Elsevier

Economics Letters

Volume 101, Issue 2, November 2008, Pages 110-112
Economics Letters

Do sunk exporting costs differ among markets? Evidence from Spanish manufacturing firms

https://doi.org/10.1016/j.econlet.2008.06.012Get rights and content

Abstract

We test whether sunk exporting costs differ among markets. Our results confirm that during 1990–2002, sunk exporting costs were relevant for Spanish firms, and differed depending on the destination market. Besides, the costs to enter/re-enter were higher in developed markets.

Introduction

Firms that enter an international market have to adapt their products to foreign demand, technical and administrative standards and find distribution networks. Imperfect information and informal or formal barriers that separate domestic and foreign markets are supposed to explain the existence of sunk exporting costs. Literature on border effect and gravity equation gives an idea of the importance of these hidden barriers, often seen as the best candidates to explain the “mystery of the missing trade”1 (Trefler, 1995).

Our contribution is twofold: first, we test whether sunk costs differ from one market to another. Secondly, we test whether the experience in one market has the same influence as the experience in other markets on the probability to export. We use a panel data probit model to study the export behaviour of a sample of Spanish manufacturing firms continuously operating from 1990 to 2002.

The existence of sunk costs may explain the fact that exporters are generally more productive than non-exporters. Melitz (2003) provides a theoretical justification for this self-selection effect. Roberts and Tybout (1997) derive an empirical model, using data from Colombia, and find evidence in favour of the hypothesis of sunk exporting costs: a firm incurs in a cost for entering (re-entering) a market, then it exports only if the expected gross profit is positive and its current export status depends on its past experience. Bernard and Jensen (2004) obtain similar conclusions with a panel of manufacturing companies from the United States for the period 1984–1992. Campa (2004) uses a sample of Spanish manufacturing firms and finds sunk costs hysteresis to be an important determinant of export market participation.

Máñez et al. (2008) find that sunk costs, labour productivity, size of the firms, R&D intensity, unobserved characteristics of products and correlations in exogenous shocks influence the firms' participation in export markets. Esteve et al. (2004) find that survival rate at exporting is positively correlated with the export intensity. Furthermore, firms exporting to closest markets export during a longer period. Barrios et al. (2003) show that R&D activities exert a determinant effect on the exporting decision and on the intensity, for national and foreign firms, moreover when exporting to EU and OECD. Finally, Fariñas and Martin-Marcos (2007) show that exporters exhibit a greater economic performance and that they self-select. In particular, firms selling a higher share of their exports in OECD markets have a higher productivity than firms mainly exporting to the Rest of the World (ROW). None of these micro-level studies have tested the hypothesis of sunk costs by destination market, as we propose in this paper.

This paper is organised as follows. In Section 2, we present the empirical model. In Section 3, we describe the data set. Our findings are explained in Section 4. Section 5 concludes.

Section snippets

Empirical model

We closely follow Roberts and Tybout's (1997) approach to model a multi-period export decision for the entry and exit with sunk costs. They consider that in each period, a firm decides to export if the increment to the expected gross profits associated with exporting is positive. Following related literature, a reduced form of the dynamic model is estimated. We assume that the expected gross profits depend on exogenous firm's characteristics (Xit), macro conditions (µt) and past exports. Let us

Data

The Encuesta sobre Estrategias Empresariales (ESEE) is an annual representative survey of Spanish manufacturing firms classified by industrial sector3 and size categories.4 The ESEE includes variables relative to the structural characteristics of the

Econometric results

We estimate the model considering the export to four sets of countries: ALL, EU, ROW, and OECD. We study the impact of the past experience in each of these markets on the current status.6 Results are reported in Table 2. Dummies for years and industries are included in the estimations, but coefficients are

Conclusions

Summarising, this paper shows that the costs to enter (and “to re-enter”) are higher in developed markets than in those of the ROW. Moreover, we find that a previous experience in the EU market (respectively the ROW) increases both the probability of exporting to the OECD and to the same market but is less relevant for the ROW (respectively for the EU). OECD countries (non-EU) appear as an intermediate case. European norms may be very specific and homogeneous among members, whereas

References (9)

  • CampaJ.M.

    Exchange rates and trade: how important is hysteresis in trade?

    European Economic Review

    (2004)
  • BarriosS. et al.

    Explaining firms' export behaviour: R&D, spillovers and the destination market

    Oxford Bulletin of Economics and Statistics

    (2003)
  • BernardA.B. et al.

    Why some firms export?

    The Review of Economics and Statistics

    (2004)
  • EsteveS. et al.

    The determinants of survival of Spanish manufacturing firms

    Review of Industrial Organization

    (2004)
There are more references available in the full text version of this article.

Cited by (21)

  • The complementarity effect of exporting, importing and R&D on the productivity of Ukrainian MNEs

    2023, International Business Review
    Citation Excerpt :

    Moreover, due to the multicotomous nature of the selection variable (i.e., firms can self-select into all possible combinations of export and import activities), the first stage of the analysis uses a multinomial logit estimator that includes lagged labor productivity, firm size (as number of full-time registered employees), and dummies indicating the possession of intangible assets and foreign ownership as explanatory variables. We also include lagged export and import status of the firm to take into account possible persistence in trade participation (Blanes-Cristóbal, Dovis, Milgram-Baleix, & Moro-Egido, 2008), and lagged annual 2-digit industry sales, export and import intensity to take into account domestic demand conditions and knowledge spillovers that might affect firms’ decisions regarding international trade participation. We use lagged export and import intensity of a given industry/year as exclusion restrictions, required to identify the sample-selection model.

  • Foreign market re-entry: A review and future research directions

    2021, Journal of International Management
    Citation Excerpt :

    For instance, firms which previously had operations in both developed and emerging markets, could shift their geographical focus by favoring opportunities in emerging markets (Cairns et al., 2010; Donzé, 2015; Lee et al., 2014; Nummela et al., 2016). In light of the sunk cost in different markets, the probability of re-entry also varies depending on different destinations (Blanes-Cristóbal et al., 2008). In terms of product scope, studies have found that firms tend to re-enter with larger scopes when they aim to increase their market share, achieve further sales growth (Bala and Subramanium, 1996; Zhang and Larimo, 2013), have extensive knowledge (Javalgi et al., 2011) or undertake innovation or R&D during time-out periods (Zhang and Larimo, 2013).

  • Export wage premium for south Brazilian firms: Interaction between export, human capital, and export destination

    2020, EconomiA
    Citation Excerpt :

    However, differences in parallel trends may be greater when we compare firms that only export to developed countries and never exporters (especially, among the smaller firms). In fact, Blanes-Cristóbal et al. (2008) showed that export self-selection effect is higher when destination of exports are developed markets. Therefore, our results must be analyzed with care and we argue that it is not possible to reach solid conclusions about the causal effects of export to developed regions on firms performance.

  • How micro, small and medium-sized enterprises are driven outward the superior international trade performance? A multidimensional study on Italian food sector

    2018, Research in International Business and Finance
    Citation Excerpt :

    All of these costs could be sunk in property. Afterwards, the essence of sunk cost not only makes the export decision dynamic (Alessandria and Choi, 2014b), but also may elucidate the fact that exporter firms are more productive than non-exporters (Das et al., 2007; Blanes-Cristóbal et al., 2008; Alessandria and Choi, 2014a), resulting on internationalization behavior of firms (Javalgi et al., 2011; Yi and Wang, 2012). Nevertheless, high sunk costs related to export activities result to more persistent of export behavior (Bernard and Jensen, 2004).

  • Global engagement and the occupational structure of firms

    2017, European Economic Review
    Citation Excerpt :

    To provide more direct evidence for this mechanism, we now study the variation in skill mix across firms that export to different destination markets. The literature has pointed out that different export markets imply different costs of entry (Blanes-Cristóbal et al., 2008; Arkolakis 2010; Gullstrand 2011). Therefore, in Table 3 we examine the extent to which the skill mix varies between firms that export to different markets.

  • Inspection, testing errors and trade in tainted products

    2015, Journal of the Japanese and International Economies
View all citing articles on Scopus

The authors gratefully acknowledge the financial support from the CICYT Projects SEJ2005-001163 and SEJ2006–11067 (Spanish Ministry of Education and FEDER), Grupos de investigación (SEJ 340 y SEJ 246), Proyectos de Excelencia (SEJ 01252 y SEJ 03261) and Fundación Centro de Estudios Andaluces (ECOD.105/034).

View full text