Social conflicts are a serious obstacle to economic growth in many Latin American societies, affecting economic activity both in the short and the long run. The purpose of this paper is twofold. First, we develop a methodology to estimate the direct, short-run impact of frequently occurring exponents of local conflicts, such as road blockings, on economic activity. The methodology is based on using high frequency (daily) data to estimate the relation between a proxy for economic activity and the occurrence of road blockings. Careful analysis of impulse-response functions traces the effect of such local social conflicts on economic activity. This information is combined with the aggregate annual time series relation between gross domestic product (GDP) and the high frequency proxy for economic activity to estimate the direct cost of local social conflicts for the economy as a whole. Second, we apply the proposed methodology to analyse the direct economic costs of road blockings in Bolivia, using detailed daily information for the year 2003. The results show that these costs are large. Stronger institutions are a critical factor in reducing the prevalence of such conflicts and to avoid the huge economic costs that they imply. Copyright © 2008 John Wiley & Sons, Ltd.