For years (since around 1940) the NSF has often required cost sharing on projects that it has funded. This was especially true for projects that bought equipment. That requirement has now been all but eliminated.
This is important because it removes a bias against smaller institutions. Equipment proposals are often large and hence the matching portion would be large. This has prevented less well off institutions from competing for important Federal funds and over the long-haul would limit the frontier across which we can advance our understanding.
There is a caveat of course. Another implication is that because NSF must now fund all of its projects, it will be able to fund a smaller number of them and thus the competition for limited funds will become more intense, but it still provides an opening for high quality efforts from smaller institituions.
A final caveat is that this move also frees up a pretty good chunk of discretionary money at larger institutions. How will that money now be used?...
Aside
If the link above has expired and you are a Chronicle subscriber, the article can be found here. If you are not a subscriber, contact me and I will get you the text...
End