We study Merton's optimal portfolio problem in a market responsive to macroeconomic regimes, as characterized by VIX. Instead of solving the traditional Hamilton-Jacobi-Bellman equation, we train an artificial neural network (ANN) to learn the optimal allocation as a feedback function. Our regime-specific strategy using a simple ANN with one hidden layer, benchmarked against the classical Merton portfolio, shows superior performance subject to realistic diversification constraints with borrowing/short selling excluded. A 35-year backtest (1990-2024), including 17 out-of-sample years, on a diversified portfolio of twelve assets plus cash, reveals that accounting for regime shifts improves both expected utility and average returns.
We introduce carbon beta, a measure of climate transition risk determined by a stock's return sensitivity to a pollutive-minus-clean portfolio. Carbon beta is higher for smaller and more leveraged firms, firms with more investments and fixed assets, as well as firms with lower R&D. Carbon betas correlate with green patent issuance and other forward-looking measures of climate risk. We study the interaction of carbon beta with shocks to climate risk to judge its hedging ability: Returns to stocks with high carbon betas are lower during months with climate risk realizations.
We propose a novel method to estimate emotional yields of collectibles based on factor-mimicking portfolios. Using up to 110 years of collectibles returns for 13 distinct asset classes, we apply machine learning techniques to address challenges from non-synchronous trading. We use these estimates to study how emotional yields affect equilibrium pricing. Emotional yield estimates for 24 of our 30 collectibles return series are positive, with an annualized mean (median) of 2.64% (2.53%). Despite various forms of underestimation, these results provide evidence that assets with positive emotional returns have lower equilibrium financial returns.
When measured with daily return data, hedge fund factor exposures are more statistically significant than when measured with monthly data. Consequently, daily data can significantly improve investors' hedge fund portfolios. Furthermore, daily data reveal that hedge funds have lower alphas and are exposed to more factors, and these exposures are more time-varying, than monthly data suggest. Analysis of hedge fund exposures with a comprehensive universe of daily variables reveals the importance of a new commodity put factor and of new liquidity timing effects for factors other than the market.
Many investors believe that the US stock market is riskier than it has been historically because a large fraction of its capitalization is concentrated in a few large technology companies. Some investors, therefore, conclude that they should rebalance their portfolios toward safer assets. The authors present clear evidence that the US stock market has indeed become more concentrated. However, they present practical and conceptual arguments along with persuasive empirical evidence that challenges the notion that investors should act to offset concentration.
Each year, a significant number of stocks transition from non-value stocks to value stocks and likewise from non-growth stocks to growth stocks. As a result, about half of value stocks and half of growth stocks are new. We find that the value premium based on new value and growth stocks is statistically higher than that based on old value and growth stocks, mostly driven by the underperformance of new growth stocks. In addition, we find that the large value premium for new value and growth stocks is more pronounced during contraction, the Federal Reserve monetary tightening cycle, subperiods with high long-term yields, and high economic uncertainty. Moreover, we show that the large value premium for new value and growth stocks is mainly the across-industry effect. Finally, international evidence further confirms the findings.
We study which firm characteristics drive the economic value of machine learning portfolios. Three results stand out. First, in-sample variable importance overfits and provides little reliable guidance, highlighting the need for out-of-sample evaluation using economic criteria. Second, conventional models are dominated by microcaps, which inflate returns and concentrate gains in costly-to-trade stocks; excluding microcaps is essential for meaningful inference. Third, some predictors carry negative importance and consistently degrade performance; removing them improves risk-adjusted returns and clarifies which characteristics matter. These findings show that only with economic restrictions can machine learning deliver robust asset pricing insights.
Conventional growth indices suffer from two important shortcomings. First, stocks that are anti-value (very expensive) are not necessarily growth stocks. The decision to include a stock in a growth index should be based on fundamental growth measures, such as growth in sales, profits, or R&D spending, rather than price-based measures. Second, when these indices are weighted by objective measures of growth, rather than by market value, performance markedly improves. Overpaying for growth is unhelpful. We also assert that some stocks with poor growth prospects and unattractive valuations may have no place in either value or growth indices.
We survey Small Business Investment Companies (SBICs) to perform a novel analysis of their performance. SBIC funds outperform comparable non-SBIC peers by around 2% to 3% as measured by internal rate of return and about 0.3x to 0.7x as measured by multiple on invested capital, depending on benchmark deployed. To mitigate sample selection bias, we also examine SBICs in the MSCI Private Capital Universe data, which shows similar, but smaller, outperformance. We analyze SBIC funds by fund strategy and amount of leverage utilized to provide a granular view of risk-adjusted performance. We believe this to be the first large-sample analysis of SBIC returns.
We examine downside protection-or defensive-strategies over more than 220 years of global financial history, covering many years in which traditional equity-bond portfolios suffer and across a wide range of economic scenarios and historical regimes. Traditional defensive equity factors-low-risk, quality, and value-consistently provide effective downside protection, whereas gold and put options prove less drawdown or cost-effective. Our long-run evidence shows that multi-asset defensive strategies, particularly a return-enhanced version of the defensive absolute return (DAR) portfolio introduced by Cavaglia et al. (2022) and trend-following, provide the most effective downside protection. DAR and trend-following are complementary across tests by diversifying each other across stages of drawdowns. Investors can improve the defensive properties and improve total portfolio outcomes of traditional portfolios by considering the deep sample evidence on defensive strategies provided in this paper.
We investigate the impact of Environmental, Social, and Governance (ESG) rating changes and daily ESG news sentiment on firm credit risk. We document a significant increase in credit default swap (CDS) spreads following ESG rating downgrades, especially for the social pillar, while we find a muted reaction to ESG upgrades. A similar asymmetrical effect is documented for ESG news. We further show that the adverse effect of ESG downgrades on the CDS market is stronger for firms with lower creditworthiness, but mitigated in the presence of positive ESG sentiment, a transparent information environment, and higher rating disagreement.
The Sharpe ratio is almost perfectly aligned with investors' welfare when borrowing is unrestricted. However, when borrowing is realistically restricted, this alignment breaks down dramatically. We show that the geometric mean (GM) provides a much better alternative for fund ranking in this case. Estimates of the ex-ante GM can be improved by first shrinking the sample gross GM and then subtracting fees. The generalized GM (GGM) captures this idea and provides a good estimate of the future net GM. We argue that mutual fund selection can be substantially improved by employing the GGM rather than the more popular Sharpe ratio or alpha.
Default and term structure risk are key drivers of fixed income performance. Ignoring this information when comparing investment strategies can be misleading. This study proposes an algorithm derived from mimicking factor portfolios to neutralize risk differences, thereby distinguishing selection from market timing. For a well-diversified portfolio, this method allows for simultaneous management of multiple risk dimensions, ensuring the final portfolio remains investable. The algorithm can be modified in such a way as to guarantee positive weights, thus offering greater flexibility compared with conventional methods. We apply it to credit sector portfolios to neutralize discrepancies in duration times spread (DTS) and find notable differences between risk-adjusted and unadjusted performance.
Over the last thirty years there has been a strong positive trend in the magnitude of amortization charges, due to both economic and accounting changes. This trend has accelerated over the last decade, following the implementation of a revised accounting standard for business combinations. Concurrent with the recent trend, managers and external users of financial statements increasingly discuss operating performance focusing on earnings metrics that exclude amortization but include depreciation. This study compares earnings before interest, taxes and amortization (EBITA) with its two more common alternatives—EBIT and EBITDA. Consistent with the amortization trend, EBITA’s advantage over EBIT in explaining market values has gradually increased over time. However, throughout the sample period, EBITDA performed substantially better than both EBITA and EBIT. In terms of predicting stock returns, the three operating income measures performed well in the 1990s and 2000s, but not over the last decade.
Since its inception in 1945, the Financial Analysts Journal (FAJ) has advanced some of the investment profession's most influential ideas by providing an outlet for innovative thinkers. We trace the FAJ's history by identifying the most prolific contributors and innovations featured over its first 80 years and in each of nine financial eras. Using the comprehensive database and rigorous methodology that we developed, this article provides rankings of the top authors and the most frequent words in titles and examines the context in which these words were used to identify seminal ideas and the authors behind them.
