Segregated funds are variable annuity products sold by Canadian Insurance companies. Essentially, these products allow investors to participate in the stock market upside, while providing a downside guarantee. Typically, these guarantees are sold as a "rider" on a mutual fund held within a pension plan.
Here is a quote from a paper we wrote in 2002 ( H. Windcliff, P.A. Forsyth, M.K. Le Roux, K.R. Vetzal, ``Understanding the behaviour and hedging of segregated funds offering the reset feature,'' North American Actuarial J., 6 (2002) 107-125.)
If one adopts the no-arbitrage perspective...in many cases these contracts appear to be significantly underpriced, in the sense that the current deferred fees being charged are insufficient to establish a dynamic hedge for providing the guarantee. This is particularly true for cases where the underlying asset has relatively high volatility. This finding might raise concerns at institutions writing such contracts.
We held a workshop on this topic in Toronto ( Workshop on Options in Financial Products, Fields Institute, December 8, 2000). The industry participants were skeptical about our results. We were told that there was no problem, since ``equity markets are never down over a ten year time frame.'', and that ``no-arbitrage pricing was inappropriate for insurance companies, since we take a long term view.''
What happened? As described in a Globe and Mail article (Report on Business, December 2, 2008, ``Manulife, in red, raises new equity,''), one of the large Canadian insurance companies, Manulife, posted a large mark-to-market writedown to account for losses associated with these segregated fund guarantees. From the Globe and Mail Streetwise Blog, November 7, 2008
Concerns that the market selloff will translate into massive future losses at Canada's largest insurer sent Manulife shares reeling last month. Those concerns were passed in part of Manulife's strategy of not fully hedging products such as annuities and segregated funds, which promise investors income no matter what markets do.
In other words, Manulife was able to generate profits for many years by collecting fees, and not hedging their exposure. This looked good for a while. But now, the shareholders get to experience the losses. What about the executives who decided that hedging was not required? They get to keep their bonuses.
So, were complex models to blame here? In this case, the models showed quite clearly that the insurance companies were taking on large risks, which they were not hedging. Why did they do this? Everybody else was selling these products at low prices, and making money. The temptation to post profits and not worry about long term risks was too great for risk managers to ignore.