Reverse stress test
Exercise that first fixes the outcome, insolvency or the failure of the business model, and works back to the scenarios that lead there, instead of measuring the loss from a scenario chosen in advance.
A standard stress test picks a scenario and measures the loss. A reverse stress test starts from the other end: it fixes the outcome, insolvency or the inability to carry on the business model, and looks for the events that lead there. What comes out is a scenario, or a family of scenarios, and a question for management: is it plausible, and what do we do before we get there? In the UK, the FSA imposed it through policy statement PS09/20 of December 2009, which created chapter SYSC 20 of its Handbook. The requirement covered banks, building societies and most insurers from December 14, 2010, and management had to approve the results. For insurers, the PRA now ties the exercise to the ORSA in its supervisory statement SS19/16. The Solvency II Directive does not use the term. The expectation comes from EIOPA's ORSA guidelines (EIOPA-BoS-14/259, applicable since January 1, 2016), under which the ORSA policy states how and how often reverse stress tests are performed, and then from EIOPA supervisory statement EIOPA-BoS-17/097 of 2017, which calls for better quality in them. The breaking point is set in advance: a coverage ratio at 100% of the SCR, the MCR breached, a rating downgrade that triggers termination clauses, reinsurers pulling out. The difficulty is mathematical. An infinite number of shock combinations reach the same boundary, and the usual choice is the most likely one under the model, or a few narratives the board can read. What a reverse stress test mostly probes is the risk team's imagination. It forces a search for the concentration nobody saw: a single reinsurer on a key treaty, margin calls on interest rate hedges, as with UK pension funds in September 2022. On a cyber portfolio, it brings up the systemic event, an outage at a major cloud provider or a destructive worm, that the standard scenario would have sized too small.
Textbook case: an insurer holds 1,200 million euros of eligible own funds against an SCR of 800 million, a ratio of 1,200 / 800 = 150%. It holds 2,000 million of equities. Each point of decline in equity markets costs it 2,000 × 1% = 20 million of own funds and cuts its SCR by 4 million, because the equity charge applies to a smaller holding. The task is to find the fall s that brings the ratio to 100%: 1,200 − 20s = 800 − 4s, so 400 = 16s, and s = 25%. Check: own funds of 1,200 − 500 = 700 million, SCR of 800 − 100 = 700 million, ratio of 100%. The risk team now has to say whether a 25% fall in equities, alone or combined with wider spreads, is an event it considers possible over its plan horizon.
reverse stress test, reverse stress testing, reverse stress test ORSA, reverse stress test insurance, SYSC 20 reverse stress testing, what is a reverse stress test, business model failure scenario
Follow the research
One letter a week. First, the week's underwriting note, when one is published: a dated, argued decision, signed by Paul Bernardski. Then what the observatory published on cyber risk, political risk and insurance. No account, no tracking.
No spam, one-click unsubscribe, no data ever sold.