It worked out for a while as society converted more of its potential scientists into actual ones, but there are only so many such people and now I would judge we are well past the point of diminishing returns. Thus the precipitous drop in the quality of research and the proportional increase in the amount of trivial work being done. Unfortunately, it seems from my experience that this cultural change seems to disincentivise the “true” scientists who are perhaps less than brilliant at navigating complex grant bureaucracies and academic politics.

New results now indicate reproducibility would be significantly boosted (ha ha, little pun there) and false positives similarly suppressed by raising the bar for “p” to .005 and .001 for significant and highly significant results, respectively.

http://www.pnas.org/content/early/2013/10/28/1313476110.full.pdf

Certainly this comes at the cost of more expensive and slower science (need more samples, you see), but maybe a more solid footing will bear out the truth in “slow is smooth; smooth is fast”.