Subject: File No. S7-14-08
From: Steven Kilgore
Affiliation: Adjunct Professor of Economics

September 10, 2008

With regards to the proposed regulation of equity indexed annuities (EIAs) I would like to make the following points:

Marketing Issues: It is apparent that complaints received relating to annuities are significant. However, the proportion of these relating to equity indexed products seems unclear and has not been well documented, it would appear. Possibly EIAs would receive fewer complaints than those received by variable annuities given their comparative simplicity – with the major point of difficulty likely being the impact of surrender charges. The commission structure is also commonly mentioned as a negative but this is essentially double counting the surrender charge issue since commissions are effectively paid by insurers out of surrender charges in the case of early surrender, rather than directly out of consumer account balances. Commissions on annuities are only relevant from a consumer point of view if there is an opinion that agent advice is advice is skewed as a result. There would seem to be no data that indicate EIA commissions are out of line with those of other insurance products.

Risk Exposure: It has been noted the EIAs present substantial risk to the investor/policy owner. The risk is in three primary areas: (1) the market index used as a benchmark may perform unfavorably and therefore the annuity holder will only receive a minimum stated return, (2) the annuity holder will need or wish to withdraw their funds prior to maturity and therefore incur substantial surrender charges, and (3) risk of insurer default on the guarantees made of either a minimum return or the matching of a market indexs return.

(1) Risk of receiving only a minimum return: Some of the more well known formulas for measuring investment risk utilize standard deviation around a mean return or deviation from a benchmark rate – with possibly this deviation weighted by the degree of risk assumed. Several the incorporate the Capital Asset Pricing Model through use of Beta as part of the formula. These include the Jensen Index and the Treynor Index. Two others relate risk not to that of the market through use of Beta but instead include standard deviation of returns as compared to a risk free rate - the Sharpe ratio and the risk adjusted performance statistic (RAP). However, a more important measure of risk adjusted return for many purposes would seem to be downside risk rather than pure variation. The Sortino ratio therefore captures only variation of returns below a specific benchmark as its measure of risk. Note that EIAs (excluding the surrender charge or insurer default possibilities) would have no downside risk at least in nominal terms and that they commonly promise at least minimal return that could be considered to approximate the risk free rate of return that could be earned on short-term treasuries. Value at Risk (VaR) measures are used by many institutional investors and also commonly focus on downside risk rather than overall variability in returns. To the degree that the distribution of returns may be skewed towards one end or another of a probability distribution, it is important to focus on downside risk specifically. VaR is commonly reported as a key metric of investment risk exposure by investment banking firms.

Less directly within Economics, game theory can be applied to some degree in the consumer investment choice context. Von Neumanns Minimax theorem holds that an optimal result may be where the potential for loss is minimized – this being possibly consistent with a Nash equilibrium in case of a zero sum game – where gains only occur at the expense of others. The relevant point of these approaches is to support the concept that the minimization of downside risk would seem to be a very critical risk related metric in certain situations. In addition, if EIAs link upside returns only to well known market indices such as the SP 500, it might be supposed to that less investment information needs to be provided to consumers to insure they have a good understanding of the product. Most consumers will be able to track the supposed progress of their investment from independent general news sources such as even local evening news shows.

(2) Surrender charges can create a substantial loss in value to annuity holders but generally the presence of illiquidity or an indirect financial market connection has not in every case been defining as to whether a product is considered a security. The case of bank CDs is similar, for example, in that there can be substantial loss with early withdrawal. Money market accounts holdings do not generally have this illiquidity issue but instead have returns that may vary unpredictably after the point when the product is sold. These are often sold by banks as being tied in some way to broad market indices such as the federal funds rate. These are not, however, considered to be securities when offered in this form rather than as true money market funds. The greater complexity of insurance products as compared to consumer level banking products is support for the requirement of licensure of insurance agents but the precedents in banking might suggest that the treatment of the product as a security is another matter when the link is only indirect to securities markets. As for EIAs, both CDs and money market accounts have no material downside risk, in nominal terms, beyond the potential for default of the issuing institution.

(3) Risk of insurer default – traditionally insurer default has been regulated by the states and insurer promises are generally backed by a guarantee fund - up to set dollar limits. Defaults among companies that primarily write annuities have not been common in recent years and various risk management strategies are available to the insurers to either assume or transfer the risk for the guarantees being made. The guarantee fund system has worked well to protect consumers in these instances where defaults have occurred.

In summary, there would seem to be a sound theoretical, precedential and practical basis for defining investment risk primarily in terms of downside risk rather than overall variation. In addition, there are only minor indications mentioned in published reports that equity indexed annuities might be more difficult for consumers to understand than standard fixed annuities or other similar products that are not considered to be securities – such as those some banking products. Problems that may exist with inappropriate sales practices would seem to be of the type that could be effectively managed within the current regulatory framework.