Тарасов_резюме_ENG (Биржевые фонды (ETF) и скрытые индексные стратегии российских паевых инвестиционных фондов (ПИФ)), страница 2

Theliterature review presented in the first chapter is a logical supplement to the threepreviously published surveys. The paper introduces an alternative literatureclassification, covers articles not included in earlier published reviews, andsuggests ETF type’s classification and systematization of the market developmentstages. The survey provides a brief history of the market and a description of thefund working mechanism. Additionally, in conclusion, some ideas for furtherresearch are proposed. To my knowledge, this is the first dissertation research inRussia, related to ETF.In the first part of the second chapter of the dissertation, I document thephenomenon of the second type of closet indexing (using one ETF) and itsinfluence on the net investor financial result. Since 2014-2015, most Russian fundof-funds have been investing only in one preselected Western ETF (Annex 1).

If aninvestor buys shares in an ETF via a mutual fund rather than doing it directly, heoverpays in fees up to one-third of the invested capital over a 10-year horizon(Annex 6). Their fee level is not perceived as different from the active fund fee11After my review from the first chapter was published, there appeared another survey (Ben-David, Franzoni, Moussawi, 2017). The focus of that review is on the impact of ETF on assetprices; besides, the authors discuss the literature on a few other topics, including "ETF vs.

indexmutual funds "and "passive investment growth".6level. The funds do not say explicitly that they track an index. An investorexpenses comparison between investing in the shares of the same ETF via a mutualfund and a discount-brokerage account is carried out according to the methodologyof comparing investments in the same index via an ETF and index mutual fund(Kostovetsky, 2003).The second part of the second chapter is devoted to the active fund closetindexing and its relation to the investors’ net financial result, researched in linewith the Active Share concept (Cremers and Petajisto, 2009).

To assess the activefund closet indexing effect on the investor's risk-adjusted net return, I construct apanel database for the period from early 2011 to mid-2016 (1190 observations on62 equity funds). The database consists only out of funds whose benchmark couldbe found among the Russian indexes.12 Thus, in the sample there are only fundsthat invest more than 50% of their NAV13 in Russian assets.

One-factor CAPMalpha is taken as the net risk adjusted financial result measure. The research resultshows negative and significant correlation coefficient between the level of fundactivity (Active Share) and investor net risk-adjusted financial result (alpha). Thissurprising result contradicts Cremers, Ferreira, Matos, Starks’ (2016) observationbased on the analysis of 32 markets. Obviously, it raises questions about thequality on the active money management service in Russia. In this dissertation Imake an assumption that most managers do not deviate from index for reasonothers than to try outperforming it.

The further investigation into other reasons,possibly, would shift the focus from managers skills problem to the agency issue.This would only add arguments to the urgent need to reconsider the whole businesspractice around this underdeveloped open-ended fund industry.As fund commissions have an important impact on the shareholder’s netprofit, I built two multifactor regressions. In the first case, the dependent variableis the alpha, and independent ones are the fund’s active share, the fund’s size, its12Due to the Active Share methodology, for all funds should be found a correspondent index13NAV – net asset value7cash inflow. In the second regression, the dependent variable is the size of fundtotal shareholders costs,14 and independent ones are the fund’s active share, thefund’s size, its cash inflow and its return.

The level of Active Share has a positiveand significant correlation with the level of total shareholder costs. Obviously, themore active fund managers try to ask for the better remunerations and, as thecorrelation shows, they apparently succeed in it. The Cremers and Petajisto’s(2009) and Cremers, Ferreira, Matos, Starks’s (2016) results do not give the clearpicture on the researched markets. Regarding the relationship between totalshareholder costs and fund size, the regression demonstrates negative andsignificant correlation coefficient.

This is in line with international studies(Khorana, Servaes, Tufano, 2009), but contradicts the papers on the Russianmarket (Abramov, Akshenseva, 2015). The reason may be due to differences in thecomposition datasets. Usually, in Russian papers the funds are not divided into twogroups, even if only half of equity funds invest in Russian assets. another half ofopen-ended mutual funds invest mostly in foreign assets.The section gives a brief description of the Russian market Active Share andshows the market shares of mutual fund pursuing all three index strategies(explicit, closet and implicit indexing).

Data are compared with selected foreignmarkets. The share of completely passive funds (explicit and implicit indexing)reaches 30%, making the Russian market one of the most passive in the world. Thelevel of closet indexing is not distinguished from the general world statistics. Thissection introduces the Active Share concept to the Russian speaking academic fieldand partly contributes to available studies on 32 markets (Cremers, Ferreira,Matos, Starks, 2016).In order to make a better understanding of the Russian active moneymanagement service, I conduct the test whether the Russian mutual fundsoutperform the main home index, MICEX, (non-risk adjusted benchmark-adjusted14TSC (total shareholder cost) as in Khorana, Servaes, Tufano (2009)8return)15 during the post crisis period (2010 - 2016).

In average, active funds do notoutperform the index funds and obviously the index. The results contradict theconclusions of Abramov, Akshenseva, Radygin (2015) and Volodin andKuznezova (2015) who show that active fund non-risk adjusted return slightlyhigher those of index funds. Once again, the reason may lie in sample difference.To my knowledge, this is the first research in Russia that do not consider mutualfund that invest mostly in foreign assets when the question is whether Russianmutual funds outperform the Russian index. Furthermore, the second part Volodinand Kuznezova’s research demonstrates fund negative result if the studied returnwould be risk adjusted.All these, together with very low development level of the Russian openended mutual fund industry (Abramov, 2011; Abramov, Akshenseva, Radygin2015; Inozemzev, 2016) allows to conclude that new approaches to the industryrevival are strongly needed.The purpose of the third chapter is to formulate possible measures ofeconomic policy aimed at reducing the frequency of investor-made mistakesassociated with closet indexing.

Since, according to the results, the active fundscloset indexing in Russian at least does not cause economically significant damageto investors, further research focuses on the funds-of-funds closet indexing.Unfortunately, the research shows that even fairly objective and simple tounderstand existing fund descriptions do not allow the majority of even financiallyhighly educated people to realize that active money management is not needed not15Volodin and Kuznetsova (2015) propose before analyzing the risk-adjusted fund return to testone without taking risk into account.

Abramov, Akshentseva and Radygin (2015) show that it isalmost no difference in Russian mutual funds result if testing the return with risk or without riskconsideration. Since the research into relationship between the return and Active Share levelaims to complete the other countries survey (Cremers, Ferreira, Matos, Starks, 2016) I follow thekey principles of their methodology, taking risk into account.9only for funds-of-funds whose portfolio consists of one pre-selected ETF shares,but also for index funds.In this study I use the experimental method whose popularity greatlyincreased in recent years.16During the period 2004 – 2010 (the period of index investment popularityexplosion), literature emerged investigating why an individual investor invests inindex funds with higher commission while there are mutual funds tracking thesame index and taking much lower commissions for the same work. The feedispersion among mutual funds investing in the same index is not smaller thanthose dispersion among actively managed funds aiming to outperform this index(Choi, Laibson, and Madrian, 2010).

Almost all authors consider the investorschoice among the approximately 80 US index funds tracking the S & P 500index.17 In general, index fund researchers argue that in the rational investor viewindex funds tracking the same index are commodity goods18 (Elton, Gruber andBusse, 2004; Hortacsu and Syverson, 2004; Boldin and Cici, 2010; Choi, Laibsonand Madrian, 2010).The researched phenomenon has two aspects. The first is the fact ofsubstantial part of investors chose the high fee index funds among funds tracking16For example, the journal "Society for experimental finance", the Journal of Behavioral andExperimental Finance, released for the first time in 2014, as early as 2016, entered the Q2 ratingof SCOPUS in the "finance" section.

Moreover, two of the last three studies of a similar topicdescribed below use the experimental method (Choi, Laibson, and Madrian, 2010; Müller,Weber, 2010).17Due to various (often technical) factors, passive funds rarely reflect the dynamics of indexprice changes with absolute accuracy. There is always a tracking error. Therefore, investorsalways see different returns of index funds tracking the same index. The average active sharelevel of index funds in the US varies around 4% (Cremers et al., 2016).18Products with the same characteristics. Practically the only criterion for choosing among themis their price. For example, Brent crude oil, one wheat type of, one type of cheese (before it ispackaged for retail), etc.10the same index. The second is the theoretical conclusion that index fund investorschose index funds because they know about poor results of active fundsperformance: it is not more than 2% of active funds that outperform theirbenchmarks (Fama, French, 2010).