Gerald W. Buetow

The Inconsistency of Return-Based Style Analysis

The authors demonstrate that the usual application of the return–based style analysis relies on commercially available indexes that exhibit extreme multicollinearity. The subsequent results are volatile and have little meaning. As a result, the authors argue that implementing return–based style analysis with commercially available indexes can result in accurate and inappropriate investment decision–making. Even with multicollinearity, however, they demonstrate that the results of the analysis can be meaningful as long as the explanatory variables properly capture the investment objective of the portfolio. The authors conclude that the only way to implement return–based analysis is to use portfolio–specific benchmarks that properly capture the investment objectives of the portfolio.

The Benefits of Rebalancing

This article analyzes the effects of various rebalancing decisions on the risk and return of a multi–asset class portfolio. The findings show that the daily monitoring of a portfolio coupled with interval rebalancing can add to performance net of costs while simultaneously controlling for risk. Implementing a daily monitoring and interval rebalancing system for the multi–asset class (and multi–managed) portfolio requires attentive professional management with the expertise to address structural obstacles in some asset classes.

An Empirical Analysis of Exchange-Traded Funds

Exchange-traded funds (ETFs) provide investors with a single, tradable security for gaining exposure to an index that presumably proxies for an asset class. In the U.S. alone, ETF assets under management have grown to nearly

Monetary policy and fixed income returns

900 billion. Buetow and Henderson analyze ETF returns to evaluate how closely ETF prices replicate exposures to their benchmark indices. The authors’ analysis highlights practical considerations that influence index replication. On average, an ETF closely tracks its benchmark index. An ETF that tracks an index composed of less liquid securities exhibits a larger tracking error, and an ETF that tracks an index composed of less liquid, or non-U.S., securities exhibits a greater correlation (relative to the benchmark index) with a U.S. equity index, suggesting that the diversification benefits of these ETFs are less than is implied by the return properties of the benchmark index.

Impact of Different Interest Rate Models on Bond Value Measures

Abstract This analysis extends the findings of previous researchers by examining the relationship between Federal Reserve monetary policy and long-term returns to various sectors of the U.S. corporate and governmental bond markets. The results of this analysis show that corporate bond market return patterns are strongly associated with Federal Reserve monetary policy periods. Specifically, bond market indexes exhibit higher returns and lower standard deviations of returns during expansive monetary policy environments. In fact, all of the corporate bond indexes analyzed in this study exhibited negative Sharpe ratios during restrictive monetary policy environments. This would indicate that investors would be better off by investing in T-bills rather than corporate bonds during restrictive monetary policy environments. At minimum, the results suggest that investors in the corporate bond market should closely monitor Federal Reserve monetary policy.

The trade-offs between alternative finite difference techniques used to price derivative securities

There is an extensive literature on ways of valuing bonds with embedded options and measuring their interest rate risk exposure. Price sensitivity measures-effective duration and effective convexity-are found by shocking rates up and down by the same number of basis points and determining the impact on the value of the security. Yet, despite the critical importance of the measures that are derived from a valuation model-option-adjusted spread, effective duration, and effective convexity-there appears to be no empirical study that examines whether interest rate models produce any significant differences for these measures. The authors examine a callable bond, a putable bond, a callable range note, and a putable range note using several popular interest rate models. The substantial differences in the measures estimated are attributed to characteristics of the interest rate model.

Monetary Policy and Interest Rate Factors

Approximating partial differential equations (PDEs) using the finite difference method (FDM) is a common occurrence in contingent claim asset pricing. Past studies have stated that an explicit scheme is being used when a mixed (a combination of an explicit and implicit) scheme is actually being implemented. This study explores the financial and mathematical consequences of using this commonly found method with other FDM approaches. The consequences on valuation and risk management are also briefly addressed.

Mutual Fund Asset Allocation and Federal Reserve Monetary Policy

Previous research establishes the stylized fact that bond returns differ across monetary policy regimes. Specifically, bond returns are higher (lower) and exhibit lower (higher) standard deviations during expan-sive (restrictive) policy periods. Using the three factors known to explain bond returns, we find that the level and slope factors drive the pattern documented in the literature. We find, however, that this pattern is not present in recent data. Our results suggest that either the linkage between monetary policy and bond returns changed or that the association documented in the extant literature arose spuriously.

The Dangers in Using Return Based Style Analysis in Asset Allocation

As individual investors have assumed more responsibility for their retirement fund decisions, the demand for easy-to-implement asset allocation strategies has grown. In this study, we develop such a strategy based on Federal Reserve monetary policy. We show that a straightforward tactical asset allocation strategy based on Fed policy would have added significant wealth over the 44-year period studied.

A Note on Common Interest Rate Risk Measures

The article begins with some background information and presenting the problem faced by the retail investor and the on-line advisor. The authors then formally set up the problem that must be solved in order to give the investor accurate advice. They suggest that, though their experience is predominately within the retirement plan arena, their framework is not limited to this area and argue that, in fact, much of their work may be even more important for taxable investments. They then offer a few examples of why strictly quantitative methodologies cannot be used blindly by on-line advisors when offering advice. They subsequently present a remedy to the drawbacks associated with the strictly quantitative approaches: a fundamental analysis focused on qualitative as well as quantitative factors. They argue that fundamental analysis establishes correct and stable style exposures of the funds being analyzed. They illustrate the potentially severe differences in outcomes depending upon which methodology is used in selecting individual funds.