I talk about game development time, how it has increased for some games and how this affects game franchises.
One con he didn't mention about long development times is how a large part of your active life is wasted on a single game that may not even have players play it.
Another con is that chances are your "fresh new idea" has become an overdone cliche by the time you finish.
Another con is that existing team members move on to other places and new members come in and nobody aside from the lead (assuming they aren't also replaced) is really invested in the game. From a programming perspective this also has the extra bonus of ossifying any system written by someone who left that has been around long enough to have gameplay designed on it.
And of course an obvious con is that the longer develpment times mean more expensive development which mean more demands for sales which mean the game needs to be designed so that it can target the largest audience. Actually i should have thought this first.
There are really lots of cons IMO. According to Masters of Doom (book about the development of id software), John Romero once thought (in the 90s) that spending a year on a game was crazy - and in a way i agree with it.
I also agree with Tim that there are things you can't do in a short time frame - but as he said, you don't need to have everything in a single game. And since he brought up Monarch, personally i'd say that making it larger than it was wouldn't make the game any better (chances are it'd make it worse).
Of course that is all pretty much against the AAA+ trends which are all about having the largest games with the largest numbers (world size, quest count, map icons, polycounts, etc) made by the largest teams with the largest budgets for the largest possible target audience. Often for the largest price and followed by the largest number of DLCs and microtransactions :-P.