1. I think passive money is generally not taking any position on whether an asset is over or under priced, and as a class tends to be very inactive. I doubt it either increases or decreases arb that much.
2. I think what would increase arb opportunities would be:
A. A critical mass of misinformed active managers
B. So few active managers that the market is gameable
However, as the number of active managers decreases, one would think that the misinformed ones would go first (A), and that (B) won't be a linear progression, but a sudden tipping point at a very low level of active management.
Actually, I sort of think the final statement is reversed. As the number of active managers declines, and the average active manager becomes more competent, it will become harder for the active managers to find profitable arbitrage opportunities.