Chen, Ren-Raw Fabozzi, Frank J. Pan, Ging-Ging Sverdlove, Ronald

The major sources of credit risk are default probability and recovery. Together with interest rate risk, they determine the price of credit derivatives. In this article, we study the relative importance of these sources by testing pair-nested structural models with data from credit default swaps. By comparing pairs of models that are nested in terms of their assumptions about these sources (one model is a special case, or restriction, of the other), we are able to understand what causes a pricing model to succeed or fail. Using credit default swap data from 2/15/2000 through 4/8/2003, we find that random interest rates and random (asset value) recovery are important assumptions for achieving accurate pricing, while continuous default times (as opposed to a single default time) is not.

Fallen Angels: A Separate and Superior Asset Class

Fridson, Martin Sterling, Karen

From Benjamin Graham's day to the present, market participants have debated the relative investment merits of fallen angels and original issue high yield bonds. Analysis over the most recent complete cycle indicates that fallen angels have performed better, under both bull and bear market conditions. The edge is not attributable to the fallen angels' higher ratings or longer average maturities. By a commonly applied standard, fallen angels constitute a separate asset class capable of contributing diversification, as well as superior risk-adjusted returns, in a multi-asset portfolio.

Non-Idiosyncratic Alpha: A Case of the Corporate Bond Market

Kozhemiakin, Alexander

The definition of alpha as the difference between a portfolio's returns and those attributable to its expected beta to the market underpins a common belief of an almost axiomatic conviction that alpha and the market returns are completely unrelated. Yet, at least hypothetically, it is quite possible for alpha to be affected by the market returns if there are more opportunities to generate alpha in certain market environments. The article argues that this is indeed the case in the corporate bond market even if tactical beta timing is not allowed. In particular, when security selection is the only possible source of alpha, a consistently skillful corporate bond manager should generate higher alpha in the bear market. This is because cyclical downturns serve as a catalyst for greater credit re-rating which can be profitably exploited. Conversely, the alpha opportunity set consisting of all potentially profitable security selection trades shrinks when there is relatively less credit re-rating as the bull market becomes more entrenched. Unless the use of leverage is permitted, the counter-cyclical fluctuations in the alpha opportunity set support a switch from fixed to floating alpha targets for corporate bond portfolios.

Volatility Skew and the Valuation of Mortgages

Bhattacharjee, Ranjit Radak, Branislav Russell, Robert A.

Mortgage-backed securities contain embedded options and are thus exposed to the uncertainty of interest rates. The Black formula, used by practitioners to specify the volatility of rates, assumes the same yield volatility for all option strikes. But swaptions struck at rates below the at-the-money rate consistently trade at volatilities higher than those struck at rates above the at-the-money rate. This feature is called off-the-money (OTM) volatility skew. It is different from the at-the-money (ATM) volatility skew, which is evidenced by the fact that at-the-money swaptions trade at higher volatilities in a low-rate environment than in a high-rate environment. Volatility skew arises from assumptions regarding the distributions of interest rates. We discuss various term-structure models, their apprehension of volatility skew and the way they impact pricing, risk analysis and hedging of MBS.

Term Structure Slope Risk: Convexity Revisited

Hodges, Stewart D. Parekh, Naru

The article provides a new and practical approach to measuring the risk of changes in the slope, and curvature of the term structure, as well as its level. This is related to the principal components commonly found in term structure movements, but obviates the need to estimate them. The technique may be applied to construct hedges and to reconcile how value changes have stemmed from term structure movements. The article establishes how the new measures are related to the existing measures of duration and convexity, and to the (generalized) moments of the (present value weighted) distribution of cash flow dates.

Separability and Pension Optimization

Bazdarich, Michael J.

The Separation or Mutual Fund Theorem of finance theory is extended to the case surplus optimization, which is relevant for defined-benefit pension plans and others. While Separation theorems in the standard case have limited empirical applications, separation results in the surplus-optimization case are directly relevant to real-world pension management, because one of the "bases" of the surplus-efficient frontier is the full hedge of the pension plan's liabilities. Surplus-optimal or efficient portfolios can be expressed as combinations of the full liability hedge and exposure in the overlay with maximum Sharpe ratio. Optimal pension allocation is thus analogous to management of a portable alpha strategy with a "beta" equal to plan liabilities. Our results also indicate that the difference in portfolio weights between asset-optimal (standard) and surplus-optimal efficient frontiers is constant across those respective frontiers, so that the impact of moving from an asset-optimal to a surplus-optimal framework is independent of target returns or risk tolerance. Finally, we explore the details of the full liability-hedge portfolio. In a number of relevant cases, there is no presumption that a dollar of liabilities can be fully-hedged with a dollar of assets, which gives rise to the concept of an effective funding ratio for a pension plan.