 2–6 of the endogenous regressors included (CASH, COLLATERAL and LEVERAGE) in addition to the exogenous regressors (SIZE, GROWTH and CASH_- FLOW) as instruments in Eq. (2).5 Instrument validity is checked via Hansen’ (1982) J-statistic, which in light ofour instrument set, reduces to a x2(12) statistic. For brevity, we only report (see Table 5) the estimation results of our extended model (Eq. (2)) considering the dependent variable EXTFIN_D + E. Results hold when regressing this model with debt or equity, separately.