No, we cannot include age as an explanatory variable in the original model. Each person in the panel data set is exactly two years older on January 31, 1992 than on January 31, 1990. This means that ∆agei = 2 for all i. But the equation we would estimate is of the form