Looks like the registration on the color has moved. This could be the sheet feeding or a different run of chips. I'd never know for sure.

You can figure that they print pages of black and then when it dries, load the same pages and print the next color. Then the circles are probably die-cut out of the page.

Personal opinion for my collection, no I wouldn't consider them variations, unless there was a font change or something that's significant. Same as I personally choose to ignore variations in UV stamps, on the same chips, where the location and rotation of the stamp can vary with each batch from Paulson.