Every fall, departments move their numbers, and every spring, applicants discover the move. Stipends shift with competition, cost of living moves underneath them, and the program that was a good deal last year is not always a good deal this year. Here is what I am watching this application season.
First, the raisers. When one prominent department bumps its stipend, peer programs feel pressure to follow, and these moves cluster in the fall. An offer that looked average in October can look below-market by March, or the reverse. Timing is not everything, but it is not nothing either.
Second, the cities, not just the stipends. A department can raise its number five percent while local rents climb ten. The nominal figure went up and your life got worse. This happens constantly and nobody announces it. You have to check it yourself, every cycle, for every city on your list.
The practical move is to run every offer through a PhD stipend database search tool by program, not a generic list of last year's averages. You want this season's numbers, per program, with cost of living attached. A solid stipend comparison tool in 2026 updates with the market instead of recycling stale figures from three cycles ago.
Third, the fine print on new guarantees. Some programs answer competition by extending guaranteed funding years instead of raising the stipend. That is often the better deal, and it never shows up in a headline number. Ask about years, not just dollars. Years are the part that protects you.
My honest advice: do not chase a single year's ranking. Rankings are snapshots, and a snapshot from the wrong angle misleads. Chase the structure. A program with solid guaranteed funding in an affordable city beats a flashy stipend that barely covers rent, this year and every year. The fundamentals move a lot slower than the headlines. Compare this season's offers with cost-of-living adjustments at phdstipend.fyi, across 798 US programs, and make the decision on structure, not hype.