Better than a glacier but worse than fresh snow. "More than 80 to 90 percent of the sunlight falling on fresh snow is reflected back into space, compared to 15 to 35 percent of the sunlight reflected by most ice."[1].

[1]https://earthobservatory.nasa.gov/features/DirtySnow/page2.p...

Is there a good chemical / spatial argument for why this is so?

I think it is because a flat surface gets one chance to reflect by refraction and predominantly only at a shallow angles. A bunch of micro surfaces in an aggregate formation mean it could bounce around multiple times and then head out, or pass through the ice medium and then bounce back out when it exits and reflects off more. Absorption wise a greater ratio of air to ice means it will absorb less often too over the same distance through snow.

This is correct, it's why things like solar panels have a surprisingly low albedo despite being mostly black, they are completely flat.