In our warming climate there is a general expectation that the variability of precipitation (P) will increase at daily, monthly and inter-annual timescales. Here we analyse observations of monthly P (1940-2009) over the global land surface using a new theoretical framework that can distinguish changes in global P variance between space and time. We report a near-zero temporal trend in global mean P. Unexpectedly we found a reduction in global land P variance over space and time that was due to a redistribution, where, on average, the dry became wetter while wet became drier. Changes in the P variance were not related to variations in temperature. Instead, the largest changes in P variance were generally found in regions having the largest aerosol emissions. Our results combined with recent modelling studies lead us to speculate that aerosol loading has played a key role in changing the variability of P.