Morningstar Advisor - February/March 2009 - (Page 34) Spotlight A Failure to Gauge Risk By Arijit Dutta The financial crisis has burned shareholders of funds once considered safe. Why didn’t they see it coming? Risk has taken center stage again. Markets were exceptionally calm for several years before 2007, and this calmness dulled most market participants’ perception of risk and led many to invest aggressively. It’s been payback time since then. Volatility returned with a vengeance after the subprime meltdown and the accompanying financial crisis, and it continues to dog just about all assets. The market’s turmoil has sparked much debate and soul-searching about risk-management techniques used by financial institutions and money managers. Indeed, we have seen several prominent cases of funds that suffered body blows to their previously solid reputation in risk management. For example, Dodge & Cox Stock DODGX and Dodge & Cox Balanced DODBX both had a great record during the 2000-2002 bear market but have paid a hefty price this time around for misjudging the downside in their financials holdings. But what’s really been shocking in this downturn is the extent of losses in investments that were considered downright safe or that supposedly had a tight leash on risk. Before the crisis, many investors believed that money-market funds were about the safest place they could put their cash, outside of a savings account in a bank. That assumption was shattered in September after one of the country’s oldest and largest money-market funds, the Reserve Primary Fund, “broke the buck” when its NAV fell to $0.97. A risky stake in Lehman Brothers bonds and a run by shareholders to redeem led to the fund’s demise. Shareholders are still waiting to get their balances out of the liquidating fund, and it’s unclear how much of their investments will be returned to them. The ultrashort-bond category, previously regarded as a safe-haven and cash-substitute investment, produced some severe blowups. SSgA Yield Plus (which liquidated in June) and Fidelity Ultra-Short Bond FUSFX, for example, suffered sharp losses during the past year and a half on a scale that would’ve been unthinkable based on their own or the category’s past record. Each fund suffered security markdowns in their subprime and mortgage-related holdings and was forced to sell holdings in a declining market to meet shareholder redemptions. That same toxic mix occurred at Regions Morgan Keegan Select High Income MKHIX, a high-yield bond fund. True, the fund is in a risky category that is not foreign to sharp losses, but veteran manager Jim Kelsoe had a stellar record in controlling downside. Over the past 12 months through Dec. 8, however, the fund posted a staggering loss—almost 80%. It’s striking how ineffective the risk-control systems of these money managers and, indeed, those of most financial institutions were in the liquidity crunch. Shareholders of these funds didn’t get much help, either: The most widely used risk measures did not flash red for investors leading up to the bust. Using the funds mentioned earlier as examples, we will look at three such measures—standard deviation, kurtosis, and Morningstar Risk—and assess how they did in foretelling the funds’ troubles. We’ll also compare how strongly these measures have factored in the downside risk since markets tumbled. A Quick Primer Standard deviation is a measure that captures how variable the return of a security is around its average over a time period. The bigger a security’s standard deviation, the more “risky” it is in the sense that the fund has exhibited large deviations from its average return. But standard deviation doesn’t paint a clear picture of the possibility of extreme returns— the “outliers” in a returns distribution. Kurtosis, which is another measure that captures how deviant a security’s returns have been from its average, is better than standard deviation at indicating whether a returns distribution has extreme outliers. A positive value of kurtosis 34 Morningstar Advisor February/March 2009
For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.