Date of Award


Document Type

Honors Thesis (Open Access)


Colby College. Economics Dept.


Guillermo Vuletin


Commodity price uncertainty imposes large costs on society. On the macro level, it results in sudden and unexpected shifts in current account imbalances and real GDP volatility, while on the micro level, it leads to allocation inefficiencies. Accurate price forecasts have the potential to remove some of this uncertainty and allow for a more efficient distribution of resources, and thus, an increase in social welfare. Despite the obvious gains to be had from accurate commodity price forecasts, few models have been able to deliver these results. Chen, Rogoff and Rossi (2008) were the first to find a promising link between exchange rates and commodity prices for commodity-currency countries. Their study shows that exchange rates can be used to accurately predict commodity prices; however, they only analyze countries with floating official exchange rates, which is an unnecessarily narrow approach. This restriction eliminates all of the emerging markets that possess a fixed official exchange rate, despite the simultaneous existence of a flexible-looking parallel exchange rate.

This paper incorporates Reinhart and Rogoff's (2003) exchange rate classification scheme to identify commodity-currency countries that have periods in which their official exchange rate is fixed, but the simultaneous existence of capital or exchange controls creates a flexible looking parallel exchange rate. During these periods, we use a modified version of Chen, Rogoff and Rossi's (2008) model to predict commodity price movements for five countries. This expansion allows the model to be applied to many more non-OECD countries, which suffer significantly more from commodity price uncertainty than their OECD counterparts.


commodity currency, forecast, exchange rate