This paper incorporates export price shocks into the analysis of the effect of aid on growth. Previous analysis of the aid-growth relationship by Burnside and Dollar found that aid is more effective in raising growth the better are policies. However, this result has been criticized for being sensitive to choice of sample and for neglecting shocks. We construct export price indices using the approach pioneered by Deaton and Miller. We locate shocks by differencing the indices, removing predictable elements from the stationary process, and normalizing the residuals. Extreme negative shocks are the bottom 2.5% tail of this distribution. Introducing these extreme shocks into the Burnside Dollar regression, we find that they are highly significant: unsurprisingly, extreme negative shocks reduce growth. Once these shocks are included the Burnside and Dollar results become robust to choice of sample. Further, the adverse effects of negative shocks on growth can be mitigated by offsetting increases in aid. Indeed, targeting aid towards countries experiencing negative shocks appears to be even more important for enhancing aid effectiveness than targeting aid to countries with good policies. However, we show that to date donors have not in aggregate used aid for this purpose.