In a sample of 154 Aggregate Bond portfolios, 15% have greater Single B credits and outperform. Another third has a greater chance of being recommended to the investor. The former involves gaming, the latter, herding. An investment intermediary penalizes 'universe-gaming' but that leads to congregated recommendation for underperforming managers. This asymmetry in utility stresses performance when creditspreads narrow and recommendation when they widen. Larger fund size investors benefit in marginal utility. In down markets investors may select to hold on to 'out-recommended' underperformers and sell outperformers, worsening return instability.
Target date funds (TDFs), which seek to provide a broad demographic of individuals with a long-term retirement investment solution, play a major role in defined contribution (DC) retirement plans, and have also attracted substantial assets from investors in other types of accounts. Further, most DC plan investment menus designate TDFs as the default investment. This paper provides an overview of the strategies and structures of TDFs, as well as current trends among major TDF providers. We also document substantial heterogeneity across TDFs and provide guidance on how to evaluate the quality of TDFs in light of this heterogeneity.
Many economists believe that asset price bubbles do not exist or that, due to a joint hypothesis, they are difficult if not impossible to empirically validate. This paper dispels this belief by providing definitive proof that bubbles exist for a set of digital assets that have no cashflows and a zero liquidation value, but trade with positive market prices.
The classic estimates of CAPM equity betas are notoriously unstable. We assume that this is mainly due to changes in firms'leverage overtime. In order to take leverage into account, we propose a new approach where asset correlations among firms are pairwise constant, while equity correlations change over time as a function of the stochastic evolution of firms' asset values. The paper closes with a simulation that helps to show the model's features.
Fund rating systems can include quantitative and qualitative considerations, and can be either backward-looking "grade cards" or forward-performance forecasting systems (or both), with different investors using them for different purposes. This paper provides a detailed description of the most widely used publicly available rating systems today: the various ratings systems of Lipper and Morningstar. The more recently introduced Zacks fund rating system is also discussed in some depth. For each of these ratings systems, the predictive ability of their ratings is discussed. Finally, this paper offers some suggestions for future improvements that may lead to better predictions of future risk-adjusted fund returns.
The investment landscape has undergone a seismic shift with the advent of commission-free trading and low-cost ETFs, empowering retail investors but also introducing new challenges. While fees have significantly decreased, hidden costs related to increased risk have emerged. This article explores Warren Buffett's critique of "casino-like" investor behavior and introduces the "risk matters hypothesis" (RMH), acorollary to John Bogle's "cost matters hypothesis"(CMH). The RMH posits that the average risk of active portfolios exceeds that of the market portfolio, leading to a lower return-to-risk ratio (Sharpe ratio) for concentrated stock portfolios than diversified index funds. The allure of "free" investing has led many investors to engage in frequent trading, often at the expense of long-term returns. This paper argues that taking extra risk is akin to paying fees, both of which reduce the expected return-to-risk ratio of an investor's portfolio. By highlighting the importance of considering both explicit and implicit costs, the article cautions investors against the seductive appeal of "free" investing and emphasizes the need for a disciplined, risk-aware approach to achieve optimal investment outcomes. In Lake Wobegon, all the women are strong, all the men are good-looking, and all the children are above average. - Garrison Keillor
We show that traditional equity factor risk models tend to overstate factor risk for two basic reasons. First, the standard industry practice of using regression weights proportional to the square root of market capitalization leads to excessively noisy estimates of factor returns, which in turn inflates the factor variance estimates. Second, traditional techniques fail to remove the idiosyncratic component from the variance estimates of the pure factor portfolios, which further inflates factor risk. In this paper, we describe solutions to both problems. In addition to yielding more accurate risk forecasts and better decomposition of portfolio risk, we show that our approach produces more efficient optimized portfolios and mitigates spurious correlations between factor returns and idiosyncratic returns.
Corporate bond markets have grown into a core component of global financial systems and institutional portfolios, yet the academic literature has progressed unevenly across pricing, empirical return behavior, and portfolio construction. This paper provides a comprehensive survey of corporate bond research, spanning structural, reduced-form, and hybrid pricing models; empirical evidence on default, recovery, liquidity, and return predictability; and portfolio-oriented approaches used in both academic and practitioner settings. We identify a central gap in the literature: while pricing theory operates largely under the risk-neutral measure and empirical studies document return behavior under the physical measure, these strands have not been fully integrated into a unified portfolio framework. In the final part of the paper, we show how the Equivalent Expectation Measures and multiverse EEMs provide analytical tools for deriving finite-horizon expected returns and variance-covariance matrices of corporate bond returns under the physical measure, enabling direct application of classical mean-variance analysis to construct efficient frontiers and ex-ante Sharpe-ratio-maximizing portfolios for risky corporate bonds.
While machine learning has revolutionized many fields such as natural language processing (NLP) and computer vision, its impact on time-series forecasting is still widely disputed, especially in the finance domain. This paper compares forecasting performance on U.S. Treasury yield curve data across econometrics/time-series analysis, classical machine learning, and deep learning methods, using daily data over 47 years. The Treasury yield curve is important because it is widely used by every participant in the bond markets, which are larger than equity markets. We examine a variety of methods that have not been tested on yield curve forecasting, especially deep learning algorithms. The algorithms include the Autoregressive Integrated Moving Average (ARIMA) model and and multiple transformers built for forecasting. ARIMA and naive econometric models outperform other models overall, except in one time block. Of the machine learning methods, TimeGPT, LGBM and RNNs perform the best. Furthermore, the paper explores whether stationary or nonstationary data are more appropriate as input to deep learning models.
This paper demonstrates that portfolio performance can be substantially enhanced by simultaneously utilizing historical factor return volatilities and option-derived market volatilities to optimize factor exposures. The improvements are particularly pronounced in regimes where option-implied market returns exhibit high volatility and right-skewness. Further gains in risk-adjusted portfolio returns are achieved by estimating model parameters separately for different regimes. Qualitatively similar results are obtained when all parameters are estimated strictly out-of-sample. These findings are not limited to a specific set of factors; comparable enhancements are observed when employing principal components derived from a broad set of factors.
We propose a method for creating a sovereign securities portfolio that gradually reduces its carbon footprint, in line with the Paris Agreement. This allows passive investors to achieve net zero (NZ) targets while maintaining risk-adjusted returns similar to a business-as-usual benchmark. From 2015 to 2021, our approach would have cut carbon intensity by 34.7% with a 7.5% yearly target, compared to just an 8.5% reduction for the benchmark. Total emissions would have dropped by 27.5%, while they would have risen by 25.4% in the benchmark. Notably, NZ portfolios match the benchmark's financial performance and creditworthiness without significant foreign exchange risks.
We use large language models (LLMs) and natural language processing (NLP) to extract environmental, social and governance (ESG) insights from real-time news, creating an expert-annotated dataset to evaluate ESG classification, firm relevance, and sentiment. Our fine-tuned models outperform pre-trained ones in ESG detection, firm impact, and sentiment analysis. Furthermore, in-context learning does not improve performance, indicating optimal tuning. Event studies and backtests show that our sentiment signals predict underperforming stocks, with higher model confidence in negative sentiment correlating to worse outcomes. These findings emphasize the value of fine-tuning models with expert-annotated data, and leveraging ESG sentiment signals to generate investment insights from qualitative data to enhance alpha generation.
As one of the founding fathers of modern finance, Harry M. Markowitz changed the way financial economists think about financial markets and institutions, and transformed the practice of finance from art to science. To honor his memory, I provide three specific examples of how portfolio theory played central roles in my own research, one involving the Adaptive Markets Hypothesis and two related to practical applications in biomedicine and fusion energy. These examples convinced me of the "unreasonable effectiveness"-to borrow a phrase from the great physicist Eugene Wigner-of portfolio theory in theory and practice.
Bruce Jacobs recounts his long professional and personal relationship with Harry Markowitz spanning more than 30 years in remarks delivered at the Spring 2024 JOIM interests and did complementary work. This led to collaboration, debate, and building upon each other's ideas and research. Their work covered topics on portfolio insurance, portfolio theory, market simulation, and risks of portfolio leverage, and helped to bridge the gap between theory and practice.
Assets in fixed income index mutual funds and exchange-traded funds (ETFs) have grown substantially in recent years. This paper examines fixed income index fund portfolio rebalancing efficiency using empirical evidence from four large fixed income index funds. We show how fund managers can preserve value in these portfolios using a wide range of dynamic portfolio management strategies while navigating the challenges posed by the general lack of liquidity and transparency in fixed income markets.
None of the explanations suggested so far for the size anomaly seems to be consistent with the empirical evidence. This paper examines under-diversification as a possible explanation for the size effect. When the portfolio weight of a stock is non-negligible, its variance is priced. As small stocks are much more volatile than large stocks, this induces a size effect. We analytically derive the relation between under-diversification and the size premium, which allows us to estimate the magnitude of the under-diversification-induced size effect. We find it to be in close agreement with the empirically measured size effect.
Using daily and intraday data from 1997 to 2023, we study strategies that stabilize volatility around a target by rebalancing between the S&P 500 and Treasury bills based on a broad set of volatility forecasts. Somewhat counterintuitively, lower forecasting errors do not necessarily result in more stable strategy volatility. Simple forecasts with fewer parameters can stabilize volatility as well as more complex models. In particular, combinations of implied volatility and simple estimators based on past returns exhibit good volatility control and lower turnover. On the implementation front, we show that the target volatility strategies we study are viable in the presence of realistic trading costs, delays between forecasting and rebalancing, or constraints on rebalancing frequency. Collectively, our findings can help design target volatility strategies that improve upon portfolios with constant target weights (e.g., a 60/40 portfolio) in achieving and maintaining investors' desired volatility exposures over time.
