Jones (1995a) gives several explanations for the contrasting relationship between
the state of technology and productivity growth, known in the literature as the
productivity paradox. These explanations are partially inspired by an article written
by Romer (1987). They are that either some offsetting effect occurs in the movement
of other variables that permanently affect economic growth, or that persistent
changes in policy measures which should have permanent effects on economic
growth, in fact do not. After a lengthy empirical analysis Jones concludes that these
explanations cannot be endorsed and that endogenous growth models are therefore
inconsistent with time series evidence.