Using small sub-sample and sparse fault data from typical environmental test bases to assess the reliability of the electricity meter is the difficulty and hot spot of the current research. This paper proposes a Bayesian assessment method of electricity meter reliability by integrating the reliability estimation with the reliability prediction results of electricity meters in typical environment as a priori information, and combining with the test failure data. Firstly, based on IEC62380, the reliability of components under typical environment profile is expected, the reliability of electricity meter is expected based on the logical relationship between components and system reliability, and then, the a priori distribution of electricity meter life distribution parameters is constructed by adopting Monte-Carlo and pivot quantities, and the hyper-parameters are determined by adopting the method of moments equivalence to construct the a posteriori distribution of the life distribution parameters of the electricity meter. The Bayes point estimates and confidence limits of the reliability measure are calculated. The results of the study are of reference value for the verification of energy meter reliability.