Abstract
We consider a market comprising a number of perfectly complementary and homogeneous commodities. We concentrate on the incentives for firms producing these commodities to merge and form a vertical syndicate. The main result establishes that the nucleolus of the associated market game corresponds to the unique vector of prices with the following properties: (i) they are vertical syndicationproof, (ii) they are competitive, (iii) they yield the average of the buyers and the sellersoptimal allocations in bilateral markets, and (iv) they depend on the traders’ bargaining power but not on their identity. The proof uses an isomorphism between our class of market games and the entire class of bankruptcy games.
Similar content being viewed by others
Notes
We refer to Schmalensee (2007) for the vast industrial organization literature that deals with the incentives for the creation and/or destruction of syndicates.
The assumption that technology markets have only three sectors is a defensible simplification, despite the fact that intermediary firms also participate in them.
Economides (2001) analyzes the case of “US against Microsoft”.
See http://www.reuters.com/article/2015/04/27/useugoogleantitrustidUSKBN0NF1YX20150427, retrieved on 3July2015.
See http://www.bbc.com/news/business18164190, retrieved on 29June2015.
We do not claim that the nucleolus and the properties characterizing are the only reasonable possibilities.
The impact of buyers’ bargaining power on their surplus was analyzed in Inderst and Wey (2007).
Given \(x,y\in \mathbb {R}^{n}\), we say \(x<_{Lex}y\) if there is some \(1\le i\le n\) such that \(x_{i}<y_{i}\) and \(x_{j}=y_{j}\) for \(1\le j<i\). Also, we say \(x\le _{Lex}y\) if \(x<_{Lex}y\) or \(x=y\).
This assumption enables us to speak properly about the \(i^\mathrm{th}\) agent of sector \(k\in M\).
For convenience, we define the bargaining power only in 2regular markets. However, the definition can be generalized to arbitrary markets in \(\mathcal {BBM}\) by using Eq. (2) and the translation vector.
Given a game (N, v) and a player \(i\in N\), the marginal contribution of i to the grand coalition is the amount \(v(N)v(N{\setminus } \{i\})\).
Theorem 2.1 is concerned with the core of the assignment game. Consequently, the translations that we present here are especially applicable when the rule that is being translated proposes core allocations. However, to study how properties in one setting are translated to the other setting, it is convenient to be able to translate any rule in \(\mathcal {BP}\) into a rule in \(\mathcal {BBM}\), and vice versa.
This condition is usually imposed on the definition of a rule. However, to compare rules and properties of bankruptcy problems and markets, it is better to consider wd as a property of a rule in \(\mathcal {BP}\). In particular, note that the translation \(\Psi \) may give rise to a rule in \(\mathcal {BP}\) that is not well defined.
We impose that \(\mathsf {f}\) satisfies cs to facilitate the explanation of the property.
For the sake of understanding the property, we do not rename this unique sector of sellers, as it should be denoted by \(N^1\) instead of \(N^k\). An analogous comment applies to the sector of buyers. Also note that both the active seller and the inactive seller pay the same price. In particular, the inactive seller may be paying prices to other sellers that are below their costs.
At this point, the question arises whether or not to require Eq. (10) to hold with a weak inequality and then use such a condition as a property instead of 2vsp. The answer to this question falls outside the scope of the current paper. In Knudsen and Østerdal (2012) merging and splitting incentives are studied separately for arbitrary cooperative games.
A point or setvalued solution concept f is anonymous if for all \((N,v)\in \mathcal {G}\) and \(i\in N\), it holds that \(f_{\pi (i)}(N,\pi v)= f_i(N,v)\), where \(\pi :N \rightarrow N\) is a permutation and \(\pi v (S) = v(\pi (S))\) for all \(S\subseteq N\).
We abuse notation and do not write the dependence of \(\mathbf {c}'\) and \(w'\) on the market \((\mathbf {c},w)\). This spares us the use of some cumbersome expressions.
In line with part (i) of Lemma 5.2, it can be verified that the reduced market in the definition of svsp is a 2regular market whenever \(\mathsf {f}\) satisfies cs and \((\mathbf {c},w)\in 2 \hbox {}\mathcal {BBM}\).
To be consistent with the assumptions made throughout this paper, dummy sellers might be added to sector k so that there is the same number of agents in all sectors.
See http://ec.europa.eu/competition/antitrust/legislation/legislation.html, retrieved on 10September2016.
References
Asch P, Seneca J (1976) Is collusion profitable? Rev Econ Stat 58:1–12
Aumann R (1973) Disadvantageous monopolies. J Econ Theory 6:1–11
Aumann R, Maschler M (1985) Game theoretic analysis of a bankruptcy problem from the Talmud. J Econ Theory 20:455–481
Bloch F, Ghosal S (2000) Buyers’ and sellers’ cartels on markets with indivisible goods. Rev Econ Des 5:129–147
Brandenburger A, Stuart H (2007) Biform games. Manag Sci 53:537–549
Dagan N (1996) New characterizations of old bankruptcy rules. Soc Choice Welf 13:51–59
Dubey P, MasColell A, Shubik M (1980) Efficiency properties of strategies market games: an axiomatic approach. J Econ Theory 22:339–362
Economides N (2001) The Microsoft antitrust case. J Ind Compet Trade 1:7–39
Einy E, Moreno D, Shitovitz B (1999) The asymptotic nucleolus of large monopolistic market games. J Econ Theory 89:186–206
Fershtman C, Gandal N (1994) Disadvantageous semicollusion. Int J Ind Organ 12:141–154
Gale D (1960) The theory of linear economic models. University of Chicago Press, Chicago
Gardner R (1977) Shapley value and disadvantageous monopolies. J Econ Theory 16:513–517
GonzálezDíaz J, SánchezRodríguez E (2007) A natural selection from the core of a TU game: the corecenter. Int J Game Theory 36:27–46
Granot D, Granot F (1992) On some network flow games. Math Oper Res 17:792–841
Guesnerie R (1977) Monopoly, syndicate, and shapley value: about some conjectures. J Econ Theory 15:235–251
Herrero C, Villar A (2001) The three musketeers: four classical solutions to bankruptcy problems. Math Soc Sci 42:307–328
Inderst R, Wey C (2007) Buyer power and supplier incentives. Eur Econ Rev 51:647–667
Kaneko M, Wooders MH (1982) Cores of partitioning games. Math Soc Sci 3:313–327
Knudsen PH, Østerdal LP (2012) Merging and splitting in cooperative games: some (im)possibility results. Int J Game Theory 41:763–774
Legros P (1987) Disadvantageous syndicates and stable cartels: the case of the nucleolus. J Econ Theory 42:30–49
Mendelson H (1982) Market behavior in a clearing house. Econometrica 50:1505–1524
Milliou C, Petrakis E (2007) Upstream horizontal mergers, vertical contracts, and bargaining. Int J Ind Organ 25:963–987
MorenoTernero JD, Villar A (2006) New characterizations of a classical bankruptcy rule. Rev Econ Des 10:73–84
Nash JF Jr (1950) The bargaining problem. Econometrica 18:155–162
Núñez M, Rafels C (2005) The BöhmBawerk horse market: a cooperative analysis. Int J Game Theory 33:421–430
Okuno M, Postlewaite A, Roberts J (1980) Oligopoly and competition in large markets. Am Econ Rev 70:22–31
O’Neill B (1982) A problem of rights arbitration from the Talmud. Math Soc Sci 2:345–371
Postlewaite A, Rosenthal R (1974) Disadvantageous syndicates. J Econ Theory 9:324–326
Quint T (1991) The core of an msided assignment game. Games Econ Behav 3:487–503
Salant S, Switzer S, Reynolds R (1983) Losses from horizontal merger: the effects of an exogenous change in industry structure on CournotNash equilibrium. Q J Econ 98:185–199
Schmalensee R (2007) Handbook of industrial organization, vol 3. Elsevier, Amsterdam
Schmeidler D (1969) The nucleolus of a characteristic function game. SIAM J Appl Math 17:1163–1170
Segal I (2003) Collusion, exclusion, and inclusion in randomorder bargaining. Rev Econ Stud 70:439–460
Shapley L, Shubik M (1972) The assignment game I: the core. Int J Game Theory 1:11–26
Sherstyuk K (1999) Multisided matching games with complementarities. Int J Game Theory 28:489–509
Sproul MF (1993) Antitrust and prices. J Polit Econ 101:741–754
Stuart HW Jr (1997) The supplierfirmbuyer game and its msided generalization. Math Soc Sci 34:21–27
Tejada O (2010) A note on competitive prices in multilateral assignment markets. Econ Bull 30:658–662
Tejada O (2013) Analysis of the core of multisided BöhmBawerk assignment markets. TOP 21:1–17
Tejada O, Núñez M (2012) The nucleolus and the corecenter of multisided BöhmBawerk assignment markets. Math Methods Oper Res 75:199–220
Thomson W (2003) Axiomatic and game theoretic analysis of bankruptcy and taxation problems: a survey. Math Soc Sci 45:249–297
Thomson W (2015) Axiomatic and gametheoretic analysis of bankruptcy and taxation problems: an update. Math Soc Sci 74:41–59
von BöhmBawerk E (1930) Positive theory of capital (translated by W. Smart). G.E. Stechert & Co., New York (original publication 1891)
Acknowledgments
This research received financial support from the Ministerio de Economía y Competitividad through Projects ECO201452340P and MTM201453395C32P as well as from Generalitat de Catalunya through Project 2014SGR40. Discussions with Julio GonzálezDíaz, Hans Gersbach, Marie Riekhof, and Clive Bell improved the paper and are gratefully acknowledged. We also acknowledge the comments received at the seminars and conferences in which this work was presented: Seminars at the Hebrew University of Jerusalem, Universitat de Barcelona, and University of Southern Denmark as well as at the SING9 and IWGTS2014 conferences in Vigo and São Paulo, respectively. Last but not least, we would like to thank the referees for their comments and suggestions which substantially helped improve a previous version of the paper. Finally, the usual disclaimer applies.
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
Proposition 8.1
The set of assignment games associated with 2regular markets, \( 2 \hbox {}\mathcal {BBG}\), and the set of sectors games, \(\mathcal {SG}\), are isomorphic, i.e.,
Proof of Proposition 8.1
Let \(N^1,\dots ,N^m\) be fixed throughout the proof, and recall that \(M=\{1,\dots ,m\}\) and \(N=N^1 \cup \cdots \cup N^m\). First, note that due to Remark 2.2,
and, by Definition 2.3,
Second, consider the following mapping:
Note that if \((N,\omega ^{\mathbf {c},w})=(N,\omega ^{\mathbf {c}',w'})\), then \((M,v^{\mathbf {c},w})=(M,v^{\mathbf {c}',w'})\), so \(\Gamma \) is well defined. Moreover, it is straightforward to check that \(\Psi \) is surjective. To check that it is also injective, let \((N,\omega ^{\mathbf {c},w}),(N,\omega ^{\mathbf {c}',w'})\in 2 \hbox {}\mathcal {BBG}\) be such that
On the one hand, suppose that \(r(\mathbf {c},w)=0\). Then, by Eq. (28) it must be that \(r(\mathbf {c}',w')=1\). Thus, \(v^{\mathbf {c},w}=v_0\) and \(v^{\mathbf {c}',w'}\ne v_0\), which implies \(\Gamma (N,\omega ^{\mathbf {c},w})\ne \Gamma (N,\omega ^{\mathbf {c}',w'})\).
On the other hand, assume that \(r(\mathbf {c},w)=r(\mathbf {c}',w')=1\). From Eq. (28) it follows the existence of \(S\subseteq N\) minimal w.r.t. to inclusion such that \(\omega ^{\mathbf {c},w} (S)\ne \omega ^{\mathbf {c}',w'}(S)\). Furthermore, from the definition of the characteristic function of an assignment game, it must be that \(S=Z^R\) for some \(R\subseteq M\). Hence,
Thus, \(\Gamma (N,\omega ^{\mathbf {c},w})\ne \Gamma (N,\omega ^{\mathbf {c}',w'})\), so \(\Gamma \) is injective.\(\square \)
Proof of Lemma 5.1
Let \(\mathsf {f}\) be a rule in \(\mathcal {BBM}\) satisfying cs, then it trivially satisfies part (ii). Part (i) then follows from Theorem 2.1. The reverse implication follows from Theorem 2.1 and the properties of the market \((\tilde{\mathbf {c}},\tilde{w})\) defined in Eq. (2).
Proof of Lemma 5.2
We only prove the first part, as the second part is trivial. Indeed, \(\vert N^{k} \vert = \vert N^{m} \vert = 2\) since \((\mathbf {c},w)\in 2 \hbox {}\mathcal {BBM}\), and
where the penultimate inequality holds because \(\mathsf {f}^{k}_2(\mathbf {c},w) = \mathsf {f}^{m}_2(\mathbf {c},w) =0\)—due to Theorem 2.1—, the fact that \((\mathbf {c},w)\) is a 2regular market, and because \(\mathsf {f}\) satisfies cs, while the last inequality holds since \((\mathsf {f}_k(\mathbf {c},w))_{k\in M}\) is a core allocation of \((N,v^{\mathbf {c},w})\).
\(\square \)
Proof of Lemma 5.3
By Lemma 5.2, the reduced market is a 2regular market, so \(r( \alpha ^{k}(\mathbf {c},w,\mathsf {f}) ,w) \le 1\). Then it suffices to realize that cs implies \(\mathsf {f}^{k}_2(\mathbf {c},w)= \mathsf {f}^{k}_2 \left( \alpha ^{k}(\mathbf {c},w,\mathsf {f}) ,w \right) = 0 \) and
where the first and the third equalities hold since \(\mathsf {f}\) yields efficient allocations for \((\mathbf {c},w)\) and \(( \alpha ^{k}(\mathbf {c},w,\mathsf {f}),w)\) respectively, the second equality holds since \(\mathsf {f}\) satisfies 2vsp, and the last equality holds since by definition of the reduced market,
\(\square \)
Proof of Claim A
Let \(\mathsf {f}\) be a rule in \( 2 \hbox {}\mathcal {BBM}\) satisfying 2cbs and let \((\mathbf {c},w),(\bar{\mathbf {c}},\bar{w})\in 2 \hbox {}\mathcal {BBM}\) be two 2sided markets satisfying \(w_1c^1_1=\bar{w}_1\bar{c}^1_1\) (same total surplus), \(c^1_2c^1_1=\bar{c}^1_2\bar{c}^1_1\), and \(w_1w_2=\bar{w}_1\bar{w}_2\) (same bargaining power). Then,
Since \(\mathsf {f}\) satisfies 2cbs, we obtain \(\mathsf {f}(\mathbf {c},w)=\mathsf {f}(\bar{\mathbf {c}},\bar{w})\) by inserting the above equations into the two expressions of Eq. (11).\(\square \)
Proof of Claim B
Let \(\mathsf {f}\) be a rule in \( 2 \hbox {}\mathcal {BBM}\) satisfying 2cbs and let \((\mathbf {c},w),(\bar{\mathbf {c}},\bar{w})\in 2 \hbox {}\mathcal {BBM}\) be two 2sided markets satisfying \(w_1c^1_1=\bar{w}_1\bar{c}^1_1\) (same total surplus), \(c^1_2c^1_1=\bar{w}_1\bar{w}_2\), and \(w_1w_2=\bar{c}^1_2\bar{c}^1_1\) (switched bargaining power). Then,
Since \(\mathsf {f}\) satisfies 2cbs, we obtain \(\mathsf {f}^1(\bar{\mathbf {c}},\bar{w})=\mathsf {f}^2(\mathbf {c},w)\) and \(\mathsf {f}^2(\bar{\mathbf {c}},\bar{w})=\mathsf {f}^1(\mathbf {c},w)\) by inserting the above equations into the two expressions of Eq. (11).\(\square \)
Proposition 8.2
The properties that characterize the Talmud Assignment Rule, \( \mathsf {T}\), are logically independent.
Proof
We consider the following four rules in \(\mathcal {BBM}\):
(i) Let the rule in \(\mathcal {BBM}\), \(\hat{\mathsf {T}}\), be defined for every \((\mathbf {c},w)\in \mathcal {BBM}\) as
Then, \(\hat{\mathsf {T}}\) satisfies 2vsp, 2abp, and 2cbs but not cs.
(ii) Let the rule in \(\mathcal {BBM}\), \(\hat{\mathsf {T}}\), assign for every \((\mathbf {c},w)\in \mathcal {BBM}\) the corecenter (see GonzálezDíaz and SánchezRodríguez (2007)) of \((N,\omega ^{\mathbf {c},w})\). From Núñez and Rafels (2005) and Tejada and Núñez (2012), we know that this allocation differs from the nucleolus (of the assignment game) for the whole class of market games in \(\mathcal {BBG}\), but that they coincide for bilateral markets. Lastly, the corecenter is an anonymous solution concept. Then, \(\hat{\mathsf {T}}\) satisfies cs, 2abp, and 2cbs, but not 2vsp.
(iii) Let CEA be the constrained equal awards bankruptcy rule. This rule is defined for every \((M,E,d)\in \mathcal {BP}\) and \(i\in M\) by \(CEA_i (M,E,d) = \min \left\{ d_i,\lambda \right\} \), where \(\lambda \) is chosen so that \(\sum _{i\in M} CEA_i (E,d) = E\). The following properties are known:

(a) CEA chooses a core allocation of the bankruptcy game \((M,v^{E,d})\)—see Theorem 2 in Thomson (2003).

(b) CEA satisfies bc—see Theorem 1 in Herrero and Villar (2001).

(c) CEA does not satisfy cg—by Theorem 5.1 and because CEA satisfies wd by construction.
First, let \(\hat{\mathsf {T}} = \Phi (CEA)\). From Theorems 2.1 and I and statement (a), it holds that \(\hat{\mathsf {T}} \) satisfies cs. Second, following the same lines as in the proof of statement (ii) in Proposition 6.2, we can show that due to statements (a) and (b), \( \hat{\mathsf {T}}\) also satisfies 2vsp. Third, the definition of \( \hat{\mathsf {T}} \) directly implies that it satisfies 2abp. Fourth and last, statement (c), statement (ii) in Proposition 6.1, statement (i) in Proposition 6.2, and the fact that \(CEA= \Psi (\Phi (CEA))\) imply that \(\hat{\mathsf {T}} \) does not satisfy 2cbs.
(iv) Consider the following subset of \( 2 \hbox {}\mathcal {BBM}\):
Since \(w_22c<0\) for all \((\mathbf {c},w)\in \mathcal {V}\), we have \(\mathcal {V}\subseteq 2 \hbox {}\mathcal {BBM}\). Let the rule in \(\mathcal {BBM}\), \(\hat{\mathsf {T}}\), be such that for every \((\mathbf {c},w)\in \mathcal {V}\),
and \(\hat{\mathsf {T}}^k_2 (\mathbf {c},w)= 0\) for all \(k\in M=\{1,2,3\}\). Note that for every \((\mathbf {c},w)\in \mathcal {V}\)
Hence, sectors 1 and 2 are symmetric players in the sectors game. Since the nucleolus is an anonymous solution, \(\mathsf {T}^1_1(\mathbf {c},w)=\mathsf {T}^2_1(\mathbf {c},w)\). Also, note that \(x_1+x_2 = w_1 \ge w_1\), so \(x \in C(M,v^{\mathbf {c},w})\), with \(M=\{1,2,3\}\). It then follows from part (i) of Theorem 2.1 that \(\hat{\mathsf {T}}(\mathbf {c},w)\in C(N,\omega ^{\mathbf {c},w})\) for all \((\mathbf {c},w)\in \mathcal {V}\). Next, let the rule in \(\mathcal {BBM}\), \( \tilde{\mathsf {T}}\), be defined for every \((\mathbf {c},w) \in \mathcal {BBM}\) by
By construction, \(\tilde{\mathsf {T}}\) satisfies 2cbs and cs. Next we focus on 2vsp. On the one hand, consider the reduced market \((N^1,N^3,\alpha ^{1}(\mathbf {c},w,\tilde{\mathsf {T}}),w)\), where \(\alpha ^{1}_1(\mathbf {c},w,\tilde{\mathsf {T}})=0\) and \( \alpha ^{1}_2(\mathbf {c},w,\tilde{\mathsf {T}})=c\). Then,
On the other hand, consider the reduced game \((N^2,N^3, \alpha ^2 (\mathbf {c},w,\tilde{\mathsf {T}}),w)\), where \(\alpha ^{2}_1(\mathbf {c},w,\tilde{\mathsf {T}})=w_1\) and \(\alpha ^{2}_2(\mathbf {c},w,\tilde{\mathsf {T}})=w_1+c\). Then,
That is, we have proved that \(\tilde{\mathsf {T}}\) satisfies 2vsp. Lastly, since \(\tilde{\mathsf {T}}\) differs from \(\mathsf {T}\), it follows from Theorem II that \(\tilde{\mathsf {T}}\) cannot satisfy 2abp.\(\square \)
Proof of Proposition 7.1
First, we show that \(\mathsf {T}\) satisfies svsp for 2regular markets. Indeed, for such a market, the conditions in the definition of svsp easily follow by repeating the lines in the proof of Proposition 6.2 and the existence part of Theorem II using consistency instead of bilateral consistency. Finally, from Lemma 5.1 it follows that \(\mathsf {T}\) satisfies svsp for arbitrary markets.\(\square \)
Proof of Proposition 7.2
First, note that if a rule in \(\mathcal {BBM}\) satisfies 2cbs, cs, 2abp, and 2vsp, by Theorem II it has to coincide with the Talmud Assignment Rule. Second, consider the bilateral 2regular market \((\mathbf {c},w)\) defined by \(\mathbf {c}=c^{1}=(1,2)\) and \(w=(3,0)\), and let \(k=1\) and \(S=\{1\}\subseteq N^{1}\). Then, by Definition 6.1,
and
which concludes the proof.\(\square \)
Rights and permissions
About this article
Cite this article
Tejada, O., ÁlvarezMozos, M. Vertical syndicationproof competitive prices in multilateral assignment markets. Rev Econ Design 20, 289–327 (2016). https://doi.org/10.1007/s1005801601931
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s1005801601931