The nice thing about compression on the master bus is that you can get away with a little bit less on the individual channels when you mix into it from the start, and it puts you in control of the mix dynamics because you can ‘lean into’ the compressor when making your fader moves and other mix decisions.
If you don’t compress the mix, the mastering guy probably will if it’s typical rock/pop/electronic stuff. But since putting a stereo compressor on the mix will change the sound of the mix - the relative volume between verses/choruses, the amount the loud stuff (like vocals, kick, snare) sits proud of the quieter stuff, and sometimes brings reverb etc up when it’s applied - there’s a pretty good argument to be made for the mixer dealing with the compression during the mix, when they can actually sculpt what’s going into the compressor
1.5:1, 1/2db GR is pretty mild and a safe bet if your main concern is that you don’t want to damage a mix that already sounds good. Depending on the compressor and material, you might even struggle to hear it doing much of anything at all.
A couple of years ago when I was trying to get a handle on how to deal with mix bus compression, I asked a few working pros with decent discographies over on another forum what kind of settings they used and how much GR they got.
I was pretty surprised how aggressive they were; 3 or 4:1, and anywhere up to around 4dB GR were pretty common answers. Attack and release times were all over the place, too - just looking at attack, someone usually used 1ms, another 2.5ms, another couldn’t deal with less than 30ms… I guess it all just depends how each individual mixer wants to deal with the relationship between channel comps and master bus comp.
At any rate, the take home message I got was not to be afraid to use it. These days I usually see the GR meter move between 1 and 3dB, sometimes in the very loudest parts of the song it takes a wild swing towards 4dB. If I’ve dialed in the mix through the compression to take advantage of the cohesion and punch it can provide, I end up in a situation where the right amount of GR is whatever sounds right. Too little, and the compressor won’t be doing the job I need it to do.
With regard to what your friend said, I can’t agree or disagree because every mixer has their own views and methods. For me personally, it’s definitely not a round of compression that isn’t needed because, for example, if I’m getting an extre 2dB GR when a chorus kicks in then that might be 2dB less compression I’ve needed to put on lots of individual tracks - the mix bus comp gives me 2dB free compression while also gluing the mix and making the whole thing punchier, what’s not to like about that? 
And I’ve never had a mastering engineer complain about the compression, though they have complained about other things… 
lol
$7800?? That’s pretty brutal for a compressor. I’d paid more than that for monitors, but monitors are way more important.
