CORPORATE SOCIAL RESPONSIBILITY (CSR) AS A MODEL OF EXTENDED CORPORATE GOVERNANCE

Transcription

1 Luc Papers n. 142, Sere Etca, Drtto ed Economa 10, suppl. a febbrao 2004 CORPORATE SOCIAL RESPONSIBILITY (CSR) AS A MODEL OF EXTENDED CORPORATE GOVERNANCE. AN EXPLANATION BASED ON THE ECONOMIC THEORIES OF SOCIAL CONTRACT, REPUTATION AND RECIPROCAL CONFORMISM Lorenzo Saccon 1. CSR: Much dsorder under the sun For a long tme corporate ethcal and socal responsblty has been a subject of analyss and concern for hghly actve and commtted but not exactly majorty groups of busness ethcs scholars, NGO s actvsts, and frms wth forward-lookng managers. More recently, on the contrary, t seems that there s no trade or busness assocaton, local or natonal publc authorty, frm or unversty throughout Europe gvng up the opportunty of organzng a conference or a meetng on the theme. Snce the end of the last century, a varety of ntatves throughout Europe 1 and the US 2 have sought to formulate standards for management systems devoted to mplementng corporate socal responsblty (henceforth CSR) wthn corporatons. These are ntatves whch have sprung from jont projects amongst research centres and unverstes, busness frms and the professons; and they nvolve cooperatves, non-governmental and I gratefully acknowledge the support obtaned under the UE research project developng a CSR framework to ntegrate Q-RES and other socal and ethcal standards. 1 By way of example, n the UK the Accountablty 1000 standard for ethcal and socal audt and reportng has been developed, followed by the Sgma Project supported by the Blar government. A Values Management System has been developed n Germany at the Unversty of Konstanz (Weland 2003), whle CSR standards modelled on the example of qualty management systems have been proposed by ndependent bodes of standardzaton n Span (Aenor) and France (Afnor). 2 In 2000 the US Ethcs Offcer Assocaton submtted to the US ISO Commttee (BCMS) a proposal for startng the settng of a new standard for corporate ethcs and socal responsblty (see Weland 2003). 1

2 Luc Papers n. 142, suppl. a febbrao 2004 non-proft organzatons, representatves of nterest organzatons, stakeholders assocatons and sometmes local and government authortes as well. In ths regard Italy has for once not lagged behnd, as testfed by two voluntary ntatves. 3 Many of these as well as other ntatves whch orgnally conssted n self-regulaton models or best practces by frms have been reflected n (but also encouraged by) the Green Paper (2001), and then the Communcaton (2002), ssued by the European Commsson to set out ts strategy for the defnton of concepts, benchmarkng experences, poolng of approaches, and development of a CSR framework at the European level. Ths has been accompaned by burgeonng publc nterest whch has been ndubtably fostered by the decson to put CSR on the agenda of governmental nsttutons, frst those of the European Unon (the UE Commsson) and then those of the member-states. At the same tme, however, unease has been expressed over the perceved superfcalty of a number of these talks, together wth a call for clarfcaton so that msunderstandngs may be prevented, and the consequent false expectatons and dsappontments forestalled. Sometmes nsttutonal ntatves do not help 4. To begn wth, n some governmental ntatves CSR rsks to be confused wth corporate gvng and wth the purpose of channellng of corporate donatons nto socal programmes set up by the Government. The mstake n part resdes n a muddlng of the whole, whch s CSR, wth the part, namely corporate gvng. But t conssts above all n tamperng wth the structural mechansm of the donaton-based fundng of the nonproft sector. Asde from dfferences n place and tme, economc theory of nonproft organsatons sees the thrd sector as a response to supramedan demand for quantty, qualty and cultural dfferentaton n the supply of welfare goods and servces that the governmental supply s unable to satsfy, The government n fact s structurally unable to cater to ths demand wth ts own programmes (fnanced out of taxaton), for these are by defnton desgned to satsfy demand by the medan voter.e. the voter whose consensus gves a poltcal coalton or a party vctory at electons (Wesbrod 1988). If ad hoc fscal and fnancal ncentves are used to channel donatons towards socal polces decded by the government (wth the obvous ntent of reducng even more fscal pressure), there s a rsk of dstortng one of the structural mechansms whch fuel the development of the thrd sector and remedy neffcences n the overall supply of welfare servces. A second rsk s msconcevng the ncentves mechansms that can promote voluntary CSR standards. Ths ensues f as the ntatve under examnaton suggests - the promse of tax relef on donatons provded these are pledged to government s preferred socal programs s combned wth the announcement of a system to ascertan how much companes are socally responsble - called socal 3 The GBS standard for socal reportng ssued n the sprng of 2001, and the Q-RES Project amng to defne a qualty standard of management systems for ethcal and socal responsblty of frms. The latter ntatve led n October 2001 to ssung the Q-RES Management Gudelnes, see (Saccon et al. 2003), and more recently to the settng of the Q-RES Norm for the Improvement of Corporate Ethcal- Socal Performances of Organsatons (March 2003) see (Saccon, decolle, Baldn, Oakley, Weland and Zadek 2003) 4 To exemplfy the msunderstandng of CSR at governmental level, n the text I make some remarks on the Italan Government s ntatve (Mnstry of Welfare) called CSR-SC project, proposed durng the semester of Italan UE Presdency. 2

3 Lorenzo Saccon, Corporate Socal Responsblty (CSR) as a model of extended corporate governance. statement. By fllng the statement companes would gve rse to a publc regster of frms elgble for tax relef and enttled to publcze themselves as socally responsble - admtted that they gve ther donaton to a publc fund managed by the Government. Ths nduces perverse ncentves, for all frms and rghtly so seek to obtan tax abatement on donatons to the thrd sector, so that n order to admt them to tax reducton a mnmal quas-compulsory and fscal verfcaton s needed. Instead, voluntary CSR certfcaton should repay wth reputaton benefts (and only wth such beneft) only those frms effectvely able to demonstrate ther complance wth hgh standards of CSR. The lnkage between CSR certfcaton amed to reputaton and mnsteral apprasal amed to concesson of fscal benefts may arouse such dscontent n the busness communty (afrad that the standard would be too hgh to many companes only nterested n obtanng some tax relef) that at end the Government s forced to reduce CSR certfcaton to smple a self-declaraton about a short lst of compulsory tems, manly concerned wth corporate communty gvng, and much less commttng and nformatve than a proper socal report. (Whch s easy to foresee: f a label s gven by a Mnstry n exchange for money contrbuted to a governmental fund, whch the Mnstry has nterest to rase as hgh as possble, there are very poor ncentves to be very selectve n grantng that label whose value result thus nflated.) At end, after much dsorder and alarm, all that turn too ndulgent towards those backward frms that see the voluntary approach to CSR as mere dscreton (the exact opposte of a moral oblgaton whch frms are asked to account for). Not only Governments and frms, however, but also dstngushed scholars are seemngly reluctant to gve the proper weght to a well-posed self-regulaton approach to CSR. To be far n accountng for the ongong dscusson on CSR, menton must be made of two opposed form of sceptcsm. The frst can be dentfed wth the poston put forward by professor Mchael Jensen (2001), whch bascally cast doubts on any mult-stakeholder approach to strategc management and governance of frms for these would destroy the unqueness of goal and drecton of corporate management, openng the route to manageral slack and self-dealng. It s sharply sympathetc wth the domnant shareholder value maxmsaton vew, and based n favour of the dea that self-nterest and proft-maxmsaton, even f n the long run, are able to accommodate any real problem whch CSR s about. I wll largely concentrate on these theses and I wll attempt to counteract them n sec. 5, 6, 8. On the other hand, there s a second stream of sceptcsm whch consders any voluntary approach to CSR, and the thess tself that frms should endorse CSR as a crteron for ther management and corporate governance, no more than cheap talk and hence nvokes stronger regulatons and mposton of hard law n order to oblge corporate actors to comply wth even the mnmal requrement of avodng conflcts of nterest. For example Gudo Ross, an outstandng Italan legal scholar and one of the consultants of the UE commsson on the reform of corporate governance, s resolutely opposed to busness ethcs as the bass for ntroducng forms of ethcal and socal self-regulaton by frms (for example, codes of ethcs) (cf. Ross 2003, chap. V). I wll deal wth ths form of sceptcsm n the next secton n order to be able of puttng asde certan prelmnary msunderstandngs when I wll step to the proper dscusson of CSR n the followng sectons. In sectons 9, 10 and 11 I wll provdes two arguments one based on reputaton, the second restng upon 3

4 Luc Papers n. 142, suppl. a febbrao 2004 conformsm whch show that self-regulaton, far from beng wthout teeth, may resort to qute effectve ncentves and motvatons, provdng for ts self-enforcement. 2. Busness ethcs and the lawyer s presumpton As a representatve of the second knd of sceptcsm on CSR, Ross needs some defnton of what hs sceptcsm s about. He fnds a defnton of busness ethcs (cf. bd. p. 116) whch locates frmly t n the sphere of ratonal ethcs as justfcaton (or refusal to justfy) by means of models of mpartal collectve choce economc nsttutons at macro level (for example the mx between state and market n a gven economc system), at meso level (the choce among alternatve forms of enterprse), and mcro level (the soluton of professonal ethcal dlemma of managers). 5 But then he entrely forgets ths approach, n whch busness ethcs s clearly rule-based ethcs, and presumes that t can only be character-based ethcs appealng to personal vrtue and conscence and consequently devod of any motvatonal effcacy and any socal sancton (bd. p. 123, 133-4). From ths follows Ross s presumpton that, snce ethcs entrely lack clarty any effectveness whch are nstead typcal of the law any recourse to self-regulaton based on busness ethcs would be msleadng. Yet ths s to omt one of the most obvous defntons of ethcs: namely that they consttute the set of socal norms commonly accepted on the bass of agreements and conventons, and therefore sustaned by the ratonal choces of multple agents. Once those norms have passed the unversalzablty test, 6 whch s what meta-ethcally dstngushes them from rules of mere prudence or from socal norms not susceptble to moral meanng, they are effectve by dnt of ther socal functon of solvng cooperaton and coordnaton problems. In sum, by forgettng the relaton wth socal norms what s entrely lost s any reference to ratonal ethcs, whch besdes Kant also owe a debt to Hume. 7 It should be borne n mnd that the relatonshp between ethcs and legal rules s a complex one. From a prescrptve pont of vew (a legal theorst would say de jure condendo), ethcs precede the law, n the sense that an endurng legal order always rests upon an overlappng consensus on consttutonal ethcal prncples (Rawls 1993) whch s mpartal wth respect to nterests and partcular conceptons of the good, so that the decson to promulgate a law must always pass a test of mpartal assessment (that s, whether t comples wth the content of the overlappng consensus ) unless t s beleved that legslaton s purely the doman of arbtrary decsons by the majorty. 8 Secondly, ethcs proceed together wth the law, especally as the spontaneous acceptance of the authorty of the law and of the legtmate consttutonal order. Wthout broad acceptance of the authorty 5 The defnton s taken from the ntroducton to Saccon (1991). 6 Cf. Hare (1981). 7 Ths s rather paradoxcal gven the moralstc tone of Ross s conclusons, where he seems to revert to the ethcs of personal vrtues (and vces) whose effcacy he has prevously dsputed. He concludes wth Luther (who n turn ctes St Paul) that the love of money s the root of all evl : cf. bd. p For Rawls (1993) overlappng consensus s the agreement stemmng from publc ethcs defned as the exercse of publc reason : that s, the mnmum ndspensable content of the socal contract whch corresponds to the (more exgent) noton of justce as farness. 4

5 Lorenzo Saccon, Corporate Socal Responsblty (CSR) as a model of extended corporate governance. of the law, and especally of the consttuton, these would be wholly mpossble to enforce. Only a small number of breaches of the legal order can be sanctoned; whle general dsobedence would make t mpossble for any legal order to survve. 9 Moreover, va nterpretatons based on ethcal theores of open norms lke the good fath rule, or wth the help of the ethcal rules arsng from self-regulaton, the judge extends and clarfes the range of applcaton of the legal rules themselves. 10 Fnally, ethcs go beyond the law. Whle numerous matters cannot be regulated by detaled rules, ths can be done only by general prncples or standards agreed by the partes concerned, or by prvate orders establshed by ntermedate groups n socety. These prncples are not necessarly enforced by legal sanctons; used nstead may be other ncentve mechansms lke socal approval, economc advantage, the threat of excluson and beng socally ostracsed, and so on. The economy s typcally the sphere n whch complete contracts ex ante coverng long-term transactons or all-compassng publc regulatons are not cogntvely possble, or else would be too costly to apply. As a consequence, used n economc affars are other forms of socal regulaton, self-constranng socal norms, codes of behavour, standards shared among groups or wthn socal groups, and so on (Posner 2000). It s evdent that the term ethcs s used here accordng to at least two dstnct meanngs. Ethcs can be understood as a normatve dscplne whch apples the rules of moral reasonng, or as a postve scence whch descrbes norms and socal rules shared n the form of socal conventons (Lews 1969) whch n addton acqure moral meanng because they are unversalzable. Thus dentfed are two roles of busness ethcs. The frst s the furnshng of mpartal justfcaton for economc nsttutonal arrangements and rules. The second concerns the socal rules of economc self-regulaton that possess the force mparted by voluntary complance and effectve socal rules, even when these are not enforced by legal sanctons. The former functon has only the force of mpartal moral reasonng and lacks strong ncentves dervng from materal nterests or psychologcal drves; the latter can nstead rely on numerous ndvdual ncentves and effcacous motvatonal force (motvaton may be drven by reputaton and trust or by conformsm and recprocty, two dstnct yet nterdependent mechansms). The two aspects do not necessarly converge. For example, the constructon of socal contract as a choce behnd the vel of gnorance, descrbed by John Rawls (the foremost poltcal phlosopher of the second half of the twenteth century), s no longer supported by personal ncentves or ndvdual motves once the vel s skpped, unless one accepts the debatable psychologcal theory whch holds that the sense of justce s nnate and evolves naturally from the more ntmate sphere of personal relatonshps to mpersonal relatons. 11 However f the theores of reputaton effects and of conformst expectatons 9 See n ths regard the statement by Savero Borrell, former Chef Publc Prosecutor of Mlan, apropos of corrupton: [ ] I do not beleve and I have never beleved that the crmnal justce system s able to cure the lls of socety [ ] the magstrates do not cure the llness wth ther nvestgatons. Ths should be done by the poltcal system, the culture [ ] Not only poltcs but also publc opnon, socety n short, has ts responsbltes. I beleve that the average Italan does not feel partcularly upset by these offences because he or she s used to fnaglng and therefore accepts that others do t as well. And when ths behavour becomes a cultural fact, the general atttude, t s even more dffcult to change ntervew wth La Repubblca, 14/10/ On ths see Prete (1993), chap. 3, secton Cf. Rawls (1971). 5

6 Luc Papers n. 142, suppl. a febbrao 2004 (see sectons 9, 10 and 11) are appled to codes of busness ethcs, t can be shown that the two classes of practcal reasons (the detached ones appealng to mpartal reason, and the personal ones consstng of ncentves and motvatonal drves) may converge n a manner whch ensures that moral rules, as well as codes of ethcs and CSR voluntary management standards, are effectve. 3. A defnton of CSR The EU Commsson s qute demandng n ts defnton of CSR: By statng ther socal responsblty and voluntarly takng on commtments whch go beyond common regulatory and conventonal requrements, whch they would have to respect n any case, companes endeavour to rase the standards of socal development, envronmental protecton and respect of fundamental rghts and embrace an open governance, reconclng nterests of varous stakeholders n an overall approach of qualty and sustanablty (Promotng a European Framework for Corporate Socal Responsblty, Green Paper, p.4, Brussels, , emphass added). Ths quotaton shows that the Commsson regards CSR as a form of corporate strategc management that sets ts standards of conduct at a level hgher than legal constrants, and envsages CSR as a system for the governance of transactons and relatons between the frm and ts stakeholders. It s clear that here governance s no longer the set of rules smply allocatng property rghts and defnng the owners control over the management of a frm. Instead t resembles the neo-nsttutonal vew whereby the frm, lke the contract and other nsttutonal forms, s a governance system whch establshes dverse rghts and oblgatons n order to reduce transacton costs and the negatve externaltes of transactons. I therefore propose the followng defnton of CSR: a model of extended corporate governance whereby who runs a frm (entrepreneurs, drectors, managers) have responsbltes that range from fulflment of ther fducary dutes towards the owners to fulflment of analogous fducary dutes towards all the frm s stakeholders. Defnton s requred of two terms n the foregong proposton: a) Fducary dutes. It s assumed that a subject has a legtmate nterest but s unable to make the relevant decsons, n the sense that s/he does not know what goals to pursue, what alternatve to choose, or how to deploy hs/her resources n order to satsfy hs/her nterest. S/he, the trustor, therefore delegates decsons to a trustee empowered to choose actons and goals. The trustee may thus use the trustor s resources and select the approprate course of acton. For a fducary relatonshp ths beng the bass of the trustee s authorty vs-à-vs the trustor to arse, the latter must possess a clam (rght) towards the former. In other words, the trustee drects actons and uses the resources made over to hm/her so that results are obtaned whch satsfy (to the best extent possble) the trustor s nterests. These clams (.e. the trustor s rghts) mpose fducary dutes on the agent who s enttled wth authorty (the trustee), whch s/he s oblged to fulfl. The fducary relaton apples n a wde varety of nstances: tutor/mnor and teacher/pupl relatonshps, and (n the corporate doman) the relaton between the board of a trust and ts 6

7 Lorenzo Saccon, Corporate Socal Responsblty (CSR) as a model of extended corporate governance. benefcares, or accordng to the predomnant opnon, between the board of drectors of a jontstock company and ts shareholders and then more generally between management and owners (f the latter do not run the enterprse themselves). By the term fducary duty, therefore, s meant the duty (or responsblty) to exercse authorty for the good of those who have granted that authorty and are therefore subject to t. 12 b) Stakeholders. Ths term denotes ndvduals or groups wth a major stake n the runnng of the frm and who are able to nfluence t sgnfcantly (Freeman and McVea 2002). However, a dstncton should be drawn between the followng two categores: Stakeholders n the strct sense: those who have an nterest at stake because they have made specfc nvestments n the frm (n the form of human captal, fnancal captal, socal captal or trust, physcal or envronmental captal, or for the development of dedcated technologes, etc.) that s, nvestments whch may sgnfcantly ncrease the total value generated by the frm (net of the costs sustaned for that purpose) and whch are made specfcally n relaton to that frm (and not n any other) so that ther value s dosyncratcally related to the completon of the transactons carred out by or n relaton to that frm. These stakeholders are recprocally dependent on the frm because they nfluence ts value but at the same tme gven the specfcty of ther nvestment depend largely upon t for satsfacton of ther well-beng prospects (lock-n effect). Stakeholders n the broad sense: those ndvduals or groups whose nterest s nvolved because they undergo the external effects, postve or negatve, of the transactons performed by the frm, even f they do not drectly partcpate n the transacton, so that they do not contrbute to, nor drectly receve value from the frm. It s evdent that these two categores cannot be sharply separated. For example, a manufacturer n a developng country who supples a component for an ndustral good assembled n a Western European country s essentally dependent on hs contract; and wth hs low labour costs (due to the customer s market power) he makes a crucal contrbuton to the European frm s profts. At the same tme, however, f a mature technology s used, he s easly replaceable by the European frm, whose dependence on the suppler s therefore lmted (n short, the recprocal dependence relaton s not symmetrc). Lkewse, a local communty may not be party to the transactons performed by a company wth a plant on ts terrtory, but t s nevertheless subject to that plant s envronmental and socal externaltes. However, f the communty has representatve nsttutons wth the power to grant or wthhold a lcence to operate, t s able to nfluence the company s creaton of value and negotate a reducton n the negatve externaltes. These decsons connected as they are wth the furnshng of nfrastructures may be vewed as nvestments ntended to select and attract producton actvtes whose postve externaltes outwegh ther negatve ones. We are now able to apprecate the scope of CSR defned as an extended form of governance: t extends the concept of fducary duty from a mono-stakeholder settng (where the sole stakeholder 12 On fducary dutes see Flanngan (1989). 7

8 Luc Papers n. 142, suppl. a febbrao 2004 relevant to dentfcaton of fducary dutes s the owner of the frm) to a mult-stakeholder one n whch the frm owes fducary dutes to all ts stakeholders (the owners ncluded). It s obvous that classfcaton of stakeholders on the bass of the nature of ther relatonshp wth the frm must be regarded as mportant n gaugng these further fducary dutes The economc bases of the dea that the frm has further responsbltes towards ts stakeholders 4.1. Theory of the frm. Let us now nqure whether economc theory provdes support for the thess that the frm has further responsbltes towards ts stakeholders. Accordng to neo-nsttutonal theory (Wllamson 1975, 1986; Grossman and Hart 1986; Hart and Moore 1990; Hart 1995; Hansmann 1996), the frm emerges as an nsttutonal form of unfed transactons governance ntended to remedy mperfectons n the contracts that regulate exchange relatons among subjects endowed wth dverse assets (captal, labour, nstrumental goods, consumpton decsons, and so on). These assets, f used jontly, are able to generate a surplus over the cost of ther use that s hgher than n the case of ther separate use by each asset-holder. However, contracts by whch these asset-holders regulate ther exchanges are ncomplete: they do not nclude provsos coverng unforeseen events, owng to the costs of draftng them, or because the cogntve lmts of the human mnd make t mpossble to predct all possble states of the world. Yet for these assets to be used n the best manner possble, specfc nvestments must be made: nvestments undertaken wth a vew to the value that they may produce wthn a dosyncratc contractual relaton. Ths entals that the surplus generated wth respect to the costs sustaned by each party to the exchange s determned by the undertakng of specfc actvtes wth specfc counterparts (supplers, customers, employees, fnancers, etc.). Let us assume that partes behave opportunstcally (that s, they are egosts who act wth astuteness). Thus, once the nvestments have been made, contractual ncompleteness means that the terms of the contract can be renegotated, so that the party n a stronger ex post poston s able to approprate the entre surplus, thereby expropratng the other stakeholders. But f agents expect to be exproprated, they wll have no ncentve to undertake ther nvestments at the optmal level. Ths expectaton of unfar treatment gves rse to a loss of effcency at the socal level. 13 At frst sght, t mght be objected that many stakeholders, n both the strct and broad senses, do not have relatons wth a frm such that they formally delegate authorty to those who run t (for example, they do not vote), wth the consequence that the fducary dutes as defned earler do not apply to them. However, n the model of the socal contract as a hypothetcal explanaton of the orgn of the frm see secton 5.2 all the stakeholders partcpate n the frm s second socal contract, wth the consequence that ther trust consttutes the authorty of the frm s owner and manager. Ths also explans how the authorty of the latter may be accepted by these subjects. Moreover, the hypothetcal socal contract s typcally used to explan how authorty that s, legtmate power may come about at both the poltcal and organzatonal levels: cf. Green (1990), Raz (1985), Watt (1982). For a dscusson of manageral authorty see MacMahon (1989) and Saccon (1991). 8

9 Lorenzo Saccon, Corporate Socal Responsblty (CSR) as a model of extended corporate governance. The frm responds to ths problem by brngng the varous transactons under control of a herarchcal authorty the authorty, that s, of the party whch owns the frm and through ownershp s enttled to make decsons over the contngences that were not ex ante contractble. Unfed governance supplements ncomplete contracts wth authorty relatons through the vertcal and horzontal ntegraton of the unts that prevously made separate contrbutons. The frm s therefore a specal contractual form: when contracts lack provsos contngent upon unforeseen events, they can be completed wth the resdual rght of control whch enttles ts holder to decde what should be done about decsons not ex ante contractble that s, decsons left over from the orgnal contract and that become avalable only when unforeseen stuatons occur. The resdual rght of control underpns authorty: those partes enttled wth resdual rght of control may threaten the other partes to the contract wth excluson from the physcal assets of the frm, thereby ensurng that ex ante non-contracted decsons are taken ex post to ther own advantage. They are thus safeguarded aganst opportunsm by the other stakeholders, and they are able to protect the expected value of ther nvestments n stuatons where contract ncompleteness provdes margns of dscreton when resdual decsons have to be taken. There s therefore an effcency ratonale for the dea of the frm as unfed governance of transactons: f one party (a class of stakeholders) has made a specfc nvestment of greater mportance than those made by the others at rsk, or f ts exercse of unfed governance dscourages opportunsm by the others to approprate the surplus, then that party should be granted the property rght and wth t the rght to take resdual decsons. Ths s also the bass for regulaton of authorty delegaton from the owners to drectors or managers by corporate governance rules, when the owners themselves are not able of drectly exercsng the entre resdual rght of control. Fducary dutes owed to the owners must guarantee that delegated exercse of resdual rghts of control by the board of drectors or managers wll mantan or mprove the effcency of ther orgnal allocaton to the selected class of stakeholders The rsk of abuse of authorty. However, one should not underestmate the rsks of the frm qua unfed governance. There s not just one sngle stakeholder at rsk because of contract ncompleteness; t s usually the case that multple stakeholders undertake specfc nvestments (nvestments n human captal, nvestments of trust by consumers, nvestments of fnancal captal, nvestments by supplers n raw materals, technologes and nstrumental goods). Contracts wth these stakeholders are also ncomplete. Yet f a frm brngs ts contracts wth certan stakeholders (labour contracts, oblgatons towards and relatons wth mnorty shareholders) under the authorty of a party to whom s allocated control over resdual decsons (for example, the controllng shareholder group) and more generally f a party s enabled by ts de facto power to exercse dscreton over ex ante non-contractble decsons concernng mplct or explct contractual relatons wth the other stakeholders (consumers, customers, supplers, credtors, etc.) what, one may ask, s there to ensure protecton of nvestments and nterests other than those of the controllng stakeholder? It s evdent that f fducary dutes attach only to ownershp, those 9

10 Luc Papers n. 142, suppl. a febbrao 2004 stakeholders wthout resdual rght of control wll not be protected by the fducary dutes of those who run the frm. The nherent rsk, therefore, s an abuse of authorty (Saccon 1997, 2000). Those weldng authorty may use t to exproprate the specfc nvestments of others by explotng gaps n contracts whch persst even under unfed governance (n fact t smply allocates to only one stakeholder the rght to fll those gaps wth ts dscretonary decsons). Those n a poston of authorty, n fact, are able to threaten the other stakeholders wth excluson from access to physcal assets of the frm, or from the benefts of the contract, to the pont that those other stakeholders become ndfferent between acceptng the expropraton and forgong the value of ther nvestments by wthdrawng from the relaton. Thus the entre surplus, ncluded that part of t mputable to efforts and nvestments made by the non controllng stakeholders, wll be approprated by the controllng party. Agan forward-lookng stakeholders wll be deterred form enterng the herarchcal transacton wth the controllng party. In general, ths wll produce a nternal crss of legtmacy between frm and stakeholders (a crss n the relatonshps between the organzatonal authortes and partcpants n the organzaton) and an external crss of trust (n relatonshps wth stakeholders that have entered nto contractual or external relatons wth the organzaton). Varous stakeholders wll ex ante have a reduced ncentve to nvest (f they foresee the rsk of abuse), whle ex post they wll resort to conflctng or dsloyal behavour (typcally possble when asymmetry of nformaton s nherent n the executon of some subordnate actvty) n the belef that they are beng subjected to abuse of authorty. In the economst s jargon, ths s a second best state of affars (less than optmum): all governance solutons based on the allocaton of property rghts to a sngle party may approxmate socal effcency, but they can never fully acheve t. Ths much s acknowledged by the theoretcans of contractual ncompleteness when they pont out that the allocaton of the resdual rght of control nduces the party protected by that rght to over-nvest, whle those not so protected are nduced to under-nvest, wth a consequent shortfall wth regard to the socal optmum (Grossman and Hart 1986; Hart 1995). On the other hand, f the stakeholder category enttled to exercse ownershp (the double rght of controllng resdual decsons and clamng resdual revenue; cf. Hansmann 1987, 1996) s selected on the bass of ts ablty to mnmze total costs dervng form the summaton of contractual costs borne by of varous stakeholders and costs of exercsng authorty, t s by no means certan that a soluton wll be found whch reduces each of those costs to the mnmum (that s, reduces opportunsm suffered by each stakeholder to the mnmum). Suffcent for ths soluton to emerge s, for example, that the governance costs of one class (the captal-holders, for example) are low enough to counterbalance a relatve ncrease n the contractual costs borne by another class (the workers, for example) compared to alternatve cases (for nstance the case n whch there s no centralzed governance, or the one n whch t s a sub-set of workers that governs, or the consumers). In ths case, too, some ncentves are nullfed, whch dstances the real-world soluton from complete (Paretan) socal effcency. The fact s that the relatve (n)effcency depends on manfest or smply expected unfarness: separaton between effcency and farness (a myth of neoclasscal economcs) s no longer feasble when we face the real-lfe problem of workng out acceptable soluton for the governance of transactons. 10

11 Lorenzo Saccon, Corporate Socal Responsblty (CSR) as a model of extended corporate governance. My suggeston s therefore that when CSR s vewed as extended governance, t completes the frm as an nsttuton of transactons governance (cf. Saccon 2000). The frm s legtmacy defct (whatever category of stakeholders s placed n control of t) s remeded f the resdual control rght s accompaned by further fducary dutes towards the subjects at rsk of abuse of authorty and deprved of the resdual control rght. At the same tme, ths s a move towards greater socal effcency because t reduces the dsncentves and socal costs generated by the abuse of authorty. From ths perspectve, extended governance should comprse: the resdual control rght (ownershp) allocated to the stakeholder wth the largest nvestments at rsk and wth relatvely low governance costs, as well as the rght to delegate authorty to professonal drectors and management; the fducary dutes of those who effectvely run the frm (admnstrators and managers) towards the owners, gven that these have delegated control to them; the fducary dutes of those n a poston of authorty n the frm (the owner or the managers) towards the non-controllng stakeholders: the oblgaton, that s, to run the frm n a manner such that these stakeholders are not deprved of ther far shares of the surplus produced from ther specfc nvestments, and that they are not subject to negatve externaltes. 14 A number of recent economc and legal models of governance support ths vew of CSR. For example, the frm can be seen as a nexus of specfc nvestments regulated by ncomplete contracts, rather than as a nexus of smple contracts, and therefore as a team of actors cooperatng to produce a surplus from those specfc nvestments (Rajan and Zngales 2000). Based on a smlar vew whch combnes dfferent theores of the frm the theory of ncomplete contracts wth that of team producton s the model of mult-stakeholder governance developed by Margaret Blar and Lynn Stout, and whch sees the purpose of corporate governance structures as beng preventon of opportunstc behavour among the N members of the team that make specfc nvestments. When appled to a publc company, ths model translates nto a board of drectors actng as a medatng herarchy: an authorty system charged wth the task of fndng 14 I have proposed n a prevous work (cf. Saccon 1991) a vew of manageral ethcs based on a smlar analyss of the theory of frm, as well as on the cooperatve game theory of the frm put forward by Masharo Aok (Aok 1984) and the noton of extended fducary dutes (cf. Saccon 2000). Surprsngly, Gudo Ross (Ross 2003, pp ), after polemcally descrbng busness ethcs as tampon ethcs, dsmsses the dea that governance of a frm should comprse fducary dutes towards multple stakeholders as a varant of the conflct of nterest epdemc (cf. Ross 2003, 124). He thus confuses medaton by the trustee among the dfferent trustors legtmate nterests wth a pathologcal conflct between the trustor s nterests, whch the trustee s oblged to pursue, and the personal ones of the trustee hmself. Ths confuson so that a board of drectors whch seeks to medate between shareholders value and other objectves would undergo to a conflct of nterest (hardly a progressve thess) may be due to Ross s gnorance of recent developments n the theory of the frm whch at least suggest that the problem should be taken serously. The contenton whch sprngs precsely from nternal crtcsm of neo-nsttutonalst theory that frms have responsbltes towards ther stakeholders (what I would call a reformst vew) may be tampon ethcs, but Ross s merely recyclng Mlton Fredman s thess that the only socal responsblty of busness s to make profts when he appears to accept the captalsm of shareholder value wthout any tampon at all, menacng anyone who argues to the contrary wth the same fate as the Wemar Republc (bd. pp ). 11

12 Luc Papers n. 142, suppl. a febbrao 2004 the approprate balance n the protecton of dverse nterests (cf. Blar, Stout 1999). The (controversal) legal bass for ths form of mpartal governance exercsed by the board of drectors and by management n the US jont-stock company s the busness judgment doctrne : the manager s use of a standard of professonal conduct whch nsulates hs/her choces aganst clams by shareholders (cf. Blar, Stout 1999, but also see Meese 2002). However, a number of unanswered questons reman whch the proponent of CSR as extended governance must necessarly address. Does there exst a crteron wth whch to gve more precse specfcaton to these extended dutes, and from whch t s possble to derve a strategc management standard of suffcent clarty such that the extended governance model cannot be accused to ental hgher governance costs than the tradtonal narrow corporate governance vew? What norms are effectve for the mplementaton of CSR? What s the role of company law wth respect to other parts of the law that mpose constrants on corporate behavour? And what role can be played by self-regulaton? 5. How can we dentfy CSR dutes (wthout arbtrarness)? 5.1. The Socal Contract as a crteron for strategc management. If a frm s a team of partcpants wth specfc nvestments, then the metaphor of a barganng cooperatve game among multple stakeholders can be appled. These stakeholders must agree on a shared acton plan (a jont strategy) whch allocates tasks among the members of the team so that the contrbuton of each of them s effcent (because t produces the maxmum surplus net of each stakeholder s costs). The barganng cooperatve game played by the stakeholders s typcally one of mxed nterests. Although t s n ther common nterest to cooperate, because ths enables them to produce a surplus that would otherwse be mpossble, conflct nevertheless perssts among the stakeholders over the dstrbuton of the value created. Governance and strategc management consequently consst n the soluton of two problems: a) Identfyng the jont strategy that the stakeholders (as the players n the cooperatve game) may utlze to coordnate themselves, n that they accept t ex ante as a voluntary agreement to cooperate so that strategc management can reduce barganng costs (tme, conflct, etc.) and the costs of gatherng nformaton on the alternatves avalable and on the ntentons of each players about cooperaton. b) Ensurng ex post that each member of the team comples wth the agreement on the jont strategy selected and does not act as a free rder wth regard to the others. Choosng the jont strategy (pont a) s equvalent to select a barganng equlbrum. It must therefore answer the queston of what s due to each stakeholder and what each of them can expect from the frm n exchange for ts contrbuton, so that each stakeholder may agree on that jont strategy. The queston thus arses as to how the stakeholders nterests can be balanced aganst each other, and what clams on the frm should be consdered the approprate bass for the management s fducary dutes. Stakeholder, n fact, s a descrptve term. It remnds us that a varety of classes of ndvduals have nterests at stake n 12

13 Lorenzo Saccon, Corporate Socal Responsblty (CSR) as a model of extended corporate governance. the runnng of the frm, and that they may sometmes advance conflctng clams. The use of the term stakeholder, however, does not provde a crteron wth whch to balance clams when they are mutually conflctng. To answer the queston we consequently need a crteron able to dentfy the balance that any whatever stakeholder would accept as the bass for ts voluntary cooperaton wth the frm: that s, an mpartal crteron. It s here that ethcs understood as a set of mpartal crtera for collectve chocemakng come nto play as part of the frm s governance and strategc management. As an ethcal crteron, therefore, I suggest the socal contract among the stakeholders of the frm (Saccon 1997, 2000). By socal contract I mean not any whatever real-lfe bargan but a touchstone from whch pont of vew to assess the dverse outcomes of day by day practcal runnng of the frm. In other words, the socal contract s the agreement that would be reached by the representatves of all the frm s stakeholders n a hypothetcal stuaton of mpartal choce. 15 Correspondng to the noton of socal contract s the followng mult-stage delberatve procedure whch generates mpartally acceptable agreements. Force, fraud and manpulaton must be set asde. Each party comes to the barganng table wth only ts capacty to contrbute and ts assessment of the utlty of each agreement or non-agreement proposed (dspensng wth any form of threat other than ts possble refusal to agree). The barganng status quo must be set at a level such that each stakeholder results mmune aganst the cost of ts specfc nvestments that s, each stakeholder must obtan from the socal contract at least rembursement of the cost of the specfc nvestment wth whch t has contrbuted to the surplus (otherwse the barganng process would permt opportunstc explotaton of the counterparty s lock-n stuaton). The dstrbuton of the surplus s regulated by the socal contract and by the correspondng delberatve procedure on the bass of ntal endowments thus defned. v Each party n turn puts tself n the poston of all the others, and n the poston of each of them he can accept or reject the contractual alternatves proposed. v If solutons are found whch are acceptable to some stakeholders but not to others, these solutons must be dscarded and the procedure repeated (whch reflects the assumpton that cooperaton by all stakeholders s recognzed as necessary). 15 It s qute evdent the debt of ths contractaran vew on the theory of frm to the works of both John Rawls (1971, 1993) and Davd Gauther (1986). For the frst formulaton of the theory of the corporate socal contract, based the revson of neo-nsttutonalst theory of frm and wth reference to the problem of the abuse of authorty vs-à-vs stakeholders, see however Saccon (1991), and latterly Saccon (1997, 2000). For a formulaton external to economc theory see Donaldson (1982) and then Dunfee and Donaldson (1995). The socal contract ethcal theores - n both the deal (Kantan) and real (Hobbesan) versons has been put forward as keys to nterpretaton of the abuse of majorty power n jont-stock companes by Dsano Prete (Prete 1992). 13

14 Luc Papers n. 142, suppl. a febbrao 2004 v The terms of the agreement reached are therefore those that each stakeholder s wllng to accept from ts partcular pont of vew: that s, the non-empty ntersecton of the jont strateges and relatve dstrbutons acceptable to each of them. Note that ths ntersecton s necessarly non-empty, for otherwse the game would not allow a cooperatve surplus. That s to say, t would not be the case that jont acton by the partes may produce somethng more than ther separate acton and that at least one surplus dstrbuton proves to be recprocally advantageous (f t must be so, then there exsts at least one agreement acceptable to all) Socal contract and the emergence of the frm. Thus far, the socal contract has been presented as a normatve delberatve procedure by whch to dentfy the terms of an agreement that would be acceptable from an mpartal standpont that s, from the pont of vew of any whatever stakeholder so that t can be adopted as a standard of behavour by, for example, the medatng herarchy proposed by Blar and Stout. However, the socal contract can also furnsh a reconstructon understood as a potental explanaton of how barganng has gven rse to a frm wth both fducary dutes towards the owners and socal responsblty (.e. further fducary dutes) towards all the stakeholders. Consder a state of nature pror to the creaton of the frm. Blateral transactons among stakeholders regulated by ncomplete contracts are subject to recprocal opportunstc behavour, wth the consequence that prohbtve barganng costs render them neffcent. At the same tme, the partes to those transactons are entrely unconcerned about the negatve external effects of ther transactons on other agents, who although they do not partcpate, are nevertheless affected. Ths s a Hobbesan scenaro n whch the lfe of economc transactons among agents s soltary, poor, nasty, brutsh, and short. 16 The stakeholders thus address the problem of creatng an assocaton whereby all ther transactons can be undertaken n accordance wth agreed-to rules and are therefore not subject to contract-costs, whle at the same tme the negatve effects on those who do not partcpate n the benefts from the transactons are reduced to the mnmum. The Frst Socal Contract of the frm (pactum unons) s nothng other than the agreement whch the stakeholders reach among themselves to set up ths assocaton (the just frm ). They negotate on the assocaton s consttuton, whch conssts n a common plan of acton (jont strategy) to whch each of them contrbutes ether by carryng out a postve effort or by smply refranng from applyng hs/her veto. Ths frst socal contract of the frm stpulates as follows: a) rejecton of shared plans of acton whch generate negatve externaltes for those not partcpatng n the cooperatve venture or, f these negatve externaltes are essental for the producton of the cooperatve surplus, a compensaton of thrd partes so that they are rendered neutral; 16 Cf. Hobbes, Levathan, (1651), part 1, chapter

15 Lorenzo Saccon, Corporate Socal Responsblty (CSR) as a model of extended corporate governance. b) producton of the maxmum surplus possble (dfference between the value of the product for ts consumers, who belong to the assocaton, and the costs sustaned by each stakeholder to produce t); c) a dstrbuton of the surplus whch s far, or ratonally acceptable to each stakeholder n a barganng process free from force or fraud and based on an equtable status quo, that s, consderng the surplus net of the specfc nvestments. However, f an attempt s made to reach ths form of an deal assocaton (the just frm ) whch elmnates all the partcpants contract-costs, they arrve n practce to an organsatonal form whch s found to be neffcent from the pont of vew of ts governance costs. The stakeholders dscover, for example, that the general assembly of all members s unable to take coherent decsons n a reasonable amount of tme. In the absence of a montorng system, once the members of the assocaton have establshed far shares of the surplus to be dstrbuted among them, they have an ncentve to act opportunstcally and not to play ther part. Coordnaton problems arse on how the jont strategy can be mplemented under changng crcumstances, whch may alter belefs and recprocal expectatons asymmetrcally. The stakeholders consequently draw up a second socal contract of the frm (pactum subjectons) 17 by whch they consttute, n the proper sense of the term, a governance structure for the assocaton. It s only now that the assocaton becomes a herarchcal structure. The second socal contract provdes that authorty should be delegated to the stakeholder most effcent n performng governance functons (the takng of resdual decsons, devsng coordnaton solutons as crcumstances change, montorng, the enactment of sanctons, excludng potental free rders, etc.). For ths reason, t can also be seen as a contract between the stakeholders and those who s gven control over the frm (socal contract wth the frm). After comparatve examnaton of the governance costs of each stakeholder, the one wth the lowest costs s selected and assgned ownershp, and s therefore the one to whch the rght of governng the assocaton s delegated (Hansmann 1996). Ths class, whch s remunerated wth the resdual s authorsed to delegate some dscretonary decsons n regard to runnng the frm to professonal drector and managers, and to appont those who are n the authorty poston of runnng the frm. Prma face, ther authorty wll be effectvely consttuted that s, the delegaton wll reman vald as long as they comply wth what I call Narrow fducary provso: the owners are remunerated wth the maxmum resdual revenue possble (n forms compatble wth the dverse nature of the controllng stakeholder: profts, returns, dscounts, mproved condtons of servce, mproved condtons of employment, and so on) n the lght of condtons obtanng n the frm s specfc market. However, t s evdent that ths provso entals that the postons of the other stakeholders change (from the just frm to just a frm). Formerly co-equal members of the assocaton, they are now subject n varous ways to the dscretonary decsons taken by the stakeholder enttled wth authorty, and by the admnstrators that t has apponted. Unlke n the standard economc theory of the frm, n the socal 17 Interestngly, also Blar and Stout (1999) adopt the analogy between the frm and the two socal contracts typcal of the socal contract tradton. 15

16 Luc Papers n. 142, suppl. a febbrao 2004 contract theory the rsk of the abuse of authorty can squarely be faced. The second socal contract s therefore conceved n a manner such that ths cost of herarchy s forestalled as well. Hence, under the second socal contract, the stakeholders agree to submt to authorty, thereby renderng t effectve, f the contract contans the whch stpulates that the frm s new governance structure must comply wth fducary dutes towards all the stakeholders (owners and non-owners). Extended fducary provso: Towards the non-owners The frm must abstan from actvtes whch mpose negatve external effects on stakeholders not party to transactons, or compensate them so that they reman neutral; The frm must remunerate the stakeholders partcpatng n the frm s transactons wth payoffs (monetary or of other knds, for example n terms of the quantty, qualty and prces of goods, servces, workng condtons, etc.) whch, taken for granted a far status quo, must contan a part ted to the frm s economc performance such to approxmate far/effcent shares of the surplus (assumng that ths s postve) as envsaged by the frst socal contract. 18 Towards the owners: The frm must remunerate the owners wth the maxmum resdual compatble wth far remuneraton as defned by the frst socal contract of the effcent contrbutons made by all the other stakeholders The Socetary nterest of the company and extended fducary dutes. What does ths hypothetcal explanaton yeld? It yelds a defnton of the socetary nterest of the company that s, the nterest that the manager actng n the name of the company must serve whch s consstent wth the contractaran model. Accordng to ths reconstructon, n fact, the manager (apponted through the second socal contract) has a specal fducary duty towards the owners (or the resdual clamant ) that has delegated authorty to hm/her (va narrow fducary provso). Ths duty apples, however, only under the constrant that the general fducary dutes are fulflled towards all the stakeholders whch s defned va the extended fducary provso. We may thus construct the corporate nterest by means of a herarchcal decson-makng procedure whch moves from the most general condtons to the most specfc ones: Frst step: mnmze the negatve externaltes affectng stakeholders n the broad sense (perhaps by payng sutable compensaton); Second step: dentfy the agreements compatble wth the maxmzaton of the jont surplus and ts smultaneous far dstrbuton, as establshed by the mpartal cooperatve agreement among the stakeholders n the strct sense; 18 Note that meant here s remuneraton n utlty and not necessarly n money. Put n economc parlance, ths remuneraton conssts of the consumer rent, the producer rent, the worker rent and so on, accrung to each of them from the frm s transactons. Ths means that some stakeholders may not want to receve monetary benefts from the frm, but rather mprovements n workng condtons or n purchasng power, n the qualty of goods and servces, of contractual condtons, etc., to whch the shares of the surplus are n any case devoted. 16