And eu, Lau a; Fo ne , Ca los; Sa o, José Luis
A icle
Window d essing in he Ac i e Sha e sco es in publicly
epo ed po olios
BRQ Business Resea ch Qua e ly
P o ided in Coope a ion wi h:
Asociación Cien í ica de Economía y Di ección de Emp esas (ACEDE), Mad id
Sugges ed Ci a ion: And eu, Lau a; Fo ne , Ca los; Sa o, José Luis (2024) : Window d essing in
he Ac i e Sha e sco es in publicly epo ed po olios, BRQ Business Resea ch Qua e ly, ISSN
2340-9444, Sage Publishing, London, Vol. 27, Iss. 3, pp. 324-345,
h ps://doi.o g/10.1177/23409444211024645
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/327047
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h ps://c ea i ecommons.o g/licenses/by-nc/4.0/
h ps://doi.o g/10.1177/23409444211024645
Business Resea ch Qua e ly
2024, Vol. 27(3) 324 –345
© The Au ho (s) 2021
A icle euse guidelines:
sagepub.com/jou nals-pe missions
DOI: 10.1177/23409444211024645
jou nals.sagepub.com/home/b q
C ea i e Commons Non Comme cial CC BY-NC: This a icle is dis ibu ed unde he e ms o he C ea i e Commons
A ibu ion-NonComme cial 4.0 License (h ps://c ea i ecommons.o g/licenses/by-nc/4.0/) which pe mi s non-comme cial use,
ep oduc ion and dis ibu ion o he wo k wi hou u he pe mission p o ided he o iginal wo k is a ibu ed as speci ied on he SAGE and
Open Access page (h ps://uk.sagepub.com/abou us/openaccess.h m).
In oduc ion
The exis ence o window d essing (WD) in po olio man-
agemen has a ac ed inc easing in e es among academics
and p ac i ione s because o i s manage ial consequences.
P e ious li e a u e desc ibes WD as a managemen s a egy
in ended o include a ac i e asse s (ho s ocks) and/o p e-
en una ac i e asse s in he epo ed po olios o d aw
highe money lows. Di e en measu es ha e been p o-
posed o cap u e his phenomenon, depending on he ea-
sons o WD beha io (see Aga wal e al., 2014).
In his s udy, we analyze a di e en ype o WD whe e
manage s a emp o inc ease he di e ence in he po olio
weigh s be ween hei epo ed po olio and he bench-
ma k. Ac i ely managed unds o e po olios ha a emp
o ou pe o m a e e ence benchma k by ac i ely de ia ing
om he benchma k weigh s o e weigh ing (unde weigh -
ing) he s ocks ha he manage conside ed good (bad)
in es men s. Ac i e mu ual unds cha ge highe ees o his
ac i e managemen . Mo eo e , in es o s could conside
la ge de ia ions om he benchma k a p oxy o skill and
consequen ly indica e highe p obabili ies o be e u u e
pe o mance. The e o e, we sugges ha und manage s
could be emp ed o a i icially inc ease he de ia ion o
hei po olio weigh s om he benchma k weigh s in he
mon hs when hei po olio holdings a e made public o
gi e an ex a image o ac i eness. Wi h his WD s a egy,
manage s could be e jus i y hei high ees and also po ay
a highe image o skill o po en ial in es o s, which could
consequen ly inc ease hei money lows.
We use he Ac i e Sha e (AS) measu e p oposed by
C eme s and Pe ajis o (2009) as an agg ega e measu e o
how und holdings de ia e om he benchma k. This
measu e adds he absolu e alue o all he und s ock
weigh de ia ions om he benchma k weigh s. Al hough
he wo k o C eme s and Pe ajis o (2009) was published
a e ou sample pe iod (2000–2006), he AS me ic is so
Window d essing in he Ac i e Sha e
sco es in publicly epo ed po olios
Lau a And eu1, Ca los Fo ne 2 and José Luis Sa o1
Abs ac
Using a unique da abase ha includes publicly disclosed und holdings a he end o he qua e as well as he holdings
in all non-publicly disclosed mon hs, we ound ha some unds could al e hei po olios in publicly disclosed mon hs
o a i icially inc ease hei Ac i e Sha e sco es and consequen ly appea mo e ac i e and ake ad an age o he posi i e
ela ionship be ween Ac i e Sha e and money lows. We show how, consis en wi h non-in o med ades, hese unds
e ode hei u u e pe o mance. Howe e , hese unds each hei objec i e o inc easing u u e money lows. Mo eo e ,
we ind ha window-d esse unds can be iden i ied by con olling he le el o acking e o . The unds wi h high Ac i e
Sha e sco es and low acking e o s ha e he highes le els o Ac i e Sha e window d essing and he wo s u u e
e u ns. Howe e , compa ed wi h less ac i e unds, hey a e able o cap u e highe money lows.
JEL CLASSIFICATION: G23; G11
Keywo ds
Ac i e Sha e, mon hly und po olio holdings, acking e o , window d essing
1 Depa men o Accoun ing and Finance, Uni e si y o Za agoza,
Za agoza, Spain
2
Depa men o Accoun ing and Finance, Uni e si y o Alican e,
Alican e, Spain
Co esponding au ho :
Lau a And eu, Depa men o Accoun ing and Finance, Uni e si y o
Za agoza, G an Vía 2, 50005 Za agoza, Spain.
Email: [email p o ec ed]
1024645BRQ0010.1177/23409444211024645Business Resea ch Qua e lyAnd eu e al.
esea ch-a icle2021
Regula Pape
And eu e al. 325
in ui i e ha we conside i a good p oxy o how in es o s
compu e agg ega e und holding de ia ions om he
benchma k in he sample pe iod.
“Ho s ocks” and “AS” WD p ac ices e e o po olio
holding mo emen s be o e he und po olio becomes
public wi h he same inal objec i e o inc easing money
lows. Howe e , hey a e di e en . No e ha when “ho
s ocks” WD manage s buy (sell) a ac i e (una ac i e)
s ocks, which a e o e weigh ed (unde weigh ed) in he
po olio, his ade will necessa ily inc ease s ock de ia-
ion om benchma k, and he e o e inc ease AS le els.
Howe e , when “ho s ocks” WD manage s buy (sell)
a ac i e (una ac i e) s ocks which a e unde weigh ed
(o e weigh ed) in he po olio, his ade could educe
he s ock de ia ion om benchma k, and he e o e educe
AS le els.
To compu e AS WD, i is necessa y o ha e a da abase
ha includes und holdings bo h in he mon hs du ing
which he und po olios a e made public o in es o s and
he mon hs du ing which hese und po olios a e no
made public. We employ a unique da abase o he Spanish
mu ual und ma ke ha includes bo h publicly disclosed
und holdings a he end o he qua e (Ma ch, June,
Sep embe , o Decembe ) and he und holdings in all non-
publicly disclosed mon hs (non-qua e -end mon hs). This
in o ma ion is no publicly a ailable o in es o s as he
Spanish o icial supe iso p o ides non-qua e -end und
holdings o esea ch pu poses only. This makes his da a-
base especially use ul o ou aim. Mo eo e , his da abase
is ee o any selec ion bias, gi en ha his in o ma ion is
a ailable o all mu ual unds and is p o ided by he o icial
supe iso a he han a p i a e supplie .
The quali y o he in o ma ion a ailable in he Spanish
ma ke allows us o de ec he po en ial WD in he AS le -
els in publicly epo ed mon hs, as we can compu e how
he AS in public qua e -end mon hs di e s om he AS in
he wo p e ious and he ollowing wo non-public mon hs.
We e e o his AS WD p oxy as AS_di e ence. This
s udy canno be ex ended o o he eu ozone ma ke s.1
Howe e , ou esul s a e ele an o he wo ldwide mu ual
und indus y. C eme s and Pe ajis o (2009) a gue ha AS
sco es based on epo ed und holdings a he end o a
qua e a e unlikely o su e any po en ial WD because he
po olio dis o ion supposes a high u no e cos .
Howe e , e idence o AS WD in he Spanish mu ual und
ma ke would mean ha und manage s wo ldwide could
ha e enough incen i es o al e hei po olios despi e he
u no e cos s. Mo eo e , he Spanish domes ic mu ual
und ma ke exhibi s some cha ac e is ics (small size o
he Spanish S ock Exchange) ha make i di icul o
manage s o de ia e om he benchma k. I Spanish man-
age s ind enough incen i es o al e hei AS despi e he
di icul ies o de ia ing om he benchma k, manage s
om o he coun ies wi h ewe di icul ies could ha e
e en mo e incen i es o window d ess he le el o AS.
This s udy’s main con ibu ion is es ing, o he i s
ime in li e a u e, whe he mu ual unds al e hei hold-
ings in publicly disclosed mon hs o appea mo e ac i e
and how his WD s a egy in luences u u e und pe o -
mance and money lows. We also con ibu e o he WD
li e a u e by examining he undisclosed in o ma ion. Few
s udies (e.g., El on e al., 2010; Mo ey & O’Neal, 2006;
Mus o, 1997; O iz e al., 2012) a e able o use bo h dis-
closed and undisclosed in o ma ion. Howe e , in mos
cases, he e exis s some epo ing selec ion bias.
In ou s udy, he incen i es o al e AS in publicly epo ed
mon hs depend on whe he he AS sco e is posi i ely ela ed
o he u u e pe o mance o und (manage skills) and u u e
money lows. In he i s pa o he s udy, we demons a e
ha he highe he AS le el, he highe he u u e und e u ns
and money lows. Mo eo e , we also in es iga e whe he his
ela ion is obus o he inclusion o acking e o (TE) as an
al e na i e measu e o und ac i eness.
I some manage s al e hei AS le els, hey will
inc ease his measu e be o e publicly disclosed mon hs
and e e o hei p e ious le els a e wa d. In his case,
manage s de ia e he po olio holdings om he bench-
ma k only o o e a empo al highe le el o ac i eness
when hey a e going o make he in o ma ion public, and
no because he manage is o e weigh ing (unde weigh -
ing) s ocks ha , acco ding o he manage ’s skill, a e going
o be winne s (lose s). This AS inc ease is a i icial and no
ela ed o skill. In addi ion, i AS is a p oxy o manage-
men skills, he skills should no di e be ween publicly
and non-publicly disclosed mon hs. The e o e, his WD
s a egy will e ode und pe o mance because o he
inc ease in u no e and expense a ios. Howe e , i some
manage s decide o modi y he AS le el despi e i s
expec ed nega i e impac on und pe o mance, i would
be because his p ac ice allows o u u e money lows o
inc ease. Consis en wi h his idea, we ind ha high le els
o AS in publicly epo ed mon hs ega ding he AS le els
in he wo p e ious and he ollowing wo non-publicly
epo ed mon hs (AS_di e ence) a e ela ed o wo se
u u e und pe o mance and be e u u e money lows.
Ou analysis also ocuses on de ec ing unds ha a e
mo e likely o dis o he AS le el. Gi en ha TE canno
be modi ied easily, we sugges ha unds wi h high AS bu
low TE le els a e mo e likely o al e hei AS le els. We
consis en ly ind ha con a y o he o he ca ego ies o
unds, on a e age, unds wi h high AS bu low TE le els
inc ease hei AS le els be o e hei holdings a e made
public and educe hem a e wa d. They he e o e show
signi ican and posi i e AS_di e ence alues. Mo eo e ,
his ca ego y o unds shows he wo s u u e pe o mance
bu a ac s highe money lows han unds wi h simila
le els o TE.
Finally, gi en he indings o Golez and Ma in (2015),
who obse e ha Spanish bank-a ilia ed mu ual unds
sys ema ically inc ease hei holdings in he con olling
326 Business Resea ch Qua e ly 27(3)
bank s ock a ound some e en s, we also analyze he pos-
sible AS WD s a egies o subsamples o non-independ-
en unds ( unds managed by managemen companies ha
a e in eg a ed in a inancial g oup conglome a e) and inde-
penden unds. We consis en ly ind ha he nega i e
impac o AS WD a qua e -ends on und pe o mance is
exclusi e o non-independen unds. This nega i e impac
is no obse ed in independen unds.
The es o his a icle is o ganized as ollows. The
“Li e a u e e iew and hypo heses” sec ion p esen s he li -
e a u e e iew and he hypo heses o he s udy. The “Da a
and measu es” sec ion desc ibes he da a and measu es used
in his s udy. The “Empi ical models and me hods” sec ion
desc ibes he empi ical models and me hods used o es he
hypo heses o his s udy. The “Main esul s” sec ion p esen s
he main indings o ou hypo heses. The “Cha ac e is ics o
s ocks aded by AS window d esse s: Is AS WD a p ac ice
di e en om ‘ho s ock’ WD?” sec ion examines he cha -
ac e is ics o s ocks aded by AS window d esse s. The sec-
ion “Examining he unds ha a e mo e p one o window
d ess he AS le el” examines he cha ac e is ics o unds ha
a e mo e likely o al e he AS le el. The “Robus ness analy-
ses” sec ion p esen s some obus ness analyses. The
“Conclusion” sec ion concludes he s udy.
Li e a u e e iew and hypo heses
The in o ma ion con en in publicly disclosed und holdings
has a ac ed he in e es o many s udies. One impo an line
o esea ch, which is ela ed o he exis ence o a con lic o
in e es be ween und manage s and in es o s, is he analysis
o WD in und holdings. Haugen and Lakonishok (1988) and
Ng and Wang (2004) conside WD in und holdings as he
main ac o ha d i es high e u ns o small and ecen lose
s ocks a e he yea -end. Mos s udies examine he in luence
o his p ac ice on p ice anomalies, such as he Janua y e ec ,
a he han examining he exis ence o his p ac ice and i s
unde lying mo i a ions (see, o example, Aga wal e al.,
2014). Speci ically, Aga wal e al. (2014) highligh ha he e
is limi ed unde s anding o he incen i es o manage s o
engage in WD. They a i m ha i WD is ewa ded wi h
highe lows, one migh wonde why no all und manage s
ollow such ac i i ies. To disen angle his p oblem, hese
au ho s sugges ha WD is a isky decision because i s suc-
cess in a ac ing highe lows depends on und pe o mance
in he delay pe iod be ween he end o he mon h when he
po olio is publicly disclosed and he da e when his in o -
ma ion is publicly disclosed o in es o s.2
Hence, he main mo i a ion o his p ac ice is he pe -
cep ion o manage s ha po olio disclosu es signi ican ly
in luence he opinions o in es o s ega ding hei p o es-
sional skills. Manage s a e emp ed o imp o e he appea -
ance o hei epo ed po olio (i.e., including/dele ing
a ac i e/una ac i e asse s) be o e p esen ing hem o
clien s o sha eholde s o a ac la ge money lows. The
exis ence o WD is p oblema ic o in es o s who seek
eliable in o ma ion on und holdings ha will help hem
alloca e hei money mo e app op ia ely.
Ano he line o esea ch ela ed o he in o ma ion con-
en in public und holding disclosu es uses hese da a o
measu e he le el o ac i e managemen and i s ela ion-
ship wi h und pe o mance. The i s s udy o analyze his
ela ion was We me s (2003) which does no measu e he
le el o ac i e managemen using holding disclosu es bu
uses e u n in o ma ion h ough he TE. Mo e ecen ly,
o he s udies ha e s a ed analyzing ac i e managemen
using po olio holdings in o ma ion, which “a p io i” p o-
ides a mo e comp ehensi e image o he ac i eness o
po olio manage s. Kacpe czyk e al. (2005) examined
ac i e managemen using po olio concen a ion. Mo e
ecen ly, C eme s and Pe ajis o (2009) analyzed how und
holdings de ia e om benchma ks h ough he AS meas-
u e. These s udies ound a posi i e ela ionship be ween
he le el o ac i e managemen and und pe o mance.
The AS measu e has a ac ed subs an ial a en ion bo h
in academia and in he managemen indus y. Mo eo e ,
mo e ac i e equi y unds and ins i u ional money manag-
e s cu en ly epo hei AS. An example o his impac is
he a icle published by F azzini e al. (2016) which
e iews he g ea impac o AS on he in es men commu-
ni y. They indica e ha ins i u ional in es o s a e mo e
ocused on asse manage s wi h a high AS and ha some
ins i u ional in es o s ha e e en embedded a high AS
equi emen in hei in es men guidelines. Ano he exam-
ple o he impac o AS in he und communi y is Caquineau
e al. (2016) om Mo nings a , who ad ise in es o s no o
ely solely on AS when selec ing unds.
In his s udy, we link bo h lines o esea ch (WD and
ac i e managemen ) by examining a di e en p ac ice o
WD, ha is, he epo o a high le el o di e en ia ion
om he benchma k on disclosu e da es (AS WD) in he
Spanish equi y ma ke .3 Gi en ha he incen i es o dis o
he AS le el depend on whe he high AS le els in publicly
epo ed po olios send a posi i e signal o po en ial
in es o s, we i s analyze whe he he US e idence o he
e ec o AS on u u e und e u ns is obse ed in he
Spanish ma ke . Hence, ou baseline hypo hesis is as
ollows:
Hypo hesis 1 (H1): AS le el posi i ely in luences sub-
sequen und pe o mance.
We hen con ibu e o he li e a u e examining he ollow-
ing no el esea ch hypo heses:
Hypo hesis 2 (H2): AS le el posi i ely in luences sub-
sequen money lows.
Hypo hesis 3 (H3): AS WD nega i ely in luences sub-
sequen und pe o mance.
And eu e al. 327
Hypo hesis 4 (H4): AS WD posi i ely in luences sub-
sequen money lows.
Da a and measu es
To analyze WD in ol ing how und holdings de ia e om
benchma ks, we use he AS measu e p oposed by C eme s
and Pe ajis o (2009), which compa es he holdings o a
mu ual und wi h he holdings o i s benchma k index.
Speci ically, he AS o a gi en mu ual und p in a gi en
mon h is de ined as ollows
AS
ww
p pi bi
i
N
,,
,
=−
=
∑
1
21
(1)
whe e wpi, and wbi, a e he po olio weigh s o asse i in
mu ual und p and benchma k b.4
We also compu e he adi ional TE measu e o und
ac i eness o be used as a con ol a iable in ou analysis
o AS WD. The TE measu e is commonly de ined as he
ime-se ies s anda d de ia ion o he di e ence be ween a
und e u n
()
,
Rp and i s benchma k e u n
()
,
Rb . The
TE o a gi en mu ual und p in a gi en mon h is calcu-
la ed as ollows
TE S de R R
p p b
,,
,
=−
(2)
The TE a he end o a gi en calenda mon h was calcu-
la ed using he ime se ies o he p e ious 120 daily e u ns.
I was la e annualized. As expec ed, he AS and TE meas-
u es we e posi i ely co ela ed (.5046). Al e na i ely, we
also calcula e he TE by eg essing he excess und e u ns
o e he isk- ee a e on excess benchma k e u ns, as
ollows5
RR RR
p pb p p ,, ,, ,
−=+−
+
αβ
ε
(3)
TE S de
p
be a
p
,,
=
ε
(4)
To compu e he AS o mu ual unds, we equi e da a on
he po olio composi ion o mu ual unds, as well as hei
benchma ks. The mu ual und holdings da abase used in
his s udy comp ises he mon hly po olio epo s o all
ac i ely managed Spanish domes ic equi y unds. This
in o ma ion di e s om he p e ious li e a u e ha ana-
lyzes qua e ly holdings, which implies he loss o in e im
ades; he e o e, i is impossible o es a change in he
und ac i i y le el in hese in e im mon hs. The egula o
o he Spanish ma ke , he Spanish Secu i ies and Exchange
Commission (CNMV), supplied us wi h his mon hly da a-
base un il Decembe 2006, a e which he in o ma ion is
p o ided on a qua e ly basis (calenda qua e -ends).6 The
da abase is ee o su i o ship bias because i also
includes dead unds. In addi ion, he da abase o e comes
any possible p oblem o epo ing selec ion bias ha may
be p esen in he sca ce esea ch using high- equency
po olio in o ma ion, whe e mu ual unds olun a ily sup-
ply hei po olio holdings o p i a e da a p o ide s (see,
o example, El on e al., 2010).
Ou main objec i e is o es he possible exis ence o
WD a he le el o AS shown by equi y mu ual unds on
publicly epo ed da es (qua e -end mon hs). We he e o e
ocus ou s udy on he ime pe iod in which he po olio
holdings o all mu ual unds a e epo ed on a mon hly
basis (1999–2006).7
The ini ial sample included 169 Spanish domes ic
equi y mu ual unds. F om his sample, we elimina e pas-
si e managemen unds, such as index unds o exchange
aded und (ETFs), and a couple o unds o ien ed o
Small o Mid-Caps. We begin ou analysis in Decembe
2000 because o he equi emen o da a in ended o es i-
ma e he momen um ac o . The inal sample comp ises
137 unds om Decembe 2000 o Decembe 2006.
The CNMV mu ual und da abase p o ides in o ma ion
on he po olio holdings and in es men oca ions o
Spanish mu ual unds. I also con ains o he cha ac e is ics
o mu ual unds such as daily and mon hly e u ns, o al
ne asse s (TNA), numbe o in es o s, annual manage-
men ees, und age, and managemen company.
Panel A o Table 1 epo s he numbe o unds exam-
ined annually, he a e age TNA o he unds, he a e age
numbe o in es o s, he a e age numbe o s ocks held by
mu ual unds, he a e age numbe o managemen compa-
nies examined, and he a e age annual u no e a io.
Speci ically, he a e age TNA o he unds in ou sample is
€68 million, and he a e age numbe o in es o s is
app oxima ely 2,800. Fu he mo e, he a e age numbe o
secu i ies held by a und is 40.
Panel B o Table 1 epo s he sha e o und po olios in
he main ypes o secu i ies (s ocks, ixed-income secu i ies,
o he mu ual und uni s, and cash and cash equi alen s)
ac oss he yea s s udied. As expec ed, acco ding o he
in es men oca ion o he unds examined, he main in es -
men is in domes ic s ocks and, mo e p ecisely, in domes ic
s ocks belonging o he Ibex35 index. This panel also shows
ha he pe cen age in es ed in ixed income and o he
mu ual und uni s is ela i ely small. In conclusion, he low
pe cen age o non-con olled secu i ies (less han 1% o he
po olios) ein o ces he quali y o ou da abase.
We include he Ibex35 index as a benchma k o all
unds because i is he mos impo an la ge-cap index o
he Spanish s ock ma ke and is, he e o e, he mos popu-
la sel - epo ed benchma k o ou sample o Spanish
domes ic equi y unds ( hese unds in es , on a e age,
mo e han 61% o he po olio in Ibex35 s ocks, as shown
in Table 1). Fo all unds, we used he mos ele an bench-
ma k o Spanish la ge-cap equi y ins ead o using he
benchma k sel - epo ed by he manage in he und
328 Business Resea ch Qua e ly 27(3)
p ospec us because manage s can epo a misleading
benchma k ha is easily bea en (Sensoy, 2009). Fo obus -
ness, we also conduc ed wo analyses: (1) conside ing he
index ha p oduced he lowes AS be ween Ibex35 index
and IGBM (Índice Gene al de Bolsa de Mad id), as in
Pe ajis o (2013) and C eme s and Pe ajis o (2009), and (2)
limi ing he analysis o unds ha sel - epo ed Ibex35 o
IGBM as he benchma k.8
We ha e his o ical mon h-end cons i uen s o he Ibex35
index and IGBM index, and he daily and mon hly e u ns o
hei cons i uen s om Da aS eam. This his o ical in o ma-
ion allows us o conside new benchma k cons i uen s and
dele ions om he benchma ks. All s ock holdings, o bo h
mu ual unds and benchma ks, we e ma ched wi h he s ock
e u ns h ough he in e na ional secu i ies iden i ica ion
numbe ing sys em (ISIN) code o each secu i y.
Table 2 shows some desc ip i e s a is ics on he dis i-
bu ion o unds along he wo dimensions o ac i e man-
agemen (AS and TE). A each calenda qua e -end
(publicly epo ing da es), he unds a e independen ly
so ed by hei AS and TE. We used 20%, 40%, 60%, and
80% AS cu s and 4%, 8%, and 12% TE cu s (% pe yea ).
Panel A epo s he minimum, a e age, and maximum
numbe o unds in his bi a ia e dis ibu ion o he sam-
ple pe iod and in he uni a ia e ma ginal dis ibu ions
along he AS and TE dimensions.
The Spanish uni a ia e AS dis ibu ion di e s sligh ly
om ha obse ed in US equi y unds. Mos o he Spanish
domes ic equi y unds (85% o he unds on a e age) ha e
an AS lowe han 60%, which is he opposi e o he dis i-
bu ion obse ed in he US ma ke . Fo example, in 2002,
only 457 o 1,678 unds (27%) had AS le els lowe han
60% in he Uni ed S a es (Table 1, C eme s & Pe ajis o,
2009) and 82% in Spain. The e o e, he majo i y o domes-
ic equi y unds in Spain all in he ca ego y ha C eme s
and Pe ajis o (2009) e e o as close indexe s, ha is,
unds wi h low AS le els ha claim o be ac i e.
Rega ding he pe cen age o asse s unde managemen ,
he igu es om he US and Spanish ma ke s also di e
signi ican ly. Fo example, in Decembe 2006, in he
Uni ed S a es, less han 40% o o al asse s we e managed
by unds wi h AS le els lowe han 60% (Figu e 5,
Pe ajis o, 2013); hese accoun ed o 81% in he Spanish
ma ke . Simila ly, Eu opean Secu i ies and Ma ke s
Au ho i y (ESMA, 2016) analyzes he Unde akings o
he Collec i e In es men in T ans e able Secu i ies
(UCITS) equi y unds domiciled in Eu opean Union (EU)
membe s a es o he pe iod om 2005 o 2012 and inds
ha 15% o unds ha e AS le els less han 60% and TE
le els less han 4%. Howe e , his ype o und ep esen ed
43% o he Spanish sample o he same pe iod.9 ESMA
(2016) also measu es he pe cen age o unds wi h an AS
Table 1. Summa y s a is ics o Spanish domes ic equi y unds o e ime.
2000 2001 2002 2003 2004 2005 2006
Panel A
No. o unds 101 101 97 97 95 102 100
TNA (in housand eu os) 72,819 59,723 50,258 45,241 66,630 79,617 86,839
No. o in es o s 2,848 2,749 2,715 2,582 2,861 2,837 3,020
A e age numbe o s ocks held 43 39 37 36 39 41 41
No. o managemen companies 55 55 53 53 51 55 53
Tu no e 58.52% 42.84% 40.44% 37.35% 35.67% 40.70% 44.64%
Panel B
S ocks 81.69% 78.32% 78.36% 77.21% 76.52% 79.56% 82.24%
Spanish 71.40% 69.39% 70.70% 71.19% 70.62% 74.60% 77.93%
Ibex 61.42% 58.92% 58.68% 60.62% 59.73% 62.73% 65.13%
Non-Ibex 9.99% 10.48% 12.02% 10.57% 10.89% 11.88% 12.80%
Eu opean 8.76% 8.08% 7.13% 5.78% 5.74% 4.81% 4.23%
O he 1.53% 0.85% 0.53% 0.24% 0.16% 0.15% 0.08%
Fixed income 4.97% 5.38% 4.16% 3.23% 3.24% 3.67% 2.13%
O he mu ual und uni s 0.39% 0.20% 0.13% 0.16% 0.09% 0.06% 0.21%
Cash and cash equi alen s 12.01% 15.10% 16.88% 18.92% 19.72% 16.31% 15.09%
Non-con olled secu i ies 0.95% 1.00% 0.47% 0.48% 0.42% 0.39% 0.33%
To al 100.00% 100.00% 100.00% 100.00% 100.00% 100.00% 100.00%
TNA: o al ne asse s.
No e: The able is spli in o Panel A and Panel B. Panel A epo s he annual cha ac e is ics o mu ual unds, such as he numbe o unds analyzed,
he TNA, he numbe o in es o s, he a e age numbe o s ocks held by he po olios, he numbe o managemen companies in ol ed each yea ,
as well as he annual u no e . Panel B epo s he po olio sha e o und po olios o he main ypes o secu i ies ac oss he yea s. The asse s
in es ed by unds a e classi ied as ollows: s ocks (Spanish, Eu opean, and o he s), ixed income, o he mu ual und uni s, cash and cash equi alen s,
and non-con olled secu i ies. Spanish s ocks a e spli in o Ibex35 s ocks and non-Ibex35 s ocks. The da a co espond o he a e age o he
po olios ha epo each yea .
And eu e al. 329
Table 2. Desc ip i e s a is ics depending on he le el o ac i e managemen .
Ac i e Sha e T acking e o (% pe yea )
Low (<4%) 4%–8% 8%–12% High (>12%) All
Panel A: Minimum/a e age/maximum numbe o mu ual unds
High (>80%) 0/0/0 0/1/4 0/0/4 0/2/6 2/3/6
60%–80% 0/0/4 0/4/9 0/1/7 0/3/11 7/10/13
40%–60% 0/9/20 1/8/16 0/2/14 0/1/6 15/21/29
20%–40% 9/33/51 0/7/22 0/2/16 0/0/4 37/44/51
Low (<20%) 5/9/18 0/1/5 0/0/2 0/0/1 6/11/19
All 14/52/79 13/22/36 0/7/30 0/7/23
Panel B: Median TNA (in housand eu os)
High (>80%) 31,959.25 31,692.60
60%–80% 23,510.54 12,022.00 3,634.73 13,239.92
40%–60% 40,934.90 22,147.63 16,898.50 3,344.75 26,154.98
20%–40% 54,646.38 38,221.38 19,844.08 44,307.14
Low (<20%) 52,448.58 15,099.00 44,019.52
All 54,462.62 24,408.86 18,555.32 6,395.82 32,321.16
Panel C: A e age annual managemen ee (%)
High (>80%) 1.62 1.61
60%–80% 1.73 1.81 1.58 1.69
40%–60% 1.83 1.86 1.84 2.04 1.84
20%–40% 1.92 1.75 1.91 1.88
Low (<20%) 2.06 1.84 2.03
All 1.94 1.79 1.84 1.66 1.85
Panel D: A e age annual u no e (%)
High (>80%) 48.02 47.91
60%–80% 44.97 63.53 58.49 55.07
40%–60% 44.08 54.39 48.24 87.60 49.96
20%–40% 30.16 37.46 32.21 32.55
Low (<20%) 15.76 13.05 15.19
All 29.32 42.84 41.79 51.01 36.78
Panel E: A e age pe cen age o independen unds
High (>80%) 56.67 54.00
60%–80% 58.50 53.33 65.50 56.14
40%–60% 24.47 29.24 8.15 90.00 26.03
20%–40% 5.67 18.15 47.38 11.71
Low (<20%) 7.40 20.00 12.89
All 7.70 32.72 37.20 63.73 22.70
Panel F: Pe cen age o unds ha cha ge ees o e pe o mance
High (>80%) 0.00 3.33
60%–80% 5.06 3.57 13.56 7.86
40%–60% 21.83 12.53 24.91 36.67 20.35
20%–40% 9.59 17.58 18.17 12.30
Low (<20%) 5.33 20.00 5.15
All 10.27 14.05 14.06 17.05%
Panel G: Ne e u n adjus ed o con ol po olios by TNA, lows, and u no e (50 h pe cen ile) (%)
High (>80%) 2.96 2.36
60%–80% −0.06 0.62 0.88 0.26
40%–60% −0.15 −0.19 −0.70 −0.11 −0.16
20%–40% −0.26 −0.12 0.04 −0.13
Low (<20%) −0.32 −0.12 −0.25
All −0.26 −0.08 −0.21 1.48
TNA: o al ne asse s.
No e: This able epo s he desc ip i e s a is ics o Spanish domes ic equi y mu ual unds, depending on hei le el o ac i e managemen , measu ed
h ough he AS and TE. A each calenda qua e -end (publicly epo ing da es), he unds a e independen ly so ed as pe hei AS and TE. We
used 20%, 40%, 60%, and 80% AS cu s and 4%, 8%, and 12% TE cu s (% pe yea ). Panel A epo s he minimum, a e age, and maximum numbe o
mu ual unds; Panel B shows he median TNA o each calenda qua e -end and pa i ion wi h a leas i e unds; Panel C shows he a e age annual
managemen ees cha ged by he unds; Panel D shows he a e age annual u no e ; Panel E shows he a e age pe cen age o unds managed by an
independen managemen company (as opposed o bank-a ilia ed unds); Panel F shows he pe cen age o unds ha cha ge ees o e pe o mance;
and Panel G shows TNA, lows, and u no e con ol po olio adjus ed e u ns. The sample pe iod is Decembe 2000 o Decembe 2006.
330 Business Resea ch Qua e ly 27(3)
le el o less han 50% and a TE le el o less han 3%, as
hese cu s could be mo e indica i e in membe s a es wi h
ela i ely small equi y ma ke s. This epo inds ha only
7% o unds all in o his ca ego y. Howe e , hese unds
ep esen 32% o he unds in ou sample.
The high ac ion o close indexe s in he Spanish
mu ual und ma ke can be explained by he smalle size o
he Spanish S ock Exchange in compa ison o o he
exchanges, such as hose in he Uni ed S a es. The a e age
low le el o ac i eness in he Spanish mu ual und indus y
can also be explained by he low le els o compe i i eness.
Acco ding o C eme s e al. (2016), ma ke s wi h less
explici indexing unds a e less compe i i e, and ac i e
unds ha e lowe AS le els. The e o e, he low supply o
explici indexing unds in Spain (only 9% o TNA as o
Decembe 2010 as s a ed by C eme s e al., 2016) is
ano he cha ac e is ic o he Spanish indus y.
The bi a ia e dis ibu ion o unds shows he expec ed
posi i e ela ionship be ween he wo ac i e managemen
dimensions. The numbe o unds wi h low AS and high
TE and ice e sa is e y low. Ne e heless, as in C eme s
and Pe ajis o (2009), bo h a iables show subs an ial dis-
pe sion in he o he dimension. The e o e, bo h ac i e
managemen measu es could ha e addi ional explana o y
powe in e ms o und pe o mance. This inding con-
i ms ha i is impo an o dis inguish be ween he wo
dimensions o ac i e managemen .
Panel B o Table 2 epo s he median TNA o unds in
he AS–TE bi a ia e dis ibu ion. The size o mu ual unds
ends o dec ease mono onically when shi ing om he
less ac i e o he mos ac i e unds. Howe e , his en-
dency disappea s o unds wi h ex eme AS alues (mo e
han 80%). This ela i ely la ge size o Spanish mu ual
unds wi h AS alues highe han 80% explains why he
pe cen age o Spanish close indexe s is less ex eme when
asse s unde managemen a e conside ed ins ead o he
numbe o unds. Howe e , mu ual unds wi h low TE al-
ues a e he la ges g oup ep esen ed in he sample, as in
C eme s and Pe ajis o (2009).
Panel C o Table 2 epo s he a e age annual manage-
men and cus odial ees cha ged by Spanish equi y mu ual
unds depending on he bi a ia e dis ibu ion. Con a y o
wha one could expec , Panel C shows ha Spanish domes-
ic equi y unds wi h he lowes AS alues cha ge highe
annual managemen and cus odial ees han he emaining
unds. This puzzling ac is somehow simila o ha ound
in Gil-Bazo and Ruiz-Ve dú (2009), whe e unds wi h poo
be o e- ee pe o mance cha ge highe ees. This inding can
be explained by he p esence o in es o s wi h di e en le -
els o sophis ica ion and deg ees o sensi i i y o pas pe -
o mance. The e o e, we can conclude ha he Spanish und
indus y is cha ac e ized by close indexe s ha cha ge high
managemen ees p o iding only index-like e u ns.
Panel D epo s an a e age po olio u no e o 37%
pe yea , which is lowe han he 95% epo ed by C eme s
and Pe ajis o (2009). The unds wi h low le els o AS and
TE a e hose wi h less u no e (lowe han 20%). Howe e ,
we do no ind he expec ed posi i e ela ion along he
en i e dis ibu ion.
Panel E shows he a e age pe cen age o unds man-
aged by independen managemen companies. This pe -
cen age o unds managed by independen managemen
companies inc eases wi h he AS and TE le els. This ind-
ing is ela ed o he wo k o Golez and Ma in (2015), who
ind ha Spanish bank-a ilia ed mu ual unds sys ema i-
cally inc ease hei holdings in he con olling bank s ock
du ing ce ain e en s, such as p ice d ops, indica ing ha
managemen company owne ship ma e s.
Díaz-Mendoza e al. (2014) ound ha pe o mance-
based ee unds pe o m signi ican ly be e han he
o he isky unds conside ed in he Spanish mu ual und
indus y. Panel F shows he pe cen age o unds ha
cha ge ees based on pe o mance. The esul s do no
suppo a ela ionship be ween und ac i eness and his
a iable.
Finally, Panel G compa es he pe o mance mean o
he mo e ac i e unds e sus he mo e passi e unds con-
olling o TNA, lows, and u no e ia “con ol po o-
lios.” Speci ically, ne e u ns a e adjus ed conside ing
he excess und ne e u n in a gi en mon h in compa ison
wi h he “expec ed” ne e u n acco ding o i s TNA,
lows, and u no e . Each mon h, we cons uc wo ca e-
go ies o each o he h ee abo e-men ioned a iables
using he 50 h pe cen ile o ob ain a su icien numbe o
po olios in each o he eigh ca ego ies. We compu e he
equally weigh ed a e age ne e u ns o each ca ego y.
We hen ma ch each mu ual und mon h wi h he TNA–
low– u no e ca ego y o which he und belongs in ha
mon h and calcula e i s adjus ed ne e u n as he di e -
ence in hei ne e u ns. We ind ha he adjus ed e u ns
o con ol po olios a e posi i ely ela ed o bo h meas-
u es o ac i eness (AS and TE) and ha he unds wi h
he highes le els o AS and TE ha e he bes
pe o mance.
To compu e he money lows o mu ual unds, we ol-
low he absolu e and ela i e money low measu es p o-
posed in p e ious li e a u e (see, o example, Gue cio &
Tkac, 2002; Si i & Tu ano, 1998)
Absolu e Flow TNATNA R
i i i i ,,,
=− ⋅+
()
−11 (5)
Rela i eFlow
TNATNA R
TNA
i
i i i
i
,
,, ,
,
=−⋅+
()
−
−
1
1
1 (6)
whe e TNAi, is he o al ne asse s o und i in mon h , and
Ri, is he e u n o und i in mon h .
Al hough he absolu e lows a e ele an om a man-
agemen pe spec i e because he as majo i y o he unds
And eu e al. 331
examined cha ged hei managemen ees based on TNA,
he p e ious li e a u e has also ecommended he use o
ela i e lows when he size o mu ual unds is qui e dis-
pe sed. To a oid po en ial biases de i ed om dispa i ies
among he a e age size o he unds acco ding o hei
le el o AS (see Panel B o Table 2), we addi ionally p o-
pose an app oach in which money lows a e adjus ed con-
side ing he excess low a ac ed by a und in a gi en
qua e in compa ison wi h he “expec ed” low acco ding
o i s size. Speci ically, each mon h, we cons uc he TNA
deciles and compu e hei equally weigh ed a e age lows.
We hen ma ch each mu ual und mon h wi h he TNA
decile in which he und belongs in ha mon h and calcu-
la e i s adjus ed money low as he di e ence in he abso-
lu e lows a ac ed.
Finally, as a p oxy o de ec AS WD, we compu e AS_di -
e ence, which is he di e ence be ween he AS le el in he
publicly epo ed mon h (end o qua e ) and he a e age AS
le el in he su ounding non-publicly epo ed mon hs
AS di e enceASAS AS AS AS
_=−
+++
−−
++
21
12
4
(7)
Empi ical models and me hods
The in luence o ac i e managemen on u u e
und pe o mance
P e ious s udies show ha he a e age mu ual und sligh ly
ou pe o ms he ma ke e u n be o e expenses and ails o
ou pe o m he ma ke e u n a e expenses (see, o
example, Daniel e al., 1997; Fama & F ench, 2010;
Jensen, 1968). Recen s udies examining mu ual und
ades ha e demons a ed he supe io abili y o ce ain
ac i e mu ual unds. Chen e al. (2000), Alexande e al.
(2007), Bake e al. (2010), and Jiang e al. (2014) ind
suppo o he hypo hesis o he ading skills o mu ual
und manage s.10
Ou pe o mance can only a ise om ac i e manage-
men ; he e o e, he e should be c oss-sec ional di e -
ences in und pe o mance depending on he le el o AS.
Ac i ely managed unds o e a po olio ha a emp s o
ou pe o m a e e ence benchma k by ac i ely de ia ing
om he benchma k weigh s by o e weigh ing (unde -
weigh ing) he s ocks ha he manage conside ed good
(bad) in es men s. Ac i e mu ual unds cha ge highe ees
o ac i e managemen . The e o e, in o ma ion ega ding
he le el o ac i e managemen (AS) is impo an . I is
impossible o bea a benchma k wi hou some le el o AS,
ha is, he lowe he le el o AS, he lowe he possibili y
o ou pe o ming he benchma k. Howe e , a high le el o
AS is necessa y bu is no a su icien condi ion o bea he
benchma k; his depends on he skill o he manage .
Acco dingly, C eme s and Pe ajis o (2009) demons a e
ha he AS le el in he US ma ke is posi i ely ela ed o
und pe o mance (manage skill).
Gi en ha he incen i es o window d ess he AS le el
depend on whe he high le els o AS in publicly epo ed
po olios send a posi i e signal o po en ial in es o s, we
i s analyze whe he he US e idence o he posi i e e ec
o AS on subsequen und pe o mance is obse ed in he
Spanish ma ke (H1). To s udy he ela ionship be ween
AS and u u e und pe o mance, we un he ollowing
pooled panel eg ession
AdjRe u
nAST
E
AdjRe u n
con ol
i i i
i
_
_
,,
,
,
+=+ +
+
+
10
12
3
αα α
α
ssi i ,,
+
ε
(8)
whe e he dependen a iable is he g oss o ne benchma k
adjus ed e u n o mu ual unds in mon h + 1, while he
independen a iables a e he AS and TE le els o mu ual
unds, he Adj_Re u n, and some con ol a iables measu ed
in mon h . Speci ically, we use he annual u no e , annual
managemen ees, log(TNA), log(TNA)2, numbe o s ocks,
und age, numbe o in es o s, and annual ela i e money
lows ( lows/TNA) as con ol a iables. Yea ixed e ec s
a e included in all he speci ica ions.
The in luence o ac i e managemen on u u e
money lows
I AS in luences u u e und pe o mance (H1), we should
analyze whe he his posi i e signal gi en by epo ing
high le els o AS makes in es o s mo e likely o in es in
hese unds and he e o e in luences he le el o u u e
money lows cap u ed by mu ual unds (H2).
To es his hypo hesis, we examine he ela ionship
be ween he AS le el and he u u e money lows ecei ed
by each mu ual und, including some con ol a iables ha
could also de e mine he magni ude o he in es men
lows, as shown in he ollowing eg ession
Flow AS TE
AdjRe u ncon ols
i i i
i i
,,,
,,
_
+=+ +
++
+
1012
3
αα α
αε
ii ,
(9)
whe e he dependen a iable is he und money low in
qua e + 1, de ined as he absolu e, ela i e, and adjus ed
money lows (see equa ions (5) and (6)). Adjus ed money
lows a e low de ia ions om hei espec i e TNA decile
lows. The independen a iables a e he le els o AS and
TE o he mu ual unds a he end o qua e , he Adj_
Re u n, and he con ol a iables explained in equa ion (8).
All he explana o y a iables a e measu ed a he end o
qua e (publicly disclosed mon hs). Yea ixed e ec s
a e included in all he speci ica ions. This pooled eg es-
sion has been conduc ed on a qua e ly basis because he
po olio holdings in he mon hs wi h no public disclosu e
a e no known o in es o s. The e o e, hey canno a ec
u u e money lows.
338 Business Resea ch Qua e ly 27(3)
∆
AS DPublic con ols
i i i i
,,,,
_=+
++
αα ε
01 (14)
whe e he dependen a iable is he change in AS in mon h
, α0 is a cons an , and D_Public is a dummy a iable ha
equals 1 o publicly disclosed mon hs (calenda qua e -
ends), and 0 o he wise. We un his eg ession o he wo
AS ca ego ies and o each o he ou AS–TE und ca ego-
ies. Once a und is anked in a gi en ca ego y in each
calenda qua e -end, i emains in he same ca ego y du ing
he ollowing 2 mon hs. The con ol a iables a e TE, Adj_
Re u n, and hose de ined in equa ion (8).
The e o e, he cons an coe icien measu es he a e -
age ∆AS in non-publicly epo ed mon hs, and he coe i-
cien o he dummy measu es how he a e age ∆AS in
publicly epo ed mon hs de ia es om he o me . Table
10 shows he esul s.
Table 9. AS le el a qua e -end mon hs and he su ounding mon hs.
Panel A TE
AS Low (80%) High (20%) All
Low (80%) Min/a e /max no. 66/69/73 3/4/8 69/73/79
High (20%) Min/a e /max no. 3/4/8 12/14/16 17/18/20
All Min/a e /max no. 69/73/79 17/18/20
Panel B TE
AS Low (80%) High (20%) All
Low (80%) AS_di e ence (%) −0.42***
(.00)
−1.12***
(.01)
−0.46***
(.00)
Mon h – 1 AS (%) 31.54***
(.01)
35.30***
(.01)
31.77***
(.00)
Ranking AS (%) 31.18 33.41 31.30
Mon h + 1 AS (%) 31.60**
(.02)
33.38
(.71)
31.71**
(.02)
Mon h + 2 AS (%) 31.63*
(.07)
33.99
(.67)
31.79*
(.06)
High (20%) AS_di e ence (%) 2.04***
(.00)
−0.08
(.60)
0.37**
(.04)
Mon h – 1 AS (%) 58.45***
(.00)
73.48
(.13)
70.21
(.11)
Ranking AS (%) 60.76 73.22 70.46
Mon h + 1 AS (%) 59.88
(.29)
73.55
(.59)
70.60
(.90)
Mon h + 2 AS (%) 58.30***
(.01)
73.07
(.32)
69.94*
(.06)
All AS_di e ence (%) −0.29**
(.02)
−0.35**
(.04)
Mon h – 1 AS (%) 33.06*
(.08)
65.03***
(.00)
Ranking AS (%) 32.84 64.33
Mon h + 1 AS (%) 33.11
(.15)
64.93
(.36)
Mon h + 2 AS (%) 33.05
(.44)
64.74
(.81)
AS: Ac i e Sha e; TE: acking e o .
No e: This able con ains wo panels. Panel A shows he minimum, a e age, and maximum numbe o unds in each ca ego y, depending on
he le el o ac i e managemen measu ed h ough he AS and TE. Panel B shows he AS le el a calenda qua e -end mon hs when po olio
holdings a e publicly epo ed ( anking AS) and he AS le el in he p e ious mon h (mon h – 1) and he nex non-publicly epo ed mon hs
(mon h + 1 and mon h + 2) o he sample pe iod. In addi ion, he able shows he AS_di e ence, de ined as he di e ence be ween he AS
le el o he public mon h and he a e age AS in he 4 non-public mon hs su ounding ha mon h. E e y 3 mon hs, he unds we e independen ly
anked acco ding o hei AS and TE le els. Then, we cons uc wo po olios o each a iable using he 80 h pe cen ile and ou po olios
om hei in e sec ions. The p alues (in pa en heses) epo he signi icance o he pai ed es o means be ween he AS in he anking mon h
and ha in he su ounding mon hs.
*, **, and *** indica e signi icance a .1, .05, and .01 le els, espec i ely.
And eu e al. 339
Table 10 shows ha , on a e age, Spanish equi y mu ual
unds end o educe he AS le el in publicly epo ed
mon hs in ela ion o he a e age AS change in non-pub-
licly epo ed mon hs (a s a is ically signi ican −0.41%).
This inding is consis en wi h he e idence p o ided in
Table 8 and wi h he esul s p o ided in Table 9. This esul
can be pa ially explained by he ac ha mos o he unds
a e classi ied as low AS and low TE. Fu he mo e, hese
unds a emp o appea less ac i e in qua e -ends, p oba-
bly because hey a e o ien ed o conse a i e in es o s
looking o low ac i e managemen .
When he di e en und ca ego ies acco ding o hei
ac i e managemen a e analyzed, we obse e ha his en-
dency o educe AS in publicly epo ed mon hs is exclu-
si e o low AS unds. The changes in he AS le els in he
high AS and low TE unds a e he opposi e. These unds
end o inc ease he AS le el in publicly epo ed mon hs
ela i e o he a e age AS change in non-publicly epo ed
mon hs (a s a is ically signi ican 3.28%). Finally, as
expec ed, unds wi h bo h high AS and TE do no exhibi
di e en beha io in hei mon hly AS changes because
his ca ego y ga he s hese mu ual unds wi h eal ac i e
managemen cap u ed by bo h me ics, ha is, AS and TE.
The esul s a e consis en when he con ol a iables and
ime ixed e ec s a e included.
In summa y, his s udy p o ides e idence o a ce ain ype
o WD o appea mo e ac i e conduc ed by mu ual unds
wi h a high le el o AS and a low le el o TE when po olios
a e publicly a ailable o inancial ma ke pa icipan s.
Robus ness analyses
High AS–low TE unds and u u e pe o mance
In he p e ious sec ion, we ound ha unds wi h high AS
bu low TE le els exhibi highe le els o AS_di e ence.
In his sec ion, we a emp o p o ide addi ional e idence
by analyzing he u u e pe o mance o he ou und ca -
ego ies. I his measu e ac ually measu es AS WD, hen
we should expec he high AS and low TE und ca ego y o
exhibi wo se pe o mance han he unds wi h a simila
le el o AS, bu wi h no dis o ion, ha is, high AS and
high TE unds.
Table 11 shows he annualized alues o he benchma k
adjus ed e u ns (Panel A) and he ou - ac o alphas (Panel
B) in qua e + 1 o each o he ou ca ego ies o unds
and o bo h g oss and ne e u ns.19 Funds wi h high AS
and low TE le els no only exhibi wo se pe o mance
han unds wi h high AS and TE, hey a e he und
Table 10. AS le el in publicly disclosed mon hs e sus non-publicly disclosed mon hs: pool eg ession.
Fund ca ego ies Low AS
Low TE
Low AS
High TE
High AS
Low TE
High AS
High TE
Low AS High AS All
Panel A: Change in AS in publicly disclosed mon hs
Cons an 0.19***
(.00)
0.63
(.21)
−1.10***
(.00)
−0.20
(.12)
0.22***
(.00)
−0.39***
(.00)
0.10**
(.04)
Dpublic −0.57***
(.00)
−2.50*
(.10)
3.28***
(.00)
−0.15
(.39)
−0.67***
(.00)
0.61**
(.02)
−0.41***
(.00)
n4,908 274 275 992 5,182 1,267 6,586
R2 (%) 0.60 4.49 6.44 0.04 0.77 0.45 0.27
Adjus ed R2 (%) 0.58 4.14 6.10 −0.06 0.75 0.37 0.25
Panel B: Change in AS in publicly disclosed mon hs wi h con ol a iables and ime ixed e ec s
Dpublic −0.53***
(.00)
−1.32
(.12)
3.05***
(.00)
−0.15
(.35)
−0.57***
(.00)
0.58**
(.02)
−0.33***
(.01)
Con ol a iables Yes Yes Yes Yes Yes Yes Yes
Time ixed e ec s Yes Yes Yes Yes Yes Yes Yes
n4,695 238 255 965 4,933 1,220 6,184
R2 (%) 1.97 9.60 11.59 1.71 1.69 1.78 1.27
Adjus ed R2 (%) 1.62 2.61 5.25 −0.06 1.35 0.39 0.99
AS: Ac i e Sha e; TE: acking e o .
No e: This able shows he impac o he public epo ing da es on he change in he AS le el o Spanish domes ic equi y mu ual unds depending
on he le el o ac i eness h ough a pooled eg ession. A calenda qua e -ends, unds a e anked acco ding o hei AS and TE le els in o ou
ca ego ies using he 80 h pe cen ile. Once a und is anked in a gi en ca ego y in each calenda qua e -end, i emains in he same ca ego y du ing
he nex 2 mon hs. Speci ically, Panel A epo s he esul s o es ima ing a pooled eg ession whe e he dependen a iable is he change in he
le el o AS o each mu ual und in mon h , DAS, while he independen a iable is a cons an and a dummy a iable, DPublic, ha equals 1 o publicly
disclosed mon hs (Ma ch, June, Sep embe , and Decembe ). Panel B epo s he es ima ion esul s, including he ollowing con ol a iables: he TE,
he und adjus ed e u n, and he con ol a iables de ined in Table 3 o mon h . The con ol a iables wi h annual equency emained cons an in
all mon hs o he calenda yea . The p alues (in pa en heses) a e based on s anda d e o s clus e ed by he und.
*, **, and *** indica e signi icance a .1, .05, and .01 le els, espec i ely.
340 Business Resea ch Qua e ly 27(3)
ca ego y wi h he wo s pe o mance. The unds included
in his ca ego y ail o each he expec ed le els o good
pe o mance ela ed o hei high le el o AS.
Table 11 also shows o he in e es ing ela ions ha a e
di e en om hose obse ed in he US ma ke . Fi s , we
ound ha nei he AS no TE can p o ide a posi i e and
Table 11. Annual pe o mance acco ding o he le el o ac i e managemen .
Panel A TE
AS Low (80%) High (20%) All High–low
Low (80%) G oss adjus ed e u n (%) −0.76
(.15)
−0.29
(.92)
−0.76
(.21)
0.47
(.85)
Ne adjus ed e u n (%) −2.66***
(.00)
−2.20
(.45)
−2.67***
(.00)
0.46
(.86)
High (20%) G oss adjus ed e u n (%) −3.28**
(.05)
5.81
(.11)
3.43
(.26)
9.08***
(.01)
Ne adjus ed e u n (%) −5.10***
(.00)
4.16
(.26)
1.75
(.56)
9.26***
(.01)
High–low TE High–low AS
All G oss adjus ed e u n (%) −0.93*
(.09)
4.24
(.22)
5.17*
(.10)
4.19*
(.10)
Ne adjus ed e u n (%) −2.83***
(.00)
2.53
(.46)
5.37*
(.08)
4.41*
(.08)
Mix H-L
High-low G oss adjus ed e u n (%) −2.52*
(.09)
6.10***
(.00)
6.56**
(.05)
Ne adjus ed e u n (%) −2.44*
(.10)
6.36***
(.00)
6.82**
(.04)
Panel B TE
AS Low (80%) High (20%) All High–low
Low (80%) G oss alpha 4 (%) −1.09***
(.00)
−2.67*
(.10)
−1.23***
(.00)
−1.57
(.30)
Ne alpha 4 (%) −2.98***
(.00)
−4.57***
(.01)
−3.12***
(.00)
−1.59
(.30)
High (20%) G oss alpha 4 (%) −4.56**
(.02)
1.88
(.34)
0.17
(.93)
6.44***
(.00)
Ne alpha 4 (%) −6.38***
(.00)
0.25
(.90)
−1.49
(.43)
6.63***
(.00)
High–low TE High–low AS
All G oss alpha 4 (%) −1.30***
(.00)
0.66
(.71)
1.96
(.24)
1.40
(.42)
Ne alpha 4 (%) −3.19***
(.00)
−1.02
(.57)
2.16
(.20)
1.63
(.35)
Mix H-L
High-low G oss alpha 4 (%) −3.47**
(.04)
4.54***
(.00)
2.97
(.11)
Ne alpha 4 (%) −3.40**
(.05)
4.82***
(.00)
3.23*
(.08)
TE: acking e o ; AS: Ac i e Sha e; Mix H-L: Mix High–Low.
No e: Panels A and B epo he annual pe o mance (benchma k adjus ed e u ns and ou - ac o alphas). A calenda qua e -ends (publicly epo ing
da e), he unds a e independen ly anked acco ding o hei AS and TE. Then, we cons uc wo ca ego ies o each a iable using he 80 h pe cen ile and
ou ca ego ies om hei in e sec ion. The unds o each ca ego y we e held 3 mon hs a e he o ma ion da e. The “Mix High–Low” s a egy shows he
di e ence be ween “High AS and TE” unds and “Low AS and TE” unds. The p alues (in pa en heses) we e based on Whi e’s s anda d e o s.
*, **, and *** indica e signi icance a .1, .05, and .01 le els, espec i ely.
And eu e al. 341
s a is ically signi ican pe o mance. This inding di e s
om ha o C eme s and Pe ajis o (2009), who show ha
AS allows in es o s o selec unds wi h good pe o -
mance. Howe e , ou esul s a e consis en wi h C eme s
e al. (2016), who s a e ha he a e age alphas gene a ed
by ac i e managemen a e lowe in ma ke s whe e close
indexing is gene alized. Second, we also ind ha he AS
le el p edic s und pe o mance among high TE unds
( he di e ence in he ou - ac o alpha be ween high AS
and low AS is abo e 4%, a igu e posi i e and s a is ically
signi ican o bo h g oss and ne ). Simila ly, he TE le el
p edic s he und pe o mance among high AS unds ( he
di e ence in he ou - ac o alpha is also abo e 4%). This
las inding di e s om ha o C eme s and Pe ajis o
(2009), who do no obse e ha TE p edic s und pe o -
mance wi hin any AS ca ego y. The di e ences in he
indings may be because we independen ly ank unds
acco ding o bo h me ics o ob ain app op ia e o hogo-
nal po olios, while hese au ho s so unds in o AS i s
and hen in o TE quin iles. So ing unds i s in o AS and
hen in o TE es s whe he TE has p edic i e powe a e
con olling o AS, bu no he con a y. To analyze he
p edic i e powe o AS con olling o TE, unds should
i s be so ed in o TE. Thi d, we can pe o m a be e
selec ion i we ocus on he unds ha simul aneously
ha e high le els o bo h ac i e measu es. Al hough unds
wi h bo h high AS and TE le els do no yield signi ican
posi i e pe o mance, i mus be highligh ed ha , in con-
as o he o he unds, on a e age, hey a e he only ca -
ego y able o gene a e enough alue o, a leas ,
compensa e o hei ees.20
High AS–low TE unds and u u e money lows
In his sec ion, we a emp o p o ide ex a suppo o H4 by
analyzing he ela ionship be ween he di e en ca ego ies
o ac i e managemen based on AS and TE and hei money
lows. Table 12 shows he mon hly a e age adjus ed low in
he 3 mon hs a e he o ma ion da e (holding pe iod) o
each o he ou po olio ca ego ies. We ind ha despi e he
wo s pe o mance o window d esse s (high AS–low TE
unds), hese unds do no exhibi signi ican ly lowe le els
o money lows han he ca ego y o unds wi h bo h high
AS and TE. Fu he mo e, hey a ac signi ican ly highe
money lows han unds wi h bo h low AS and TE.
The e o e, window-d esse unds each he objec i e o
a ac ing money lows. This e idence suppo s H4.
AS WD and u u e pe o mance: Mon hly basis
eg ession and independen unds subsample
In addi ion o he qua e ly based eg ession o equa ion
(11), we also analyzed he impac o AS WD in publicly
disclosed mon hs on u u e pe o mance in a mon hly basis
eg ession, as shown in he ollowing equa ion
AdjRe u ns DPublic
AS DPublic AS
i i
i i
__
_
,,
,,
+
=+
++ ×
101
23
αα
αα
ii
i i
con ols
,
,,
++
ε
(15)
whe e he dependen a iable is he g oss and ne adjus ed
e u ns o mu ual unds in he nex mon h + 1, while he
Table 12. Adjus ed in es men lows acco ding o he le el o ac i e managemen .
AS TE
Low (80%) High (20%) All High–low
Low (80%) −300.82***
(.00)
509.38***
(.00)
−262.58***
(.00)
810.20***
(.00)
High (20%) 697.60*
(.06)
1,094.71***
(.00)
996.74***
(.00)
397.11
(.36)
High–low TE High–low AS
All −259.57***
(.00)
980.46***
(.00)
1,240.03***
(.00)
1,259.32***
(.00)
Mix H-L
High–low 998.42***
(.01)
585.33***
(.01)
1,395.53***
(.00)
TE: acking e o ; AS: Ac i e Sha e; Mix H-L: Mix High–Low; TNA: o al ne asse s.
No e: The able shows he mon hly a e age money lows adjus ed by TNA decile po olios in he 3 mon hs a e he o ma ion da e (holding
pe iod) ob ained by Spanish domes ic equi y mu ual unds, depending on he le el o ac i eness measu ed h ough he AS and TE measu es. A
calenda qua e -ends (publicly epo ing da e), he unds a e independen ly anked acco ding o hei AS and TE le els. Then, we cons uc
wo ca ego ies o each a iable using he 80 h pe cen ile and ou ca ego ies om hei in e sec ion. The “Mix High–Low” s a egy shows he
di e ence be ween “High AS and TE” unds and “Low AS and TE” unds. The p alues (in pa en heses) we e based on Whi e’s s anda d e o s.
*, and *** indica e signi icance a .1, and .01 le els, espec i ely.
342 Business Resea ch Qua e ly 27(3)
independen a iables a e as ollows: DPublic_ is a
dummy a iable ha equals 1 o publicly disclosed
mon hs, and 0 o he wise; AS o mu ual unds in mon h ;
and he in e ac ion be ween DPublici
_, and ASi ,.
The con ol a iables a e TE, Adj_Re u ns, and hose
de ined in equa ion (8).
Hence, he in e ac ion a iable, DPublic AS
i i
_
,,
×,
measu es de ia ions o AS in publicly disclosed mon hs
om non-publicly disclosed mon hs and, he e o e, could
be conside ed a p oxy o changing he holdings o appea
mo e ac i e. I he p ac ice o a i icially inc easing he
le el o AS in publicly disclosed mon hs de e io a es und
pe o mance, hen we should expec nega i e coe icien s
o his a iable.
Mo eo e , we sugges ha he nega i e pe o mance
impac o changing po olio holdings in publicly disclosed
mon hs depends on how his p ac ice is conduc ed.
Acco dingly, Golez and Ma in (2015) ind ha Spanish
bank-a ilia ed mu ual unds (non-independen unds) make
holding changes based on he bank’s in e es . Based on his
p e ious e idence, we hypo hesize ha independen unds
a e, on a e age, less p one o pe o ming WD, and i hey
inc ease he AS le el in publicly epo ed mon hs (high le -
els o AS_di e ence), i will mo e likely be ela ed o
in o med ansac ions a he han AS WD. The e o e, hey
will no damage pe o mance. In con as , non-independen
unds a e, on a e age, mo e p one o pe o ming WD (sub-
jec o he in e es o he bank hey a e a ilia ed wi h). I
hey inc ease he AS le el in publicly epo ed mon hs, i
will p obably be ela ed o WD a he han o in o med ad-
ing. The e o e, hey will damage pe o mance.
Table 13 shows he esul s o unning equa ion (15) o
he ull sample and o independen and non-independen
Table 13. AS WD and u u e und e u ns: independen e sus bank-a ilia ed unds.
G oss adjus ed e u ns +1 Ne adjus ed e u ns +1
Full sample Independen Bank-a ilia ed Full sample Independen Bank-a ilia ed
Public 0.3414***
(.01)
−0.0925
(.80)
0.3785***
(.01)
0.3416***
(.01)
−0.1025
(.78)
0.3850***
(.01)
AS 1.1595***
(.00)
2.5667***
(.00)
0.2679
(.41)
1.2047***
(.00)
2.4965***
(.00)
0.3721
(.25)
Dpublic AS
i
_,×−1.0311***
(.01)
−0.3323
(.65)
−1.0925**
(.02)
−1.0376***
(.01)
−0.3105
(.67)
−1.1204**
(.02)
n6,311 1,361 4,950 6,311 1,361 4,950
R2 (%) 4.98 8.26 5.37 5.26 8.54 5.74
Adjus ed R2 (%) 4.69 6.93 4.99 4.98 7.21 5.37
AS: Ac i e Sha e; WD: window d essing; TE: acking e o .
No e: This able epo s he esul s o he analysis ha examines he pe o mance impac o inc easing he AS le el in publicly disclosed mon hs.
The dependen a iables a e he g oss and ne benchma k adjus ed e u ns o mu ual unds in mon h + 1, while he independen a iables a e
as ollows: he AS le els o mu ual unds in mon h ; D_Public, which is a dummy a iable ha equals 1 o publicly disclosed mon hs (Ma ch, June,
Sep embe , and Decembe ), and 0 o he wise; he in e ac ion be ween AS and D_public; TE; he und adjus ed e u ns; and he con ol a iables
explained in Table 3 measu ed in mon h . In addi ion, he able shows he esul s o he ull sample spli in o independen unds and non-
independen unds (bank-a ilia ed mu ual unds). Yea ixed e ec s a e included in all he speci ica ions. The p alues (in pa en heses) a e based on
s anda d e o s clus e ed by he und.
**, and *** indica e signi icance a .05, and .01 le els, espec i ely.
(bank-a ilia ed) subsamples. The esul s o he ull sub-
sample show ha inc easing he le el o ac i i y a qua e -
ends nega i ely in luences und pe o mance (coe icien
o D_publicxAS), which con i ms H3. Mo eo e , his neg-
a i e impac is no obse ed in he independen unds sub-
sample and only he bank-a ilia ed unds subsample
shows a s a is ically signi ican coe icien on D_pub-
licxAS. This esul suppo s ou in ui ion ha he inc ease
o AS in publicly disclosed mon hs is p obably ela ed o
in o med ades o he independen unds subsample and
p obably ela ed o WD o he non-independen unds
subsample.
Conclusion
In his s udy, we use a unique da abase ha includes non-
publicly disclosed po olios o es whe he mu ual unds
al e hei holdings in publicly disclosed mon hs o appea
mo e ac i e by epo ing a high le el o AS o a ac u u e
money lows. No s udy in he li e a u e has examined
whe he he le el o de ia ion om he benchma k can be
al e ed by po olio manage s (AS WD). We compu e how
und holdings de ia e om he benchma ks using he AS
me ic o C eme s and Pe ajis o (2009).
We examine he ela ionship be ween AS and u u e
und pe o mance, and he ela ionship be ween AS and
money lows, and demons a e ha he highe he AS le el,
he highe he u u e und e u ns and lows. Hence, we
hypo hesize ha ce ain po olio manage s ha e incen-
i es o show a high le el o AS when he po olios a e
publicly a ailable. The e o e, we s udy he exis ence o AS
WD and how his p ac ice in luences u u e und pe o -
mance and money lows. We de ine he AS_di e ence as
And eu e al. 343
he di e ence in he le el o AS in publicly epo ed
mon hs as opposed o he le el in he su ounding mon hs
and use i as a p oxy o AS WD. We ind ha high AS_di -
e ence alues a e ela ed o poo u u e und pe o mance
bu high u u e money lows.
Finally, we analyze unds ha a e mo e p one o con-
duc ing AS WD. We ind ha , con a y o he o he ca ego-
ies o unds, on a e age, unds wi h high AS and low TE
le els inc ease hei AS le els be o e hei holdings a e
public and educe hem a e wa d, hence showing a sig-
ni ican and posi i e AS_di e ence. Mo eo e , his ca e-
go y o unds exhibi s he wo s u u e pe o mance bu
a ac s highe money lows han unds wi h simila le els
o TE. This inding also indica es ha his s a egy o de i-
a ing mo e om benchma ks when holdings a e public
makes economic sense because he highe he money lows
a e, he highe he incomes o he managemen companies,
gi en ha mos Spanish mu ual unds cha ge hei ees
based on TNA. Howe e , we should no o ge ha hese
in es men s a egies conduc ed by ce ain unds e ode
in es o und pe o mance, pa icula ly in he case o
bank-a ilia ed unds.
We belie e ha he esul s p esen ed he e a e compel-
ling enough o wa an u he analysis. The abo e-men-
ioned esul s p o ide e idence o he implica ions o his
esea ch o policymake s who should conside he neces-
si y o mon hly po olio holdings ins ead o qua e ly
holdings o limi he disc e ion o po olio manage s.
Al hough we a e awa e o he di icul ies in ob aining
bo h publicly and non-publicly a ailable po olio hold-
ings o all unds in a gi en in es men ca ego y, u u e
esea ch in his a ea could ex end his s udy o o he coun-
ies in he eu ozone and he US ma ke . The exis ence o
publicly and non-publicly a ailable holdings o all
mu ual unds in he di e en eu ozone coun ies will
allow esea che s o examine whe he he AS WD phe-
nomenon is jus a cha ac e is ic o he Spanish ma ke o
i i is a widesp ead phenomenon. Ano he line o esea ch
could in ol e a mo e de ailed analysis o he in es o s’
eac ions, o examine whe he hey a e awa e o his man-
age ial p ac ice.
Decla a ion o con lic ing in e es s
The au ho (s) decla ed no po en ial con lic s o in e es wi h
espec o he esea ch, au ho ship, and/o publica ion o his
a icle.
Funding
The au ho (s) disclosed eceip o he ollowing inancial suppo
o he esea ch, au ho ship, and/o publica ion o his a icle:
This wo k was suppo ed by he Spanish go e nmen and he
Eu opean Union FEDER unds RTI2018-093483-B-I00.
ORCID iD
Lau a And eu h ps://o cid.o g/0000-0003-3661-5448
No es
1. Mo nings a , o ins ance, p o ides mon hly po olio hold-
ings o ce ain unds in o he eu ozone coun ies. Howe e ,
he e is a selec ion bias because his in o ma ion is no
always a ailable o all mu ual unds. In addi ion, his in o -
ma ion is publicly a ailable o o he ma ke compe i o s,
in es o s, and so on. This ac p e en s ex ending he analy-
sis o o he eu ozone coun ies. Simila ly, he o icial supe -
iso in Spain does no p o ide mon hly po olio holdings
om 2007 onwa d. The e o e, his limi s ou ime pe iod o
s udy om 2000 o 2006.
2. I he manage pe o ms well du ing he delay pe iod,
in es o s a e less likely o a ibu e po olio mo emen s
o window d essing (WD) and mo e likely o imp o e
secu i y selec ion. Consequen ly, subsequen o he delay
pe iod, manage s may bene i om inc emen ally highe
lows han ha jus i ied by he und’s pe o mance.
Con as ingly, i he pe o mance du ing he delay pe iod
is bad, in es o s a e mo e likely o a ibu e he po o-
lio mo emen s o WD and he e o e cause he manage o
incu he cos o lowe lows.
3. Some s udies on he Spanish mu ual und indus y exam-
ine WD (in equi y and bond unds) and he Janua y e ec
and con i m ha in es o s eac o po olio holdings in o -
ma ion in his ma ke (see, o example, O iz e al., 2010,
2012, 2015, among o he s). The e a e also s udies ha
examine o he esea ch ques ions in he Spanish indus y
ha show ha in es o s use in o ma ion based on disclosed
po olio holdings o assess manage ial abili y in conjunc-
ion wi h pas pe o mance (see, o example, Ál a ez
e al., 2014; And eu e al., 2017).
4. The sum is aken o e he uni e se o equi y posi ions
only, as we apply he measu e exclusi ely o all-equi y
po olios.
5. The co ela ion coe icien be ween AS (Ac i e Sha e) and
TEbe a is .3775. We p esen he esul s o acking e o (TE;
wi h highe co ela ion wi h AS) in he ex as we aim o ana-
lyze whe he AS has explana o y powe beyond TE measu es.
The esul s using TEbe a a e a ailable upon eques .
6. The mon hly po olio holdings p o ided by Spanish
Secu i ies and Exchange Commission (CNMV) con ain
o icially disclosed and non-o icially disclosed po olios
because managemen companies in Spain mus epo o
in es o s on a qua e ly basis, which is mo e equen han
he Eu opean Union’s equi emen o semi-annual po o-
lio epo s. Because he iscal yea o Spanish managemen
companies is he calenda yea , epo s a e manda o y a
he end o each calenda qua e . Consequen ly, om 1999
o 2006, he da abase was compiled om o icial CNMV
ilings, as well as non-o icial disclosu es p o ided om
CNMV o he au ho s exclusi ely o esea ch pu poses.
The e o e, und manage s could no an icipa e he elease o
hese undisclosed in a-qua e po olios.
7. The qua e -end disclosed holdings o icially p o ided by
CNMV om Ma ch 2007 onwa d could be complemen ed
wi h mon hly po olio holdings p o ided by Mo nings a .
Howe e , his uno icial in o ma ion is no a ailable o
all mon hs and o all mu ual unds, implying a selec ion
bias. Mo eo e , he mon hly po olio holdings p o ided
by Mo nings a a e publicly a ailable, and he goal o ou
344 Business Resea ch Qua e ly 27(3)
es is o check he di e ences be ween publicly and non-
publicly a ailable holdings.
8. The esul s ob ained when using he Spanish benchma k
ha p oduces he lowes AS a e qui e obus along he di -
e en analyses. Simila ly, when we limi he analysis o he
93 unds ha sel - epo Ibex35 o IGBM (Índice Gene al
de Bolsa de Mad id), he esul s a e obus . The esul s o
bo h o he obus ness analyses a e no epo ed in he s udy
due o b e i y easons. These esul s a e, howe e , a ailable
om he au ho s upon eques .
9. Al hough we only p esen esul s o he pe iod om
Decembe 2000 o Decembe 2006 because o he a ailabil-
i y o mon hly po olio holdings o ocus ou s udy on he
analysis o WD, we ha e qua e ly AS and TE da a un il
June 2014 o all he mu ual unds o he sample.
10. This ac has a ac ed he a en ion o esea che s who com-
pa e passi e unds and ac i ely managed unds (see, o exam-
ple., De zle & Wiggins, 1997; Doshi e al., 2015; Fa ooqi
e al., 2019; Pás o e al., 2015; Pás o & S ambaugh, 2002).
11. The e u n gap is compu ed as he di e ence be ween he
ac ual e u ns and he hypo he ical e u ns based on he p e-
ious holdings (see, o example, Kacpe czyk e al., 2008).
12. We use he e u ns o he Ibex35 To al Re u n Index as
he equi y benchma k and he e u ns o one-day Spanish
T easu y bill epos as he isk- ee a e. The ac o s o size,
book o ma ke , and momen um ha e been calcula ed ol-
lowing he same p ocedu e de ailed on he websi e o
Kenne h F ench conside ing he s ocks aded in he Spanish
s ock ma ke (see, o example, h p://mba. uck.da mou h.
edu/pages/ acul y/ken. ench/da a_lib a y.h ml).
13. The esul s a e obus when TE is no included in he equa-
ion. The esul s a e a ailable om he au ho s upon eques .
14. I is di icul o pu he size o ou esul s o AS WD in pe -
spec i e o o he e idence o WD because we a e analyz-
ing a di e en kind o WD. Howe e , i we compa e ou
a iable AS_di e ence dummy alue highe >3% wi h he
BHRG10% dummy (de ined as 1 i ‘backwa d holding e u n
gap’ is in he op 10 h pe cen ile) o Aga wal e al. (2014), ou
–2.06% AS_di e ence dummy coe icien o ne Ca ha
alphas in ou Table 6 is highe han he –0.58% BHRG10%
dummy coe icien o Table 12(3) in Aga wal e al. (2014).
This compa ison should be conside ed wi h cau ion because
ou con ol a iables, al hough simila , a e no exac ly he
same as hose o Aga wal e al. (2014).
15. The disposi ion e ec , ha is, he endency o hold on o
losses and o sell winne s oo soon, has been documen ed in
he US ma ke (see, o example, Cici, 2012; Odean, 1998)
and in he Spanish mu ual und indus y (see, o example,
And eu e al., 2020).
16. See Banda chuk and Hilsche (2013) o an example o how
condi ional anks can d i e alse conclusions.
17. The main conclusions, al hough wi h less signi icance, hold
wi h he 50 h pe cen ile. These esul s a e a ailable om he
au ho s upon eques .
18. As a obus ness analysis, we ied o e ine he po olio ha
consis s o mu ual unds cha ac e ized by a high alue o AS
and a low alue o TE. To do ha , we use o he pe cen iles
di e en om he 80 h pe cen ile o classi y a mu ual und
as one wi h “low TE.” The p oblem is ha e en wi h he 60 h
pe cen ile, he high AS–low TE po olio was emp y in many
ime pe iods.
19. We ha e also checked he obus ness o Table 11 using he
o al ne asse s (TNA), lows, and u no e con ol po olio
adjus ed e u ns used in Panel G o Table 2. The esul s sup-
po ou e idence ha unds wi h high AS and low TE le els
ha e he wo s pe o mance. The au ho s can p o ide hese
esul s upon eques .
20. Non- epo ed es s also show ha he pe o mance di e -
ence be ween high and low ac i e unds is signi ican ly
highe (a he 1% le el) when he wo ac i e measu es (AS
and TE) a e used (Mix High–Low) han when only one
ac i e measu e is used (high–low AS o high–low TE).
Re e ences
Aga wal, V., Gay, G. D., & Ling, L. (2014). Window d essing in
mu ual unds. The Re iew o Financial S udies, 27, 3133–
3170. h ps://doi.o g/10.1093/ s/hhu045
Alexande , G. J., Cici, G., & Gibson, S. (2007). Does mo i a ion
ma e when assessing ade pe o mance? An analysis o
mu ual unds. The Re iew o Financial S udies, 20, 125–
150. h ps://doi.o g/10.1093/ s/hhl008
Ál a ez, J., And eu, L., O iz, C., & Sa o, J. L. (2014). A
nonpa ame ic app oach o ma ke iming: E idence
om Spanish mu ual unds. Jou nal o Economics and
Finance, 38, 119–132. h ps://doi.o g/10.1007/s12197-012-
9228-9
And eu, L., Ma eos, L., & Sa o, J. L. (2017). The alue added
by ading based on alua ion c i e ia. In e na ional Re iew
o Finance, 17, 327–352. h ps://doi.o g/10.1111/i i.12097
And eu, L., O iz, C., & Sa o, J. L. (2020). Disposi ion e ec
in und manage s. Fund and s ock-speci ic ac o s and
he upsho o in es o s. Jou nal o Economic Beha io
& O ganiza ion, 176, 253–268. h ps://doi.o g/10.1016/j.
jebo.2020.04.002
Bake , M., Li o , L., Wach e , J. A., & Wu gle , J. (2010).
Can mu ual und manage s pick s ocks? E idence om
hei ades p io o ea nings announcemen s. Jou nal
o Financial and Quan i a i e Analysis, 45, 1111–1131.
h ps://doi.o g/10.1017/S0022109010000426
Banda chuk, P., & Hilsche , J. (2013). Sou ces o momen um
p o i s: E idence on he i ele ance o cha ac e is ics.
Re iew o Finance, 17, 809–845. h ps://doi.o g/10.1093/
o / 036
Caquineau, M., Mö öla, M., & Schumache , J. (2016). Ac i e
Sha e in Eu opean equi y unds: The ac i eness o la ge-
cap Eu opean und manage s h ough he lens o Ac i e
Sha e. Mo nings a .
Ca ha , M. M. (1997). On pe sis ence in mu ual und pe o -
mance. The Jou nal o Finance, 52, 57–82. h ps://doi.
o g/10.1111/j.1540-6261.1997. b03808.x
Chen, H. L., Jegadeesh, N., & We me s, R. (2000). The alue
o ac i e mu ual und managemen : An examina ion o
he s ockholdings and ades o und manage s. Jou nal o
Financial and Quan i a i e Analysis, 35, 343–368. h ps://
doi.o g/10.2307/2676208
Cici, G. (2012). The p e alence o he disposi ion e ec in
mu ual unds’ ades. Jou nal o Financial and Quan i a i e
Analysis, 47, 795–820. h ps://doi.o g/10.1017/S002210901
2000348
C eme s, K. J. M., Fe ei a, M. A., Ma os, P. P., & S a ks, L. T.
(2016). Indexing and ac i e und managemen : In e na ional
And eu e al. 345
e idence. Jou nal o Financial Economics, 120, 539–560.
h ps://doi.o g/10.1016/j.j ineco.2016.02.008
C eme s, K. J. M., & Pe ajis o, A. (2009). How ac i e is you
und manage ? A new measu e ha p edic s pe o mance.
The Re iew o Financial S udies, 22, 3329–3365. h ps://
doi.o g/10.1093/ s/hhp057
Daniel, K., G inbla , M., Ti man, S., & We me s, R. (1997).
Measu ing mu ual und pe o mance wi h cha ac e is ic-
based benchma ks. The Jou nal o Finance, 52, 1035–1058.
h ps://doi.o g/10.1111/j.1540-6261.1997. b02724.x
De zle , M. L., & Wiggins, J. B. (1997). The pe o mance o
ac i ely managed in e na ional mu ual unds. Re iew o
Quan i a i e Finance and Accoun ing, 8, 291–313. h ps://
doi.o g/10.1023/A:1008219106132
Díaz-Mendoza, A. C., López-Espinosa, G., & Ma ínez, M. A.
(2014). The e iciency o pe o mance-based ee unds.
Eu opean Financial Managemen , 20, 825–855. h ps://doi.
o g/10.1111/j.1468-036X.2012.00654.x
Doshi, H., Elkamhi, R., & Simu in, M. (2015). Manage ial
ac i eness and mu ual und pe o mance. The Re iew o
Asse P icing S udies, 5, 156–184. h ps://doi.o g/10.1093/
aps u/ a 005
El on, E. J., G ube , M. J., Blake, C. R., K asny, Y., & Ozelge,
S. O. (2010). The e ec o holdings da a equency on con-
clusions abou mu ual und beha io . Jou nal o Banking
& Finance, 34, 912–922. h ps://doi.o g/10.1016/j.jbank-
in.2009.10.002
Eu opean Secu i ies and Ma ke s Au ho i y. (2016). Supe iso y
wo k on po en ial close index acking. h ps://www.esma.
eu opa.eu/si es/de aul / iles/lib a y/2016-165_public_s a e-
men _-_supe iso y_wo k_on_po en ial_close _index_
acking.pd
Fama, E. F., & F ench, K. R. (2010). Luck e sus skill in he
c oss-sec ion o mu ual und e u ns. The Jou nal o
Finance, 65, 1915–1947. h ps://doi.o g/10.1111/j.1540-
6261.2010.01598.x
Fa ooqi, J., Jo y, S., & Ngo, T. (2019). Ac i e und manag-
e s and ea nings managemen a po olio companies.
Re iew o Accoun ing and Finance, 19, 48–82. h ps://doi.
o g/10.1108/RAF-11-2017-0209
F azzini, A., F iedman, J., & Pomo ski, L. (2016). Deac i a ing
Ac i e Sha e. Financial Analys s Jou nal, 72, 14–21.
h ps://doi.o g/10.2469/ aj. 72.n2.2
Gil-Bazo, J., & Ruiz-Ve dú, P. (2009). The ela ion be ween p ice
and pe o mance in he mu ual und indus y. The Jou nal
o Finance, 64, 2153–2183. h ps://doi.o g/10.1111/j.1540-
6261.2009.01497.x
Golez, B., & Ma in, J. M. (2015). P ice suppo by bank-a il-
ia ed mu ual unds. Jou nal o Financial Economics, 115,
614–638. h ps://doi.o g/10.1016/j.j ineco.2014.10.008
Gue cio, D. D., & Tkac, P. A. (2002). The de e minan s o he
low o unds o managed po olios: Mu ual unds s. pen-
sion unds. Jou nal o Financial and Quan i a i e Analysis,
37, 523–557. h ps://doi.o g/10.2307/3595011
Haugen, R. A., & Lakonishok, J. (1988). The inc edible Janua y
e ec . Dow Jones-I win.
Jensen, M. C. (1968). The pe o mance o mu ual unds in he
pe iod 1945–1964. The Jou nal o Finance, 23, 389–416.
h ps://doi.o g/10.1111/j.1540-6261.1968. b00815.x
Jiang, H., Ve beek, M., & Wang, Y. (2014). In o ma ion con en
when mu ual unds de ia e om benchma ks. Managemen
Science, 60, 2038–2053. h ps://doi.o g/10.1287/mnsc.2013
.1847
Kacpe czyk, M., Sialm, C., & Zheng, L. (2005). On he indus y
concen a ion o ac i ely managed equi y mu ual unds. The
Jou nal o Finance, 60, 1983–2011. h ps://doi.o g/10.3386/
w10770
Kacpe czyk, M., Sialm, C., & Zheng, L. (2008). Unobse ed
ac ions o mu ual unds. The Re iew o Financial S udies,
21, 2379–2416. h ps://doi.o g/10.1093/ s/hhl041
Mo ey, M. R., & O’Neal, E. S. (2006). Window d essing in bond
mu ual unds. Jou nal o Financial Resea ch, 29, 325–347.
h ps://doi.o g/10.1111/j.1475-6803.2006.00181.x
Mus o, D. K. (1997). Po olio disclosu es and yea -end p ice
shi s. The Jou nal o Finance, 52, 1563–1588. h ps://doi.
o g/10.1111/j.1540-6261.1997. b01121.x
Ng, L., & Wang, Q. (2004). Ins i u ional ading and he u n-o -
he-yea e ec . Jou nal o Financial Economics, 74, 343–
366. h ps://doi.o g/10.1016/j.j ineco.2003.05.009
Odean, T. (1998). A e in es o s eluc an o ealize hei losses?
The Jou nal o Finance, 53, 1775–1798. h ps://doi.
o g/10.1111/0022-1082.00072
O iz, C., Ramí ez, G., & Vicen e, L. (2010). Qua e ly e u n
pa e ns in he Spanish s ock ma ke . Applied Financial
Economics, 20, 1829–1838. h ps://doi.o g/10.1080/09603
107.2010.528366
O iz, C., Ramí ez, G., & Vicen e, L. (2015). Mu ual und ading and
po olio disclosu es. Jou nal o Financial Se ices Resea ch,
48, 83–102. h ps://doi.o g/10.1007/s10693-014-0198-2
O iz, C., Sa o, J. L., & Vicen e, L. (2012). Po olios in dis-
guise? Window d essing in bond und holdings. Jou nal o
Banking & Finance, 36, 418–427. h ps://doi.o g/10.1016/j.
jbank in.2011.07.017
Pás o , L., & S ambaugh, R. F. (2002). In es ing in equi y mu ual
unds. Jou nal o Financial Economics, 63, 351–380.
h ps://doi.o g/10.1016/S0304-405X(02)00065-X
Pás o , L., S ambaugh, R. F., & Taylo , L. A. (2015). Scale and
skill in ac i e managemen . Jou nal o Financial Economics,
116, 23–45. h ps://doi.o g/10.1016/j.j ineco.2014.11.008
Pe ajis o, A. (2013). Ac i e Sha e and mu ual und pe o -
mance. Financial Analys s Jou nal, 69, 73–93. h ps://doi.
o g/10.2469/ aj. 69.n4.7
Sensoy, B. A. (2009). Pe o mance e alua ion and sel -des-
igna ed benchma k indexes in he mu ual und indus y.
Jou nal o Financial Economics, 92, 25–39. h ps://doi.
o g/10.1016/j.j ineco.2008.02.011
Si i, E. R., & Tu ano, P. (1998). Cos ly sea ch and mu ual und
lows. The Jou nal o Finance, 53, 1589–1622. h ps://doi.
o g/10.1111/0022-1082.00066
We me s, R. (2003). A e mu ual und sha eholde s compensa ed
o ac i e managemen be s? [Wo king pape ]. Uni e si y
o Ma yland.
