Although that might be based on the perceived likelihood of success. After all, pre-Wiles, I doubt there was much chance of a grad student getting funding to prove Fermat. Nor is a grad student likely to get funded to prove the Riemann Hypothesis.