Economy & Finance

Report Team

Description
Report Team
Published
of 4
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
   Ball and Shivakumar, Earnings quality at initial public offerings, 2008 Ball and Shivakumar (2008) research earnings management around IPO’s. Results o the research o Ball and Shivakumar (2008) are that the discretionar! accruals o IPO irms in !ear "2 and in !ear "# did not signiicant dier rom non"IPO irms. $videncealread! commented %eore in &aragra&h '.' is that the average income o IPO irms in the &ros&ectus as not signiicantl! restated and the &arameters o the regression hich indicate theearnings inlation ere not signiicant. *onse+uentl! is concluded that IPO irms do noto&&ortunisticall! inlate their earnings. ,urthermore the! communicated great dou%t a%out the conclusions %! -eoh et al. (#8a)that managers to inluence share &ricing o&&ortunisticall! inlate their earnings. /hile themonitoring on &u%lic com&anies is much higher than on &rivate com&anies %! Ball andShivakumar it is im&lausi%le that managers ould have the o&&ortunit! to inlate their earningson such large scale. -he! re&licate the research o -eoh et al. (#8a) hoever this time ith adierent model the corrected modiies ones model %ut ith the same 1S data. -o measure thediscretionar! accruals the corrected modiied ones model determines accruals %! etracting datain the cash lo statement hile -eoh et al. (#8a) used data in the %alance sheet. -here&lication o the research o -eoh et al. (#8a) did not ound evidence o s!stematic use o earnings inlation. -his outcome contri%utes to the develo&ment o the theor! o the im&act o management on the IPO’s over"&ricing and o&ens a ne ield or research and de%ate.Ball and Shivakumar (2008) eamine the h!&othesis o artiicial earnings inlation driven %! the incentive to inluence IPO &rices. -he authors give another reason to the increase indiscretionar! accruals surrounding IPOs3 the irms 4use o their IPO &roceeds to make changesto their orking ca&ital. Prior to going &u%lic &rivate irms ma! ace"under investment inreceiva%les and inventor! due to resource constraints hich disa&&ear ater going &u%lic. -husan increase in discretionar! accruals ma! re&resent the relie o resource constraints. -he argument o Ball and Shivakumar (2008) ails to e&lain the negative relationshi& %eteen earnings management and &ost"issue stock &erormance documented %! -eoh et al.(#8). I an increase in discretionar! accruals is caused %! inancial constraints then it is note&ected to ind earnings management ith strong negative im&acts on the su%se+uent &erormance o the stocks olloing such oerings.  -o the contrar! Ball and Shivakumar (2008) 5hereater BS6 suggest that as irms gothrough the IPO &rocess the! transition rom a &eriod o o&acit! to a &eriod o increasedscrutin! regarding earnings. -hus the! have incentives to re&ort conservativel!. 1sing a s&ecialsam&le rom 1.7. IPO irms the! ind su&&ort or their h!&othesis. ,urthermore the! +uestionthe results o -// and theoreticall! sho that *90 is a %iased measure o &re"IPO earningsmanagement.:iven this ongoing de%ate regarding the earnings +ualit! o IPO irms this stud! eaminesa sam&le o ;<2 IPOs rom 28 countries attem&ting to anser the olloing +uestions.#) o irms engage in earnings management &rior to an IPO and i so to hat degree=2) Is *90 a %iased measure o &re"IPO earnings management=') oes the degree o &re"IPO earnings management aect the irm’s long"term &erormance=9nal!>ing the earnings +ualit! o IPO irms in an international setting is useul in severala!s. ,irst it allos us to calculate *9"# or a large sam&le o IPO irms. -he use o *9"#as a &ro! or &re"IPO earnings management is eem&t rom most o the criticism o BS (?ote 2)and it allos us to %etter understand the degree o &re"IPO earnings management. Second thecom&arison %eteen *9"# and *90 in an international setting allos us to investigate theseemingl! contradictor! claims o -// and BS using an out"o"sam&le test. -hird %! anal!>ingthe degree o earnings management in several countries e can %e more conident that theresults are not driven %! characteristics that are s&eciic to a single countr!.-he results sho that the degree o earnings management varies %! countr!. In mostcountries neither the median nor the mean value o *9"# is signiicantl! dierent rom 0.,urthermore the countries hich re&ort *9"# as signiicantl! dierent rom 0 are evenl!divided %eteen those hich re&ort signiicantl! negative and signiicantl! &ositive *9"#. -hisresult is consistent ith the h!&othesis o BS and shos that on average IPO irms don’t engagein earnings management. @oever hen *90 is used as a &ro! or earnings management theresults sho su&&ort or the earnings management h!&othesis3 not onl! do most countries(ece&t ,inland) sho a &ositive *90 on average %ut a%out hal o them sho signiicantl! &ositive *90. ,urthermore in almost ever! countr! (ece&t Belgium enmark and ,inland)*90 is larger than *9"#. Statistical tests sho that the mean (median) *90 is signiicantl!greater than the mean (median) *9"# in #0 (#2) out o 28 countries. -his result su&&orts thecriticism o BS and shos that the use o *90 eaggerates the degree o &re"IPO earningsmanagement. Interestingl! :reat Britain rom hich the sam&le o BS is taken shos negative   &re"IPO a%normal accruals (consistent ith their vie that IPO irms have a disincentive tomanage earnings) %ut signiicantl! &ositive *90 (consistent ith their criticism). ?evertheless these results are also consistent ith the results o -// in that most irms re&ortsigniicantl! &ositive *90.-o test hether a%normal accruals aect the long"run &erormance o IPO irms Ball andShivakumar (2008) use to methods. ,irst olloing -// the! divide irms into +uartiles %ased on their level o *9 and com&are the long"term &erormance o each +uartile. Secondthe! regress IPO long"term a%normal returns on *9 several control varia%les that hel& &redictuture &erormance and countr! dummies to control or countr! ied eects. *onsistent ith-// the! ind that &re"IPO earnings management hel&s e&lain the long"termunder&erormance o IPO irms. ?ot onl! do the IPO irms in the highest +uartile o *9under&erorm those in the loest +uartile %ut the coeicient or the *9 varia%le is alsosigniicantl! negative in the regression models. -hese results are ro%ust to several methods o measuring under&erormance (*9R B@9R and 9l&ha) over several time hori>ons (' and A!ears) to the to measures o earnings management (*9"# or *90) and hold even ater controlling or several irm" and countr!"s&eciic characteristics that could e&lain IPOunder&erormance. -his result su&&orts the earnings management h!&othesis o -//.Overall the! inter&ret the results o the stud! as consistent ith %oth seemingl!contradictor! h!&otheses o BS and -//. *onsistent ith BS on average IPO irms do notengage in &re"IPO earnings management and *90 is a %iased measure o &re"IPO earningsmanagement. @oever consistent ith -// some irms do engage in &re"IPO earningsmanagement. -hese irms are a%le to deceive investors and receive a higher IPO &rice thusunder&erorm over the long run. 9lthough the use o *90 is a %iased measure o &re"IPOearnings management the consistenc! o the results across %oth measures o earningsmanagement (*9"# and *90) in e&laining IPO under&erormance shos that *90 ma! %e used ith some caution as a &ro! or earnings management. Perha&s uture studies coulddevelo& methods to immuni>e *90 rom the criticism o BS.*riti+ue anal!sis o the &a&er alloed us to &oint out several uncertainties o Ball andShivakumar’s research.#) 9uthors use &re"IPO data o &rivate irms. -his sam&le includes re&orts hich asreleased in !ears %eore IPO. Such methodolog! ma! have disadvantages as irm start &re&aretheir &ros&ects according to &u%lic re+uirements %eore IPO !ear. -hereore dierence in re&ortsma! %e not signiicant. Researchers tr! to diminish this &ro%lem using several sam&les ("# !ear"2 !ear and "' !ear). -his methodolog! reduces %ut not eliminates such &ro%lem. In addition  data o "' !ear have interesting statistics. ,or eam&le mean o ?et Income included in &ros&ectus is negative. Such statistics attract attention and raises some +uestions.2) Researchers stud!ing international variation in market and regulator! regimesencounters a thicket o &otentiall! correlated omitted varia%les. -he omitted varia%les &ro%lem issomehat mitigated ith IPOs %ecause it is the same irm in the same industr! that changes to &u%lic status. 1sing international data have on cons as dierent countries have dierentre+uirements or re&orts. -hus it ill %e slightl! diicult to ind all signiicant ommitedvaria%les. 9lso authors use sam&le %eteen #2 and #. It include several !ears and re&ortstandards could changed through this time &eriod.') Coreover sam&le o stud! include irms rom dierent industries. -his method ma! risediiculties ith estimation as dierent industries have on s&eciic and thereore somes&eciic charecteristics in &ros&ectus hich is demanded rom the &u%lic.;) In addition reaserchers tr! to estimate re&utation eects cost o ca&ital eects andmonitoring %! internal and eternal auditors %oards anal!sts rating agencies the &resslitigators and other &arties. -his is slightl! diicult and thus the! ma! miss some varia%leshich is higl! correlated ith such eects.
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks