Repricing a loss-making age band in individual PMI
You are the chief actuary of a UK private medical insurance (PMI) provider. The individual book has 18,500 policyholders, mostly aged 45-70, and is community-rated in five-year age bands. In 2024 the book loss ratio rose from 68% to 79%. The 55-59 band is loss-making at 112% and is cross-subsidised by the 30-44 bands. Two developments drive the deterioration: oncology claims are up 22% in frequency and severity, and a new hospital group contract raises tariffs by 9% from 1 January 2025. Medical inflation is 7%. The excess-of-loss reinsurer, covering £50,000 per person, has quoted a 15% rate increase at renewal, or alternatively a £75,000 attachment for a 5% increase. The largest group scheme, 2,100 members, renews on 1 April 2025 and requires 30 days' notice. The board also wants to grow the 30-44 segment by 30% over two years. Capital coverage is 165% of the regulator's requirement.
1. Recommend how to deal with the 55-59 band: full repricing, continued cross-subsidy, or withdrawal. Support your recommendation with the arithmetic and identify the regulatory and commercial risks.
Read the worked answer
Arithmetic. Before repricing, the band's expected 2025 loss ratio is roughly 112% x 1.09 (tariff) plus residual severity drift, say 122%. Hitting the book target of 75% requires a premium increase of 122/75 - 1 = 63%, which is not credible for a retail product with price elasticity around -1.5; a 63% rise could halve persistency and destroy more premium than it recovers.
Options. (i) Full repricing is arithmetically correct but commercially self-defeating. (ii) Continued cross-subsidy is defensible only if the subsidy is quantified, disclosed in fair-value assessments, and bounded, because the 30-44 target segment is precisely the group whose future claims will not materialise for 15-20 years, so it cannot fund the subsidy indefinitely. (iii) Withdrawal removes the loss but leaves a book concentrated in ages 60+, worsening future cross-subsidy and shrinking scale against fixed costs.
Preferred route. Reprice the band by around 25-30%, simultaneously restructure the product (raise the excess from £250 to £500, introduce a co-payment for outpatient diagnostics, tighten moratorium underwriting on oncology history), and cap the residual subsidy at a defined percentage of premium. That narrows the gap to target without a lapse spiral, and the product changes shift some oncology cost back to the policyholder rather than to the 30-44 cohort. Document the fairness evidence and the lapse monitoring trigger before renewal notices are issued.
2. Evaluate the reinsurance choice facing the provider, given the oncology trend and the desire to grow a new segment.
Read the worked answer
Cost comparison. Assume from the portfolio's claim distribution that roughly 1 in 400 policyholders exceeds £50,000 and 1 in 900 exceeds £75,000, with average excess claims of £85,000 and £100,000 respectively. On 18,500 lives: about 46 claims above £50,000, costing 46 x £35,000 = £1.61m, plus the 15% increase, so about £1.85m. At the higher attachment: about 21 claims above £75,000, costing 21 x £25,000 = £0.53m, plus 5%, so about £0.55m.
Retained risk. At £50,000 the insurer retains 46 x £50,000 = £2.3m of large-claim cost; at £75,000 it retains 21 x £75,000 = £1.58m but each large claim hurts more and the number of large claims is more volatile relative to the small book. The difference in expected reinsurance cost (£1.3m) is real, but so is the capital relief: with capital coverage already at 165% and a 30% growth plan in a younger, lower-claim segment, the marginal capital strain of a single £300,000 oncology claim is more affordable under the lower attachment.
Recommendation. Given rising oncology severity, a small numerator and a growth plan that consumes capital, take the £50,000 attachment with the 15% increase, but negotiate a claims-limit reinstatement and a profit-share so the cost is partly recovered if experience is good. The higher attachment only becomes attractive once the 30-44 block is large enough to diversify large-claim volatility.