Commodity price shocks, factor intensity and non-primary production growth: Evidence from Colombia

Abstract

This paper studies the differential effects of commodity price fluctuations on non-primary production. I build a simple multisectoral model of price shocks that is used to derive several predictions. These predictions motivate the empirical strategy that leverages i) variation across municipalities (counties) in the suitability to produce coffee and historical patterns in oil production; and ii) temporal differences in commodity prices. The results show that, for a municipality with the median suitability to grow coffee, an exogenous increase in the growth rate of internal coffee price in a standard deviation leads to a reduction in the growth rate of per-capita industrial production of 0,28 standard deviations. On the other hand, in the municipality with a median historical production of oil (conditional on producing at all), the effect is a small but positive increase in industrial production (0.007 standard deviations). The effects on the non-tradable sector, services, in this case, are both positive but small in magnitude (0.03 and 0,001 standard deviations for coffee and oil, respectively). Furthermore, I show that these heterogeneous effects on industry growth are stronger in municipalities with export-oriented industries. I argue that these results are in line with the model's predictions and point to the fact that, to understand the effects of commodity price shocks on other economic sectors, is important to consider the technology used in production and the degree of openness in industrial sectors.