Testing the growth effects of structural change

Against the backdrop of Baumol's model of ‘unbalanced growth’, a recent strand of literature has presented models that manage to reconcile structural change with Kaldor's ‘stylized fact’ of the relative constancy of per-capita real GDP growth. Another strand of literature goes beyond this, arguing that the expenditure shifts toward Baumol's ‘stagnant’ sector stimulate rather than dampen long-term economic growth because of the human capital-accumulating nature of major ‘stagnant’ services (like health care and education). This paper tests the relationship between structural change and economic growth empirically by means of a Granger-causality analysis of a panel of 18 OECD countries.

► The paper tests the relationship between structural change and economic growth empirically by means of a Granger-causality analysis of a panel of 18 OECD countries.
► Structural change is defined as shifts of expenditures toward the stagnant – yet human capital forming – sectors education and health care.
► No evidence was found that structural change so defined raises economic growth endogenously.
► When Japan is excluded from the sample; there is a statistically significant and long-lasting negative impact of structural change on economic growth.
► This finding is more in line with the predictions of Baumol's model of unbalanced growth than with models of balanced economic growth in the aggregate.