there should be a downfall (or at least a stagnation) in the ratio of patents granted per unit R&D over time if firms indeed have switched their R&D efforts from non-equivalent to equivalent innovations, especially since the mid 1980s.
figure 3.1 obviously sustains this: the ratio has decreased steadily over the years. There appears to be arise in the ratio of the US and (a lesser extent) the G7 at the end of the sample period though, which is probably induced by the general resurgence of Western economies that occurred around that time.