Factor investing stands as a cornerstone methodology in the sphere of portfolio management, where the selection of securities is guided by identifiable and quantifiable characteristics - referred to as factors - that are empirically linked to potential excess returns. These factors, among which value, size, momentum, quality, and volatility are most prominent, serve as the bedrock for constructing investment strategies that aim to achieve superior risk-adjusted performance compared to the broader market benchmarks.

Single factor models are strategies that concentrate on exploiting the return potential of one specific factor. The advantage of such a focused approach lies in its clarity and ease of implementation where investors can distinctly attribute the performance of their portfolio to the behavior of the selected factor. Moreover, the simplicity inherent in single factor models allows for straightforward attribution analysis and rebalancing procedures. However, these models are not without their limitations. The reliance on a singular factor exposes the investor to a heightened degree of cyclical risk, whereby the factor may exhibit varying degrees of performance through different economic phases. This can potentially lead to periods of significant underperformance. Furthermore, such models harbor concentration risks, as the portfolio may be unduly exposed to sector-specific shocks or macroeconomic trends that disproportionately affect the chosen factor.

In contrast, multi-factor models present a more nuanced and sophisticated investment strategy. By combining various factors, the multi-factor models strive to construct a portfolio that captures a more comprehensive set of factor risk premiums, potentially leading to a more consistent and stable performance over time. The diversification achieved through the combination of low-correlated factors reduces overall portfolio volatility, thereby offering a smoother investment journey. Nonetheless, the intricate nature of multi-factor models introduces complexity to the investment process. The risk of overfitting, a scenario where a model is excessively tailored to historical data, thus impairing its future predictive power is a pertinent concern. Additionally, the interplay between different factors may lead to a dilution effect, where the strong performance of one factor is offset by the weaker performance of another, potentially muting the overall return profile of the portfolio.

Source: Bloomberg. The correlations mentioned above are the average of daily rolling 10-year correlation for the period starting from 05 July 1995 to 31 December 2024. The correlations are calculated using the daily excess return over the S&P 500 Total Return Index.

Source: Bloomberg. The correlations mentioned above are the average of daily rolling 10-year correlation for the period starting from 15 July 2014 to 31 December 2024. The correlations are calculated using the daily excess return over the S&P Europe 350 Total Return Index.

Source: Internal research, CMIE, National Stock Exchange of India, NJ’s Smart Beta Platform (in-house proprietary model of NJAMC). The correlations mentioned above are the average of daily rolling 10-year correlation for the period starting from 30 September 2006 to 31 December 2024. The correlations are calculated using the daily excess return over the Nifty 500 total return index.