Consider the relationship between Power ($1 - \beta$) and the effect size in a hypothesis test. If all other factors (sample size, variance, and significance level) remain constant, what happens to the power of the test if the true population mean $\mu_a$ moves further away from the null hypothesis mean $\mu_0$?