Purpose The aim of the study was to evaluate the cost-effectiveness of capecitabine plus bevacizumab compared with capecitabine alone in elderly patients with metastatic colorectal cancer (CRC) from a Chinese societal perspective. Methods A decision-analytic Markov model was conducted to simulate the process of metastatic CRC. Three distinct health states: progression-free survival (PFS), progressive disease and death were included. Clinical data were derived from the AVEX trial. Health effectiveness was denoted in quality-adjusted life years (QALYs) and health utilities were derived from previously published studies. Incremental cost-effectiveness ratio (ICER) was regarded as the primary endpoint and willingness-to-pay (WTP) threshold was set at $26,753.37/QALY (3 × per capita GDP of China, 2017). One-way sensitivity analyses and probabilistic sensitivity analysis were also performed to explore the parameters uncertainty in the study. Results Over a 10-year life horizon, capecitabine plus bevacizumab gained 1.14 QALYs at an average cost of $21,609.48, while the effectiveness and cost of capecitabine group were 0.99 QALYs and $7274.83, respectively. The ICER between the two groups was $95,564.33/QALY. Parameters that mostly influenced the results of the model were utility of PFS state, duration of PFS state for capecitabine plus bevacizumab, total cost of PFS state for capecitabine plus bevacizumab and price of bevacizumab. The probabilities of capecitabine plus bevacizumab and capecitabine as the dominant option were 0% and 100% at the WTP threshold of $26,753.37/QALY. Conclusions The results of the study showed that capecitabine plus bevacizumab is unlikely to be a cost-effective treatment option for elderly patients with metastatic CRC.