I looked at that idea, as well as aluminum foil. Mirrors use rear surface Aluminum Coatings. Aluminum reflect 89%, while the glass transmits 96%.
http://img40.imageshack.us/img40/4215/alumspect.jpg Actually I'm reflecting 82% (.96x.96x.89). As far as spectrum reflected, the soda-lime glass blocks about 40% of UV light below the 380μm range, which is outside the PAR curve. I'm starting seeds inside for transplant outside, so I really want more blue light.
http://imageshack.us/a/img560/3693/daylightcflcurve.jpg
I ran a spectrograph on gloss white paint and found it reflected between 60% and 35% within the PAR curve. I didn't run a curve for flat white. I have to admit that, even though the reflectivity is much lower, the diffusing effect would provide better coverage but significantly less lumens.
As a retired optical engineer, I have spent significant time evaluating and running curves on commercial growlights, and as a byproduct of that effort I looked at different mediums for reflection. Actually Aluminum Foil works extremely well, except it is difficult to keep clean.
And by the way, a combination of daylight 6500k and Soft White 2800k fluorescent bulbs provide the best spectral output relative to PAR, and definitely the most bang for the buck.
Hotwired NY 5b