The incidence rates of cancers in men differ by countries of the world. We compared the incidence rates of three of the most common cancers (prostate, lung, and colon) in men residing in 164 different countries with the population-weighted light at night (LAN) exposure and with several developmental and environmental indicators, including per capita income, percent urban population, and electricity consumption. The estimate of per capita LAN exposure was a novel aspect of this study. Both ordinary least squares (OLS) and spatial error (SE) regression models were used in the analysis. We found a significant positive association between population exposure to LAN and incidence rates of prostate cancer, but no such association with lung cancer or colon cancer. The prostate cancer result is consistent with a biological theory and a limited number of previous studies of circadian disruption and risk. The LAN-prostate cancer connection is postulated to be due to suppression of melatonin and/or disruption of clock gene function. An analysis holding other variables at average values across the 164 countries yielded a risk of prostate cancer in the highest LAN-exposed countries 110% higher than in the lowest LAN exposed countries. This observed association is a necessary condition for a potentially large effect of LAN on risk of prostate cancer. However, it is not sufficient due to potential confounding by factors that increase the risk of prostate cancer and are also associated with LAN among the studied countries.