Do Mechanical Value Investing Strategies Beat The Market In The Nordics?

Abstract

In this thesis, we study whether three mechanical value investing strategies consistently generate excess risk-adjusted returns (alpha) on the Nordic exchanges for a Norwegian investor (returns are reflected in NOK). We backtest: l) Piotroski (2000)'s selection method, 2) Greenblatt (2006)'s "magic formula", and 3) a mechanical strategy employed by the new Norwegian mutual fund First Veritas (FV). We employ the CAPM, Fama and French's threefactor model (FF3F), and Carhart's four-factor model (C4F) to measure alpha. The data coverage allows for backtests from July 2008 to the end of 2021. Before accounting for transaction costs, the "magic formula" generates statistically significant alpha (on the 5% level) with the C4F, and the FV strategy generates significant alpha with all models. The Piotroski method's alpha is statistically indistinguishable from zero with all models. When controlling for transaction costs (i.e., bid-ask spreads and commission fees), the FV strategy generates significant alpha with the CAPM and C4F, while the "magic formula" portfolio's alpha becomes insignificant with all models. Furthermore, when assessing the FV strategy in a mutual fund setting where we exclude companies below 2000 MNOK market capitalization and account for fees, the alpha is statistically insignificant with all models.