VHIS Premium Index
Tracking premium levels across 579 Voluntary Health Insurance Scheme plans and 20 age × gender profiles in Hong Kong. Based at 2019 (= 100).
2026 Q2
Premiums were broadly flat in 2019–2021, then accelerated after 2024 — over 70% of the seven-year rise is concentrated in the last two years.
PREMIUM-RISE RANKING
Which insurers and plan series raised premiums the most?
The same premium-index method, broken down to every insurer and every plan series. The left table is ranked by each insurer's annualized rise; the right table by each plan series' annualized rise. Both use the same measure, so they are directly comparable. All figures are computed from the Insurance Authority's published premium tables; click through to see the year-by-year repricing record for any insurer or series.
- 1+5.5%/yr
- 2+5.5%/yr
- 3+4.7%/yr
- 4+4.5%/yr
- 5+4.3%/yr
- 6+4.0%/yr
- 7+3.9%/yr
- 8+3.3%/yr
- 9+2.9%/yr
- 10+2.7%/yr
- 1+9.7%/yrannualized
- 2+8.0%/yrannualized
- 3+7.9%/yrannualized
- 4+7.7%/yrannualized
- 5+7.6%/yrannualized
- 6+7.4%/yrannualized
- 7+7.2%/yrannualized
- 8+6.9%/yrannualized
- 9+6.9%/yrannualized
- 10+6.7%/yrannualized
Note: the plan-series ranking collapses each series' deductible / additional-medical-benefit / co-insurance variants into a single row, taking the highest annualized rise among them. Both tables are sorted by annualized rise (%/yr), so they are directly comparable. Starting price and coverage value matter just as much — the rise is only one data point. Bar length is the relative proportion within each table; the depth of colour indicates how high the rise is.
METHODOLOGY
How is the Premium Index calculated?
Data-integrity note · the same “matched-pairs” method as the government CPI
本指數用政府編製消費物價指數（CPI）的同一套「配對比較」方法：等權（每個計劃一票）、以幾何平均（Jevons）鏈式接駁、涵蓋 20 組年齡×性別（0/10/…/90 歲 × 男女）、每季更新、以 2019 年第 2 季 ＝ 100 為基準。新計劃的處理：計劃的推出價格「不會」被當作通脹計算——計劃要到有上一期可作比較（即第二次觀測）時才接駁入指數，此後貢獻其自身的保費變化（若無調價則為 0%），與官方 CPI 採用的配對比較完全一致。性別軸轉系列亦一併納入：每條新的男／女保費腿以其之前的「不分性別」舊價作配對基準，並保留「軸轉」標記以作提示。美元計劃一併納入：同幣值保費比率不受匯率影響（尺度不變）。
Representativeness note · market “stickiness”
在 579 個追蹤計劃中，約 12.8%（74 個）自推出以來從未調整過保費——反映市場相當「黏性」，升幅主要由曾經調價的計劃帶動。
Limitations of this method
- 1.
Equal-weighted, not weighted by number of policyholders: Each plan carries the same weight, reflecting the rise of “the average plan” rather than “the average policyholder's” experience; a plan bought by many people is not counted more heavily.
- 2.
Premium only, not adjusted for coverage changes: The index tracks premium rises but is not adjusted for changes in the coverage itself. If a premium rises while coverage is simultaneously expanded or trimmed, the index reflects only the premium portion.
- 3.
A market average, not your plan: This is a whole-market indicator; the actual rise on your own plan can differ substantially — use the search / per-insurer lookup above to check yours.
- 4.
Data source and updates: Based on the Insurance Authority's publicly archived premium tables (vhis.gov.hk). The index is updated quarterly; the most recent quarter's figure may be revised slightly as the industry gradually archives its premium tables, while already-published historical figures stay fixed.
- 5.
Handling of new plans: A new plan's launch price is not counted as inflation (following the “matched-pairs” approach used by the Consumer Price Index — compiled by the Census and Statistics Department); a plan only links into the index once it has a second observation.
Aggregation: For each plan, arithmetic-average its 20 profile price-relatives (one plan = one vote); then geometric-mean (Jevons) across plans; chain-link the quarterly steps; rebase 2019 Q2 = 100. Premium “as-of value” rule: Each plan's premium as-of a quarter-end = its latest filed version on or before that date (carry-forward between reprices). Data source: the Insurance Authority's Voluntary Health Insurance Scheme website (vhis.gov.hk), published premium tables.