In a simple linear regression $Y = \beta_0 + \beta_1 X + \epsilon$, if the sample size $n \to \infty$, what happens to the variance of the OLS estimator $\hat{\beta}_1$?