Swedish House of Finance Research Paper No 14-04
Swedish Equity Mutual Funds: Performance,
Persistence and Presence of Skill
Harry Flam
Stockholm University - Institute for International Economic Studies
Roine Vestman
Stockholm University and Swedish House of Finance
Swedish House of Finance (SHoF) is a research center focusing on financial markets, and is jointly supported by the Stockholm School of Economics,
SIFR, Vinnova, and the financial industry. The goal of SHoF is to produce and disseminate financial research through providing financial support,
organizing PhD courses, hosting a financial data center, and organizing seminars, conferences, and visitors programs for both academics and
practitioners.
!
Swedish(Equity(Mutual(Funds:((
Performance,(Persistence(and(Presence(of(Skill∗ (
!
Harry!Flam!
Institute!for!International!Economic!Studies,!Stockholm!University,!and!CESifo!
!
Roine!Vestman!
Department!of!Economics,!Stockholm!University,!and!SIFR!
!
!
February,!2014!
!
!
Abstract(
Actively!managed!Swedish!equity!mutual!funds!generated!an!average!positive!4Ifactor!alpha!of!
0.9!per!cent!per!year!before!expenses!and!a!negative!alpha!of!I0.5!per!cent!after!expenses!in!
1999I2009.!There!is!practically!no!persistence!in!returns.!When!funds!are!ranked!on!past!
performance,!their!returns!converge!to!the!mean!in!about!two!years.!There!is!furthermore!
practically!no!evidence!of!true!management!skill.!The!actual!4Ifactor!alphas!of!most!funds!
before!and!after!expenses,!including!those!with!the!highest!alphas,!do!not!differ!significantly!
from!bootstrapped!alphas!constructed!under!the!null!hypothesis!that!alpha!is!zero!for!all!funds.!!(
(
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
!!We!are!grateful!to!Hanna!Mühlrad!for!research!assistance.!!We!thank!the!Data!Center!at!the!Swedish!House!of!
Finance,!and!in!particular!Erik!Eklund,!for!the!provision!of!stock!market!and!company!data.!
!
!
∗
1!
!
(
1.!!Introduction!
Most!Swedes!own!shares!in!Swedish!equity!mutual!funds.!Vestman!(2013)!reports!a!stock!
market!participate!rate!outside!the!pension!system!equal!to!63!percent!at!the!household!level!
and!an!average!equity!fund!holding!of!SEK!79,000!(USD!12,000).!In!addition,!employers!make!
mandatory!contributions!to!the!funded!part!of!the!public!pension!system!amounting!to!2.5!per!
cent!of!gross!wages!and!wage!earners!can!choose!to!invest!the!contributions!in!more!than!800!
mutual!funds.!Based!on!historical!rates!of!return,!the!funded!part!of!the!public!pension!system!
may!come!to!contribute!as!much!to!wage!earners’!pensions!as!the!nonIfunded,!transfer!part!of!
the!system.!In!addition,!three!quarters!of!Swedish!households!make!voluntary!investments!in!
broad!Swedish!equity!mutual!funds.!Consequently,!the!choice!between!different!equity!mutual!
funds!may!be!quite!important!for!most!Swedes.!
!
Most!investors!in!equity!mutual!funds!presumably!have!little!or!no!knowledge!about!the!
equities!that!funds!hold!and!the!corporations!that!have!issued!them;!they!simply!want!to!have!
some!exposure!to!the!stock!market.!When!the!financial!industry!or!various!media!advise!
investors!on!the!choice!of!funds,!they!frequently!recommend!actively!managed!funds!with!
superior!past!performance,!based!on!the!belief!that!superior!performance!can!be!attributed!to!
stock!picking!skill!and!that!such!skill!is!persistent.!In!contrast,!when!a!financial!economist!in!
academia!is!asked!for!advice,!he!or!she!is!likely!to!recommend!a!lowIcost!index!fund,!based!on!
the!assumption!that!actively!and!passively!managed!funds!can!be!expected!to!earn!the!same!
return!as!the!stock!market!before!expenses,!but!the!latter!are!likely!to!do!so!at!a!lower!cost.!!
!
Most!of!the!empirical!evidence!based!on!data!on!the!U.S.!mutual!fund!market!support!
the!efficient!market!hypothesis.!Carhart!(1997)!finds!that!most!funds!underperform!by!about!
the!magnitude!of!their!investment!expenses,!Kosowski!et!al!(2006)!that!the!average!net!riskI
adjusted!(excess)!return!per!year!is!I1.2!per!cent,!Fama!and!French!(2010)!that!it!is!–1!per!cent,!!
Barras!et!al!(2010)!!I0.5!per!cent,!and!Berk!and!Binsbergen!(2012)!!–0.7!per!cent.!!
Although!Sweden!has!among!the!highest!equity!fund!penetration!rates!in!the!world,!
there!is!very!little!evidence!on!the!performance!of!Swedish!equity!mutual!funds.!Only!two!
studies!have!been!published.!The!first,!by!Dahlquist!et!al.!(2000),!separates!equity!funds!
according!to!the!taxation!of!returns!to!investors.!The!study!finds!that!funds!without!preferential!
tax!treatment!have!positive!but!insignificant!average!and!median!excess!net!returns!of!0.5!and!
0.1!per!cent!per!year,!and!that!funds!with!preferential!tax!treatment!have!average!and!median!
net!excess!returns!of!–1!and!–0.7!per!cent!per.!!The!second!study,!by!Engström!(2004),!finds!an!
average!net!excess!return!of!1.7!per!cent.!!
2!
!
!
Our!study!covers!a!more!recent!and!longer!time!period.!We!find!average!equalI
weighted!gross!excess!returns!of!0.90!per!cent!per!year!and!net!excess!returns!of!I0.45!per!
cent.!This!seems!to!indicate!the!existence!of!some!skill!in!stock!picking,!but!also!that!fund!
managers!on!average!are!unable!to!compensate!investors!for!the!cost!of!management.!!
We!also!estimate!excess!returns!of!index!funds,!a!product!segment!that!is!still!small!but!
growing!in!Sweden.!We!find!that!index!funds!on!average!deliver!returns!equal!to!their!
respective!benchmarks!before!costs,!but!that!large!differences!exist!across!funds.!When!funds!
are!evaluated!against!the!whole!stock!market,!we!find!substantial!underperformance.!!The!
average!equalIweighted!gross!and!net!excess!returns!are!I0.92!and!I1.31!per!cent!per!year.!
Further,!the!median!index!fund!has!performed!worse!than!the!median!actively!managed!fund.!
This!underperformance!can!be!explained!by!the!fact!that!a!majority!of!index!funds!use!
benchmarks!for!a!subset!of!the!stock!market!consisting!of!the!largest!and!most!traded!
companies,!whose!performance!has!been!below!average.!!!
!The!finding!that!actively!managed!funds!have!higher!net!excess!returns!than!index!
funds!on!average!means!that!the!advice!to!investors!to!choose!a!lowIcost!index!fund!over!an!
actively!managed!fund!must!be!qualified.!The!index!fund!should!not!be!picked!at!random.!Even!
paying!attention!to!the!tracking!error!and!the!fee!is!not!sufficient!–!the!index!fund’s!benchmark!
index!also!matters.!!
!
Although!actively!managed!funds!do!not!earn!positive!net!excess!returns!on!average,!
some!do.!!This!is!commonly!taken!as!evidence!of!superior!stock!picking!skills.!However,!we!find!
little!evidence!of!persistence!in!returns!and!consequently!little!to!indicate!the!presence!of!skill.!
Regardless!of!whether!funds!rank!high!or!low!based!on!past!performance,!their!returns!
converge!very!quickly!and!become!similar!after!two!or!three!years.!This!accords!well!with!the!
findings!of!Carhart!(1997)!for!U.S.!equity!mutual!funds,!except!that!he!finds!that!the!bottom!
decile!of!funds!underperforms!persistently.!
!
The!presence!of!persistently!high!returns!may!indicate!the!presence!of!skill,!but!it!can!
also!be!the!outcome!of!luck.!We!make!a!direct!investigation!of!whether!superior!and!inferior!
performance!should!be!attributed!to!superior!and!inferior!skill!or!to!good!and!bad!luck!by!
employing!a!version!of!the!bootstrap!method!in!Kosowski!et!al!(2006).!Simulated!crossI
sectional!distributions!of!gross!and!net!returns!are!created!repeatedly!under!the!assumption!
(null!hypothesis)!that!gross!and!net!excess!returns!are!zero.!Actual!gross!and!net!returns!are!
then!compared!to!the!simulated!average!gross!and!net!returns.!We!find!that!actual!gross!
returns!are!insignificantly!different!from!the!simulated!gross!returns,!except!for!approximately!
ten!percent!of!all!the!funds!(13!of!105!funds)!in!the!second!and!third!decile.!In!other!words,!
there!is!little!evidence!of!general!presence!of!skill.!It!follows!that!it!is!even!less!likely!that!some!
managers!possess!sufficient!skill!to!deliver!positive!net!excess!returns!to!investors.!This!is!
3!
!
indeed!the!case;!no!actual!net!excess!return!is!statistically!significantly!greater!than!the!
corresponding!simulated!return.!A!handful!of!funds!deliver!so!poor!net!excess!returns!that!it!is!
very!unlikely!that!bad!luck!alone!would!be!the!cause.!
!
!
2.!Data!sources!and!data!construction!!
Our!fund!universe!consists!of!115!actively!managed!mutual!funds!holding!a!broad!set!of!
equities!of!corporations!listed!on!the!Stockholm!Stock!Exchange,!plus!15!passively!managed!
funds!tracking!a!variety!of!indexes!related!to!the!same!exchange.!Specialized!funds,!for!example!
funds!that!invest!in!specified!industries!or!in!small!cap,!are!excluded.!It!should!be!noted!that!
some!funds!may!have!a!small!share!of!their!assets!in!equities!listed!abroad!and!that!funds!must!
hold!a!minimum!amount!of!liquidity!for!transaction!purposes;!the!amount!can!vary!over!time.!
Our!principal!period!of!investigation!is!January!1999!to!December!2009.!We!lack!data!to!
adjust!returns!for!size!and!value!bias,!as!proposed!by!Fama!and!French!(1993),!for!later!years,!
but!we!can!show!that!simple!betaIadjusted!excess!returns!change!little!when!the!period!of!
investigation!is!extended!to!August!2013.!The!minimum!number!of!consecutive!monthly!
returns!required!for!a!fund!to!be!included!is!36.2!Funds!that!meet!the!requirement!but!have!
been!closed!down!are!included!to!avoid!survival!bias.!!
!
!The!data!on!net!fund!returns!were!supplied!by!MoneyMate!for!the!period!1999I2009!
and!by!Morningstar!for!the!period!1999I2013,!both!of!whom!use!primary!data!supplied!by!the!
Swedish!Investment!Fund!Association.!The!fund!returns!include!reIinvested!dividends!and!are!
net!of!expenses.!Some!funds!are!sold!both!in!the!retail!market!and!in!the!market!for!various!
pension!plans.!In!such!cases,!only!the!data!pertaining!to!the!retail!market!are!included.!!
!!
Gross!returns!are!obtained!by!adding!back!costs!as!given!by!the!total!expense!ratio!
(TER)!to!net!returns.!TER!is!reported!annually!to!the!Swedish!Financial!Supervisory!Authority!by!
most!but!not!all!funds.3!We!have!obtained!TER!from!the!annual!financial!statements!for!105!of!
115!funds.4!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2
!We!exclude!28!actively!managed!funds!and!5!index!funds!that!have!35!or!fewer!return!observations.!
!We!note!that!the!Swedish!measure!of!total!expenses,!TKA,!is!more!inclusive.!In!addition!to!fees!and!trading!costs,!
it!also!includes!commissions!and!resultsIbased!fees!in!per!cent!of!the!average!value!of!assets!under!management!
during!the!year.!The!TER!for!the!105!funds!for!which!we!have!data!is!1.34!and!it!is!1.59!per!cent!for!the!TKA.!
3
4
!Our!sample!includes!some!funds!which!are!domiciled!abroad!and!that!therefore!do!not!report!TER!and!TKA.!Five!
funds!are!registered!in!Luxembourg,!one!in!Norway,!one!in!Ireland,!one!in!Great!Britain!and!two!have!no!domestic!
jurisdictition!(ISIN!code)!reported!in!the!Money!Mate!data!set.!
4!
!
Data!for!the!construction!of!systematic!risk!factors!–!share!returns,!market!capitalization!
and!book!values!–!were!kindly!delivered!by!the!Data!Center!at!the!Swedish!House!of!Finance.!
The!data!were!originally!constructed!by!SIX!Telekurs!before!2004!and!by!NASDAQ!OMX!for!the!
period!2004I2009.!We!construct!the!size!(SMB)!and!value!(HML)!factors!of!Fama!and!French!
(1993)!and!of!Carhart’s!(1997)!momentum!(MOM)!factor!as!follows.!We!first!take!account!of!
the!fact!that!Sweden!has!a!multiple!share!class!system.!For!each!corporation,!we!determine!the!
largest!share!class!in!terms!of!market!value!and!use!only!that!class.!We!then!attribute!the!entire!
market!value!of!the!corporation!to!this!class.!For!each!12Imonth!period!starting!in!April!of!year!
t"1!and!ending!with!March!of!year!t!we!require!12!recorded!monthly!observations.!We!then!
match!the!book!and!market!values!from!March!of!year!t!with!the!return!series!from!April!of!
year!t!to!March!of!year!t"1.!We!also!control!for!which!particular!stock!market!list!that!the!stock!
was!traded!on!in!December!of!year!t"1!by!excluding!stock!markets!that!are!judged!to!be!too!
small!and!!illiquid!to!be!suitable!as!investment!targets!for!a!mutual!fund.5!The!SMB!and!HML!
factors!were!then!constructed!as!in!Fama!and!French!(1993).!Unfortunately,!the!stock!data!that!
we!need!to!construct!the!SMB!and!HML!factors!are!available!only!until!2009.!The!MOM!factor!
was!constructed!by!first!sorting!all!stocks!on!their!12Imonth!lagged!returns,!and!then!using!the!
bottom!20!percent!to!construct!a!portfolio!to!go!short!in,!and!the!top!20!percent!to!construct!
another!portfolio!to!go!long!in.!!
!!
3.!Performance!
The!CAPM,!the!3Ifactor!model!by!Fama!and!French!(1993)!and!its!extension!to!four!factors!by!
Carhart!(1997)!are!consistent!with!models!of!market!equilibrium!with!one,!three!or!four!
systematic!risk!factors.!They!can!also!be!interpreted!as!models!for!performance!attribution;!!we!
will!use!them!as!such.!!
The!following!time!series!regressions!attribute!returns!in!excess!of!the!riskIfree!interest!
rate!to!one!(CAPM),!three!(Fama!and!French)!and!four!(Carhart)!systematic!risk!factors!
respectively:!! !!
!
!!!!!!!!!!!!!!!!! ! − !"#$%&1! = ! + !(!"#$%# − !"#$%&1!) + !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!(1)!
!!!!!!!!!!!!!!!!! ! − !"#$%&1! = ! + !! !"#$%# − !"#$%&1! + !!!"# + !!!"# + !!!!!!!!(2)!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
5
!The!stock!market!lists!are:!Aktietorget,!Externa!listan,!Innovationsmäklarna,!Inofficiella!noteringar,!Nordic!Growth!
Market,!Nordic!Otc!Market,!Nya!Marknaden,!Stockholms!Börsinfo,!Stockholmsbörsen,!Stockholmsbörsen!A2Ilistan,!
Stockholmsbörsen!First!North,!Stockholmsbörsen!Large!Cap,!Stockholmsbörsen!Mid!Cap,!Stockholmsbörsen!OI
listan,!Stockholmsbörsen!OTCIlistan,!Stockholmsbörsen!Small!Cap,!Stockholmsbörsen!Utländska.!Among!those,!we!
keep!the!six!underlined!ones.!!
!
5!
!
! ! − !"#$%&1! = ! + !! !"#$%# − !"#$%&1! + !!!"# + !!!"#!!!!!!!!!!!!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!+!!!!"! + !!!!!!!!!!!!!!(3),!
where$! ! ! ! = !, ! !is!the!gross!or!net!return!of!fund!i$at!time!t$(fund!and!time!subscripts!are!
omitted),!STIBOR1M$is!the!Stockholm!1Imonth!interbank!lending!rate,!α!is!the!return!left!
unexplained!by!the!benchmark!model,!β,!!, !!!and!!!are!factor!loadings,!!"#$%#!is!a!valueI
weighted!!index!for!all!companies!listed!on!the!Stockholm!Stock!Exchange,!including!reinvested!
dividends,!where!the!weight!of!a!single!company!is!capped!at!ten!percent!to!reflect!a!regulation!
that!applies!to!mutual!fund!portfolios,!!"#!is!our!Swedish!version!of!the!HighIMinusILow!
bookItoImarket!value!factor,!!"#!is!our!Swedish!version!of!the!SmallIMinusIBig!market!
capitalization!factor,!!"!!is!our!Swedish!version!of!Carhart’s!momentum!factor,!which!is!long!
on!priorIyear!winners!and!short!on!priorIyear!losers,!and!!!is!the!regression!residual.!6!
!
Equations!(1)!–!(3)!are!estimated!for!all!equity!mutual!funds!with!a!minimum!of!36!and!a!
maximum!of!132!monthly!return!observations!from!January!1999!and!including!December!
2009.!Equation!(1)!is!also!estimated!for!the!period!January!1999!–!August!2013.!Requiring!a!
minimum!return!history!of!36!months!means!that!we!have!practically!no!survival!bias!in!our!
estimates,!but!also!that!estimates!for!the!shortestIlived!funds!tend!to!have!lower!precision.!We!
therefore!report!actual!and!bootstrapped!tIstatistics!in!Table!7.! !!
!!
3.1!Actively!managed!funds!
Consider!first!some!descriptive!statistics!for!our!fund!universe!of!actively!managed!funds!in!
Table!1.!The!equallyIweighted!average!absolute!net!return!–!without!risk!adjustment!–!of!the!
115!actively!managed!funds!was!lower!than!that!of!the!market!portfolio!by!18!basis!points!
during!the!period!1999I2009.!But,!as!reported!in!Table!2,!the!average!total!expense!ratio!was!
1.32!per!cent!and!the!average!total!cost!share!–!which!includes!trading!costs!in!addition!to!
expenses!I!was!1.59!per!cent.!This!means!that!the!gross!return!of!the!average!equity!mutual!
fund!was!substantially!higher!than!the!stock!market!return.!!We!cannot,!however,!conclude!
that!this!is!due!to!true!stock!picking!skills!of!fund!managers!before!adjusting!fund!returns!for!
systematic!risk.!
!
[TABLE!1]![TABLE!2]!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
6
!There!is!controversy!about!whether!the!average!SMB,!HML!and!MOM!returns!are!rewards!for!risk!or!the!result!of!
mispricing.!Regardless,!it!remains!true!that!fund!managers!can!implement!passive!strategies!to!capture!returns!to!
size!and!value!bias!and!to!momentum,!and!that!stockIpicking!ability,!i.e.!active!management,!should!show!up!in!
the!intercept!(alpha).!
6!
!
!
Table!3!reports!net!riskIadjusted!returns!above!the!riskIfree!interest!rate!–!net!excess!
returns!–!of!actively!managed!funds!as!given!by!estimated!alphas!from!equations!(1)!–!(3)!for!
the!period!1999I2009,!and!from!equation!(1)!for!the!period!1999IAugust!2013.!!
!
[TABLE!3]!
For!the!period!1999I2009,!the!equalIweighted!average!net!excess!return!of!all!funds!is!
negative,!regardless!of!the!number!of!systematic!risk!factors!adjusted!for.!The!variation!across!
deciles!of!funds!is!large,!the!top!decile!having!a!positive!a!net!excess!4Ifactor!return!of!5.9!per!
cent!per!year!and!the!bottom!decile!a!negative!return!of!–6.5!per!cent.!The!table!also!reports!
excess!net!returns!of!the!top!and!bottom!five!funds.!Among!those,!the!variation!is!even!larger,!
the!best!and!worst!performing!fund!having!a!4Ifactor!net!return!alpha!of!13.6!and!–15.3!per!
cent!respectively.!Table!A.1!in!the!Appendix!reports!estimated!net!excess!returns!year!by!year!
to!show!that!the!variation!over!time!is!considerable.!
We!lack!sufficient!stock!market!data!to!extend!the!estimation!of!3I!and!4Ifactor!gross!
and!net!excess!returns!to!the!most!recent!years.!We!can!however!estimate!1Ifactor!gross!and!
net!excess!returns!for!the!period!January!1999!up!to!and!including!August!2013.!The!results!are!
presented!in!the!first!column!of!Table!3.!They!should!be!compared!with!estimated!net!excess!
returns!for!the!longer!time!period!in!the!fourth!column!of!Table!3.!A!comparison!reveals!that!
the!average!returns!are!quite!similar:!average!net!returns!in!the!shorter!and!longer!period!are!I
0.68!and!I0.70!per!cent!per!year,!and!the!corresponding!average!gross!returns!are!0.63!and!
0.56!per!cent!per!year.!A!comparison!of!the!distribution!across!deciles!and!across!top!and!
bottom!funds!reveals!that!the!distribution!for!the!shorter!period!is!wider!than!that!of!the!
longer!period.!Based!on!this,!we!see!no!reason!to!believe!that!estimates!of!3I!and!4Ifactor!
excess!returns!for!the!longer!period!would!differ!substantially!from!the!estimates!for!the!
shorter!period.!RV:!ADD!HOW!MANY!DECILES!THAT!ARE!NEGATIVE?!
!
Net!return!is!the!relevant!performance!measure!for!the!investor,!but!to!ascertain!
whether!fund!managers!have!true!skill!in!picking!stocks!we!need!to!estimate!gross!excess!
returns.!Table!4!reports!gross!1I,!3I!and!4Ifactor!excess!returns!for!the!time!period!1999I2009,!
and!gross!1Ifactor!excess!returns!for!the!time!period!1999IAugust!2013.!
![TABLE!4]!
For!the!shorter!period,!the!average!equalIweighted!gross!4Ifactor!alpha!for!the!105!funds!is!a!
positive!0.9!per!cent!per!year.!!In!fact,!more!than!half!of!the!funds!have!positive!gross!alphas.!
The!top!fund!has!a!gross!alpha!of!more!than!15!per!cent!per!year!and!the!bottom!fund!a!
negative!gross!alpha!of!I13.8!per!cent.!!
7!
!
!
A!comparison!of!1Ifactor!gross!excess!returns!for!the!shorter!and!longer!time!periods!in!
Table!4!again!reveals!that!the!estimates!are!quite!similar.!The!mean!gross!alpha!for!all!funds!is!7!
basis!points!lower!over!the!longer!period,!and!the!variation!across!deciles!and!across!the!five!
top!and!bottom!funds!is!smaller.!The!top!six!deciles!have!positive!gross!excess!returns!in!both!
the!shorter!and!longer!period.!!
Estimated!net!and!gross!alphas!in!Tables!3!and!4!are!quite!large!at!the!tails!of!the!
distribution.!The!question!is,!to!what!extent!–!if!at!all!–!are!estimated!excess!returns!
significantly!different!from!zero?!Figure!1,!panel!a)!and!b)!shows!estimated!net!and!gross!alphas!
of!all!funds!together!with!conventional!twoIsided!confidence!bands!constructed!under!the!
hypothesis!that!excess!returns!are!zero.!Somewhat!surprisingly,!only!a!small!number!of!funds!
have!significantly!positive!or!negative!excess!returns.!Looking!first!at!gross!excess!returns!in!
panel!b),!it!can!be!seen!that!only!the!top!funds!plus!a!handful!with!lower!rankings!have!positive!
estimated!returns!that!are!significant!at!the!strictest!confidence!level.!Similarly,!very!few!have!
significant!negative!gross!excess!returns.!In!other!words,!very!few!funds!seem!to!possess!true!
skills!in!picking!stocks.!Looking!next!at!net!excess!returns!in!panel!a),!one!can!see!that!only!a!
couple!of!top!funds!are!able!give!investors!significant!positive!net!excess!returns,!while!a!larger!
number!at!the!bottom!of!the!ranking!give!investors!significantly!negative!net!excess!returns.!
The!latter!seem!to!have!truly!negative!skills!in!picking!stocks,!or!a!truly!inefficient!management!
style.!
![FIGURE!1]!
!
Superior!performance!should!enable!fund!managers!to!extract!relatively!high!fees!from!
investors!and!still!give!investors!superior!net!excess!returns.7!For!this!reason!we!expect!to!see!a!
positive!relation!between!net!excess!returns!and!fees.!Table!5!reports!regressions!of!net!excess!
returns!on!fees!and!expenses!or!costs.!!
!
[TABLE!5]!!
Contrary!to!expectation,!the!relation!between!net!returns!and!fees!or!costs!tends!to!be!
negative.!In!the!regression!where!fees!and!costs!explain!net!excess!returns!jointly,!all!loadings!
are!negative!but!insignificant.!Likewise,!total!expense!ratios!(TER)!and!total!cost!ratios!(TKA)!
have!negative!but!insignificant!loadings.!!We!conclude!that!high!fees!signal!management!
inefficiency!or!inferior!skill!if!anything.!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
7
!Berk!and!van!Binsbergen!(2013)!argue!that!management!skill!is!reflected!in!the!difference!between!gross!and!net!
excess!returns!multiplied!by!assets!under!management.!In!other!words,!a!manager!with!a!smaller!grossInet!margin!
but!greater!assets!under!management!can!be!more!skilled!than!a!manager!with!a!higher!gross!excess!return!and!
greater!margin,!but!smaller!assets!under!management.!We!stick!to!the!traditional!view,!that!gross!excess!return!
differences!measure!skill!differences.!
8!
!
!
3.2!Index!funds!
We!have!return!series!for!15!index!funds!with!a!minimum!of!36!consecutive!monthly!return!
observations!for!the!period!1999I2009!and!for!12!funds!for!the!period!1999I2013.8!The!funds!
track!different!indices.!Of!the!set!of!15!funds,!five!track!all!or!nearly!all!listed!companies!on!the!
Stockholm!Stock!Exchange,!two!track!the!80!to!100!largest!and!most!traded!companies,!and!
eight!track!the!30!largest!and!most!traded!companies.9!
!
An!obvious!approach!to!evaluate!the!performance!of!index!funds!is!to!compare!their!
gross!and!net!returns!to!that!of!their!respective!benchmark.!Many!funds!use!benchmarks!
without!reinvested!dividends!despite!the!fact!that!dividends!are!included!in!the!returns.!These!
funds!should!therefore!be!evaluated!against!the!stated!benchmark!but!including!dividend!
returns.10!11!!!
Investors!should!expect!to!receive!a!return!equal!to!that!of!the!benchmark!including!
dividends!and!minus!management!costs.!This!is!rarely!the!case!because!of!tracking!error.!
Tracking!error!can!depend!on!a!number!of!different!factors,!such!as!the!timing!and!frequency!of!
adjusting!index!weights,!the!handling!of!inflows!and!outflows,!and!the!method!of!replicating!
the!benchmark.!Costs!can!also!vary!greatly!between!funds.!They!!depend!on!management!style!
and!importantly!on!fund!size!–!there!are!considerable!economies!of!scale!in!fund!management.!
For!larger!funds,!it!is!cost!effective!to!obtain!index!returns!by!holding!the!component!stocks,!for!
smaller!funds!to!replicate!the!index!by!holding!index!futures.!!
!
Another!approach!to!evaluate!index!fund!performance!is!to!treat!them!in!the!same!way!
as!actively!managed!funds,!i.e.!to!evaluate!them!against!the!market!portfolio!including!dividend!
returns!(and!taking!account!of!the!10!per!cent!cap!on!the!index!weight!of!any!individual!stock!
required!by!regulation).!In!taking!this!approach!we!assume!that!the!investor!wants!exposure!to!
the!stock!market!at!the!highest!riskIadjusted!return!net!of!costs.!!
We!follow!both!approaches.!Following!the!first!approach,!Table!6!reports!the!1Ifactor!
net!and!gross!excess!returns!of!each!index!fund!against!its!own!benchmark!including!reinvested!
dividends.12!!A!positive!or!negative!gross!excess!return!–!a!positive!or!negative!tracking!error!–!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
8
!We!have!excluded!two!leveraged!exchange!traded!funds!(ETFs).!!
!The!benchmark!index!for!each!fund!is!reported!in!Table!6.!
10
!!
11
!One!management!company!of!an!index!fund!does!not!charge!any!fee!but!instead!appropriates!dividend!returns.!
It!is!evaluated!against!its!stated!benchmark!and!the!dividend!return!is!treated!as!a!fee.!!
12
!Index!funds!hold!all!equities!in!its!benchmark!(or!a!derivative!of!the!benchmark!itself)!and!therefore!take!no!
systematic!risk!in!the!form!of!value!or!growth!bias!or!exploit!momentum!by!holding!only!a!subset!of!the!equities!
included!in!the!benchmark.!!
9
9!
!
means!that!the!fund!has!performed!better!or!worse!than!its!benchmark!including!reinvested!
dividends.!!
[TABLE!6]!
Net!excess!returns!vary!greatly!across!the!index!funds,!from!1.35!per!cent!for!the!top!to!
I4.12!per!cent!for!the!bottom!fund.!Differences!in!net!excess!returns!can!be!attributed!to!
differences!in!both!tracking!error!and!expenses.!As!can!be!seen,!tracking!error!–!gross!alphas!–!
and!expenses!–!measured!by!TER!–!vary!greatly!across!funds.!It!is!evident!that!an!investor!must!
consider!tracking!error!as!well!as!expenses!when!choosing!between!index!funds.!Which!
benchmark!a!fund!has!seems!to!play!a!role:!four!of!the!top!five!funds!and!only!one!of!the!
bottom!five!funds!in!terms!of!net!excess!returns!have!a!narrow!benchmark.!On!average,!and!
excluding!the!anomalous!bottom!fund,!index!funds!give!investors!stated!benchmark!returns!
including!dividends!and!incur!less!than!a!third!of!the!expenses!of!actively!managed!funds.!!!
Table!7!reports!net!and!gross!4Ifactor!alphas!when!all!index!funds!are!evaluated!against!
the!market!portfolio!including!reinvested!dividends.!!
[TABLE!7]!!
The!gross!excess!return!should!be!zero!for!a!fund!that!tracks!the!whole!stock!market!
and!therefore!takes!no!systematic!risk,!provided!that!it!has!no!tracking!error!and!that!all!costs!
are!added!back!to!the!net!return.!We!find!that!this!more!or!less!is!the!case;!the!four!index!funds!
that!track!the!market!portfolio!have!an!average!gross!excess!return!of!I0.06!per!cent.!!Index!
funds!that!track!the!subset!of!the!30!most!traded!stocks!have!negative!4Ifactor!excess!gross!
returns!on!average.!We!know!from!Table!1!that!the!smallIversusIlarge!capitalization!factor!has!
a!positive!loading,!consistent!with!lower!gross!excess!returns!for!index!funds!focused!on!large!
caps.!!
It!is!clear!that!an!investor!who!follows!the!advice!to!select!a!random!index!fund!instead!
of!a!random!actively!managed!fund!may!not!be!better!off.!In!terms!of!4Ifactor!alphas,!the!
median!index!funds!maps!into!the!sixth!or!seventh!decile!among!the!actively!managed!funds.!A!
better!strategy!is!to!be!selective!both!in!terms!of!the!fund’s!benchmark!index!and!fees.!It!is!not!
sufficient!to!choose!a!broad!over!a!narrow!index!fund.!The!only!fund!with!a!positive!net!excess!
return!alpha!is!a!narrow!fund,!benchmarking!against!the!SIX30RX!index,!which!in!part!is!due!to!
the!fact!that!the!manager!carries!all!costs.!!
!
4.!Persistence!!
10!
!
The!top!four!deciles!of!all!actively!managed!funds!have!obtained!positive!net!excess!returns!on!
average!over!a!period!of!eleven!years,!and!the!top!five!funds!achieved!excess!net!returns!of!
more!than!5!percent!per!year.!Financial!advice!on!investing!in!mutual!funds!is!often!based!on!
past!performance.!The!presumption!is!that!superior!past!performance!is!due!to!skill!and!that!
skill!is!persistent.!!
To!examine!the!presence!of!funds!with!persistent!superior!performance!in!our!fund!
universe,!we!sort!funds!into!deciles!according!to!performance!and!then!estimate!their!
performance!during!subsequent!years,!following!Carhart!(1997).!We!note!that!this!approach!is!
subject!to!possible!model!misspecification,!since!the!same!performance!attribution!model!is!
used!to!rank!funds!and!to!measure!performance.!If!the!model!has!a!bias!in!risk!adjustment,!for!
example!due!to!an!omitted!variable!or!in!the!construction!of!risk!factors,!the!bias!in!ranking!will!
also!affect!the!subsequent!measurement!of!performance.!
We!choose!to!rank!funds!on!their!performance!during!a!three!year!period!rather!than!
on!a!shorter!period!to!reduce!noise.!Our!first!ranking!period!is!1996I1998.!Based!on!their!net!
excess!return!alphas,!funds!are!grouped!into!deciles,!the!first!decile!having!the!highest!and!the!
tenth!the!lowest!performance.!We!then!estimate!the!performance!of!deciles!in!each!year!
during!the!period!1999I2003.!Moving!one!year!forward!at!a!time,!this!is!repeated!for!six!more!
ranking!and!evaluation!periods.!The!last!ranking!period!is!2002I2004!and!the!last!evaluation!
period!2005I2009.!The!weights!on!each!fund!and!the!composition!of!deciles!are!adjusted!for!
the!entry!and!exit!of!funds!over!time.!We!then!calculate!average!performance!for!the!seven!
fiveIyear!ranking!and!evaluation!periods.!The!result!is!presented!in!Figure!2.!!
!
[FIGURE!2]!
!!
Performance!varies!greatly!across!deciles!in!the!ranking!period.!The!top!decile!has!a!
positive!net!excess!return!of!about!16!per!cent!and!the!bottom!decile!a!negative!return!of!
about!I8!per!cent!per!year.!The!absence!of!persistence!in!the!evaluation!period!is!quite!striking.!
If!anything,!a!tendency!towards!a!reversal!of!fund!rankings!is!visible!during!the!first!year;!the!
top!decile!becomes!the!ninth,!and!the!seventh!becomes!the!first.!Decile!returns!have!more!or!
less!converged!to!the!mean!in!the!second!year!and!remain!close!in!the!following!years.!!!
!
The!same!pattern!of!rapid!convergence!and!a!tendency!of!reversal!in!rankings!in!the!
first!year!are!visible!for!the!top!and!bottom!five!funds,!see!Figure!3.!
[FIGURE!3]!
Note!that!the!net!excess!return!of!the!top!fund!remains!substantially!below!those!of!the!
other!four!funds!in!the!evaluation!period,!and!that!the!second!worst!performing!fund!remains!
substantially!above!the!other!four.!
11!
!
The!conclusion!that!can!be!drawn!from!Figure!2!and!3!is!that!the!rapid!reversion!to!the!
mean!of!fund!excess!returns!is!a!strong!indication!that!fund!managers!lack!stockIpicking!skills!
and!that!excess!returns!–!positive!or!negative!–!are!due!to!pure!chance.!
!
5.!Skill!or!luck?!
To!test!whether!the!performance!of!active!funds!is!truly!superior!or!inferior,!especially!at!the!
tails!of!the!distribution,!we!employ!a!bootstrap!procedure!that!yields!a!distribution!of!pseudoI
returns!for!each!fund!when!true!alpha!of!every!fund!is!set!to!zero.!With!gross!excess!returns,!
setting!alpha!equal!to!zero!presumes!that!funds!obtain!the!same!excess!return!as!the!market!
portfolio.!With!net!excess!returns,!setting!alpha!equal!to!zero!presumes!that!funds!can!obtain!
returns!that!cover!the!costs!of!active!management.!!Actual!fund!excess!returns!are!compared!to!
bootstrapped!excess!returns.!If!actual!gross!excess!returns!differ!from!bootstrapped!returns!
with!statistical!confidence,!we!conclude!that!the!abnormal!excess!returns!are!due!to!superior!or!
inferior!skill.!If!not,!we!cannot!rule!out!that!actual!gross!returns!are!due!to!good!and!bad!luck,!
and!not!to!superior!or!inferior!skill.!If!actual!net!excess!returns!differ!from!bootstrapped!returns!
with!statistical!confidence,!we!conclude!that!abnormal!positive!excess!returns!belong!to!funds!
that!can!more!than!compensate!investors!for!the!costs!of!active!management,!and!that!
abnormal!negative!excess!returns!belong!to!funds!that!have!abnormally!high!management!
costs!or!an!inefficient!management!style.!
To!generate!returns!under!the!hypothesis!of!zero!alphas,!we!subtract!each!fund’s!excess!
return!as!measured!by!alpha!from!equations!(1)I(3)!from!its!monthly!gross!and!net!returns!for!
the!part!of!1999I2009!it!has!been!in!existence.!A!simulation!run!is!a!random!sample!with!
replacement!drawn!from!the!132!calendar!months!from!January!1999!to!December!2009,!the!
same!random!sample!of!months!for!each!fund.!Each!time!a!certain!month!is!drawn,!all!funds!in!
existence!that!month!are!given!the!corresponding!adjusted!return!and!the!corresponding!factor!
returns.!!We!estimate!equations!(1)I(3)!for!each!fund!on!the!simulation!draw!of!adjusted!fund!
returns!and!factor!returns.!Each!simulation!run!produces!a!crossIsection!of!alpha!estimates!
using!the!same!random!sample!of!months.!This!is!done!5,000!times!to!produce!a!distribution!of!
alphas!at!each!point!in!the!ranking!distribution!(under!the!assumption!of!zero!alphas).!The!
distribution!for!the!top!fund!is!constructed!as!the!distribution!of!the!maximum!alpha!generated!
across!all!bootstraps,!the!distribution!for!the!second!best!fund!as!the!second!best!alpha!across!
all!bootstraps,!and!so!on.!!
!
Note!that!in!a!given!simulation!run,!funds!will!have!the!same!random!sample!of!months!
as!when!they!overlap!in!existence.!This!has!the!advantage!that!the!simulations!will!capture!any!
crossIcorrelation!in!fund!returns.!We!show!in!figure!A.1!that!crossIcorrelation!exists!and!is!
12!
!
important.!The!joint!sampling!of!fund!and!factor!returns!has!the!added!advantage!of!capturing!
any!correlated!heteroscedasticity!of!the!explanatory!returns!and!the!disturbances!of!the!
benchmark!model.!Sampling!the!same!months!for!all!funds!has!the!disadvantage!that!a!fund!
may!show!up!in!a!simulation!run!for!more!or!less!than!the!number!of!months!it!has!actually!
existed!since!any!given!month!may!be!sampled!more!than!once!or!never.!Presumably,!overI
sampling!of!some!funds!will!be!balanced!by!underIsampling!of!others!in!each!simulation!and!
over!the!5,000!runs!used!to!make!inferences.!There!is!a!caveat,!however.!We!discard!funds!with!
a!return!history!of!less!than!36!months,!with!the!result!that!we!may!end!up!with!a!bit!more!
overI!than!underIsampling.!
!
There!are!two!more!caveats.!Using!the!same!month’s!return!for!all!funds!in!a!simulation!
run!preserves!the!crossIcorrelation!of!fund!returns,!but!eliminates!any!effects!of!
autocorrelation.!This!seems!to!be!a!minor!problem,!see!e.g.!Fama!(1965)!or!Kosowski!et.al.!!
(2006).!Also,!since!we!randomly!sample!months,!we!lose!any!effects!of!time!variation!in!the!
regression!slopes!of!(1)I(3).!Ferson!and!Schadt!(1996)!argue!that!time!variation!should!be!
allowed,!but!we!rule!it!out.!!
!
We!are!now!ready!to!turn!to!simulation!results,!and!start!with!bootstrapped!alpha!
distributions!of!net!returns.!Remember!that!setting!true!alpha!equal!to!zero!in!our!simulations!
to!estimate!pseudoIalphas,!we!make!the!assumption!that!fund!managers!are!able!to!achieve!
net!returns!that!are!sufficient!to!cover!the!costs!of!active!management.!Figure!4,!panel!a)!
shows!bootstrapped!net!4Ifactor!alpha!distributions.!Points!in!the!distributions!marking!
different!confidence!levels!have!been!joined!to!form!bands!for!different!confidence!levels!
around!the!estimates.!It!should!be!pointed!out!that!the!confidence!bands!are!not!based!on!an!
assumed!parametric!form!of!the!distributions,!but!on!actual!counts!of!point!estimates!over!the!
5,000!simulations.13!Estimates!of!actual!net!4Ifactor!alphas!are!also!shown.!(Similar!
distributions!of!net!1I!and!3Ifactor!pseudoIalpha!distributions!are!presented!in!Figures!A2!and!
A3.)!
[FIGURE!4]!
The!estimated!actual!net!return!alphas!are!higher!than!the!estimated!pseudoIalphas!for!
the!top!30!funds!and!lower!for!the!remaining!85!funds.!It!seems!that!a!majority!of!fund!
managers!are!unable!to!fully!compensate!for!the!costs!of!active!management,!but!that!a!
minority!is!able!to!do!so.!This!is!consistent!with!the!estimate!of!average!net!excess!return!of!!!!!!
I0.45!per!cent.!However,!there!are!no!significant!positive!net!excess!returns!at!any!of!the!three!
levels!of!confidence.!At!the!other!end!of!the!distribution,!there!are!some!funds!significant!
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
13
!Kosowski!et!al!(2006)!stress!that!the!actual!distribution!may!differ!from!the!normal!distribution.!
13!
!
negative!net!excess!returns:!32!funds!at!the!10!per!cent,!9!at!the!five!per!cent,!and!2!funds!at!
the!2.5!per!cent!oneIsided!confidence!level.!!
This!does!still!not!answer!the!question!whether!some!active!managers!possess!true!
stockIpicking!skills,!i.e.!can!obtain!higher!gross!excess!returns!than!the!market!portfolio.!Figure!
4,!panel!b)!shows!confidence!bands!around!simulated!gross!4Ifactor!excess!returns!for!the!105!
funds!for!which!we!have!expense!data!and!estimated!actual!gross!excess!returns.!(Similar!
figures!for!1I!and!3Ifactor!gross!excess!returns!are!shown!in!figures!A4!and!A5.)!
!
A!majority!of!funds!–!72!out!of!105!–!have!positive!estimated!actual!gross!excess!
returns,!consistent!with!the!estimated!average!gross!return!of!+0.90!per!cent.!Some!funds!have!
significant!gross!excess!returns:!funds!7!to!37!at!10!per!cent,!12I30!at!5!per!cent,!and!15!plus!
17I28!at!a!2.5!per!cent!oneIsided!confidence!level.!In!other!words,!we!find!significant!skill!not!
for!the!very!top!funds,!but!for!some!funds!in!the!second!and!third!deciles.!!There!are!no!funds!
that!have!significant!negative!excess!returns!at!any!level!of!confidence.!We!conclude!that!a!
small!share!of!the!105!active!funds!in!our!sample!seem!to!possess!true!stockIpicking!skills,!but!
that!the!top!funds!have!obtained!their!high!returns!by!being!lucky.!!
Funds!that!are!shortIlived!have!less!precise!estimates!of!excess!returns!–!estimates!with!
greater!standard!errors!–!than!funds!that!exist!the!whole!time!period.!By!using!the!tIstatistics!of!
the!estimates!of!excess!returns,!one!can!control!for!the!effect!of!different!lifeItimes!of!funds!
and!the!subsequent!precision!of!the!estimates.!This!is!the!preferred!approach!of!both!Kosowski!
et.al.!(2006)!and!Fama!and!French!(2010).!!Table!7!shows!–!in!the!rightImost!column!–!what!in!
effect!are!pIvalues!for!the!probability!that!bootstrapped!gross!4Ifactor!excess!returns!are!
greater!in!absolute!value!than!actual!estimated!gross!4Ifactor!returns.!It!can!be!seen!that!no!
decile!of!funds!have!a!pIvalue!lower!than!10!per!cent,!corresponding!to!a!twoIsided!80!per!cent!
confidence!level.!Thus,!the!use!of!tIstatistics!of!alphas!instead!of!the!alphas!themselves!yields!
more!negative!findings!for!the!presence!of!skill.!It!is!only!when!we!examine!pIvalues!for!the!
very!top!and!bottom!funds!that!we!find!a!trace!of!evidence!for!superior!and!inferior!skills!at!the!
80!per!cent!confidence!level:!at!this!level!the!second!and!fourth!fund!at!the!top!have!significant!
positive!excess!returns,!and!the!fifth!fund!from!the!bottom!has!significant!negative!excess!
return.!!
!
6.!Summary!
Most!investors!and!financial!advisers!seem!to!believe!that!some!fund!managers!possess!true!
skill!in!picking!stocks.!Financial!advisers!commonly!recommend!that!investors!look!at!past!
returns!of!funds!as!a!guide!to!choose!which!fund!to!invest!in.!This!means!that!they!not!only!
14!
!
believe!in!the!existence!of!superior!skills,!but!also!that!skills!are!persistent!and!that!return!
unadjusted!for!systematic!risk!is!a!good!measure!of!skills.!
!
We!examine!whether!these!beliefs!are!warranted!by!analyzing!performance!and!
persistence!of!Swedish!equity!mutual!funds!in!the!period!1999I2009.!We!start!as!late!as!1999!
because!the!Swedish!market!for!mutual!funds!developed!quite!late!and!the!new!pension!
system!was!launched!in!2000.!Performance!is!measured!as!the!net!or!gross!return!in!excess!of!
the!returns!to!systematic!risk!factors:!bias!towards!a!smaller!or!larger!part!of!the!stock!market,!!
small!or!large!corporations,!companies!with!high!or!low!bookItoImarket!value,!or!holding!a!
portfolio!of!stocks!with!high!past!performance.!Returns!net!of!costs!is!the!relevant!performance!
measure!for!investors,!while!gross!returns!is!relevant!for!determining!whether!fund!managers!
have!superior!or!inferior!skills.!
!
We!find!that!the!average!equalIweighted!net!excess!return!of!all!115!actively!managed!
funds!in!existence!for!at!least!36!months!is!negative.!The!variation!across!deciles!is!large,!the!
top!decile!having!a!positive!net!excess!return!of!5.9!per!cent!per!year!and!the!bottom!decile!a!
negative!return!of!–6.5!per!cent.!The!variation!across!individual!funds!is!considerably!larger,!the!
top!fund!having!an!excess!net!return!of!13.6!per!cent!and!the!bottom!fund!–15.3!per!cent!per!
year.!
!
The!average!equalIweighted!gross!alpha!for!the!105!funds!for!which!we!have!data!is!on!
the!other!hand!a!positive!and!significant!0.9!cent!per!year.!More!than!half!of!the!funds!have!
positive!gross!alphas,!and!the!top!fund!has!a!gross!alpha!of!15.2!per!cent!per!year.!A!fairly!large!
number!of!funds!have!managers!who!seem!to!possess!true!skill!–!their!gross!alphas!are!
significantly!positive!–!and!quite!a!few!are!able!to!more!than!compensate!for!costs!to!give!
investors!positive!net!excess!returns.!At!the!same!time,!some!managers!have!substantially!
negative!gross!excess!returns.!They!seem!to!have!inferior!skills!and/or!a!costly!management!
style.!
!
We!also!examine!the!performance!of!15!index!funds.!Adherents!of!the!efficient!market!
hypothesis!tend!to!recommend!investors!to!choose!an!index!fund!instead!of!an!actively!
managed!fund!on!the!premise!that!expected!gross!returns!are!the!same!(and!equal!to!the!stock!
market’s)!but!that!cost!are!lower!and!net!returns!higher!of!index!funds.!We!follow!two!
approaches!to!evaluate!passive!funds.!In!one,!we!treat!them!in!the!same!way!as!active!funds!
and!estimate!net!and!gross!excess!returns.!On!average,!index!funds!have!lower!net!and!gross!
excess!returns!than!actively!managed!funds.!The!main!reason!is!that!a!majority!of!the!passive!
funds!take!a!relatively!high!market!risk;!they!hold!only!the!30!most!traded!and!largest!
corporations!in!capitalization!on!the!Stockholm!stock!exchange.!In!the!other!approach,!we!
evaluate!the!index!funds!against!their!own!benchmarks,!but!adjusted!for!the!fact!that!all!funds!
15!
!
receive!dividend!returns.!We!find!that!the!funds!track!their!benchmarks!quite!well!on!average,!
but!that!individual!funds!miss!their!benchmarks!substantially.!!
!
The!finding!that!more!than!half!of!the!actively!managed!funds!have!positive!excess!gross!
returns!indicates!that!they!possess!superior!stock!picking!skills.!If!they!do,!they!should!tend!to!
exhibit!persistency!in!their!returns.!We!examine!persistence!by!ranking!funds!on!threeIyear!
performance!and!evaluating!their!performance!during!the!subsequent!five!years.!Our!short!
time!series!permits!us!to!obtain!the!average!of!seven!ranking!and!evaluation!periods.!Given!this!
limitation,!we!find!a!very!rapid!convergence!of!net!excess!returns!to!the!mean!of!all!deciles,!and!
also!of!the!top!and!bottom!five!funds.!If!anything,!there!is!a!tendency!for!a!reversal!in!ranking!in!
the!first!year!of!the!evaluation!period.!Hence,!there!is!practically!no!evidence!of!persistence!
and!therefore!of!skill.!
!
Looking!at!persistence!of!performance!only!gives!indications!of!the!presence!of!truly!
superior!and!inferior!performance.!To!obtain!evidence!with!statistical!precision,!we!investigate!
directly!the!presence!of!true!skills.!Employing!a!bootstrap!procedure,!we!construct!a!very!large!
number!of!pseudoIexcess!returns!for!each!actively!managed!fund.!By!construction,!the!returns!
are!generated!under!the!assumption!that!no!fund!possesses!superior!or!inferior!skill.!The!
bootstrap!yields!a!distribution!of!pseudoIexcess!net!and!gross!returns!for!each!fund.!We!then!
test!whether!the!actual!excess!return!of!each!fund!lies!within!the!distribution!generated!by!
lucky!or!unlucky!draws.!We!find!that!good!and!bad!luck!rather!than!superior!and!inferior!skill!
can!explain!most!of!actual!gross!excess!returns!across!funds,!except!for!a!range!of!funds!below!
the!very!top,!which!seem!to!possess!stock!picking!skill.!However,!when!we!control!for!the!
precision!of!estimates!of!alpha!due!to!different!lifeItimes!of!funds,!we!find!practically!no!
evidence!of!superior!skill.!Whatever!skill!that!may!exist,!they!are!dissipated!by!costs.!There!is!no!
evidence!significant!positive!net!excess!returns!and!at!the!same!time!evidence!of!negative!net!
excess!returns.!
!
In!conclusion,!there!is!practically!no!evidence!of!true!stock!picking!skill!among!managers!
of!Swedish!equity!mutual!funds,!including!managers!that!have!achieved!the!highest!average!
gross!excess!returns.!Gross!excess!returns!at!both!ends!of!the!distribution!can!be!obtained!by!
good!and!bad!luck,!and!not!by!superior!and!inferior!skill.!Investors!wanting!exposure!to!the!
Swedish!stock!market!through!mutual!funds!should!be!aware!of!the!lack!of!persistence!in!fund!
returns!and!choose!a!passively!managed!fund!with!low!or!no!fees!rather!than!an!actively!
managed!fund.!!
!
!
!
16!
!
References(
Berk,!Jonathan!B.!and!Jules!H.!van!Binsbergen,!2013,!Measuring!skill!in!the!mutual!fund!
industry,!manuscript!dated!November!4.!
Carhart,!Mark!M.,!1997,!On!persistence!in!mutual!fund!performance,!Journal$of$Finance!
52,!57I82.!
Cuthbertson,!Keith,!Dirk!Nietzsche!and!Niall!O’Sullivan,!2008,!Journal$of$Empirical$
Finance!15,!613I634.!
Dahlquist,!Magnus,!Stefan!Engström!and!Paul!Söderlind,!2000,!Performance!and!
characteristics!of!Swedish!mutual!funds,!Journal$of$Financial$and$Quantitative$Analysis!35,!409I
423.!
Engström,!Stefan,!2004,!Investment!strategies,!fund!performance!and!portfolio!
characteristics,!SSE/EFI Working Paper Series No. 554.!
Fama,!Eugene!F.,!1965,!The!behavior!of!stock!market!prices,!Journal$of$Business!38,!34I
105.!
Fama,!Eugene!F.!and!Kenneth!R.!French,!1993,!Common!risk!factors!in!the!returns!on!
stocks!and!bonds,!Journal$of$Financial$Economics!33,!3I56.!
Fama,!Eugene!F.!and!Kenneth!R.!French,!2010,!Luck!versus!skill!in!the!crossIsection!of!
mutual!fund!returns,!Journal$of$Finance!65,!1915I1947.!
Ferson,!Wayne!E.!and!Rudi!W.!Schadt,!1996,!Measuring!fund!strategy!and!performance!
in!changing!economic!conditions,!Journal$of$Finance!51,!425I462.!
Kosowski,!Robert,!Allan!Timmerman,!Russ!Wermers,!and!Hal!White,!2006,!Can!mutual!
fund!“stars”!really!pick!stocks?!New!evidence!from!a!bootstrap!analysis,!Journal$of$Finance!61,!
2551I2595.!
!
!
!
!
17!
!
!
Table(1.((Fund(and(factor(returns(
! !
!
!
! !
(
!!!
!!!
1999!I!2009!
!
!
!
Portfolio!
! No.!of!
funds!
Mean!!
!
All!funds!
!
! !
! 8,8%!
!
! !
!
!
(
130!
! !
(
! !
!!!
!
1999!I!2013!
!
!
Std!
! No.!of!
funds!
Mean!
Std!
! 22,0%!
!
! !
113!
! 9,2%!
! 21,7%!
101!
9,2%!
21,8%!
12!
8,7%!
20,7%!
!
Index!funds!
!!!
Panel(A.(EquallyBweighted(portfolio(returns(
!
!
! !
!
(
! !
Actively!managed!
funds!
!!! !!!
!
115!
8,9%!
21,9%!
15!
8,1%!
22,5%!
!
!
Panel(B.(Factor(returns(
!
! !
!
!
!
1999!I!2009!
!
!
! !
!
!
!
Factor!
!
Mean!
Std!
! !
!
!
!
STIBOR1M!
! !
! !
! 3,2%!
! 0,3%!
! !
! !
!
!
!
!
SIXPRX!
! !
9,1%!
21,2%!
! !
!
!
SMB!
! !
2,1%!
23,5%!
! !
!
!
HML!
! !
20,7%!
31,9%!
! !
!
!
MOM!
! !
10,1%!
43,4%!
! !
!
!
! !
! !
!
! !
!
!
! !
Notes:!!1999I2013!includes!returns!until!September!2013.!
!
!
!
!
!
Table(2.((Fees(and(expenses,(1999(B(2009(
!
!
!
!
Annual!fee!
Entry!fee!
! ! Actively(managed(funds(
!
!
! No.!of!funds!
Mean! Std!
!
! !
! 1,3%! !0,4%!
100!
!
20!
2,3%! 2,3%!
!
! !
!
Index(funds(
! No.!of!funds!
Mean!
!
! !
! 0,5%!
14!
!
2!
2,2%!
!
!
Std!
!0,2%!
0,8%!
18!
!
Exit!fee!
Performance!fee!!
TER!
!
TKA!
!
Other!costs!
!
! !
!!
47!
8!
105!
99!
44!
1,2%!
18,1%!
1,3%!
1,6%!
0,2%!
1,5%!
6,9%!
0,4%!
0,6%!
0,2%!
!
!
!
!
!
!
!
! !
7!
0!
14!
13!
3!
0,6%!
I!
0,5%!
0,5%!
0,1%!
!
0,3%!
I!
0,2%!
0,2%!
0,1%!
!
Notes:!!
TER!!=!!total!expense!ratio,!the!sum!of!management!costs,!interest!costs!and!other!costs!and!taxes!
recorded!in!the!annual!statement,!as!a!percentage!of!the!average!net!asset!value!(NAV),!which!is!
defined!as!the!market!value!of!the!fund’s!assets!minus!the!annual!fee,!divided!by!the!number!of!fund!
shares.!Net!asset!value!excludes!trading!charges!(courtage)!and!results!based!fees.!
TKA!=!total!cost!share,!TER!plus!trading!charges!and!results!based!fees,!as!a!percentage!of!the!average!
net!asset!value.!
Other!costs!=!external!accountant!fees,!fees!to!the!Swedish!Financial!Supervisory!Authority,!costs!for!
asset!keeping!and!some!minor!administrative!costs.!
!
!
19!
!
Table(3.((Net(excess(returns((net(alphas)(for(actively(managed(
funds(
!
(Portfolio!
! !1999B2013(
(
1Ifactor!
!
!
!
All!funds!
!!!
I0,70%!
1st!decile!
4,48%!
!
2nd!decile!
2,06%!
!
3rd!decile!
0,73%!
!
4th!decile!
I0,45%!
!
5th!decile!
I0,83%!
!
6th!decile!
I1,24%!
!
7th!decile!
I1,53%!
!
8th!decile!
I2,06%!
!
9th!decile!
I3,24%!
10th!decile!
!!!
I4,96%!
1st!fund!
8,89%!
!
2nd!fund!
7,68%!
!
3rd!fund!
4,34%!
!
4th!fund!
4,08%!
!
5th!fund!
3,65%!
5th!from!last!
!
fund!
I4,79%!
4th!from!last!
!
fund!
I6,07%! I6,05%! I6,33%!
I4,81%!
3rd!from!last!
!
fund!
I6,15%! I6,14%! I6,50%!
I4,95%!
2nd!from!last!
!
fund!
I6,56%! I6,88%! I6,99%!
I5,39%!
I
I
I
!
Last!fund!
15,15%! 12,93%! 15,30%!
I7,09%!
!
!
!
Notes:!The!1999I2009!sample!consists!of!115!funds,!implying!that!each!decile!contains!eleven!or!twelve!
funds!rounded!to!nearest!integer.!The!1999I2013!sample!comprises!of!101!funds,!implying!that!each!
decile!contains!ten!or!eleven!funds.!The!sample!includes!returns!until!September!2013.!
!
! !
!
!
1999B2009(
( 1Ifactor! 3Ifactor! 4Ifactor!
!
!!! I0,68%!
!
!
!
I0,53%!
I0,45%!
5,18%!
5,11%!
5,91%!
! 3,02%!
3,00%!
3,13%!
! 1,31%!
1,56%!
1,58%!
! 0,04%!
0,19%!
0,12%!
! I0,57%! I0,36%! I0,47%!
! I1,13%! I1,00%! I0,86%!
! I1,86%! I1,78%! I1,51%!
! I2,78%! I2,52%! I2,36%!
! I3,99%! I3,69%! I3,91%!
!!! I6,28%! I6,12%! I6,49%!
10,29%! 9,88%! 13,60%!
! 8,66%!
8,04%!
8,11%!
! 5,46%!
5,68%!
8,00%!
! 5,06%!
5,28%!
5,50%!
! 5,03%!
5,11%!
5,18%!
!
I5,48%! I5,89%! I5,84%!
!
20!
!
!Table(4.(((Gross(excess(returns((gross(alphas)(for(actively(
! !
!
!
! !
managed(funds(
!
!
!
!
!
! !1999B
1999B2009(
2013(
(
( 1Ifactor! 3Ifactor! 4Ifactor! ( 1Ifactor!
Portfolio!
!
!
!
!!! 0,63%!
!
!
!
!
All!funds!
0,78%!
0,90%!
!!! 0,56%!
1st!decile!
6,00%!
5,99%!
7,07%!
5,40%!
!
!
2nd!decile!
4,46%!
4,48%!
4,57%!
3,55%!
! 2,76%!
3rd!decile!
2,92%!
2,97%! ! 1,91%!
! 1,45%!
4th!decile!
1,67%!
1,44%! ! 0,91%!
! 0,84%!
5th!decile!
0,93%!
0,80%! ! 0,47%!
! 0,17%!
6th!decile!
0,32%!
0,45%! ! 0,04%!
! I0,48%! I0,33%! I0,12%! ! I0,33%!
7th!decile!
! I1,37%! I1,13%! I0,85%! ! I0,86%!
8th!decile!
! I2,67%! I2,36%! I2,38%! ! I1,86%!
9th!decile!
10th!decile!
!!! I5,17%! I5,06%! I5,35%! !!! I3,64%!
1st!fund!
10,41%! 9,79%! 15,21%!
9,32%!
!
!
2nd!fund!
6,96%!
6,78%!
9,86%!
5,59%!
! 6,28%!
3rd!fund!
6,14%!
7,00%! ! 5,57%!
! 5,71%!
4th!fund!
6,11%!
6,25%! ! 5,39%!
! 5,59%!
5th!fund!
5,76%!
5,93%! ! 4,90%!
!
5th!from!last!fund! I4,40%! I4,19%! I4,42%! ! I3,80%!
4th!from!last!fund! I4,62%! I4,79%! I4,98%! ! I3,86%!
3rd!from!last!fund! I4,86%! I5,24%! I5,41%! ! I4,15%!
2nd!from!last!fund! I5,06%! I5,28%! I5,52%! ! I4,40%!
I
I
I
!
Last!fund!
13,62%! 11,39%! 13,76%!
I4,66%!
!
!
!Notes:!The!1999I2009!sample!consists!of!105!funds,!implying!
! !
!
!
! !
that!each!decile!contains!ten!or!eleven!funds!rounded!to!
nearest!integer.!The!1999I2013!sample!consists!of!91!funds,!
implying!that!each!decile!contains!nine!or!ten!funds.!The!sample!
includes!returns!until!September!2013.!!
!
!
!
21!
!
Table(5.(((BenchmarkBbased(1Bfactor(net(and(gross(alphas(for(index(
funds(
!
!
Fund!
!1st!fund!
2nd!fund!
3rd!fund!
4th!fund!
5th!fund!
6th!fund!
7th!fund!
8th!fund!
9th!fund!
10th!fund!
11th!fund!
12th!fund!
13th!fund!
14th!fund!
15th!fund!
!
Avg.,!all!funds!
Avg.,!excl.!15th!
fund!
! !
!
!
1999I2009!
!
Net!
alpha!!
TER! Gross!alpha!
!
! ! 1,35%! !0,50%! ! 1,85%!
!
0,28%! 0,49%!
0,77%!
! I0,17%! 0,73%!
0,56%!
! I0,28%! 0,83%!
0,54%!
! I0,29%! 0,00%!
I0,29%!
! I0,35%! 0,40%!
0,05%!
! I0,35%! 0,74%!
0,39%!
! I0,38%! 0,50%!
0,12%!
! I0,60%! 0,30%!
I0,30%!
! I0,66%! 0,00%!
I0,66%!
! I0,81%! 0,31%!
I0,50%!
! I1,00%! 0,53%!
I0,46%!
! I1,51%! 0,30%!
I1,21%!
! I1,67%! 0,30%!
I1,37%!
! I4,12%! 0,00%!
I4,12%!
!
! ! I0,70%! !0,40%! ! I0,31%!
!
I0,46%! 0,42%!
I0,04%!
! !1999I2013!
!
Net!alpha!
!
! ! 0,49%!
!
0,10%!
!
I0,08%!
!
I!
!
I0,36%!
!
I0,48%!
!
I0,32%!
!
I0,42%!
!
I0,50%!
!
I0,71%!
!
I0,82%!
!
I0,75%!
!
I1,31%!
!
I!
!
I!
!
! ! I0,43%!
!
! !
!Note:!Funds!are!sorted!on!their!net!1Ifactor!alphas.!The!correlation!
! !
!
!
! !
between!net!and!gross!alpha!in!the!period!1999I2009!is!0.985.!
!
!
!
22!
!
Table(6.(((Net(and(gross(4Bfactor(alphas(for(index(funds,(1999(
B(2009(
!Portfolio!
! !Net!alpha! ! TER! !Gross!alpha!
!
!
! ! I1,31%! !0,40%! ! I0,92%!
All!funds!
!
SIX30RX!benchmark!
0,37%!
0,00%!
0,37%!
!
OMXS30!benchmark!
I1,47%! 0,33%!
I1,14%!
!
SIXRX!benchmarks!
I0,71%! 0,66%!
I0,06%!
!!! I1,90%! 0,41%!
Other!benchmarks!
I1,49%!
1st!fund!
0,37%!
0,00%!
0,37%!
!
2nd!fund!
I0,35%! 0,73%!
0,38%!
!
3rd!fund!
I0,51%! 0,40%!
I0,11%!
!
4th!fund!
I0,64%! 0,53%!
I0,11%!
!
5th!fund!
I0,73%! 0,50%!
I0,23%!
!
6th!fund!
I0,94%! 0,50%!
I0,44%!
!
7th!fund!
I1,06%! 0,74%!
I0,32%!
!
8th!fund!
I1,13%! 0,30%!
I0,83%!
!
9th!fund!
I1,13%! 0,31%!
I0,82%!
!
10th!fund!
I1,51%! 0,83%!
I0,68%!
!
11th!fund!
I1,55%! 0,30%!
I1,25%!
!
12th!fund!
I1,85%! 0,00%!
I1,85%!
!
13th!fund!
I2,02%! 0,49%!
I1,53%!
!
14th!fund!
I2,60%! 0,30%!
I2,30%!
!
15th!fund!
I4,01%! 0,00%!
I4,01%!
!
!
! !
!
!
Note:!The!first!row!(All!funds)!reports!the!EW!average!of!all!
15!funds.!The!second!row!reports!the!EW!average!for!the!
single!fund!that!uses!SIX30RX!as!benchmark!index.!The!third!
row!reports!the!EW!average!for!the!sevens!funds!that!use!
some!version!of!OMXS30!as!their!benchmark.!The!fourth!row!
reports!the!EW!average!for!the!three!funds!that!use!SIXRX!as!
benchmark.!The!fifth!row!reports!the!EW!average!for!the!
remaining!four!funds.!
!
!
!
!
!
23!
!
Table(7.((Actual(and(bootstrapped(tBstatistics,(4Bfactor(model,(gross(returns,(
actively(managed(funds((1999B2009)(
!
Portfolio!
Actual!t!
Bootstrapped!t!
|Actual|!<!|Bootstrapped|!
1st!decile!
2,39!
1,73!
11,8%!
2nd!decile!
1,56!
1,01!
11,2%!
3rd!decile!
1,18!
0,65!
11,2%!
4th!decile!
0,88!
0,37!
11,6%!
5th!decile!
0,58!
0,12!
13,6%!
6th!decile!
7th!decile!
8th!decile!
9th!decile!
10th!decile!
0,27!
I0,07!
I0,54!
I1,21!
I2,22!
I0,13!
I0,38!
I0,65!
I1,00!
I1,67!
18,2%!
59,0%!
57,7%!
33,1%!
14,9%!
1st!fund!
3,37!
2,47!
11,1%!
2nd!fund!
3,30!
2,21!
5,89%!
3rd!fund!
2,58!
1,95!
13,2%!
4th!fund!
2,51!
1,80!
9,92%!
5th!fund!
2,19!
1,70!
17,2%!
101th!fund!
I2,35!
I1,66!
9,34%!
102th!fund!
I2,42!
I1,77!
10,9%!
103th!fund!
I2,43!
I1,92!
16,4%!
104th!fund!
I2,47!
I2,13!
25,7%!
105th!fund!
I3,29!
I2,54!
13,7%!
!
!
!
!
Note:!Each!decile!consists!of!10!or!11!funds!(rounded!to!nearest!integer).!
!
!
24!
!
!
Table(A1.(((EquallyBweighted(net(excess(returns((net(
alphas)(for(actively(managed(funds(
(((((((((((((((((((((
!
Year!
!
1999!
2000!
!
67!
84!
2001!
90!
2002!
2003!
94!
102!
2004!
104!
2005!
105!
2006!
108!
2007!
106!
2008!
102!
2009!
101!
2010!
90!
2011!
2012!
2013!
86!
79!
68!
!Mean,!1999!I!
2009!
!
!
!
! No.!of!
funds!
!
! 1I
! 3I
! 4I
factor! factor! factor!
I!
7,95%!
5,39%!
I
1,08%!
I
2,22%!
1,93%!
I
5,63%!
I
0,71%!
I
0,40%!
I
2,05%!
I
2,43%!
2,15%!
I
2,73%!
I
1,21%!
14,6%!
0,42%!
!
96,6!!
I
1,18%!
I!
0,18%!
4,26%!
I
1,93%!
I
3,04%!
7,17%!
I
8,44%!
I
2,59%!
I!
4,27%!
3,07%!
0,76%!
I
2,70%!
I
1,63%!
I
0,68%!
0,56%!
I
4,42%!
I!
I!
I!
I!
I!
I!
I!
I!
0,76%!
I
3,02%!
7,77%!
I
9,35%!
I
2,48%!
0,65%!
I
1,54%!
! I
! I
0,82%! 1,12%!
!
!
!
!
!
Notes:!!Each!fund!alpha!is!estimated!on!calendar!year!data!
(i.e.!between!two!and!twelve!monthly!return!
observations).!2013!includes!returns!until!September.!
Yearly!returns!are!the!arithmetic!mean!of!monthly!returns.!
Variable! &
&
(1)!
;!
0.839***!
(0.0037)!
;!
!
!;0.0009!
&
(2)!
!
;!
(0.0038)!
;!
!
0.0003!
&
(3)!
&
(4)!
!
;!
&
(5)!
&
(6)!
!
;!
&
(7)!
!
;!
! (8)!
!
;!
! (9)!
! (10)!
!
;!
!!
! (11)!
!
!
!
!
!
!
0,000
!
!;!
!
!
!
;0.0052! !;0.0081! !0.016!
6!
0.0038! !;0.0058! ;0.0052! !;0.0059!
(0.0038 (0.0041 (0.013
(0.0047 (0.0037 (0.0076
)!
)!
)!
;0,012! (0.011)!
)!
)!
)!
;!
;!
;1.460!
;!
;!
;!
;!
(0.916
!
)!
!
!
;!
Table&A2.&&&Regressions&of&actively&managed&funds'&net&4;factor&alphas&on&fees&and&expenses,&1999&;&2009&
!
Constant!
!
Annual!fee!
;!
!
0.0014!
(0.011
)!
;0.489!
(0.839
)!
!
Entry!fee!
;0.740!
(0.539
)!
!
;!
(0.266)!
!
;!
!
;!
;!
0.932***!
!
;!
;!
!
;!
!
;!
!
;!
!
Exit!fee!
;!!
!
!
!
0.0297!
;!!
(0.289)!
;!
;!!
;!!
;!!
;!!
;!!
!
Performance!fee!
!
3.484*!
!
!
(0.069)!
;!
;!!
;!!
;0.254!
(0.582
)!
;0.038!
(0.072
)!
3.042!
(2.034
)!
;!
;!!
;!!
(2.035)!
;!
!
;0.400!
;0,824!
;!
!
;0.0744!
(0.639)!
;!
;!!
;!!
;!!
;!!
;!!
!
;!!
!
Other!costs!
;!!
;!!
;!!
!
;!!
;!!
;!!
;!!
!
;!!
;!!
;!!
;!!
!
;!!
!
TER!
;!!
;!!
!
;!!
;!!
;!!
!
;!!
!
TKA!
;!!
!
;!!
!
TKA;TER!
!
!
!
!
!
!
!
!
!
!
!0.317!
(1.219)!
!
!
!
!
TKA;TER;Perf.fee!
!
Transaction!cost!
!
!
R;squared!
!
Observations!
!
;!!
!
!
0.096!
!
;!!
!
!
0.002!
!
;!!
!
!
0.029!
!
;!!
!
!
0.141!
!
;!!
!
0,002
!
4!
!
;!!
!
99!
!
!
0.0001!
!
;!!
!
99!
!
!
0.0007!
!
;!!
!
99!
!
!
0.0005!
;0.016!
(0.071)!
;!
!
91!
!
0.0001!
1!
!
;!!
!
!
0.092!
!
100!
!
!
;0.508!
(4.47)!
!
!
0.0035!
!
100!
!
!
100!
!
!
100!
!
!
100!
!
!
100!
!
!
100!
!
!
!
!
!
!
!
Note:!Types!of!fees!not!charged!have!been!set!to!zero!(rather!than!missing).!
!
Figure!1!)!Net!and!gross!alphas!with!parametric!confidence!intervals!
25
25
20
20
15
15
10
10
Gross Alpha (% pear year)
Net Alpha (% per year)
Point estimate
80% confidence
90% confidence
95% confidence
5
0
−5
5
0
−5
−10
−10
−15
−15
−20
−20
−25
10
20
30
40
50
60
70
Fund Ranking (1 to 115)
80
90
100
110
−25
10
20
30
40
50
60
70
80
90
100
Fund Ranking (1 to 105)
!
!
!
1!
!
!
Figure!2!)!Performance!persistence!by!ranking!period!decile!
20%%
Decile%1%
15%%
Decile%2%
Decile%3%
10%%
Decile%4%
05%%
Decile%5%
Decile%6%
00%%
Decile%7%
Decile%8%
!05%%
Decile%9%
!10%%
Decile%10%
Ranking%
period%
!
1%year%
2%years%
3%years%
4%years%
5%years%
!
!
2!
!
Figure!3!)!Performance!persistence!among!the!five!best!and!five!worst!funds!
30%%
25%%
1st%fund%
20%%
2nd%fund%
15%%
3rd%fund%
10%%
4th%fund%
05%%
5th%fund%
00%%
5th%worst%fund%
!05%%
4th%worst%fund%
!10%%
3rd%worst%fund%
!15%%
2nd%worst%fund%
Worst%fund%
!20%%
Ranking%
period%
!
1%year%
2%years%
3%years%
4%years%
5%years%
!
!
3!
!
Figure!4!)!Actual!and!bootstrapped!t)statistics!of!4)factor!alpha!(net)!
5
Point estimate
80% confidence
90% confidence
95% confidence
4
3
t of net alpha
2
1
0
−1
−2
−3
−4
−5
10
20
30
40
50
60
70
80
90
100
110
Fund Ranking
!
!
!
4!
!
Figure!5!)!Actual!and!bootstrapped!t)statistics!of!4)factor!alphas!(gross)!
5
Point estimate
80% confidence
90% confidence
95% confidence
4
3
t of gross alpha
2
1
0
−1
−2
−3
−4
−5
10
20
30
40
50
60
70
80
90
100
Fund Ranking
!
!
!
5!
!
Figure!6!)!Actual!and!bootstrapped!4)factor!alphas!(net)!
30
Point estimate
80% confidence
90% confidence
95% confidence
Net Alpha (% per year)
20
10
0
−10
−20
−30
10
20
30
40
50
60
70
80
90
100
110
Fund Ranking (1 to 115)
!
!
!
6!
!
Figure!7!)!Actual!and!bootstrapped!4)factor!alphas!(gross)!
30
Point estimate
80% confidence
90% confidence
95% confidence
Gross Alpha (% pear year)
20
10
0
−10
−20
−30
10
20
30
40
50
60
70
80
90
100
Fund Ranking (1 to 105)
!