Multi-product exporters: Product churning, uncertainty and export discoveries

Leonardo Iacovone, Beata Javorcik 01 August 2010

a

A

Virtually all successful development stories in the last few decades have been accompanied by massive increases in exports. It is not surprising therefore that the question of "how do firms start exporting?" is one of great interest to policymakers.

Yet answering that question is difficult since exporting is endogenous to several other poorly understood phenomena such as learning about new markets, development of new products, entry of new firms, etc... Quite simply, it's hard to know what is causing what in most situations.

One natural experiment that provides some traction on the reverse causality problem is the formation of the North American Free Trade Agreement (NAFTA) in 1994. This free trade agreement, whose implementation was under question right up to its final passage by the US Congress, resulted in a massive Mexican export boom that most analysts agree was triggered by the agreement.

Decomposing an export boom: Lessons from NAFTA

An examination of the Mexican export boom (a 300% increase in exports between 1993 and 2002) which took place after Mexico joined NAFTA in 1994 suggests that:

  • the Mexican export boom was mostly driven by incumbent exporters expanding foreign sales of their existing export products;
  • the expansion of the range of export products on the part of incumbent exporters accounted for only 10% of the observed export growth;
  • the emergence of new exporters (producers who did not export before the NAFTA period) played a marginal role in the total export growth (see Figure 1).

Figure 1. Decomposition of the Mexican export boom

The importance of uncertainty

The limited role of the latter two channels is consistent with the predictions of theoretical models where uncertainty and information asymmetries shape international trade flows (Rauch and Watson 2003, Albornoz et al. 2008). According to Rauch and Watson (2003), starting small allows importers to test the credibility of their foreign partners or the suitability of their products for the export market. This view is supported by the patterns found in the Mexican data we examine in Iacovone and Javorcik (2010):

  • Mexican producers tend to enter foreign markets with a single product and a very small volume of exports (relative to their domestic sales of the product). Both the number of products and the export volume expand as the producers’ experience in export market increases, which suggests that they are testing the suitability of their products for the foreign market (see Figure 2).

Figure 2. New exporters start small

Note: The graph depicts the value of exports of a product newly introduced in export markets by a Mexican producer relative to the producer’s total sales of the product (expressed in %). Period 0 corresponds to the year when the product is first introduced into export markets.

  • Although not all new relationships start small, most of them do, as indicated the shape of the distribution of export revenues relative to total sales in “new trade relationships.” As seen in Figure 3, the distribution is skewed to the left which is consistent with the existence of uncertainty.

Figure 3. Distribution of export revenues relative to total sales in the first year of a product being exported by its producer

  • The cautious approach with respect to entry into foreign markets is warranted, as about 60% of newly introduced products are withdrawn from export markets within one year. The survival rate increases with the product’s tenure in the export market.

The consequence of uncertainty: Coping strategies

Having provided some evidence on the role of uncertainty in shaping export patterns, we describe coping strategies adopted by exporters.

To limit the uncertainty, most producers enter export markets with products already tested at home – especially exporters breaking into foreign markets for the first time (see Table 1). While on average 68% of new export products were previously sold at home, the corresponding figure for firms entering export markets for the first time is 85%. This pattern is not surprising as new exporters are subject to higher risks, given that they have not yet tested themselves, nor are they familiar with export markets.

Table 1. Firms tend to export product varieties previously sold at home
 Year
Share of new export varieties that were previously sold at home
All firms
New exporters
Existing exporters
 
 
 
 
1995
82%
90%
75%
1996
69%
87%
65%
1997
69%
87%
67%
1998
65%
80%
66%
1999
63%
68%
69%
2000
70%
89%
72%
2001
67%
86%
69%
2002
72%
83%
78%
2003
73%
78%
75%
 
 
 
 
Average 1997-2003
68%
85%
69%

Note: Varieties previously sold at home are defined as those sold on the domestic market at t-1.

Innovation and imitation

The presence of uncertainty facing future exporters manifests itself in other ways. A theoretical model incorporating uncertainty and opportunities for imitation (Hausmann and Rodrik 2003) predicts a limited amount of export discovery, i.e., introducing into export markets products that have not yet been exported from Mexico. The theory predicts that firms potentially capable of producing a certain good will hold off until someone else introduces the product first and, having witnessed their success, they will imitate quickly.

Consistent with these predictions, the Mexican data indicate that export discovery is a relatively rare phenomenon. Most firms introduce products to export markets that others have already been exporting (see Table 2). Furthermore, once a firm has introduced an export product, other firms follow quickly. On average, a discovery is simultaneously introduced by more than one firm (that is, between one and two firms introduce a variety that is an export discovery). For approximately half of the new discoveries, an additional firm starts exporting the same product in the following year. Three years after the discovery the rate of diffusion slows down but continues for at least 8 years, the longest time period considered in our study.

Table 2. New export products and export discoveries

Year
New export varieties
Export discoveries
Discoveries as a share of all new export varieties
1995
1,072
352
33%
1996
999
319
32%
1997
828
253
31%
1998
699
172
25%
1999
481
97
20%
2000
426
100
23%
2001
367
101
28%
2002
309
80
26%
2003
426
113
27%
 
 
 
 
Total
5,607
1,587
28%

Conclusion

Although uncertainty and information asymmetries are not the only factors preventing diversification of exports in developing countries, they do appear to play a significant role, and their existence creates space for policy interventions that lower the costs of obtaining information and support diffusion of knowledge about export markets.

This column presents the views of the authors and should not be understood as necessarily representing the views of their employers.

References

 

Albornoz, F, H Calvo Pardo, G Corcos, and E Ornelas (2008), “Sequential exporting”, Mimeo, London School of Economics.

Hausmann, R and D Rodrik (2003), “Economic development as self-discovery”, Journal of Development Economics.

Iacovone, L and B Javorcik (2010). “Multi-product Exporters: Product Churning, Uncertainty and Export Discoveries”, Economic Journal.

Rauch, J and J Watson (2003), “Starting small in an unfamiliar environment”, International Journal of Industrial Organization.

 

a

A

Topics:  Development International trade

Tags:  exports, NAFTA, Mexico

Senior Economist in the Innovation, Technology and Entrepreneurship Global Practice, Financial and Private Sector Development Network, World Bank

Beata Javorcik

Professor of International Economics, University of Oxford; and Research Affiliate, CEPR