Document Type

Article

Publication Date

January 2016

Abstract

PurposeThe purpose of this study is to identify the factors that impact the exchange-traded funds net fund flow changes on a daily basis.Design/methodology/approachA total of 1,212 different exchange-traded funds with a proprietary daily net fund flow data and logistic regressions were studied because the majority of the 1,212 exchange-traded funds have mostly zero daily net fund flow changes.FindingsIt was documented that in the period December 22, 2005 to July 28, 2010 autocorrelation at the daily frequency is not universally present for the 1,212 exchange-traded funds that we study, despite the fact that this is the case in the monthly data documented in prior studies. No support was found for the feedback trading hypothesis but some support was found for the contrarian investor hypothesis on daily basis, even though the opposite is ascertained for both in the prior literature monthly data. Also, it cannot be concluded that tracking error prompts net fund flow changes and thus arbitrage activity.Originality/valueThe paper contributes to the ongoing analysis of the factors influencing investment companies fund flow changes, which has mostly focused on open-end funds and monthly data so far. Considering the increased scope and relevance of exchange-traded funds in today’s financial markets, this study fills a void in the fund flow changes literature.

Comments

This is the Accepted Manuscript of an article that appeared in Studies in Economics and Finance, 33, 2, 2016. The Version of Record is available at the following link.

Share

COinS