My main guess is that they want to give the students a specific target at the start of their senior year, It is also possible that when you combine the time to process and compare and then get the information in the programs and on the diplomas, it is faster if the target is already set, although with computers being so fast now I doubt that is the issue, but still it could be for reasons I am not thinking of. In any case the year to year variation is extremely small I am sure, as the Duke data would indicate. I doubt Tulane is any different.
As far as summa always being top 5%, the answer is no for Tulane and no in general, there is no accepted definition. Schools are free to use whatever criteria they want, although in my limited searching the top 5% or systems that come close to that seem to be the most common. Tulane actually used to be a lot easier to get summa as far as GPA, and this answers your first question about the old system. Until 2 years ago (or maybe it was 3 years ago) summa was a flat 3.80 and magna a flat 3.60, BUT you had to be in the HP as it was constituted at the time and that required doing a senior thesis as well, as well as taking some honors courses, but there weren’t that many.
I never liked that system. To now answer your last question, there has indeed been a lot of grade inflation in those intervening years and at most schools. Some of it is justified in the sense that so many students come in better prepared than in the past, with a lot more AP courses and even college courses at a local school. This subject has been argued to death over the last 20 years give or take, so we won’t beat it up again in this thread. But again I point to the Duke data, where the minimum GPA for summa has been running about 3.935. When you are bumping up that close to the 4.0 maximum, 35/100th’s is a lot.
