Political Contributions, Subsidy and Mergers AbstractWe examine, in a partial oligopolistic equilibrium model, the eﬀects of mergers andinternal groups (lobbies) in shaping national subsidy policies. Domestic and foreignﬁrms compete in the market for a homogeneous good in a host country, then theoptimal output of the foreign ﬁrms can be aﬀected ambiguously by the governmentsubsidy policy in the host country. Domestic ﬁrms oﬀer political contributions to thegovernment that are tied to the government’s policy decision. The government setsthe policy (subsidies) to maximize a weighted sum of total contributions and aggregatesocial welfare taking itno account merger of domestic ﬁrms.JEL Classiﬁcation: F12, F13Keywords: Foreign Direct Investment, Mergers, Lobby.
1 Political Contributions, Subsidy and Mergers By M. Ozgur Kayalica† and Rafael Espinosa-Ramirez‡ AbstractWe examine, in a partial oligopolistic equilibrium model, the eﬀects of mergers andinternal groups (lobbies) in shaping national subsidy policies. Domestic and foreignﬁrms compete in the market for a homogeneous good in a host country, then theoptimal output of the foreign ﬁrms can be aﬀected ambiguously by the governmentsubsidy policy in the host country. Domestic ﬁrms oﬀer political contributions to thegovernment that are tied to the government’s policy decision. The government setsthe policy (subsidies) to maximize a weighted sum of total contributions and aggregatesocial welfare taking itno account merger of domestic ﬁrms.JEL Classiﬁcation: F12, F13Keywords: Foreign Direct Investment, Mergers, Lobby.† Department of Economics, Sakarya University, Adapazari, Turkey.(firstname.lastname@example.org)‡ Department of Economics, University of Guadalajara, Guadalajara, Mexico.(email@example.com)Mailing address: Rafael S. Espinosa-Ramirez, Department of Economics, University ofGuadalajara, Mexico. TEL: +52-33-37703300 ext. 5579, FAX: +52-33-37703300 ext. 5214,e-mail: firstname.lastname@example.org
21 IntroductionAs an important element of global economic activity, Foreign Direct Investment (FDI)has received enormous attention from scholars worldwide.1 Bulk of the literature con-siders the host country government as a social welfare maximising agent. In reality,governments design their policies not only according to welfare concerns, but also inresponse to the interests of organised lobby groups.2 The Trade Related InvestmentMeasures (TRIM) agreement that is based on the GATT principles on trade in goodsand regulates foreign investment does not govern the entry and treatment regulationsof FDI, but focuses on the discriminatory treatment of imported and exported prod-ucts and not the services. This suggests that national governments can encourage ordiscourage foreign investors in a discriminatory manner by choosing the policy toolsthat do not have a direct eﬀect on international trade. Therefore, the political pro-cesses generating economic policy is likely to be aﬀected by pressure groups as far asforeign investment is concerned. There are many models in the international trade literature that uses politicalprocess. These include the tariﬀ-formation function approach of Findlay and Wellisz(1982), the political support function approach of Hillman (1989), the median-voter 1 See, for example, Brander and Spencer (1987), Ethier (1986), Hauﬂer and Wooton (1999), Help-man (1984), Horstmann and Markusen (1987), and (1992), Itagaki (1979), Janeba (1995), Lahiriand Ono (1998a) and (1998b), Markusen (1984), Markusen and Venables (1998), Motta (1992), andSmith (1987). 2 For instance, almost all countries have well-organised local producers (such as the automobileindustry) who lobby the government for higher levels of protection against imported goods or againstthe goods of the foreign-owned plants producing in the country.
3approach of Mayer (1984), the campaign contributions approach of Magee et al (1989),and the political contributions approach of Grossman and Helpman (1994a).3 The use of political competition in the theory of FDI stems back to Bhagwati(1985) and his notion of quid pro quo (protection-threat-induced FDI) in which hestudies how the protection threats aﬀect FDI entry. Takemori and Tsumagari (1997)focus on whether FDI is helpful in reducing the protectionism, and thus achieving freetrade. Hillman and Ursprung (1993) also explore how the presence of FDI aﬀects theemergence of protection. They develop a model where both national and multinationalﬁrms lobby for protection in the jurisdictions where they have plants. Unlike the authors above, Ellingsen and W¨rneryd (1999) deploy a model where athe domestic industry does not want the maximum protection and lobby for lessprotection. They argue that this is because a high level of protection could induceFDI to jump the trade barriers and even may be more harmful for the domestic ﬁrms.Konishi et al (1999), using common agency framework, construct political economymodel in which the choice of protection (between tariﬀ and voluntary export restraint)is endogenously determined. Grossman and Helpman (1994b) combines quid pro quoFDI with their political contributions approach and develop a model in which tradepolicy and FDI are endogenously and jointly determined. In all these works, tradepolicies, such as tariﬀ and quota, are the only policy instruments available to thegovernment. In this work we develop a partial equilibrium model of an oligopolistic industry 3 The literature has been surveyed in several works, including Magee, Brock and Young (1989),and Rodrik (1995).
4in which a number of domestic and foreign ﬁrms compete in the market for a homoge-neous good in a host country. It is assumed that the optimal output and competitionstrategies of foreign ﬁrms can be aﬀected by government policy in the host country.The host country government uses two types of per unit subsidies to impact the opti-mal output of foreign ﬁrms. This distinguishes our model from the works mentionedin the previous paragraph since we allow the government to use subsidies instead ofdirect trade policies like tariﬀ and quota. Moreover, we allow uniform policies tosee how the behavior of lobby group changes when receiving the same beneﬁt as theforeign ones. Lobbying is modeled following the political contributions approach. Domesticﬁrms oﬀer political contributions to the government which are tied to the government’spolicy choices. Then, the government sets the policy to maximise a weighted sumof total contributions and aggregate social welfare. Lobbying in our model has thestructure of the common agency problem explored by Bernheim and Whinston (1986),which is later used by Grossman and Helpman (1994a) to characterise the politicalequilibrium under trade protection and ﬁnally generalised by Dixit, Grossman andHelpman (1997) for wider economic applications. On the other hand, during the period between 1990-2000, most of the growth ininternational production has been via cross-border Mergers and Acquisitions (M&As)rather than greenﬁeld investment. The total number of all M&As worldwide (cross-border and domestic) has grown at 42 per cent annually between 1980 and 1999. Thevalue of all M&As (cross-border and domestic) as a share of world GDP has risen from0.3 per cent in 1980 to 8 per cent in 1999 UNCTAD (2000).
5 Governments’ policy measures regulating M&A activities aﬀect the welfare ofbillions of consumers, as discussed in Benchekroun and Chaudhuri (2006), as well asthe welfare of other economic agents such as employees and employers. For example,Bhattacharjea (2002) claim that if foreign mergers and export cartels can be treatedas a reduction in the eﬀective number of foreign ﬁrms, this can actually reduce homewelfare below the autarky level, as the free-rider beneﬁts that greater concentrationbestows on domestic ﬁrms who are not party to the merger are insuﬃcient to com-pensate for the loss inﬂicted on domestic consumers. This is a very serious regulatoryissue in the world economy. The countries should pursue local and international poli-cies in order to regulate possible unfair competitive strategies in case of mergers. Thisquestion has been addressed by Bhagwati (1993), Gatsios and Seabright (1990) andNeven (1992). These researches claim that the regulatory policies should be subjectto international negotiations or assigned to higher levels of government.4 Bulk of thestudies in the literature analyse the aﬀect of foreign mergers on welfare. Domestic ﬁrms also merge for several reasons, for instance in order to obtaincompetitive advantage against foreign rivals. Mergers of domestic ﬁrms appear to bea surviving strategy. Following this line, Collie (2003) develops a signiﬁcant paper on 4 In 2002, The Fair Trade Commission (FTC) of the South Korea government announced that itwould introduce regulations by the end of that year. The FTC claimed that this would allow it to trackmergers between foreign ﬁrms which could seriously impair relevant domestic industries. FTC signedan agreement with Australia in 2003 for the mutual application of Korea’s fair competition law andwould pursue similar agreements with the United States, European Union and Japan. Similarly, theEuropean Commission has regulated mergers between foreign ﬁrms when they are aﬀecting negativelythe European interests.
6mergers of local and foreign ﬁrms and trade policy under oligopoly. Ross (1988) showsthat a domestic merger driven by ﬁxed cost savings leads to lower price increases inthe face of unilateral tariﬀ reduction than otherwise. In a two country oligopolisticmodel, Long and Vousden (1995) show that bilateral tariﬀ reductions increase theproﬁtability of a domestic merger when the asymmetry between the merging ﬁrmsis large enough. Benchekroun and Chaudhuri (2006) show that trade liberalizationalways increases the proﬁtability of a domestic merger (regardless of the cost-savingsinvolved). Espinosa and Kayalica (2007) analyse the interface between environmentalpolicies and domestic mergers externalities. Despite these works, domestic mergershave been an issue not explored enough by the economic literature. Under the above speciﬁcation, we examine aspects of the political relationshipbetween the government and the domestic ﬁrms under diﬀerent degrees of corruption.In other words, the optimal policies in the absence of lobbying are also analysed tosee how policies change by pressure from the interest group. The basic structure isgiven in the next section where we use a lobbying framework that follows Grossmanand Helpman (1994a) and Dixit et al (1997). In the third section we analyse thecomparative static of discriminatory and uniform subsidy. The optimal discriminatorysubsidy and merger in domestic ﬁrms is analysed in the fourth section. In this sensein the ﬁfth section we consider merger and uniform subsidy. We conclude in the lastsection.
72 The Basic Framework: LobbyingWe consider an economy in which there are m identical domestic ﬁrms and n identicalforeign ﬁrms. The domestic ﬁrms form a lobby group whose political contributionschedule is deﬁned by C(s), where s is a per unit subsidy determined by the govern-ment, which we examine in detail in the next section5 . Each domestic ﬁrm has thefollowing utility, V d = π d − C, (1)where π d is the proﬁt of a domestic ﬁrm. Consumers have identical quasi-linear prefer- ¯ences and are given some exogenous level of income, Y 6 . We denote the consumptionof the non-numeriare good by D, while function f is increasing and strictly concave ¯in D. Hence, with income Y each individual consumes D = g(p) of the non-numeriare ¯good and y = Y − pg(p) of the other goods (where p is the price of non-numeriaregood). We can then derive the consumers’ indirect utility. ¯ V c = CS + Y (2) p=p∗where CS is the consumer surplus (CS = p=0 f (g(p)) − pg(p)). The governmentcollects the subsidy cost from consumers by taxing. We denote the total cost of thesubsidy by T R. The government’s objective can be written as G = ρmC + (V d m + V c − T R) (3) 5 The model is adapted from Kayalica and Lahiri (2007) which developed a similar framework. 6 The preferences of the consumers are represented by u(y, D) = y + f (D) where y is the consump-tion of a numeriare good produced under competitive conditions with a price equal to 1. There isalso just one factor of production whose price is determined in the competitive sector.
8where ρ > 1 is a constant parameter we call corruption level, so the ﬁrst term in (3)is the political contribution impact of the m ﬁrms on government objective function.7The second term in (3) is the total social welfare. The political equilibrium can be determined as the result of a two-stage gamein which the lobby (representing domestic ﬁrms) chooses its contribution schedule inthe ﬁrst stage and the government sets the level of subsidy in the second. Then, thepolitical equilibrium consists of a political contribution schedule C ∗ (s), that maximisesthe proﬁts of all the domestic ﬁrms given the anticipated political optimisation by thegovernment, and a subsidy level, s∗ , that maximises the government’s objective givenby (3), taking the contribution schedule as given. As discussed in Dixit et al (1997), the model can have multiple sub-game equi-libria, some of which may be ineﬃcient. Dixit et al (1997) develop a reﬁnement thatselects truthful equilibria that result in Pareto-eﬃcient outcomes.8 Stated formally, 0 0let C 0 (s0 , V d ), s0 be a truthful equilibrium in which V d is the equilibrium utility 0 0level of each domestic ﬁrm. Then, C 0 (s0 , V d ), s0 , V d is characterised by 0 C(s, V d ) = Max(0, A) (4) 7 Using equations (1) and (2) government’s objective function can also be written as G = ρCm + ¯ ¯(π d m−Cm+CS+ Y −T R). Reorganizing the equation, we get G = (ρ−1)Cm+(π d m+ Y +CS−T R).Hence, government attaches a positive weight to contributions provided that ρ > 1. In other words,there is no political relationship between the government and the domestic ﬁrms when ρ = 1. Theweight that the government attaches to social welfare is normalised to one. 8 Bernheim and Whinston (1986) develop a reﬁnement in their menu-auction problem. Followingthis, ﬁrst Grossman and Helpman (1994a) and later Dixit et al (1997) develop a reﬁnement (as inBernheim and Whinston (1986)) for the political contribution approach, that selects Pareto-eﬃcientactions.
9 0 0 s0 = Argmaxs ρC(s, V d )m + V d m + (V c − T R)(s) (5)and V d (s1 )m + V c (s1 ) − T R(s1 )) = 0 0 ρC(s0 , V d )m + V d (s0 )m + V c (s0 ) − T R(s0 ) (6)where V c is deﬁned in (2) and 0 Vd = πd − A (7) s1 = Argmaxs V d (s)m + V c (s) − T R(s) (8)Equation (4) characterises the truthful contribution schedule chosen by the lobby,where A can be interpreted as the compensation variation. Hence, equation (4) (to- 0gether with (7)) states that the truthful contribution function C(s, V d ) relative to 0the constant V d is set to the level of compensating variations. In other words, undertruthful contribution schedules the payment to the government is exactly equal to thechange in domestic ﬁrms’ proﬁts that is caused by a change in policy s (see Dixit etal (1997, p.760)). Equation (5) states that the government sets the subsidy level tomaximise its objective, given the contribution schedule oﬀered by the domestic ﬁrms. Equation (6) implies that in equilibrium the contribution of the lobby has toprovide the government at least the same level of utility that the government could getif it did not accept any contributions. The lobby pays the lowest possible contributionto induce the government to set s0 deﬁned by (5). Then, the government will beindiﬀerent between implementing the policy (s1 ), by accepting no contributions andimplementing the equilibrium policy (s0 ) and accepting contributions. In the ﬁrst
10case, contribution would be zero and the government would maximise its objectivefunction as if the domestic ﬁrms were politically unorganised.9 Totally diﬀerentiating (3) we get dG = ρmdC + dCS − dT R (9)where, diﬀerentiating (4) (and (7))10 dC = dπ d (10)When ρ = 1 equation (9) serves for the case in which the government refuses theﬁrms’ contributions, and simply maximises the social welfare. That is, when ρ = 1 weobtain s1 deﬁned by (8). Equation (9) helps us to examine the public policy outcomeof political relationship between the government and the lobby. After analysing theequilibrium subsidy level, we focus on the eﬀects of mergers and the optimal respondof the government facing a decrease in welfare. It is a well known fact that dCS = −Ddp. (11) Having described the political equilibrium, we shall now introduce the rest ofthe model. We consider an oligopolistic industry with m identical domestic ﬁrms andn identical foreign ﬁrms. The marginal costs of the domestic and foreign ﬁrms arecd and cf respectively. These marginal costs are assumed to be constant, and thusthey also represent average variable costs. The domestic and foreign ﬁrms compete in 9 Using (1) to (3) it can be seen that the government does not accept any contribution at all whenρ = 1. 10 Assuming A > 0 we have A(·) = C(·).
11the domestic market of a homogeneous good. The inverse demand function for thiscommodity is given by11 p = α − βD, (12)where D is the sum of outputs by domestic and foreign ﬁrms, i.e., D = mxd + nxf , (13)where xd and xf are the output of a domestic and a foreign ﬁrm. We examine opti-mal subsidy levels when the government imposes discriminatory and uniform subsidypolicies. Proﬁts of a domestic and a foreign ﬁrm are respectively given by π d = (p − cd + sd )xd (14) π f = (p − cf + sf )xf (15)where sd and sf are respectively the per unit subsidies imposed on the domestic andforeign ﬁrms, with negative values of s representing taxes. It is assumed that the domestic and foreign ﬁrms behave in a Cournot-Nashfashion. Each ﬁrm makes its output decision by taking as given output levels setby other ﬁrms, the number of ﬁrms, and the subsidy level set by the government.The equilibrium is deﬁned by a three-stage model: ﬁrst, the government chooses thesubsidy level taking everything else as given; in the second stage, the number of foreignﬁrms is determined given the level of subsidy and output levels; ﬁnally, output levelsare determined. 11 The inverse demand function is derived from one speciﬁc case of the preferences mentioned inthe beginning of this section. That is, u(y, D) = y + αD − βD2 /2.
12 Using (14) and (15) we ﬁnd the ﬁrst order proﬁt maximisation conditions as βxd = (p − cd + sd ), (16) βxf = (p − cf + sf ), (17) Using (12) to (17) we ﬁnd the following closed form solutions π f = β(xf )2 , (18) π d = β(xd )2 , (19) α + m(cd − sd ) − (m + 1)(cf − sf ) xf = , (20) β∆ α + n(cf − sf ) − (n + 1)(cd − sd ) xd = , (21) β∆ Where ∆ = m + n + 1 > 0. Now we have the backbone of our analysis. Weshall proceed to analyse the eﬀect of subsidies on government objective function.3 Comparative StaticAfter setting the model, we are going to consider the eﬀect of subsidies on domesticproﬁts (and therefore on political contribution), consumer surplus and the subsidycost. In this section we discuss the case of comparative static of discriminatory anduniform (sd = sf = sU ) subsidies. Discriminatory subsidy to domestic ﬁrms it is not from our interest given thestrategy feature of domestic merger. In other words, if merger is a competitive strategyof domestic ﬁrms, there is nos sense to talk about a merger in response to a domesticﬁrm subsidy. We shall now totally diﬀerentiate (14) and using the closed form solutions(16-21) to get
13 2nxd f dπ d = − ds (22) sd =0 ∆ 2xd U dπ d = ds . (23) sf =sd =sU ∆ Equation (22) states that when only foreign ﬁrms are subsidised, the proﬁts ofthe domestic ﬁrms go down. This is because subsidising foreign ﬁrms gives them acompetitive advantage over the domestic ﬁrms due to a cost reduction. On the other hand, equation (23) states that a uniform subsidy will increasethe domestic proﬁts. Even when a subsidy to foreign ﬁrms gives them a competitiveadvantage, the positive impact on cost reduction on domestic ﬁrms is larger than thaton foreign ﬁrms. Next, the eﬀect on consumer surplus can be found by using (11), (12) and theclosed form solutions (16-21) as nD f dCS|sd =0 = ds , (24) ∆ (n + m)D U dCS|sd =sf =sU = ds . (25) ∆ Subsidising the foreign ﬁrms reduce the cost of foreign ﬁrms and increase thecost of domestic ﬁrms. However, the cost reduction in foreign ﬁrms is larger thanthe cost increase in the domestic ﬁrms and it will increase the total output consumedand therefore the market price is reduced. A discriminatory subsidy to foreign ﬁrmswill increase the consumer surplus. On the other hand a uniform subsidy will reducecost in both foreign and domestic ﬁrms increasing the output and reducing the price,increasing the consumer surplus.
14 Finally, the total cost of ﬁnancing per unit subsidy is deﬁned as T R = sd mxd + sf nxf . (26)From total diﬀerentiation of (26) and using again (16-21) we get the following generalexpression sf n(m + 1) dT R|sd =0 = nxf + dsf , (27) β∆ sU (m + n) dT R|sd =sf =sU = D+ dsU . (28) β∆ Needless to say, subsidising foreign and domestic ﬁrms (in a discriminatory oruniform way) increases the total cost of subsidy. So far, it is clear that subsidisingthe ﬁrms has opposing eﬀects on the various components of government’s objectivefunction or welfare, as we will mention in the following sections.4 Discriminatory Subsidy and MergersHaving described the general framework above, in this section we shall begin ouranalysis with the case when the government uses a discriminatory policy, namelysubsidising the foreign ﬁrms but not the domestic ones. Substituting (22), (24), (27)in (9) we ﬁnd dG n = −2ρmxd + D − ∆xf − sf (m + 1) (29) dsf sd =0 ∆ As discussed above, subsidising the foreign ﬁrms has opposing eﬀects on welfarethrough its various components. The above equation reﬂects this ambiguity. Clearly asubsidy to foreign ﬁrm will reduce the beneﬁt of the domestic ﬁrms and therefore thecontribution made by them. It can be seen in the ﬁrst term inside the square brackets
15in (29). On the other hand, a foreign subsidy will increase the consumer surplusaccording to the second term in (29). Finally the last two terms tell us that ﬁnancingthe subsidy to foreign ﬁrms produces a negative impact on welfare. Assuming G tobe concave in sf , we get the optimal subsidy equalizing (29) to zero as m sf ∗ = βxd (1 − 2ρ) − βxf < 0. (30) m+1 From (30) it is clear that the cost of subsidising foreign ﬁrms plus the loss inpolitical contribution is larger than the beneﬁt in consumer surplus. In this case theoptimal subsidy will be unequivocally negative and taxing foreign ﬁrms will be theoptimal policy. Stating the above results formally,Proposition 1 In the absence of any policy towards the domestic ﬁrms, the optimalsubsidy to the foreign ﬁrms is negative Intuitively speaking from the domestic ﬁrms point of view, a discriminatorysubsidy seems to be an unfair policy for them. Even they may not know aboutthe optimal policy chosen by the government, the perception is that they must dosomething in order to compensate the political advantage given to foreign ﬁrms. Oreven if they know the optimal setting of the political policy, the domestic ﬁrms mayreact strategically in order to get some competitive advantage. One of the competitivestrategies used to gain some advantage over the competitors is merging. We shall now analyse the eﬀect of local merger when the optimal policy hasbeen set by the domestic government. It will be useful to review the eﬀect of mergeron welfare when he domestic government pursues an optimal per unit subsidy policy.12 12 In terms of value, about 70 per cent of cross-border Mergers and Acquisitions are horizontal (see
16Following Salant, Switzer and Reynolds (1983) the horizontal merger is modeled as anexogenous reduction in the number of domestic ﬁrms.13 We will analyse the eﬀect of achange in the number of ﬁrms m on welfare. This change is given by the diﬀerentiationof (9) with respect to m as dG dπ d dCS dT R = ρπ d + ρm + − (31) dm dm dm dm The ﬁrst and second term in the right hand of (31) show the change in thepolitical contribution given by the merger (the change in the domestic proﬁt and thecontributing number of ﬁrms). The third and fourth terms are the changes in theconsumer surplus and the cost of subsidising ﬁrms respectively. From (11)-(21) and(26) we get the eﬀect of merger in each component as dπ d 2π d dCS Dβxd dT R nxd =− < 0; = > 0; = −sf . (32) dm ∆ dm ∆ dm ∆ The eﬀect of domestic ﬁrms’ merger on domestic ﬁrms’ proﬁts is positive asmerger increase the market share for domestic ﬁrms. A reduction in the number ofdomestic ﬁrms will reduce the consumer surplus due to a reduction in the amount ofoutput available to consume and therefore the price increases. Less domestic ﬁrmsmeans more subsidy to be paid by the government because of the increase in themarket share, so a merger will increase the expense in subsidy made by government.14Substituting (32) in (31) and using (30) we getUNCTAD (2000, p.xix.)). 13 Although the number of domestic and foreign ﬁrms take an integer value, it will be treated as acontinuous variable. 14 In our case, with a negative optimal subsidy, a merger means more tax revenue.
17 dG πd = [ρ(1 − m) + m] . (33) dm m+1 Once the optimal policy has been set, there are opposite eﬀects of merger onwelfare. First of all the political contribution presents an ambiguous result in thepresence of domestic merger. From the two ﬁrst terms in (31) and using (32) we have d(ρmπ d ) ρπ d = [n + 1 − m] . (34) dm ∆ In this case a merger will reduce the number of contributing ﬁrms but increasethe proportion each remaining ﬁrm contributes to the political lobby. The net eﬀectwill depend on the number of competing domestic and foreign ﬁrms in the market.When larger is the number of remaining domestic ﬁrms respect to foreign ﬁrms, thecontribution oﬀered by the remaining domestic ﬁrms is larger than the loss in contri-bution given by a reduction in m. In opposition to this, a larger number of foreignﬁrms over domestic ﬁrms will reduce the share of contribution that each domestic ﬁrmoﬀers to the government and the merger will reduce the political contribution. On the other hand, as mention before, a merger will reduce the consumersurplus because the amount of ﬁrms producing the consumed output is reduced andso the total production, increasing the price and reducing the consumer surplus. In thesame sense a merger means a reduction in competing ﬁrms, so in this case the amountof output produced by foreign ﬁrms increase and, given that the optimal policy is aper unit subsidy, the amount of tax revenue increases. After this explanation, from (33) we can see that the net eﬀect of merger onwelfare is going to depend on the number of domestic ﬁrms. In an extreme case, if
18the merger leads us into a monopoly (m = 1), the monopolist would be unable tooﬀer a larger contribution than that oﬀered by two or more ﬁrms. In this case amerger will reduce the beneﬁt in welfare and this result is independently of the levelof corruption.15 Only with a suﬃciently large amount of domestic ﬁrms over foreign ﬁrms, themerger will increase the political contribution according to (34). In this case, thebeneﬁt in political contribution and tax revenue is larger than the loss in consumersurplus. A merger will increase the welfare. Formally we can sayProposition 2 When the government applies discriminatory subsidy to foreign ﬁrms,a merger of domestic ﬁrms will increases the welfare when m >> n. On the otherhand, it will be reduced when m = 1. Finally, to ﬁnish this section we follow the analysis made by Collie (2003).When a local merger reduces the welfare, the government tries to correct this negativeexternality using the policy instrument. In this case, when the government pursuesan optimal subsidy to foreign ﬁrms, how should the domestic country governmentrespond to a local merger? In order to solve this question, we obtain the comparativestatic of a reduction in the number of local ﬁrms on the optimal subsidy policy suchthat dsf ∗ πd = [(1 − 2ρ)(n + 2ρ) + (m + 1)] . (35) dm (m + 1)∆ 15 This result can be also reached when the corruption is suﬃciently small (ρ 1), and the politicalcontribution is negligible making the loss in consumer surplus larger than the beneﬁt in tax revenue.However, we do not consider this case here as we assume ρ > 1 since the beginning.
19 Taking into account that the government is going to respond politically to anylocal merger as long as it aﬀects negatively the welfare, the conditions under whichhappen this situation is when merger lead us into a domestic monopoly in the country(m = 1).16 Assuming these values on (35) we can rewrite it as dsf ∗ πd = [(1 − 2ρ)(n + 2ρ) + 2] . < 0. (36) dm 2∆ This result is unequivocally negative and, since the optimal subsidy is negative(a tax), the optimal response is a tax reduction over the foreign ﬁrms. Formally wecan sayProposition 3 When the government applies discriminatory subsidy to foreign ﬁrms,the optimal response of the domestic country to a local merger is to decrease the taxlevied to foreign ﬁrms. The intuition behind is quite straightforward. Once the optimal policy hasbeen set by government, evaluating not only the impact on consumer surplus and thetotal proﬁts of domestic ﬁrms but also the beneﬁt on tax revenue, the domestic ﬁrmsreact and merge in order to get better proﬁts by obtaining monopolistic advantages.Then the government is willing to reduce the tax levied to foreign ﬁrms in order tostimulate the competition and increase the consumer surplus by reducing prices. Theconsumer surplus is the most important consideration since the contribution has anegative impact on welfare given by the monopoly condition. 16 As mentioned before when government objective function increase with a merger of local ﬁrms,the government does not have incentives to change the optimal policy and therefore we ignore theanalysis.
205 Uniform Subsidies and MergersHaving described the case in which we have a discriminatory subsidy addressed toforeign ﬁrms, we shall follow our analysis with the case when the government uses auniform subsidy. As we mentioned before, we are not going to analyse the case ofdiscriminatory domestic subsidy as the local merger is a competing strategy that doesnot ﬁt with a domestic subsidy. A uniform subsidy is a fair policy to both kind of ﬁrms. Diﬀerent to thediscriminatory subsidy, where the lobbying made by domestic ﬁrms determine thepolitical contribution in clear opposition to a discriminatory subsidy in favor of foreignﬁrms, in the case of uniform subsidy the lobby eﬀort is made in order to receive moresubsidy even it is uniformly equal between foreign and domestic ﬁrms. More subsidymeans more contribution oﬀered by domestic ﬁrms despite the foreign ﬁrms beneﬁt. Substituting (23), (25), (28) in (9) (with (10)) we ﬁnd dG 1 = 2ρmβxd + (m + n)βD − β∆D − sU (m + n) (37) dsU sd =sf =sU β∆ As mentioned previously, subsidising uniformly to both kind of ﬁrms has op-posing eﬀects on G through its various components. The above equation reﬂects thisambiguity. Clearly a uniform subsidy will increase the beneﬁt of the domestic ﬁrmand therefore the contribution made by them. It can be seen in the ﬁrst term insidethe square brackets in (37). On the other hand, a uniform subsidy will increase theconsumer surplus according to the second term in (37). Finally the last two terms in(37) tell us that ﬁnancing the subsidy to both ﬁrms produces a negative impact on
21government objective function. Assuming G to be concave in sU , we get the optimaluniform subsidy equalizing (37) to zero as β sU ∗ = mxd (2ρ − 1) − nxf . (38) m+n From (38) we have an ambiguous vale of the optimal uniform subsidy. It is clearthat the cost of subsidising ﬁrms is contrary to the beneﬁt in political contributionand consumer surplus. We can see that a larger corruption level will produce a largerpositive perception about political contribution by government. If it is the case, thebeneﬁt given by political contribution and consumer surplus will be larger than theloss given by ﬁnancing subsidy and the optimal policy will be a subsidy. However, if the corruption level is small enough and so the political contri-bution, the optimal uniforms subsidy will depend on the eﬃciency of domestic andforeign ﬁrms. When the domestic ﬁrms are suﬃciently more eﬃcient than the foreignﬁrms, the government will adopt a subsidy since the beneﬁt of the domestic proﬁtsand consumer surplus is larger than the cost for subsidising uniformly both kind ofﬁrms. On the other hand, when the foreign ﬁrms are suﬃciently more eﬃcient thanthe domestic ones, the domestic proﬁts are small and, despite the beneﬁt in consumersurplus, the cost for subsidising ﬁrms uniformly is larger than the beneﬁt in consumersurplus and domestic proﬁts. In this case the optimal subsidy will be negative andtaxing uniformly will be the optimal policy. Stating the above results formally,Proposition 4 In the presence of a uniform subsidy to domestic and foreign ﬁrms,the optimal subsidy will be if ρ >> 1, then sU ∗ > 0,
22 if ρ 1 and cd << cf (cd >> cf ), then sU ∗ > 0 (sU ∗ < 0). Even when a uniform subsidy is a fair policy, the domestic ﬁrms may take advan-tage of their local position and set a strategic behavior against the foreign competitors.As the last section, we consider merging as the competitive strategy implemented bydomestic ﬁrms once the optimal policy has been set. All the explanation and intuitionused in the last section respect to consumer surplus and political contribution eﬀectsof merger apply in this case. The only diﬀerence comes from the eﬀect of merger onthe cost of subidising. The cost of a uniform subsidy can be seen as T R = DsU . Diﬀerentiation of this expression with respect to m we get the eﬀect of mergeron the cost of subsidising as dT R xd = sU . (39) dm ∆ From (39) we can see that a merger will reduce the cost of subsidy both ﬁrmsas soon as the optimal policy is positive. Otherwise a merger will increase the taxrevenue. Substituting (38) in (39) and together with (32) (where apply) in (31) weget dG βxd = xd (ρ(n − m) + m) + nxf . (40) dm (m + 1) The eﬀect of a merger on the government objective function is ambiguous andit is going to depend on the level of corruption and the number of domestic and foreign
23ﬁrms. If merger leads us into a situation in which the number of foreign ﬁrms is largeror equal than the number of domestic ﬁrms (n ≥ m), then a merger will reduce the 17welfare. On the other hand, a merger can increase welfare if the number of domesticﬁrms is larger than the number of foreign ﬁrms and the corruption level is suﬃcientlylarge. Formally we can sayProposition 5 When the government applies uniform subsidies to foreign and do-mestic ﬁrms, a merger of domestic ﬁrms will reduce the government objective functionwhen n ≥ m. On the other hand, the government objective function will increase whenρ >> 1 and/or n < m. Intuitively speaking, with a merger the consumer surplus will be reduced un-equivocally. However in the ﬁrst case (n ≥ m), and according to (34), the policycontribution is reduced by merger because the amount of contributing ﬁrms is re-duced. The reduction in consumer surplus and policy contribution is larger than thereduction in the cost of subsidising ﬁrms. In this case with a merger in domestic ﬁrms,the welfare will be reduced On the other hand, when m > n and ρ >> 1, a merger of domestic ﬁrmswill reduce consumer surplus as the previous case, but the political contribution willincrease given by the larger market share enjoyed by the large number of domesticﬁrms despite the reduction in contributers according to (34). In brief, in the secondcase, a merger will promote a reduction in consumer surplus in smaller proportionthan the increase in the beneﬁt obtained by political contribution and the reduction 17 Although the same result can be obtained with no corruption level (ρ = 1), we just consider thatρ > 1 because for any level of corruption the condition n ≥ m holds.
24in the cost of subsidy. With a merger in domestic ﬁrms the welfare will increase. As in the previous section, we wonder how the government is going respond interms of the political policy as a result of a welfare’s decreasing local merger. Again,we will diﬀerentiate the optimal policy function (38) with respect to m, and we get dsU βxd nβxf = (2ρ − 1)(n2 + n − m2 ) + n(m + n) + (41) dm (m + n)2 ∆ (m + n)2 Considering only the condition under which the government objective functionis reduced by merger (n ≥ m), it is clear that (41) is positive. In this sense the optimalgovernment responses to a domestic merger is to decrease the optimal uniform subsidy.Formally we can sayProposition 6 When the government applies uniform subsidy to foreign and domesticﬁrms, the optimal response of the domestic country to a local merger is to decrease theuniform subsidy. Intuitively speaking, the fall in consumer surplus and political contribution willbe compensated by a reduction in subsidy cost. Even when a reduction in subsidymay aﬀect negatively the output produced by both ﬁrms through an increasing pro-duction cost, aﬀecting negatively the consumer surplus and the amount of contribution(already decreased by the merger of domestic ﬁrms), the government is willing to re-duce the cost of subsidy reducing the optimal uniform subsidy. This result is quiteinteresting as we may suppose that the optimal respond would be to increase the sub-sidy in order to beneﬁt from consumer surplus and political contribution. Howeverit seems that the beneﬁt produced by the reduction in the cost for subsidising over-
25comes the possible beneﬁt of increasing consumer surplus and political contributionindependently of the political corruption level.6 ConclusionIn this work we develop a partial equilibrium model where the foreign and domesticﬁrms compete under oligopolistic conditions. The government is endowed with perunit proﬁt subsidies (taxes) to impose on both groups of ﬁrms (discriminatory anduniformly) while facing political pressure from a special interest group representingthe domestic ﬁrms. Under this structure, the government maximises a weighted sumof the total political contributions from interest groups and aggregate social welfare. Using the above framework, we determine optimal policies in the presence oflobbying. We found that in the case of discriminatory subsidy for foreign ﬁrms, theoptimal policy is to tax foreign competitors when the government receives politicalcontributions from the domestic ﬁrms. In the case of uniform subsidies, when thegovernment is highly corrupted we show that the optimal subsidy is unequivocallypositive and a subsidy will be given to both types of ﬁrms. However, when the levelof corruption is suﬃciently small, there is practically an absence of lobbying. Thegovernment is only concerned with maximising the aggregate social welfare. In thiscase the optimal uniforms subsidy is going to depend on the relative eﬃciency of bothgroup of ﬁrms. In particular, we found that the optimal uniform subsidy is positive(negative), if the foreign ﬁrms are less (more) eﬃcient than the domestic ones. We also analyse how the mergers of domestic ﬁrms change the equilibriumlevels of subsidies and contribution payments. Our results show that, in the presence
26of lobbying, a merger of domestic ﬁrms is going to have diﬀerent results according tothe subsidy structure. In the case of a discriminatory subsidy, if merger leads us into amonopoly in domestic ﬁrms, then the welfare would be reduced given by a reduction incontribution and consumer surplus. This result is identical in the absence of corruptiongiven clearly by the null eﬀect of contribution in the government objective function.On the other hand, when the number of domestic ﬁrms is larger enough with respectto foreign ﬁrms, then the welfare will increase by a merger due to mainly an increasein contribution and the low level of monopolistic distortions. In the presence of a uniform subsidy, the eﬀect of a merger on welfare willdepend again on the number of foreign and domestic ﬁrms as well as the corruptionlevel. A merger will reduce welfare as soon as the number of foreign ﬁrms are equalor larger than the number of domestic ﬁrms. Diﬀerent to the discriminatory case, itis not required to have a domestic monopoly to have a welfare reduction, it is enoughif the foreign ﬁrms are at least the number of domestic ﬁrms. The explanation isthe same than in the discriminatory case. On the other hand, a merger will increasewelfare if the number of domestic ﬁrms is larger than the number of foreign ﬁrms,as in the discriminatory case, and the level of corruption should be large enough. Itmakes the political contribution signiﬁcant in the policy decision. Finally, we consider the optimal policy response of the government facing awelfare’s decreasing situation due to a merger in domestic ﬁrms. The result is quiteinteresting as we have an contrary responses in both cases. At the discriminatorycase, a merger in domestic ﬁrms will be answered decreasing the tax levied to foreignﬁrms (as the negative subsidy means a tax to foreign ﬁrms). The interest to reduce
27the monopolistic distortion seems to be the key consideration in the political decision.On the other hand, in the case of uniform subsidy, the optimal response will be toreduce the uniform subsidy to both types of ﬁrms. It seems that the cost of ﬁnancingis larger than the loss in consumer surplus and political contribution.
28 ReferencesBenchekroun, H. and Chaudhuri, A.R., 2006. Trade Liberalization and the Proﬁtabil-ity of Mergers: a Global Analysis. Review of International Economics, 14(5), pp.941-957.Bernheim, B. and Whinston, M., 1986. Menu auctions, resource allocation, and eco-nomic inﬂuence. Quarterly Journal of Economics, 101, pp. 1-31.Bhagwati, J.N., 1985. Investing abroad. Esme Fairbairn Lecture. Lancaster: Univer-sity of Lancaster Press.Bhagwati, J.N., 1993. Fair trade, reciprocity and harmonization: the novel challengeto the theory and policy of free trade. In: D. Salvatore, ed. 1993. Protectionism andworld welfare. Cambridge: Cambridge University Press. Ch.2.Bhattacharjea, A., 2002. Foreign Entry and Domestic Welfare: Lessons for DevelopingCountries. The Journal of International Trade and Economic Development, 11(2), pp.143-162.Brander, J.A. and Spencer, B.J., 1987. Foreign direct investment with unemploymentand endogenous taxes and tariﬀs. Journal of International Economics, 22, pp. 257-279.Collie, D.R., 2003. Mergers and trade policy under oligopoly. Review of InternationalEconomics, 11, pp. 55-71.Dixit, A., Grossman, G.M. and Helpman, E., 1997. Common Agency and Coordi-nation: General Theory and Application to Government Policy Making. Journal ofPolitical Economy, 105, pp. 752-769.Ellingsen, T. and W¨rneryd, K., 1999. Foreign direct investment and the political a
29economy of protection. International Economic Review, 40(2), pp. 357-379.Espinosa, R. and Kayalica, M.O., 2007. Environmental policies and mergers’ exter-nalities. Economia Mexicana: Nueva Epoca, XVI(1), pp. 47-74.Ethier, W.J., 1986. The multinational ﬁrm. Quarterly Journal of Economics, 101, pp.805-833.Findlay, R. and Wellisz, S., 1982. Endogenous tariﬀs, the political economy of traderestrictions, and welfare. In: J.N. Bhagwati, ed. 1982. Import competitionand andresponse. Chicago: University of Chicago Press. Ch. 8.Gatsios, K. and Seabright, P., 1990. Regulation in the European Community. OxfordReview of Economic Policy, 5(2), pp. 37-60.Grossman, G. and Helpman, E., 1994a. Protection for sale. American EconomicReview, 84, pp. 833-850.Grossman, G. and Helpman, E., 1994b. Foreign investment with endogenous protec-tion. National Bureau of Economic Research. Working Papers 4876.Hauﬂer, A. and Wooton, I., 1999. Country size and tax competition for foreign directinvestment. Journal of Public Economics, 71, pp. 121-139.Helpman, E., 1984. A simple theory of trade with multinational corporations. Journalof Political Economy, 92, pp. 451-471.Hillman, A. L., 1989. The political economy of protection. Chur: Harwood AcademicPublishers.Hillman, A.L. and Ursprung, H., 1993. Multinational ﬁrms, political competition, andinternational trade policy. International Economic Review, 34, pp. 347-363.Hortsman, I. and Markusen, J., 1987. Strategic investments and the development of
30multinationals. International Economic Review, 28, pp. 109-121.Hortsman, I. and Markusen, J., 1992. Endogenous market structures in internationaltrade (natura facit saltum). Journal of International Economics, 32, pp. 109-129.Itagaki, T., 1979. Theory of the multinational ﬁrm: An analysis of eﬀects of govern-ment policies. International Economic Review, 20(2), pp. 437-448.Janeba, E., 1995. Corporate income tax competition, double taxation, and foreigndirect investment. Journal of Public Economics, 56, pp. 311-325.Kayalica, M. O. and Lahiri, S., 2007. Domestic Lobbying and Foreign Direct Invest-ment. The role of policy instruments. Journal of International Trade & EconomicDevelopment, 16(3), pp. 299-323.Konishi, H., Saggi, K. and Weber, S., 1999. Endogenous trade policy under foreigndirect investment. Journal of International Economics, 49(2), pp. 289-308.Lahiri, S. and Ono, Y., 1998a. Foreign direct investment, local content requirement,and proﬁt taxation. The Economic Journal, 108, pp. 444-457.Lahiri, S. and Ono, Y., 1998b. Tax policy on foreign direct investment in the presenceof cross-hauling. Weltwirtschaftliches Archiv, 134, pp. 263-279.Long, N.V. and Vousden, N., 1995. The eﬀects of trade liberalization on cost-reducinghorizontal mergers. Review of International Economics, 3, pp. 14155.Magee, S.P., Brock, W.A. and Young, L., 1989. Black hole tariﬀs and endogenouspolicy theory. Cambridge: Cambridge University Press.Markusen, J.R., 1984. Multinationals, multi-plant economics, and the gains fromtrade. Journal of International Economics, 16, pp. 205-226.Markusen, J.R. and Venables, A.J., 1998. Multinational ﬁrms and the new trade
31theory. Journal of International Economics, 46, pp. 183-203.Mayer, W., 1984. Endogenous tariﬀ formation. American Economic Rewiev, 74, pp.970-985.Motta, M., 1992. Multinational ﬁrms and the tariﬀ jumping argument: a game the-oretic anlysis with some unconventional conclusions. European Economic Review, 36,pp. 1557-1571.Neven, D., 1992. Regulatory reform in the European Community. American EconomicReview, 82, pp. 98-103.Rodrik, D., 1995. Political economy of trdae policy. In: G. Grossman and K. Rogoﬀ,eds. 1995. Handbook of International Economics, 3. Amsterdam: North-HollandPublishing House. Ch. 28.Ross, T.W., 1988. On the price eﬀects of mergers with free trade. InternationalJournal of Industrial Organisation, 6, pp. 23346.Salant, S.W., Switzer, S., and Reynolds, R.J., 1983. Losses due to merger: the eﬀectsof exogenous change in industry structure on Cournot-Nash equilibrium. QuarterlyJournal of Economics, 98(2), pp. 185-200.Smith, A., 1987. Strategic investment, multinational corporations and trade policy,European Economic Review 31, pp. 89-96.Takemori, S. and Tsumagari, M., 1997. A political economy theory of foreign invest-ment: an alternative approach. Japan and the World Economy, 9, pp. 515-531.UNCTAD, (2000). World investment report 2000: Cross-border Mergers and Acquisi-tions and Development, Geneva: United Nations Publications.