We used band-recovery data from 2 populations of greater sage-grouse (Centrocercus urophasianus), one in Colorado, USA, and another in Nevada, USA, to examine the relationship between harvest rates and annual survival. We used a Seber parameterization to estimate parameters for both populations. We estimated the process correlation between reporting rate and annual survival using Markov chain Monte Carlo methods implemented in Program MARK. If hunting mortality is additive to other mortality factors, then the process correlation between reporting and survival rates will be negative. Annual survival estimates for adult and juvenile greater sage-grouse in Nevada were 0.42 ± 0.07 (x ¯ ± SE) for both age classes, whereas estimates of reporting rate were 0.15 ± 0.02 and 0.16 ± 0.03 for the 2 age classes, respectively. For Colorado, average reporting rates were 0.14 ± 0.016, 0.14 ± 0.010, 0.19 ± 0.014, and 0.18 ± 0.014 for adult females, adult males, juvenile females, and juvenile males, respectively. Corresponding mean annual survival estimates were 0.59 ± 0.01, 0.37 ± 0.03, 0.78 ± 0.01, and 0.64 ± 0.03. Estimated process correlation between logit-transformed reporting and survival rates for greater sage-grouse in Colorado was ρ  =  0.68 ± 0.26, whereas that for Nevada was ρ  =  0.04 ± 0.58. We found no support for an additive effect of harvest on survival in either population, although the Nevada study likely had low power. This finding will assist mangers in establishing harvest regulations and otherwise managing greater sage-grouse populations.