In his response, he hypothesizes that the "price" for insurance offered by multi-company multi-line organizations would be lower, due to the ability of the group to "walk away" from an insolvent subsidiary. The implication is that default risk is higher for such companies. This runs completely counter to both general belief and past insolvency track records. For this reason, I skimmed through the paper referred to by Mr. Phillips.
The paper he referenced is 49 pages long. Due to time constraints, I did not read the entire paper. Instead, I skimmed the section where he discussed his empirical data tests and his definition of insurance price measures.
The paper defines price as the ratio of liabilities to premium. It was not readily apparent from the paper whether these numbers were before (gross of) reinsurance or after (net of) reinsurance, but I suspect it was net of reinsurance. As such, one problem with the analysis is the impact of company net reinsurance retentions.
(To explain, an insurance company typically protects itself from excessive volatility in insurance results through the purchase of reinsurance. The level of reinsurance is dependent on risk tolerance and the existing risk profile of the company's business before reinsurance. Companies with a greater internal spread, or diversification, of risk, such as is accomplished by writing multiple lines, will generally have a higher risk tolerance for an individual line. Hence they need less reinsurance.
The type of reinsurance may vary, between straight proportional to high level excess reinsurance. If a company has a high risk tolerance for an individual line, it will tend to buy high level excess reinsurance rather than low level excess or proportional reinsurance. As larger losses tend to have a much longer duration than the smaller losses, this means that the insurer will have a much longer tail, or duration for its liabilities, than a company with a lower risk tolerance by line (that cedes its long tail losses to others). This will lead to a deeper discount when converting its liabilities or losses from a nominal to an economic basis. Therefore, a company with a higher net retention will have a lower "price", or ratio of premium to net economic losses, strictly due to the level of losses it retains and NOT DUE TO ANY DEFAULT RISK.)
Another potential pitfall of the analysis referred to was the use of daily stock price volatility to measure insolvency risk. During the time in question, the oldest and most established multi-line companies were being punished by the stock market for perceived future earnings drags from Environmental and Asbestos. As subsequent history has born out, the biggest danger due to this problem was earnings, not insolvency. The stock markets are much more focused on earnings than insolvency. Therefore, stock price volatility may not be a good measure of insolvency risk.
A third problem alluded to above is the high correlation during the time-period in question between multi-company groups and Environmental & Asbestos problems. This factor may overwhelm any other correlation being tested during the period in question between multi-line multi-company groups and any other factor.
Lastly, I would greatly appreciate it if future discussions on this forum could be done in a concise manner. Referencing a 49 page paper, augmented by a multi-sentence single paragraph technical abstract, is not conducive to effective communication. Many people on this Net are business people, or otherwise busy people who cannot afford to read long technical papers. As such, the tendency is either to rely on the conclusion without testing or to ignore the work entirely. Neither is desirable. Therefore, while I greatly appreciate an extended and diverse discussion from all areas, I plead with all those entering into the discussion to remember the entire audience and its needs. Thank you.
Visit the CAS Web Site at http://www.casact.org
===============================================
To subscribe or unsubscribe from CASNET:
Send an e-mail to caslists@lists.casact.org
Type in the body join casnet to subscribe
or leave casnet to unsubscribe.