Investment liberalisation and firm selection process: A welfare analysis from a host-country perspective

Abstract

This paper analyses the welfare and market structure effects of investment liberalisation allowing for intra-industry firm-heterogeneity to account explicitly for the firm selection process induced by foreign firm entry and its interactions with productivity spillovers. Using a two-stage oligopolistic model in which a foreign firm decides whether and how to enter the host-country market (export versus foreign direct investment) while two asymmetrical local firms decide on their exit/stay strategy, it is shown that even where productivity spillover effects are absent and the entry of the multinational firm leads to the crowding-out of local firms, foreign direct investment can still improve host-country welfare.

Keywords:

• foreign direct investment
• welfare
• firm selection process
• market structure
• crowding-out
• productivity spillovers

JEL Classifications:

• F23
• L11
• D6

Acknowledgements

The author is particularly grateful to M. Motta and A. Kumar for invaluable advice and comments. The author has benefited greatly from P.J. Buckley's and J.R. Markusen's suggestions on an earlier draft of the paper. Finally, the author would like to thank three anonymous referees for helpful comments. Any remaining errors are the author's.

Notes

 1. Generally, productivity spillovers are said to take place ‘when entry or presence of a MNE affiliates lead to productivity or efficiency benefits in the host country's local firms, and the MNEs are not able to internalize the full value of these benefits’ (Blomström and Kokko 1998).

 2. The exceptions are Glass and Saggi (2002) and Görg (2000), who assumed asymmetrical firms within an industry; however, they focused on the rate and magnitude of innovation under the FDI and licensing entry mode (Glass and Saggi 2002) and in the greenfield versus acquisition decision of an MNE (Görg 2000) rather than on the firm selection process induced by the entry of an MNE.

 3. However, within the given partial equilibrium setting, the welfare analysis cannot capture the whole complexity of welfare effects of inward FDI; in particular, it doesn't allow addressing employment effects of domestic industry displacement.

 4. This is in line with the usual assumption in the FDI literature whereby multinational firms are typically assumed to possess a firm-specific ownership advantage (a product or production process that gives the firm a market power advantage in foreign markets) that allows them to offset the significant costs of doing business abroad relative to local firms. In the model, I assume that the firm-specific asset takes the form of superior production technology, which is reflected in lower marginal costs. However, the existence of a cost advantage of a foreign firm is neither a necessary nor a sufficient condition for a firm to become a multinational (as demonstrated, for example, by Graham 1998).

 5. The model corresponds more to those industries characterised by relatively high entry barriers, such as one-off fixed costs where foreign firms (MNEs) are in a superior position compared with new potential domestic entrants (but not compared with existing domestic firms that have been already operating in the market) because they have already established production in their home countries, whereas other potential domestic entrants still have to incur start-up costs.

 6. However, the assumption that spillovers only occur in the case of direct investment and thus only have a local effect might not seem that straightforward. Several arguments for such an assumption are stressed in the literature. Fosfuri, Motta, and Rønde (2001) argued that where spillovers arise due to the mobility of workers previously trained and employed in MNE affiliates, it is more difficult for the local firm to identify and attract trained workers from abroad than from the local affiliate. Other potential sources of spillovers, such as demonstration and vertical linkage effects, also tend to be local in nature due to easier and cheaper communication in the case of a local presence.

 7. As pointed out by Görg (2000), due to the assumption of production of a homogeneous good, Bertrand competition would yield prices that are equal to marginal costs. In this case, the low-cost firm simply prices the high-cost firm out of the market. Hence, the Cournot model seems more appropriate for focusing on the effects of market entry on market structure.

 8. This means I assume that export costs are not prohibitively high. This implies assuming the profitable entry of firm M into the host-country market via exports, which is, according to Table 1, ensured by condition in case both host-country firms are operating in the market, and by in case firm H exits the market.

 9. However, a switch from [M^EX, L] to [M^FDI, L, H] is not possible in the absence of productivity spillovers, as from it follows that also .

10. With reference to Figure 1, λ^C _{[FDI,L,H]/[EX,L,H]} is defined by the boundary between [M^FDI, L, H] and [M^EX, L, H], λ^C _{[FDI,L]/[EX,L,H]} by the boundary between [M^FDI, L] and [M^EX, L, H], and λ^C _{[FDI,L]/[EX,L]} by the boundary between [M^FDI, L] and [M^EX, L].

11. The [M^EX, L, H] – [M^EX, L] boundary is a vertical line given by firm H s zero profit condition () which is independent of γ.