Privatization, strategic foreign direct investment and host-country welfare

ARTICLE IN PRESS European Economic Review 53 (2009) 775–785

Contents lists available at ScienceDirect

European Economic Review journal homepage: www.elsevier.com/locate/eer

Privatization, strategic foreign direct investment and host-country welfare Arijit Mukherjee a,, Kullapat Suetrong b a

University of Nottingham, and The Leverhulme Centre for Research in Globalisation and Economic Policy, School of Economics, University Park, University of Nottingham, Nottingham NG7 2RD, UK Bureau of Policy and Planning, Department of Business Development, Ministry of Commerce, Bangkok, Thailand

b

a r t i c l e in fo

abstract

Article history: Received 29 November 2007 Accepted 20 February 2009 Available online 19 March 2009

Recent evidence shows that developing countries and transition economies are increasingly privatizing their public firms and at the same time experiencing rapid growth of inward foreign direct investment (FDI). We show that there is a two-way causality between privatization and greenfield FDI. Privatization increases the incentive for FDI, which, in turn, increases the incentive for privatization compared to the situation of no FDI. The optimal degree of privatization depends on the cost difference of the firms, and on the foreign firm’s mode of entry. & 2009 Elsevier B.V. All rights reserved.

JEL classification: D43 F12 F13 F23 L13 L33 O25 Keywords: Privatization Greenfield FDI Welfare

1. Introduction Empirical evidence shows that many developing and transition economies are privatizing state-owned enterprises across several sectors and also experiencing significant inflow of foreign direct investment (FDI). In an earlier study on Latin America, Baer (1994) notes that the presence of foreign capital has increased as the presence of state has declined. It is documented in UNCTAD (2002) that along with a combination of several reform measures such as improved investment climate, openness to trade and FDI, macroeconomic stability, etc., privatization has increased FDI inflow over the 1990s. Using annual data of eight Asian and nine Latin American and Caribbean countries for 1990–1999, Gani (2005) shows that privatization is positively correlated to FDI. Focusing on the Central and Eastern European countries (CEECs), Merlevede and Schoors (2005) show that privatization history positively affects FDI. It has been found that, during 2000–2003, China accounted for almost 90% of the privatization proceeds1 in East Asia and the Pacific and it is, at the same time, the biggest

 Corresponding author. Fax: +44 115 951 4159.

E-mail address: [email protected] (A. Mukherjee). Privatization proceeds are defined to include all monetary receipts to the government resulting from partial and complete divestitures (via asset sales or sale of shares), concessions, leases, and other arrangements. The data do not cover management contracts, new greenfield investments, investments committed by new private operators as part of concession agreements, and ‘‘voucher’’ privatizations (Kikeri and Kolo, 2005). 1

0014-2921/$ - see front matter & 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.euroecorev.2009.02.004

ARTICLE IN PRESS 776

A. Mukherjee, K. Suetrong / European Economic Review 53 (2009) 775–785

FDI recipient in the region. India also shares a similar story on FDI and privatization proceeds. Other regions, such as Latin America, Europe and Central Asia, also recorded the same trend of FDI and privatization proceeds.2 While the empirical evidence suggests a positive correlation between privatization and FDI, the causality between these two is not immediate. To the best of our knowledge, there is no theory which helps us to understand whether privatization is the cause or consequence of FDI. We develop a simple open-economy mixed oligopoly model to understand the causality between privatization and greenfield FDI. We find complementarity between these two. We show that privatization increases the incentive for greenfield FDI, and the possibility of greenfield FDI increases the optimal degree of privatization if the degree of privatization that is required to attract FDI is not very high. If the degree of privatization that is required to attract FDI is very high, the host-country may not prefer to induce FDI through privatization. In this situation, the optimal degree of privatization is the same with and without FDI. Thus, our analysis suggests that reforms allowing FDI and reducing state ownership can help to increase welfare by complementing each other. The cost difference between the firms, the fixed cost of undertaking FDI and the demand parameter play important roles in determining the optimal degree of privatization. We also show that partial privatization is the optimal strategy of the host-country. In other words, neither complete privatization nor complete nationalization maximizes the host-country welfare in the presence of foreign competition. This result is in line with Maw (2002), which shows that partial privatization of the public firms are mostly observed in transition countries while their economies are increasingly open to foreign competitions.3 The reasons for our results are as follows. Nationalization of the domestic firm serves as a credible commitment to increase output beyond the profit-maximizing level. Thus, in an imperfectly competitive market, nationalization acts as a disciplining device by increasing the industry output. However, if the foreign firm is more cost efficient than the domestic firm, nationalization creates production inefficiency by reducing the output of the foreign firm. If the degree of nationalization reduces, it reduces the domestic firm’s weight on social welfare, and induces the foreign firm to increase its output by creating a more level playing field of competition. However, since the foreign firm’s cost is lower under FDI compared to exporting, the foreign firm’s gain from privatization is higher under the former, thus increasing the incentive for FDI. While privatization reduces the output of the domestic firm, it increases the output of the foreign firm, which is more cost efficient than the domestic firm. The optimal degree of privatization balances these effects, and makes partial privatization as the optimal strategy of the domestic country. If privatization induces FDI, it increases the benefit of privatization by inducing the foreign firm to choose a more cost-efficient production strategy. Hence, unless the degree of privatization which is required to attract FDI is not very high, the FDI attracting role of privatization increases the incentive for privatization compared to the situation with no FDI. However, if the degree of privatization which is required to attract FDI is very high, the negative effect of privatization due to the lower output of the domestic firm becomes the important factor, which eliminates the incentive for attracting FDI through privatization. In this situation, the degree of privation remains the same with and without the possibility of FDI. Therefore, while privatizing in the presence of FDI, a government needs to balance the loss of the disciplining effect of nationalization and the gain in cost efficiency due to FDI. The remainder of the paper is structured as follows. The next section reviews the related literature. Section 3 describes the basic model. The effects of privatization on the incentives for FDI and welfare are shown in Section 4. Section 5 concludes. 2. Review of the related literature A tradition of the theoretical literature on privatization is to consider mixed oligopoly models where the markets are characterized by a small number of firms and the objective function of at least one firm is not to maximize profit but to maximize welfare, in the case of a completely nationalized firm, or a combination of profit and welfare, in the case of a partially privatized firm.4 The works on privatization gets momentum with De Fraja and Delbono (1989), yet the earlier literature mainly focuses on closed economies to explain the relationship between privatization and several other important aspects, such as incentive delegation (Barros, 1995), endogenous market structure (Anderson et al., 1997), entry deterrence (Fershtman, 1990), cost asymmetry (Matsumura, 1998) and innovation (Ishibashi and Matsumura, 2006).5 The consideration of privatization in an open economy has started to attract attention only in recent years. Pal and White (1998) show the effects of privatization under strategic trade policies such as production subsidies and import tariffs. Ohori (2004) considers the effects of privatization on tariffs and environmental taxes. Ohori (2006) considers export competition between the public firms in the presence of strategic government policies. Fjell and Heywood (2002) consider privatization in a Stackelberg leader-follower structure, and show that the effects on firms’ outputs, profits and welfare depend on the relative number of domestic and foreign firms. However, a common feature of these works is to ignore FDI 2

Kikeri and Kolo (2005) provide details on privatization in developing countries. Besides the decision on complete or partial privatization, often the governments need to decide whether to privatize all the state-owned enterprises simultaneously or sequentially. See Gupta et al. (2008) for a recent discussion on this latter issue. 4 De Fraja and Delbono (1989) provide definition of mixed oligopoly in more details. 5 See Harris and Wiens (1980), Beato and Mas-Colell (1984) and Cremer et al. (1989) for other works on privatization. 3

ARTICLE IN PRESS A. Mukherjee, K. Suetrong / European Economic Review 53 (2009) 775–785

777

by the foreign firms, while, in this era of globalization and the tremendous growth of FDI, it is certainly important to see how privatization affects and is affected by inward FDI. We take up this issue in this paper. It is worth mentioning that our paper is also related to the vast literature on FDI explaining the rationale for preferring FDI over the other modes of foreign-market entry such as exporting and technology licensing. Instead of reviewing the vast literature on FDI, we refer to Pack and Saggi (1997) and Saggi (2002) for recent surveys on FDI. However, a common feature of the existing FDI literature is to focus on the profit-maximizing private firms, thus ignoring the issue of privatization, which is a main ingredient of this paper. 3. The model We consider a two-country model, which consists of a home country and a host-country, and assume that there is a firm in each country. The firms produce a homogenous product. The firm in the home country is called firm M, who wants to serve the host-country market either through FDI or through export. The firm in the host-country is a public (or stateowned) firm, which is called firm P. An important difference between firms M and P is about their objective functions. While the former firm maximizes profit, the latter firm maximizes a convex combination of profit and the host-country welfare depending on the share distribution between the government and the private owners of the host-country. Following the literature on privatization6 and the practice of many countries, we consider privatization as the process of a change in the structure of the public firm. Privatization reflects the transfer of the public firm’s ownership from the government to domestic investors of the hostcountry.7 In the following analysis, we will assume that, at the beginning, firm P is completely nationalized, which means that, to start with, the objective function of firm P is to maximize welfare of the host-country. We consider the following cost structures of the firms. We assume that the constant marginal cost of firm P is cp. The constant marginal costs of firm M under export and FDI are, respectively, cx and cf, where cfocx. For simplicity, we normalize cf to 0. We assume that firm M is more cost efficient than firm P with the following relationship: 0 ¼ cfocxpcp. We assume that, under FDI, firm M needs to incur a fixed cost f. Assume that the inverse market demand function in the host-country is p ¼ aQ, where P is the price and Q the total output and a4ci where i ¼ x,f,p. We will consider the following game in the next section. At stage 1, the host-country government decides on the level of privatization. At stage 2, firm M decides whether to export or to undertake FDI. At stage 3, the firms compete in the product market like Cournot duopolists. We solve the game through backward induction. 4. The effects of privatization on FDI and welfare 4.1. Privatization and the incentive for FDI The profit-maximizing firm M maximizes the following expression to determine its output:

pm ¼ ða  qm  qp  cm Þqm  K

(1)

where qm and qp denote the outputs of firms M and P, respectively. We have cm and K equal to cx and 0, respectively, under export, and equal to 0 and f, respectively, under FDI. The objective function of firm P depends on the share distribution between the government and the private owners of the host-country. Following the existing literature (e.g., Fershtman, 1990), we assume that firm P maximizes a convex combination of profit and social welfare, where the weights on profits and social welfare are given by the fractions of shareholdings by the host-country investors and the host-country government. The justification for this objective function of the public firm follows Fershtman (1990), which argues that the behavior of a partly nationalized firm results from a conflict of interests between the directors representing the private investors’ interests and the government’s interest. This conflict of interests is assumed to be resolved through a compromise. Consequently, the firm’s output choice is a compromise between the output that maximizes profit and the output that maximizes welfare. If a indicates the level of privatization, i.e., the fraction of shareholding by the host-country investors, the objective function of firm P is Objp ¼ app þ ð1  aÞw

(2)

where aA[0,1]. Note that complete nationalization and complete privatization are the special cases of (2). If firm P is completely privatized, a becomes one, and if firm P is completely nationalized, a becomes zero. As a increases, it reduces the fraction of shareholding by the government and firm P moves more towards profit maximization. Eq. (2) can be 6

We refer to Vickers and Yarrow (1991), Schmidt and Schnitzer (1997) and Pal and White (1998) for the overviews of the privatization literature. For example, before 1996, the auction of shares of the Indian public sector enterprises was restricted to dispersed domestic investors (Kapur and Ramamurti, forthcoming). In many situations, countries restrict foreign individuals and firms to acquire domestic firms, or apply special restrictions to foreign firms in certain industries, as is the case in Malaysia and the Republic of Korea, for example. Though the practice of the countries in this respect changes over time, the government policies still favor greenfield investment (UNCTAD, 2000). 7

ARTICLE IN PRESS 778

A. Mukherjee, K. Suetrong / European Economic Review 53 (2009) 775–785

expanded to "

ðqm þ qp Þ2 Obj ¼ aða  qm  qp  cp Þqp þ ð1  aÞ ða  qm  qp  cp Þqp þ 2 " # 2 ðqm þ qp Þ ¼ ða  qm  qp  cp Þqp þ ð1  aÞ 2

#

p

(3)

We find that the equilibrium outputs of firms P and M are, respectively, ðað2  aÞ þ acm  2cp Þ=ð2 þ aÞ and ðaa  ð1 þ aÞcm þ cp Þ=ð2 þ aÞ. The total output is ð2a  cm  cp Þ=ð2 þ aÞ. Substituting the equilibrium outputs into both firms’ objective functions yield the following: Objp ¼

pm ¼

2ðaa  cp ð1 þ aÞ þ cm Þðað2  aÞ  2cp þ acm Þ þ ð1  aÞð2a  cm  cp Þ2 2ð2 þ aÞ2

  aa  ð1 þ aÞcm þ cp 2 K 2þa

(4)

(5)

Note that under complete privatization, cp must be less than a/2 to ensure a duopoly market structure. We assume that this condition holds throughout our analysis. It is worth pointing out that, if the degree of privatization is not very high, the profit generated in the (partially) privatized firm is negative. Hence, it is important to discuss why the private sector is interested to buy the shares of the public firm. There are at least two ways to induce shareholding in the public firm by the private sector. First, the government can induce the private investors to acquire shares of the public firm by offering them a lump-sum payment, which can be generated by imposing lump-sum tax on the consumers. Since this lump-sum payment simply represents a redistribution of surplus between consumers and private producers, the equilibrium outputs and the degree of privatization are not affected. In other words, if a properly chosen degree of privatization maximizes social welfare, while creating negative profit in the privatized public firm, there is always the case for compensating the private investors through a non-distortionary subsidy. Another way to induce the private sector investors in acquiring shares of the public firm is to impose a minimum profit requirement for the public firm. However, as evident from Saha and Sensarma (2004), this constraint on the profit of the public firm induces the government to privatize in a way that generates lower welfare compared to situation with no such constraint. Hence, the first procedure dominates the second one, and we assume that there exists a tax-subsidy mechanism that induces the private investors to buy shares of the public firm. Let us now consider the optimal production strategy of firm M. Given the level of privatization, firm M prefers to undertake FDI over export provided

pm;p 4pm;p x f 

aa þ cp 2þa

2

  aa  cx ð1 þ aÞ þ cp 2  f4 2þa

cx ð1 þ aÞð2aa þ 2cp  ð1 þ aÞcx Þ ð2 þ aÞ2

 f p ðaÞ4f

(6)

(7)

where pfm,p and pxm,p represent the profits of firm M under FDI and under export, respectively. Condition (7) suggests that firm M undertakes FDI if and only if the fixed cost of FDI (i.e., f) is lower than the critical value fp(a), where fp(a) represents the difference between the gross profits of firm M under FDI and under export. A higher fp(a) implies a higher difference in the gross profit between FDI and exporting and, therefore, a higher incentive for FDI. Let us now see how the incentive for FDI changes with respect to the degree of privatization, i.e., how fp(a) changes with respect to a. We find that X

@f p ðaÞ 2cx ½ð2aa þ 2cp  ð1 þ aÞcx Þ þ ða  cp Þð2 þ aÞ ¼ 40 @a ð2 þ aÞ3

(8)

The following proposition follows from the above discussion. Proposition 1. As the degree of privatization increases, it increases the incentive for FDI. If the host-country government increases the degree of privatization, the objective of firm P moves more from welfare maximization to profit maximization. As a result, given the output of firm M, a higher degree of privatization reduces the optimal output of firm P. Hence, a higher degree of privatization increases the equilibrium output of firm M but reduces the equilibrium output of firm P. However, the gain in market share by firm M due to a higher degree of privatization increases

ARTICLE IN PRESS A. Mukherjee, K. Suetrong / European Economic Review 53 (2009) 775–785

779

as its marginal cost falls,8 which, in turn, implies that a higher degree of privatization increases the outputs and profits of firm M more under FDI than under export. Hence, a higher degree of privatization increases firm M’s incentive for FDI. It is worth noting that the derivative of fp(a) with respect to cx is positive, suggesting that a fall in cx reduces fp(a). If cx involves trade costs, the relationship between cx and fp(a) implies that ceteris paribus, trade liberalization, which helps to reduce the trade cost, reduces the incentive for FDI. This is consistent with the well-known ‘‘tariff jumping’’ argument, which states that a lower trade cost reduces the incentive for FDI. 4.2. Privatization and the host-country welfare Let us now determine the optimal degree of privatization for the host-country. Following Proposition 1, which suggests that a higher degree of privatization increases the incentive for FDI, it is immediate that FDI occurs irrespective of the degree of privatization if fofp(0) and FDI never occurs irrespective of the degree of privatization if f4fp(1). Therefore, depending on the fixed cost of FDI, we have the following three possibilities: (i) (F,F),9 i.e., firm M undertakes FDI irrespective of the degree of privatization, and it occurs for fofp(0), (ii) (X,X), i.e., firm M exports irrespective of the degree of privatization, and it occurs for f4fp(1) and (iii) (X,F), i.e., privatization induces firm M to undertake FDI, and it occurs for fp(0)ofofp(1). Given the degree of privatization, the host-country welfare is wðaÞ ¼

2ðaa  cp ð1 þ aÞ þ cm Þðað2  aÞ  2cp þ acm Þ þ ð2a  cp  cm Þ2 2ð2 þ aÞ2

(9)

where cm ¼ cx if firm M exports and cm ¼ 0 if firm M undertakes FDI. 4.2.1. When the foreign firm always undertakes FDI Let us first consider the situation where the fixed cost of FDI is sufficiently small so that fofp(0), and we have (F,F). In this situation, given the degree of privatization, the host-country welfare is wfp ðaÞ ¼

2ðaa  cp ð1 þ aÞÞðað2  aÞ  2cp Þ þ ð2a  cp Þ2 2ð2 þ aÞ2

(10)

Differentiating wpf with respect to a, we obtain A

@wfp ð2a  cp Þðcp þ 2acp  3aaÞ ¼ @a ð2 þ aÞ3

(11)

Differentiating A with respect to cp yields @A 3aa þ 2ð1 þ 2aÞða  cp Þ ¼ 40 @cp ð2 þ aÞ3

(12)

which suggests that A is positively related to cp. We also find that (i) if cp ¼ 0 and a ¼ 0, A ¼ 0, (ii) if cp ¼ 0 and a40, Ao0, (iii) if cp ¼ (a/2) and a ¼ 0, A40, (iv) if cp ¼ (a/2) and a ¼ 1, Ao0, (v) at any given cp, A reduces with higher a and (vi) A is concave with respect to cp,10 and the value of cp that maximizes A is greater than a/2. Hence, A shows a positive slope for cpA(0,a/2). The information obtained above enables us to construct Fig. 1, which illustrates the relationship between A, cp and a. It is clear from Fig. 1 that if cp is equal to 0, any degree of privatization reduces the host-country welfare if firm M always undertakes FDI. However, for any cp such that 0ocpoa/2, there always exists a value of aA(0,1) such that A ¼ 0, which implies that partial privatization is the optimal strategy of the host-country government. For example, if cp ¼ cp1, the optimal degree of privatization is a ¼ a1*. Further, a*, which indicates the degree of privatization that maximizes the hostcountry welfare for a given cp, increases with cp(a2*4a1* as cp24cp1). It is also clear from Fig. 1 that complete privatization is never optimal if firm M always undertakes FDI. The above discussion is summarized in the following proposition. Proposition 2. Given that 0 ¼ cfocxpcp, if the fixed cost of FDI is very small so that the foreign firm undertakes FDI irrespective of the degree of privatization, partial privatization is the optimal strategy of the host-country government for any cost disadvantage of the public firm, i.e., for cpA(0,a/2). The reason for the above finding is as follows. Since a higher degree of privatization reduces the public firm’s weight on welfare maximization, it tends to lower consumer surplus by reducing the output of the public firm, thus creating a negative impact on the host-country welfare. However, the output reduction by the public firm helps to reduce the cost of 8 The equilibrium outputs of firm M under export and under FDI are, respectively, ðaa þ cp Þ=ð2 þ aÞ  ð1 þ aÞcx =ð2 þ aÞ and ðaa þ cp Þ=ð2 þ aÞ. Since ð1 þ aÞ=ð2 þ aÞ increases with a, a higher degree of privatization (i.e., a higher a) increases the output of firm M more under FDI than under export. 9 The first (second) term in the bracket indicates firm M’s mode of entry before (after) privatization. 10 We get @2 A=@2 cp ¼ ð2ð1 þ 2aÞÞ=ð2 þ aÞ3 o0.

ARTICLE IN PRESS 780

A. Mukherjee, K. Suetrong / European Economic Review 53 (2009) 775–785

A

 =0

0

cp1

cp2

1* 2* a cp 2  =1

Fig. 1. The effect of privatization on the host-country welfare when the foreign firm always undertakes FDI.

production for firm P and the fall in output is also being partially compensated by the higher output of the foreign firm, thus reducing the negative impact of lower output of firm P on the host-country welfare. Further, as the cost difference between the firms increases, it helps to reduce the loss of consumer surplus due to privatization. Moreover, a higher degree of privatization increases the profit of the public firm, thus creating a positive impact on the host-country welfare. If the public firm is (almost) completely nationalized, the significantly higher weight on welfare maximization induces the public firm to produce a large amount of output. Hence, a slight privatization does not have a significant negative effect on consumer surplus, while it helps to increase the profit of the public firm. Therefore, if the public firm is (almost) completely nationalized, the effect of higher domestic profit due to privatization dominates the loss of consumer surplus, thus creating an incentive for privatization. On the other hand, if the public firm is almost completely privatized, the output of the public firm is not very large, and a further reduction of the public firm’s output due to privatization creates a significant negative impact on consumer surplus. In this situation, the loss of consumer surplus due to privatization dominates the gain in profit in the public firm, thus reducing the incentive for privatization. Therefore, for any cost difference between the public and the foreign firms, there exists an optimal degree of privatization that balances the positive effect of the higher profit in the public firm and the negative effect of the loss of consumer surplus. Furthermore, as the cost efficiency of the foreign firm compared to the public firm increases, it reduces the loss of consumer surplus for a given degree of privatization, thus increasing the optimal degree of privatization.

4.2.2. When the foreign firm always exports Let us now consider the situation where the fixed cost of FDI is sufficiently high so that f4fp(1), and firm M exports irrespective of the degree of privatization. In this situation, the host-country welfare is wxp ðaÞ ¼

2ðaa  cp ð1 þ aÞ þ cx Þðað2  aÞ  2cp þ acx Þ þ ð2a  cp  cx Þ2 2ð2 þ aÞ2

(13)

Looking at the relationship between the host-country welfare and the degree of privatization, we get that @B=@cp 40, where B  @wxp =@a (see Appendix A for the derivation), which suggests that B and cp are positively related. At any given a, the relationship between B and cp is concave11 and the value of cp which maximizes B is greater than a/2. Hence, B is positively sloped with respect to cp over the interval [cx,a/2]. We also find that the qualitative relationship between B, cp and a is similar to the relationship between A, cp and a shown in Fig. 1. The relationship between B, cp and a is shown in Fig. 2. Hence, we get the following proposition immediately from Fig. 2. Proposition 3. Given that 0 ¼ cfocxpcp, if the fixed cost of FDI is very high so that the foreign firm exports irrespective of the degree of privatization, complete nationalization is the optimal strategy of the host-country government if the public firm is equally efficient to the private firm (i.e., cx ¼ cp), and partial privatization is the optimal strategy of the host-country government if the public firm is more cost inefficient than the foreign firm (i.e., cxocpoa/2). The intuition for the above result is similar to that of Proposition 2. In Fig. 2, a higher difference between cp and cx implies a higher degree of privatization. Hence, ceteris paribus, a fall in cx, which may be the outcome of trade liberalization, increases the host-country government’s incentive for privatization. Propositions 2 and 3 extend the results of Matsumura (1998). However, the most interesting situation arises when privatization affects firm M’s incentive for FDI. 11

We get @2 B=@c2p ¼ 2ð1 þ 2aÞ=ð2 þ aÞ3 .

ARTICLE IN PRESS A. Mukherjee, K. Suetrong / European Economic Review 53 (2009) 775–785

781

B

 =0 cx

x cp1

x cp2

a 2

* x1 * x2 cp  =1

Fig. 2. The effect of privatization on the host-country welfare when the foreign firm always exports.

4.2.3. FDI inducing privatization Let us now consider the situation where the fixed cost of FDI is moderate so that fp(0)ofofp(1). In this situation, the foreign firm exports without any degree of privatization, whereas it undertakes FDI under a suitable degree of privatization. Hence, privatization can induce the foreign firm to switch its production strategy from export to FDI. However, it remains to see whether attracting FDI through a privatization policy is worth for the host-country. It follows from (6) that, given the fixed cost of FDI, there exists a minimum a, say, af1, such that firm M is indifferent between FDI and exporting at this minimum a, and if a is greater than af1, firm M finds FDI profitable than exporting. Furthermore, as the fixed cost of FDI increases, af1 increases. We also find that, for a given a, cp and cx, the host-country welfare is higher under FDI than under exporting by firm M. However, the maximum host-country welfare under exporting by firm M is higher than the host-country welfare ‘‘if firm M undertakes FDI and there is complete privatization’’.12 Depending on the fixed cost of FDI, which determines af1, the welfare analysis of this subsection can be summarized into three possible cases. Fig. 3 shows the situation where the fixed cost of FDI is moderate but sufficiently small so that af1 is less than the value of a that maximizes the host-country welfare under FDI (say, af*). In this situation, the host-country government prefers to privatize up to af*, since it creates the maximum possible welfare for the host-country (which occurs under FDI by firm M) and also induces firm M to undertake FDI by satisfying condition (6). It also follows from Fig. 3 that if FDI was not an option to firm M, i.e., if firm M could serve the host-country only through exporting, the optimal degree of privatization would be ax*, which maximized the host-country welfare under exporting by firm M. Therefore, the possibility of FDI induces the host-country government to increase the degree of privatization compared to the situation with no FDI (af*4ax*). Next, consider the situation where the fixed cost of FDI is such that af14af* and the host-country welfare under FDI at af1 is greater than the maximum host-country welfare under exporting. This is shown in Fig. 4. In this situation, the optimal degree of privatization is af1. If the host-country government privatizes less than af1, firm M serves the host-country through export, and the hostcountry welfare follows the curve w(X,X) in Fig. 4, where the first (second) argument in w is firm M’s mode of production before (after) privatization. If firm M serves the host-country through export, the maximum possible welfare of the hostcountry is given by point A, which corresponds to the degree of privatization ax*. However, if the degree of privatization is equal to or more than af1, firm M undertakes FDI and the host-country welfare follows the curve w(F,F) from point B. Since the welfare function w(F,F) is negatively sloped with respect to a for aXaf1, the degree of privatization that maximizes the host-country welfare conditional on FDI by firm M is af1. Further, given that the host-country welfare at af1 (which is given by point B) is greater than the maximum host-country welfare under exporting by firm M (which is given by point A), the optimal degree of privatization is given by af1. Note that, even if af1 is not the degree of privatization that maximizes w(F,F), it maximizes the host-country welfare taking into account the equilibrium production strategy of firm M. Lastly, consider Fig. 5, which considers the situation where af14af* and the host-country welfare under FDI at af1 (which is shown by B) is lower than the maximum host-country welfare under exporting (which is shown by A). In this situation, following the above discussion, if the host-country government wishes to attract FDI, it would privatize up to af1. However, given that the welfare at af1 conditional on FDI by firm M is lower than the maximum welfare under exporting by firm M,

12 If cp ¼ cx, we get that w(F,F,a ¼ 1)4w(X,X,a ¼ 1) but w(F,F,a ¼ 1)ow(X,X,a ¼ 0), where the first (second) argument in w shows firm M’s mode of production before (after) privatization, and the third argument in w shows the degree of privatization. Since welfare is continuous in aA[0,1], it implies that there exists a such that the maximum host-country welfare under exporting by firm M is higher than the host-country welfare ‘‘if firm M undertakes FDI and a ¼ 1’’. Further, welfare under exporting by firm M increases with lower cx, implying that if cx falls from cp, it increases the possibility of getting ‘‘maximum host-country welfare under exporting by firm M’’ to be higher than the ‘‘host-country welfare when firm M undertakes FDI and a ¼ 1’’.

ARTICLE IN PRESS 782

A. Mukherjee, K. Suetrong / European Economic Review 53 (2009) 775–785

W

w(F, F) w(X, X)

0

*x α f1

α*f



1

Fig. 3. Privatization attracting FDI, when af1oaf*.

W

B A w(F, F) w(X, X)

0

*x

α*f

αf1



1

Fig. 4. Privatization attracting FDI, when af14af*.

W

A B w(F, F) w(X, X)

0

*x

α*f

αf1 1



Fig. 5. Privatization attracting FDI, when af1 is very large.

the optimal degree of privatization in Fig. 5 is ax*. Even if privatization could attract FDI, the host-country would prefer to privatize in a way that would not attract FDI and would generate maximum host-country welfare under exporting by firm M. Since the degree of privatization which is required to attract FDI is very high, the negative impact of privatization (which is the loss of consumer surplus) becomes significant and is not outweighed by the higher domestic profit. Hence, even if privatization could induce firm M to choose a more cost-efficient production strategy, the host-country would prefer to sacrifice that benefit of privatization, and would privatize only up to the point which would maximize host-country welfare under exporting by firm M. Therefore, in this situation, the optimal degree of privatization remains the same with and without the possibility of FDI. In sum, the above analysis shows that whether privatization that brings FDI improves the host-country welfare is not immediate and depends on the extent of privatization which is required to attract FDI. It is possible that w(X,X) is higher than w(X,F) and the host-country government prefers firm M to export than to undertake FDI. We also show that the

ARTICLE IN PRESS A. Mukherjee, K. Suetrong / European Economic Review 53 (2009) 775–785

783

host-country government will privatize its state-owned firm at least up to ax* and it will privatize beyond this point only if w(X,F) is higher than maximum w(X,X). Hence, the possibility of FDI can increase the incentive for privatization. The above discussion is summarized in the following proposition. Proposition privatization privatization privatization

4. Whether privatization that induces FDI improves the host-country welfare depends on the extent to which is required to attract FDI. The possibility of FDI increases the incentive for privatization, unless the amount of that is required to attract FDI is not high enough to create lower host-country welfare under FDI inducing compared to the maximum host-country welfare under exporting by the foreign firm.

The above result implies that, if the host-country government can reduce the cost of FDI, there are situations where the host-country government may prefer to adopt both privatization and FDI policies (which helps to reduce the cost of FDI) to attract FDI. More specifically, for Figs. 4 and 5, the host-county government would prefer to privatize up to af* and to reduce the cost of FDI in a way so that the foreign firm prefers FDI than exporting at af*. 4.3. The effects of cost reduction under privatization It has been noted that one of the main aims of privatization is to promote efficiency in the firms and to raise revenue for the states.13 The high costs of production in the public firms compared to its private competitors may be due to the limited provision on the firm’s R&D resources and/or managerial slackness. Privatization may help to correct this inefficiency in the public firm. However, to show the relationship between the effects of privatization and FDI in the simplest way, we have abstracted the above analysis form the possibility of public firm’s cost reduction due to privatization. Mukherjee and Sinha (2007) show that cost reduction in a domestic profit-maximizing firm reduces a foreign firm’s incentive for FDI by making the domestic industry more competitive. Hence, it follows that if there is a cost reduction in the public firm, the net effect of privatization on the incentive for FDI depends on the relative strengths of higher private shareholdings, which tends to increase the market share of the foreign firm, and the cost reduction in the public firm, which tends to reduce the market share of the foreign firm. If the cost reduction in the public firm due to privatization is significantly large, privatization may reduce the incentive for FDI. In other words, Proposition 1 remains if the costreduction effect due to privatization is not very strong. It should also be clear that higher cost efficiency in the public firm due to privatization would increase the incentive for privatization if the degree of privatization did not affect the foreign firm’s mode of production. But, if the degree of privatization affects the production strategy of the foreign firm, it is not so straightforward whether higher cost efficiency in the public firm due to privatization increases the host-government’s incentive for privatization. It would depend on the cost reduction in the public firm due to privatization (which would also affect the incentive for FDI) and the cost change in the foreign firm due to its change in the production strategy following privatization. If the cost reduction in the public firm due to privatization does not reduce the incentive for FDI, cost reduction in the public firm is likely to increase the incentive for privatization. 5. Conclusion Though many developing and transitional economies are privatizing state-owned firms and also experiencing inflow of FDI, the existing theoretical literature did not capture both these aspects together. We take up this issue in this paper, and show the interaction between privatization and greenfield FDI. We show that there is complementarity between privatization and greenfield FDI. Privatization increases the incentive for FDI, which, in turn (generally), increases the incentive for privatization. However, it is not necessary that a country would always prefer to privatize up to a point that attracts FDI. If the degree of privatization that is required to attract FDI is sufficiently high, the host-country may not find it beneficial to attract FDI through privatization. Instead, it will privatize up to the point at which the host-country welfare is maximized under exporting by the foreign firm. We show that whether or not the degree of privatization affects the foreign firm’s mode of production, partial privatization is the optimal strategy of the host-country. Both the cost difference between the firms and the effect of privatization on the foreign firm’s production strategy play important roles in determining the privatization policy. There are, however, some important remarks need to be made. First, in our analysis, we have focused on the effect of privatization on the foreign firm’s production strategy but abstracted our analysis from entry of the host-country firms. Privatization may attract host-country firms by reducing the output of the public firm, thus leaving more residual demand for the potential host-country firms. Higher competition in the host-country market due to host-country firms’ entry reduces the residual demand for the foreign firm and may adversely affect the foreign firm’s incentive for FDI. Hence, the effects of domestic-entry on the foreign firm’s incentive for FDI following privatization and the corresponding welfare implications will be similar to the effects of cost reduction in the public firm, which has been discussed in Section 4.3. Second, we have considered production decision of a single foreign firm, thus ignoring competition for FDI. If there are multiple foreign firms, which serve the host-country either through FDI or through export irrespective of the privatization 13

See Megginson and Netter (2001) for a recent survey on privatization, profitability and efficiency of the firms.

ARTICLE IN PRESS 784

A. Mukherjee, K. Suetrong / European Economic Review 53 (2009) 775–785

policy of the host-country, while firm M in our analysis chooses its mode of production, our qualitative results will remain.14 Privatization will increase the outputs of all foreign firms. However, since only firm M decides on exporting and FDI and the marginal cost of firm M is higher under exporting than under FDI, firm M’s benefit due to privatization is higher under FDI than under exporting, thus increasing the incentive for FDI under privatization. Further, the possibility of FDI will increase the incentive for privatization since it will induce firm M to choose a more cost-efficient production option, unless the degree of privatization (which asks for sacrificing the disciplining effect of nationalization) that is required to attract FDI is not very high. The presence of other foreign firms in the market will only affect the equilibrium values by affecting the residual demand for firm M. A more interesting situation arises if many foreign firms decide on exporting and FDI. In this situation, a foreign firm’s production decision creates a strategic effect on other foreign firms’ plant location decision, along with the effects shown in this paper. Since this problem requires a complete analysis and is outside the scope of this paper, we leave it for future research. Third, following the literature on privatization and the practice of many countries, we have assumed that the shares of the public firm are sold to the domestic investors. However, there are situations where, in the case of privatization, the foreign firms buy shares of the domestic public firms. Hence, foreign acquisition of the public firm can be an area for future research. Acquisition of the domestic public firm by the foreign firm may be viewed as the foreign firm’s strategy to eliminate competition in the domestic market. Since the market becomes more attractive for investment after privatization, the foreign firm may have higher incentive for greenfield FDI compared to the situation where acquisition by the foreign firm is prohibited in the privatization policy. Finally, as a first step in explaining the relationship between privatization and FDI, we follow the tradition of the mixed oligopoly literature and ignore agency problems in the firms. In the presence of agency problems, privatization may affect the organizational structure of the public firm, and may have significant impacts on the contracts between the agents, which may have important implications on the outputs, profits and welfare.15 We intend to take up these and the related issues in our future research.

Acknowledgments We thank two anonymous referees and an Associate Editor of this journal, Hartmut Egger, Catia Montagna, Zhihong Yu, the seminar participants at the school of Economics of the University of Nottingham, CCER of the Peking University and the 6th GEP postgraduate conference for useful comments and suggestions. Arijit Mukherjee has benefited from an informal discussion with Bibhas Saha. The views presented here are solely of the authors and not necessarily of their institutions. The usual disclaimer applies. Appendix A The derivation for @B=@cp 40: We get 3 2 ð2 þ aÞða  cp Þðað2  aÞ  2cp þ acx Þ 7 6 ð2 þ aÞða  c Þðaa  c ð1 þ aÞ þ c Þ x p x 7 6 @wxp 1 7 6 ¼ B¼ 6 @a ð2 þ aÞ3 4 2ðaa  cp ð1 þ aÞ þ cx Þðað2  aÞ  2cp þ acx Þ 7 5 ð2a  cp  cx Þ2 and @B 1 ¼ ½2ða  cp Þ þ að7a  3cx  4cp Þ40 @cp ð2 þ aÞ3

References Anderson, S.P., Palma, A.d., Thisse, J.-F., 1997. Privatization and efficiency in a differentiated industry. European Economic Review 41, 1635–1654. Baer, W., 1994. Privatisation in Latin America. The World Economy 17, 509–528. Barros, F., 1995. Incentive schemes as strategic variables: an application to a mixed duopoly. International Journal of Industrial Organisation 13, 373–386. Beato, P., Mas-Colell, A., 1984. The marginal cost pricing rule as a regulation mechanism in mixed markets. In: Merchand, M., Pestieau, P., Tulkens, H. (Eds.), The performance of public enterprises. North-Holland, Amsterdam. Cremer, H., Marchand,, M., Thisse, J.-F, 1989. The public firm as an instrument for regulating an oligopolistic market. Oxford Economic Papers 41, 283–301. De Fraja, G., Delbono, F., 1989. Alternative strategies of a public enterprise in oligopoly. Oxford Economic Papers 41, 302–311. Fershtman, C., 1990. The interdependence between ownership status and market structure: the case of privatization. Economica 57, 319–328. Fjell, K., Heywood, J.S., 2002. Public Stackelberg leadership in a mixed oligopoly with foreign firms. Australian Economic Papers 41, 267–281.

14 15

The confirmation of the qualitative results of our Propositions 2 and 3 may be anticipated from Matsumura and Kanda (2005). See Schmidt (1996) for a survey on the incomplete contract approach to privatization.

ARTICLE IN PRESS A. Mukherjee, K. Suetrong / European Economic Review 53 (2009) 775–785

785

Gani, A., 2005. Foreign direct investment and privatization. USPEC Working Paper, No. 2005/6, Department of Economics, University of the South Pacific. Gupta, N., Ham, J.C., Svejnar, J., 2008. Priorities and sequencing in privatization: evidence from Czech from panel data. European Economic Review 52, 183–208. Harris, R.G., Wiens, E.G., 1980. Government enterprise: an instrument for the internal regulation of industry. Canadian Journal of Economics 13, 125–132. Ishibashi, I., Matsumura, T., 2006. R&D competition between public and private sectors. European Economic Review 50, 1347–1366. Kapur, D., Ramamurti, R., forthcoming. Privatization in India: the imperatives and consequences of gradualism. In: Srinivasan (Ed.), India After a Decade of Economic Reforms: Retrospect and Prospects. Stanford University Press. Kikeri, S., Kolo, A.F., 2005. Privatization: trends and recent developments. World Bank Policy Research Working Paper, No. 3765. Matsumura, T., 1998. Partial privatization in mixed duopoly. Journal of Public Economics 70, 473–483. Matsumura, T., Kanda, O., 2005. Mixed oligopoly at free entry markets. Journal of Economics 84, 27–48. Maw, J., 2002. Partial privatization in transition economies. Economic Systems 26, 271–282. Megginson, W.L., Netter, J.M., 2001. From state to market: a survey of empirical studies on privatization. Journal of Economic Literature 39, 321–389. Merlevede, B. Schoors, K., 2005. How to catch foreign fish? FDI and privatization in EU accession countries. Working Paper, No. 785, William Davidson Institute. Mukherjee, A., Sinha, U.B., 2007. Welfare-reducing domestic cost reduction. Review of International Economics 15, 294–301. Ohori, S., 2004. Environmental tax, trade, and privatization. The Kyoto Economic Review 73, 109–120. Ohori, S., 2006. Optimal environmental tax and level of privatization in an international duopoly. Journal of Regulatory Economics 29, 225–233. Pack, H., Saggi, K., 1997. Inflows of foreign technology and indigenous technological development. Review of Development Economics 1, 81–98. Pal, D., White, M.D., 1998. Mixed oligopoly, privatization, and strategic trade policy. Southern Economic Journal 65, 264–281. Saggi, K., 2002. Trade, foreign direct investment, and international technology transfer: a survey. World Bank Research Observer 17, 191–235. Saha, B., Sensarma, R., 2004. Divestment and bank competition. Journal of Economics (Zeitschrift fu¨r Nationalo¨konomie) 81, 223–247. Schmidt, K.M., 1996. Incomplete contracts and privatization. European Economic Review 40, 569–579. Schmidt, K. M. Schnitzer, M., 1997. Methods of privatization: auctions, bargaining and give-aways. CEPR Discussion Paper, No. 1541. UNCTAD, 2000. World Investment Report, United Nations, New York and Geneva. UNCTAD, 2002. World Investment Report, United Nations, New York and Geneva. Vickers, J., Yarrow, G., 1991. Economic perspectives on privatization. Journal of Economic Perspectives 5, 111–132.