By Stephan Klasen, University of Goettingen; Maria C. Lo Bue, University of Goettingen; Vincenzo Prete University of Goettingen
Sustained economic growth and substantial progress in reducing poverty have taken place in many parts of the developing world over recent decades. Yet, large parts of the population in many countries continue to be highly vulnerable to poverty. A key challenge to establishing a development strategy for reducing vulnerability to poverty lies in the interplay between distribution and growth.
This paper considers how shocks can affect the distributional pattern of growth and thus poverty reduction, thereby bringing together the literature on the distributional pattern of growth with the literature on shocks and their impacts.
Standard growth incidence curves (GICs) describe how growth episodes impact on the overall income distribution. Yet, nothing can be said about the drivers of such episodes.
To this regard, on may note that, by construction, GICs may give a misrepresentation of true growth due to measurement errors. At the same time, the variation in the incidence of growth across the different percentiles can also be the effect of some shocks that hit these percentiles in different ways.
By relying on a non-anonymous axiom, we compare actual growth episodes at each percentile of the initial personalized distribution with a counterfactual pattern of predicted growth which, by relying on the assumption of time-constant marginal returns of individual endowments, rules out the presence of shocks. The comparison between the observed and the counterfactual GICs allows understanding to what extent the way that growth shaped individual income trajectories has been the outcome of unexpected changes in the marginal return of individual socio-economic characteristics which substantially changed individual ranking in the income distribution.
Using empirical illustrations based on longitudinal survey data, we show to what extent the degree of the pro-poorness of growth in Indonesia has been the outcome of patterns of mobility and vulnerability. Our results show that growth has been pro-poor over the period 2000-2014, with the incidence of growth in the poorest quintile being larger than expected.
