Horizontal Mergers and the Structure -Conduct-Performance Paradigm

European Journal of Economics, Finance and Administrative Sciences
ISSN 1450-2887 Issue 15 (2009)
© EuroJournals, Inc. 2009
http://www.eurojournals.com/EJEFAS.htm

Horizontal Mergers and the Structure –Conduct-Performance
Paradigm


Ebrahim Hosseini Nasab
Department of economics, Tarbiat Modares University
P.O. Box: 14155-4838, Tehran, Iran
Tel: +98-21-8288-4640
E-mail: ebhn23@hotmail.com

Somayeih Azami
Department of Economics, Tarbiat Modares University
P.O. Box: 14155-4838, Tehran, Iran
E-mail: sazami_econ@yahoo.com

Abbas Asari
Department of Economics, Tarbiat Modares University
P.O. Box: 14155-4838, Tehran, Iran
E-mail: asari_a@modares.ac.ir


Abstract
In the traditional horizontal mergers, the merging firms (the insiders) cooperate and
maximize their joint profit without changing their post-merger behavior so that in the
quantity-setting games, the merger is profitable for the outsiders, unprofitable for insiders,
socially not beneficial. We suggest a different approach in that we allow for the
cooperating firms to change their behavior after integration in accordance with the
“structure –conduct– performance” paradigm while still seeking maximizing their joint
profit. We use the model to study the most important automobiles manufactures in Iran,
namely: Iran Khodrou and Saipa, as represented in Tehran’s stock exchange market. Our
results indicate that firstly, the integration of these two car manufacturers increases the
consumer welfare and secondly, it will harm the outsiders. The integration of these two car
manufactures is not profitable for the participants in the integration.


Keywords: Horizontal Integration, Horizontal Mergers, the Structure-Conduct-
Performance-Paradigm, Private Profitability of Integration, Social profitability
of Integration.
JEL Classification Codes: L11, L13, L41, L62, D21, D43.

1. Introduction
Industrial economists began to focus on integration or mergers since the 1950s with the following
questions:
a) Is integration profitable for the participating firms in the merger (for the insiders)?
b) Is integration profitable for the non- participating firms in the merger (for the outsiders)?
c) How does integration affect the consumers of the commodities?

---

135
European Journal of Economics, Finance And Administrative Sciences - Issue 15 (2009)

The literature on horizontal integration does not offer a unanimously agreed upon answer to
these questions. For example, Salant et al (1983) conclude that in the quantity-setting games, outsiders
benefit from the mergers, while insiders and society as a whole loose. This conclusion, however, was
not totally convincing to other industrial economists who saw a positive side to the profitability of
mergers for the insiders and sought to find incentives that make mergers profitable to insiders as well.
Perry and Porter (1985) and Denecker and Davidson (1985) argue that an integrated entity is
different from its constituent parts and that holding this view of an integrated entity can affect the
outcome of horizontal integration. Denecker and Davidson argue that in a price-setting game in an
industry with a homogeneous product, horizontal integration of several firms within an industry results
in a firm that is larger than its constituent firms and this makes the merger profitable to the participants
in the integration. Perry and Porter show that horizontal integration among a number of firms within an
industry leads to formation of a firm that due to having a higher capital share, is larger and in different
form than its constituent elements and this makes integration profitable for insiders.
Salant and Gaudet (1991) conclude that the number of the participating firms and the
“adjustment factor” are among the factors that influence the profitability for the insiders of integration.
The characteristic of demand and cost functions (the slope of the inverse demand function, the slope of
the cost function, the second derivative of the marginal cost function and the second derivative of the
demand function) constitute the adjustment factor.
Bru and Fauli-Oller (2002) show that horizontal integration among a number of firms within an
industry increases the purchasing power of those firms and this effect can compensate for the outsider
response. In other words, it can compensate for the effect that originates from the increase in
production of outsider firms due to the increase in prices resulting from integration that lead to non-
profitability of integration to participants in the integration, as noted by Salant et al. Hence integration
becomes profitable for the participants in the integration.
Lommerud et al (2005) argue that firms after integration gain additional purchasing power in
the input market. They can acquire inputs from input suppliers at lower prices. Hence integration can
be profitable for the firms participating in the integration.
Eckbo(1983) and Fee and Thomas(2004) have an analytic discussion of the profitability of
integrations; their focal point is the efficiency and collusion effects that result from participants in the
integration.
In effect the dominant thinking in the literature on horizontal integration as reflected in these
papers is maximization of the joint profits. According to this thinking, the firms participating in the
integration (the insiders) cooperate and maximize their profits. Therefore, the outsiders benefit and
consumers loose.
However, Crean and Davidson (2004) argue that it would be difficult to believe that outsiders
will benefit from collusion among their competitors, particularly in markets characterized by strategic
substitutes (the case where each firm’s best response function is sloping downwards). Furthermore,
according to this argument, society as a whole does not necessarily loose from these integrations.
Crean and Davidson argue that integrated firms can be viewed as a multidivisional firm that do
not necessarily cooperate after integration and hence do not maximize their joint profits. An integrated
firm by dividing its parts into two groups: leader and follower in making quantity and price decisions
follows the stakelberg model. Following this type of approach, the authors conclude that integrations
harm the outsiders while from the social standpoint they increase consumer’s welfare. Furthermore the
insiders benefit from the integration.
We too follow the joint profit maximization thinking from the view point of the” structure –
conduct– performance” (SCP) paradigm to study horizontal mergers. Thus, we use the SCP paradigm
in analysis of the post- merger game. If we view integration as a change in the market structure,
according to the SCP paradigm, it is expected that integration can change a firm’s behavior towards
other firms and this leads to a change in the performance of the insider and outsider firms. With firms’
prices and productions changed due to changes in the firms’ behavior with respect to each other, the

---

136
European Journal of Economics, Finance And Administrative Sciences - Issue 15 (2009)

profitability of the firms also changes. The index we have in mind for reflecting the market behavior is
“consistent conjectural variation” and we allow consistent conjectural variation and production in the
two games before integration and after it to be determined endogenously within the system. In other
words, we can determine behavior and performance simultaneously within the system.
In short, we can place the approach taken in this paper in the following context:


Salant et al
Joint profit maximization approach
Perry and Porter, Denecker & Davidson,
Lommerud et al

The multidivisional firm approach
Crean & Davidson

The Structure-Conduct-Performance approach
This Paper

Due to complexities of the formulas derived in this paper following the SCP approach, it would
be difficult to arrive at a unique conclusion from the formulas. Therefore, we attempt to test our
theoretical findings using the case of automobile industry in Iran. More specifically, we consider
integration of the two largest automobile manufactures in Iran, namely: Irankhodrou and Saipa and try
to predict the outcome of this integration in the light of the theoretical findings. The exercise consists
of pooling time series and cross section data and use a panel estimation methodology based on the data
over the 1998-2007 time period.
The paper is organized as follows: In the next section (section 2) after this introductory section,
we will review briefly concepts related to consistent conjectural variation (CV) and consistent
equilibrium and show the derivation of CV and production in the pre-merger and post-merger game.
Section 3, presents an analysis of the effect of mergers based on theoretical findings and demand
estimates. Finally, section 4, summarizes and brings the paper to its conclusions with recommendations
for further research.


2. Theoretical Grounds
2.1. Consistent Conjectural variation
Behavior is one of the three elements of the market in the manner of SCP that firms adopt to adjust
themselves to the market. Beliefs determine behavior and behavior determines performance. Firm’s
behavior in realty is firm’s conjecture of how will its competing firm respond to its decision to change
production. The firm’s belief about its competitor, determines its production choice vis avis its
competitor. In the “decision rule” this behavior is expressed as a “reaction function”. There are two
questions of interest here:
a) How does firm’s conjecture measure?
b) is firm’s conjecture consistent?
Frisch (1933), considers firm’s belief about competitor’s response to a change in its decision as
conjectural variation. Hence, Frisch uses conjectural variation to represent competitors’ expected
response. (Cornout assumes that conjectural variation is zero). Hence it might be said that conjectural
variations are measures of believes and conjectures. It is noted, however, that not all conjectural
variations are consistent and rational and the problem determining consistent conjectural variations
remains.
Fellner (1949), point out that in general ad hoc conjectural variations are not consistent with
rational behavior expect in equilibrium points. Firms form expectations of how their competitors will
behave but these expectations do not necessarily need to be correct; In fact, any adhoc conjectural
variation is in general inconsistent with out -of –equilibrium behavior of the firms. This critique
motivated a search for consistent conjectures, i.e. conjectures consistent with the firm’s response to
actual production (price) decision of the ith firm.

---

137
European Journal of Economics, Finance And Administrative Sciences - Issue 15 (2009)

Various scholars such as Laitner (1980), Bresnahan (1981), Perry (1982), Kamien and
Schwartz (1981), Boyer and Moreaux (1983) have tried to characterize consistent conjectural
variations. This paper uses the definition of Kamien and Schwartz to see how consistent conjectural
variations can be explained.
Assume that firms choose their production levels to confront the maximum profit criteria and
the production levels chosen by their competing firms. The reaction function expresses this relation
provided that the firm’s profit maximizing production level is a function of the production levels of the
firm’s competitors. In a duopoly model, the slope of this response function represents the rate at which
the profit minimization production (denoted by q1) changes in response to changes in the competitor’s
production (denoted by q2), ∂q1/∂q2.
On the other hand, firm 2 may guess that its competitor, firm 1 for example, has a response
function. This response function is unknown to firm 2, but it can form conjectures about it based on the
slope of the competitor’s response function. These conjectural slopes are called conjectural variations
denoted by (Δq1/Δq2).
Now the consistency condition can be stated as follows: (Δq1/Δq2) = ∂q1/∂q2, i.e., The belief
and the conjecture of Firm 2 about the response function of firm 1 must be equal to the slope of the
response function of firm1.
In an oligopoly model, the consistency condition requires equality of the actual rate of change
in the production of ith firm in response to the production change of the nth firm given the nth firm
conjecture of the ith firm’s response. This condition can be stated as: (Δqi/Δqn)= ∂qi/∂qn,i.e. the
conjecture of the nth firm about the response of the ith firm must be equal to the derivative of the
response function of the ith firm with respect to the production of the nth firm.
Since in an oligopoly situation more than two firms are involved, it is not easy to calculate the
partial derivative of ith firm with respect to production of the nth firm, holding the production of the
remaining 2-2 firms constant.
Holt (1980) argues that the ith firm in responding to the change in the production of nth firm, not
only needs to take account of the change in the production of the nth firm, but also the maximizing
profit adjustments of the remaining n-2 firms. Therefore the rate of change of production of each firm
in response to a change in the production of the nth firm, given the assumption of erogeneity of the nth
firm production, and given simultaneous adjustments by all the n-2 competitors is a profit
maximization rate. This consistent joint rate of the change in production of each firm in response to a
change in production of others can be denoted by k.
Even though it appears that consistent conjectures equilibrium (CCE) was founded in 1980’s,
more accurately it was in fact Wassily Leontieff that founded this concept in the year 1936. His
analysis in a paper entitled: Stakelberg in monopoly competition published in the Journal of Political
Economy in 1936 shows that he had a clear vision the CCE concept. Even though his argument is not
as rigorous as some of the present day arguments and his study does not have the breath of today’s
papers, but his early attempts at discovering the concept deserve some credit.
Friedman (1985) criticizes analysis of conjectural variations in static models. Riordan (1985)
introduces the concept of dynamic conjectural variations and relate this to equilibrium behavior in a
Cornout Model in which firms have perfect information. Firms do not observe the production of
competitors, but they can draw conclusions regarding the position of the demand curve from observing
past prices.

2.2. Consistent Equilibrium.
Consistent equilibrium refers to an equilibrium derived from the two conditions of consistency and
equilibrium. A consistent condition is a condition that leads to consistent conjectural variation. An
equilibrium condition is the first order condition that micro economists use to derive equilibrium price
and production. In general, the consistency and equilibrium conditions are fuction of production q and
conjectural variation k; combining these two conditions yield a consistent k and an equilibrium

---

138
European Journal of Economics, Finance And Administrative Sciences - Issue 15 (2009)

production corresponding to consistent conjectural variation. It is clear that before integration, there
exists a consistent equilibrium from which q and k can be derived. After integration, there exist another
consistent equilibrium from which we can calculate equilibrium q and k.
Consider an industry with N firms producing a non-homogenous product and each firm makes
production its decision variable. Firms face constant marginal costs and the following demand
function:
Pi= P(q1,…,qi,…,qn) (3)

2.2.1.The no-merger game 1.
Firms determine their production and price levels through maximizing their profits. Hence:
Max [P(q1,…,qi,...,qn)-c]qi (4)
∂π
n
n
i/∂qi = 0, Pi – c +(p1+(N-1)p2kn)qi = 0
(5)2
where Pi is the ith firm price, qi is the ith firm production and c is constant marginal costs.
p1 = ∂Pi/∂qi and p2= ∂Pi/∂q-i.
We assume there are n symmetric firms and differentiate (5) to arrive at:
[2p
n
n
n
1 + (N-1)p2kn +qi (p11 +(N-1) p21kn)]dqi +[p2 + qi (p12 +(N-1)p22kn)]Σ2 dqj=0 (6)
Now we divide the two sides by dqn and write:
kn = dqj/dqn,
j ≠ n
to arrive at:
(N-1)(kn)2 [p2 + qn (p21 +(N-2)p22 )] + kn[2p1 + (N-2)p2 + qn (p11 +(N-2)p12 +(N-1)p22)]
+p2 +p12 = 0
(7)
Equation (7) is the consistency condition. The method used here is due to Kamien and
Schwartz. We also could have used Perry’s method that in the final analysis yields identical results.
If the demand function (3) follows a linear functional form like:
Pi = A – Bqi – DQ
(8)
Then the consistency condition will be as follows:
(N-1)D(kn)2 + (2B + ND ) kn + D = 0
(9)
Therefore, given the linear functional form of the demand, the consistency condition is a
function of k only and from this condition we can calculate the consistent kn ( kn is the conjectural
variation in pre-merger game):
kn = -Y + ( Y2 – 1/N-1)1/2; Y = (2B+ND)/(2N-1)D = (2γ + N)/2(N-1); γ=B/D (10)
where 1/γ approximates the difference in commodities, 0 < 1/γ <1.
Given the amount of consistent kn, entering it in the equilibrium condition gives symmetric
equilibrium qn (the symmetric equilibrium production in the pre –merger game). The amount of
production in the consistent equilibrium is:
qn = (A-C)/[γ +(N-2)/2+(γ2 +Nγ+(N-2)2 /2)] ½
(11)

2.2.2. The post- merger game
Now we consider the case where M from N firms in the industry merge. Since there would be two
types of firms, i.e. firms that have cooperated in the integration (insiders) and firms that have not
(outsiders), there will be two types of equilibrium condition and two types of consistency condition.
The equilibrium and consistency conditions for the outsiders in the post-merger game remain the same.

1 Derivation of the consistency condition in the no-merger game follows Kamien and Schwartz, but derivation of the game
conditions in the post-merger is solely the work of the authors.
2 N indicates no-merger game

---

139
European Journal of Economics, Finance And Administrative Sciences - Issue 15 (2009)

But the equilibrium and consistency condition for the insiders need to be derived afresh. To do this, we
write the profit function of the merged firm and maximize it for a hypothetical insider firm such as firm
1. The joint profit function can be written as:
πm = ΣM
p
p
j=1Pj qj
(12)3
where πm is the profit of the merged firm, P p
p
j is the post-merger price of the jth insider firm and qj the
post merger production of jth insider firm. Now we maximize πm for a hypothetical insider firm such as
firm 1:
∂πm /∂q p
p
p
p
p
p
p ∂ p
p
p
p
1 =0, P1 + q1 (∂P1 /∂q1 )+ P1 (∂q-1 / q1 )+ q1 (∂P-1 /∂q1 )=0
P p
p
p
p
p ∂ p
p
1 + q1 (p1 + (N-1)p2 kI )+P1 (M-1) (∂qj / q1 ) + q1 (M-1)p2 = 0
(13)
where: k p
p ∂
p
I is the post-merger conjectural variation of the insiders; ∂q-1 / q1 is the effect of the
production of the insider firm 1 on the production of other insider firms in the post-merger game; ∂qj
p ∂ p
/ q1 (with j≠1 ) is the post merger effect of the production of the insider firm 1 on the production of
one of the insiders in the post merger game;
∂P- p
p
1 /∂q1 is the effect of the production of the insider firm 1 on the price of other insider firms
in the post-merger game; And the subscript -1 refers to all insider firms except firm i. But the question
is how much is the quantity of ∂q p ∂
p
j / q1 (with j≠1)?
A review of the previous research on modeling mergers in industries with non-homogeneous
products reveals that in deriving the post-merger insider firm’s equilibrium condition, the production
effect of a firm participating in the merger on the price of other firms participating in the merge is
taken into effect, while the production effect of the insider firms on each other are not.
The new contribution of this paper to the studies on consistent behavior in mergers is taking
account of the effect of both p
p ∂ p
2 and ∂qj / q1 (with j≠1). The authors find taking the effect of and ∂qj
p ∂ p
/ q1 (with j≠1) into account is considered a logical new step taken in addition to the work done so far
in the existing literature, as this allows behavior to vary.
Now it may be asked: do the firms participating in mergers have conjectures and beliefs about
each other? This question can be answered in positive, because firms participating in the merger do
follow the leader-follower model, as in the case of Crean and Davidson (op cited). There is no reason
why we cannot say that the firms participating in the merger have conjectures and beliefs about each
other’s production. If this is true, then, we can consider the following two states:
∂q p ∂ p
p
p ∂ p
j / q1 (with j≠1) = kI and ∂qj / q1 (with j≠1) = 0
If ∂q p ∂
p
j / q1 (with j≠1) = 0, then the equilibrium condition a hypothetical insider firm such as
firm 1 is:
P p
p
p
p
1 + q1 (p1 + (N-1)p2 kI ) + q1 (M-1)p2 = 0
(14)
If ∂q p ∂
p
p
j / q1 (with j≠1) = kI , then:
P p
p
p
p
p
p
1 + q1 (p1 + (N-1)p2 kI ) + P1 (M-1) kI +q1 (M-1)p2 = 0
(15)
Now we can derive the consistency condition from the equilibrium condition. After replacing
P p
1 with quantity in equations (14) and (15) [(in accordance with equation (8), since we have different
q’s q p
p
p
o (qoutsider), q1 , qj (with j≠ 1)], we will have three types q’s and three equations for deriving the
consistent conjectural variation shown in Table (1&2). Actually, to derive the consistency condition,
we divide (15) into dq p
p
p
o , dq1 and /or dqj that depending on the particular dq, we arrive at a particular
consistency condition.

3 -P indicates the post-merger game

---

140
European Journal of Economics, Finance And Administrative Sciences - Issue 15 (2009)

Table 1:
Derivation of the insider consistency condition.

dq p
p2
p
o
[(M-1)/γ+(2MN-N-M2)] kI +[(N-2)+2/γ+2(M-1)] kI +1=0
∂ p
p
p
dq p
p2
p
q /∂q = k
1
[(M-1)(M-2)+2(N-M)(M-1)+(M-1)(1+1/γ)] kI +[2N-2M-2] kI +2(1+1/γ)=0
j
1
I
j≠1
dq p
p2
p
j
[2MN-M2-N+1+(M-2)(1+1/γ)] kI +[(N-2)+3(1+1/γ)+ (M-2)] kI +1=0
dq p
p2
p
o
[N-M] kI +[2/γ+N+(M-1)] kI +1=0
p/∂q p
1 =0
dq p
p
1
kI =-2/(N-M)
∂qj
j≠1
dq p
p
j
kI = -(1+1/γ)/(N-M)

Table 2:
Derivation of the insider consistent conjecture variation using the equations written in table 1.

dq p
p
o
kI =-[1/2(M+2MNγ -NM2γ-1)][-4γ+2Mγ+Nγ±(16γ2-16Mγ2-8Nγ2-4NMγ2-
12γ+4Mγ+4Nγ+4M2γ2+4+4NM2γ2+N2γ2)1/2]
dq p
p
1
kI =[-1/γM2-2MNγ-M+1-γ+2Nγ ][-γ+Mγ+Nγ± (-γ2 - 2Mγ2+2Nγ2+3M2γ2-2NMγ2+2γM2-
∂ p
p
p
2Mγ+4Nγ-2M+2+γ2N2-4MNγ)1/2]
qj /∂q1 = kI
p
j≠1
dqj
k p
I = -1/2(γM2-Mγ-M+2+γ+Nγ-2MNγ)
[-γ+Mγ+Nγ+3 ± (5γ2 – 6Mγ2+2Nγ2+5M2γ2-
6NMγ2+N2γ2+2Mγ+9+6Nγ+2γ)1/2]
dq p
p
o
kI =[ -1/2γ(M-N)] [-γ+Mγ+Nγ+2 ± (γ2 – 2Mγ2—6Nγ2+M2γ2+ 2MNγ2+N2γ2+4Mγ+ N2γ2
∂q p
p
j /∂q1 =0
+4Mγ+4Nγ-4γ+4)1/2]
j≠1
dq p
p
1
kI =-(1+1/γ)/(N-M)
dq p
p
j
kI =-2/(N-M)

But the outsider consistency condition still is:
(N-1)k p2
p
o
+(2γ+N) ko +1=0
(16)
Where kop is the consistent outsider conjectural variation in post merger game( Table 3).

Table 3:
Derivation of the outsider consistency condition and outsider consistent conjectural variation.

(N-1) k p2
p
o
+ (2γ + N ) ko +1 = 0
k p
o = -Y + ( Y2 – 1/(N-1))1/2; Y = (2γ + N) / 2(N-1)

Up to this point, we have tried to derive the consistent conditional variations for insider firms in
the post –merger game. Since we have assumed a linear demand curve, the consistency condition is
simply a function of k and from here we could derive the consistent conditional variation that is not
necessarily the same as the consistent conditional variation of the insider firms. Now, the question is:
how can we proceed to determine the amount of production (q)?
To determine the amount of production (q) we can make use of the equilibrium condition and
derive the insider production (q p
p
p
I ) and the outsider production (qo ) as a function of the amount of kI
and k p
o . This is shown in tables (4) and (5).

---

141
European Journal of Economics, Finance And Administrative Sciences - Issue 15 (2009)

Table 4:
The insider firm’s equilibrium condition as a function of k, q in the post merger game.

The insider equilibrium condition:
[1+(M-1) k p
p
p
p
I (A-C)/D]- qI [(1+1/γ)+(N-1) kI +(M-1)+(1+(M-1) kI )(M-
∂q p
p
p
j /∂q1 = kI
1+1/γ)]- q p
p
o [(N-M)(-1-(M-1) kI )]=0
j≠1
The outsider equilibrium condition:
[(A-C)/D]- q p
p
p
I [M-1]- qo [(N-M)-(N-1) ko -1]=0
The insider equilibrium condition:
p
p
[(A-C)/D]- q p
p
p
∂
I [-1/γ-(M-1)-(1/γ+1)-(N-1) kI +(M-1))]- qo [N-M]=0
qj /∂q1 =0
j≠1
The outsider equilibrium condition:
[(A-C)/D]- q p
p
p
I [M-1]- qo [(N-M)-(N-1) ko -1]=0

Table 5:Derivation of qop and qIp

∂
p
p
p
q p
p
p
= [2(N-M)-(N-1) k -1][1+(M-1) k ] (A-C )/DΔ′
j /∂q1 = kI
qI
o
I
q p =[(1/ γ (2+(M-1) )k p+M+(N-1) k p](A-C)/DΔ′
j≠1
o
I
I
∂
p
p
q p
p
=[ k (1-N)-1](A-C)/DΔ
j /∂q1 = 0
qI
o
q p =[ k p (N-1)+M+2/γ](A-C)/DΔ
j≠1
o
I
Where: Δ and Δ′ in table (5) are determinants of the matrix of qI and q0 coefficients


3. The Empirical Findings
In this section, we present the results obtained from an empirical application of the theoretical
apparatus, presented in the previous section above (tables 1-5), to a study of the Iranian vehicle
industry. We show how the private and social profitability of Irankhodrou and Saipa (the two largest
manufacturers of vehicles) merger is affected in the event that consistent conjectural variation
(behavior) and the production of firms or industry vary. We calculate the consistent amount of k for the
insiders and the outsiders in the pre-merger and post merger games, considering two alternative
functional forms for the demand for vehicles. And finally, we use k to derive outsider production (qop)
and insider production (qIp).
The Chamberlain Demand for vehicles is estimated in two functional forms: linear and semi-
logarithm and the use of pooled data method.
Pi=A-Bqi-DQ (14)
′
′
′
LPi=A -B qi-D Q (15)
The estimated demand functions and derivations of the relevant variables are given in tables 6,
7 and 8 below.

Table 6:
Estimates of demand for automobiles (Linear and semi- logarithm forms)

Equation form parameters
Estimates Standard
Deviation
R2 SEE
D-W
A
56×106




Linaer
B -3.7 3


C -5.6
1.6
0.58

0.89
A 15.6
0.32


Semi-log
B
-6.9×10-7 3.7×10-7



C
-2.59×10-7 4.8×10-7
0.73 0.48 2.57

---

142
European Journal of Economics, Finance And Administrative Sciences - Issue 15 (2009)

Table 7: kIp, kop, qop, qIp in consistent equilibrium in the pre-merger and post-merger games
(Linear demand form)

No-merger game
Post-merger game
kn=-0.1 k p
p
I =-0.13,
qI =0.3×107
qn=0.11××107
k p
p
o =-0.1,
qo =0.14×107

k p
p
I =-0.9,
qI =0.27×107
k p
p
o =-0.1,
qo =0.15×107

Table 8:
kIp, kop, qop, qIp in consistent equilibrium in the pre-merger and post-merger games
(Semi-logarithmic demand form).

No-merger game
Post-merger game
kn=-0.14 k p
p
I =-0.12,
qI =2.9×107
qn=0.08× 107
k p
p
o =-0.14,
qo =0.47×107

k p
p
I =-0.16,
qI =2.15×107
k p
p
o =-0.14,
qo =0.63×107


4. Summary and Conclusion
The study makes it obvious that the post-merger production of the outsider and the insider firms i.e.
industry’s output increases in the case of both linear and semi-log functional forms. With a high
estimate of the post-merger output, but a low estimate of the post merger prices, we see that we have a
sharp decrease in the post-merger- profit. Therefore according to this study, Irankhodrou and Saipa
would constitute a socially profitable merger, but the merger would harm those not participating in the
merger. This result may help to resove a paradox prevalent in the relevant literature, namely: how can
outsiders benefit form collusion among two firms? On the other hand we found no indication of the
merger’s profitability for the insiders.
Our results here also compare favorably, with those of Crean and Davidson’s (op cited) who
have used a multidivisional firm approach to answer the following three questions:
d) Is integration profitable for the participating firms (for the insiders)?
e) Is integration profitable for the non- participating firms (for the outsiders)?
f) How does integration affect the consumers of the commodities?
We have tried to answer the same questions, using the structure-conduct- performance approach
and resolve the paradox that exists in the horizontal mergers. In fact, the multidivisional firm approach
was proposes as an alternative to the joint profit maximization approach. It can be said that the merger
and behavior approach both draw on the joint profit maximization view with the difference that the
former uses the SCP approach and allows for the post-merger change in behavior. An application of
the SCP approach and the multidivisional –firm to vehicle manufacturing industry in Iran yields
identical results, although the latter application is not reported in this paper.

---

143
European Journal of Economics, Finance And Administrative Sciences - Issue 15 (2009)

References
1]
Bresnahan, F., 1987. "Competition and collusion in the American automobile industry: the
1955 price war", The Journal of industrial economics, 4: 457-482.
2]
Boyer, Marcel and Michael Moreaux, 1983.''Conjectures, rationality and duopoly theory'',
International Journal of Industrial Organization, 23-42.
3]
Bru, L. and Fauli-Oller, 2008. “Horizontal mergers and buyer power”, Economics Bulletin, 12,
1-7.
4]
Creane, A. and Davidson, C., 2001. "Multidivisional firms, internal competition and the merger
paradox”, Manuscript, Michigan state university.
5]
Deneckere, R. and Davidson, C., 1985. "Incentives to form coalitions with Bertrand
competition”, Rand Journal of Economics 16:473-486.
6]
Ekbo, B.E., 1983."Horizontal mergers, collusion, and stockholder wealth", Journal of Financial
Economics 11:241-273.
7]
Fellner, W.J., 1949. Competition Among the Few, Knopf, New York
8]
Fee, C.E. and Thomas, S., 2004."Sources of gains in horizontal mergers: evidence from
customers,supplier,and rival firms", Journal of Financial Economics 74:423-460
9]
Friedman, J.W., 1983. Oligopoly Theory, Cambridge University Press, Cambridge, MA
10]
Fisher, I., 1898. “On Mathematical economics,'' Quarterly Journal of Economics, vol.12,,119-
132
11]
Gaudet, G., and Salant, S.W., 1991."Increasing the profits of a subset of firms in oligopoly
models with strategic substitutes", American Economic review 81:658-665.
12]
Holt, C.A. 1980. ''Equilibrium models of tacit collusion in oligopoly experiments with price
setting firms'', Center for economic Research, university of Minnesota,
13]
Kamien. Schwartz, L., 1981."Conjectural variations", Managerial Economics and Decision
Sciences, Northwestern University, USA.
14]
Lommerud, K.E., Straume, O.R. and Sorgard, L., 2005."Downsream merger with upstream
market power".European Economic Review 49,717-743.
15]
Leontief, Wassily, 1936. “Stackelberg on monopolistic competition'', Journal of Political
Economy,,554-559
16]
Laitner, J., 1980.''Rational duopoly equilibrium'', Quarterly Journal of Economics, 641-662
17]
Perry, M.K. and Porter, R.R., 1985."Oligopoly and incentives for horizontal mergers".American
economic review, 75:219-227.
18]
Perry, K., 1982."Oligopoly and consistent conjectural variation", The Bell Journal of
Economics, 13:197-205.
19]
Riordan, M., 1985."Imperfect information and dynamic conjectural variations", Rand Journal
of Economics, 16:41-50.
20]
Salant, S., Switzer, S. and Reynolds, R.J., 1983."Losses from horizontal merger: The effects of
an exogenous change in industry structure on Cornuot-Nash equilibrium", Quarterly Journal of
Economics, 98:185-199.
21]
Stigler, George J., 1950. ''Monopoly and Oligopoly by Merger’’, American Economic Review
Proceeding, 23-34.
22]
www.irbourse.com
23]
www.cbi.ir

---