The Efficiency of Mutual Fund Families: Applications to ... · he mutual fund industry has...

The Efficiency of Mutual FundFamilies: Insights from theSpanish Market

This page intentionally left blank

The Efficiency of MutualFund Families: Insightsfrom the Spanish Market

By

Carlos Sánchez-GonzálezUniversidad EIA, Colombia

United Kingdom � North America � Japan � India � Malaysia � China

Emerald Publishing LimitedHoward House, Wagon Lane, Bingley BD16 1WA, UK

First edition 2018

Copyright r 2018 Emerald Publishing Limited

Reprints and permissions serviceContact: [email protected]

No part of this book may be reproduced, stored in a retrieval system,transmitted in any form or by any means electronic, mechanical,photocopying, recording or otherwise without either the prior writtenpermission of the publisher or a licence permitting restricted copying issuedin the UK by The Copyright Licensing Agency and in the USA by TheCopyright Clearance Center. Any opinions expressed in the chapters arethose of the authors. Whilst Emerald makes every effort to ensure the qualityand accuracy of its content, Emerald makes no representation implied orotherwise, as to the chapters’ suitability and application and disclaims anywarranties, express or implied, to their use.

British Library Cataloguing in Publication DataA catalogue record for this book is available from the British Library

ISBN: 978-1-78743-800-2 (Print)ISBN: 978-1-78743-799-9 (Online)ISBN: 978-1-78743-983-2 (Epub)

Certificate Number 1985ISO 14001

ISOQAR certified Management System,awarded to Emerald for adherence to Environmental standard ISO 14001:2004.

http://[email protected]

I want to dedicate this book to my wife Doris and my sons,Nicolas and Juan Diego; their love and unconditional supportwere essential to fulfill this project. Thank you God for beingpresent in our lives and allowing us to have so many blessings.

I also want to dedicate this achievement to my mother and myfather (†2017); without the adventurous vision that they gave

me, I could not have fulfilled my dreams.

Contents

List of Figures xi

List of Tables xv

About the Author xix

Preface xxi

Acknowledgments xxiii

Part I The Efficiency of Mutual FundFamilies: Using a Slacks-basedMeasure

CHAPTER 1 Introduction to Part I 3

CHAPTER 2 Background 72.1. Basic Concepts 72.2. Basic DEA Models 112.3. Applications to Financial Institutions 23

2.3.1. Context of Study 232.3.2. Review of Frontier Studies in Financial

Institutions 252.3.3. Discussion on the Production and Asset

Approaches 30

CHAPTER 3 Model and Variables 353.1. Multi-management Stages Model in a Mutual Fund

Management Company (MFMC) 353.2. Variables of the Model: The Case of the

Spanish Market 39

vii

CHAPTER 4 Data and Empirical Analysis:The Case of the Spanish Market 47

4.1. Data 474.2. Empirical Analysis 524.3. Influence of the Variable-returns-to-scale on the

Efficiency Rankings 884.4. Robustness of the More Controversial

Variables of the Model 91

CHAPTER 5 Conclusions and Summary of theApplications in the Spanish Market 97

Appendix A 99

Part II Further Evaluation of Efficiency ofMutual Fund ManagementCompanies

CHAPTER 1 Introduction to Part II 119

CHAPTER 2 Variations of the Slacks-based Measure(SBM) of Efficiency 123

2.1. Basic Concepts of the New Measures 1232.2. Comparative Evaluation between the SBM

Original Model and SBM Variation I 1322.3. Random Searches for Facets (Variation IV) 1362.4. The Search for Locally Efficient Companies Applying

Variation III 143

CHAPTER 3 Persistence in Funds ManagementCompanies: Do the Best Winners andLosers Usually Repeat? 165

3.1. Persistence Studies in the Mutual Fund Industry 1653.2. Methodology of Performance Persistence 1673.3. Empirical Results of Efficiency Persistence:

The Case of the Spanish Market 188

viii CONTENTS

CHAPTER 4 Conclusions and Summary of theApplications in the Spanish Market 201

Appendix B 205

References 225

Index 235

Contents ix

List of Figures

Part I: The Efficiency of Mutual Fund Families:Using a Slacks-based MeasureChapter 2

Figure 1 Relationships between Productivity,Efficiency, and Effectiveness. . . . . . . . . . . 9

Figure 2 Production Frontier and Technical Efficiency. . 10

Figure 3 Differences of Returns to Scale. . . . . . . . . 17

Chapter 3

Figure 4 MFMC Interactions Based on Berkowitz andQiu (2003). . . . . . . . . . . . . . . . . . . 37

Figure 5 Multi-management Stages Model for anMFMC. . . . . . . . . . . . . . . . . . . . . 38

Figure 6 Multi-management Stages Model for anMFMC (Including Variables). . . . . . . . . . 40

Chapter 4

Figure 7 Evolution of the Mutual Fund Industry(Example Spanish Market). . . . . . . . . . . 50

Part II: Further Evaluation of Efficiency of MutualFund Management CompaniesChapter 2

Figure 8 Convex Set. . . . . . . . . . . . . . . . . . . 124

Figure 9 Convex Set with Two Shps. . . . . . . . . . . 125

Figure 10 Random Search around Efficient DMUs(Tone, 2010). . . . . . . . . . . . . . . . . . 131

Figure 11 Comparative Graphs between SBM andVariation I Scores for 2005. . . . . . . . . . . 134

xi

Figure 12 Comparative Graphs between SBM andVariation I Scores for 2009. . . . . . . . . . . 135

Figure 13 Distinction between Agglomerative and DivisiveTechniques. . . . . . . . . . . . . . . . . . . 146

Figure 14 Dendrogram (Divisive Analysis) DIANA for2005.. . . . . . . . . . . . . . . . . . . . . . 151

Figure 15 Dendrogram (Divisive Analysis) DIANA for2006 . . . . . . . . . . . . . . . . . . . . . . 152




Chapter 3

Figure 19 Dendrogram DIANA of Efficiency Scores forStage 1 of 2005. . . . . . . . . . . . . . . . . 170










Figure 29 Dendrogram DIANA of Efficiency Scores forOverall Efficiency of 2005. . . . . . . . . . . . 180

xii LIST OF FIGURES





List of Figures xiii

List of Tables

Part I: The Efficiency of Mutual Fund Families:Using a Slacks-based MeasureChapter 2

Table 1 Comparison between Basic DEA Models. . . . 24

Chapter 3

Table 2 Set of Inputs and Outputs. . . . . . . . . . . . 40

Chapter 4

Table 3 Descriptive Statistics of the Data(31st December). . . . . . . . . . . . . . . . . 48

Table 4 Efficient Companies by Stage and Variable (%). 53

Table 5 SBM-efficiency under CRS for 2005. . . . . . . 54





Table 10 Average CRS-SBM Scores and Spearman RankCorrelation. . . . . . . . . . . . . . . . . . . 87

Table 11 Average VRS-SBM Scores and Spearman RankCorrelation. . . . . . . . . . . . . . . . . . . 90

Table 12 Spearman Rank Correlation between CRS-SBMand VRS-SBM Eff. Scores. . . . . . . . . . . . 90

Table 13 Stage 1 — Robustness Analysis Based on LaborData. . . . . . . . . . . . . . . . . . . . . . . 93

Table 14 Stage 1 — Robustness Analysis Based on theCapital Effort Data. . . . . . . . . . . . . . . 94

Table 15 Overall Efficiency — Robustness Analysis Basedon the Capital Effort Data.. . . . . . . . . . . 95

xv

Appendix A

Table 16 Comparative CRS-SBM Scores for 2009 withDifferent Measures of Labor . . . . . . . . . . 100

Table 17 Comparative CRS-SBM Scores for 2009with Different Measures of Capital. . . . . . . 106

Table 18 Comparative CRS-SBM Scores for 2009 withDifferent Measures of Capital . . . . . . . . . 112

Part II: Further Evaluation of Efficiency of MutualFund Management CompaniesChapter 2

Table 19 Comparative Findings between SBM andVariation I Scores. . . . . . . . . . . . . . . . 133

Table 20 Average CRS-variation I Scores and SpearmanRank Correlation. . . . . . . . . . . . . . . . 137

Table 21 Maximal Friends Found for 2009. . . . . . . . 139

Table 22 Friends and Maximal Friends (Variation IV). . 140

Table 23 Comparative Findings between Variation I andVariation II Scores.. . . . . . . . . . . . . . . 141

Table 24 Average CRS-variation II Scores and SpearmanRank Correlation. . . . . . . . . . . . . . . . 142

Table 25 Cluster Classification by Year. . . . . . . . . . 157

Table 26 Descriptive Statistics of Clusters. . . . . . . . . 158

Table 27 Efficient Companies by Stage. . . . . . . . . . 159

Table 28 Efficient Companies by Clusters. . . . . . . . . 160

Chapter 3

Table 29 Efficiency Persistence for Stage 1 Based on 2× 2Contingency Tables. . . . . . . . . . . . . . . 189

Table 30 Efficiency Persistence for Stage 2 Based on 2× 2Contingency Tables. . . . . . . . . . . . . . . 189

Table 31 Efficiency Persistence for Overall EfficiencyBased on 2× 2 Contingency Tables. . . . . . . 190

Table 32 Performance Persistence for Stage 1 Based on4×4 Contingency Tables. . . . . . . . . . . . 191

Table 33 Performance Persistence for Stage 2 Based on4×4 Contingency Tables. . . . . . . . . . . . 192

xvi LIST OF TABLES

Table 34 Performance Persistence for Overall EfficiencyBased on 4×4 Contingency Tables. . . . . . . 193

Table 35 Aggregate Mean and Deviation Value byNumber of Employees. . . . . . . . . . . . . . 195

Table 36 Aggregate Mean and Deviation Value byShareholder Equity. . . . . . . . . . . . . . . 196

Table 37 Aggregate Mean and Deviation Value by AssetsManaged. . . . . . . . . . . . . . . . . . . . 197

Table 38 Aggregate Mean and Deviation Value byNumber of Funds. . . . . . . . . . . . . . . . 198

Table 39 Comparison of Mean Values between“Top W�Top W” and “Bott L�Bott L”Companies. . . . . . . . . . . . . . . . . . . 199

Appendix B

Table 40 Variation I Efficiency under CRS for 2009 . . . 206

Table 41 Variation II Efficiency under CRS for 2009 . . 212

Table 42 Variation III Efficiency under CRS for 2009 . . 218

List of Tables xvii

About the Author

Dr. Carlos Sánchez-González is a Full-time Professor at theUniversidad EIA, Colombia. He holds a first-class honors bache-lor’s degree in Financial Engineering from the UniversidadAutónoma de Bucaramanga — UNAB, Colombia. Then, heobtained a competitive scholarship by Santander Bank for youngResearch Professor from Latin America, and he completed hisMSc studies in Accounting and Finance, and presented his PhDdissertation in Finance Program in Spain. He obtained theFinancial Research Award (Accésit) “Antonio Dionis Soler”2015 by the Foundation of Financial Studies (FEF) and SpanishInstitute of Financial Analysts (IEAF). His research on efficiencymeasurement in organizations has been presented at Congressesin Colombia, Ecuador, and Spain. He has published paperson recent innovations in the DEA methodology, such as “Theefficiency of mutual fund companies: Evidence from an innova-tive Network SBM Approach,” in the Omega Journal ofManagement Science, under the co-authorship with ProfessorVicente and Professor Sarto (University of Zaragoza).

He served as a Research Professor in the ESAI Business School,Universidad de Especialidades Espiritu Santo — UEES; AssistantProfessor at the Faculty of Administrative Sciences, UniversidadTecnológica Equinoccial — UTE; Visiting Professor at theBusiness School, Universidad San Francisco de Quito; VisitingResearcher at the College of Science and Mathematics, MontclairState University; and Visiting Professor at University of Zaragoza.

He worked as Chief Financial at Automotores Continental CompanyS.A. (the largest Chevrolet dealer in Ecuador), Head of Budgets atProinco Finance Corporation S.A., Ecuador, and Financial SupportEngineer at Bucaramanga Corporation Entrepreneurship —

Incubator of Technology-based Firms, Colombia.

xix

Preface

The mutual fund industry has witnessed remarkable growthworldwide during the last decade. A significant effect onthe savings patterns of the countries has been promoted

by investment alternatives offered by mutual funds. Mutual fundmanagement companies have suffered both great expansion andcontraction periods in the same manner as other financial institu-tions. The need to analyze the efficiency of financial institutionshas instigated an important line of research in the financial litera-ture using alternative approaches. However, the emphasis ofthese studies has been typically on the banking and insuranceindustry rather than other specific companies related to theinvestment business. In addition, research on the mutual fundindustry has been primarily focused on mutual funds rather thanon their management companies. For that reason, this book aimsto fill this gap in the empirical knowledge of the efficiencyobtained by mutual fund management companies. In this sense,within the practical part the book will contrast the efficiency levelachieved by companies in the management and distribution ofmutual funds to establish their core competence. The book ana-lyzes the potential interaction between this core competency andthe overall efficiency results obtained by the fund company. Inthe empirical sections, the book shows research that analyzes theSpanish case, which is one of the most relevant fund industries inthe Euro market. Therefore, to better understand the rationalebehind the mutual fund industry, this study offers new insightsinto issues that have not been studied before in the Europeanmarket.

This book develops an innovative model that considers thedifferent management stages of mutual fund families, also knownas mutual fund companies in the context of study, thereby over-coming the traditional dispute between the different approachesused in banking and insurance research. However, the variablesconsidered in this model are unique and specific to mutual fund

xxi

companies rather than a mere replication of the variables used inthe banking and insurance literature.

Data Envelopment Analysis (DEA) is the methodology usedto evaluate this model. This frontier method is one of the mostwidely used methods in studies of technical efficiency, but thespecific technique used in this study is one of the most recentinnovations in this field and overcomes the problem of the highmarket concentration in the Spanish mutual fund industry.Specifically, this book is one of the first empirical applicationsof slacks-based measure (SBM) variations, which representsan important and recent contribution to DEA methodology.Additionally, the persistence analysis of the efficiency achieved bythe different management stages considered in the fund companymodel is revealed to be an original contribution. Moreover, non-parametric statistics based on contingency tables have been devel-oped based on an innovative cluster analysis of efficiency scores.

The book consists of two parts. The first part initiallyincludes the basic concepts, a brief explanation of the basic DEAmodels, and a review of the most important applications to finan-cial institutions. We then formulate the multi-management stagemodel and the unique set of fund industry-specific variables usedin this research. Subsequently, the efficiency of Spanish mutualfund families is evaluated using the SBM approach. Finally, thispart concludes and summarizes the major findings of the effi-ciency analysis.

The second part of the book reviews the major concepts ofthe variations in the original SBM approach and the more accu-rate empirical results of these SBM variants in the Spanishmutual fund industry. Subsequently, this part illustrates the per-sistence phenomenon to further determine whether any relevantfactors may drive the efficiency results previously obtained.Finally, the second part presents conclusions and a summary ofthe major findings.

xxii PREFACE

Acknowledgments

Thanks to my mentors, Jose Luis Sarto and Luis Vicente,for their constant support, enthusiasm, dedication, andvaluable contributions. I also want to thank Laura

Andreu and Cristina Ortiz for their important assistance in thesecond part of the study.

I am grateful to the University of Zaragoza, the Accountingand Finance Department, for its financial support, and theSantander Bank for mobility grants to Latin American studentsin doctoral programs during the period 2010�2013. Finally, Iwant to thank the Universidad EIA for the great opportunity itgave me to continue this project.

xxiii

PART IThe Efficiency of Mutual FundFamilies: Using a Slacks-based

Measure

CHAPTER

1 Introduction to Part I

From the early 1990s through the middle of 2007, themutual fund industry saw remarkable growth worldwide,reaching a maximum of 26.13 trillion U.S. dollars in

December 2007 (source: European Fund and Asset ManagementAssociation, EFAMA). This successful expansion involved theproliferation of Mutual Fund Management Companies and theinception of a large number of mutual funds, both of whichplaced greater demands on the business structure of the industry.Then, when the markets experienced the subprime mortgagefinancial crisis, the mutual fund industry suffered a significantdecline in the total assets under management, especially in 2008,primarily driven by a fall in financial investments by householdsdue to an increase in risk aversion sparked by the high marketvolatility and a drop in confidence in financial instruments as aresult of the crisis.1 This market picture stopped the creation ofnew management companies and, in some cases, led to themerger or closure of other firms, thereby rearranging the compet-itive map of the industry. This drop in assets under managementand the number of investors is explained not only by the poorimage of mutual funds but also by a lack of confidence in theglobal markets where these funds allocate their primary

1Mutual fund net assets worldwide decreased by approximately 27%in 2008. Europe and America experienced quite similar trends, whileAsia and the Pacific showed even more dramatic decreasing rates.Source: EFAMA, International Statistical Releases.

3

investment positions. Unfortunately, despite all financial adviceabout diversification, equity investors had no place to hide in thiscrisis (Aït-Sahalia, Andritzky, Jobst, Nowak, & Tamirisa, 2012;Bartram & Bodnar, 2009). In many cases, this situation forcedfinancial institutions to rethink their production structures andto create new products not only on their own initiative butalso driven by new legislative and regulatory requirementsimposed by supervisory authorities who were trying to restoreconfidence in financial markets (Bernanke, 2009; Kirkpatrick,2009; McCauley, McGuire, & Von Peter, 2012).

More recently, mutual fund assets worldwide have beenslowly returning to the successful figures achieved before the cri-sis, i.e., 28.4 trillion U.S. dollars managed in September 2012,representing over one-third of the world GDP. This industryemploys skilled labor, has spillover effects on other sectors andtax returns, and provides important liquidity to the financial sys-tem and wealth for retail and institutional investors. The analysisof the efficiency of mutual fund industry is quite similar to theextensive literature addressed in other financial sectors such asbanking and insurance. If banks and insurance companies aremore productive, then these financial sectors should obtain betterperformance, thereby offering new and safer products to theircustomers at lower prices. According to this argument, MutualFund Management Companies should work to offer a wide rangeof top-performing funds with diverse investment characteristicsfor different types of investors while keeping fees and expenses aslow as possible.2 The expansion of these appropriate standardsof management should result in higher levels of overall efficiencyin the mutual fund industry. It is worth noting that mutual fundactivities reduce the exposure of banks to financial-services indus-try risk and increase scale economies and bank profitability,thereby improving the operating performance of banks (Asaftei,2008; Gallo, Apilado, & Kolari, 1996). In addition, the analysisof productivity differences across these companies in recentyears could make it possible to identify the success or failure of

2Referring to this last subject, there has been a gradual reduction in theannual management fees of funds since the mid-1990s, leading to amajor impact on fund-manager income during the last financial crisisdue to the drastic fall in assets managed. For instance, in the Spanishmarket, the management fees dropped from 1.65% in 1994 to 0.83%in 2009. Source: INVERCO, Impact of Subprime crisis in the IIC, June2010.

4 THE EFFICIENCY OF MUTUAL FUND FAMILIES

management initiatives and might also highlight the differentstrategies undertaken by the companies during the financialcrisis.

While extensive research has been devoted to productivity infinancial institutions, as far as we know, only Zhao and Yue(2010) and Medeiros (2010) have studied the efficiency of mutualfund companies and pension fund companies, respectively. Onone hand, Zhao and Yue (2010) examine the core efficiencyof Chinese fund firms by analyzing both the investment/researchand the marketing/service subsystems. On the other hand,Medeiros (2010) analyzes the changes in total productivity of asample of Portuguese pension fund companies from 1994 to2007 by means of a DEA-Malmquist index. A potential explana-tion for this scarce literature may be the difficulty of identifyingspecific variables for the appropriate evaluation of these compa-nies without merely replicating the previous studies focused onfinancial institutions, such as banks and insurance companies. Todevelop appropriate evaluation models for mutual fund compa-nies, it should be desirable to have a range of possibilities forspecific industry variables that would complement those modelsanalyzed in other financial sectors. Therefore, these fund indus-try-specific proposals should contain relevant managementinputs/outputs for mutual fund companies instead of merelyusing the general approach previously considered in banking andinsurance.

This study fills this gap in the financial literature and aims toshed additional light by analyzing the efficiency of mutual fundcompanies in Spain, which is one of the most relevant fund indus-tries in the Euro market. The significant market concentration ofSpanish mutual fund companies is the most challenging featurewhen obtaining an appropriate evaluation for this industry.That is, the coexistence of a few, very large, and well-diversifiedMutual Fund Management Companies together with a hugenumber of small managers specialized in fund strategy per sectorand/or geographical area makes it difficult to obtain appropriateevaluations of the industry as a whole because of the strikingdifferences between competitors. Therefore, a question arises asto the selection of an accurate methodology and managementvariables to appropriately analyze so heterogeneous a set ofSpanish mutual fund companies.

To conduct this analysis, we apply Data EnvelopmentAnalysis (DEA), which has been one of the most popular meth-ods over the past decades for evaluating efficiency in the financial

Introduction to Part I 5

industry (e.g., Berg, Førsund, & Jansen, 1991; Berg, Førsund,Hjalmarsson, & Suominen, 1993; Casu, Girardone, & Molyneux,2004; Cummins & Xie, 2008; Cummins, Rubio-Misas, & Zi,2004; Cummins, Weiss, Xie, & Zi, 2010; Holod & Lewis, 2011;Mlima & Hjalmarsson, 2002; Schaffnit, Rosen, & Paradi,1997).3 and we examine the performance of institutional portfo-lios as an alternative approach to the traditional performancemeasures; portfolio performance works with the functional rela-tionships between return and risk associated with behavioralassumptions (e.g., Basso & Funari, 2001; Eling, 2006; Gregoriou,Sedzro, & Zhu, 2005; Lozano & Gutiérrez, 2008a, 2008b;Murthi, Choi, & Desai, 1997).

This first part of study employs an original model and aunique set of fund industry-specific variables that complementthe traditional models in banking and insurance, thereby allow-ing for an accurate and comprehensive evaluation of the overallefficiency of Mutual Fund Management Companies.

We used this innovative approach to address a number ofquestions regarding the efficiency of Mutual Fund ManagementCompanies in the Spanish market and to further discuss theimplications of the results obtained in this analysis. What are thekey management stages driving the efficiency of a mutual fundcompany? Is efficiency robust across the different managementstages within a mutual fund company? How does scale affect theefficiency results?

This first part is organized as follows: Section 2 provides areview of the early DEA literature, a brief explanation of thebasic models, a short review of major contributions to efficiencyin financial institutions, and a discussion of the two most popularapproaches. Section 3 describes the proposed theoretical modeland the variables used in our analysis. Section 4 illustrates thedata, the empirical analysis and results, the influence of the vari-able-returns-to-scale, and robustness analyses. Finally, Section 5concludes and summarizes the primary findings.

3Some other studies are Drake and Howcroft (1994), Yeh (1996),Thompson, Brinkmann, Dharmapala, Gonzalez-Lima, and Thrall(1997), Athanassopoulos (1997), Sherman and Rupert (2006), Tortosa-Ausina, Grifell-Tatje, Armero, and Conesa (2008), Chen, Gregoriou,and Rouah (2009), Eling and Luhnen (2010).

6 THE EFFICIENCY OF MUTUAL FUND FAMILIES

The Efficiency of Mutual Fund Families: Applications to ... · he mutual fund industry has...

Documents

Transcript of The Efficiency of Mutual Fund Families: Applications to ... · he mutual fund industry has...