The EU's own analysis of the matter is that the difference is due to a difference in private sector spending and that it should be made up through increased private sector spending on research and the development of new products (rather than shifting more public sector spending from education to research, which may solve one statistical/accounting problem but creates 10 other very real problems).

The difference in private sector spending points at something real. Of course, the figures need to be broken down, as you indicate. It may be that in the case of the US it is merely spending on military development and patent trolling. That's an interesting topic for further research.

I think that there is a fairly continuous line from fundamental research to applied research to new product development, and therefore it makes sense to group them together when doing an analysis in an economic context.