The Internationalization Process of Firms: from Exports to FDI?

Transcription

1 The Internationalization Process of Firms: from Exports to FDI? Paola Conconi Université Libre de Bruxelles (ECARES) and CEPR André Sapir Université Libre de Bruxelles (ECARES) and CEPR Maurizio Zanardi Université Libre de Bruxelles (ECARES)

2 Introduction In recent decades, more and more companies have started to operate internationally, either by exporting their products to foreign customers or by engaging in horizontal foreign direct investment (FDI). Vast literature in international business from Johanson and Vahlne (1977): Market-specific knowledge (e.g., cultural patterns, consumers tastes, local regulations) can only be gained through experience. Firms follow a process of gradual involvement in foreign markets, first exporting and eventually, in some cases, establishing foreign subsidiaries.

3 Our paper Simple theoretical model in which firms enter the foreign markets via exports and switch to FDI if they discover that they are profitable enough. Empirical analysis of firms decisions of how to serve foreign markets over time, using data from NBB covering exports and FDI in individual destinations for all companies registered in Belgium during

4 Outline of the presentation Related literature The model Datasets and variables Empirical analysis Conclusions

5 Related literature Since Johanson and Vahlne (1977), international business studies argue Market-specific knowledge (e.g., cultural patterns, consumers tastes, business regulations) can only be gained through experience. To acquire such knowledge, firms serve first foreign markets via exports; eventually, in some cases, they establish foreign subsidiaries. We develop a simple theoretical model to formalize these ideas and provide systematic empirical evidence of firms internationalization process.

6 Related literature The literature on the proximity-concentration tradeoff examines firms choice to serve a foreign market via exports or FDI. Markusen (1984) Brainard (1997) Markusen and Venables (2000) Helpman, Melitz and Yeaple (2004) Ramondo, Rappoport and Ruhl (2010) Rob and Vettas (2003) Our paper: focus on the dynamics of firms internationalization choices, showing that uncertainty can lead them to switch from exports to FDI.

7 Related literature (cont.) The literature on real options shows that, if investments are irreversible and market conditions are uncertain, firms may prefer to minimize current investments but secure an option to invest later (e.g., McDonald and Siegel, 1986; Dixit and Pindyck,1994, Guiso and Parigi, 1999). Our paper shows that, when faced with the choice on how to serve foreign markets, firms may optimally choose to export first, waiting to invest until they learn about foreign demand and supply conditions.

8 Related literature (cont.) An increasingly vast literature on firms export dynamics emphasizes the role of uncertainty in export decisions: Rauch and Watson (2003) Eaton, Eslava, Kugler, and Tybout (2008) Ruhl and Willis (2008) Lawless (2009) Aeberhardt, Buono, and Fadinger (2009) Eaton, Eslava, Krizan, Kugler, and Tybout (2010) Freund and Pierola (2010) Albornoz, Calvo-Pardo, Corcos, and Ornelas (2012) Our paper: dynamics of firms exports and FDI choices in foreign markets.

9 Model setup To focus on the role of uncertainty in foreign markets, we consider a representative domestic firm, abstracting from firm heterogeneity. Variable costs: - unit cost of production (normalized to zero) - unit cost c ik of selling good k in country i (foreign distribution costs) - unit trade costs τ ik (transport costs and barriers to trade) Fixed costs: - of exports F ik E (e.g., learning about customs procedures) - of FDI F ik I (e.g., setting up a foreign production plant), with F ik I > F ik E Demand in the foreign market: q ik (p ik ) = a ik - p ik

10 Model setup (cont.) Ex-ante, the firm does not know the distribution costs in the foreign market (c ik ) and foreign consumers willingness to pay for its product (a ik ). Uncertainty about foreign market profitability (Albornoz et al, 2012): where µ ik is a random variable with continuous cumulative distribution function G(.) on the support

11 Timing t = 1: the firm chooses between exports, FDI or not entering the foreign market. If it enters via exports (FDI), it pays F E (F I ) and chooses how much to sell. At the end of the period, if the firm has sold a positive amount, it infers its profitability. t = 2: if the firm has not entered at t = 1, it decides whether or not to do so. If the firm has entered, it decides whether to exit, serve the foreign market under the same mode, or switch mode.

12 Firm s strategies Entry via exports: at t = 1, the firm pays the fixed cost F E, exports to the foreign market and discovers its profitability; at t = 2, it decides whether to continue serving the foreign market through exports, switch to FDI, or exit. Entry via FDI: at t = 1, the firm pays the fixed cost F I, sets up a foreign subsidiary and discovers its profitability; at t = 2, it decides whether to continue serving the foreign market through FDI, switch to exports, or exit. No entry in the foreign market at t = 1.

13 Period t = 2, following entry via exports at t = 1 Second-period export profits: Second-period FDI profits:

14 Export and FDI profits at t = 2, following entry via exports at t = 1

15 Model setup (cont.) Assumption 1: the fixed costs of setting up a foreign subsidiary are large enough compared to trade costs that FDI does not always dominate exports:

16 Period t = 2 Entry via exports at t = 1: Second-period export profits: Second-period FDI profits:

17 Export and FDI profits at t = 2 following export entry at t = 1

18 Result 1: After entering the foreign market via exports and discovering its profitability µ, the firm will exit if µ < τ, will continue serving the foreign market via exports if τ < µ < µ EI and will switch to FDI if µ >µ EI.

19 Period t = 2 Entry via FDI at t = 1 Second-period FDI profits: Second-period export profits: A firm entering the foreign market via FDI at t = 1 will never switch to exports or exit the market at t = 2.

20 Export and FDI profits at t = 2 following FDI entry at t = 1

21 Entry via exports at t = 1 Period t = 1 Ex-ante expected profits from entering the foreign market via exports: Entry via FDI at t = 1 Ex-ante expected profits from entering the foreign market via FDI:

22 Result 2: The first-period entry decision depends on expected protability in the foreign, Eµ. If Eµ µ E, the firm may experiment, exporting an arbitrarily small amount at a loss to discover whether serving the foreign market is actually profitable.

23

24 Policy implications Uncertainty give rise to dynamic complementarity between exports and FDI. This can explain some trends in FDI: Studies on the determinants of FDI fail to find a significant effect of trade protection (e.g. Grubert and Mutti, 1991; Blonigen, 1997). Notwithstanding major reductions in trade barriers and transport costs in recent decades, foreign affiliates sales have not fallen, but have actually expanded faster than exports of goods and services (Helpman, 2006). In contrast to static models of the proximity-concentration tradeoff, Trade liberalization may actually foster FDI, by lowering the cost of export experimentation FDI liberalization can lead some firms to start exporting, by increasing the option value of export entry

25 Empirical predictions Uncertainty creates an option value of entry the option value of export entry: firms are willing the pay the fixed cost associated with export entry to find out whether serving the foreign market is actually worthwhile. Prediction 1: Uncertainty leads firms to follow a gradual internationalization process, testing a foreign market via exports, before engaging in FDI. Export experimentation should also lead new exporters to exit the foreign market, if they discover that their profitability is too low to justify the trade costs. start by exporting small amounts; following an initial trial period, conditional on surviving as exporters, their export volumes should expand.

26 Empirical predictions (cont.) In our model, all uncertainty is resolved as soon as the firm starts serving the foreign market: export experimentation occurs only during the entry period after which, if realized profitability is high enough, the firm switches to FDI. The model can be easily extended to multiple periods, to allow for information acquisition to take time. In this setting, there will be an option value of waiting to invest (e.g. McDonald and Siegel, 1986; Dixit and Pindyck,1994): new exporters will prefer to continue serving the foreign market via exports until they are certain that it is worthwhile to establish foreign subsidiaries. Prediction 2: uncertainty delays FDI entry decisions by new exporters.

27 Export-supporting FDI The logic of our model can be applied to a setting in which exporting firms must choose between using a foreign distributor (higher variable costs) and setting up its own distribution network (higher fixed costs). Uncertainty can lead to a process of gradual involvement in foreign markets: a firm may initially test a foreign market by exporting a small amount and using local distributors; if it discovers that it is profitable enough, it will invest in setting up its own distribution network. Main differences between export supporting and horizontal FDI: Exports increase following FDI Higher trade barriers decrease the likelihood that a firm invests abroad

28 Datasets and variables Unique data from the National Bank of Belgium (NBB), allowing us to study the dynamics of firm-level exports and FDI in individual foreign markets, for all companies registered in Belgium (information from different firm-level datasets can be linked through a unique identification number). Data on exports from NBB Foreign Trade dataset, covering firm-level trade transactions since Data on FDI from the NBB Survey on Foreign Direct, covering all companies holding at least 10 percent of the social capital of foreign firms, providing information on their investment stocks and flows in individual foreign markets since 1997.

29 Main sample Destinations Countries outside the EU Single Market (SM) Belgian firms are more likely to face uncertain market conditions No issue of reporting thresholds in exports to non-sm countries Members of the WTO Firms Manufacturing industries At least 5 employees Exporting and investing in at least one country outside SM during

30 Descriptive statistics Sorting patterns, already highlighted in the literature (e.g. Helpman et al, 2004):

31 Export and FDI entry Export entry fit is equal to 1 if firm f starts exporting to country i in year t, not having exported to that country in any of the previous 5 years. Advantages of using this stringent definition of export entry: Very few re-entries (only 3.4 percent of export entries) Focus on firms that start to learn about conditions in a foreign market No left-censoring (export data since 1993, 5 years before first FDI entry) FDI entry fit is equal to 1 if firm f starts investing to country i in year t, not having invested in that country in the previous year.

32 Uncertainty measures Export experience fit measures ln (1 + number of years of positive exports) of firm f in country i following export entry. The higher is export experience, the more information the firm should have acquired in the foreign market. BIT it is equal to 1 if country i has implemented a bilateral investment treaty with Belgium at time t. Following the implementation of these treaties, Belgian new exporters should thus face less uncertainty in the destination market. Export exit i measures the average exit rate of Belgian new exporters in country i (in the year following export entry). It provides a direct measure of the exten of uncertainty and experimentation in a foreign market. Variance regulation i is the variance of the World Bank's regulatory index (over the period), which captures how unpredictable business conditions are.

33 An example Exports of firm f to market i (1 = positive exports): FDI of firm f in market i (1 = positive FDI): Export entry: 1998 FDI entry: 2003 Export experience1 at time of FDI entry: 4 years

34 FDI entry is almost always preceded by export entry Table 1 provides support for the first empirical prediction of our model.

35 Characteristics of new exporters Most firms do not survive in new destination (exit rate of 47% at t +1), but survival probability increases over time.

36 Characteristics of new exporters (cont.) Surviving firms expand their export volumes in the years following entry.

37 Uncertainty and FDI entry decisions of new exporters Empirical studies of real option theories have used survival analysis (also called duration analysis) to verify whether uncertainty delays investment decisions (e.g. Hurn and Wright, 1994; Favero et al.,1994; Kogut and Chang, 1996). We use survival analysis the second empirical prediction of our model, according to which uncertainty should delay FDI entry decisions of new exporters. We use a Cox proportional hazard model to estimate the hazard rate h fit (t), the probability that new exporter f starts investing in country i at time t: h fit (t) = h 0 (t) exp (β X fit ) h 0 (t): baseline hazard rate X ijt : explanatory regressors β: vector of coefficients to be estimated

38

39

40

41

42

43

44 Uncertainty and FDI entry decisions of new exporters Robustness checks: Weibull model (specific functional form for baseline hazard rate) Excluding re-entries (first entry spells only)

45

46 Uncertainty and FDI entry decisions of new exporters Robustness checks: Weibull model (specific functional form for baseline hazard rate) Excluding re-entries (first entry spells only)

47 Uncertainty and FDI entry decisions of new exporters Robustness checks: Weibull model (specific functional form for baseline hazard rate) Excluding re-entries (first entry spells only) Excluding vertical FDI entries Including countries that have acceded to SM (e.g. Bulgaria, Romania) Falsification exercise: old members of SM (e.g. France, Germany)

48

49 Conclusions Firms export and FDI decisions must be understood as part of a broader dynamic strategy to serve foreign markets in the face of uncertainty. Although exports and horizontal FDI are substitutes from a static perspective since hey represent alternative ways of serving a foreign market they may be complements over time since the knowledge acquired through export experience can lead firms to invest abroad.