The unfolding of events connected with a change in the economy, as a consequence of the multiplier effect. Cumulative causation can be set in motion where the expansion of the cluster, through attraction of firms with complementary competencies, adds to the initial attractiveness of the cluster (Maskers and Malmberg (2007) J. Econ. Geog. 7, 5).

Kaldor (1981) Econ. appliquée 34, 4 argues that any initial differences in productivity between regions are sufficient to produce progressive, self-reinforcing divergence—but this demonstrates how the nature of ‘historicity’ in the new economic geography models is very circumscribed (Martin and Sunley (2006) J. Econ. Geog. 6, 4). Improvements made through cumulative causation are made at a cost to some other part of the economy; although underdeveloped regions offer the advantage of low-wage labour, these benefits tend to be offset by the agglomeration economies in the industrialized regions. ‘Therefore in order to overcome the [negative] effect of cumulative causation intervention on the part of government is necessary’ (Cibulskiené and Butkus (2007) Jahrbuch für Regionalwissenschaft 27). The process is not limitless and can be reversed through negative externalities, such as congestion costs, or reductions in trade costs. Simulation models tend to show that large changes in trade costs may be required to cause deconcentration and that at intermediate levels of trade costs concentration remains high (Weiss (2005) ADB Institute Discuss. Paper 2005/04).

Cumulative causation