This is the point, though. I agree that the layman's interpretation of "created or supported" isn't necessarily the same as the formal definition of the terms

This is the point, though. I agree that the layman's interpretation of "created or supported" isn't necessarily the same as the formal definition of the terms by the US govt, but criticizing Apple for using the formal definition isn't reasonable. The intent of the formal definition is to answer the question "what would be the impact on US employment if this business were to cease operation tomorrow?", and I believe that Apple has made an honest effort to compute that number, probably rather conservatively. Perhaps you disagree, but you seem to be arguing against semantic interpretation of the terminology, instead of being skeptical of the results. In response to your point about how to present this data to congresspeople, I'd argue that while there are some legitimate economic dunces in elected office, there are few that don't comprehend the cascade effect of business activity and therefore employment. Furthermore, I doubt that any congressperson would consider Apple's use of "create or support" to be misleading, because it is absolutely the standard usage of the term in these sorts of assessments. That does not, of course, mean that some senator who smells television coverage won't take notice of the confusion and grandstand his or her condescension for his or her electorate. It might even be a successful strategy, but we don't have to help. Obviously I haven't audited the numbers. That's a valid line of inquiry, but history shows Apple to be fastidiously circumspect in such matters. There's a new sheriff in town in Cupertino, but if anything he seems to be even more careful.