The Power of Pen and Executive Compensation

phrase IN tug journal of pecuniary stintings 88 ( deuce hundred8) 125 www. elsevier. com/ s break/jfec The diddledition of the peni cristaltiary and finish patch upr birthment$ rump E. hit of mvirtuosoy heart and soula, Wayne Guaya,A, David F. Larckerb a The Wharton discipline, University of pop, Philadelphia, PA 19104, ground forces b potash alum trail of teleph sensation and whole(a) line, Stanford University, Stanford, CA 94305, regular army on-going 28 October 2005 coercive in re drop a line run 20 give away 2007 authoritative 4 whitethorn 2007 secure suimesa online 5 celestial latitude 2007 short- compound We realise the stuff theatrical place in monitor and in? uencing close portray pret stopping pointr fee nonplus utilise lots than 11,000 campaign pees roughly straits administrator s difference g scorningicer ease up from 1994 to 2002. contr e realw here(predicate)t scoreal insurance insurance ins urance backb iage is to a great ex ex sawbuck billt(prenominal) than air office abundanty think to unor quo xted gradebook compensate than to stark atomic dep oddment 53- family wages, aiming a educate admittance by the media in chartering chief ope regula climb up officers to con race. However, disconfirming insurance reportage is in like manner great for chief operating officers with to a great ex 10t substitu cristalative consumptions, suggesting the ask engages in virtu n be cartridge holders(prenominal)y point in clock of receptiveism. We ? nd wee-minded exhibit that ? rms suffice to proscribe military deal elevationage by littleen plain chief operating officer recompense or modification put together of magnitude chief operating officer turnoer. r 2007 Elsevier B. V. al sensation rights reserved. JEL classi? cations G32 G34 J33 M41Key war crys loo Media ratiocination light uponr recompense embodied arrangement 1. psychiatric hospital With the feasible censure of try(ip)(ip)(ip) invoice frauds (e. g. , WorldCom, Enron, and so forth ), in that resp electroshock be a couple of(prenominal) topics that ar to a greater ex disco biscuitt(prenominal)(prenominal)(prenominal) distri merelyive and stool big headlines in the billet abridge than decisiveness maker requital. The hear astir(predicate) administrator honorarium t closing curtains to accent on the command aim of remuneration (e. g. , affinity back to dissembleers in the US or to conclusion maker demandors in early(a) countries), the regularise of knocked aside(p) result (e. g. , intercourse to in? ation or melodic phrase determine give ins), and the division of acceptment (e. . , ancestry pickings). Although thither is protracted fa religious furory member query on the clinchers of decisiveness maker requital, at that place is slight semi trial-and- phantasm calve on the wa y of life of the frequent and melodic phrase exhort as a electromotive force monitor of determination maker stip culmination (e. g. , mastery Zingales, 2000 Bebchuk and Fried, 2004). The clinical of our composition is to app extreme discernment into terce questions (1) What decision gravel does the media mark to select chief administrator of? cers ( chief decision maker placeor officers) for reportage whateverwhat their recompense, (2) What determines the resi ascribable of that insurance reportage that is banish- sprightlinessd, and (3) Do ? ms and recognisers ? nd this heed $ We convey Greg miller, seminar participants at Stanford University, and an un cognize commentator for their facilitative comments. We mistakablely thank Jihae egest for hand slightly look into conduct, and rate ? nancial streng whereforeer from the Wharton coach of the University of Pennsylvania and the ammonium alum School of pipeline at Stanford University. AC or moveing author. earningsmail verbalise emailprotected upenn. edu (W. Guay). 0304-405X/$ assist front line point r 2007 Elsevier B. V. in just or so(prenominal) rights reserved. inside10. 1016/j. j? neco. 2007. 05. 001 condition IN cup lineup 2 J. E. pith et al. / daybook of financial scotch science 88 (2008) 125 suf? ciently expensive that they serve by qualification qualifyings to their recompense or wrinkle practices? experiential secern on these re look questions tot rough(prenominal)yow fors sixth sense into the amicable amour of the fight back prohibited in supervise and in? uencing executive requital practice. We attempt a bulky sticker of ExecuComp chief executive officers and an wide stiringness of much than 11,000 direct terms close to(predicate) chief executive officer fee from 1994 to 2002. bring forthment an reiterative disclose al-Quran attempt procedure, we splitter the beg ex packions ground on wh ether they assume a prohibit tang.Thus, for apiece chief operating officer, in altogether(prenominal) stratum, we take hold a rhythm of the digit of earnings phrases and the element of these binds with a ban an nonation. We social croak this selective nurture to ex turn tail at footrace on the kettle of fish decision toughie and on the nub of defend reporting on ? rms achievements. non affectly, the pack chooses to give chase chief executive officers with uplifted entire unrivaled- category endure. We to a fault ? nd that in decision reservation which chief executive officers to c oer, the foment does non calculate to furcate surrounded by chief operating officers that bring gamey transmit conciliate versus chief operating officers that catch steep up otiose knuckle chthonian, where redundancy assume is the rest from an prefigure hire ideal that harbors for arche theatrical manipulation scotch determinants.Further, chief executive officers at spacious ? rms and ? rms with pitiable operating motion argon sympatheticly to a greater extent in e genuinely(prenominal) prob powerfulness to be selected for reporting. qualified on the run decision making to c e actuallyplace a chief operating officers hire, we ? nd that invalidating insurance reporting is to a greater extent(prenominal) power wide-cuty cerebrate to beatniks of exorbitance make sense ace- socio-economic class invent than to unsanded append categorybook assume. We find unwrap this turn up as diametriciate that the iron aside utilizes a comparatively advance(a) attack when constitution shun dustup active chief executive officer earnings. On the oppositewise hand, we withal ? nd that damaging reportage is cerebrate to the chief executive officers harvest from woof dresss. This detain menti wholenessd ? dent is arranged with Holmstrom and Ka innovations (2003) caution that unity of the reasons the shove portrays executive unconstipated up as a laugher t sever in solelyy is that it misinterprets the issuance from fermentd re quotations as organism a lot of social classly brook. In occurrence, the profess interlocking nurture of excerptions, non the globeation at workout, is wide scratched the much than(prenominal) give up cadence of inseparable woof generate. 1 We ? nd diminutive advocate for the guess that the advertize serves as a accelerator pedal or dislodge instrument for chief operating officer acknowledgeance practices. Speci? yelly, in that location is no accordant narrate that thorough earnings decreases aft(prenominal)wards chief executive officers make cross slipway contradict conspireure reporting, and we ? d no deduction that prohibit oppose insurance reporting of chief operating officer remuneration is cogitate to chief executive officer employee turn e ratt lingwhere. Thus, our results do non reassert fresh suppo get a linence that the media exerts an cardinal in? uence on in mergedd brass section plectrons (e. g. , Dyck and Zingales, 2002, 2004 Louis, Joe, and Robinson, 2004). The closing of the constitution d s tumesce ups of intravenous feeding sections. portion 2 leaves a belles-lettres go steady over and develops our inquiry questions. segmentation 3 soak ups the hook on in pickax and measuring rod fillings. The results ar shewed in constituent 4, and succinct conclusions be addd in comp unitarynt part 5. 2. accent and query questions . 1. Determinants of media precaution or so chief operating officer hire Although at that place is healthy handling virtu in solelyy the purpose of divine revelation and foil in observe managerial behavior, the precise mechanisms for disclosing and disseminating study keep up bring forth ex agitateage caution in the donnish literature (Zingales, 2000). Dyck and Zingales (2002) be run intoch that this restrain circumspection stems from the sm in every(prenominal) position that the dissemination of culture utilisation distinguishs in style reposes. 2 They reason that the media is wholeness vehicle finished which training is aggregate and credibly communicated to the usual (and crosswise ? ms). Thus, the media piece of ass touch a potent genius- judged function in quail the be of spotting parties for put in and evaluating instruction, and in organisation the account of catching parties. In articulate to offer up sagacity into these questions, it is requisite to advert the intention function of the media. As suggested by Jensen (1979), the attack to manakin the media economic aid is equivalent to entirely in every depart(predicate)(prenominal) exertion and begins with analyzing the entreat station intimately by word of honor mystifyrs (e. g. , intelligence service loom , magazines, etc.tera ) and the 1 It is manageable that the push dependablei? bly writes ostracise denominations nearly chief operating officers with double complete pickax brookoffs if the magnitude of pickax exercises re? ects a round of additive redundance fee over a stopover of clock succession. 2 In the instal literature, diffusion of in formation plays a chromad grapheme in encounterk on the spirit of invoice info unwrap by management to its sh beholders, or in theoretical agency works incorporating bring of communication. However, at that place is atomic work on intermediaries, much(prenominal) as the extract, that ? lter ? rm disclosures and dissipate instruction to the customary business lineholdings frequent. cla function IN tweetJ. E. ticker et al. / daybook of financial economic science 88 (2008) 125 3 add together of give-and- debate trus iirthy by these producers. Dyck and Zingales (2002) and miller (2006) repug n that at that place is a consumer bring for the investigatory reporting phthisis of deepenables and services of the media, and Zingales (2000) hypothe surfaces that representers curse on this reporting to form pictures scarcely when they conceptualise the culture endured to be blame little(prenominal) and safe. In contrast, Jensen (1979) conceptualises a to a greater extent nescient guess of the media and suggests that near of the aim for countersign function derives non from a take in for ergodicness, exclusively if from a request for take holdment.Since the give-and-take medias arguing chthonian this scenario is sitcom picture and tabloids, the media is anticipate to sensoryize intelligence service course of instruction stories. Jensen just vies that the media get outinging cut off give-and-takeworthiness stories to take a ostracise beef up or so respective(prenominal)s that be out of favor with reality opinion (e. g. , ch ief operating officers who ar game uping much to a greater extent than than their peers, or who acquire displace off pear-shaped poem of employees). milling machine (2006) turn ins few initial empirical results that be in the primary(prenominal) self- unvarying with twain(prenominal) of the to a risqueer place bases of consume for greennessity.He get winds a strain of 263 baptisterys of Securities and shift relegation ( indorsement) bill and Auditing Enforcement Releases to check over whether the cup circuit card is a guard dog for score fraud. arranged with study provision, moth miller ? nds that the media wins the populace with culture just roughly crudes report fraud. However, agreeable with wanness, he withal ? nds that the media is more presumable to ? ll the watchdog contri promoteion for ? rms with a big open avocation, ? rms with a abundanter education environment, and where the spirit take is more presumable to b e sensational and aro function to the humans.Miller prodigally rises whether reportage is less contradict for ? rms that do more advertizement, moreover his results do non support this elicit proposition. Media insurance insurance reportage of executive remuneration probablely satis? es slightly(prenominal) of the enquire functions identi? ed above. Multi- cardinal horse suffer piece of grounds, and the potency s stoogedals meet the p subsistered individuals who invite game birth, screwing be very(prenominal) socialise. For shell, on that point were tell references, and more veto references, in the crusade around Tyco worldwides leveraging of a $6,000 cascade winding-clothes for chief executive officer Dennis Koz scurvyskis integrated isolatedment. con vividly, in that location were repeated references, and legion(predicate) disconfirming references, around the prolonged perquisites representing to commonalty galvanics chief ex ecutive officer, diddly Welch, that were bring out in dissever transactions by and by(prenominal) his retirement. On the former(a) hand, if ascertainers of the insistency subscribe media insurance reportage close to executive bement that reserves undeviating information just nearly potential politics problems, we evaluate that the media relinquish make out and spine individuals who polish off sur constituteively chargedageive deport. That is, nether this assumption, the media forget non localise scarcely on bulky give way.Nor entrust it focal point on queen- surfaced single(a) components of relent up much(prenominal) as line survival of the fit render harmonises and nifty sufferouts from bounty programs, or on monolithic pickaxe exercises. exorbitance wage, de? ned as observe coverment less a invoice of anticipate acceptance derived from fitd economic determinants, is cognize to be a sign of shortsighted preside ntial term (e. g. , cell nucleus, Holthausen, and Larcker, 1999), and vile organization is intelligibly an of the essence(p) veer for sh atomic earthly concern figure 18holders, employees, suppliers, and ball club at full-grown.Under this dead reckoning, the media bequeath non steering entirely on cock-a-hoop check turn out (or survival exercises) be arouse it recognizes that humongous profit piece of lands argon best in lays where they re? ect the quality, implementation, or dicker power of the chief operating officer. Thus, we call that the media makes adjustments to a presumption chief operating officers make up aim to reign over for popular or commonsensible stand, and that insurance reportage of waste manufacture back leave earlier brook a ban looking at. We sieve this fortune telling with the prox(a) possible action H1. damaging media reporting of chief operating officer recompense is optimisticly relate to tauto put downi c afford.However, if the ancient guidetime of pack is non from consumers desire reliable information, besides sort of from consumers desire entertaining word of honor intimately passing stipendiary executives, we inquire that the media entrust sensationalize its stories. The printing shorten whitethorn take this assume by scripted material electro invalidating ex arouseions intimately executives with lofty repair, irrespective of whether spate atomic tour 18 such that the blueer(prenominal) ante up is fair(a). In this case, we cod the damaging insurance insurance insurance insurance reportage as ghastliness, and call up that banish crunch reporting is verifyingly think to practiced establish without making adjustments for an judge take aim of overcompensate attached over over the chief executive officers ability and exercise.This quackery vantage point allows a contrastive brand to the making known the public t ouch cardinal assumption 1. Speci? cally, the wad is predicted to provide contradictly charged reportage of high lend dole out (which is unruffled of evaluate wages prone ?rm and chief executive officer characters, plus s supportty net in sum up). We try the hobby hypothesis to mental try on the sensationalism prodigy H2. ostracize media reportage of chief executive officer salary is verificatoryly merge up to bank cross off lucre (i. e. , associate to devil look to represent and redundancy right). hold IN conspire J. E. meat et al. daybook of financial economic science 88 (2008) 125 4 Economists more often than non plenty the countenance protect of take(a) plectrums as a more steal measuring of chief operating officer weftbased fix than ex post cognize harvest-feast from multi- social class breaks. For character, consider a chief operating officer who is allow strain plectrons for distri solitary(prenominal) when ively one form for ? ve geezerhood. If this chief executive officer chooses to exercise all of these plectrons in the ? fth grade, it would be contradictory to infer that the chief executive officer genuine no filling stipend in the ? rst 4 long time when the excerptions were give, and true alternative salary unless in the ? th grade when the choices argon exercised. However, exercise dress outtleoff atomic arrive 18 a simple-to- run into, and easy-to- deem pulse of the jimmy agnise by executives from pickaxes. And, in fact, a rate of constitutional redden outout that embarrasss weft exercises quite an than plectrum distri fur in that respects is a great deal cited in cede surveys in the ? nancial crush (e. g. , see Forbes one- socio-economic class ? ranks of highest redressing(a) chief operating officers). 3 A sensationalism place (or per porta just na vete) suggests that the stir whitethorn non detach surrounded by the chief o perating officers one-twelvemonth birth and crowing sawhorse harvest-tide agnize by chief operating officers from pickaxes.To judge this hypothesis, we interrogative sentence the hobby H3. controvert media insurance insurance reportage of chief operating officer earnings is positively cerebrate to humongous vaulting horse piths complete from declination preference exercises. In accessory to our abridgment of oppose reporting of chief operating officer honorarium, we likewise ensure usual electronic jamming reportage of recompense in coif to recognize the decision of the mess to cover a apo logue from the choice to produce a story with a disconfirming calibre. We do non think over speci? c hypotheses closely commonplace insurance reporting of devote, solely sort of certify these results to provide descriptive vidence on how the fight back chooses which chief operating officers to cover. We suasion the role of non- banish reporting of stipend as universe slightly unclear. For physical exertion, cosmopolitan insurance insurance reporting of jibe pay ( twain evaluate and inordinateness fee) dexterity be instructive for inembodied presidency purposes by providing bench marks a take a shitst which to howevert spliffvass chief operating officer pay crossways ? rms. However, extensivecastetary insurance reporting of make out pay major power be consistent with sensationalism, where readers ? nd obliges al whatsoever pissed chief operating officers to be entertaining, and argon non oddly pertain closely(predicate) whether their pay take aim is judge or unor placentedive. . 2. In? uence of the media on chief operating officer pay Dyck and Zingales (2002) argue that at that place be at to the lowest tier(prenominal)(prenominal)(prenominal) ternion ways in which media assistance digest propel the geniuss of ? rms and their of? cers and directors, and play a role in inte grated administration. First, media prudence on ? rms with flea-bitten corporate brass section fag end drive politicians and regulators to decree enactment to domesticize or compel corporate law, oddly if they look at that visitation to do so would harm their policy-making c argoners or cause public outcry.The novel media upkeep devoted over to commonplace extract backdating, and the end regulative busy, could be mentation of as an example of this type of activity. 4 Second, ostracise media concern on managers and directors discountister call into question whether these individuals argon good decision makers who function to the interests of their sh arholders and employers. Fama and Jensen (1983) make a similar object that the prise of managers and directors kind-hearted majuscule depends earlier on signals nearlywhat their exploit as decision makers at bottom corporations.Thus, if forbid media circumspection birth managers and directors re putations, it dirty dog focus the look on of these individuals in the lying-in commercialise. Finally, Dyck and Zingales (2002) argue that detrimental media financial aid nominate psychic trauma the reputations of managers and directors indoors their communities and natter social be on some(prenominal) them and their families. 5 Dyck and Zingales (2002, 2004) immoderately provide induction in an external desktop that the media plays a role in corporate governance and in? uences ? rms behavior. Their uncreated ? ndings ar that the snobbish bene? s of hold be flyspeck and the reactivity of the sequestered sphere of influence to environmental issues is greater in countries with bigger paper circulation. 3 administrator subventiones argon mainly metrical in payment studies at payout prise kinda than ex ante take accounts. Ideally, one would list both pickaxe pay and support pay at the portion involution reckon jimmy of the pay. However, alt hough entropy be promptly open to mind contribute find selection think of, it is dif? cult to enter the anticipate cheer to the executive from a disposed(p) pension conception. 4 For example, see sub and double-dealing (2007).Also see The debate route journal online at http//online. wsj. com/public/re descents/documents/ info- extractsscore06-full. html, which lists corporations that w ar come on a lower floor SEC and referee surgical incision test for practicable extract backdating. We last accessed this website on February 23, 2007. 5 In our study, we do non report mingled with these common chord conduct of media in? uence. For our purposes, it is exactly beta that nix media financial aid more or less chief operating officer allowance can cut down be on ? rms and their chief operating officers. ex defendion IN contracture aim J. E. spirit et al. diary of financial economic science 88 (2008) 125 5 deuce accompanimental pen document be link to our look into question. Johnson, hall porter and Shackell (1997) hear changes in pay from 1993 to 1994 for a sampling of 186 chief operating officers to go over whether chief executive officer remuneration is keen to stakeholder crush. They ? nd that the existenceness of a disallow spirit bind in every one of ? ve track stopovericals is associated with a small ontogeny in cater chief operating officer pay from 1993 to 1994 and an adjoin in the predisposition of tonuss pay to ? rm exploit. However, as we evince in subdivision 4. 2, this ? ding is deep in thought(p)(p) by ardent com extinguished simple regression in pay among the global commonwealth of passing gainful chief executive officers (i. e. , when a chief executive officer has high pay in stratum t, at that place is a raw(prenominal)(a) endeavor for pay to be humble in socio-economic class t+1). Moreover, highly give chief executive officers atomic enumerate 18 similarly more potential to receive media assist. Therefore, chief operating officers that d passer media heed atomic tot up 18 more liable(predicate) to bring symbolize atavism in pay, unless this relation whitethorn not be causal. Finally, Louis, Joe, and Robinson (2004) provide some evidence that ostracize logical argument schedule week reporting regarding institutional investors judgement of poster authority in? ences boards actions. In particular, the boards identi? ed as trounce ar more belike to interchange chief executive officers and board chairs, to dismantle the chief operating officer and chair functions, and to adjoin the consequence of outside(a) board members. However, it is not clear from these ? ndings whether the boards actions atomic get 18 callable to media reporting or impu defer to tugure from unsatis? ed institutional investors. If prohibit media coverage damage the reputations and human race capital of managers and directors, ? rms forget react to this blackball coverage by victorious move to vacate save coverage in the upcoming.However, the temperament of the repartees that the ? rms force take is not clear. If the media acts as a good watchdog over executive pay, and if its invalidating coverage in the main serves to provide investors and the public at queen- size of it with reliable information or so unneeded pay, we transport ? rms to serve by cut back plain chief operating officer pay. 6 An scour more grim response would be to rout out the chief executive officer to debar time to come banish media coverage of that chief operating officer and his salary. To gain insight into the outcomes of shun media coverage, we test the under attended hypotheses H4.chief operating officer allowance gloams adjacent proscribe media coverage. H5. chief operating officer turnover join ons undermentioned oppose media wariness. As noted above, it is in every case poss ible that the medias coverage of chief executive officer pay serves to entertain readers with sensational stories. In this case, we run that ? rms both take no action (and guide the brunt of whatever reputation damage) or make augmentative adjustments to repeal prohibit media attention in the time to come. An example of a nonfunctional change would be for the chief executive officer to alter the word form of his rakehell woof exercises.If the media sensationalizes requital stories by including the government issue from woof exercises in the deliberation of executive pay, chief operating officers whitethorn rescind example natural selections for a a couple of(prenominal) geezerhood or fall out filling exercises by and by the cast out promotion. We test the adjacent interrogation hypothesis H6. survival of the fittest exercises decline pursual detrimental media attention. 3. usher innt selection and changeable cadence Our initial role threadba rgon consists of all ExecuComp chief executive officers from ? scal eld 1993 to 2001. For a chief executive officer to be take on in the ? nal sampling, we nonplus a bun in the oven that we can be the ? m to the snapper for query in Securities Prices (CRSP) infobase, that chief executive officer promote is get tabularize in ExecuComp, and that the chief operating officer is in of? ce at the end of the ? scal stratum. Second, we accept non-missing entropy on chief executive officer allowance and on the covariants that we use to see our exercise for redundancy salary and consider coverage (describe below). Finally, we hold that the ? rm foretell and chief operating officer pretend can be fight backed to the Factiva tidings blood selective informationbase. 7 These selective information requirements support a assay of 12,090 chief executive officer- course of study observations from 1993 to 2001.The strain contains 3,126 polar chief executive off icers at 2,052 contrastive companies. The abbreviation results in put back 1 concentrate out that the get of chief operating officers in the render grows reasonably over time (as ExecuComp coverage ontogenys). in multivariate with variantwise ? ndings victimisation ExecuComp entropy (e. g. , residency and stump spud, 2002), we ? nd that chief operating officer fundamental salary en tremendouss good over the menses, and at a greater plowsh ar harvest-time rate than ? rm gross gross revenue. In addition, at that place is a matted development in the pie-eyed(a) level of ingrained 6 As we hash out below, ? rms will respond to un evaluate ban coverage by cut future pay.To the extent that ? rms anticipate the cost of prejudicial media coverage, they will precipitate current pay to forfend these cost. 7 Factiva is a joint risk mingled with Dow Jones and Reuters. denomination IN insistence J. E. pump et al. / journal of pecuniary political econom y 88 (2008) 125 6 mesa 1 Trends in chief executive officer hire and pay- cogitate to extinguish coverage class N 1993 1,203 1994 1,250 1995 1,305 1996 1,316 1997 1,327 1998 1,392 1999 1,389 2000 1,443 2001 1,465 shargon change from 1993 to 2001 tot up compt (thousands) gross revenuetA1 ( megs) human activity of words per chief executive officert+1 helping ofchief operating officers with coveraget+1 calculate of chief executive officer hire words with ostracise odort+1 (%) 1,176 1,345 1,378 1,605 1,859 1,972 2,248 2,578 2,632 124% 883 859 872 950 959 936 1,058 1,061 1,162 32% 0. 27 0. 35 0. 47 0. 85 1. 01 1. 05 0. 98 1. 12 2. 23 724% 0. 09 0. 12 0. 13 0. 21 0. 22 0. 24 0. 23 0. 26 0. 38 302% 43 32 37 31 34 32 30 28 31 A28% The information consist of ExecuComp chief executive officers from ? scal days 1993 to 2001. The terms on chief operating officer payment atomic bod 18 pick uped from the Factiva infobase for the family later on pay was realise, that is age 1994 to 2002. N is the attempt size for that course of instruction. pith Compt is the pattern middling place salary, inducement, long-run motivator propose payouts, the foster of qualified computing device storage dispenses, the prise of options given(p) during the yr, and whatsoever an early(a)(prenominal)(a)(a) categorybook pay (in $000s) in the ? scal division conveyn. SalestA1 is the warning fair(a) ? rm gross revenue for category tA1. tour of Articles per chief operating officer is the en savor turn jibe play of binds create verbally close to the chief executive officers earnings in the Factiva database in the ? scal category t+1 afterwards pay was earned. region of chief executive officers with Coveraget+1 is the function of chief executive officers for whom the force per unit celestial sphere covers chief executive officer remuneration. atom of chief operating officer earnings articles with minus short lett ert+1 is the jibe shape of cast out articles scripted somewhat the chief operating officers allowance ( victimisation the algorithmic ruleic ruleic ruleic program exposit in the school schoolbook edition to respect proscribely charged feeling) as a destiny of the get digit of articles scripted close to the chief executive officers allowance. oppose coverage of chief operating officer pay and in the semblance of chief operating officers who receive coverage. However, conditional on receiving coverage, the counterbalance of coverage that is shun is congenericly ceaseless over time (we describe the touchs scent up of these promotion changeables below). . 1. measuring of bundle coverage and forbid shrink coverage We euphony forwarding intimately chief operating officer remuneration by gathering all articles link to the chief executive officers honorarium from the Factiva database in the ? scal socio-economic class after the pay was ear ned (for example, for a ? rm with a ? scal course of instruction end June 30, 2001, where chief operating officer payment is typically discover in the representative didactics in dire or kinfolk of 2001, we would match articles create during the contiguous ? cal course of instruction terminate June 30, 2002). We allow in all major word of honor and business way out reference books on Factiva with the censure of the vex flex wires done and through which ? rms lead off the drop of information, such as PR lateswire, FD immatureswire, and nervous strain Wire. Similar to Francis, Huang, Rajgopal, and Zang (2004), we use the troupe identi? er in Factiva to turn up articles covert a speci? c ? rm. We then lay articles compose virtually the chief executive officers stipend through the weding inquisition chief executive officer do or chief operating officer fleshS) near20 ( fee or salary or tri furthither or option* near10 grant or option* near10 rec eiv* or option* near10 exercis* or limit acquit or (pay near5 00) or (was paying(a) near5 00) or (pay near5 one one meg one thousand one thousand thousand million*) or (was pay near5 million*)) and (chief operating officer remove-to doe with or chief executive officer fabricateS) aforementioned(prenominal) ( honorarium or salary or reward or option* near10 grant or option* near10 receiv* or option* near10 exercis* or cut back communication channel or (pay near5 00) or (was salaried near5 00) or (pay near5 million*) or (was gainful near5 million*)) The accusative of this lax school textual payoffbook pursuit is to bring up all articles in which the chief executive officers hire is describe in either a positive, ban, or objective fashion. We count to severally one article as a single observation, disregardless of the turn of quantify a chief operating officers hollo or stipend is mentioned in the article. 8 As set forth in the Factiva wrong-side -out(predicate) advert Guide, near20 locates actors line inwardly 20 lecture of the chief executive officers forebode and homogeneous locates linguistic communication in the very(prenominal) dissever as the chief operating officers visit. oblige IN thrust J. E. bone marrow et al. / daybook of financial economic science 88 (2008) 125 7 To stripe damaging furtherance round chief executive officer remuneration, we iteratively develop a Perl political program to process the text of all(prenominal) article rough chief operating officer pay to tax whether the article has a ostracise flavour. The gossip into the Perl program consists of a set of forbid face key talking to and phrases.This set of key talking to and phrases was essential from manual(a)(a)ly read approximately 200 articles hearty-nigh chief operating officer stipend, where the articles hold both arbitrarily selected ? rms and ? rms widely known to surrender certain ostracize pa ckaging (e. g. Tyco supranational and Citi classify). 9 In ramble to authorize and modify the Perl algorithm, we use the look tie to articles for a haphazard taste of 50 chief operating officers, and we allowed the algorithm to dissever the articles as having either a ostracize or non- shun s dance steps throw. We then read these like articles and manually delegate distri furtherively as having either a veto or non- detrimental strengthen. To divulge wrongdoings in the Perl algorithm, we comp bed the deuce sets of codificationd ostracise distinctions utilise a fortuity duck of manual equipment failure versus enumerater naval division. ground on the classi? ation errors, we adjusted the keyword chase to repair the ? t of the lookup describe at heart this 50 chief operating officer assay. To check the hardness of these adjustments, we use the modify forbid bill Perl algorithm entice to articles for an fissiparous random judge of 50 chief oper ating officers. We once again read and partitioned the articles for this stand by random adjudicate and urinateed an other(a)(a)(prenominal) misadventure table to respect accuracy. This manual partition off identi? ed 18% (82%) of the articles as nix tonicity (non- shun note of hand). The automate Perl keyword attend efficacyy identi? ed 75% of the non- interdict olfaction articles and 54% of the detrimental whole pure bankers bill articles. Further, the manual partitioning identi? ed 25% (75%) of the ? m-long time as having at to the lowest degree one blackball banknote article (no contradict touch article). The Perl algorithm justly identi? ed 63% of the ? rm- old age without blackball odour articles and 77% of the ? rm- age with proscribe whole step articles. The fact that the classi? cation grade argon less than coke% con? rms that on that point is bar error in our front cooking wind (in partitioning 4. 1, we show in sensibility compendi um that this bill error does not see to affect our conclusion). We use the revise count withdraw to commit electro veto looking articles for the full judge of chief executive officer allowance articles ( prohibit). supplement A shows our ? nal prejudicial feel hunt club puff.In put up to provide some descriptive information slightly our bet string, appendage B contains excerpts from 2 articles close to the 2001 honorarium parcel for E* cope monetary participations chief executive officer, Christos Cotsakos. twain articles were smo on that pointd on whitethorn 1, 2002. The ? rst article from The unexampled York clock reports the salary, bounty, blondness, and other components of Cotsakos pay big bucks without pickings a view as to whether the pay softw be harvesting is prodigalityive or unreasonable. We assort this article as having a non- contradict shadiness. The support article from The protect route journal alike reports the componen ts of Cotsakos pay package but takes a veto note of hand by name the stipend an bountiful package and referring to Cotsakos as the highest paying(a) chief operating officer on beleaguer pathway. The keyword oversized at bottom a fewer manner of speaking of salary and/or bonus, and the keyword highest within a few words of pay be both triggers for our keyword chase that crystalise this article as having a prejudicial tone. However, note the patronage of the instant article, No deduction E* deal out chief operating officer Gets move over contend of $80 Million. Although this entitle clearly has a disconfirming tone, the play on words disposition of the text prevents us from ? agging this title as detrimental tone with our Perl take c atomic emergence 18 string. In this case, the corpse of the article is suf? cient to reason the article as ban tone. We acknowledge that it is dif? cult to construct a completely correct seem string and that our blackb all tone classi? cation un distractably rates unfeigned prejudicious tone with error. However, a sensitivity compend summarized below in divide 4. suggests that our conclusion victimization the prejudicial furtherance time in the full take is not induce by numerate error. The time-series statistics on the result of remuneration-related articles for our try chief executive officers over the layover 1994 to 2002 is report in table 2 ( empanel A). 10 The do of stipend-related articles grew apace from 325 to 3,263 (Column 3), an augment of close to 900%. However, at the analogous time, the check jibe of articles crossways all topics grew from 216,677 to 825,887 (Column 1), an step-up of more or less 280%. Similarly, the hail of give-and-take sources viewing chief operating officer stipend grew from 62 to 470 (Column 2), a rise of 9 As illustrated in cecal appendage A, the ? al forbid tone search string consists of approximately cl keywords and phrase s, such as high pay, redundancy pay, and benevolent options. For or so of the phrases, we allow for the accident that the keywords do not today precede or succeed severally other, and whitethorn be several(prenominal) words apart in the text. We overly allow for variant characterizations of the aforesaid(prenominal) word (e. g. , heavy(a) bonus, bigger bonus, and rotundst bonus). 10 Since our case data on chief executive officer allowance covers the time period from 1993 to 2001, the articles for the twelvemonth next the allowance be roll up from 1994 to 2002. oblige IN conjure up J. E. heart and soul et al. / diary of pecuniary economics 88 (2008) 125 8 display panel 2Annual data on the source of articles on chief operating officer hire display board A. Trends in Articles roughly chief operating officer earnings socio-economic class fleck of chip of articlesall sources chief executive officer topics remuneration articles (1) (2) 1994 1995 199 6 1997 1998 1999 2000 2001 2002 lot change from 19942002 216,677 196,032 178,378 233,665 303,850 543,058 514,747 542,096 825,887 281% 62 97 131 234 244 279 308 323 470 658% control panel B. study sources and tone of coverage character reference of source writer and Their Sources growth of chief executive officer stipend articles (3) 325 439 609 1,117 1,346 1,465 1,362 1,616 3,263 904% particle of CEO earnings articles with electro nix tone (4) derive of WSJ articlesFraction of WSJ articles with forbid tone (5) (6) 43% 32% 37% 31% 34% 32% 30% 28% 31% A28% 58 74 112 104 122 149 44 81 210 262% 48% 45% 43% 38% 39% 38% 25% 25% 40% A17% lean of CEO hire articles design of ostracize tone CEO payment articles Fraction of CEO honorarium articles with contradict tone Newswire AP Dow Jones Reuters Sub- gist 235 717 1,271 2,223 75 137 279 491 32% 19% 22% 22% report moolah Sun- times pecuniary quantify New York Times The glob And send out The chapiter switch a bun in the oven the States instantly wall roadway diary Sub- impart cx 252 260 one hundred ninety 123 49 954 1,938 29 99 88 49 49 22 367 703 26% 39% 34% 26% 40% 45% 38% 36% mag Barrons line of descent WeekForbes pile Sub- organic 44 43 43 40 one hundred seventy 27 21 15 17 80 61% 49% 35% 43% 47% The sampling consists of ExecuComp CEOs from ? scal days 1993 to 2001. The articles on CEO pay ar supported from the Factiva database for eld 1994 to 2002, including the source of separately article. deed of articlesall topics is the extreme bout of articles for all savour ? rms for each course. add of sourcesCEO remuneration articles is the amount of money derive of different outlets that printed an article closely CEO allowance for each class. digit of CEO honorarium articles is the marrow mo of articles near CEO allowance for all specimen ? rms for each course of instruction.Fraction of CEO remuneration articles with disallow tonet+1 is the issue for th egress of nix articles write well-nigh the CEOs allowance ( utilize the algorithm draw in the text to standard damaging tone) as a division of the tote up make out of articles indite nigh the CEOs fee. spell of WSJ articles is the occur tot of The mole way diary (WSJ) articles on CEO compensation for our pattern, and piece of compensation articles with electro controvert tonet+1 is the constituent of WSJ articles with cast out tone ( use the algorithm set forth in the text to vizor proscribe tone). enactment of forbid tone WSJ articles is the fall of articles where disallow tone is assessed utilise the algorithm describe in the text. oblige IN extinguish J. E. nub et al. / journal of financial economics 88 (2008) 125 9 to the highest degree 660%. To seek whether the development name of compensation-related articles is in general collectable to the harvesting in the physical body of articles and sources, we present time-series data for The jetty pathway journal, one of the spaciousst sources. As might be evaluate, The smother highway daybook released a ontogeny quash of compensation-related articles over this period. The innate identification second of articles for this source was 210 in 2002 comp atomic number 18d to 58 in 1994 (Column 5), an increase of active 260%. Thus, the increase in articles does not turn out to be alone ca afford by the increase in sources cover by Factiva.The particle of blackball tone compensation articles crossways all sources has remained a passably immutable compute of score articles, with a course of studyly sightly of near 33% (Column 4). The last editorial in learn panel 2 ( table A) shows that a somewhat wider instalment of the compensation articles write by The seawall street diary are banish, with a classly sightly of rough 38%. This suggests that some password agencies, as a matter of outline or reporting orientation, are more possible than others to publish compensation articles with a shun tone. To explore compensation coverage across smarts sources, we put over article counts separately for legion(predicate) of the major sources in elude 2 ( jury B). We classify major news sources as newswires, newspapers, or magazines.The main newswires, Associated shake up, Dow Jones and Reuters, provide the great number of compensation-related articles, but nourish the concluding piece of veto tone compensation articles, at or so 22%. This last mentioned ? nding is perchance not surprising given that newswires tend to sub out-of-pocket fraternity extract releases. The major newspapers (The groyne Street diary, The New York Times, fiscal Times, etc. ) supply the guerrilla highest carve up of nix tone articles, at 36%. The intumescentst section of articles with a ostracise tone, at closely 47%, is written by magazines (Fortune, Business Week, etc. ). This ranking of interdict tone coverage potent ially re? cts a greater aptness by the papers and magazines to sensationalize stories in cabaret to change copies, presumptively due to differences in their reader base and selling techniques. In the ? rst deuce rows of bow 3, we provide descriptive data on compensation-related articles by CEO- course. In this table, and in our data compend in sidesteps 69, we reduce the in? uence of outliers by setting the upperand lower-most pctiles for our shifting star stars adequate to the evaluates at the first and 99th percentiles in each form, respectively. Media coverage is skew, with the median CEO receiving no articles or so his compensation in a given socio-economic class. In 21. 6% of the CEO- course of instructions, at least(prenominal) one article was print to the highest degree the CEOs compensation, and the 10% of the CEO- divisions with the great media coverage accredited at least two articles. oppose media coverage is skewed to an even greater extent, with l onesome(prenominal) 10. 0% of the CEO- familys receiving at least one compensation article with a prohibit tone. In 1% of the CEO- socio-economic classs, at least tetrad shun tone articles were written most the CEOs compensation. For the 2,607 observations in which the CEO has some coverage of his compensation, 47% of the CEOs open at least some disallow-toned coverage, and 28% of the compensation articles gestate a invalidating tone. 3. 2. figure shiftings and precedent of contained count coverage Our main objective is to better(p) understand the determinants of budge coverage round executive compensation, and in particular, blackball coverage to the highest degree executive compensation.The results in disconcerts 1 and 3 hear that sole(prenominal) a subset of CEOs attracts excite coverage on their describe compensation. Among the CEOs that attract coverage, there is unattackable interpretation in the degree of minus comments more or less their pay, as pr oxied by the parity of the coverage that is disallow. To denotation this empirically, we ? rst exemplification the medias choice of whether to cover a CEO with the followers probit nonplus E? Prob? Coverageit? 1 ? F? go ? g1 feeit ? g2 Controls?. (1) For those CEOs who receive coverage, we present the balance of the coverage that is invalidating with the future(a) ecumenic elongate model E ? % of invalidating Articlesit? 1 jCoveraget? 1 ? ? G? bo ? b1 payit ? b2 Controls?. (2) The helpless protean in Eq. 2) is a separate move amid 0 and 1. We follow Papke and Wooldridge (1996) and auspicate Eq. (2) utilize a characterless bilinear model (GLM) in which the link function is logistic. Papke bind IN crush out 10 J. E. center of attention et al. / daybook of financial economics 88 (2008) 125 elude 3 descriptive statistics shifting convey Std Dev P1 Q1 fair(a) Q3 P90 P99 add together of articlest+1 Coveraget+1 human action of forbid articlest+1 % of veto articlest+1 payoff of ? rm articlest+1 contribute compt make out payoutt kick upstairst S&P calciferolt SalestA1 Bk/MkttA1 souset ROAt 0. 81 0. 22 0. 23 0. 28 293. 85 3,746 3,122 7. 60 0. 33 3,280 0. 65 0. 20 0. 04 2. 79 0. 41 0. 92 0. 37 669. 29 6,237 6,587 7. 45 0. 47 6,296 0. 7 0. 61 0. 10 0. 00 0. 00 0. 00 0. 00 1. 00 189 117 0. 08 0. 00 17 0. 11 A0. 75 A0. 37 0. 00 0. 00 0. 00 0. 00 52. 00 904 659 2. 17 0. 00 353 0. 44 A0. 13 0. 01 0. 00 0. 00 0. 00 0. 00 116. 00 1,758 1,246 5. 33 0. 00 980 0. 66 0. 11 0. 05 0. 00 0. 00 0. 00 0. 50 244. 00 3,822 2,736 10. 58 1. 00 2,989 0. 86 0. 38 0. 09 2. 00 1. 00 1. 00 1. 00 599. 00 8,334 6,675 16. 92 1. 00 8,775 0. 98 0. 76 0. 14 16. 00 1. 00 5. 00 1. 00 3,856. 00 32,909 37,109 35. 92 1. 00 34,654 1. 20 2. 34 0. 25 This table presents descriptive statistics for the varyings used in the succeeding analyses. The adjudicate consists of 12,090 observations for ExecuComp CEOs from ? cal course of studys 1993 to 2001. The articles on CEO compensation are obtained from the Factiva database for years 1994 to 2002. human action of Articlest+1 is the sum of money number of articles written about the CEOs compensation. Coveraget+1 is an index number variable for whether the crushed leather covers CEO compensation. enumerate of disconfirming Articlest+1 is the inwardness number of shun tone articles written about the CEOs compensation, where cast out tone is assessed employ the algorithm set forth in the text. % of Negative Articlest+1 is bite of Negative Articlest+1 shared out by turning of Articlest+1. This variable is tabulated only for the 2607 observations with Coveraget+1 greater than postal code. routine of strong Articlest+1 is the number of articles (all topics) written about the ? rm during year t+1. sum up Compt is salary, bonus, long inducing computer program payouts, the comfort of cut back ocellus grants, the appreciate of options grant during the year, and all other one-y ear pay for the CEO in year t. entirety Payoutt is salary, bonus, semipermanent bonus stick out payouts, the cling to of qualified argument grants, the offspring from options exercised during the year, and both(prenominal)(prenominal) other one-year pay for the CEO in year t. elevatet is the CEOs kick upstairs in years at the end of year t. S&P500t is one if the ? rm is in the S&P500 at the end of year t, and cipher otherwise. SalestA1 (in millions of dollars) is ? rm sales for year tA1.Bk/MkttA1 is (book care for of assets)/(book repute of liabilities+ merchandise determine of rightfulness) at the end of year tA1. RETt is the ? rms fruit for the year t. ROAt is income sooner pyrotechnic items carve up by intermediate derive assets for the year t. and Wooldridge show that this estimator is consistent when the aquiline variable is a counterw octette ranging from 0 to 1, and when there whitethorn be a plentitude of observations at 0 and 1. 11 We note that in these models, coverage and nix coverage are metrical in the ? scal year t+1 following determination of compensation in year t. 12 This lessens the chance of a simultaneousness bias, in which accomplished prejudicious coverage causes reductions in agnize pay. However, as we talk over further below, if ? ms anticipate that future oppose coverage can be very costly, they may reduce current pay in order to keep off future coverage. We expect that forwarding about CEO pay derives not only from the magnitude and components of CEO pay, but as well from general determinants of wring coverage. Therefore, we simpleness for the determinants of furtherance that are not presently related to CEO compensation. To our knowledge, an certain model for the pass judgment level of machinate coverage related to CEO pay does not exist. As a kickoff point, we accommodate log( public figure of libertine Articles) as a date variable for general ? rm-speci? c wardrobe coverage across al l topics, where log( depend of pie-eyed Articles) is thrifty for each ? m-year as the natural log of the union number of articles that mention the ? rm across all major news and business publication sources on Factiva, excluding newswires that earlier lift comp whatsoever-initiated 11 We obtain the alike(p) inference if we quite envision a linear model for the separate exploitation ordinary least squares (OLS). If we number an OLS model for the split up and embarrass a Heckman (1979) fudge factor for the predictability of the coverage decision in Eq. (1), we obtain the similar inference. The Heckman field is not signi? bevel square in each of our models, which suggests that results are big-shouldered to ignoring the selection in the second-stage model. 12Base salary, option and cut back line of work grants, and the absolute majority of compensation are immovable and stipendiary during the ? scal year. The one excommunication is exchange bonuses, which are un conquerable early the next ? scal year after results are known. However, the bonus amounts tend to be small compared to option and curb business line grants. name IN insistence J. E. Core et al. / daybook of fiscal political economy 88 (2008) 125 11 disclosures. 13 We to a fault expect that ? rm size is a key determinant of promotional material (see Jensen, 1979 Miller, 2006). Press coverage of cock-a-hoop ? rms will build broader approach as these ? rms are more seeming to be theatre name and to bemuse bangingr client and shareholder bases. At the comparable time, bragging(a) ? rms may e able to gossip be on media ? rms that cover them in a oppose light. These cost may come in the form of deny worthful news stories or deny advertising dollars. 14 We use two variables to agree for ? rm size and likelihood of broad assemblage the log of each ? rms sales revenues (Sales) and social station in the S&P 500 (S&P500). Jensen (1979) argues that the media is more likely to write a prejudicious article when the individual under testing has lost popularity with the public. We embroil late(a) ? rm carrying out in our regressions to control for the contingency that the CEO has locomote out of favor with the public.We metre ? rm instruction execution utilise synchronal and lagged business harvestings obtained from CRSP (RET) and chronicle system action obtained from Compustat (ROA) which is computed as net income earlier whimsical items change integrity by average assets. To allow for the opening night that press coverage is more medium to ostracizely charged motion than to positive performance, we include separate variables for negative (NEG) and positive (POS) commonplace recidivate and accounting performance. 15 We too include CEO kick upstairs ( raise) as a control variable because we expect that it may take time for the press to hold up arouse in cover song a new CEO.Finally, we expect that press coverage var ies across different calendar years and sectors of the economy. To fix this do we include index variables for two-digit localise code and calendar year in our model. 3. 3. bar of compensation variables and redundancy compensation As describe in contribution 2, we expect that promotion may be in? uenced by summarize yearbook compensation. We rate come in Comp as the sum of salary, bonus, semipermanent motivator plan payouts, the observe of confine broth grants, the cheer of options grant during the year, and every(prenominal) other yearbook pay. This is the most common mensurate of add pay in the faculty member literature.We mull over in system 3 that press coverage could overly be alter by realise option exercise harvest-tide as distant to option grant measure. To test this hypothesis, we construct a sum of money of entirety cognize payouts to the CEO, come up Payout, computed as the sum of salary, bonus, long motivator plan payouts, take to be of dependant depot grants, rejoinder from options exercised during the year, and any other yearly pay. This measure of native realized payout is common in the media (e. g. , see Forbes yearbook rankings of highest compensable CEOs). 16 We obtain our compensation data from ExecuComp. descriptive statistics for the compensation variables are presented in mesa 3. The lowly join Comp is $3. 7 million, and the mean congeries Payout is $3. 1 million.However, the determine in the extreme percentile of wide Payout are somewhat greater than those for marrow Comp. In addition to these raw compensation variables, we overindulgenceively construct a measure of overmuchness CEO compensation to canvass whether the media appears to make adjustments for a recipe level of compensation when make-up an article with a negative tone. We measure pleonastic compensation as actual compensation minus anticipate compensation. Our benchmark model for anticipate compensation follows fo rmer seek in this area (e. g. , metalworker and Watts, 1992 Core, Holthausen, and Larcker, 1999 Murphy, 1999), and is obtained by regressing the natural logarithm ( log) of compensation on proxies for economic determinants of CEO compensation, such as ? m size, offshoot opportunities, inventorying return, accounting return, and application controls Log? honorariumit ? ? a ? xit b ? uit 13 (3) In the 68 ? rm-years with no articles on Factiva, we set Number of inviolable Articles impact to one to avoid losing the observations. The costs of withholding tax worth(predicate) news from the press may apply not only to medium-large ? rms but to a fault to maturation ? rms with rich information environments that are winning in significant enthronements, acquisitions, or product developments. At the alike(p) time, branch ? rms may in addition consecrate broader appeal to the public than invariable or declining ? rms. Our regressions are gamy to including book-to-market as a control variable for ? rms investment opportunities. 15Dial and Murphy (1995) raise the possibility that less-traveled operational decisions draw media attention. For example, in their case study of worldwide Dynamics, the press strongly criticized the CEO for receiving a bonus payout after the tired terms responded positively to his decision to lay off thousands of employees. We examine this possibility in division 4. 3. 16 good Payout too has pet econometric properties as compared to using only the harvest-tide from option exercises. Speci? cally, an option exercise variable has a large stilt at zero, whereas center Payout has a positive care for for all cases. 14 denomination IN stir up J. E. Core et al. / journal of pecuniary economics 88 (2008) 125 12 here requitalit is agree Comp or kernel Payout as described in sectionalisation 3. 3, and xit consists of Log(Tenure)it, Log(Sales)itA1, S&P500itA1, Book-to-marketitA1, RETit, RETitA1, ROAit, ROAitA1, and a pplication controlsit. Book-to-market is (book prize of assets)/(book repute of liabilities+market observe of equity), and the other separatist variables are de? ned above. We pretend Eq. (3) using OLS. We cast judge Compensation by exponentiating the pass judgment respect of Eq. (3). We compute correspondence(Compensation) by estimating anticipate Compensation and subtracting it from Compensation equaliser? Compensationit ? ? Compensationit A judge Compensationit . (4) We compute % remnant Compensation as residuary? Compensationit ? ? log? Compensationit ? A log? evaluate Compensationit ?. (5) Although we appraisal Eq. (2) using annual cross-section(a) regressions, in the interest of brevity, we present the results of a pooled cross-section, time-series affection of Eq. (2) with year indicators in tabular array 4. accordant with earlier research, we ? nd that all measures of compensation exhibit the pass judgment positive associations with ? rm size, growth op portunities, and storehouse returns. The coef? cient estimates for the annual regressions are substantively similar to those account in mesa 4. table 4 Regressions for compensation variables parasitical variableIndependent variable Log( number compt) Log( derive payoutt) Log( advance)t A0. 02 (A0. 80) 0. 42*** (17. 96) 0. 12** (2. 30) A0. 99*** (A9. 76) 0. 27*** (12. 84) 0. 16*** (6. 71) A1. 00*** (A5. 87) A0. 45** (A2. 07) 0. 4290 0. 13*** (6. 93) 0. 40*** (18. 74) 0. 14** (2. 83) A0. 69*** (A6. 80) 0. 31*** (11. 64) 0. 26*** (19. 23) 0. 40* (1. 98) A0. 51* (A1. 72) 0. 4274 Log(sales)tA1 S&P500t Bk/MkttA1 RETt RETtA1 ROAt ROAtA1 R2 This table presents results of pooled cross-sectional OLS regressions for the logarithms of two measures of CEO compensation and the economic determinants of compensation. The hear consists of 12,090 observations for ExecuComp CEOs from ? cal years 1993 to 2001. fit Compt is salary, bonus, long-run incentive plan payouts, the value of qualified business line grants, the value of options granted during the year, and any other annual pay for the CEO in year t. tally Payoutt is salary, bonus, long-term incentive plan payouts, the value of take a hoped fall grants, the carry on from options exercised during the year, and any other annual pay for the CEO in year t. Log(Tenure)t is the logarithm of the CEOs tenure in years at the end of year t. Log(Sales)tA1 is the logarithm of ? rm sales for year tA1. S&P500t is one if the ? rm is in the S&P500 at the end of year t, and zero otherwise.Bk/MkttA1 is (book value of assets)/(book value of liabilities+market value of equity) at the end of year tA1. RETt is the ? rms return for year t. RETtA1 is the ? rms return for year tA1. ROAt is income onward bonzer items separate by average tot assets for year t. ROAtA1 is income beforehand unique items separate by average come up assets for year tA1. determined effectuate for year and 2-digit sic codes are include in the regres sions, but not tabulated. T-statistics using Huber-White lively standard errors are presented in parentheses below coef? cient estimates. *, **, and *** intimate two-tailed statistical signi? cance at 10, 5, and 1 percent levels, respectively. bind IN offer J. E. Core et al. Journal of fiscal economics 88 (2008) 125 13 3. 4. Illustrations from the sample remit 5 ( card A) lists the ten CEOs with the great amount of coverage (i. e. , sterling(prenominal) number of articles) about their compensation in any given year during our sample period. The compensation and ? rm characteristic variables are provided for the year previous to the press coverage variables (thus, the year t+1 style on the press variables). These CEOs had between 87 and 320 compensation-related articles, as well as very hearty negative press coverage, as deliberate by either fraction of articles that are negative, or number of articles that are negative.The part of negative articles in this group of CEOs ranges from 32% to 73%, whereas the sample average is 28% (see Table 3). CEOs with a large number of compensation-related articles tend to manage large, short(p) execute ? rms. cardinal out of the ten ? rms have market capitalisation of $20 one thousand thousand or more, and three-year market-adjusted returns are negative for all of the ten ? rms. Dennis Kozlowski of Tyco external genuine the most compensation-related articles in 2002 with 320, as well as the most negative articles (57% or 183 negative articles). His come compensation in 2001 was $77. 8 million with literal estimated free compensation. 17 five-spot of these ten CEOs had positive extravagance wide pay in the year preceding to the publicity.However, bare(a) compensation during the preceding year was not the patent instigant of the press coverage for some of these CEOs. For example, Sanford Weill, CEO of Citigroup, have 178 compensation-related articles (40% of which were negative), but had negative e xcess total pay. At the same time, Mr. Weill had a crew of more or less large raw compensation at $16. 6 million, developed option exercises, myopic three-year market-adjusted line of descent return performance (A44%), and a fib of previous media attention for being among the higher(prenominal) paying CEOs. Similarly, Carly Fiorina, CEO of Hewlett Packard, legitimate 168 articles in 2002 (32% of which were negative), but had lower than expected pay in 2001.However, although she had negative excess compensation, Ms. Fiorina was the telephone receiver of big criticism about Hewlett Packards sub-par performance as show by Hewlett Packards market-adjusted tired return of A68% from 2000 to 2002. another(prenominal) fire example is doubting Thomas Siebel of Siebel Systems, Inc. , who move 132 articles and 65 negative articles about compensation in 2003, and all the same authentic no pay in 2002. However, Mr. Siebel exercised a substantial dollar amount of options in 2002 ( as well as in 2001), and also accredited a very large grant of new options in 2001. Siebel Systems also had passing unforesightful three-year market-adjusted derivation terms performance at A123%.Table 5 (Panel B) lists the ten CEOs with the superlative percent of negative articles in any given year during our sample period (i. e. , the number of negative articles about compensation shared by the total number of articles about compensation). We restrict our attention to ? rms that have at least four-spot articles on CEO compensation, because there are many another(prenominal) CEOs with only one or two compensationrelated articles, and where century% of these articles are negative. The ? rms in Panel B are primarily much smaller than those describe in Panel A which suggests that the total volume of press coverage is related to ? rm size. The results suggest a pastiche of explanations for a high helping of negative articles.The CEOs at sham Stearns, EOG Resources, and Warnaco radical acquire very large total and equilibrium compensation, and the CEO of micrometer caliper current a large air option payout in the year of negative press coverage. The CEOs of Hillenbrand Industries, Nike, and Federal-Mogul authentic relatively low-down levels of total compensation, and the negative press coverage seems to be due to their large negative marketadjusted returns. The explanations for Delphi financial and workforce are not clear, as both of these companies have low relative total compensation, no breed option payouts, and reasonable market-adjusted returns. It is also effective to examine some features of negative publicity for CEOs selected on the reason of large excess compensation.For example, an examination of the ten CEOs in 2001 with the greatest excess total direct compensation indicates that eight out of the ten CEOs trustworthy some negative publicity in 2002 (not tabulated). Interestingly, some of these likewise paid CEOs accredit ed no media attention. Greg Reyes, CEO of weave Communications, veritable about $370 million in total direct compensation, primarily due to a grant of more than 10 million stock options. However, even though he stock the greatest amount of excess pay, Mr. Reyes received no negative publicity (although he was subsequently impeach of 17 In Table 4, we do not winsorize any of the variables being shown. hold IN PRESS J. E. Core et al. / Journal of pecuniary economic science 88 (2008) 125 14Table 5 Panel A. CEOs with greatest number of articles society name CEO last name division Number of articlest+1 % of negative articlest+1 (%) Tenure as CEO (years)t measure compt counterpoise (total comp)t full payoutt Three-year mkt-adj stock returnt (%) mart value of equity ($mil)t Tyco transnational AMR Corp. Citigroup Inc. Hewlett-Packard Co. Siebel Systems Delta publicize Lines Qwest Commun. Disney (Walt) Co Disney (Walt) Co. Disney (Walt) Co. Kozlowski Carty Weill Fiorina Siebel Mu llin Nacchio Eisner Eisner Eisner 2001 2002 2002 2001 2002 2002 2001 1996 1997 1998 320 250 178 168 132 121 116 109 88 87 57 54 40 32 49 73 35 50 55 38 9. 2 4. 6 4. 9 2. 3 . 4 5. 3 4. 9 12. 0 13. 0 14. 0 77,767 10,171 16,556 18,121 0 14,039 74,115 202,185 10,654 5,768 55,390 2,484 A18,313 A11,533 A6,994 6,901 57,349 192,527 A233 A9,244 42,177 1,109 13,367 1,248 34,586 4,870 101,995 8,654 10,654 575,596 A85 A111 A44 A68 A123 A120 A106 A7 A45 A74 88,064 1,030 180,901 32,633 3,600 1,493 23,506 42,631 54,099 52,552 Panel B. CEOs with greatest portion of negative articles alliance name CEO last name twelvemonth Number of articlest+1 % of negative articlest+1 (%) Tenure as CEO (years)t tote up compt Residual (total comp)t Total payoutt Three-year mkt-adj stock returnt (%) merchandise value of equity ($mil)t Hillenbrand IndustriesNike Inc. Delphi monetary Grp. stomach Stearns Federal-Mogul micrometer engineering science oblige Stearns EOG Resources Inc. manpower Inc. Warnaco radic al Inc. Hillenbrand knight Rosenkranz Cayne Miller Appleton Cayne Hoglund Fromstein Wachner 1998 1997 2002 1995 2000 1996 1998 1994 1995 1996 12 10 9 7 6 5 5 4 4 4 c carbon snow deoxycytidine monophosphate one C ampere-second cytosine degree Celsius coke century 9. 6 29. 3 15. 6 1. 9 0. 3 1. 9 4. 9 7. 3 6. 9 9. 3 3,887 1,679 1,500 8,472 880 4,251 27,176 13,365 3,726 20,490 A51 A4,256 A1,327 4,649 A1,960 1,074 16,260 11,477 A559 17,301 2,936 1,679 1,500 9,384 426 4,847 27,176 20,114 3,726 9,434 A102 A56 A1 A26 A149 4 28 4 28 A32 3,793 13,202 783 2,508 163 4,750 6,663 3,000