Consider the following two regression models for the log earnings of German and foreign male workers: Model 1: Model 2: Yi= a + Fi + Ei, Yi a2 + 3 Fi + 22, +Ei, where F is a dummy variable that takes on the value 1, if worker i is a foreign worker and 0, otherwise. yi represents log gross earnings and zi is a vector of other explanatory variables such as schooling and experience. 1. What is an interpretation for the coefficients and 32 in the two models? 2. Consider now a third specification: Model 3: Yi = a3 +33 Fi + ziy + (Fizi)'d3 + Ei (i) How can you interpret the coefficient vector 63? (ii) What is the difference between specifications 2 and 3? 3. Show that Model 3 can be estimated by two separate regressions.