Well, part of the purpose of that article is not "current state" but future state (like up to 20 years). This is important to do. It is also important not to completely equate today's reality with projections of 20 years from now. Any look at demographics will tell you how important that is. My group is 50% composed of people who are planning to retire in the next 10 years or so (although not in the next 2-3 years). Other groups in my area also have a sizeable contingent of 60 year old pathologists. While no doubt some will keep working until they are 75, if you actually talk to most of these people most do not have the will or desire to work that long. Can that change? Yes. But in terms of my group several are already sketching out their timelines. Some are putting out 65 as the date, and that is not moving. Others are saying 67-68.
One major key is the ability of these older pathologists to change their duties. If groups allow senior pathologists to gradually phase out, eliminating the things they find objectionable and just focusing on what they want to without significant decline in compensation, people are going to retire later. Our group is actively discussing this issue. Older pathologists shouldn't be able to cherry pick duties (like say, they just want to do GI biopsies now) and dump all the admin on younger pathologists while still getting paid the same way. I suspect this is going to happen in a lot of other places though, especially ones where younger pathologists do not have as much influence in group workings. Older pathologists could potentially bleed the rest of the group dry.
Younger pathologists definitely need to be aware of this. You don't want to be stuck in a group where the senior partners 1) Get more vacation time, 2) Don't do offsite duties, 3) don't do much management activities, 4) cherry pick the standard straightforward cases, but 5) get paid the same.
But yes, I am curious about the demand article also. Maybe it is more complicated (which it should be!) because there were factors they had not initially considered. Supply is much easier to forecast.