In insurance, survival functions are never used for hedging; expected shortfalls are used instead.

The claim is absolute ('never'). True: insurance hedging, reinsurance and risk practice rely on expected shortfall or tail-expectation measures, and survival functions or exceedance probabilities are not used for hedging. Mostly true: expected-shortfall-type measures (expected losses beyond a threshold, such as in excess-of-loss pricing or Swiss Solvency Test tail value-at-risk) dominate, and survival-probability measures are used only marginally for hedging. Mixed: both kinds of measure are commonly used. Mostly false: the claim overstates a real pattern, for example where expected-loss measures are important but exceedance probabilities, return periods or value-at-risk quantiles (such as the Solvency II 99.5% VaR standard or catastrophe exceedance-probability curves) are also in standard use. False: survival-probability or quantile measures are the main basis and expected shortfall is rarely used.

Set before the research began.

1. E1 IAA Risk Book Catastrophe Risk International Actuarial Association, Insurance Regulation Committee · October 2025

   “In such circumstances, this point estimate, considered by the Solvency II prudential regime, is also referred to as the 99.5% VaR measure. These numbers are used to set capital requirements and determine how much reinsurance to buy, and for formal risk tolerance statements”

2. E2 Technical document on the Swiss Solvency Test Federal Office of Private Insurance (FOPI), Switzerland · 2 October 2006

   “The risk measure used in the SST is the expected shortfall or the TailVaR.”

3. E3 Directive 2009/138/EC of the European Parliament and of the Council of 25 November 2009 on the taking-up and pursuit of the business of Insurance and Reinsurance (Solvency II) (recast) (Text with EEA relevance) legislation.gov.uk (text of EU Directive 2009/138/EC) · 25 November 2009

   “Each of the risk modules referred to in paragraph 1 shall be calibrated using a Value-at-Risk measure, with a 99,5 % confidence level, over a one-year period.”

4. E4 IAA Risk Book Catastrophe Risk International Actuarial Association, Insurance Regulation Committee · October 2025

   “Rather than using the PML, a “value-at-risk” (VaR) measure, some insurers are moving to metrics capturing more of the EP curve, such as “tail-value-at risk” (TVaR), for their marginal impact decisions.”

E2 shows that the Swiss Solvency Test uses expected shortfall (TailVaR), which fits the expected-shortfall part of the claim. E3 shows that Solvency II calibrates its risk modules with a 99.5% Value-at-Risk measure, which is a quantile measure, not an expected shortfall. E1 says that 99.5% VaR figures are used to set capital and to decide how much reinsurance to buy, which is a hedging decision. E4 describes only some insurers moving from PML/VaR toward TVaR. So expected-shortfall measures are in real use, but quantile and exceedance-probability measures were also standard, and the absolute "never" overstates the pattern.