Integrity labels: HISTORICAL ANALYSIS / COMPARATIVE EVIDENCE / CURRENT FCM GOVERNING ALTERNATIVE
Evidence base: "Never Again a Head Tax" (fiscal-constitution essay, 2026-09-23): Liu Shougang 《财政中国三千年》 (highlighted passages, pp. 34, 117, 209–210, 216); ROC (Taiwan) fiscal and civil-service practice since the 1950s; comparative cases (UK, US, Meiji Japan, South Korea, New Zealand, Singapore).
Evidence grading: VERIFIED / PROBABLE / UNVERIFIED per source, the section convention. The historical and comparative analysis of bureaucratic scale is developed in the companion essay; this dossier states the current-policy question it leaves open.
Policy question
How many people does the mainland administrative state employ, what do they cost, and who decides whether the number is right? These are elementary questions for any fiscal constitution — and they are among the hardest to answer with public data. The companion essay establishes the historical pattern: imperial China repeatedly built bureaucracies that were loyalty-purchase machines rather than administrative instruments, from the Song 官多阙少 (more officials on the payroll than posts to fill, produced by the 阶官/职官 rank-post separation) to the systematic gap between authorized establishment （编制） and actual headcount （实有人数）. This dossier asks whether the present system has escaped that pattern, and what evidence would be needed to know.
What the historical record establishes
The essay's thesis is structural, not moral: dynasties did not fail at fiscal reform for lack of ideas but because three traps were never escaped — revenue pushed onto heads instead of capacity, a bureaucracy that purchased loyalty, and no fiscal constitution to bind either (see companion essay, §1). On staffing specifically, the Song case is the archetype: the "养官" (support-the-officials) policy plus rank/post separation produced a payroll that grew independently of administrative need. The comparative cases show the alternative is possible: ROC (Taiwan) post-1950s practice, the UK, US, Meiji Japan, South Korea, New Zealand, and Singapore each at some point bound administrative scale to defined functions and published establishments.
The essay grades its forward-looking numbers explicitly: figures for a future PRC transition are UNVERIFIED estimates — targets are given as benchmarked ratios and methods, not asserted facts. That discipline applies here a fortiori.
What is not publicly established
No current, comprehensive, independently auditable figure for the mainland administrative workforce's total scale and cost was located in the project corpus. Authorized-establishment figures, actual headcounts, and the boundary between civil servants, public-institution （事业单位） staff, and state-enterprise personnel are reported inconsistently across sources, and no single official series reconciles them. This dossier therefore does not assert a headcount. It records the gap as the finding: a state that cannot publish a reconciled establishment table cannot be held to a staffing policy, and the absence of the table is itself the accountability failure.
Related transparency gaps tracked elsewhere in this project: no budget-transparency / open-budget evidence packet exists in the corpus; the 2003 abolition of custody-and-repatriation after Sun Zhigang shows administrative systems can be dismantled under pressure, but it is the exception that marks the rule.
FCM governing alternative
A Third Republic should treat the establishment table as a founding fiscal document:
- Publish a reconciled establishment. Within the first year, release a unified table of authorized vs. actual personnel across civil service, public institutions, and Party organs performing state functions — the 编制/实有人数 gap closed by disclosure.
- Bind scale to function. Adopt the essay's benchmarked-ratio method: set staffing ceilings as ratios to population and to defined functional portfolios, published and legislated, not left to annual bargaining.
- Separate the Party payroll from the state payroll. Positions performing state functions are funded, counted, and audited as state positions; Party-organizational staffing is disclosed separately.
- Make the table permanent. Annual establishment reporting, independently auditable, as a standing fiscal-constitution requirement — so the Song pattern (payroll growing independent of need) becomes structurally impossible rather than merely criticized.
How to measure success
- Publication of a reconciled authorized-vs.-actual establishment table (binary: exists or not).
- Administrative personnel per 1,000 residents, by province, published annually.
- Personnel expenditure as a share of total fiscal expenditure, published in budget documents.
- Independent audit findings on establishment compliance, published.
Sources
- Companion essay: "Never Again a Head Tax: What Two Millennia of Chinese Fiscal Failure Teach a Future Free China" (2026-09-23) — staffing analysis §§2–3; evidence base: Liu Shougang 《财政中国三千年》; ROC (Taiwan) post-1950s practice; UK/US/Meiji Japan/South Korea/New Zealand/Singapore comparators.
- Corpus note: no current mainland staffing-statistics packet located; no open-budget evidence packet located (recorded gaps, not oversights).
Open questions this dossier will track
A reconciled establishment series if one is ever published; the 事业单位 boundary question; comparative staffing ratios for a transition government (UNVERIFIED until sourced).