No abstract

PurposeThe purpose of this paper is to investigate whether downside-risk measures help to explain why households largely refrain from investing in Exchange Traded Funds that replicate broad and internationally diversified market indices, so-called XTFs, although studies frequently recommend to do so.Design/methodology/approachThe paper analyzes whether evaluating risk in terms of downside-risk measures which reflect households' interpretation of risk closer than the standard deviation (SD) of returns, yields less risk-return-enhancements, and thus, fewer incentives for households to invest in XTFs. Household portfolios are compiled by combining stylized portfolio compositions that involve multiple asset classes and German households' security holdings. The data set covers the period from January 2014 to December 2016 and includes 47,388 securities.FindingsThe results indicate that none of the downside-risk measures can help to explain the reluctance of households to invest in XTFs. On the flip side, the results show that all stylized household portfolios can enhance the risk-return position from employing XTFs, regardless of the underlying risk measure. This supports the advice to invest in XTFs and extends it upon households that evaluate risk in terms of downside-risk.Originality/valueTo the best of the authors' knowledge, this study is the first to investigate risk-return-enhancements from XTFs while simultaneously considering various downside-risk measures and multiple asset classes of household portfolios.

In this doctoral thesis, XTFs are considered as plain-vanilla Exchange Traded Funds which pursue a passive investment approach and replicate a broad, internationally diversified market index. The thesis pursues two research goals. First, to investigate whether XTFs enhance risk and return of household portfolios when taking multiple relevant asset classes into account – not only stocks. Second, to examine whether employing XTFs in household portfolios is reasonable when including practical constraints and risk measures that reflect households’ actual investment situation and interpretation of risk more closely compared to previous literature. To meet these research goals, three empirical analyses are conducted. The analyses are built on the foundations of neoclassical finance theory, new institutional economics, market microstructure theory, financial intermediation, as well as behavioral finance and economics. For the empirical investigations, two data sets of the German central bank (Deutsche Bundesbank) are combined: The Panel on Household Finances-survey and the Securities Holdings Statistics-base. The thesis ends by discussing the empirical analyses' results, possible limitations to the findings' generalizability, and implications for different stakeholders.