If movie camera shutters weren't fairly accurate, we'd have more problems with flicker and whatnot -- anyway, any variations from exactly 1/48th of a second at 24 fps with a 180 degree shutter is probably smaller than a light meter can be manually offset to compensate for (less than 1/3 of a stop) and less than would be visible to the eye. I wouldn't be surprised that the lenses and perhaps even density from processing are greater variables to final exposure/density results -- not to mention human error in exposing.
If the camera is running exactly at 24 fps and a half-circle disk spins once around during that time, then the gate is open to light 50% of the time. I don't think the extra exposure lost as the disk moves to its most open position and then back to its closed position is significant.
Well, I'm sure there are a lot of discrepancies that creep up with shutter speeds and older equipment, especially equipment that isn't maintained, but being that movie cameras are run off of electrical current or, more and more rarely these days, off of clockwork motors, the error in timing should be constant from frame to frame to frame , save maybe at the beginning of a shot as the camera is gaining speed.
You can certainly still see pulsing or strobing in highlights, often in skies, from some sort of irregularity I think is due to processing in modern films. There's a variation in density or uneven development within a single frame that accounts for this, I think. This probably comes up in release printing more than with developing the negatives, as I'm sure release print standards today have reachd an all-time low.