The “source” link goes to an article that talks about a possible 2% figure. I avoided using the combined 3.4% figure as many have said it might be a high-side estimate, so went instead with a lower 2% figure. However, in practice the CFR (case fatality rate) is proving to be much higher than 3.4% in many countries now tracking spread, in part because testing has been so constrained. It is possible that over time CFR will trend down to below 1%, but that won’t make hospitals less full right now, so I generally have tried to downplay the importance of the mortality rate.