1-2 (10%) Please compare $\text{var}(\hat{\beta})$ with $\text{var}(\hat{\beta})$, the variance of least squares estimator $\hat{\beta}$. Does this result contradict the general result that ordinary least squares is efficient? (8)