Asset pricing with unsystematic risk
File(s)
Author(s)
Dello Preite, Massimo
Type
Thesis
Abstract
For decades, asset pricing has operated under the foundational assumptions that unsystematic, diversifiable risk is not compensated. That assumption has shaped both theory and empirical work. Within this framework, researchers have long sought to explain differences in the cross-section of expected returns. Although these differences are understood to reflect the risk-return trade-off of each asset, identifying the relevant sources of risk has proven challenging. Despite the proliferation of systematic risk factors, a substantial portion of expected returns remains unexplained, suggesting that the traditional view may be incomplete. In this thesis, my coauthors and I take a different perspective. Rather than searching for additional, previously unmodeled sources of systematic risk, we challenge the widespread belief that only systematic risk carries a risk premium. This shift represents a conceptual break from traditional asset pricing. We work within the framework of the Arbitrage Pricing Theory which allows for non-zero alphas while remaining fully consistent with no-arbitrage.
In Chapter1, we demonstrate that allowing for compensation of unsystematic risk significantly improves the pricing performance of factor models and reveals that unsystematic risk plays a central role in explaining expected returns. We also present a simple equilibrium model in which unsystematic risk is compensated.
In Chapter2, we study the estimation and econometric implications of incorporating compensation for unsystematic risk alongside a, possibly non-exhaustive, set of candidate, observable factors in the APT framework. We also show how to assess and compare these enriched models using both pricing-based and portfolio-based performance metrics.
In Chapter3, we investigate the consequences of priced unsystematic risk for portfolio construction. Once unsystematic risk is recognized as compensated, it is no longer optimal to diversify it away. Instead, efficient portfolios must combine the traditional factor-based portfolio with a second portfolio that loads only on unsystematic risk showing that unsystematic risk enhances portfolio efficiency.
In Chapter1, we demonstrate that allowing for compensation of unsystematic risk significantly improves the pricing performance of factor models and reveals that unsystematic risk plays a central role in explaining expected returns. We also present a simple equilibrium model in which unsystematic risk is compensated.
In Chapter2, we study the estimation and econometric implications of incorporating compensation for unsystematic risk alongside a, possibly non-exhaustive, set of candidate, observable factors in the APT framework. We also show how to assess and compare these enriched models using both pricing-based and portfolio-based performance metrics.
In Chapter3, we investigate the consequences of priced unsystematic risk for portfolio construction. Once unsystematic risk is recognized as compensated, it is no longer optimal to diversify it away. Instead, efficient portfolios must combine the traditional factor-based portfolio with a second portfolio that loads only on unsystematic risk showing that unsystematic risk enhances portfolio efficiency.
Version
Open Access
Date Issued
2026-01-15
Date Awarded
2026-06-01
Copyright Statement
Attribution-NonCommercial 4.0 International Licence (CC BY-NC)
License URL
Advisor
Zaffaroni, Paolo
Publisher Department
Business School
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)