John Major, Senior Vice President, Instrat
Guy Carpenter’s Firm-Value Risk Modeling (FVRM) approach, described in two previous GCCapitalIdeas.com articles, takes Dynamic Financial Analysis (DFA) a step beyond existing techniques by modeling the impact of risk on the shareholder value of the (re)insurer. This puts the two dimensions of DFA - risk and reward - on the same scale: value. The output of an FVRM analysis is the M-curve, that relates the (re)insurer’s book value (surplus) to its market (shareholder) value. When the (re)insurer is in financial distress, with insufficient surplus to cover its risks adequately, its market value reflects the possibility of imminent liquidation, and will typically be not much greater than book value. On the other hand, when the (re)insurer holds a generous capital buffer, its market value reflects a going-concern potential to generate a stream of profits and dividends, and will include some franchise value above and beyond its book value. There is a point, however, where the (re)insurer has so much capital that adding more does nothing to enhance its franchise value.
The shape of the M-curve has implications for raising capital. If, by adding an increment of book value through an equity issue, the (re)insurer can raise its market value enough to cover the transaction costs of raising that capital, then it should do so. This applies strongly in the transition area where financial distress can be alleviated by an infusion of surplus. When the (re)insurer’s book value is such that a relatively small increase in surplus will put it in a decisively better position to generate sustainable profits, its market value will increase disproportionately. Investors will see this opportunity for capital gains and bid up the share price.
On the other hand, raising capital when the firm is awash with it does nothing to enhance franchise value, and investors will shy away from such an issue, depressing the share price.
These observations apply where the true state of the (re)insurer’s finances are understood by the capital markets, but, the insurance sector is notoriously opaque largely because of the status of reserves. And because reserves are liabilities, uncertainty in the level of reserves translates directly into uncertainty about the level of surplus the (re)insurer is holding.
This uncertainty would be no different in principle than the uncertainty as to whether a devastating hurricane or earthquake would rock the (re)insurer’s balance sheet next year if not for the phenomenon of information asymmetry. Investors are likely to view reserve estimates as something that management insiders understand much better than the general public ever could. The current market value of the (re)insurer reflects the market’s consensus about the adequacy of reserves (among other things), which may differ from management’s superior understanding - or at least, investors are entitled to suspect this may be the case.
When there is information asymmetry, an announcement of a new equity issue provides information to the marketplace: information that the current stock price is sufficiently high as to make an equity issue advantageous to the (re)insurer. This tends to cause investors to reassess and (more likely than not) lower the share price, effectively introducing another cost of raising capital. If the price goes down too much, then the equity issue may no longer be economically viable. Management needs to anticipate the drop in share price and only plan to issue if that drop is acceptable. Investors, in turn, understand that an issue announcement will only occur when their reaction is anticipated by management to result in a share price still sufficiently high. This provides even more information to them, and leads to an even bigger downward adjustment which in turn needs to be anticipated by management. This chicken-and-egg problem would seem to lead to an infinite regress, but what is known as an equilibrium solution can be solved mathematically. The goal of the research was to calculate the probability of success, and effective cost, of raising capital via seasoned equity offering.
Based on an FVRM model with a plausible M-curve and parameters representing the U.S. property and casualty insurance industry as a whole as if it were a single firm, effective costs peaking in the range of 100 percent to 400 percent were obtained as well as probabilities in excess of 10 percent - and sometimes quite higher - that issue will not succeed. In states of high surplus, information effects drive the probability of successful issue to nearly zero; in states of covert insolvency (negative book value that has not yet been recognized by regulatory authority), however, there is a significant probability - 30 percent to 50 percent - of a successful issue that transfers wealth from new to existing shareholders.
For more detailed information on the research findings, contact John A. Major.