Tariff jumping and joint ventures.

Beladi, Hamid ; Marjit, Sugata ; Chakrabarti, Avik 等

1. Introduction

A major incentive for foreign direct investment (FDI) in countries that restrict trade is to jump tariff and non-tariff barriers. Local production helps in reducing costs due to protection, and therefore "tariff jumping" stands out as a strong motivation behind FDI. Apart from locational advantages arising out of technology, marketing, or distributional factors, trade policy itself may be a reason that multinational firms would be lured by protected markets. When market size is large it makes sense to put additional investment in place so that basic advantages enjoyed by local firms can also be shared by the foreign competitors. In recent years a large number of foreign investors have entered the Indian automobile market, which still enjoys a significant degree of protection. Most of such investments are in the form of joint ventures, in which well-known foreign firms tie up with major Indian companies to grab a share of the large domestic market. Recently India has increased the ceiling on foreign equity holdings and liberalized foreign investment regulations to attract more foreign investment for the success of the ongoing reform program. It is logical to argue that the higher the existing tariff rate, the greater must be the incentive to jump tariff. This in turn implies a greater possibility of FDI if high tariffs push profits from exports to a low level. The purpose of this paper is to show that this is not necessarily true. In fact, we argue that very high tariffs may actually discourage FDI in the presence of a strong local competitor. There might be situations in which the only feasible form of investment is through a joint venture between the foreign firm and the local competitor. In the presence of a high tariff the success of such a joint venture is impossible to guarantee.

Generally speaking, the issue of FDI in transition economies is quite appealing. Countries are often forced to reduce trade barriers gradually while trying hard to woo foreign investors in the meantime. This may naturally lead to a situation in which tariffs are still in place but the barriers to FDI have been lifted. Analysis of investment strategy of the foreign firms in that situation is of interest to us and, we believe, to a number of people interested in policy issues. The industrial economics of ownership in developing countries has been the subject of much debate, but the economic literature on this issue is relatively sparse. Saggi and Pack (2006), in their critical survey of the analytical literature on industrial policy, shed light on this debate. The tariff discrimination hypothesis, dating back to Mundell (1957), holds that to avoid . obstacles in trade, resulting from the imposition of a tariff, foreign investment is undertaken in the country to which it is difficult to export because of the tariff obstacle; trade liberalization allows goods to move freely and, hence, is expected to reduce the amount of international investment. (1) The literature on joint ventures is growing, (2) though analytical models dealing with trade policy and joint ventures, to the best of our knowledge, are rare. (3) This paper is intended to be a contribution in this area.

A few early papers by Smith (1987) and Motta (1992) set the grounds by highlighting the effect of trade policy on joint ventures. In particular, they show that though high tariffs may deter foreign equity participation, there exist a large number of possibilities depending on the parameters of the models. Smith (1987) showed that, depending on the nature of the oligopolistic equilibrium, tariffs may or may not induce FDI, they may or may not change the market structure, and they may have pro- or anti-competitive effects. Motta (1992) demonstrated that no simple relationship emerges between the trade costs and the choice between direct investment and export: The existence of a tariff may cause a shift away from foreign investment or else may induce tariff-jumping investment. Among more recent contributions, Marjit, Beladi, and Kabiraj (2007) have shown how a tariff on a reputed brand product affects the conditions for joint venture. Beladi and Chakrabarti (2008) demonstrate how an interaction between ownership rules and trade policy provides a rationale for a host government to impose restrictions on a foreign multinational.

Our paper is distinct from these contributions on various counts. First, we analyze joint ventures between a local firm and a foreign firm and demonstrate that high levels of tariffs do deter joint ventures under very general conditions. Second, most of the existing contributions focus on "incumbent-entrant"--type models and the effects of foreign investment and tariffs on the entry of the local firms. Our primary concern in this paper is with a tough local competitor who may be weakened by the removal of a tariff, but who is in the market. In the earlier papers, higher anticipated tariffs may have induced local firms to enter and hence may have deterred FDI. But if the local firm exists independently of the tariffs, a higher tariff increases the incentive for FDI. (4) We rule out the possibility of "go-alone" foreign investment to focus on joint ventures. High tariffs, as we shall show, would always deter joint ventures. This result does not depend on parametric configurations. Third, in the standard cases, if tariffs deter FDI, there is no rationale as to why the firms cannot tie up in a joint venture. Our paper argues and demonstrates that such contracts, even if enforceable, may not work out with high tariffs, but they do materialize for a certain range of tariffs.

The rest of the paper is organized as follows. Section 2 describes our basic model and propositions, section 3 introduces asymmetries in cost between the foreign firm and the local firm, and section 4 concludes the paper.

2. The Basic Model

Our model starts with two firms: "i" denotes the foreign firm and "2" the local firm. Initially the foreign firm exports the product to the local market, which is also served by the local competitor. Firms are engaged in a duopolistic Cournot game. To start with, let us assume that firm 1 faces a per-unit tariff, t, and that the firms have the same constant marginal costs, c. The reduced form profit functions (5) for the ith firm (i = 1, 2) is given by

[[pi].sub.i] = [[pi].sub.i](c + t,c), i= 1, 2, (1)

where Equation 1 has the following properties:

[partial derivative][[pi].sub.1]/[partial derivative]t < 0, [partial derivative][[pi].sub.2]/[partial derivative]t > 0.

The process of FDI requires a sunk cost, I, to be incurred by firm 1. Local production will imply an increase in [[pi].sub.1] as t is effectively reduced to zero. The [[pi].sub.10] denotes the profit of the foreign firm before a reduction in t and [[pi].sub.1] denotes its profit after a reduction in t. FDI is profitable if the following conditions hold:

[[??].sub.1]- [[pi].sub.10] > I. (2)

From Equation 2 it is obvious that the higher the value of t, the greater the incentive for FDI. Higher t reduces [[pi].sub.10] and increases the profitability of FDI vis-a-vis direct exports. This is the standard tariff-jumping argument. Let us now consider t = T, such that [[pi].sub.10](c + [bar.t], c) = 0 (i.e., [bar.t] is the prohibitive tariff). We now make the following assumption:

[[??].sub.1] < I < ([[pi].sub.m] - [[??].sub.1]), (3)

where [[pi].sub.m] is the monopoly profit of firm 1 in the local market and [[??].sub.1] is the duopoly profit of firm 1 when t = 0. Equation 3 indicates that regardless of the initial tariff rate the foreign firm would never go for FDI, since the net duopoly payoff, even with an initial prohibitive tariff, is not good enough to cover the sunk cost of investment. However, the monopoly profit could recover the sunk investment cost and the opportunity cost even with zero tariffs. This assumption basically rules out "go-alone" FDI and makes the standard tariff-jumping argument irrelevant. Even when ([[pi].sub.m] -[[??].sub.1] > I), because the firms face the same costs, [[pi].sub.m] cannot be an equilibrium outcome in the post-FDI game. Also, for all tariff rates, FDI would be impossible, and, hence, tariff-jumping does not give enough incentive for FDI.

Now consider an alternative arrangement in which the foreign firm and the local firm can enter into a joint venture contract only at a certain cost, which in this case, is incurred by the organization. For illustration, after the formation of a joint venture, a manager in the subsidiary may have lower incentives to come up with good ideas to reduce production costs or to raise quality because this investment is expropriated by the parent firm. Second, there may be a loss in information about the performance in the subsidiary and therefore less incentive to make improvements because a joint venture reduces or eliminates the fluidity of the market for the stock of the subsidiary. Third, there may be legal costs associated with the formation of the joint venture. As in Hart et al. (1990), we represent these costs, managerial and/or institutional, by a fixed amount (I).

The nature of the joint venture contract is described by the following clauses: (i) Each firm would bear a portion of I and would derive a share of profits; and (ii) Neither firm would compete with the joint venture firm through independent production. It is obvious that if collusion is enforceable in the court of law, the foreign firm does not have to invest anything. We are assuming that such an arrangement is not enforceable. Governments allow joint ventures provided that they bring in some new investments. Also, we assume that the joint venture contract is enforceable in the court of law such that both of the above clauses are adhered to.

The question, then, is whether such a joint venture contract would be agreed upon. Note that a successful joint venture would bring in foreign investment that was not available through tariff jumping and FDI.

Let [lambda] [member of] (0, 1) be the share of I to be borne by the local firm. For a successful joint venture contract, the following incentive constraints must be satisfied:

[lambda]([[pi].sub.m] - I) [greater than or equal to] [[pi].sub.20](c + t,c) (4)

and

(l - [lambda])([[pi].sub.m] - I) [greater than or equal to] [[pi].sub.10](c + t,c). (5)

By rearranging Equations 4 and 5 we obtain

1 > [lambda] [greater than or equal to] [[[pi].sub.20]/[[pi].sub.m] - I], (6)

0 < [lambda] [less than or equal to] [1 - [[pi].sub.10]/[[pi].sub.m] - I]. (7)

Equations 6 and 7 indicate that the necessary and sufficient condition for such a [lambda] to exist is given by

[1 - [[pi].sub.10]/[[pi].sub.m] - I] > [[[pi].sub.20]/[[pi].sub.m] - I],

which can be written as

[[pi].sub.m] - ([[pi].sub.10] + [[pi].sub.20]) > I. (8)

Let us now define

[[pi].sub.m] - ([[pi].sub.10] +[[pi].sub.20]) [equivalent to] [OMEGA](t), (9)

where [[pi].sub.10] and [[pi].sub.20] both depend on t; whereas, [[pi].sub.m] is independent of the tariff rate. The next question is how [OMEGA](t) behaves with respect to changes in the tariff rate. An increase in t reduces [[pi].sub.10] but increases [[pi].sub.20]. There is no prior presumption as to how the sum of Cournot profits would move along with t. Demand is based on quasi-linear preferences generating a linear demand function,

P = a - ([q.sub.1] + [q.sub.2]), (10)

and the cost function,

c = [cq.sub.i], i = 1, 2, (11)

so that we have

([[pi].sub.10] + [[pi].sub.20]) = [[a- 2(c + t) + c].sup.2]/9 + [[a- 2c + c + t].sup.2]/9 (12)

It is evident from Equation 12 that

[partial derivative]([[pi].sub.10] + [[pi].sub.20]/[partial derivative]t] [greater than or equal to] [less than or equal to] 0 iff t [greater than or equal to] [less than or equal to] [a - c/5]. (13)

Therefore,

[OMEGA]*(t) [greater than or equal to] [less than or equal to] 0 iff t [less than or equal to] [greater than or equal to] [a - c/5] = [??]. (14)

[OMEGA]/(t) has been depicted in Figure 1. Note that for t = 0, [OMEGA](t) represents the difference between the monopoly and symmetric duopoly payoff. At [bar.t], the local firm is a monopolist and the difference vanishes. The intuition behind the non-monotonic response of ([[pi].sub.10] + [[pi].sub.20]) with respect to changes in t is as follows. (6)

Changes in t affect the effective marginal cost of the foreign firm. At t = 0, we have a symmetric duopoly situation. A rise in t increases the cost of one of the duopolists and hurts the total industrial profits because, to start with, [q.sub.1] is quite significant. An increase in [[pi].sub.2], as a result of the strategic effects, fails to compensate for the fall in [[pi].sub.1]. If it is near [bar.t], [[pi].sub.1] falls and, hence, [q.sub.1] is already quite small. Therefore, further rise in t increases [[pi].sub.2] more than a decline in [[pi].sub.1]. The proof with a general demand function is provided in the Appendix.

For the general demand function of the form

P = f([q.sub.1] + [q.sub.2]), (15)

with f* < 0 and f** < 0, we obtain

[OMEGA]*(t) [greater than or equal to] [less than or equal to] 0 iff [f*[q.sub.2] + ([q.sub.2] - [q.sub.1])(f* + f**[q.sub.2])/3f* + f** ([q.sub.1] + [q.sub.2]] [less than or equal to] [greater than or equal to] [q.sub.1]. (16)

[GRAPHIC OMITTED]

It is easy to check that with a linear demand function Equation 16 is reduced to Equation 14. At t = 0, [q.sub.1] = [q.sub.2], and it is obvious that [OMEGA]'(t) > 0. For t [greater than or equal to] [bar.t], [q.sub.i] = 0 and [OMEGA]* (t) < 0. At this stage it can be said that for very low tariff rates [OMEGA](t) rises as t increases, and for very high tariff rates it falls with t. While in the linear case we could guarantee unique inflection of the [OMEGA](t) function, in the general case it may not be true. But as we shall see, our proposed result would follow even if we cannot characterize the entire [OMEGA](t) function. We shall first prove our result with single inflection of [OMEGA](t). (7)

PROPOSITION 1. Suppose ([[pi].sub.m] - I) < ([[pi].sub.10] + [[pi].sub.20]) at t = 0, then either (i)joint ventures would not be feasible under very low or very high tariffs, or (ii) joint ventures would not be feasible under any tariff. (8)

PROOF. We know that [OMEGA](t) reaches a maximum at t = [??] Suppose [OMEGA]([??])[greater than or equal to] I. We also know that [OMEGA](t) < I at t = 0. Hence, [there exists][t.sub.1] such that [OMEGA]([t.sub.1]) = I as [OMEGA](t) is increasing for t [member of] [0, [??]]. Hence, [for all]t < [t.sub.1], in which case joint ventures would not be feasible. Similarly, one can define [t.sub.2], where [t.sub.2] > [??] such that [OMEGA]([t.sub.2]) = I and [for all]t > [t.sub.2], [OMEGA](t) < I. This proves (i). Suppose now that [OMEGA]([??]) < I; then there does not exist any tariff rate for which joint venture is feasible. This proves (ii). QED.

This result is shown in Figure 1. For I = [I.sub.1] the feasibility region is between [t.sub.1] and [t.sub.2], whereas for [I.sub.2] no such region exists. It is straightforward to argue that a region such as [Ot.sub.1] would vanish provided that I is fairly low and less than OA. But a region such as [t.sub.2][bar.t] would always exist, no matter how low I is, provided it is not zero. The condition stipulated in the proposition guarantees that OA < [I.sub.1]. This indicates that for very high tariff rates joint ventures will not be feasible. This qualifies the standard tariff-jumping argument. (9)

In our setup, "go-alone" FDI is not possible, but firms are allowed to form a joint venture. In the case of "go-alone" FDI, a higher tariff provided a greater incentive for FDI. But in a joint venture, the local partner's reservation payoff depends on the initial tariff rate. If t is very high, the local partner is already close to being a monopolist. If one looks at the incentive constraint (ii), it is revealed that for a [[pi].sub.20] close to [[pi].sub.m], a feasible [lambda] would not exist. The local firm would be in a stronger postion to bargain and the joint venture will not mature. In a sense, pre-existing high tariffs imply a strong bargaining position for the local partner. It is quite possible that lower tariff rates would induce investment through the formation of joint ventures. This would be true if [I.sub.1] < OA.

PROPOSITION 2. Even for a [OMEGA](t) function with multiple inflections, very high tariffs will discourage joint ventures.

PROOF. Note that even with multiple inflections, a stretch such as [t.sub.2][bar.t] (in Figure 1) would always exist as [[pi.sub.2](c + [bar.t] c) = [[pi.sub.m](c). Now, for any positive L however small, the following condition must be true:

[OMEGA]([bar.t]) = 0 < I.

Then one can easily choose [t.sub.2] < [bar.t], such that [OMEGA]([t.sub.2]) = I. Hence, so long as I is positive, a prohibitive tariff will always deter joint ventures. Another way of proving it for the general case is to manipulate Equation 6. Note that the minimum [lambda] needed to induce the local firm is given by

[[lambda].sub.min] = [[[pi].sub.20](t)/[[pi].sub.m] - I]. QED. (17)

Let [[pi].sub.20]([??]) = ([[pi]sub.m] - I); this is always possible, as [[pi].sup.'.sub.20]([??]) > 0, [[pi].sub.20]([??] = 0) = [[??].sub.1] < ([[pi].sub.m] - I), and [[pi]sub.20](t = [bar.t]) = [[pi]sub.m] > ([[pi]sub.m] - I)- Hence, [for all]t [member of] ([??], [bar.t]), [[lambda].sub.min] > 1. This makes joint ventures unfeasible.

Our propositions complement (10) the following contributions that have demonstrated a negative relationship between trade cost and FDI. For illustration, Smith (1987) argued that a tariff may play a role in strengthening the position of a local firm facing foreign competition, inducing the local firm to enter the market. Once entry has occurred, the behavior of the multinational may change. In the absence of the tariff, the multinational would have chosen to invest and produce in the country and would have secured a monopoly position. Given the entry of a local rival, the multinational may then prefer to produce elsewhere and to export. Motta (1992) extended this line of argument by showing that the existence of a tariff may cause a shift away from foreign investment in a setup in which potential entry by domestic producers is allowed. Motta and Norman (1996) showed that improved market accessibility leads to export-platform FDI. Neary (2002) highlighted two influences, distinct from the tariff-jumping argument, of internal trade liberalization by a group of countries on the level and pattern of inward FDI. First, the export platform motive favors FDI with only a single union plant relative to exporting and may induce a firm that has never exported to invest. Second, reduced internal tariffs increase competition from domestic firms, which dilutes the other motives and may induce multinationals to leave even though external tariffs are unchanged.

Some natural qualifications apply to the propositions we have derived in this section. For illustration, if a distinction is made between joint ventures and a monopoly, suppose one of the main characteristics of a joint venture is that it has control problems that a monopoly firm does not face. One can formalize this by assuming that joint ventures face a moral hazard problem in the provision of inputs, so that the joint venture payoffs would be lower compared to those under a monopoly that does not face such issues. (11) This implies that the range of parameter values (for which Propositions 1 and 2 hold) shrinks as [OMEGA](t) is negative. Alternatively, suppose that there are two foreign firms, and the domestic government only allows for entry by one such firm. In that case, Equation 2 is no longer independent of t: Instead, the profit from FDI will then be increasing in t, since an increase in t now hurts one of the competitors (the foreign firm excluded from entry). For high t, it is then the case that Equation 4 is likely to be violated so that FDI becomes feasible. Similar issues can be envisioned if more than one domestic firm is allowed for.

3. Asymmetric Costs

In this section we introduce asymmetric costs between firms 1 and 2. Since our goal is to develop an analytical structure to highlight situations in which the local market is protected, it will not be unreasonable to assume that [c.sub.2] > c = [c.sub.1]. There is now an interesting caveat to the existing problem. One has to know how [[[pi]sub.10]([c.sub.2] + t, c) + [[pi]sub.20] ([c.sub.2] + t, c)] behaves for a given t as one changes [c.sub.1]. This would determine the position of the new [OMEGA](t) function relative to the old one.

Following the example developed in the earlier section one can write

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]. (18)

This leads us to the following proposition.

PROPOSITION 3. Starting from the case of symmetric costs, the range of tariffs that implements joint ventures under asymmetric costs expands if

[c.sub.2] < [a + 4(c + t)/5]

and contracts if

[c.sub.2] > [a + 4(c + t)/5].

PROOF. Following Figure 2, the new [OMEGA](t) function lies above the old one; whereas, [??] is raised, and T also has to be increased now, as [c.sub.2] > c. This increases the feasible range of tariffs. QED.

The fact that [c.sub.2] needs to be small or reasonably close to c for [[pi]sub.10]([c.sub.2] + t, c) + [[pi].sub.20]([c.sub.2] + t, c)] to be a decreasing function of [c.sub.2] can be proven under general demand conditions. The proof and the intuition are very similar to the tariff case. If [c.sub.2] is pretty close to c, a rise in [c.sub.2] hurts [[pi].sub.20] more than it helps [[pi]sub.10]. Hence, total profit goes down and [c.sub.2] < [(a + 4c)/5], implying that the new [OMEGA](t) function would be uniformly higher. But still, very high tariffs would discourage joint ventures. If

[c.sub.2][member of](a + 4c/5, a + 4(c + [bar.t])/5).

then there will be a stretch of the new [OMEGA](t) function, which may lie below the old one for some tariff levels. Note that (c + [bar.t]) = ((a + [c.sub.2])/2), and given this, one can show that [c.sub.2] cannot exceed

(a + 4(c + [bar.t]/5 = (3a + [2c.sub.2]/5),

and

([c.sub.2] - 3a + [2c.sub.2]/5) = (3([c.sub.2] - a)/5) < 0. (19)

Therefore, there will always be a stretch of the new [OMEGA](t) function that lies above the original [OMEGA](t) function with [c.sub.2] = c. If [c.sub.2] is very high, the exporting firm already enjoys a good share of profits and therefore may not be induced to write a joint venture contract. This is likely to happen when tariffs are low, thus shrinking the feasibility region by [t.sub.1][t.sup.".sub.1] in Figure 2. (12)

[GRAPHIC OMITTED]

Investment costs in a joint venture have been assumed to be the same as in FDI. However, it may seem appropriate to change investment costs, since the joint venture can use local production facilities of the local firm. But there are reasons why a change in investment costs may not be possible. The foreign technology, even if yielding the same c, may entail automated plants, and then one needs new investments. There may be sunk organizational costs of joint ventures. If c and [c.sub.2] are different, the adoption cost due to technological difference may be more compelling.

We have thus shown how the technological difference between the local firm and the foreign firm enhances or curtails the scope for joint ventures. It may be noted that our result--that relatively high tariff rates tend to jeopardize the possibility of a joint venture--continues to hold. If firms are different but close enough, the prospect of a joint venture increases. If the local firm is significantly backward, it does not work, and the feasibility is somewhat recovered through higher tariffs. Again, for very high tariffs such a possibility disappears.

4. Concluding Remarks

In this paper, we propose a possible relationship between high tariff rates and joint ventures when "go-alone" FDI is impossible because of the existence of a strong local firm. High tariffs imply a large reservation payoff to the local firms, which therefore claim a significant share in a joint venture deal. We have shown that very high tariffs will fail to generate a joint venture contract, thus reducing the possibility of FDI. There do exist certain tariff rates for which FDI may not be forthcoming; although, joint ventures would take place. It is our hope that the model presented in this paper will encourage researchers to look at the empirical relationship between tariffs and foreign equity participation, particularly for countries with significant experience with protected industrialization, such as India. As the governments launch liberalization policies and tariffs start coming down, contrary to the tariff-jumping argument, this may help foreign investment through the formation of joint ventures. On the other hand, China, one of the world's largest destinations of FDI, maintains an average (trade-weighted) tariff rate of more than 26%. Shortly after liberalizing its foreign investment regime in 1995, allowing 100% foreign ownership, Nigeria began to modify its tariff lines (most recently in 2001), resulting in an upward revision by 25% on as many as 70 tariff lines. For these reasons, countries such as India, China, and Nigeria offer natural experiments for the testable hypothesis that emerges from our model: Are joint ventures more dominant in less protected industries after controlling for foreign capital inflows and firm heterogeneity?

Our analysis can be extended to an oligopolistic setting with many foreign firms when a few of them are allowed to come in and form joint ventures. Another interesting extension would be to consider the differentiated product case when domestic and foreign brands differ in terms of their quality. In this paper, we have not discussed the optimal tariff policy of the government to induce foreign investment. However, note that once the foreign firm jumps the tariff, further lowering of the tariff does not alter the consumer surplus; however, the government can increase the bargaining position of the local firm by keeping a high initial tariff. If one follows Figure 1, it is obvious that [t.sub.2] is the optimal tariff rate, because for t > [t.sub.2] investment does not take place, and for t < [t.sub.2] consumer surplus does not change, but by increasing the tariff up to [t.sub.2] the government can maximize the home firm's profit in the joint venture.

Appendix

Characterization of the [OMEGA](t) function with symmetric costs is given by

[OMEGA](t) = [[pi].sub.m] - |[[pi].sub.1](t) + [[pi].sub.2](t)],

where

[[pi].sub.1] = [[q.sub.1]f([q.sub.1] + [q.sub.2]) - (c + t)[q.sub.1]], (1A)

and

[[pi].sub.2] = [[q.sub.2]f([q.sub.1] + [q.sub.2]) - [cq.sub.2]]. (2A)

From the usual first-order conditions we get

[[q.sub.1]f' + f - (c + t)] = 0 (3A)

and

[[q.sub.2]f' + f - c] = 0 (4A)

Therefore,

[(2f' + [q.sub.1]f"([dq.sub.1]/dt) + (f' + [q.sub.1]f")([dq.sub.2]/dt)] = 1

and

[(f' + [q.sub.2]f"([dq.sub.1]/dt) + (2f' + [q.sub.2]f")([dq.sub.2]/dt)] = 0.

Hence,

([dq.sub.1]/dt) = (2f' + [q.sub.2]f")/[DELTA], ([dq.sub.2]/dt) = -(f' + [q.sub.1]f"/[DELTA],

where

[DELTA] = [(2f" + [q.sub.2]f")(2f' + [q.sub.1]f") - (f' + [q.sub.1]f")(f' + [q.sub.2]f")] > 0,

because f' < 0 and f" < 0.

Now we have

[OMEGA]'(t) = -([partial derivative][[pi].sub.1]/[partial derivative]t + [partial derivative][[pi].sub.2]/[partial derivative]t)

= -[[q.sub.1]f'([partial derivative][q.sub.2]/[partial derivative]t) + [q.sub.2]f'([partial derivative][q.sub.1]/[partial derivative]t) - [q.sub.1]],

so that

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII],

or

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII].

Substituting for A and simplifying, we obtain

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]. (5A)

References

Al-Saadon, Yousef, and Satya P. Das. 1996. Host country policy, transfer pricing and ownership distribution m international joint-ventures: A theoretical analysis. International Journal of Industrial Organization 14:345-64.

Beladi, Harold, and Avik Chakrabarti. 2008. Foreign equity participation under incomplete information. Journal of Economic Behaviour and Organization 67:279-95.

Bertrand, Olivier, and Habib Zitouna. 2006. Trade liberalization and industrial restructuring: The role of cross-border mergers and acquisitions. Journal of Economics & Management Strategy 15:479-515.

Chakrabarti, Avik. 2003. A theory of the spatial distribution of foreign direct investment. International Review of Economics and Finance 12:149-69.

Chao, Chi-Chur, and Eden S. H. Yu. 1996. Are wholly foreign-owned enterprises better than joint ventures? Journal of International Economics 401:225-37.

Chao, Chi-Chur, and Eden S. H. Yu. 2000. Domestic equity controls of multinational enterprises. Manchester School 68:321-30.

Chowdhury, Indrani R., and Prabal R. Chowdhury. 2001. A theory of joint venture life-cycles. International Journal of Industrial Organization 19:319-43.

Damania, Richard. 2003. Protectionist lobbying and strategic investment. Economic Record 79:57-69.

Das, Satya P., and Seiichi Katayama. 2003. International joint venture and host-country policies. Japanese Economic Review 54:381-94.

Dinopoulos, Elias, and Kar-yiu Wong, 1991. Quid pro quo foreign investment and policy intervention. In International trade and global development: Essays in honor of Jagdish Bhagwati, edited by K. A. Keokkoek and C. B. M. Mennes. London: Routledge, pp. 162-90.

Eicher, Theo, and Jong W. Kang. 2005. Trade, foreign direct investment or acquisition: Optimal entry modes for multinationals. Journal of Development Economics 77:207-28.

Falvey, Rod. 1998. Mergers in open economies. The World Economy 21:1061-76.

Gang, I., and Shubhashis Gangopadhyay, 1989. Foreign direct vs. foreign indirect investment: The issue of control and optimal government policy. Rutgers University Bureau of Economic Research Working Paper No. 21.

Gang, I., and Shubhashis Gangopadhyay. 1994. Foreign investment and control of transnational enterprises. Journal of International Trade and Economic Development 3:213-28.

Grossman, Gene M., and Elhanan Helpman. 1994. Foreign investment with endogenous protection. NBER Working Paper No. 4876.

Hagedoorn, John, Danielle Cloodt, and Hans V. Kranenburg. 2005. Intellectual property rights and the governance of international R&D partnerships. Journal of International Business Studies 36:175-86.

Harris, Richard G., and Nicolas Schmitt. 2001. Strategic export policy with foreign direct investment and import substitution. Journal of Development Economics 64:293-312.

Hart, Oliver, Jean Tirole, Dennis W. Carlton, and Oliver E. Williamson. 1990. Vertical integration and market foreclosure. Brookings Papers on Economic Activity Special Issue:205-82, Microeconomics.

Herander, Mark G., and Brad Kamp. 2003. Tariff policy and the foreign entry decision in the presence of a cost-based informational asymmetry. Review of International Economies 11:845-59.

Horstmann, Ignatius J., and James R. Markusen. 1996. Exploring new markets: Direct investment, contractual relations and the multinational enterprise. International Economic Review 37:1-19.

Jones, Ronald W. 1999. Protectionism, export requirements, and joint ventures. Review of Development Economics 3:231-41.

Lin, Ping, and Kamal Saggi. 2004. Ownership structure and technological upgrading in international joint ventures. Review of Development Economics 8:279-94.

Liu, Hui-Ling, and Jiunn-Rong Chiou. 2003. International competition and tariff-jumping FDI with uncertainty. Academia Economic Papers 4:605-36.

Long, Ngo V., and Neil Vousden. 1995. The effects of trade liberalization on cost reducing horizontal mergers. Review of International Economics 2:141-55.

Luo, Yadong. 2005. Transactional characteristics, institutional environment and joint venture contracts. Journal of International Business Studies 36:209-30.

Luo, Yadong, and Seung H. Park. 2004. Multiparty cooperation and performance in international equity joint ventures. Journal of International Business Studies 35:142-60.

Marjit, Sugata. 1990. Rationalizing public-private joint ventures in an open economy: A strategic approach. Journal of Development Economics 33:377-83.

Marjit, Sugata, Hamid Beladi, and Tarun Kabiraj. 2007. Brand name collaboration and optimal tariff. Economic Modelling 24:636-47.

Marjit, Sugata, Udo Broll, and Indrajit Mallick. 1995. A theory of overseas joint ventures. Economics Letters 47:367-70.

Marjit, Sugata, and Prabal R. Chowdhury. 2004. Asymmetric capacity costs and joint venture buy-outs. Journal of Economic Behavior and Organization 54:425-38.

Motta, Massimo. 1992. Multinational firms and tariff jumping argument: A game theoretic analysis with some unconventional conclusions. European Economic Review 36:1557-71.

Motta, Massimo, and George Norman. 1996. Does economic integration cause foreign direct investment? International Economic Review 37:757-83.

Mukherjee, Arijit, and Sarbajit Sengupta. 2001. Joint ventures versus fully owned subsidiaries: Multinational strategies in liberalizing economies. Review of International Economics 9:163-80.

Mundell, Robert A. 1957. International trade and factor mobility. American Economic Review 47:1-24.

Neary, J. Peter. 2002. Foreign direct investment and the single market. Manchester School 70:291-314.

Neary, J. Peter. 2003. Globalization and market structure. Journal of the European Economic Association 1:245-71.

Neary, J. Peter. 2007. Cross-border mergers as instruments of comparative advantage. Review of Economic Studies 4:182-91.

Pontes, Jose P. 2007. A non-monotonic relationship between FDI and trade. Economics Letters 95:69-73.

Purkayastha, Dipankar. 1993. Firm-specific advantages, multinational joint ventures and host country tariff policy. Southern Economic Journal 60:89-95.

Saggi, Kamal, and Howard Pack. 2006. The case for industrial policy: A critical survey. World Bank Policy Research Working Paper No. 3839.

Smith, Alasdair. 1987. Strategic investment, multinational corporations and trade policy. European Economic Review 31:89-96.

Svejnar, Jan, and Stephen C. Smith. 1984. The economics of joint-ventures in less developed countries. Quarterly Journal of Economics 119:149-67.

(1) More recent works along this line include those of Dinopoulos and Wong (1991), Grossman and Helpman (1994), Horstmann and Markusen (1996), Harris and Schmitt (2001), Neary (2002), Chakrabarti (2003), Damania (2003), Herander and Kamp (2003), Liu and Chiou (2003), and Eicher and Kang (2005).

(2) While an extensive review of the literature on joint ventures is beyond the scope of this paper, some notable contributions include the work of Svejnar and Smith (1984), Gang and Gangopadhyay (1989, 1994), Marjit (1990), Purkayastha (1993), Marjit, Broil, and Mallick (1995), Al-Saadon and Das (1996), Chao and Yu (1996), Jones (1999), Chao and Yu (2000), Mukherjee and Sengupta (2001), Das and Katayama (2003), Lin and Saggi (2004), Luo and Park (2004), Marjit and Chowdhury (2004), Marjit et al. (2007), Hagedoorn, Cloodt, and Kranenburg (2005), and Luo (2005).

(3) See Marjit, Beladi, and Kabiraj (2007) and Das and Katayama (2003) for an assessment of the state of the literature on the role of host-country policies in international joint ventures.

(4) If one looks at the Indian joint ventures, they always indicate tie-ups of a multinational with a local partner who already enjoys a significant position in the market. The Indian big names, such as the Tatas and Mahindras, are examples of this type of venture.

(5) For the general demand function P = f([q.sub.1] + [q.sub.2]), f* < 0, and f* + [q.sub.1] f** < 0 ([for all]i = 1, 2) ensures the existence and uniqueness of a pure strategy Cournot equilibrium.

(6) The joint Cournot-Nash industry profits usually move non-monotonically with the cost differentials of the participating firms. This has been used in the literature on international technology transfer and horizontal mergers by Long and Vousden (1995).

(7) Proposition 1 holds for functions [OMEGA](t) that have a unique maxima. The properties of linearity of demand and cost functions are sufficient, though not necessary, for the proof of this proposition because the result follows even when it is not possible to characterize the entire [OMEGA]t).

(8) Proposition 1 indicates that there may be a non-monotonic relationship between t and the formation of joint ventures. In a recent paper, Pontes (2007) demonstrates a non-monotonic relationship between green-field FDI and trade.

(9) A joint venture often brings synergic benefits that can be captured by considering the cost of production of the joint venture firm below c, in which case [OMEGA] > 0 at [bar.t].

(10) A comparison of our propositions may also be drawn with those in the literature on cross-border mergers. For illustration, a model (along the lines of the small but rapidly growing literature on cross-border mergers in oligopolistic markets, exemplified by Long and Vousden [1995], Falvey [1998], and Bertrand and Zitouna [2006]) of mergers in oligopolistic markets can explain how trade liberalization encourages rather than discourages FDI. See Neary (2003, 2007) for a thorough analytical review of this strand of literature.

(11) See Chowdhury and Chowdhury (2001), who construct a dynamic theory of a joint venture life cycle that relies on synergy, organizational learning, and moral hazard.

(12) One can compare [t.sub.1], [t.sub.2] and [[t.sup.".sub.1], [t.sup.".sub.2]] to see that the feasibility region expands by [t.sub.2][t.sup.".sub.2] and shrinks by [t.sub.1][t.sup.".sub.1].

Hamid Beladi, Department of Economics, College of Business, University of Texas at San Antonio, One UTSA Circle, San Antonio, TX 78249-0633, USA; corresponding author.

Sugata Marjit, Centre for Studies in Social Sciences, Calcutta 700-094, West Bengal, India: E-mail smarjit@hotmail.com.

Avik Chakrabarti, Department of Economics, 816 Bolton Hall, College of Leners and Science, University of Wisconsin Milwaukee, P.O. Box 413, Milwaukee, W1 53201, USA; E-mail chakra@uwm.edu.

Received February 2007; accepted February 2008.

We wish to thank anonymous referee(s) for insightful comments and suggestions on an earlier version of this paper. The usual disclaimer applies.