Smart Beta Research Articles

Do smart beta ETFs deliver persistent performance?

This paper analyses smart beta ETF performance and provides the first evidence on the funds’ performance persistence. Our sample is comprised of 152 US equity smart beta ETFs over the period June 2000 to May 2017. We found that as per the risk-adjusted performance, about 40% of smart beta ETFs outperformed their related traditional ETFs after expenses. The analysis of performance persistence conducted based on the relative performance of smart beta ETFs showed that the performance of winners and losers does persist in the year ahead. The persistence in performance was documented in seven out of nine peer categories.

A novel dynamic asset allocation system using Feature Saliency Hidden Markov models for smart beta investing

The financial crisis of 2008 generated interest in more transparent, rules-based strategies for portfolio construction, with smart beta strategies emerging as a trend among institutional investors. Whilst they perform well in the long run, these strategies often suffer from severe short-term drawdown (peak-to-trough decline) with fluctuating performance across cycles. To manage short term risk (cyclicality and underperformance), we build a dynamic asset allocation system using Hidden Markov Models (HMMs). We use a variety of portfolio construction techniques to test our smart beta strategies and the resulting portfolios show an improvement in risk-adjusted returns, especially on more return-oriented portfolios (up to 50% of return in excess of market adjusted by relative risk annually). In addition, we propose a novel smart beta allocation system based on the Feature Saliency HMM (FSHMM) algorithm that performs feature selection simultaneously with the training of the HMM, to improve regime identification. We evaluate our systematic trading system with real life assets using MSCI indices; further, the results (up to 60% of return in excess of market adjusted by relative risk annually) show model performance improvement with respect to portfolios built using full feature HMMs.

Improving the Robustness of Trading Strategy Backtesting with Boltzmann Machines and Generative Adversarial Networks

This article explores the use of machine learning models to build a market generator. The underlying idea is to simulate artificial multi-dimensional financial time series, whose statistical properties are the same as those observed in the financial markets. In particular, these synthetic data must preserve the probability distribution of asset returns, the stochastic dependence between the different assets and the autocorrelation across time. The article proposes then a new approach for estimating the probability distribution of backtest statistics. The final objective is to develop a framework for improving the risk management of quantitative investment strategies, in particular in the space of smart beta, factor investing and alternative risk premia.

The Smart Beta Mirage

We document and explain the sharp performance deterioration of smart beta in- dexes after the corresponding smart beta ETFs are launched for investment. While smart beta claims to deliver excess returns through factor exposures, the market-adjusted return of smart beta indexes drops from about 3% “on paper” before ETF listings to about −0.50%~−1% after ETF listings. This performance decline cannot be explained by variation in factor premia, strategic timing, or diminishing returns to scale. Instead, we find strong evidence of data overexploitation in constructing smart beta indexes, which helps ETFs attract flows as investors respond positively to stellar backtests.

The macroeconomic drivers in hedge fund beta management

We investigate how macroeconomic indicators alter the dynamic risk exposure of different hedge fund style strategies. We implement a multifactor model to estimate the unobservable time-varying risk exposure conditional on macroeconomic information and a VAR to measure the impact of macroeconomic predictors on different time horizons. Using monthly returns on a cross-section of 10 different style indices from February 1997 to August 2019, we find that, on average, macroeconomic indicators explain approximately 30%, 55%, and 75% of the variability of betas at 1-, 6-, and 36-month horizons, respectively. Although macroeconomic predictors play a critical role at every horizon, at 1 month, the dominating effect comes from idiosyncratic shocks, which indicates that in the short run, hedge fund managers rely mostly on their own reallocation signals. Moreover, consistent with the fundamental drivers of the smart beta factors, we find that the interest rate level and GDP growth similarly impact hedge fund exposures across styles.

Do Smart Beta ETFs Deliver Persistent Performance?

This paper analyses Smart Beta ETF performance and provides the first evidence on the funds’ performance persistence. Our sample is comprised of 152 US equity smart beta ETFs over the period June 2000 to May 2017. We found that as per the risk-adjusted performance about 40% of Smart Beta ETFs outperformed their related traditional ETFs after expenses. The analysis of performance persistence conducted based on the relative performance of Smart Beta ETFs showed that the performance of winners and losers does persists in the year ahead. The persistence in performance was documented in 7 out of 9 peer categories.

Stock Characteristics and Stock Returns: A Skeptic’s Look at the Cross Section of Expected Returns

The correlation between stock characteristics and the cross section of stock returns plays a central role in empirical implementations of modern asset pricing models and has important implications for investment management. This remains true whether the correlation is due to investor preferences regarding the characteristics directly or whether the characteristics are proxies for state variables, the risk of which investors are attempting to hedge. This article asks what we know about the relation between these characteristics and the cross section of returns. The skeptic’s answer is, not much. A combination of lack of persistence in the characteristics and problems caused by model uncertainty, data snooping, and nonstationarity means that our knowledge is sketchy at best. Investors should be forewarned when considering any strategies such as smart beta that are premised on the correlation between characteristics and the cross section of returns. <b>TOPICS:</b>Portfolio management/multiasset allocation, performance measurement <b>Key Findings</b> • Many investment strategies, including smart beta, are premised on the existence of predictable correlations between stock characteristics and expected returns. • The skeptic’s view is that those correlations are not predictable owing to a combination of lack of persistence, model uncertainty, data snooping, and nonstationarity. • Empirical results and a review of the published literature offer support for the skeptic’s view and serves as a warning for investors.

Maximizing an equity portfolio excess growth rate: a new form of smart beta strategy?

It has been claimed that, for dynamic investment strategies, the simple act of rebalancing a portfolio can be a source of additional performance, sometimes referred to as the volatility pumping effect or the diversification bonus because volatility and diversification turn out to be key drivers of the portfolio performance. Stochastic portfolio theory suggests that the portfolio excess growth rate, defined as the difference between the portfolio expected growth rate and the weighted-average expected growth rate of the assets in the portfolio, is an important component of this additional performance (see Fernholz [Stochastic Portfolio Theory, 2002 (Springer)]). In this context, one might wonder whether maximizing a portfolio excess growth rate would lead to an improvement in the portfolio performance or risk-adjusted performance. This paper provides a thorough empirical analysis of the maximization of an equity portfolio excess growth rate in a portfolio construction context for individual stocks. In out-of-sample empirical tests conducted on individual stocks from 4 different regions (US, UK, Eurozone and Japan), we find that portfolios that maximize the excess growth rate are characterized by a strong negative exposure to the low volatility factor and a higher than 1 exposure to the market factor, implying that such portfolios are attractive alternatives to competing smart portfolios in markets where the low volatility anomaly does not hold (e.g. in the UK, or in rising interest rate scenarios) or in bull market environments.

Factor tracking: A new smart beta strategy that outperforms naïve diversification

This paper proposes a factor tracking strategy that minimizes the variance of the difference in returns between a target portfolio and return-driving factors. We establish and solve single- and multi-factor tracking models, and analyze the properties of the optimal portfolios. We find that the optimal factor tracking portfolios consist of factor mimicking portfolios and the minimum-variance portfolio determined in the traditional mean-variance model. Moreover, we derive the condition under which the multi-factor tracking portfolio is a linear convex combination of the corresponding single-factor tracking portfolios. Using various datasets from the US market, we show that the factor tracking portfolios outperform the equally-weighted portfolio, measured by both risk and Sharpe ratio. The superior performance of the factor tracking portfolios are attributed to these portfolios’ better upside participation and downside protection.

Practical Applications of Business Cycle–Related Timing of Alternative Risk Premia Strategies

Practical Applications In Business Cycle–Related Timing of Alternative Risk Premia Strategies, from the Spring 2020 issue of The Journal of Alternative Investments, Bernd Scherer and Matthias Apel relate how they developed a business cycle model to explore its predictive power in a tactical allocation framework. They created an aggregate of economic indicators that identifies economic regimes in order to determine the sensitivities of a host of risk premia strategies across various business cycle phases. Many of these sensitivities, measured as conditional Sharpe ratios, are statistically different from their unconditional, overall risk-adjusted returns. During each regime, a long–short tactical overlay strategy takes a long position in the four alternative risk premia strategies with the highest Sharpe ratios and shorts the four with the lowest. The authors show that the overlay strategy improves the performance of a risk-parity portfolio and provide evidence that the observed return patterns are not a beta-driven result. While their primary interest lies within a universe of investable alternative risk premia strategies, cross-validation tests also applied the model within three additional investment universes: global asset classes, smart betas, and Fama–French risk factors.

Is Factors Timing Overrated

In this paper we investigate two factor rotation approaches, performed directly on the MSCI smart beta indices that are Value, Quality, Momentum, Low Volatility and Size, over US and European Markets. Both approaches use the same indicators built on a macroeconomic signal (PMI), a market sentiment signal based on (VIX, credit spreads), and a momentum signal (time-series, cross-sectional). While the first approach is rule-based and mostly inspired by already known factor rotation frameworks, our work explores those by using our own specifications and it also seeks to check whether a style rotation works at the indices level. Our results show that our framework outperforms a simple equal-weight factor exposure in spite of application of transaction costs. On a stand-alone basis the PMI based rotation fol- lowed by the time-series momentum exhibit the strongest returns. Then we explore if machine learning techniques (tree-based) outperform equal-weight and the rule based strategies particularily after counting for transaction costs.

ESG Investing: From Sin Stocks to Smart Beta

Research on socially responsible investment in equity markets initially focused on sin stocks. Since then, the availability of data has been extended substantially and now covers environmental, social, and governance (ESG) criteria. Using ESG scores of firms belonging to the MSCI World universe, the authors measure the impact of score-based exclusion on both otherwise passive investment and smart beta strategies. They find that exclusion leads to improved scores of initially standard portfolios without deterioration of the risk-adjusted performance. Smart beta strategies exhibit a similar pattern, often in a more pronounced way. Moreover, the results demonstrate that exclusion also implies regional and sectoral tilts as well as (possibly undesirable) risk exposures of the portfolios. TOPICS:Portfolio theory, portfolio construction, ESG investing Key Findings • The authors show that environmental, social, and governance (ESG) screening can substantially improve ESG scores for both otherwise passive and smart beta portfolios without reducing risk-adjusted returns. • Starting from initially passive multicountry portfolios, ESG screening may lead to substantial regional tilts, such as overweighting Europe and underweighting the US and emerging countries or sectoral bets, for instance in favor of information technology and against financial and energy stocks. • Although the broad conclusion of improved ESG profile without affecting risk-adjusted performance also holds for smart beta portfolios, aggressive exclusion of ESG low-scoring firms may lead to some reduction in exposure to targeted factors.

Smart Beta: The Good, the Bad, and the Muddy

The authors describe the background of factor investing in its smart beta form and discuss why factor investing has become a popular investing style. They also discuss a number of reasons for skepticism regarding ex ante expected factor returns. TOPICS:Portfolio management/multi-asset allocation, factors, risk premia, style investing Key Findings • Factor investing is typically discussed using the language and machinery of efficient-markets models, yet investors are primarily expecting anomalous excess returns. • There is reason to be skeptical of claims that factor investing is delivering gross investment returns similar to that of traditional active managers but with lower fees. • For factors with risk-based explanations, even in the presence of significant factor premiums, the market portfolio is still likely to be the optimal portfolio for most investors.

Practical Applications of Trade-Off in Multifactor Smart Beta Investing: Factor Premium and Implementation Cost

<h3>Practical Applications Summary</h3> In <b>Trade-Off in Multifactor Smart Beta Investing: <i>Factor Premium and Implementation Cost</i></b>, from the 2019 Quantitative Special Issue of <b><i>The Journal of Portfolio Management</i>, Feifei Li</b> and <b>Joseph (Yoseop) Shim</b> (both of <b>Research Affiliates</b>) investigate the effects of implementation costs on multifactor portfolio construction. They examine the performance impact of rebalancing multifactor portfolios by looking at portfolio characteristics such as volume, tilt, turnover, and turnover concentration. They find that multifactor portfolios that included all six style factors under consideration—value, low beta, profitability, investment, momentum, and size—yielded the highest information ratios, both before and after trading costs, with a very small negative impact on Sharpe ratios compared with multifactor portfolios composed of fewer factors. The authors also determine that a uniform concentration level of 25% for all factors provides the greatest performance in the presence of implementation costs. Overall, they suggest investors pursue multifactor portfolios composed of the top 25% of stocks in each of the six style factors. <b>TOPICS:</b>Analysis of individual factors/risk premia, portfolio construction, VAR and use of alternative risk measures of trading risk

Stock Characteristics and Stock Returns: A Skeptic’s Look at the Cross-Section of Returns

The correlation between stock characteristics and the cross-section of stock returns plays a central role in empirical implementations of modern asset pricing models and has important implications for investment management. This remains true whether the correlation is due to investor preferences regarding the characteristics directly or whether the characteristics are proxies for state variables the risk of which investors are attempting to hedge. This paper asks what do we know about the relation between these characteristics and the cross-section of returns? The skeptic’s answer is not much. A combination of lack of persistence in the characteristics along with problems caused by model uncertainty, data snooping, and nonstationarity means that our knowledge is sketchy at best. Investors should be forewarned when considering any strategies such as smart beta that are premised on the correlation between characteristics and the cross-section of returns.

The Future of Investment Management

Investment management is in flux, arguably more than it has been in a long time. Active management is under pressure, with investors switching from active to index funds. New “smart beta” products offer low-cost exposures to many active ideas. Exchange-traded funds are proliferating. Markets and regulations have changed significantly over the past 10–20 years, and data and technology — which are increasingly important for investment management — are evolving even more rapidly. In the midst of this change, what can we say about the future of investment management? What ideas will influence its evolution? What types of products will flourish over the next 5–10 years? I use a long perspective to address these questions and analyze the modern intellectual history of investment management — the set of ideas that have influenced investment management up to now. One central theme that emerges is that investment management is becoming increasingly systematic. Our understanding of risk has evolved from a general aversion to losing money to a precisely defined statistic we can measure and forecast. Our understanding of expected returns has evolved as the necessary data have become more available, as our understanding of fundamental value has developed, and as we have come to understand the connection between return and risk and the relevance of human behavior to both. Data and technology have advanced in parallel to facilitate implementing better approaches. With an understanding of the ideas underlying investment management today, including several insights into active management, I discuss the many trends currently roiling the field. These trends, applied to the current state of investment management, suggest that investment management will evolve into three distinct branches — indexing, smart beta/factor investing, and pure alpha investing. Each branch will offer two styles of products: those that focus exclusively on returns and those that include goals beyond returns.

Is Alpha Minus Beta A Simple Rule for A Smart Beta Strategy?

Thematic investment portfolios have been of interest to portfolio managers. They can take on a variety of names including Smart Beta. One such model was suggested by Minami and Wakatsaki. They suggested an investment rule of Alpha Minus Beta as a good method to build a factor tilt portfolio which is efficient in an absolute risk-return space. This paper addresses their model by empirically examining market data. The S&amp;P 1500 was segmented into quintiles based on Alpha Minus Beta for a fifteen year period. The suggested model had value as the Coefficient of Variation was favorable compared to the index.

Optimal Glide Path Selection for Low-Volatility Assets

Target-date funds (TDFs) have become a popular choice for retirement plans, and the concept of a glide path is essential for TDFs. One of the key assumption behind glide paths is that the main component of a retirement plan should be growth assets invested in the overall equity market—for example, the S&P 500 Index. Researchers have recently challenged this assumption and argued for using smart beta factors for retirement purposes. Among them, the factor of low volatility may be uniquely suitable for retirement investing. This article studies the use of low-volatility assets for the purpose of retirement planning and the choice of ideal glide paths. This study is agnostic about the means by which the low-volatility risk/return profile is achieved and analyzes four series with diminished volatility constructed with different methodologies. The article concludes that low-volatility assets may indeed help retirement investors achieve their objectives, though such investors must choose ideal glide paths carefully to suit the selected low-volatility series. When equipped with proper low-volatility assets and carefully chosen glide paths, retirement plan managers may both improve the odds that their plans succeed and increase the expected final wealth levels. TOPICS:Mutual fund performance, retirement, other real assets, wealth management Key Findings • Both historical and de novo Monte-Carlo simulations confirm that low-volatility assets can increase the certainty of achieving investing objectives for retirement plans. • The three tested glidepath choices, hockey stick, curved, and straight line, have similar investment outcomes. • Investors with low-volatility assets may use more aggressive glidepaths to not only reduce failure rates but also increase expected final wealth levels.

How Smart Is the Real Estate Smart Beta? Evidence from Optimal Style Factor Strategies for REITs

This paper has a twofold objective. First, we contribute to the stream of literature that investigates whether traditional asset pricing factors show any predictive power for the cross-section of Real Estate Investment Trust (REIT) returns. In particular, we investigate the existence of a premium associated to the Value, Size, Momentum, Investment, and Profitability factors over the period 1993-2018. We find support for all the pricing factors but for the Profitability one. Second, we investigate whether a set of smart beta strategies, based on the combination of the identified factors, may outperform similar allocation techniques that do not exploit factors. We find that all the proposed factor-based strategies display a higher risk-adjusted out-of-sample performance than a simple buy-and-hold investment in the real estate market (proxied by the FTSE NAREIT All REITs Index). In addition, we find that when factor-based strategies are implemented, REIT-only portfolios display risk-adjusted performances comparable to those of diversified portfolios that include equity, bond, and commodities.

Practical Applications of Intelligent Selection of Smart Beta Mutual Funds

<h3>Practical Applications Summary</h3> In <b>Intelligent Selection of Smart Beta Mutual Funds</b>, which appeared in the Summer 2019 issue of <b><i>The Journal of Wealth Management</i>, C. Edward Chang</b> of <b>Missouri State University, Thomas M. Krueger</b> of <b>Texas A&amp;M University-Kingsville</b>, and <b>Cedric Tresor Mbanga</b> of <b>Missouri State University</b> investigated two research questions: 1) whether smart-beta mutual funds (SBFs) have outperformed the market in recent years and 2) whether investors can outperform their peers by limiting the money spent on SBF selection. The authors found that, in general, SBFs have not outperformed the market on a risk-adjusted basis; the increased risk offset the enhanced returns. However, they did find that investing in SBFs with lower expense ratios improved performance. Based on these findings, the authors suggest that investors limit the money spent on adding SBFs to their portfolios. <b>TOPICS:</b>Mutual fund performance, factor-based models, performance measurement