This is related to the variance: on one hand, we define ${\rm Var}(R)=E[(R-E[R])^2]$, which is (obviously?) nonnegative. On the other hand, you should expand the above expression in order to show that ${\rm Var}(R)=E[R^2]-E[R]^2$. Since this is nonnegative, you must have $E[R^2]-E[R]^2\geq 0$, and this is it. So you need to justify why the first expression of ${\rm Var}(R)$ is nonnegative, and why it equals the second one.