About 2/3 to 1 stop in my experience and also according to most sources. I.e. Foma 100 is ca. 64, the 200 is really around 125 and the 400 is around 200-250 for most people. "Several stops" really is excessive; I never observed that with any Foma film.
If you check the Foma data sheets for all 3 pan films, 100, 200, and 400, Foma posts times and curves for a variety of developers. . Foma 400 comes close to 400 with ILford Microphen, with Fomadon a bit slower but still close to box speed and D76?ID 11 200 to 250, Foma 200 is closes to box speed with most developers while 100 seems to exceed box speed with Mircophen and at box speed with others tested
Are you taking into account the fact that the datasheets show a range of G-bars? If you take 0.6 as an approximation of 'normal contrast', you'll find that the Foma 400 product really achieves 200-250. It's not close to 400 in any developer shown in the datasheet. It does peak out at around 320, but this is only for significantly higher contrast; i.e. essentially a push process scenario.
The 200-speed product reaches approx. 160 in 3 out of 4 developers listed, but interestingly for D76 no g-bar 0.6 rating is provided; extrapolating it would arrive at around 125.
Foma 100 reaches, at least according to the datasheet, 80-100.
However, curve shape and tonality IMO should be taken into regard as well, and it turns out that the 100 and 400 speed products produce a linear curve especially if you hold back on development. This drops effective speed a little, which I suspect explains why also the 100-speed film is often regarded as performing best if rated a little lower than box speed. Note that this isn't true for the 200-speed film, which remains linear across a broader range of development times. The 400-speed film shoulders off quite readily and those who have tried to use it for alt. process printing can attest to this; it can be hard to build the required density range for certain UV-sensitive processes (salted paper, carbon etc.)
I haven't had time to cross check by processing a roll of Kentemyere in F76+, still not sure what happened or why. I don't think it is the film, no one else has reported a loss of speed. My process has been standardized for decades. I use a dial thermometer calibrated against a Kodak Color thermometer, distilled or deiononized water, for one or two rolls Patterson tank for larger batches Unicolor film drum times adjusted -15%. Times are the same, in the past I found F76+ times are spot on with times for D76. F76+ 1:9 is the same as stock D76, I have used at 1:19 with same times as D76 1:1.
I just tried a roll of Arista EDU 400 developer in DK-60a 1:3 at 9 min. 20/68 deg. and it looked like like a 200-250 film to me. I developed it in a tank that had a roll of hp5+ exposed at the same time and it looked a bit over developed and looked like a 400-500 speed film. The HP5+ was a bit dense so I think 7.5 min would have been fine. After the overcast conditions leave I will re try both of them. The emulsion # is 012556-1 exp 5 2028.