Institutional ownership and simultaneity of strategic financial decisions: an empirical analysis in the case of Pakistan Stock Exchange

172 2019, XXII, 1 Finance DOI: 10.15240/ ul/001/2019-1-012 In oduc ion The adi ional in e p e a ion o co po a e i nance is cha ac e ized by owne ship. Al hough, hei igh s a e widely dis ibu ed among indi idual s ockholde s, bu can be managed by ew manage s. Hence, con l ic o in e es is a isen among manage s and sha eholde s and his esul s in an agency p oblem (Fama, 1980; Fama & Jensen, 1983). A numbe o empi ical s udies also con i med he owne ship concen a ion o i ms, especially hose domina ed by ew la ge owne s o block-holde s (La Po a e al., 1999). The concen a ed s uc u e o owne ship also con ibu es owa ds agency con l ic be ween block-holde s and mino i y sha eholde s. F om ano he pe spec i e, he block-holde s can bene i mino i y sha eholde s by hei ole in moni o ing manage s and also can be haza dous i hey s i e o achie e hei own p i a e goals (Shlei e & Vishny, 1997). The h ee main aspec s o owne ship, which ha e been widely discussed in he pas h ee decades, include concen a ed owne ship by block-holde s, amilies and o he g oups, manage ial owne ship, and ins i u ional owne ship. The hi d aspec has gained impo ance as sha eholding by ins i u ional in es o s has inc eased in he US om 17% in 1970 o nea ly 70% in he p e ious decade (Bushee & Noe, 2000). Meanwhile, in he case o Pakis an, nea ly 25% o he common s ock is owned by local and o eign ins i u ional in es o s (Eas e ly, 2001). Ins i u ional owne ship is de i ned in he li e a u e as he pe cen age o i m’s sha es owned by ins i u ional in es o s and i can also be de i ned as one minus pe cen age o sha es held by indi idual in es o s (Fi h e al., 2016). Consequen ly, ins i u ional in es o s play an e ec i e moni o ing ole in he in es ed i ms. Ini ially, ea ly esea ch mainly ocused on he analysis o he ela ionship be ween concen a ed owne ship and i m pe o mance (McConnell & Se aes, 1990; Duggal & Milla , 1999). La e on, he ela ionship be ween ins i u ional owne ship and di e en domains o co po a e go e nance ha e opened addi ional esea ch ho izon, i.e. Ka po e al. (1996), Johnson and G eening (1999), Mak and Li (2001). Ins i u ional in es o s ha e di e se p e e ences o he i ms in which hey in es . A numbe o s udies ca ied ou o de e mine he p e e ences o ins i u ional in es o s in e ms o i ms’ co po a e go e nance and o he policies. Bad ina h, Kale, and Ryan (1996) in es iga e he idea ha ins i u ional in es o s a o s ocks ha ha e highe ma ke liquidi y and lowe e u n ola ili y. O he s also con i med ha ins i u ional in es o s alue he s ocks o companies wi h supe io disclosu e (Bushee & Noe, 2000), hose ha would pay cash di idends o epu chase sha es (G ins ein & Michaely, 2005), and also hose demons a e be e manage ial pe o mance (Pa ino e al., 2003). Ne e heless, Cull and Xu (2005) ha e s essed he ole o ins i u ional in es o s in in es men decisions, G ins ein and Michaely (2005) claimed he e ec s o di idend policy decisions and o he s in le e age o capi al s uc u e decisions (Bokpin & A ko, 2009; Chung & Wang, 2014). Mos impo an ly, his s udy aims a de e mining he simul aneous e ec s o ins i u ional owne ship on i ms’ s a egic i nancial decisions. Thus, he in e dependence o i ms’ s a egic decisions se a p oblem o endogenei y, leading INSTITUTIONAL OWNERSHIP AND SIMULTANEITY OF STRATEGIC FINANCIAL DECISIONS: AN EMPIRICAL ANALYSIS IN THE CASE OF PAKISTAN STOCK EXCHANGE Rabeea Sada , Judi Oláh, Józse Popp, Domicián Má é EM_1_2019.indd 172EM_1_2019.indd 172 8.3.2019 9:16:258.3.2019 9:16:25 173 1, XXII, 2019 Finance o a causal wo-way ela ionship be ween hem. Le e age o capi al s uc u e decisions a e a ec ed by di idend decisions and hese choices also ha e an in l uence on he le e age decisions o a i m (Al-Najja & Taylo , 2008). Howe e , s udies ha e conside ed he endogenei y be ween ins i u ional owne ship and payou policy (Chang, Kang, & Li, 2016), owne ship and i ms’ alue (A za & Nazi , 2015) and i ms’ pe o mance (Maquiei a, Espinosa, & Viei o, 2011), his phenomenon seems o be pa icula ly in e es ing in he case o he Pakis ani, whe e he le el o ins i u ional owne ship is high and conside able. The main pu pose o his s udy is o analyze ela ionships among ins i u ional owne ship and he i ms’ s a egic decisions ela ing o he le e age, he capi al s uc u e, di idend decisions and ela ed in es men decisions. This pape is o ganized as ollows. Sec ion II highligh s p e ious li e a u e and p oposes ou hypo heses. Sec ion III de ails he sample and esea ch design used o analysis. Sec ion IV includes he esul s o empi ical examina ion and he discussion o i s consequences. Sec ion V concludes and highligh s he impo ance o ins i u ional owne ship in i ms’ decision. 1. Li e a u e Re iew and Hypo hesis De elopmen The agency heo y sugges s ha op imal in es o s ha e a s ong in e es in moni o ing i ms’ managemen capi al s uc u e and owne ship s uc u e, which suppo i ms o minimize hei agency cos s (Jensen, 1986). Agency cos s a e a ibu ed o he a ise con l ic o in e es . Jensen and Meckling (1976) iden i i ed wo main ypes o con l ic s, i.e. con l ic s o in e es be ween he sha eholde s and manage s, and con l ic s be ween he sha eholde s and deb holde s. Keeping manage s’ absolu e in es men in i ms’ cons an , an inc ease in he a io o deb i nancing inc eases he manage s’ sha e o equi y and he e o e, i educes he loss om any con l ic be ween manage s and sha eholde s. Mo eo e , since he deb equi es he i m o pay ou cash as a cos o deb , his educes he amoun o ee-cash a ailable o manage s and in u n educes he con l ic o in e es . P e ious li e a u e on ins i u ional owne ship has p oposed hese solu ions in o de o gain bene i s by enhancing i ms’ alue (Shlei e & Vishny, 1997). Ne e heless, ins i u ional in es o s and deb can be subs i u ed one ano he as an al e na i e o moni o ing i ms. This hypo hesis is con i med by a numbe o empi ical s udies in he li e a u e (Li, Yue, & Zhao, 2009). Consis en ly, ag eeing wi h Al- Najja and Taylo (2008), we will p opose ha he ela ionship be ween ins i u ional owne ship and i ms’ le e age can be exp essed: H1: The deg ee o s ock owne ship by ou side ins i u ions is nega i ely ela ed o he le e age o he i ms. Since, Mille and Modigliani (1961) a gued ha di idend policy does no a ec he alue o he i m, di e en empi ical s udies ha e been conduc ed o in es iga e he di idend puzzle. T uong and Heaney (2007) also epo ed ha he i ms pay di idends and a e inclined o pay mo e di idends when hey ha e high le els o p o i abili y and low le els o in es men oppo uni ies. The classical agency heo y pe spec i e holds he iew ha i ms a e likely o sha e mo e o hei p o i s wi h in es o s when hey ace lowe moni o ing cos s (Jensen, 1986). I also holds ha he la ges sha eholde may educe agency cos s by educing he amoun o ee cash l ow o manage ial disc e ion by inc easing i ms’ payou s. Meanwhile, he li e a u e p o ides some e idence on he ela ionship be ween ins i u ional owne ship and he di idend decisions o i ms. Fi h e al. (2016) and Sho e al. (2002) con i med a posi i e co ela ion be ween ins i u ional sha eholding and he di idend payou s o i ms. G ins ein and Michaely (2005) ound a posi i e ela ionship be ween sha e epu chases and ins i u ional holdings. Acco ding o hei i ndings, i ms ha epu chase mo e sha es a ac mo e ins i u ional in es men s. Thei esul s also sugges ha ins i u ional in es o s p e e i ms ha epu chase sha es egula ly. Based on his discussion, ou second p oposi ion ega ding he ela ion be ween ins i u ional owne ship and di idend is he ollowing: H2: The pe cen age o s ock owne ship by ins i u ional in es o s is posi i ely ela ed o he di idend paymen o he i ms. Bushee (1998) demons a es ha he sho - e m ocus o many ins i u ional in es o s induces some i ms o educe R&D when ea nings a e expec ed o decline. On he basis o he in es men ho izon and p e e ences, ins i u ional in es o s a e classi i ed as ‘ ansien ins i u ions’, highligh ed manage s’ myopic beha io . The o he wo ypes o hese EM_1_2019.indd 173EM_1_2019.indd 173 8.3.2019 9:16:258.3.2019 9:16:25 174 2019, XXII, 1 Finance ins i u ions a e ‘dedica ed 1 and ‘quasi-indexe ’ ins i u ions. These ins i u ional in es o s ha e s able owne ship in i ms and a e less ocused on sho - e m ea nings. The ela ion be ween he in es men decisions o i ms and ins i u ional owne ship as p oposed in he li e a u e is no so s aigh o wa d. Howe e , in es men is one o he mos impo an aspec s h ough which ins i u ional owne s can a ec a i m. A posi i e ela ion be ween in es men and owne ship was con i med by Pindado and To e (2006), whe eas Richa dson (2006) epo s ha manage s o i ms wi h la ge ins i u ional owne ship a e less likely o o e in es su plus cash, due o he moni o ing go e nance ac i i ies o ins i u ions. Conside ing Bushee’s (1998) myopic in es o hypo hesis o be mo e ele an in he con ex o Pakis ani ins i u ional in es o s (in e ms o he sho - e m ho izon and lack o in o ma ion), he ela ionship is p oposed as, H3: The pe cen age o s ock owne ship by ins i u ional in es o s is nega i ely ela ed o he in es men o he i ms. Deb and di idend can subs i u e o complemen one ano he in educing agency cos s. These kinds o s a egies auxilia y well i con e gence o in e es is s ong (Roze , 1982). I he en enchmen hypo hesis o Fa inha (2002) is e ec i e hese decisions a e complemen . Belie ing on his, ou s udy p oposes a nega i e ela ion be ween le e age and di idends. H4: The le e age o he i m is nega i ely ela ed o he di idend. Some ea ly s udies examined how i ms’ op imum deb p e e ence a ec s in es men decisions. Smi h and Wa ne (1979) a gued ha deb can bound a i m’s abili y o employ an asse subs i u ion, while Be ko i ch and Kim (1990) discuss ha p ojec i nance and secu ed deb suppo o esol e in es men incen i e p oblems. Hackba h, Hennessy and Leland (2007) indica ed ha placing bank deb a he op o he i m’s p io i ies ully exploi s ax shield bene i s o in e es s. S udying in e ac ions be ween in es men s and i nancing decisions, hey examined he idea ha a dynamic ade-o be ween p io i y, capi al s uc u e and in es men incen i es yields impo an addi ional insigh s and u he empi ical p edic ions. Based on he li e a u e we assumed ha : H5: Le e age is nega i ely ela ed o i ms’ in es men decisions. Howe e , he in es men o i ms also a ec s hei di idend policy decisions. The ela ionships among he di idend and in es men decision policies a e e idenced om he heo e ical backg ound o Mille and Modigliani (1961). This heo y a gues ha in a pe ec capi al ma ke , op imal in es men decisions by a i m a e independen o how such decisions a e i nanced. This co e heo y has also an impo an ou come as in es men decisions should no be de e mined by di idends, and di idend decisions need no be a ec ed by in es men decisions. In his pe spec i e, Fama (1974) p o ided empi ical e idence o iolence o his heo em. Since hen, he e is no e idence o an exis ed ela ion be ween di idend and in es men decisions ha equi e ea ing hem ia simul aneous equa ions models (SEM). Fo hose i ms which ha e g ea in es men oppo uni ies, paymen o di idends mus be balanced wi h he long- e m goals o i ms (Mye s & Majlu , 1984). C u chley e al. (1999) ound a nega i e wo-way ela ionship be ween di idend and in es men decisions. An inc eased di idend can lead o educed unds a ailable o in es men and hence esul s in a dec eased di idend p obabili y o he u u e (Cye e al., 1996). Hence, we can also assume ha : H6: The di idend is nega i ely ela ed o he in es men decisions o i ms. Ou co esponding hypo hesis ega ding he ela ion among ins i u ional owne ship and le e age, di idend and in es men decisions a e summa ized in Fig. 1. 2. Sample and Resea ch Design The da a used o his s udy is comp ised o a sample aken om non- i nancial i ms lis ed on he Pakis an S ock Exchange (PSE). To al s a is ics o his s udy include all lis ed i ms in 33 di e en sec o s. A sample o 170 i ms belonging o eigh di e en sec o s we e conside ed o analysis be ween 1994 and 2014. The selec ion o he sample depends upon he a ailabili y o all he equi ed da a. Financial i ms, i ms wi h nega i e equi y and i ms whose ele an da a is incomple e o no a ailable a e excluded om his sample. Mo eo e , ou analyses a e based on annual equency o da a, in o de o align i nancial s a emen s esul s and ins i u ional owne ship a iables. In his s udy, a sec o al app oach is also used, ollowing King and San o (2008). EM_1_2019.indd 174EM_1_2019.indd 174 8.3.2019 9:16:268.3.2019 9:16:26 175 1, XXII, 2019 Finance The a iables ex ac ed om he Balance Shee Analysis (BSA) publica ion o he S a e Bank o Pakis an (SBP) (2018) include capi al s uc u e decisions cap u ed by he ollowing ac o s:  le e age (LEV), which measu ed by deb o equi y a io (Hassan & Bu , 2009),  di idend payou s (DPO), as he a io o di idend pe sha e o ea nings pe sha e (A za & Nazi , 2015),  in es men decisions (INV) measu ed by he a io o Change in Fixed Asse s o To al Asse s in place o R&D expendi u es (Jensen e al., 1992),  Re u n on Equi y (ROE), as he a io o ne income o sha eholde s’ equi y (Hillman & Dalziel, 2003),  size o i m (Size) equals wi h he na u al loga i hm o he book alue o o al asse s is used (Lin & Chang, 2011),  angibili y (TANG) o asse s is measu ed as he a io o i xed asse s o he o al asse s o i m (Liu e al., 2011),  Sales g ow h (Sales_GR) is calcula ed as he annual pe cen age change in sales o a company (Lin & Chang, 2011),  age o i m (Age) is de i ned as he log o numbe o yea s elapsed since a i m was lis ed (Roy, 2015),  ins i u ional owne ship (INST), as he pe cen age o sha es owned by ins i u ional in es o s o he o al numbe o sha es ou s anding (Michaely & Vincen , 2013). INST is aken om annual epo s o indi idual companies, epo ed unde he Code o Co po a e Go e nance o Pakis an (CCGP),  and he di idend decisions o i ms a e cap u ed by di idend yield (DY), as he a io o he di idend pe sha e o ma ke p ice pe sha e (B ad o d e al., 2013). In he me hod iden i i ed o his s udy, he e is a po en ial causali y o endogenous ela ionship among le e age decisions, di idend decisions and in es men decisions. A simple o dina y leas squa e (OLS) es ima ion o cap u e he ela ionship among hese a iables will c ea e biased and inconsis en es ima es, as gi en by (Demse z & Villalonga, 2001). Hence, he e is a need o explo e a mo e sophis ica ed econome ics echnique o analysis. The e a e di e en ways o add ess he issue o biased and inconsis en es ima es. One o hem is 2SLS, he o he s a e 3SLS and GMM. 3SLS has ad an ages o e 2SLS, as wi h he o me o cap u e c oss equa ion impac s o e o e ms and he sys em o equa ion is supposed o be co ela ed in 3SLS (Zellne & Theil, 1962). Fig. 1: The in e ela ion amewo k among ins i u ional owne ship and s a egic le e age, di idend and in es men decisions o i ms Sou ce: own based on he au ho ’s assump ions EM_1_2019.indd 175EM_1_2019.indd 175 8.3.2019 9:16:268.3.2019 9:16:26 176 2019, XXII, 1 Finance This s udy is based on he 3SLS me hodology in o de o analyze he simul aneous de e mina ion o i nancial decisions and hei possible wo-way causali y. 3SLS is always p e e ed in e m o he inhe en e i ciency o i s es ima es o e 2SLS (Kap eyn & Fiebig, 1981). 3SLS is he mos app op ia e echnique o his da a se i a sys em es ima o is conside ed a he han one equa ion. This me hod is designed o cap u e a ela ion whe e equa ions in a model ha e endogenous a iables as exogenous. Since some o he explana o y a iables a e endogenous a iables, he e o e ms o he equa ions a e co ela ed, which simply iola es he assump ions o O dina y Leas Squa e (Bal agi, 2008). In o de o analyze he impac o ins i u ional owne ship on a ious s a egic decisions o i ms, he ollowing eg ession models a e speci i ed as: (1) (2) (3) whe e le e age (LEV), di idend yield (DY) and in es men (INV) a e dependen a iables in hese equa ions, showing a possible causali y ( wo-way) ela ionship among hem, since hey also appea on he igh side o he equa ion as exogenous a iables. (e) ep esen s he e o e ms o he equa ions 1, 2 and 3, and hey a e also assumed o be co ela ed. ROE, Size, TANG, Sales_GR, Size and Age a e addi ional con ol a iables in hese equa ions, as o e ed Bokpin and A ko (2009); Chang e al., (2016). In o de o con i m he obus ness o ou esul s, he abo e equa ions a e speci i ed wi h some dummy a iables o cap u ing he c oss indus y-e ec s (T uong & Heaney, 2007). Indus y speci i c dummies a e combined wi h he o mal Code o Co po a e Go e nance o Pakis an in 2002, and all lis ed companies in Pakis an should ollow a ull ep esen a ion o he CCGP. Conside ing he ime e ec s, a ious yea dummies a e also added o accoun o he impac o ins i u ional owne ship and s a egic decisions o e ime. 3. Empi ical Analysis and Resul s Tab. A.1 epo s in Appendix he desc ip i e s a is ics o all a iables. This able epo s he mean o a e age alue, he s anda d e o o he mean, he median, he s anda d de ia ion, skewness and ku osis o 1,502 obse a ions. The mean o a e age alue o LEV is 1.69 wi h a s anda d de ia ion o 1.48. Le e age alue is highe and i shows a g ea e eliance o i ms lis ed in he PSE on ex e nal sou ces o i nancing, as epo ed ea lie by A za and Nazi (2015). Thus, he a e age alue o INV is .006%, wi h a s anda d de ia ion o .00046. The a e age ins i u ional owne ship in Pakis ani i ms is epo ed as 32.14% o o al sha es ou s anding wi h a s anda d e o o 0.63. Ne e heless, in o de o check he he e oscedas ici y, a ious Whi e es s a e applied. The esul s o hese es s a e epo ed in Tab. A.2 (Panel A) and he chi-squa e alues con i m he p esence o he e oscedas ici y in ou model. Panel B also he p obabili y o es s a is ics con i ms he p esence o se ial co ela ion. The co ela ion be ween he a iables (Tab. A.3) o he sample selec ed is analyzed using he Pea son co ela ion. The esul shows a posi i e co ela ion be ween ins i u ional owne ship and di idend yield. Thus, ins i u ional owne ship shows a sligh nega i e co ela ion wi h le e age, whe eas i seemed no co ela ion wi h in es men . The ollowing Tab. 1 epo s he co esponding esul s a e 3SLS analysis o each models. The ou comes lead o he ollowing implica ions. In model 1, he coe i cien o di idend payou s (DPO) a io (-0.711) is also nega i e and signi i can a one pe cen . This con i ms he simul aneous de e mina ion o di idend and le e age. Consequen ly, we also claimed ha i ms use le e age and di idend as al e na i e moni o ing de ices. In o he wo ds, i ms paying highe di idends i nd deb a less a ac i e sou ce o i nancing (Ogden & Wu, 2013). Mo eo e , i ms wi h highe i nancial cos s a e no eady o pay di idends. A highe i m le e age will lowe he po en ial di idend payou o sha eholde s (T uong & Heaney, 2007). The le e age o i ms (LEV) is nega i ely ela ed o ins i u ional owne ship and he esul EM_1_2019.indd 176EM_1_2019.indd 176 8.3.2019 9:16:268.3.2019 9:16:26 177 1, XXII, 2019 Finance is highly signi i can . This esul shows ha ins i u ional owne s, ega dless o hei ype, a e hesi an o in es subs an ial s akes in i ms ha a e highly le e aged. This nega i e and signi i can ela ionship also suppo s he a gumen ha ins i u ional owne s may ac as a subs i u e o he moni o ing ole o deb in he capi al s uc u e o i ms (Moh’d e al., 1998; Chung & Wang, 2014). In his pe spec i e, he eluc ance o ins i u ional owne s o in es in highly le e aged i ms may be due o hei in en ion o a oid isk (C u chley e al., 1999). The p o i abili y o a i m is also nega i ely ela ed o he i m’s le e age, and he coe i cien is signi i can a 95%. This esul is in line wi h he pecking o de heo y (Mye s & Majlu , 1984), sugges ing a nega i e ela ion due o he eliance o i ms’ in e nally gene a ed unds. Essen ially, i ms’ size has a posi i e ole in de e mining he le el o le e age. Consequen ly, bigge i ms a e mo e le e aged han smalle ones. The sales g ow h has posi i e and signi i can esul s, as claimed Al-Najja and Taylo (2008). These i ndings con adic agency heo y, suppo ing he nega i e ela ion be ween Va iables LEV (Model 1) DY (Model 2) INV (Model 3) Cons an 1.3430*** -0.00467 -0.0001 SE -0.3358 -0.0093 -0.0001 INV -0.0009 -0.0000 SE -0.0010 0.0000 LEV -0.00010 0.0000 SE -0.0002 0.0000 DY 0.0000 SE -0.0005 DPO -0.711*** -0.711*** 0.0744*** 0.0744*** SE -0.1402 -0.1402 -0.0039 -0.0039 INST -0.0041*** -0.0041*** 0.0008* 0.0008* -0.0003 SE -0.0014 -0.0014 0.0000 0.0000 -0.0001 ROE -0.0154*** -0.0154*** 0.0004*** 0.0004*** -0.0000 SE -0.0016 -0.0016 0.0000 0.0000 0.0000 Size 0.1827*** 0.1827*** -.00006 -.0000 SE -0.0261 -0.0261 -0.0007 0.0000 SALES_GR 0.0017** 0.0017** 0.0000 0.0000*** 0.0000*** SE -0.0009 -0.0009 -0.0001 0.0000 0.0000 TANG -0.4372*** -0.4372*** 0.0022 0.0001** 0.0001** SE -0.1808 -0.1808 -0.0050 -0.0001 -0.0001 Age -0.2489 0.0142*** 0.0142*** 0.0001*** 0.0001*** SE -0.1857 -0.0051 -0.0051 0.0000 0.0000 Adj. R-squa ed 0.1030 0.3213 0.0580 Chi-Sq. 172.29*** 710.87*** 92.70*** Hausman Tes S a is ics: Chi-Sq.= 1.41, P ob.>Chi-Sq.= 0.9941 Sou ce: own based on (S a e Bank o Pakis an, 2018) and au ho ’s own calcula ions No e: *** deno es 1%, ** 5% and * 10% le el o signi i cance. SE is obus s anda d e o . Model 1, 2 and 3 co esponds o equa ions 1, 2 and 3 espec i ely. Tab. 1: Resul s o 3SLS Reg essions Based on Equa ion 1, 2 and 3 EM_1_2019.indd 177EM_1_2019.indd 177 8.3.2019 9:16:268.3.2019 9:16:26 178 2019, XXII, 1 Finance owne ship and g ow h due o ac ha g owing i ms end no o ans e hei weal h o c edi o s. Thus, angibili y has a nega i e and signi i can ela ionship wi h le e age. The nega i e ela ion can be a ibu ed o he p esence o ins i u ional o block-holde owne ship which esul s in close ies wi h lende s, hus educing he need o mo e colla e al (Deesomsak e al., 2004). The esul s o second models highligh ha he coe i cien o in es men (INV) and deb (LEV) wi h di idend is al hough nega i e, bu insigni i can in ou models. Thus, he coe i cien s o ins i u ional owne ship (INST) a e posi i ely and signi i can ly co ela ed wi h di idend yield (DY). The eason o inc eased di idend le els can be he ole o ins i u ional sha eholde o ing igh s o highe di idends o enhance manage ial moni o ing (Fa inha, 2002). Thus, one uni inc ease in p o i abili y a ios inc eased he le el o di idends. This suppo s ha mo e p o i able i ms wi h ce e is pa ibus highe le els o ins i u ional owne ship end o pay mo e di idends han he less ones (T uong & Heaney, 2007). Examining he signi i can con ol a iables, only Age shows a posi i e and subs an ial coe i cien a 1%. These i ndings consis en wi h he i ndings o Thana awee (2012) ega ding he i ms’ endency o paying inc eased di idends. Ne e heless, he ela ionship o ins i u ional owne ship (INST) and he i m’s in es men (INV), as sugges ed by he hi d model, is nega i e. Al hough he esul is s a is ically insigni i can , he nega i e coe i cien is in acco dance wi h he i ndings o Richa dson (2006). The sho e m ocus o ins i u ional in es o s may cons ain he manage o educe in es men (Bushee, 1998) o a oid misp icing caused by disappoin ed ins i u ional in es o s’ selling. The posi i e and signi i can ela ion be ween age and in es men a iables also p o ides suppo o he same p oposi ion. The coe i cien o angibili y is signi i can and posi i e, indica ing ha capi al-in ensi e i ms a e s ill in he p ocess o expansion. The same phenomenon is con i med by he signi i can and posi i e ela ionship be ween sales g ow h and in es men . The esul s indica e high sales g ow h equi es he i m o place mo e money in expansion o p ojec /p oduc ion acili ies. This consequence co esponds o he i ndings o Jensen e al. (1992). Howe e , he e is no e idence o simul anei y in di idend and in es men decisions o i ms, bu esul s a e seemed o consis en wi h he i ndings o Fama (1974). Tab. 2 demons a es he esul s o simul aneous equa ions whe e indus y speci i c dummies a e inco po a ed in o he model. The omi ed con ol dummy a iable ep esen s Enginee ing, which becomes a e e ence o all o he indus ies. All he epo ed esul s emain he same in e ms o hei sign and signi i cance, wi h some excep ions. The coe i cien o age in he i s le e age (LEV) model becomes signi i can a e he addi ion o dummies. In o he wo ds, one uni inc ease in age o i ms seemed o dec ease deb o equi y a io. The e o e, ageing i ms a e less le e ed in Pakis an. Thus, he nega i e impac o ins i u ional owne ship on i ms’ le e age becomes mo e p onounced a e including indus y speci i c dummies. O e all esul s a e consis en wi h bo h analyses, sugges ing ha he speci i ca ions o ou models a e obus . The explana o y in l uence o he models has also inc eased a e including indus y speci i c a iables. Va iables LEV (Model 1) DY (Model 2) INV (Model 3) Cons an 1.104*** -0.0180* -0.00006 SE -0.3623 -0.0101 -0.0001 LEV -0.00007 -0.0000 SE -0.0002 0.0000 INV -0.0006 0.0000 SE 0.0010 0.0000 DY -0.0005 SE -0.0006 Tab. 2: Resul s o 3SLS Reg essions wi h sec o al dummies (Pa 1) EM_1_2019.indd 178EM_1_2019.indd 178 8.3.2019 9:16:268.3.2019 9:16:26 179 1, XXII, 2019 Finance Va iables LEV (Model 1) DY (Model 2) INV (Model 3) DPO -0.5715*** -0.5715*** 0.0820*** 0.0820*** SE -0.1447 -0.1447 -0.0040 -0.0040 INST -0.0044*** -0.0044*** 0.0001** 0.0001** -0.0000 SE -0.0015 -0.0015 0.0000 0.0000 0.0000 ROE -0.0147*** -0.0147*** 0.0005*** 0.0005*** -0.0000 SE -0.0016 -0.0016 0.0000 0.0000 0.0000 Size 0.2043*** 0.2043*** -0.0000 -0.000005 SE -0.0286 -0.0286 -0.0008 0.0000 SALES_GR 0.0016* 0.0016* 0.0000 0.0000*** 0.0000*** SE -0.0009 -0.0009 0.0000 0.0000 0.0000 TANG -0.4607** -.4607** 0.0002 0.0001*** 0.0001*** SE -0.1902 -0.1902 -0.0053 -0.0001 -0.0001 Age -0.3422* -0.3422* 0.0135** 0.0135** 0.0053* 0.0053* SE -0.1919 -0.1919 -0.0054 -0.0054 -0.0001 -0.0001 CHEM -0.1604 -.00431 0.0000 SE -0.1609 -0.0045 -0.0001 CONS -0.2422 -0.0018 -0.0001* -0.0001* SE -0.2012 -0.0056 -0.0001 -0.0001 PAPER 0.1432 0.0036 -0.000015 SE -0.2423 -0.0068 -0.0001 ENERGY 0.3315 0.0036 -0.0000 SE -0.2067 -0.0058 -0.0001 FOOD 0.5973*** 0.5973*** 0.0112** 0.01120** -0.0001** -0.0001** SE -0.1606 -0.1606 -0.0045 -0.0045 -0.0001 -0.0001 PERSONAL 0.3197** 0.3197** 0.0170*** 0.0170*** -0.0000 SE -0.1469 -0.1469 -0.0041 -0.0041 0.0000 MISC -0.6593*** -0.6593*** 0.0063 -0.0000 SE -0.2168 -0.2168 -0.0061 -0.0001 Adj. R-squa ed 0.1479 0.3486 0.0711 Chi-Sq. 260.66*** 803.53*** 112.85*** Hausman Tes S a is ics: Chi-Sq.= 1.31, P obabili y > Chi-Sq.=1.00 Sou ce: based on (S a e Bank o Pakis an, 2018) and au ho ’s calcula ions No es: *** deno es 1%, ** 5% and * 10% le el o signi i cance. SE is obus s anda d e o . Models co espond o he equa ions (1, 2 and 3) a e inco po a ing indus y dummies espec i ely. CHEM is chemical indus y; CONS is cons uc i- on and ma e ial indus y; PAPER is pape & boa d indus y; ENERG is uel and ene gy sec o ; FOOD is ood p oduce indus y; PERSONAL is pe sonal goods indus y and MISC is miscellaneous indus ies. Tab. 2: Resul s o 3SLS Reg essions wi h sec o al dummies (Pa 2) EM_1_2019.indd 179EM_1_2019.indd 179 8.3.2019 9:16:278.3.2019 9:16:27 180 2019, XXII, 1 Finance Essen ially, addi ional ime dummies a e added o cap u e he e ec s o pa icula yea s a ec s he di idend equa ion. As a esul , he size o i ms (Size), he sales g ow h (SALES_ GR) and angibili y (TANG), which we e p e iously insigni i can , a e subs an ial wi h he di idend model now, as epo ed in Tab. 3. Namely, i he sales g ow h is inc eased mo e di idends seemed o pay o s ockholde s. Howe e , i he e is one uni inc ease in size and angibili y o i ms less disbu semen a e pu chased o he owne s. These esul s a e simila wi h (Lin & Chang, 2011) i ndings. Gene ally, ou esul s in he case o LEV and INV ha e less changed and emained insigni i can in he second and hi d ables, sugges ing he obus ness o he esul s. Adjus ed R-squa ed alues o DY and INV ha e inc eased, as well. A he bo om o each epo ed ables a e he app op ia ed adjus ed R-squa ed and Chi-squa e alues. Mo eo e , addi ional Hausman es s a e also epo ed a each able ega ding 3SLS analysis. These s a is ics a e pe o med o analyze he expec ed di e ence be ween he coe i cien s using 2SLS and 3SLS me hods. The epo ed alues o hem a e 1.37, 1.39 and 0.69 espec i ely. These coe i cien s a e insigni i can in all models, and indica ing no signi i can di e ence be ween he wo models o pa icula se s o sys em (Bal agi, 2008). P e ious s udies (Chang e al., 2016) also con i med ha he p esence o ou lie s a ec s he o e all explana o y powe o he examined model. This 3SLS me hod applied on he ‘winso ized’ da ase o elimina e he e ec o ex eme alues esul s in inconsis ency (Wilson, 1993). A e emo ing hese ou lie s, he esul s ecen ly show imp o ed explana o y powe . Howe e , gene al conclusions a e gi en only i u he de e minan s will be aken in o conside a ion o de e mine hei e ec s on i ms’ s a egic decision-making. The e o e, he alidi y o ou esul s is limi ed by he bias caused by he exclusion o he omi ed a iables o ou models. Va iables LEV (Model 1) DY (Model 2) INV (Model 3) Cons an 0.7010* 0.0442*** -0.0006*** SE -0.3699 -0.0096 -0.0001 LEV 0.0001 -0.0000 SE -0.0002 0.0000 INV -0.0011 0.0000 SE -0.0010 0.0000 DY 0.0004 SE -0.0006 DPO -0.8026*** -0.8026*** 0.0759*** 0.0759*** SE -0.1401 -0.1401 -0.0036 -0.0036 INST -0.0049*** -0.0049*** 0.0001*** 0.0001*** -0.0000 SE -0.0015 -0.0015 0.0000 0.0000 0.0000 ROE -0.0152*** -0.0152*** 0.0005*** 0.0005*** -0.0000 SE -0.0017 -0.0017 0.0000 0.0000 0.0000 Size 0.2131*** 0.2131*** -0.0026*** -0.0026*** -0.0000 SE -0.0267 -0.0267 -0.0007 -0.0007 0.0000 SALES_GR 0.0008 0.0000** 0.0000** 0.00000*** 0.00000*** SE -0.0010 0.0000 0.0000 0.0000 0.0000 TANG -0.610 -0.0097** -0.0097** 0.0001** 0.0001** SE -0.1819 -0.0047 -0.0047 -0.0001 -0.0001 Tab. 3: Resul s o 3SLS Reg essions wi h ime dummies (Pa 1) EM_1_2019.indd 180EM_1_2019.indd 180 8.3.2019 9:16:278.3.2019 9:16:27 187 1, XXII, 2019 Finance DY LEV INV INST ROE Size TANG S_GR Age DPO DY 1 LEV -0.21*** 1 INV -0.05* 0.055** 1 INST 0.062** -0.048* -0.013 1 ROE 0.380*** -0.23*** -0.019 -0.006 1 Size 0.108*** 0.094*** 0.094 0.15*** 0.20*** 1 TANG -0.171*** 0.047* 0.045* 0.009 -0.23*** 0.03 1 S_GR 0.063** 0.011 0.22*** 0.009 0.165*** 0.11 -0.022 1 Age 0.092*** -0.04 0.042 -0.014 0.094*** 0.036 -0.17*** 0.01 1 DPO 0.511*** -0.174 -0.029 .061** 0.310*** 0.15*** -0.23*** -0.003 0.02 1 Sou ce: own based on (S a e Bank o Pakis an, 2018) and au ho ’s own es ima ions No e: Co ela ion is signi i can * a 0.10 le el, ** a 0.05 le el, *** a 0.01 le el Tab. A.3: Co ela ion ma ix o he examined a iables EM_1_2019.indd 187EM_1_2019.indd 187 8.3.2019 9:16:288.3.2019 9:16:28 188 2019, XXII, 1 Finance Abs ac INSTITUTIONAL OWNERSHIP AND SIMULTANEITY OF STRATEGIC FINANCIAL DECISIONS: AN EMPIRICAL ANALYSIS IN THE CASE OF PAKISTAN STOCK EXCHANGE Rabeea Sada , Judi Oláh, Józse Popp, Domicián Má é The adi ional in e p e a ion o co po a e i nance is cha ac e ized by owne ship igh s a e widely dis ibu ed among indi idual s ockholde s, bu can be managed by ew manage s and esul ed in an agency p oblem. The p ima y objec i e o his esea ch s udy is o in es iga e he ela ionship be ween ins i u ional owne ship and i ms’ s a egic decisions. These s a egic decisions include i.e. le e age, di idend and in es men decisions. The examined da a is used om 170 non- i nancial Pakis ani lis ed i ms, cha ac e ized by a la ge pe cen age o ins i u ional in es o s, wi h a mul iple equi y s ake in di e en i ms ac oss a wide i eld o indus ies. This s udy is also able o show wo impo an no el ies. Fi s ly, he ac ha p e ious esea che s ha e al eady concen a ed on he impac o ins i u ional owne ship on indi idual s a egic decisions, as di idend o le e age policies and se e al unanswe ed ques ions emain. Consequen ly, he impac o ins i u ional owne ship has explo ed collec i ely on a ious s a egic decisions. Secondly, his s udy also ecognizes he de e mina ion o s a egic decisions by conside ing he endogenei y p oblem wi h a Th ee-S age Leas Squa e (3SLS) me hod. Essen ially, he e ec s o ins i u ional owne ship on i ms’ le e age becomes mo e p onounced a e including indus y speci i c and ime dummies in eg ession models. Based on he esul s, he case o inc eased ins i u ional owne ship o i ms has a signi i can nega i e e ec on le e age, and a posi i e e ec on di idend decisions. Hence, ins i u ional in es o s a e seemed o p e e low le e aged and high di idend-paying i ms. Mo eo e , his s udy has no able o i nd signi i can wo-way ela ions be ween ins i u ional owne ship and in es men decisions, so ins i u ional in es o s a he ocus on co po a e go e nance and in e nal con ol o i ms. Indeed, ins i u ional in es o s should de elop he e i ciency o i ms’ managemen o suppo mo e adequa e co po a e go e nance policies, and no only o eme ging ma ke s. Key Wo ds: S a egic decisions, endogenei y, ins i u ional owne ship, 3SLS. JEL Classi i ca ion: G31, G32, G11. DOI: 10.15240/ ul/001/2019-1-012 EM_1_2019.indd 188EM_1_2019.indd 188 8.3.2019 9:16:288.3.2019 9:16:28