mutual fund in world

Mutual funds are gaining importance worldwide and the mutual fund industry has registered a spectacular growth in the entire world. The industry has doubled its size from 9.6 trillion in 1998 to 18 trillion in 2005, corresponding to an annual growth rate of 9% according to data of the Investment Company Institute1 (ICI). Although the largest industry is still located in the USA, European and Asian countries are closing the gap. The growth rate of the European (12%) and Asian (10%) mutual fund industry in the same period has surpassed that of the USA (7 % ). Despite its importance, the mutual fund industry outside the USA has received little academic attention. This study intends to fill this gap. Using a new database, this paper compares the mutual fund industry worldwide. The first part of the research uses unique features of the database and characterises the structure of the world mutual fund industry assessing aspects such as: age, asset shares, charges, industry concentration

to entry are positively associated with a larger industry, and concomitantly with more efficiency in terms of returns and fees.

1 Introduction

Mutual funds are gaining importance worldwide and the mutual fund industry hasregistered a spectacular growth in the entire world The industry has doubled its sizefrom $9.6 trillion in 1998 to $18 trillion in 2005, corresponding to an annual growth

largest industry is still located in the USA, European and Asian countries are closingthe gap The growth rate of the European (12%) and Asian (10%) mutual fund industry

in the same period has surpassed that of the USA (7%)

Despite its importance, the mutual fund industry outside the USA has received littleacademic attention This study intends to fill this gap Using a new database, thispaper compares the mutual fund industry worldwide The first part of the research usesunique features of the database and characterises the structure of the world mutual fundindustry assessing aspects such as: age, asset shares, charges, industry concentration

I would like to thank to an anonymous referee, John Doukas (the editor), Miguel Ferreiraand Ant´onio Freitas Miguel for helpful comments All errors are of my responsibility.Financial support from Fundacao para a Ciencia e Tecnologia is greatly acknowledged(POCI/EGE/57002/2004)

and competition, among others The second part tries to understand the explanation forthe cross-country differences of the world mutual fund industry.

This research draws on two major groups of previous studies on the mutual fundindustry: first, studies that address individual countries (McDonald (1973) and Dermineand Roller (1992) for France, and Ward and Saunders (1976) for the UK), second, studiesthat have already addressed international differences in the industry For instance, Ottenand Schweitzer (2002) compare the industry in the USA and five European countries in

the late nineties Klapper et al (2004) study the determinants of mutual fund growth in the period 1992–98 for 40 developed and developing countries Khorana et al (2005)

analyse cross-country differences in the relative size of the mutual fund industry in asample of 56 countries They find that the mutual fund industry is larger in developedcountries and countries with a strong institutional environment The industry is smaller

in countries where entry barriers are higher, measured by the effort to set up a new fund.The fund industry is larger in countries with a more educated population, and where theindustry itself is older

This study departs from previous ones by using a novel database of more than 50,000mutual funds around the world Although the data does not cover a large number of

countries as other studies (e.g Khorana et al., 2005), it provides a deeper look into the

industry, offering new insights into the industry structure and documenting importantcountry and regional characteristics

The main results are as follows: maturity (age) is an important factor explainingindustry development, measured by the ratio of the mutual fund industry to GDP or thevalue of mutual funds per capita The investment industry needs large and liquid financial

markets (Klapper et al., 2004), which in turn need time to grow as they benefit from

positive network externalities However, cross-country differences in fund availability orwith the weight of the industry to capital markets do not show association with age Thisweak association with age is likely to come from the fact that in most countries, financialreforms fostering capital market and mutual fund industry development happened at thesame time; therefore, when adjusting the development of the mutual fund industry forcapital market development, the relation with age vanishes, as both are age dependent.More developed industries are characterised by a higher share of equity mutual funds.They also show less home bias, which might imply that larger industries enjoy betterthe benefits of international diversification and those funds achieve a better risk-returntrade-off

Countries where the industry is larger are associated with lower annual charges,consistent with mutual fund economies of scale However, more developed industriesseem to impose larger initial and/or redemption charges This finding suggests thatindustries in an early stage of development have ‘fewer entry barriers’ in order toattract investors to the industry and by dissuading investors from redeeming shares, i.e

‘increasing barriers to departure’, a mutual fund is able to invest in a more risky portfoliothus enhancing performance

More developed industries show a large market share of funds-of-funds, index funds,institutional funds, bond funds as well as new products such as ethical funds, and asmall share of close-end funds

The results highlight strong regional differences Europe presents higher levels ofindustry concentration than Asia and the USA In spite of concentration, Europe has thelargest number of competing firms per million population, 5.2 against 1.95 in Asia and2.19 in the USA, and a high level of industry contestability: the highest ratio of new fundsand new companies per million population, which may be a result of the Investment

Services Direct that allows European Union (EU) companies to have an EU ‘passport’and making the process of registering in a new country easier Notwithstanding, thehigh level of competition is not reflected in charges, as the annual average mutual fundcharge in European and Asian countries is twice that of the USA The gap in feesmay be explainable by lower average size of funds, larger distribution costs and less

‘investor-driven’ competition

Concerning industry determinants, both the mutual fund industry and the sector ofequity mutual funds are larger in countries with a wealthier population and where theindustry is older Similar to other studies, legal factors like investor protection andinsiders trading laws are important; however, industry age and country wealth havegreater explanatory power In addition, there is a strong association between industrycompetition and industry development Larger industries have a larger number ofcompeting companies and more contestability Therefore, the existence of entry barriers

in the industry seems to hinder industry development

Overall, the evidence suggests that investors are better off in more developed industries

as they provide a better risk-return trade-off Mainly funds in more developed industriescharge lower annual charges due to economies of scale and are likely to enjoy thegains of international diversification The results have important policy implications byemphasising the role of competition and contestability in industry development Fewerentry barriers are positively associated with a larger industry, and concomitantly withmore efficiency for investors in terms of returns and fees

The paper is organised as follows Section 2 describes the data and the main indicators

of this study Section 3 characterises the features of the world mutual fund industry.Section 4 describes the potential determinants of the industry’s development Section 5analyses cross-country differences in the industry and their determinants Section 6concludes

2 Data and Summary Statistics

Data is drawn from Lipper Hindsight, an extensive database of more than 80,000 shore and off-shore mutual funds To characterise the mutual fund industry around theworld, I use information about more than 50,000 funds of 20 countries The samplecountries are: Austria, Belgium, China, Finland, France, Germany, India, Italy, Japan,(South) Korea, Malaysia, Poland, Portugal, Singapore, Spain, Switzerland, Taiwan,Thailand, the UK and the USA) The information collected is from the year 2005.Individual funds are then aggregated by domicile country in order to give the size

on-of the national mutual fund industry Some caveats are in order on this issue First, themutual fund industry has global scope and many management companies sell funds inseveral countries It is also quite common that companies have funds domiciled abroad

as off-shores Second, investors’ holdings in a country can be larger than the size of thenational mutual fund industry since investors can hold mutual funds domiciled in other

countries Therefore, the definition of size of the mutual fund industry adopted in this

study refers to funds domiciled in the country, not to the assets managed by firms in the

Table 1 describes the main features of this sample, like the aggregated value of theindustry and the number of funds in each country The total size of the mutual fund

industry is the sum of the net assets of all open-end funds domiciled in the country.

industry is larger in the USA, accounting for 65% of the value of assets, followed byFrance that represents 9% The European industry represents around 28% of the value

of the assets, while Asia only 6% The mutual fund industry is smaller in countrieslike Malaysia, Thailand and Poland Results are consistent with previous studies thatfind that the European mutual fund industry is lagging behind the American (Otten andSchweitzer, 2002)

Lipper distinguishes between primary funds and sub-funds This distinction arisesbecause many countries allow funds to break up into several classes Usually funds thatoffer multiple share classes represent ownership interests in the same portfolio, but have

a different fee structure The total number is 52,286 funds of which 33,182 funds areprimary Again, the USA has the largest number of funds, 35% of the total number offunds (20% of primary), followed by the UK and Korea, while Poland has the smallestnumber of funds European countries account for 40% of the total number of funds.Previous studies have already reported that the number of funds was much larger inEurope than in the USA in 1997, which results in a smaller average size of the funds inEurope (see e.g Otten and Schweitzer, 2002; Otten and Bams, 2002)

Table 1 also reports the number and market value of Undertakings for CollectiveInvestment in Tradable Securities (UCITS) The UCITS are collective funds that can besold across national boundaries within the European Union (EU) in accordance with the

‘Undertakings for Collective Investment in Tradable Securities’ Directive These fundscan be marketed within all countries that are a part of the European Union, provided thatthe fund and fund managers are registered within the domestic country The majority offunds domiciled in EU countries are also UCITS and account for 77% of the value ofthe EU funds Only Poland has a small number of UCITS, explained by their late entryinto EU

3 Characteristics of the Mutual Fund Industry Worldwide

This section explores information on industry age, asset structure, charges, geographicfocus, industry concentration and competition and fund strategies of the mutual fundindustry

Development of the mutual fund industry

Industry size The figures presented in Table 1 need to be compared with the wealth and

financial development of the countries The development of the mutual fund industry ismeasured using the following indicators: the ratio of the size of mutual fund industry to

the Gross Domestic Product (MF/GDP), which takes into account the total wealth of the country; the mutual fund value per capita (MFpc), which considers the dollar amount

that each individual invests on average in the domestic industry; the ratio of the size of

the mutual fund industry to the size of bond and stock markets (MF/SBM) that compares

the industry with the size of the securities markets, adjusting for the general development

funds provided by the Investment Company Institute (see www.ici.org) The total net assetvalue in our sample is $11.5 trillion, which compares with $13.1 trillion ICI for the countriesincluded in our sample

Table 1Mutual fund industry around the world: descriptive statistics.

This table presents the sample main descriptive statistics on (1) the industry value, which equals the total net asset value of all open-end funds (in USD million), (2) the number of funds, distinguishing between primary and non-primary funds (Total refers to sum of primary and non-primary funds), (3) the Undertakings for Collective Investment in Tradable Securities (UCITS) presenting the number and total value of funds (in USD million) The latter information refers only to EU countries The last rows present descriptive statistics Data is from Lipper Hindsight.

of financial markets and the number of funds per million inhabitants (N Funds/Pop),

that compares fund supply and fund diversity in a country

Table 2 provides an overview of the development of the mutual fund industry around

the world An analysis of the importance of the mutual fund industry in GDP or MFpc

reveals that it is most developed in the USA but is at an earlier stage in China In Europeancountries like France, Austria and Switzerland, the mutual fund industry represents alarge proportion of GDP and presents a large per capita value In Asia, the countrieswith a more developed industry are Singapore and Korea

By weighting the size of the mutual fund industry against the size of securities markets,

a correction is made for countries whose securities markets are small in relation to GDP.From this point of view, Austria has the largest industry and Malaysia the smallest.Note that a large value of this ratio might be indicative of a high level of investment in

Table 2The development of the mutual fund industry around the world.

This table presents descriptive statistics on indicators of the development mutual fund industry: (1)

MF/GDP – The ratio of the value of the mutual fund industry to the GDP (in %); (2) MFpc – The

mutual fund value per capita (in USD); (3) MF/SBM – The ratio of the value of the mutual fund industry to the value of bond and stock markets (in %) and (4) N Funds/Pop – The total number funds

per million inhabitants The last rows present descriptive statistics For two countries there is no data

on the size of stock and bond markets Data sources are in Appendix A.

The mutual fund industry is also compared with the development of financialmarkets Indicators are Private Credit by Deposit Money Banks and Other Financial

Institutions to GDP (Private Credit/GDP), the ratio of Stock Market Capitalisation to GDP (SMC/GDP), the ratio of Private Bond Market Capitalisation to the GDP (Private Bond/GDP) and the ratio of Public Bond Market Capitalisation to the GDP (Public

Bond/GDP) and stock and bond market capitalisation to GDP (SBM/GDP) All ratios

are from Beck et al (2000) database on international development of financial markets.

Table 3The development of the world mutual fund industry and financial markets.

This table presents the correlation between the indicators of the development of mutual fund industry

and the development of financial markets Columns are: (1) MF/GDP – The ratio of the value of the mutual fund industry to the GDP (in %); (2) MFpc – The mutual fund value per capita (in USD); (3)

MF/SBM – The ratio of the value of the mutual fund industry to the value of bond and stock markets

(in %) and (4) N Funds/Pop -The total number funds per million inhabitants Rows are: Private credit by deposit money banks and other financial institutions to the GDP (Private Credit/GDP), the ratio of stock market capitalisation to the GDP (SMC/GDP); the ratio of private bond market capitalisation to the GDP (Private Bond/GDP); the ratio of public bond market capitalisation to the GDP (Public Bond/GDP) and the ratio of stock and bond market capitalisation to GDP (SBM/GDP).

Data sources are in Appendix A.

Table 3 reports the correlation between mutual fund industry and the financial

indicators Correlations show that larger values of MF/GDP and MFpc are associated

with credit and securities markets development Thus, countries where credit marketsand stock markets are more developed also have a more developed mutual fund industry

Although positive, correlation is lower for the indicators MF/SBM and N Funds/Pop,

indicating that cross-country differences in those indicators are less likely to be related

to cross-country differences on the general development of financial markets

Asset structure

Mutual funds are also categorised according to the asset class Categories include bond,equity, mixed assets, money market, real estate and other mutual funds Asset sharesare computed aggregating funds by each asset type in each country Table 4 presentsshares of asset classes in the country industries The distribution of asset types withincountries varies greatly; however, on average bond and equity funds are always the mostimportant classes of assets

Bond mutual funds have an average weight of 27% in our sample Investors in Thailandand Taiwan seem to prefer these funds as bond funds present market shares of 70%,which is well above the average, while in China and Korea the average weight is smaller

at below 7%

Equity mutual funds have an average weight of 28% In the UK and the USA, equityfunds represent more than 50% of the industry assets, while in China and Portugal they

Table 4Weights of asset classes in the mutual fund industry.

This table presents the relative weight of each asset class in a country’s mutual fund industry in Panel A Panel B reports the descriptive statistics for the sample and Panel C reports the correlations

with the indicators of the development of the mutual fund industry: (1) MF/GDP – The ratio of the value of the mutual fund industry to the GDP (in %); (2) MFpc – The mutual fund value per capita (in USD); (3) MF/SBM – The ratio of the value of the mutual fund industry to the value of bond and stock markets (in %) and (4) N Funds/Pop -The total number funds per million inhabitants Data is

from Lipper Hindsight.

have an average weight of 5% and 7% Otten and Schweitzer (2002) also report that inAnglo-Saxon countries equity funds tend to dominate the market, reporting a figure of

with 28% and 27% of European and Asian countries respectively

Funds that mix several kinds of assets have an average weight of 16% But this figure

is higher in countries like Poland (30%) and smaller in Japan, Germany, the USA,Taiwan, and India (less than 10%) Money market funds represent a share of 18% onaverage in a country’s mutual fund industry, and are very important in China and Francewith market shares around 50% In contrast, the importance of money market mutualfunds is very small in Japan, Taiwan, the UK and Singapore, with less than 3% Realestate mutual funds have an average weight of 3% although the weight is heavier incountries like Germany, Portugal and Singapore

Finally, Panel C analyses the correlation with industry development indicators Moredeveloped industries tend to have a higher share of equity mutual funds and a lowershare of bond and mixed assets funds

Geographic focus

Lipper identifies the geographic focus of the mutual fund, which can be a country, aregion or even the ‘world’ like global funds Table 5 reports the percentage of net assetsinvested geographically The table is divided into two panels, one reporting when thegeographic focus is an individual country (Panel A) and the second when it is a region(Panel B) Columns add up to 100%

Funds in China and India focus exclusively on their home country Funds in countrieslike Korea, Malaysia, Poland, Taiwan and Thailand invest more than 80% of their assets

in their home market Many companies cannot have large holdings in financial assetsabroad or are even constrained by their governments from investing outside the country

On the contrary, the industry in Austria, Belgium, Finland, France, Germany, Italy andSingapore invest only a small fraction of their assets in the home country

Except for countries like Poland, the UK and Switzerland, European funds show alow level of home bias, which maybe explained by the fact they share the euro currency

On the other hand, Asian countries have a high level of home bias, the exception areeconomically open countries such as Singapore and Japan The USA also has a lowlevel of home bias

Panel B highlights regional preferences For instance, USA mutual funds invest 38%domestically and more than 50% in North America Funds in European countries tend

to invest substantially in Europe, except for Poland, the UK and Switzerland where, aspreviously reported, they tend to invest domestically In Asia, only Singapore diversifiessubstantially in other Asian countries

Industries located in small countries like Austria, Belgium or Singapore makesubstantial investments globally but Japan also invests substantially in global funds.The correlation between the degree of home bias and the diversification in ownregion with the development of the industry (not reported in the table) shows that less

different equity culture, a strong presence of banks and a different pension system

more developed industries have a higher geographical focus in their own region (averagecorrelation 0.51).

Overall, the analysis shows that more developed industries tend to diversify moreinternationally, although they also show a preference for investing in their own region;this might imply that funds from more developed industries are likely to enjoy morethe benefits of international diversification and offer investors a better risk-return trade-off

Fund strategies

Lipper also assigns a strategy to each fund This subsection analyses the informationabout fund strategies such as closed-end funds, bond funds, ETFs, futures and options,institutional and insurance funds, ethical funds, funds-of-funds and index tracking funds.Table 6 reports market shares of these strategies in countries and shows that countriespresent great heterogeneity on this issue

Closed-end funds are those with a limited number of units/shares on issue and aninvestor wishing to vacate the fund will sell them to another investor in a manner similar

to the sale of a share of a corporation Closed-end funds have an important market share

in countries like Thailand, the USA, China and Singapore A regional analysis revealsthat closed-end funds are mainly present in Asian countries where the average share is11.8%, while in the USA and Europe it is 3.1% and 2%, respectively

Mutual funds whose main strategy is investing in corporate bonds are important inAustria, Spain and the UK, whereas those that invest in government bonds are important

in Austria, Italy, Korea and Switzerland Overall, European countries reveal a preferencefor bond funds They represent an average share of 8.6% against 4.1% in the USA and3.3% in Asia The preference of continental European countries for bond funds hasalready been reported in a previous industry analysis (see Otten and Schweitzer, 2002)

In many countries institutional investors use mutual funds extensively Institutionalfunds have the same structure as other collective investment schemes but they areestablished for the use of institutional investors and usually require higher minimuminvestment Institutional mutual funds are major products in the USA, Switzerland,France, India and the UK

In the last decades, new products like ethical funds and ETFs have emerged Ethicalfunds are important in Belgium and the UK ETFs are well developed in countries likeJapan, the USA, Singapore and Germany Mutual funds that invest in futures and optionshave a significant market share in Spain Overall, these funds tend to be present in moredeveloped industries but still have lower market shares

Funds-of-funds are mutual funds that invest in other mutual funds The fund mayinvest in a mutual fund run by the same group or by an unrelated company Externalfunds-of-funds are important in Austria, Spain and Japan, while internal funds-of-fundshave a considerable weight in Austria, Belgium and Singapore

Despite the low level of administration costs of index-tracking funds, they are still not

so popular as actively managed funds (Gruber, 1996) Index tracking funds are important

in Japan, Switzerland and the USA

Panel B reports the association between the development of the mutual fund industryand attributes of funds Less developed industries tend to have a higher market share

of closed-end funds, while more developed industries are more associated with highermarket shares of institutional funds, index-tracking funds, funds-of-funds, and bondmutual funds

Table 7Average age of mutual fund industry.

This table presents the average age of funds in the industry reported to the end of 2005 and the launch date of the oldest fund in a country Panel B reports the descriptive statistics for the sample and Panel C reports the correlations with the indicators of the development of the mutual fund industry:

(1) MF/GDP- The ratio of the value of the mutual fund industry to the GDP (in %); (2) MFpc – The mutual fund value per capita (in USD); (3) MF/SBM- The ratio of the value of the mutual fund industry to the value of bond and stock markets (in %) and (4) N Funds/Pop -The total number funds

per million inhabitants Data is from Lipper Hindsight.

Average age of funds (years) Launch date of oldest fund in the country

Other characteristics of funds

Launch date Each fund has a launch date in the database Table 7 displays the average

age of the existing funds in a country as well as information on the country’s oldest fund

in the database

Funds around the world have an average age of 5.8 years (if reporting to the end of2005) China has the youngest industry at 1.49 years, while Germany’s industry is theoldest with an average of 9.34 years In France, Italy, Portugal, the UK and the USA,the average age of the industry is over seven years, well above the median; this mightindicate that the industry started earlier, or that funds last for a long time, or even that themutual fund industry allows few new entries Korea has a young mutual fund industry,

an average age of almost two years Overall, the industry is younger in Asia, followed

by Europe and the USA

The last column presents the launch date of the oldest fund in the sample Interestingly,funds launched in 1923 in the USA, in 1941 in the UK or 1948 in Switzerland still exist

at the end of 2005

The last rows report the relation between industry age and industry size Age is

positively associated with the MF/GDP and MFpc, indicating that older industries are

also larger ones This is due to the fact that financial markets need time to developdue to network externalities However, the association between fund availability and theratio of mutual fund industry to stock and bond market capitalisation and age is weaker.Reforms on the financial system boosting the development of securities market and thedevelopment of the mutual fund industry tend be simultaneously Therefore it is likelythat when adjusting the development of the mutual fund industry for capital marketsdevelopment, the age dependence tends to disappear

Mutual fund charges Fund companies charge investors fees in order to compensate

for the management costs of the fund (annual charges) In addition, mutual funds oftencharge a sales surcharge when the investors purchase (initial charges) or sell shares(redemption charges)

Table 8 aggregates this information by country and presents country average chargesfor each category Annual charges range from 0.54% in the USA to 1.77% in Poland.Charges are higher than the average in Finland, France, Italy, Portugal, Spain, the UK,China, Japan and Malaysia Initial charges are more varied across countries The USAcharges 3.91% on average while the minimum is 0.01% in Spain The UK, Germanyand Singapore also have large initial charges, while China and Korea have low initialcharges Redemption charges are higher in the USA, well above the great majority of

countries that have values close to zero In accordance with Khorana et al (2006), total

charges are defined as the sum of annual charges plus initial and redemption chargesdivided by five Total charges are higher in Poland, around 1.9%, and 0.86% lower inKorea

The correlation analysis with the indicators of industry size leads to interestingconclusions First, larger industries tend to have lower annual charges consistent with

evidence on mutual fund economies of scale (see e.g Baumol et al., 1990; Barber et al., 2005; Khorana et al 2006) However, more developed industries tend to impose larger

initial charges This finding suggests that industries in an early stage of developmenthave ‘fewer entry barriers’ to encourage investors, as investors are more sensitive to

salient ‘in-your-face fees’ like front-end loads than to operating expenses (Barber et al.,

find a negative relation between fund flows and front-end loads and no relation betweenoperating expenses and fund flows

Table 8Average charges of the mutual fund industry.

This table presents average charges of country industries: Annual charges, initial charges and redemption charges Total charges are annual charges plus initial and redemption charges divided

by five Data is in percentages Panel B reports the descriptive statistics for the sample and Panel

C reports the correlations with the indicators of the development of the mutual fund industry: (1)

MF/GDP- The ratio of the value of the mutual fund industry to the GDP (in %); (2) MFpc – The

mutual fund value per capita (in USD); (3) MF/SBM- The ratio of the value of the mutual fund industry

to the value of bond and stock markets (in %) and (4) N Funds/Pop – The total number funds per

million inhabitants Data is from Lipper Hindsight.

More developed industries seem also to charge higher redemption charges Chordia(1996) argues that investors who redeem fund shares impose externalities on thosewho do not, from liquidation of securities; this obliges the fund to have a higher cashposition reducing performance Thus, by making redemptions expensive, a mutual funddissuades investors from redeeming shares and it is able to invest in a more risky portfolio

to obtain higher performance Redemption charges work as a ‘barrier to departure’ fromthe industry This can be seen in the common practice of funds of charging a maximumvalue on redemption charges, i.e they are only levied if the investor holds the fund forless than an agreed period Given that in developed industries investors prefer long-terminvestments as equities funds, the real impact of redemption fees on net performance issmaller

Overall, the annual average charge in European and Asian countries is twice that of theUSA, 1.2%, 1.07% and 0.54%, respectively Several factors are likely to explain these

other hand, USA funds have significantly greater assets under management, thereforethey can more easily spread fixed expenses across a larger asset base For instance,USA mutual funds frequently employ ‘breakpoints’ on annual management fees set as

a function of fund size

Second, fund companies advocate that the rise in management fees is often due to theincrease on the cost of distribution Empirical work corroborates this argument; fundsdistributed simultaneously in several countries seem to present higher charges (Khorana

et al., 2006) Hence, EU UCITS distributed simultaneously in several countries are

likely to present higher fees

Third, the existence of different fund and investment cultures Fees in the USA areunder closer inspection from regulators and media Only recently have the EU authorities

‘investor-driven competition’ as investors can only exert fee pressure if they are fullyaware of their level For instance, the USA has registered an increase in the recent years

on demand for funds with lower fees Investors now hold most of their stock and bondfund assets in funds charging below-average operational and management expenses (ICI,2006a)

Previous studies report average charges in 1997 of 1.2% for a sample of Europeancountries and 1.4% for the USA (Otten and Schweitzer, 2002) Therefore, whilemanagement fees in the USA seem to have decreased substantially, the same trend

is not found in Europe (Finland is an exception according to Korkeamaki and Smythe(2004)) The trend of decreasing fees in the USA is confirmed in a recent analysis forthe USA reporting refer that mutual fund fees and expenses fell to their lowest levels inmore than a quarter century during 2005 (ICI, 2006b)

Industry concentration and competition This subsection analyses data on portfolio

manager companies To measure and compare the degree of concentration and tition in the industry, the following measures are used (a more detailed analysis can beseen in Ferreira and Ramos, 2006):

Funds (p 6)

• Average market share in a country (Av mk sh) In general, if a country has a large

number of companies, the average market share is smaller

The higher the market share of the largest companies, the greater the industryconcentration

variable indicates the number of available companies per capita (i.e., standardised bythe country population) The larger the number of companies per capita, the greaterthe expected level of competition

Entry barriers are also measured by the level of contestability The next indicatorsanalyse the entry of new funds and new companies

describes the degree of innovation in companies, which might be indicative of thelevel of fund competition

expresses how many of the existing funds are recent, which is indicative of industryturnover or age

more indicative of the level of industry contestability standardised by the countrypopulation

companies (New firms/N firms) expresses how many of the existing companies are

recent, which is indicative of industry turnover or age

ratio will be more indicative of the level of contestability of the industry

Table 9 reports concentration and competition indicators for portfolio manager nies There is a total of 2,031 portfolio manager companies and the average number ofportfolio companies per country is 102

compa-The average market share is larger in Poland, followed by Portugal, and smaller in theUSA Industry concentration is greater in Belgium, where the five largest companieshave 89% of market share On the other hand, the UK and the USA have the lowestindustry concentration with only 28.5% and 34%, respectively

Europe presents higher levels of concentration; the average CR5 is 65% as opposed to

Asia where it is 50% and the USA with 34% However, average market share is higherfor Asia 2.6% due perhaps to it being a younger industry, whereas it is 2.1% for Europeand 0.2% for the USA

Table 9 also reports measures of competition When considering portfolio ment companies per million population, Austria has the highest number, followed bySwitzerland, Singapore and Finland

manage-Despite concentration, the figure for competing firms in Europe is high (5.2);this contrasts with 1.95 in Asia and 2.19 in the USA and it may be an out-come of the Investment Services Direct that allows EU companies to have a

EU ‘passport’, making it easier to register with national financial supervisoryauthorities

The next indicators analyse industry contestability Countries like China, India, andPoland show a high percentage of new funds in actual existing funds because theseindustries are young But in general, the differences between countries are not relevant.The number of new funds per company is higher in Korea, Japan and Belgium and new