ETF Cost Obfuscation

Sammanfattning: Index-tracking ETFs have gained popularity by both retail and institutional investors over the past years while costs in the form of fees have declined due to competitive pressures. Index-tracking funds are relatively homogenous products with only the goal of replicating an index as close as possible. However, the funds differ in their levels of costs. We question why competitive pressure has not led to fee-convergence and hypothesize that the funds vary in their ability to track their underlying indices. Retail investors might focus on the fund's stated fees without thinking about more complex measures such as tracking ability. Thus, fund managers might be able to lower fees by putting less effort into tracking their benchmarks. In order to test this hypothesis, we study the relationship between costs and tracking errors for American and European ETFs during the period 2012-2021. Failing to find evidence of a negative relationship for American ETFs, we turn to ETFs listed in Germany tracking European indices. There, we observe a negative relationship between fees and tracking errors for ETFs tracking the STOXX 50 Net Return Index, suggesting lower tracking errors for higher levels of Total Expense Ratios. The finding supports the hypothesis that passive ETF managers obfuscate costs by putting less resources into tracking their benchmark. We extend our analysis to also include other determinants of expense ratios. Fund size is found to have varied effects on expense ratios, while the synthetic ETFs tracking the STOXX 600 Net Return Index are found to have significantly lower expense ratios and a weaker relationship between expense ratios and tracking errors than the physical ETFs.