Are shocks to copper consumption persistent?

Abstract

This paper investigates whether shocks to copper consumption for 37 countries over the period 1967–2010 are transitory or persistent. A variety of time-series unit root tests is first employed. This is followed by several generations of panel data unit root tests. The presence of structural breaks is taken into account while performing those tests. In addition, cross-sectional dependence is detected and effectively controlled when applying some new generations of panel unit root tests. Copper consumption is found to follow a non-stationary process for about 86 % of the countries. There is also overwhelming evidence of a similar process when panel unit root tests are applied. Generally, shocks to copper consumption are found to be persistent.