A note on the seasonality of stock returns on the Johannesburg Stock Exchange

B ad ield, D. J. A icle A no e on he seasonali y o s ock e u ns on he Johannesbu g S ock Exchange Sou h A ican Jou nal o Business Managemen P o ided in Coope a ion wi h: Uni e si y o S ellenbosch Business School (USB), Bell ille, Sou h A ica Sugges ed Ci a ion: B ad ield, D. J. (1990) : A no e on he seasonali y o s ock e u ns on he Johannesbu g S ock Exchange, Sou h A ican Jou nal o Business Managemen , ISSN 2078-5976, A ican Online Scien i ic In o ma ion Sys ems (AOSIS), Cape Town, Vol. 21, Iss. 1/2, pp. 7-9, h ps://doi.o g/10.4102/sajbm. 21i1.909 This Ve sion is a ailable a : h ps://hdl.handle.ne /10419/218040 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/4.0/ S.A .J.Bus.Mgm .1990,21(1) 7 A no e on he seasonali y o s ock e u ns on he Johannesbu g S ock Exchange D.J. B ad ield Depa men o Ma hema ical S a is ics, Uni e si y o Cape Town, Rondebosch, 7700 Republic o Sou h A ica E idence om s udies on he majo s ock exchanges wo ld-wide sugges s ha s ocks lis ed on hese ma ke s ea n abno mally high e u ns in he mon h o Janua y. In his a icle he seasonali y o s ocks on he J?h~~nesbu g S ock Exchange is empi ically in es iga ed. Su p isingly no Janua y e ec s a e ound, howe e , a s1gm 1can Decembe seasonal e ec is documen ed. A plausible explana ion o his inding is o e ed. Ge uienis ui s udies op e ek ebeu se dwa soo die we eld dui aan da aandele abno male e diens es gedu ende Janua ie c dien. In hie die a ikel wo d die moon likheid an 'n soo gelyke e skynsel op die Johannesbu gse E ek e Beu s empi ics onde soek. Ve basend wo d geen Janua ie-e ek ge ind nie, maa wel 'n be ekenis olle Descmbe -e ek. Moon like e duidelikings i hie die be inding wo d oo ges el. In oduc ion P ac i ione s wo ld-wide ha e ecognized he impo an ole played by Capi al Ma ke Theo y in he ield o mode n inance. Decision make s in his ield ha e undoub edly ollowed he empi ical indings ela ing o Capi al Ma ke Theo y wi h keen in e es . The as quan i y o ecen li e a u e dealing wi h misspeci ica ions o he Capi al Asse P icing Model (CAPM) is e idence o he demand by p ac i ione s o esea ch in his a ea. O he ecen indings, one in pa icula has cap u ed he in e es o he in e na ional in es men communi y, namely, he o e whelming e idence o a Janua y seasonal e ec mani es ed in he es ima e o he CAPM-based isk p emium. This phenomenon has come o be known as he Janua y e ec and has been documen ed on mos o he majo s ock ma ke s wo ld-wide. In his a icle he seasonali y o s ocks on he Johannesbu g S ock Exchange (JSE) is in es iga ed and compa ed o he exis ing in e na ional e idence. Roze & Kinney (1976), Keim (1983), and Roll (1983) we e among he i s esea che s o iden i y ha he New Yo k S ock Exchange (NYSE) s ock ma ke e u ns a e, on a e age, highe in Janua y han any o he mon h o he yea . A en ion subsequen ly shi ed o in es iga ing his e ec on o he s ock exchanges as well. A summa y o esul s om in e na ional s ock ma ke s was p esen ed by Co hay, Hawawini & Michel (1987). In pa icula hey examined da a om he New Yo k (NYSE), he London (LSE), he Pa is (PSE), and he B ussels (BSE) s ock exchanges. They ound seasonali y in e u n exis ed in all o hese exchanges in Janua y. In addi ion, signi ican seasonali y was ound in o he mon hs as well. Fo example he LSE exhibi ed addi ional seasonali y in Ap il, he PSE in July, and he BSE in Feb ua y, Ap il, June, July and Oc obe . A emp s o explain his Janua y e ec ha e been, a mos , pa ially success ul. The mos popula hypo hesis a ibu es he e ec o yea -end ax-loss selling. This hypo hesis assumes ha sha es whose p ices dec eased du ing he yea would be subjec o selling p essu e owa ds he end o he yea . Consequen ly, p ices a e dep essed p io o he end o he yea and ebound a he beginning o Janua y. Keim (1986) e iewed e idence in he li e a u e which a emp s o explain he Janua y e ec and sugges ed ha liquidi y o pay oll e ec s may in luence mon hly e u ns. Tes ing me hodology Some eseache s conce ned wi h in es iga ing seasonali y in s ock e u ns ha e conside ed his concep in conjunc ion wi h he size e ec . Howe e , B ad ield (1988, 1989) ound no e idence o a size e ec on he JSE (using bo h uni a ia e and mul i a ia e es ing p ocedu es). Hence, in his in es iga ion he hypo hesis o seasonali y in JSE s ock e u ns will be in es iga ed sepa a ely. The es ing p ocedu e ou lined below aims a in es iga ing whe he seasonali y exis s in equally weigh ed indices cons uc ed o his pu pose. The a ionale behind his app oach is ha e u ns o s ock ma ke indices (minus he isk- ee a e) a e usually used as su oga es o he excess ma ke e u n (i.e. he isk p emium) in he CAPM. Owing o he unique na u e o he JSE, se e al indices we e conside ed, namely an equally weigh ed mining sha e index, an indus ial sha e index, and an all-sha e index. These indices we e cons uc ed om he eco d o ading s a is ics o e he pe iod Janua y 1974-Decembe 1984. All sha es ha had eliable eco ds o e his pe iod we e included in he espec i e indices. This amoun ed o 112 mining sha es, 357 indus ial sha es, and 469 sha es in o al in he espec i e indices. Weekly e u ns o each o he abo e-men ioned indices we e compu ed, and subsequen ly pa i ioned and a e aged wi hin each o he ele an mon hs o he yea . This amoun ed o 11 da a poin s (i.e. 11 yea s) associa ed wi h each o he 12 mon hs o each o he h ee indices. This p ocedu e is simila o he p ocedu e ou lined by Co hay, Hawawini & Michel (1987). The JSE as a whole The esul s 1 o he equally weigh ed all-sha e index cons uc ed om he 469 JSE s ocks a e shown in Table 1. Resul s om a ious o he s ock exchanges conduc ed o e he 1970-1983 pe iod (quo ed in Co hay, Hawawini & Michel, 1987) a e also included in Table 1 . o compa a i e pu poses. 8 Table 1 Seasonal esul s Numbe o lis ed NYSE LSE PSE BSE JSE i ms 1490 2170 504 197 671 Sample size 782 527 112 170 469 Janua y 5,08" 5,49" 4,10" 3,99" 2,45 (2,10) (2,04) (2,40) (4,08) (1,09) Feb ua y 0,78 2,21 0,41 1,86" 2,12 (0,74) (1,58) (0,36) (2,72) (1,52) Ma ch 1,51 0,73 1,59 0,37 0,41 (0,94) (0,42) (0,75) (0,49) (0,22) Ap il 0,57 4,19" 1, 17 1,93" 0,78 (0,36) (3,31) (I, 19) (2,71) (0,76) May -0,70 -0,48 -0,69 -0,08 2,19 (-0,50) (-0,41) (-0,45) (-0,13) ( 1,41) June 0,70 -1,39 -1,56 1,06" 0,68 (0,59) (-0,96) (-1,11) (2,06) (0,32) July 0,92 1,22 3,92" 1,76" 2,89" (0,67) (1, 18) (2,76) (2,98) (2,02) Augus 1,06 1,13 1,96 0,28 2,57 (0,66) (0,83) ( 1,52) (0,36) (1,21) Sep embe 0,31 -1,04 -0,40 -0,94 1,39 (0,23) (-0,52) (-0,21) (-1,17) (0,67) Oc obe -0,91 -0,07 -1,76 -1,48" 1,25 (-0,42) (-0,06) (-1,24) (-2,48) (1,34) No embe 2,25 -0,06 -0,27 -0,70 -0,36 (1,22) (-0,03) (-0,22) (-0,98) (-0,26) Decembe 1,48 1,62 0,37 1,58" 2,78" ( 1,31) (1,06) (0,41) (1 ,58) (2,96) s a is ics a e in pa en hesis. They a e compu ed as (R) = R>,ln I Vu(R) whe e n = numbe o yea s • signi ican a he 5% le el o a one-sided al e na i e · The esul o JSE s ocks in Table 1 is indeed su p ising. I shows ha an insigni ican Janua y e ec exis s o JSE s ocks in he ace o o e whelming e idence o a Janua y e ec in in e na ional ma ke s. A signi ican July and Decembe seasonal e ec is, howe e , e iden o JSE s ocks. Se e al hypo hesis ba e been o e ed o he Janua y e ec on he NYSE, and i is hus also concei able ha he signi ican Janua y seasonals in he o he in e na ional s ock ma ke s could be explained by simila hypo heses. Fu he mo e, he July seasonal on he PSE is likely o be caused by di idend paymen s (acco~ding o Hamon, 1986). App oxima ely wo- hi ds o all di idend paymen s in F ance a e paid in July. The high incidence o signi ican mon hly seasonals on he BSE, is howe e , di icul o explain. A plausible cause o his may be ela ed o he ela i e size o he s ock ma ke s. Rela i e o he wo ld ma ke capi aliza ion2 he BSE comp ises 0,32% in con as o 40, 6, and 1 % o he NYSE, LSE, and PSE, espec i ely. Fu he mo e, subs an ially ewe sha es a e quo ed on he BSE, namely 197 compa ed o 1490, 2171, and 504 on he NYSE, LSE, and PSE, espec i ely. Issues like hin and lacklus e ading on smalle ma ke s may he e o e ha e an in luence on he a iabili y o he ou come o S.-A .Tydsk .Bed y sl.1990,21(1) es s on hese ma ke s. Explana ions o he seasonal e ec s on he JSE shown in Table 1 will be o e ed in he inal sec ion o his a icle. The mining and indus ial sec o s on he JSE The esul s o he equally weigh ed mining and indus ial indices a e shown in Table 2. Resul s o he all-sha e index a e included o compa a i e pu poses. F om Table 2 i can be seen ha he Decembe seasonal e ec is signi ican o all h ee da a se s, and ha he seasonal e ec o July is signi ican a he 2,5% le el o he mining index, whe eas ha o Oc obe is signi ican a he 5% le el o he indus ial index. S ic ly speaking he es should ha e a wo-sided al e na i e hypo hesis o inco po a e signi ican ly poo mon hs in he al e na i e hypo hesis. This change would mean ha he abo e signi icance le els a e e ec i ely doubled4• Thus, he only o he signi ican mon h a he 5% le el besides Decembe , occu s o mining s ocks du ing July. Explana ions Since di idends we e no included in his s udy, he s a is ical signi icance is unlikely o be caused by di idends. Fu he mo e, di idend announcemen s a e usually made in Feb ua y o Ma ch o mid-yea di idends and Augus o Sep embe o end-o - he-yea di idends. Hence, he seasonal e ec is unlikely o be caused by issues ela ing o announcemen s. I is possible ha he July e ec e iden o mining s ocks is a consequence o he pa icula sample pe iod. Howe e , he signi ican Decembe e ec occu ing in all h ee indices is unlikely o be explained away as easily. I is well known ha hin and lacklus e ading on he JSE is cha ac e is ic o Decembe , which is adi ionally he holiday season in Sou h A ica, and ha his impac s on he ola ili y o s ocks o e his mon h. In pa icula , hin ading gi es ise o non-synch onous p ices being eco ded, causing he a iance o hese e u ns o be unde es ima ed o e hese pe iods. Fo example, i a pa icula sha e has no been aded o se e al weeks, i s p ice will be eco ded daily a he las ansac ion · p ice, whe eas he ' ue' p ice a ies acco ding o he economic in luences. Hence, he ' eco ded' e u n se ies will ha e subs an ially less a iabili y han he ' ue' se ies. Al hough his a gumen can easily explain why he a iance o daily o weekly e u ns in Decembe , ela i e o o he mon hs, is unde es ima ed, i does no ully explain why he se ies o mon hly e u ns o each Decembe o e he sampled pe iod exhibi ela i ely less ola ili y. Howe e , he ac ha la ge-scale selling o buying p essu e o e he Decembe holiday pe iod is unlikely may imply less a iabili y in mon hly Decembe e u ns. F om Table 2 i is seen ha o he indus ial index, Decembe mon h had he smalles s anda d de ia ion. Fo he all-sha e index he second smalles s anda d de ia ion occu ed in Decembe , and o he mining index he Decembe s anda d de ia ion was smalle han mos o he mon hs, al hough no he smalles . In conclusion, i could be a gued ha he S.A .J.Bus.Mgm .1990,21(1) Table 2 Mon hly e u ns o all-sha e, mining, and indus ial indices Equally weigh ed indices All-sha e Mon hs .R% (.R) u(R)% .R% Janua y 2,45 1,09 7,45 1,36 Feb ua y 2,12 1,52 4,64 2,34 Ma ch 0,41 0,22 6,22 0,65 Ap il 0,78 0,76 3,43 0,52 May 2,19 1,41 5,13 1,86 June 0,68 0,32 7,03 0,70 July 2,89 •2,02 4,74 5,99 Augus 2,57 1,21 6,88 2,59 Sep embe 1,39 0,67 6,85 3,99 Oc obe 1,25 1,34 3,10 -0,70 No embe -0,36 --0,26 4,66 -1, 15 Decembe 2,78 2,96 3,11 4,56 • signi ican a he 5% le el o a one-sided al e na i e h signi ican a he 2,5% le el o a one-sided al e na i e signi ican seasonal e ec in Decembe is mo e likely o be a esul o ela i ely less ola ili y han subs an ial e u n in Decembe . No es 1. The a e age weekly e u ns in each mon h we e con e ed o mon hly e u ns o ease o compa ison wi h he in e na ional e idence. 2. Sou ce: Co hay, Hawawini & Michel (1987). 3. Ba & B ad ield (1988) o e a de ailed explana ion o why he a iance o hinly aded s ocks a e usually unde es ima ed. 4. The one-sided al e na i e alues a e used o be consis en wi h he Co hay e al. e idence. Re e ences Ba , G.D.I. & B ad ield, D.J. (1988). Po olio selec ion in hinly aded en i onmen s - a case s udy. Manage. Decision Econ., ol. 9, 287-290. B ad ield, D.J., A leck-G a es, J.F. & Ba , G.D.I. (1988). Asse p icing in small ma ke s - The Sou h A ican case. S. A . J. Bus. Mgm ., ol. 19, 11-21. Mining Indus ial (.R) u(R)% .R% (.R) u(R)% 0,38 12,00 2,83 1,44 6,54 0,88 8,84 2,10 1,67 4,16 0,24 9,26 0,39 0,23 5,72 0,28 6,09 0,87 0,90 3,20 0,91 6,74 2,28 1,55 4,90 0,22 10,40 0,71 0,38 6,23 >z,84 7,00 1,88 1,22 5,ll 0,66 13,02 2,47 1,43 5,72 0,99 13,38 0,52 0,31 5,64 --0,35 6,62 l,n 01,87 3,15 -0,56 6,84 --0,23 -0,16 4,80 >z,28 7,23 2,01 >z,42 2,77 B ad ield, D.J. (1989). Applica ions and ex ensions o inancial models o small ma ke s: The Sou h A ican case. Unpublished Ph.D. hesis, Uni e si y o Cape Town. Co hay, A., Hawawini, G. & Michel, P. (1987). Seasonali y in he isk- e u n ela ionship. Some empi ical e idence. J. Fin., ol. 42, 49-68. Hamon, J. (1986). The easonal cha ac e o mon hly e u ns on he Pa is Bou se (in F ench). Finance, ol. 7, 57-74. Keim, D.B. (1983). Size- ela ed anomalies and s ock e u n seasonali y: Fu he empi ical e idence. J. Fin. Econ., 12, 13-32. Keim, D.B. (1986). The CAPM and equi y e u n egula i ies. Fin. Anal. J., ol. 42, 19-34. Roll, R. (1983). Was is das? The u n o he yea e ec and he p emium o small i ms. J. Po olio Manage., ol. 9, 18-28. Roze , M. & Kinney, W. (1976). Capi al ma ke seasonali y: The case o s ock e u ns. J. Fin. Econ., ol. 3, 379-402. 9