Governing the Unsuppressible Tail: China’s 17-Year MRMP Epidemic and the 2024 National Resistance Consensus

China has hosted the world's largest and longest epidemic of macrolide-resistant Mycoplasma pneumoniae (MRMP) and, in 2024, issued what is to our knowledge the first national consensus document dedicated exclusively to its governance. We reconstructed this seventeen-year trajectory (2008–2024) and structurally compared three national systems. Genotypic resistance rose from ~70% (Beijing, 2008) to >90% (97.4% in Beijing, 2019), driven by the A2063G 23S rRNA mutation on the ST3/ST14 clonal pair. Zero-COVID then collapsed transmission by >90%, yet the resistant clones entered the trough entrenched and emerged unchanged, rebuilding dominance within months of the 2023 resurgence — an 'unsuppressible tail' of clonal persistence, not a fixed resistant proportion (§3.2). The comparators sharpen the lesson: Japan achieved a fall from peak values of 80–90% to 14.3%, alongside sustained stewardship and clonal turnover, while Taiwan extinguished circulation entirely yet saw clinical macrolide non-response return at 54.4% by 2024–2025 — suppression starves resistant clones; only governance — or clonal turnover — replaces them. The 2024 consensus resolved the paediatric tetracycline dilemma through a genotype- and age-gated algorithm, but left two gaps: no coupling to antibiotic consumption, and no enforcement mechanism. Comparison with Japan and the United States (macrolide-first empiricism, no MRMP surveillance) distils a transferable six-element playbook — sentinel resistance surveillance, genotype- and age-gated treatment, age-stratified second-line positioning, consumption coupling, diagnostic linkage, and iterative revision. Countries entering their own MRMP epidemics should run this sequence — measure, stratify, legislate, couple, iterate — before their own crisis arrives.