Performance Comparison of CAPM and Fama–French Factor Models Across Developed and Emerging Markets

Abstract

The asset pricing models are very important to financial economics in the modern day, and they can help in estimating expected returns, assessing investment performance, and making portfolio management and corporate financial decisions. The Capital Asset Pricing Model (CAPM), the Fama–French Three-Factor Model (FF3F), and the Fama–French Five-Factor Model (FF5F) are some of the significant models in the evolution of asset pricing theory among the above. This study provides an empirical and theoretical analysis of these models that contrasts them.This study synthesizes theoretical developments and empirical evidence from developed and emerging markets and compares these models. The study uses a semi-systematic (integrative) literature review method by searching for peer-reviewed articles from the leading academic databases (Scopus, Web of Science, ScienceDirect, SpringerLink, Emerald Insight, Taylor & Francis, Wiley Online Library, and Google Scholar) that appeared throughout the year between 1993 and 2026. The analysis explores the models in terms of their theoretical assumption, factor structure, explanatory power, empirical performance, application to practice, strengths and limitations. The results suggest that although CAPM is a simple and widely used benchmark it does have a limitation because it uses only market beta to explain the returns. The FF3F model consists of the FF model with the addition of size and value factors while the FF5F model is the FF model with the addition of profitability and investment factors, both of which substantially improve the asset pricing properties of the model, and the FF5F model improves explanatory power further. The performance of these all-factors models, however, is somewhat different on developed and emerging markets because of the differences in market efficiency, institutional quality, investor behavior and macroeconomic conditions. Moreover, long-term anomalies like momentum, liquidity and sustainability-related risks are not well understood in current models. The study aims to highlight the theoretical, methodological and geographical research gaps and to suggest future research directions combining behavioral finance, environmental, social and governance (ESG) factors, alternative data and machine learning techniques. The results offer an overview for researchers, practitioners, and policy makers interested in learning about the development, use, and prospects of asset pricing models.