City Research Online

Abstract

A vast literature on the cross-section of expected returns documents that many firm characteristics predict stock returns in ways that cannot be explained by the CAPM. A central question in asset pricing is whether these cross-sectional relationships reflect compensation for systematic risk, or whether they instead arise from mispricing that persists because arbitrage is limited. The three essays in this thesis examine this question from complementary perspectives, focusing primarily on the distinction between intertemporal risk and arbitrage risk as competing explanations for return predictability.

The first essay examines whether securities lending markets can be used to distinguish genuine mispricing from noise in anomaly signals. Anomaly signals are widely used to identify undervalued and overvalued stocks, but they are inherently noisy, producing false positives on both the long and short sides. The essay shows that mispricing is concentrated in stocks with high idiosyncratic volatility, where benchmark-driven tracking-error constraints limit arbitrage. Within this region, shorting demand helps identify where benchmark-unconstrained investors actively take positions against genuine mispricing. Conditioning jointly on anomaly signals, idiosyncratic volatility, and shorting demand allows the essay to identify false longs and false shorts, and to construct a refined long-short factor that remains profitable after shorting fees.

The second essay tests whether the value premium reflects intertemporal risk or arbitrage risk. Existing evidence struggles to distinguish between these explanations because both predict a positive relation between the value premium and idiosyncratic volatility. The essay identifies the mechanism using a sign test: behavioural limits-to-arbitrage models predict opposite-signed returns for value and growth portfolios sorted on idiosyncratic volatility, while risk-based models predict zero or same-signed returns. Consistent with the behavioural interpretation, the essay finds a sign reversal concentrated among small stocks, where arbitrage constraints are most binding, and absent among large stocks. Decomposing the value factor, the essay shows that the premium is driven by small, high-idiosyncratic-volatility stocks, while post-publication evidence and securities lending data suggest that the value premium is best understood as mispricing that survives precisely where arbitrage is most difficult.

The third essay uses country equity indices as a laboratory to distinguish between risk-based and limits-to-arbitrage explanations for cross-sectional return anomalies. Country equity indices are large, liquid assets predominantly held by institutional investors, which materially reduces the stock-level frictions and arbitrage risk that underpin mispricing explanations. This setting therefore provides a cleaner test of whether anomaly returns reflect broadly priced risks. The essay finds that the profitability premium is strong and statistically significant, consistent with a systematic risk interpretation, while the value premium is weak over the sample period. The negative idiosyncratic volatility and skewness effects reported among individual stocks disappear in the country-index sample, pointing to arbitrage risk as their primary driver. The essay also documents that cash-flow beta commands a higher price of risk than discount-rate beta at the country-index level, providing out-of-sample support for a central prediction of Campbell and Vuolteenaho (2004).

Publication Type:	Thesis (Doctoral)
Subjects:	H Social Sciences > HB Economic Theory H Social Sciences > HD Industries. Land use. Labor > HD61 Risk Management H Social Sciences > HG Finance
Departments:	Bayes Business School > Bayes Business School Doctoral Theses Doctoral Theses

Export

Downloads