Multi-sourcing as an entry deterrence strategy☆

Arijit Mukherjee a,b, Yingyi Tsai c,⁎a University of York, UKb CESifo, Germanyc National University of Kaohsiung, Taiwan

a r t i c l e i n f o a b s t r a c t

Article history:Received 19 August 2011Received in revised form 18 April 2012Accepted 19 April 2012Available online 21 June 2012

This paper studies the rationale for multiple sourcing. In a simple model of outsourcing thatembodies technology transfer and the threat of competition from the supplier(s) due toimitation, we show that multiple sourcing helps to deter entry by the suppliers into the finalgoods market and enhances profitability of the outsourcing firm. Our explanation for multipleoutsourcing differs from the standard arguments, which either reduce the double marginal-ization problem or eliminate supply bottlenecks.

© 2012 Published by Elsevier Inc.JEL classifications:F12F23L13L24

Keywords:EntryImitationMultiple sourcingVertical technology transfer

1. Introduction

The tremendous growth of international outsourcing, which allows firms in the developed-countries to purchase intermediategoods and components from the developing countries, often involves vertical technology transfer. The sharing of knowledge in avertical context between the buyers from the developed economies and their suppliers in the developing countries has been welldocumented. For example, in a study with extensive interview over Korean firms in the 1970s, Rhee, Ross-Larson, and Pursell(1984) show that almost half of the firms in the sample responded as having benefitted from the technical information providedby the foreign buyers. Keesing (1982) reports, in the 1970s, that foreign corporations had located a large number of staffs sharingtheir manufacturing skills with the local suppliers in Korea and Taiwan. Hou and Gee (1993) also confirm a substantial level oftechnology transfer to the producers in the newly industrialized economies from their buyers in the developed countries. Otherempirical evidences abound. Hobday (1995) shows that vertical knowledge transfer occurs as firms from industrialized countriespurchase components and intermediate products from Asian newly industrialized countries. Pack and Saggi (2001), using RadioShack and Texas Instruments as examples, also show that firms from newly industrialized countries have been contracted toproduce components and even the end products. More evidences on vertical technology transfer can also be found in Javorcik(2004) and Blalock and Gertler (2008).

International Review of Economics and Finance 25 (2013) 108–112

☆ The authors thank an anonymous referee of this journal for valuable comments and suggestions. The authors thank Justin Yifu Lin for his intellectual inputs inan earlier version of this paper entitled “Why do firms engage in multi-sourcing?” and Hideshi Itoh for valuable comments. Yingyi Tsai gratefully acknowledgespartial financial support from the National Science Council, Taiwan under research grant no. “NSC99-2410-H-390-009”. The usual disclaimer applies.⁎ Corresponding author at: Department of Applied Economics, National University of Kaohsiung, Nan-Tze 811, Kaohsiung, Taiwan. Fax: +886 7 216 9365.

E-mail address: yytsai@nuk.edu.tw (Y. Tsai).

1059-0560/$ – see front matter © 2012 Published by Elsevier Inc.doi:10.1016/j.iref.2012.04.003

Contents lists available at SciVerse ScienceDirect

International Review of Economics and Finance

j ourna l homepage: www.e lsev ie r .com/ locate / i re f

A closer observation of the development in international outsourcing suggests that quite often a sourcing firm (or buyer) maycontract with several suppliers. For example, Mattel, a premier toy brand based in the US, purchases from a large number of third-party manufacturers across countries such as the US, Mexico, Brazil, Asia (including China and India), New Zealand, and Australia.Further, it was reported that Motorola, in 2008, will outsource 75 million to 80 million handsets to original design manufacturers(ODMs), with more than 50 million devices outsourced to Compal Communications and the remainder contracted to Chi MeiCommunications System (Digitimes, 2007).

The standard argument for multiple outsourcing follows from Pack and Saggi (2001), suggesting that multiple outsourcingreduces the double marginalization problem.1 In contrast, we consider that the outsourcing firm can design the contracts in a wayso that it can extract the entire surplus from the suppliers, thus eliminating the double marginalization problem. Nonetheless, itremains an important question as to why an outsourcing firm prefers sourcing out to multiple firms.

Andrabi, Ghatak, and Khwaja (2006) argue that subcontracting with multiple suppliers can be the equilibrium outcome in theabsence of hold-up problem,2 even though there are supply bottlenecks. They have shown, in the presence of demand uncertainty,that a buyer may want to contract with several suppliers in order to reduce the problem of supply bottlenecks.

We provide a new rationale for multiple outsourcing. We show that even if the outsourcing firm can extract the entire surplusfrom the suppliers, thus eliminating the double marginalization problem, and in the absence of supply bottlenecks, multipleoutsourcing can occur since it helps to protect the outsourcing firm by eliminating the suppliers' incentive for entry in the finalgoods market. Thus, multi-sourcing acts as an entry deterrence strategy.

In light of the fact that existing literature did not paymuch attention to vertical technology transfer and multiple sourcing, whichare the focuses in the present study, we can relate, therefore, this paper to the literature on international outsourcing (Antràs &Helpman, 2004; Deardorff, 2005; Glass & Saggi, 2001; Grossman & Helpman, 2002, 2003; Horgos, 2009; Jones, 2005; Long, 2005;Marjit & Mukherjee, 2008; Matsuoka & Fukushima, 2010; Shy & Stenbacka, 2003; Tomiura, 2009).

The remainder of the paper is organized as follows.We describe themodel and derive the results in Section 2. Section 3 concludes.The proof is relegated to Appendix A.

2. Outsourcing to a single supplier

Consider a two-period model of outsourcing with firms a and b, each located in country A and country B, respectively. Assumethat, at the beginning of period 1, firm a owns a technology for the production of a particular product. Firm a can produce itself thecommodity at a constant marginal cost cm, and, thus, earns πat(cm) in each period t, where t=1, 2. For analytical simplicity, weassume away the discount of future profits, and that the market demand for the product in both periods is identical. Hence, total

profit of firm a is given byP2t¼1

πta cmð Þ ¼ 2πa cmð Þ.

Furthermore, we assume that firm b in country B does not have the technology to produce the goods; nevertheless, should itobtain firm a's technology, firm b is able to produce at a constant marginal cost cb, where cm>cb and cb=0 for simplicity. Hence,the incentive for outsourcing by firm a underlies in the lower production cost of firm b.

If firm b produces the product, the monopoly profit generated in each period is πat(0), with πat(0)>πat(cm). We normalize thereservation payoff of firm b to zero, for simplicity. Hence, firm a can outsource the product to firm b against a fee of πat(0) to bepaid in each period. We assume that firm a gets the payment separately in each period. Financial constraint faced by firm b mayjustify this assumption.

2.1. Outsourcing with no threat of competition

In the absence of imitation and entry by firm b subsequent to outsourcing, it is optimal for firm a to outsource production to

firm b against a per-period payment of πat(0). Under such circumstance, the total profit of firm a isP2t¼1

πta 0ð Þ ¼ 2πa 0ð Þ, and there is

no need for multi-sourcing in the absence of imitation. This is different from Pack and Saggi (2001), where the outsourcing firmcannot extract the entire surplus from outsourcing and the incentive for multiple-sourcing remains under no imitation by thesupplier. The incentive for multi-sourcing in Pack and Saggi (2001) is due to the double marginalization problem, which is notpresent in our analysis.

2.2. Outsourcing due to the threat of competition

Now we consider the possibility of imitation by the supplier, which provides room for the supplier to enter product marketand to compete with the outsourcing firm. We assume that firm b requires one period to learn and imitate the technology of firma ex-post outsourcing. After its successful learning of the technology, firm b is able to compete with firm a in the product marketwith a homogeneous product. Firm b has to incur an entry cost of I should it decide to enter the market.

1 Pack and Saggi (2001) do not consider multiple technology transfers explicitly. Instead, they show that knowledge spillover in the supplier's market increasesthe profit of the outsourcing firm, thus creating the incentive for multiple technology transfers. Dinda and Mukherjee (2009) consider vertical technology transferin the presence of strategic government policy.

2 See Grossman and Hart (1986) for an important work on the hold-up problem in the property right literature.

109A. Mukherjee, Y. Tsai / International Review of Economics and Finance 25 (2013) 108–112

We consider the following game. At the beginning of period 1, firm a offers an outsourcing contract to firm b specifying thepayments for periods 1 and 2, should firm b use the technology of firm a. Firm b decides whether or not to accept the offer. If firmb accepts the offer, it uses firm a's technology for the production of the final output, and makes the payment to firm a after period1. At the beginning of period 2, firm b decides, after having learned the technology during period 1, whether or not to enter theproduct market and, thus, breaks the contract with the buyer firm a.3 Clearly, firm b enters the market when its net profit exceedsthat which it obtains otherwise. If firm b rejects firm a's offer in period 1, firm a then produces itself the product.4 We solve thegame through backward induction.

Under imitation by firm b, the threat of competition restricts profit extraction by firm a. Hence, while offering the contract atthe beginning of period 1, firm a needs to design a period-specific contract which discourages firm b from entering the market,since competition in the product market reduces the industry profit and the profit of firm a.

Let us first determine the profit of firm b in period 2, when it produces with firm a's technology in period 1 but competesagainst firm a in period 2. The net profits of firms a and b in period 2 are πa2(cm,0) and πb2(cm,0)− I>0, respectively. The first(second) argument in πi2(cm,0), i=a, b, shows the marginal cost of firm a (firm b). It follows that the outsourcing contract shouldleave firm bwith a positive profit of πb2(cm,0)− I in order to prevent firm b from entering the market. Therefore, the price paid byfirm b in period 2 should not exceed πa1(0)−(πb2(cm,0)− I). Note that πa1(0)−πb2(cm, 0)>πa2(cm,0), i.e., the gross industry profit(which includes the entry-cost) is greater under monopoly than it is under duopoly, and this is always true for homogeneousproducts.

Since there is no imitation in period 1, the payment under outsourcing for period 1 would be πa1(0). The profit of firms a and bin period 1 is πa1(0) and 0, respectively.

Therefore, at the beginning of period 1, firm a offers a contract specifying that firm b pays πa1(0) and πa1(0)−πb2(cm,0)+ I in periods1 and 2, respectively. Note that firm b cannot be better off either by rejecting this offer in period 1 or by canceling the contract inperiod 2 for entering the product market. Total net profit of firms a and b in both periods is given by 2πa1(0)−πb2(cm,0)+ I andπb2(cm,0)− I, respectively.

Since it takes one period before imitation can occur, firm amay decide to outsource only in period 2. Under such circumstance,firm a earns a total profit of πa1(cm)+πa2(0), while the total profit of firm b is zero. However, outsourcing at the beginning of period1 dominates deferred-outsourcing if

π1a 0ð Þ þ I > π2

b cm;0ð Þ þ π1a cmð Þ: ð1Þ

Eq. (1) suggests, in the presence of imitation by firm b, that regardless of firm a's outsourcing choice made in period 1 or inperiod 2, the total profit of firm a under outsourcing is lower as compared to the situation of no imitation by firm b.

3. Outsourcing to multiple suppliers

Now assume that firm a decides to outsource in period 1, but, instead of outsourcing to firm b only, it outsources to twosymmetric firms, b and c. Neither b nor c has the technology to produce the product of firm a, but both of them are able to produceafter having obtained firm a's technology. Both firms b and c require one period to imitate the technology of firm a, and each ofthem requires the entry-cost I to enter the product-market.

We consider the following game. At the beginning of period 1, firm a offers period-specific contracts to both suppliers b and c,who decide non-cooperatively whether or not to accept the contract. For any supplier accepting the offer, it imitates thetechnology during period 1, and decides at the beginning of period 2 whether to enter the market or to continue with firm a'scontract. If both suppliers b and c have accepted the offer, they make the market-entry decision in period 2 simultaneously andnon-cooperatively. If either supplier b or c rejects the offer proposed by firm a at the beginning of period 1, only the supplieraccepting the offer makes the market-entry decision in period 2. We solve the game by backward induction.

Proposition 1. If πb1(0,0)=πc1(0,0)> I>πb1(cm,0,0)=πc1(cm, 0,0),5 firm a offers the following contract to both suppliers b and c at thebeginning of period 1: in both periods 1 and 2, each supplier b and c produces one half of the monopoly output corresponding to zeromarginal cost, and each pays πit(0)/2, i=b, c, to firm a.6 Both firms b and c accept this contract.

Proof. See Appendix A.

Intuitively, under multi-sourcing, if the market size and the cost of entry are such that the net profits of the suppliers turn to benegative when both entering the market in period 2, it creates a coordination problem between the suppliers. Hence, each of

3 For simplicity, we normalize the cost of breaking up a contract to zero. If there is a cost of breaking up the contract, given the possibility of imitation, firm b canalways contract with firm a separately in each period, if the cost of contracting (which is also normalized to zero in our analysis, for simplicity) is not very high.

4 Since imitation is not a credible threat if the offer is given in period 2, and because firm a can extract the entire profit generated in firm b, if firm b rejects theoffer in period 1, there is no reason for it to accept an offer in period 2. Hence, if the offer in period 1 is rejected, there is no offer in period 2.

5 The argument of cm, 0, and 0 in πi1, i=b, c, represents the marginal cost of firms a, b, and c, respectively.6 Note that πit(0)=πjt(0) where i, j=b, c, and i≠ j.

110 A. Mukherjee, Y. Tsai / International Review of Economics and Finance 25 (2013) 108–112

them randomizes the entry decision, and the net expected equilibrium profit of each supplier is the same under entry and underno-entry. Thus, even if outsourcing creates knowledge transfer (or spillovers), multi-sourcing prevents entry of the suppliers bycreating competition between them, and helps to protect the market of the outsourcing firm from the suppliers.

If the condition in Proposition 1 holds, the profit of firm a under multi-sourcing is the same to that of under no imitation.However, if the condition in Proposition 1 does not hold but there are a large number of potential suppliers, given the market sizeand the entry-cost, firm a can choose the number of suppliers suitably so that, in the entry game of period 2, the expectedequilibrium payoff of each supplier is zero, which is the profit of a supplier under no-entry. Hence, by reducing the gross profit ofthe suppliers, multi-sourcing protects the market of the outsourcing firm by deterring entry of the suppliers.

3.1. The rationale for employing a mixed strategy7

It is straightforward that if firm a outsources to both firms b and c, there are two pure strategy equilibria (E, NE) and (NE, E)along with the mixed strategy as considered above. This happens since we are considering the situation with πb1(0,0)=πc1(0,0)>I>πb1(cm,0,0)=πc1(cm,0,0). Hence, it is worth discussing the rationale for looking into the mixed-strategy in the above analysis.

Our approach is similar to the one considered by Cabral (2004) in a different context.8 In our analysis, firms b and c aresymmetric, yet their entry decisions create two asymmetric pure strategy equilibria. While firm b benefits from the equilibrium(E, NE), firm c benefits from the equilibrium (NE, E), thus may encourage both firms to take the action E. However, if both firmstake the action E, both firms will be worse off as is compared to the situation when at least one of them takes the action NE. Hence,it is not immediate which actions will be employed by firms b and c, thus creating an uncertainty about the actions taken by firmsb and c. Following Rasmusen (1989, p. 72), we can then suggest that the best prediction in this situation is perhaps given by themixed-strategy, “… because its symmetry makes it a focal point of sorts, and does not require any differences between theplayers.” Alternatively, following Dixit and Skeath (1999, p. 139), we can argue that, if a game between the symmetric agentscreate multiple pure strategy asymmetric equilibria where different agents benefit from different equilibria, the agents mayprefer to adopt the “keep-the-opponent-indifferent method”, which can be achieved in the mixed-strategy equilibrium. Thesejustifications naturally lead us to consider mixed-strategy equilibria when there are two asymmetric pure strategy equilibria inthe entry-game played by firms b and c.

4. Conclusion

There are ample evidences of multi-sourcing, yet the theoretical literature has not paid much attention to this issue at hand.We provide a strategic reason for multi-sourcing. We show that, in the presence of imitation under outsourcing, multi-sourcingacts as an entry deterrence strategy by creating an entry game between the suppliers, thus helping the outsourcing firm to extractgreater surplus than it does under the scenario of single outsourcing. In this respect, our analysis explores the implications foroutsourcing of market size and the cost of entry.

Appendix A

Proof of Proposition 1. At the beginning of period 2, firms b and c decide simultaneously whether or not to enter the market. Theentry-game between firms b and c generates the payoff functions shown in Table 1. In Table 1, the strategies of firms b and c arecharacterized by E (i.e., entry and breaking the contract with firm a) and NE (i.e., no-entry and continuing the contract with firm a).

Assume that firms b and c enter (resp. do not enter) the market with probabilities pb and pc (resp. (1−pb) and (1−pc)),respectively. The expected payoff of firm b under entry is

pc π1b cm;0;0ð Þ−I

� �þ 1−pcð Þ π1

b 0;0ð Þ−I� �

; ðA1Þ

and its profit is zero if it does not enter the market. Setting Eq. (A1) as equal to zero, the equilibrium value of pc is given by

pc ¼π1b 0;0ð Þ−I

π1b 0;0ð Þ−π1

b cm;0;0ð Þ ; ðA2Þ

where pc∈(0,1) for πb1(0,0)=πc1(0,0)> I>πb1(cm,0,0)=πc1(cm,0,0).

7 We thank an anonymous referee for suggesting the elaboration of this issue explicitly.8 See, for example, the “grab-the-dollar entry” in Cabral (2004).

111A. Mukherjee, Y. Tsai / International Review of Economics and Finance 25 (2013) 108–112

Similarly, we can solve for the equilibrium values of pb as

pb ¼ π1c 0;0ð Þ−I

π1c 0;0ð Þ−π1

c cm;0;0ð Þ ; ðA3Þ

where pb∈(0,1) for πb1(0,0)=πc1(0,0)> I>πb1(cm,0,0)=πc1(cm,0,0).

It follows from Eqs. (A2) and (A3) that, if πb1(0,0)=πc1(0,0)> I>πb1(cm,0,0)=πc1(cm,0,0), the expected payoffs of firms b and cin period 2 are

pbð Þ2 π1b cm;0;0ð Þ−I

� �þ pb 1−pbð Þ π1

b 0;0ð Þ−I� �

¼ 0; ðA4Þ

and neither firm b nor firm c is better off by breaking up the contract with firm a to enter the market.

Both firms b and c produce for firm a in both periods. Hence, it is optimal for firm a to design the outsourcing contracts in a wayso that the monopoly output is divided equally to both suppliers b and c, thus allowing firm a to extract πit(0)/2 from eachsupplier, i=b, c. This contract is accepted by both firms b and c, since neither of them can be better off by rejecting it.□

References

Andrabi, T., Ghatak, M., & Khwaja, A. I. (2006). Subcontractors for tractors: Theory and evidence on flexible specialization, supplier selection, and contracting.Journal of Development Economics, 79, 273–302.

Antràs, P., & Helpman, E. (2004). Global sourcing. Journal of Political Economy, 112, 552–580.Blalock, G., & Gertler, G. (2008). Welfare gains from foreign direct investment through technology transfer to local suppliers. Journal of International Economics, 74,

402–421.Cabral, L. M. B. (2004). Simultaneous entry and welfare. European Economic Review, 48, 943–957.Deardorff, A. V. (2005). A trade theorist's take on skilled-labor outsourcing. International Review of Economics and Finance, 14, 251–271.Digitimes (2007). Available at http://www.digitimes.com/Tornado/V4/SearchEnd.asp, accessed 10 May 2008Dinda, S., & Mukherjee, A. (2009). International outsourcing, tax and patent protection. Journal of Public Economic Theory, 13, 139–154.Dixit, A., & Skeath, S. (1999). Games of strategy. New York, USA: W.W. Norton & Company, Inc.Glass, A. J., & Saggi, K. (2001). Innovation and wage effects of international outsourcing. European Economic Review, 45, 67–86.Grossman, S. J., & Hart, O. D. (1986). The costs and benefits of ownership: A theory of vertical and lateral integration. Journal of Political Economy, 94, 691–719.Grossman, G. M., & Helpman, E. (2002). Integration versus outsourcing in industry equilibrium. Quarterly Journal of Economics, 117, 58–119.Grossman, G. M., & Helpman, E. (2003). Outsourcing versus FDI in industry equilibrium. Journal of the European Economic Association, 1, 317–327.Hobday, M. (1995). Innovation in East Asia: The challenge to Japan. London: Edward Elgar.Horgos, D. (2009). Labor market effects of international outsourcing: How measurement matters. International Review of Economics and Finance, 18, 611–623.Hou, C. M., & Gee, S. (1993). National systems supporting technical advance in industry: The case of Taiwan. In R. R. Nelson (Ed.), National innovation systems: A

comparative analysis (pp. 384–413). New York: Oxford University Press.Javorcik, B. S. (2004). Does foreign direct investment increase the productivity of the domestic firm? In search of spillovers through backward linkages. American

Economic Review, 94, 605–627.Jones, R. W. (2005). Immigration vs. outsourcing: Effects on labor markets. International Review of Economics and Finance, 14, 105–114.Keesing, D. B. (1982). Exporting manufactured consumer goods from developing to developed economies: Marketing by local firms and effects of developing country

policies. Washington, DC: World Bank.Long, N. V. (2005). Outsourcing and technology spillovers. International Review of Economics and Finance, 14, 297–304.Marjit, S., & Mukherjee, A. (2008). International outsourcing and R&D: Long run implications for consumers. Review of International Economics, 16, 1010–1022.Matsuoka, Y., & Fukushima, M. (2010). Time zones, shift working and international outsourcing. International Review of Economics and Finance, 19, 769–778.Pack, H., & Saggi, K. (2001). Vertical technology transfer via international outsourcing. Journal of Development Economics, 65, 389–415.Rasmusen, E. (1989). Games and information: An introduction to game theory. Cambridge, Massachusetts, USA: Blackwell Publishers.Rhee, Y., Ross-Larson, B., & Pursell, G. (1984). Korea's competitive edge: Managing entry into world market. Baltimore: Johns Hopkins University Press.Shy, O., & Stenbacka, R. (2003). Strategic outsourcing. Journal of Economic Behavior and Organization, 50, 203–224.Tomiura, E. (2009). Foreign versus domestic outsourcing: Firm-level evidence on the role of technology. International Review of Economics and Finance, 18,

219–226.

Table 1Payoffs to firms b and c in period 2.

Firm cE NE

Firm b E πb2(cm,0,0)− I, πc2(cm, 0,0)− I πb2(0,0)− I, 0NE 0, πc2(0,0)− I 0, 0

112 A. Mukherjee, Y. Tsai / International Review of Economics and Finance 25 (2013) 108–112