Can Norwegian multi asset-class funds time the market?: an empirical study

Abstract

In this master thesis, we examine if Norwegian multi asset-class mutual funds (kombinasjonsfond) have the ability to time the market. Market timing skill is the ability to make successful bets on the future returns of an asset class. Norwegian multi asset-class funds have not been the subject for much research, and it is not clear if the market timing activities really adds value to the product. The market timing of multi asset-class mutual funds is particularly interesting, since they can invest in, and therefore time, multiple asset classes. Our extensive dataset consists of monthly holdings and returns data for 22 funds over 8 years. We use four returns based models; we extend a Treynor & Mazuy model and a Henriksson & Merton model in order to search for timing ability in multiple asset classes. In addition, we incorporate more risk-factors into these models. Then, we use a method developed by Clare et al. that measure the relationship between changes in portfolio holdings in one asset class and subsequent asset class returns. In addition, we present a case study of timing performance during the financial crisis. We interview multi asset-class fund managers to shed light on industry practice. Our results using the Treynor & Mazuy and Henriksson & Merton models show that a few funds seem to have market timing skill. Some of these few funds also seem to have market timing ability according to the holding based methods. The case study shows how funds on average increased their holdings of Norwegian equity at attractive levels after the financial crisis. We find that the portion of funds with timing skill in our study is higher than in most previous research from other markets. However, the most important finding is that regardless of method, the majority of funds in the sample do not have the ability to time the market.