Asset Composition and Performance of Swedish Listed Mutual Funds

Abstract: Fund investments are very popular in Sweden. However, we have the impression that despite this popularity, the average fund investor in Sweden does not pay much attention to the importance and possible link of fund’s asset composition features (e.g. Asset class, Holdings, and Geo-exposure) to fund’s performance. Instead, S/he relies on factors such as fees, risk levels, historical performance, etc. in her/his investment decisions. Similarly, academic studies mainly focus on attributes such as funds fees, size, and manager’s skill to explain fund’s performance. Thus there are limited premier academic studies on the relationship between fund’s performance and its asset composition features. The main purpose of this study is to investigate possible causal relationship between the performances of funds with their assets composition features. We study the whole population of 346 Swedish listed mutual funds older than five years for the period 2009-2013. The results of the study provides the investors and analysts with additional decision-making and investment-analysis tools to assist them in making more informed judgment on funds and their expected returns. The results are also useful for fund managers to improve their strategies by refining the combinations of their funds’ asset composition attributes in order to improve the absolute risk-adjusted performance of their funds. Our research philosophy has been based on positivism and objectivism along with functionalist paradigm and we have applied deductive approach to test the theories. We have used quantitative method and collected the funds’ data from public business databases and chosen Jensen’s alpha and Treynor ratio as funds’ risk-adjusted performance measures. We performed Correlation tests and Regression with robust techniques on our data to answer the research question from three aspects, namely asset class (equity, bond, and mixed assets); geo-exposures (Sweden, Global, Europe, and Nordic) and Top-ten holdings’ measures (asset concentration and Treynor of each fund’s passive top-ten sub-portfolio). We conclude that correlations between funds’ risk-adjusted performance and assets composition features are likely to exist. Stronger correlations are observed between the explanatory measures and fund’s relative risk-adjusted performance (fund’s Treynor) as compared to fund’s absolute risk adjusted performance (fund’s Jensen’s alpha). Asset concentration in top-ten holdings and bond asset class are more likely to be in casual relationship with fund’s risk-adjusted performance, whereas Treynor ratio of top-ten holdings’ passive sub-portfolio as well as fund’s geo-exposure do not seem to have strong explanatory power for funds’ absolute performance.