Digital ISO is much more consistent and linear than film. It's easy to check as well - just take a shot and wind up the ISO at the same time as raising the shutter speed - mid-tone density will stay the same - in the middle of the histogram. If you did the same with film (heck of a lot of hassle!) you'd often find that fast emulsions failed to hit their marked speed, and might also vary batch to batch.
Having said that, I do a lot of testing of different cameras for my work that involves shooting identical scenes at same exposure values. When they're then imported to Lightroom at default settings, some slight variables are common. But it's not massive, maybe half a stop or so.
Edit: there's another problem with film too - long exposure reciprocity failure. Everything's okay down to about 1sec exposure, but then a theoretical 4secs exposure might actually need 6secs, and a 10secs exposure might need 20secs, then things get rapidly worse so that really long exposures become impractical. There's more - with colour film, the different colour layers have different reciprocity failure characteristics, meaning that a colour cast soon develops and then turns to what's known as 'crossed curves', meaning that it's impossible to correct for accurate colour. That rules out long exposure landscapes using a ten stops ND filter with colour film, and things like star trails are almost impossible. Digital has really revolutionised astro photography.