Abstract
Other-regarding preferences have been documented in many strategic settings. We provide a model in which the managers of firms in an oligopoly have preferences for both consumer welfare and own income. We find that profit sharing can function as a facilitating practice. Managers must receive a sufficiently large share of profits for collusion to be sustained, and the optimal collusive price increases with the degree of profit sharing. Thus, restrictions on performance-based compensation may be consistent with the objectives of antitrust policy. We also find that an increase in industry concentration can harm consumers even if the firms were already successfully colluding.
Similar content being viewed by others
Notes
See Cooper and Kagel (forthcoming) for a survey of the experimental work on other-regarding preferences.
In principle, other-regarding preferences are not inconsistent with an economic approach that, according to Becker, “assumes that individuals maximize welfare as they conceive it, whether they be selfish, altruistic, loyal, spiteful, or masochistic” (1993, p. 386).
Engelmann and Muller make a similar point: “...using simulated buyers facilitates collusion as sellers with other-regarding preferences might refrain from colluding if buyers become substantially worse off” (2011, p. 293).
This disutility can be selfish rather than altruistic. For example, Benabou and Tirole (2006) analyze a model in which prosocial behavior may be both intrinsic as well as reputational.
Spagnolo (2000) examines an oligopoly model with self-interested managers. He shows that stock-based compensation makes collusion easier to sustain relative to profit-sharing. This is “because the value of the shares of a firm that deviates from a collusive agreement does not increase as much as short-run profits in the period in which the deviation occurs.”(p. 28) An other-regarding manager who receives equity compensation continues to face a tradeoff between own consumption and consumer surplus since the market value of the firm ultimately depends on present and expected future profits. Therefore, allowing managers to receive equity compensation would not change the basic insights of our model, although the incorporation of a stock market would reduce the tractability of our model.
Hay (2008) defines facilitating practices as “actions taken by would-be competitors in a market to enable them to better coordinate their actions.”
Some of the factors that have been considered are uncertain demand (Green and Porter 1984; Rotemberg and Saloner 1986), multi-market contact (Bernheim and Whinston 1990), licensing contracts (Lin 1996), team size (Bornstein et al. 2008), and focal points (Knittel and Stango 2003; Engelmann and Muller 2011).
See also Marini and Zevi (2011) who study the effects of consumer cooperatives in mixed markets.
We assume all subsequent objective functions are quasiconcave so that all maximization problems yield unique solutions. A sufficient condition for quasiconcavity of these objectives is that the ratio \(D'( P)/D( P)\) is decreasing. If consumer’s have preferences of the form \(V_{i}(\theta _{i}, P)= \theta _{i}- P\), where \(\theta _{i}\) is consumer \(i\)’s willingness to pay for the product and is distributed according to \(F(\theta _{i})\), then it is sufficient to assume that \(F(\theta _{i})\) has an increasing hazard rate, which is a common assumption.
Engel (2011) performs a meta-analysis of 129 dictator game experiments conducted between 1992 and 2010. In a dictator game, one player dictates the transfer of wealth from him or herself to a recipient. Regardless of whether participants are known or anonymous or considered more or less worthy of generosity, contributions to recipients are non-zero in fully 99 % of treatments conducted. There is also evidence from the field that individuals derive utility from transfers to others, even if they do not know them. For example, Chuan and Samak (2013) conduct a field experiment with an Illinois charity and find that donors give the same amount regardless of social distance. Harbaugh et al. (2007) find in a neurological analysis that neural activity consistent with pleasure increases when people give voluntarily.
As discussed in the Appendix, we assume that (5) holds strictly for self-interested managers, which is equivalent to assuming \(N( {1-\delta })<1\).
In general, it is not possible to rule out \(\bar{P}= \infty \). However, a sufficient condition for \(\bar{P}\) to be finite is that \(D( {P_0 })=0\) for some \(P_0 >0\).
Of course, a priori there is no reason to assume that owners are purely self-interested. Further, our model allows for owner-managers who receive all of the firm’s profit (that is, \(B\) = 1).
References
Andreoni J, Erard B, Feinstein J (1998) Tax compliance. Journal of Economic Literature 36:818–860
Armstrong M, Steffen H (2011) Behavioral economics as applied to firms: a primer. Antitrust Chronicle 1.
Bandiera O, Barankay I, Rasul I (2005) Social preferences and the response to incentives: evidence from personnel data. Q J Econ 120:917–962
Bandiera O, Barankay I, Rasul I (2010) Social incentives in the workplace. Rev Econ Stud 77:417–458
Becker G (1993) Nobel lecture: the economic way of looking at behavior. J Polit Econ 101:385–386
Benabou R, Tirole J (2006) Incentives and prosocial behavior. Am Econ Rev 96:1652–1678
Bernheim BD, Whinston M (1990) Multimarket contact and collusive behavior. RAND J Econ 21:1–26
Blanco M, Engelmann D, Normann H (2011) A within-subject analysis of other-regarding preferences. Games Econ Behav 72:321–338
Bolton G, Ockenfels A (2000) ERC: a theory of equity, reciprocity, and competition. Am Econ Rev 90:166–193
Bolton G, Katok E, Zwick R (1998) Dictator game giving: rules of fairness versus acts of kindness. Int J Game Theory 27:269–299
Bornstein G, Kugler T, Budescu DV, Selten R (2008) Repeated price competition between individuals and between teams. J Econ Behav Organ 66:808–821
Bruner D, McKee M, Santore R (2008) Hand in the cookie jar: an experimental investigation of equity-based compensation and managerial fraud. South Econ J 75:261–278
Chaudhuri A (2011) Sustaining cooperation in laboratory public goods experiments: a selective survey of the literature. Exp Econ 14:47–83
Chuan A, Samak AC (2013) Feel the warmth’ glow: a field experiment on manipulating the act of giving. Working paper. Available at http://www.spihub.org/site/resource_files/workingpapers/spi_wp_003_Chuan_Feel_the_Warmth_2013_03_01.pdf
Cooper D, Kagel J (forthcoming) Other regarding preferences: a selective survey of experimental results. In: Kagel J, Roth A (eds) The Handbook of Experimental Economics, vol 2.
Engel C (2007) How much collusion? A meta-analysis of oligopoly experiments. J Compet Law Econ 3(4):491–549
Engel C (2011) Dictator games: a meta study. Exp Econ 14:583–610
Engelmann D, Muller W (2011) Collusion through price ceilings? In search of a focal-point effect. J Econ Behav Organ 79:291–302
Fehr E, Schmidt KM (1999) A theory of fairness, competition and cooperation. Q J Econ 114:817–868
Fehr E, Fischbacher U (2002) Why social preferences matter—the impact of non-selfish motives on competition, cooperation, and incentives. Econ J 112:1–33
Fershtman C, Judd K (1987) Equilibrium incentives in oligopoly. Am Econ Rev 77:927–940
Goldman E, Slezak SL (2006) An equilibrium model of incentive contracts in the presence of information manipulation. J Financ Econ 80:603–626
Green E, Porter R (1984) Noncooperative collusion under imperfect price information. Econometrica 52:87–100
Harbaugh WT, Mayr U, Burghart DR (2007) Neural responses to taxation and voluntary giving reveal motives for charitable donations. Science 316:1622–1625
Hay G (2008) Facilitating practices. Issues Compet Law Policy 2:1189–1218
Hernandez P, Minor D, Sisak D (2011) Social preferences or collusion? Experimental evidence. Working Paper.
Ledyard JO (1995) Public goods: a survey of experimental research. In: Kagel J, Roth A (eds) Handbook of Experimental Economics. Princeton University Press, Princeton, pp 111–194
Lin P (1996) Fixed-fee licensing of innovations and collusion. J Ind Econ 44:443–449
Knittel C, Stango V (2003) Price ceilings as focal points for tacit collusion: evidence from credit cards. Am Econ Rev 93:1703–1729
Marini M, Zevi A (2011) Just one of us: consumers playing oligopoly in mixed markets. J Econ 104:239–263
Peng L, Roell AA (2008) Manipulation and equity-based compensation. Am Econ Rev 98:285–290
Philipson T, Posner R (2009) Antitrust in the not-for-profit sector. J Law Econ 52:1–18
Rotemberg JJ, Saloner G (1986) A supergame-theoretic model of price wars during booms. Am Econ Rev 76:390–407
Skivlas S (1987) The strategic choice of managerial incentives. RAND J Econ 18:452–458
Sobel J (2009) Generous actors, selfish actions: markets with other-regarding preferences. Int Rev Econ 56:3–16
Spagnolo G (2000) Stock-related compensation and product-market competition. RAND J Econ 31:22–42
Vesterlund L (2012) Voluntary giving to public goods: moving beyond the linear VCM. Available at: http://www.isites.harvard.edu/fs/docs/icb.topic1002778.files/LVchapter.pdf
Willner J (2013) The welfare impact of a managerial oligopoly with an altruistic firm. J Econ 109:97–115
Acknowledgments
We wish to thank the Editor and three anonymous referees for their very helpful comments.
Author information
Authors and Affiliations
Corresponding author
Appendix: Proofs of the Results
Appendix: Proofs of the Results
Proof of Proposition 1: First observe that \({\Pi {'}}( P)=( {P-k})D {'}( P)+D(P)\) and \(CS {'}( P)=-D( P)\). It follows that \(\left. {\frac{du_i ( P)}{dP}} \right| _{P=k} =\frac{B{\Pi } {'}(k)}{N}+\lambda CS'( k)=\frac{B}{N}D(k)-\lambda D(k)\). Given the quasiconcavity of \(u_i ( P)\), it follows that the optimal collusive price \(P^C\) is greater than \(k\) if and only if \(\left. {\frac{du_i ( P)}{dP}} \right| _{P=k} =\left[ {\frac{B}{N}-\lambda } \right] D( k)>0\) which is true if and only if \(\frac{B}{N}>\lambda \).\(\square \)
Proof of Lemma 1: Since \(CS'( P)<0\), (4) implies \({\Pi '}( {P^U})>0\). It then follows from (2) and the quasiconcavity of \({\Pi }( P)\) that \(P^{U}<P^{\uppi }\). Since \(N >\) 1 and \({\Pi '}( {P^U})>0\), (4) and (3) imply \(\left( {\frac{B}{N}}\right) {\Pi {'}}( {P^U})+\lambda CS'( {P^U})<0\). Thus, (3) and the quasiconcavity of \(u_i ( P)\) implies \(P^U>P^C\).\(\square \)
Proof of Lemma 2: First, it is clearly never optimal for a defecting manager to charge a price greater than the collusive price \(P^*\). For collusion at \(P^*>P^U\), the best defection price is \(P^U\) since, by definition, \(P^U\) maximizes \(\left[ {B{\Pi }( P)+\lambda CS( P)} \right] \), which is the manager’s utility when all other firms charge higher prices. For any price \(P<P^U\) we know that \(\left[ {B{\Pi }( P)+\lambda CS( P)} \right] \) is increasing in \(P\). Therefore, if the collusive price is \(P^*\le P^U\), a manager that defects should charge a price just under \(P^*\).
Now suppose that collusion at \(P^C\) is not sustainable. We begin by showing that collusion at \(P^*\in (P^C,P^U]\) cannot be sustained. Thus, for \(P^*\in (P^C,P^U]\) we need to show
By the definition of \(P^C\), we have
for any \(P^*\ne P^C\). By (5), if \(P^C\) is not sustainable, then
The quasiconcavity of \(B{\Pi }( P)+\lambda CS( P)\) implies that it is increasing in \(P\) over the range \((P^C,P^U]\). It follows that
for \(P^*\in (P^C,P^U]\). Clearly, (7), (8), (9) imply (6) which is what we needed to show. Thus, collusion cannot be sustained at \(P^*\in (P^C,P^U]\).
We now show that collusion cannot be sustained at any \(P^*>P^U\). For \(P^*>P^U\) a defecting manager will choose \(P^U\), so we need to show
By the definition of \(P^U\),
since \(P^C\ne P^U\), by Lemma 1. Recall that by (5), if \(P^C\) is not sustainable, then
And, by the definition of \(P^C\), we know
since \(P^*\ne P^C\). Clearly, (11), (12), and (13) imply (10). Thus, collusion cannot be sustained at \(P^*>P^U\).\(\square \)
For the remaining proofs it will be convenient to define the following slack function:
By construction, \(\sigma ( {P,\lambda ,B,N})\) is positive if and only if (5) holds strictly. Thus, in order for collusion to be sustained at some \(P^*\le P^U\) we must have \(\sigma ( {P^*,\lambda ,B,N})\ge 0\). Observe also that in order to have \(\sigma ( {P,\lambda ,B,N})>0\) for \(P>k\) and \(\lambda \ge 0\) we must have \(N( {1-\delta })<1\), and this assumption is maintained in all that follows to allow easier exposition of the results.
As discussed in the text, let \(\bar{P}\equiv \hbox {Max}\left\{ {P: \sigma (P,\lambda ,B,N)\ge 0} \right\} \) denote the highest price that causes (5) to hold with equality, so that
As discussed in footnote 11, we assume that \(\bar{P}\) is finite, which be true under very mild conditions. Finally, note that we must have
otherwise there would exist a \(P'>\bar{P}\) such that \(\sigma ( {P{'},\lambda ,B,N})>0\) contradicting the definition of \(\bar{P}\).
Proof of Lemma 3: Given that \(CS( P)<CS( k)\) for \(P>k\), it follows that \(\sigma ( {P,\lambda ,B,N})<\sigma ( {P,0,B,N})\) for all \(\lambda >0\) and \(P>k\).\(\square \)
Proof of Proposition 2: We first show the “if” part. By definition of the slack function, \(\sigma ( {P^*,\lambda ,B,N})\ge 0\) is sufficient for collusion to be sustainable at some \(P^*\in (k,P^C]\). Since \(\sigma ( {k,\lambda ,B,N})=0\), it is sufficient to show \(\left. {\frac{\partial \sigma (P,\lambda ,B,N)}{\partial P}} \right| _{P=k} >0\). Using the fact that \({\Pi '}( P)=( {P-k})D{'}( P)+D(P)\) and \(CS{'}( P)=-D( P)\) we have
Therefore, \(\lambda <\frac{B}{N}\left( {\frac{1-N(1-\delta )}{\delta }}\right) \) is sufficient for collusion to be sustainable at some \(P^*\in (k,P^C]\).
Next we show the “only if” part. Suppose that \(\lambda \ge \frac{B}{N}\left( {\frac{1-N(1-\delta )}{\delta }}\right) \). From the preceding paragraph we have
Moreover, for any \(P>k\) we have
since \(D{'}( P)<0\). Together with \(\sigma ( {k,\lambda ,B,N})=0\), it follows that \(\sigma ( {P^*,\lambda ,B,N})<0\) for any \(P^*>k\), which means collusion is not sustainable for any \(P^*\in (k,P^C]\). Therefore, by Lemma 2, since collusion at \(P^C\) cannot be sustained, collusion at \(P^*>P^C\) cannot be sustained. \(\square \)
Proof of Proposition 4:
Part (i). We show that both \(P^C\) and \(\bar{P}\) are decreasing in \(\lambda \). First, implicit differentiation of (3) yields
where the sufficient second-order condition implies that the denominator is negative. Second, using arguments similar to the ones used in the proof of Proposition 2, we can show that \(\lambda <\frac{B}{N}\left( {\frac{1-N(1-\delta )}{\delta }}\right) \) implies \(\bar{P}>k\). Implicit differentiation of (14) yields
where the sign of the denominator follows from the fact that \(\sigma ({P,\lambda ,B,N})\) must be decreasing at \(\bar{P}\).
Part (ii). We show that both \(P^C\) and \(\bar{P}\) are increasing in the bonus. Implicit differentiation of (3) yields
where the numerator is negative since \(P^C<P^{\uppi }\), by Lemma 1, and the sufficient second-order condition implies that the denominator is negative.
Now let \(\bar{P}( B)\) be the \(\bar{P}\) corresponding to \(B\), and let \(B{'}>B\). We know \(\sigma (P,\lambda ,B,N)\) increases in \(B\) when \(P^*>k\),
The fact that \(\sigma ( {\bar{P} (B),\lambda ,B',N})>0\) implies \(\bar{P}( {B'})>\bar{P} (B)\).
Part (iii). We show that both \(P^C\) and \(\bar{P}\) are decreasing in \(N\). First, implicit differentiation of (3) yields
where \({\Pi {'}}( {P^C})>0\) since \(P^C<P^{\uppi }\), by Lemma 1, and the sufficient second-order condition implies that the denominator is negative. Second, the condition \(\lambda <\frac{B}{N}\left( {\frac{1-N(1-\delta )}{\delta }}\right) \) implies we must have \(\bar{P}>k\). Implicit differentiation of (14) yields
where the sign of the denominator follows from the fact that \(\sigma ( {P,\lambda ,B,N})\) must be decreasing at \(\bar{P}\). \(\square \)
Rights and permissions
About this article
Cite this article
Santore, R., Li, Y. & Cotten, S.J. Colluding with a conscience. J Econ 114, 255–269 (2015). https://doi.org/10.1007/s00712-014-0390-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00712-014-0390-8