Grant proposals don't cover things that don't directly cover the project. Journal subscriptions, facilities, copy paper, administrative staff, etc. all aren't eligible, and so get rolled into the vague notion of indirect rates.
When an NIH grant is awarded for a particular direct cost amount, the indirect rate is then added on. So for my institution, if I get a $100,000 grant, my institution gets $153,000, because we have a 53% indirect rate. These are periodically renegotiated with the Federal government, and used widely.
Private research companies can and do just tuck this into their costs, but universities aren't allowed to do that.
A 15% rate is, to be blunt, catastrophic. It's "Every University in the U.S. Loses Money on Every Single NIH Grant" levels of bad.
And yes, this is the rate a lot of foundations charge. That's not because it's a reasonable rate - it's because it's a charitable foundation, and we all sort of indulge them because of their missions, and their own need to be efficient. But they're also a minuscule amount of the funding landscape, and often small grants. We tell ourselves its find - they're worthy causes, and loss leaders, etc. But applying that to the NIH will be a massive blow to biomedical research in the United States.