Putting aside for a moment the arguments relating to the definition of green consumption, previous research into green consumerism has been dominated by rural sociologists and geographers. Clearly not all consumers are willing to buy grenn products. Some consumers may be reluctant to purchase green products. In other words, who would prefer to buy green products? When and why? Firstly, we tested the moderation of temporal distance between green consumption values and relative purchase intention. The results showed that the overall model was very significant (R2=0.34, F=27.32, p<0.001). The main effect of time reference and green consumption values were significant. The interaction between time reference and green consumption values was significant (B=-0.51, p=0.02), the explanatory power of the three interactionsΔR2=0.03. Simple effect analysis revealed that, for the subjects with low level of green consumption values, the time reference had positive predictive effect on the purchase intention of green products (B=0.84, SE=0.22, t=3.74, p<0.00), but for the subjects with high level of green consumption values, the time reference had not predictive effect on the purchase intention of green products (B=0.01, SE=0.22, t=0.06, p>0.05).The results shows that the purchase intention of green consumers were more stable and less likely to be influenced by time reference, but non-green consumers tended to buy green products in the future rather than now. Secondly, we tested the moderation of temporal distance between green consumption values and perceived value. The results showed that the overall model was very significant (R2=0.19，F=12.48，p<0.001). The main effect of time reference and green consumption values were significant. The interaction was significant (B =-0.47, p=0.03), the explanatory power of interactionsΔR2=0.02. Simple effect analysis revealed that, for the subjects with low level of green consumption values, the time reference had predictive effect on the perceived value of green products (B =1.04, SE=0.24, t=4.26, p<0.001), but for the subjects with high level of green consumption values, the time reference had not perceived value on the purchase intention of green products (B=0.29, SE=0.24, t=1.21, p>0.05).The results shows that the perceived value of Green consumers were more stable and less likely to be influenced by time reference. A mediated moderation model was represented in the form of a statistical diagram in Figure 1. The total effect of the product of time reference and green consumption values was -0.51, The indirect effect of them was -0.07, This indirect effect was statistically different from zero according to a 95% bias-corrected bootstrap confidence interval (-0.198 to -0.0034). Finally, the direct effect of them was also statistically significant, B=0.44，p=0.03. So the effect of mediated moderation was partly. Conclusion: Green consumers were more stable and less likely to be influenced by time reference, but non-green consumers tended to buy green products in the future rather than now.