When Harry meets Sally: The buyer margins of firms’ exports

Jerónimo Carballo, Gianmarco I.P. Ottaviano, Christian Volpe Martincus, 11 September 2013



Locating a buyer is one of the most important barriers for firms trying to penetrate new markets (e.g., Kneller and Pisu 2011). Investment in gathering these data may be socially suboptimal because information can spill over to other companies. Unsurprisingly, governments around the world implement programmes to help firms find buyers. For example, the US Department of Commerce assists exporters in identifying and qualifying lead for potential buyers, both through general online matchmaking mechanisms and customised services (e.g., Gold Key Matching Service). The number of buyers is even used as a metric according to which the performance of some international organisations is assessed. For instance, one of the indicators of achievement for the International Trade Center’s goal, ‘strengthen[ing] international competitiveness of enterprises through ITC training and support” is “increas[ing the] number of enterprises having met potential buyers and, as a result, having transacted business” (ITC 2011).

Both exporters and policymakers see buyers as lead actors in international trade and specifically as a key margin for export expansion. However, buyers have played almost no role in the empirical analyses of firms’ exports and their determinants.1 In particular, there is still very little evidence on the number of actual partners for exporting firms across products and destination countries as well as on the distribution of firm-level sales and prices across partners. The reason is simple: typically data that simultaneously identify buyers and sellers in international transactions are hardly available.

The number of buyers matters for aggregate exports

Recently, the veil of ignorance has started to be raised in the case of a limited number of (small and mostly Latin American) countries (Blum et al. 2010, 2012; Eaton at al. 2013; Bernard et al. 2013; Carballo et al. 2013). In our own work, we use highly disaggregated annual firms’ export data virtually covering the entire population of exporters from three countries – Costa Rica, Ecuador and Uruguay – over the period 2005-08 to provide a precise characterisation of countries’ and firms’ export margins. More precisely, these data make it possible to uncover the fact that expansions in countries’ foreign sales to given markets can be traced back not only to increases in the number of exporting firms (firm extensive margin), the number of exported products (product extensive margin) or average exports of already existing flows (intensive margin), but also in the number of buyers (buyer extensive margin).

The decompositions of exports in their different margins suggest that reaching a larger number of buyers is a relevant channel of export expansion, at the country level as well as at the firm level, both across destinations and within destinations over time. This buyer extensive margin is at least as important as the firm and the product extensive margins for aggregate bilateral exports as well as the firms’ product extensive margin for firms’ destination-specific exports. This implies that, due to data limitations, the so far typical decomposition of exports tends to overestimate the importance of the intensive margin and misrepresent that of the firm and product extensive margins. This is shown in Figure 1 where black and grey bars respectively show the share of aggregate exports accounted for by the different margins with and without attention paid to the buyer extensive margin.

Figure 1. Decomposition of aggregate bilateral exports in their margins

Source: Carballo et al. (2013).

The number of buyers and the shares of the main buyers depend on the characteristics of destination markets

How do the number of buyers and their purchases vary across foreign markets? The buyer extensive margin behaves much like the other traditional extensive margins: it decreases with the distance to and increases with the size and the per capita income of the destination markets (Figure 2). The actual intensive margin, which in this case incorporates the buyer dimension (i.e., sales of specific products to already served buyers in given destinations), reacts qualitatively in the same way to distance and market size, albeit it appears to be less responsive. Moreover, it does not seem to be systematically related to the destination’s income per capita.

Figure 2. Responses of aggregate bilateral exports and their margins to trade enhancers/barriers

Source: Carballo et al. (2013). Note: Based on a gravity regression also including common language, colonial relationships, and regional trade agreements among the explanatory variables. The estimated coefficients are not reported.

Most exporters serve few foreign buyers, whereas few exporters serve several foreign buyers. Thus, half of the exporters have no more than two buyers, but the top 10% (5%) interact with more than 11 (20) buyers. Consistently, the main buyer accounts for a large portion or directly all foreign sales of a relatively large number of exporting firms. In contrast, for the top 10% (5%) exporters in terms of the number of buyers, the average share of the main buyer does not exceed 40% (38%) and is as low as 32% (25%) in the case of Uruguay.

Disaggregated data specifically indicate that most exporters sell abroad only a few products to a few buyers in a few destinations. The other side of the coin is that the few firms that export several products to several destinations and, on top, to several buyers account for large shares of total exports. Noteworthy, this is also true relative to firms that also sell several products in several destinations but only to a few buyers.

As expected, firms exhibit significant differences in the distribution of their sales across buyers, also at different levels of aggregation. In this sense, the distribution of firm-product-destination sales across buyers is particularly skewed and responds to the characteristics of the destination markets: skewness increases with the accessibility and the size of the destination and the toughness of competition in this market (see Figure 3).

Figure 3. Response of concentration of sales across buyers to trade enhancers/barriers and toughness of competition

Source: Carballo et al. (2013). Note: Based on gravity regressions. Specification 1 also includes common language, colonial relationships, and regional trade agreements among the explanatory variables. The estimated coefficients are not reported.

Also the price dispersion across buyers depends on the characteristics of destination markets

A series of empirical studies indicate that prices of given products, even from the same firm, differ across markets (e.g., Hummels and Klenow 2005, Hallak 2006, Manova and Zhang 2012). Using our data it is possible to go one step further and examine the distribution of prices across buyers within specific destinations. Even in the same destination different buyers pay different prices for the same product sold by the same firm, that is, there is price dispersion at the firm-product-destination level, and these price differences are sizeable for differentiated products. In particular, buyers that account for larger shares of a firm’s sales of a given product tend to pay lower prices for that product. This price dispersion decreases with distance to the destination market and increases with the size of the latter as well as with the intensity of competition therein (Figure 4).

Figure 4. Response of price dispersion across buyers to trade enhancers/barriers and toughness of competition

Source: Carballo et al. (2013). Note: Based on gravity regressions. Specification 1 also includes common language, colonial relationships, and regional trade agreements among the explanatory variables. The estimated coefficients are not reported.

Adjustments along the buyer margin can be a channel of trade-related welfare gains

Incorporating the buyer dimension provides a more complete empirical picture and also hints to potential new channels of welfare gains. In Carballo et al. (2013), we develop a simple model of selection with heterogeneous sellers and buyers consistent with the above findings in which the reduction of various types of trade barriers (including time lost in transit) induces a better alignment between consumers' ideal products and firms' core competencies due to enhanced competition. This leads to higher productivity and higher welfare.

From better knowledge to more effective policies

The main message that comes out from the analysis of the exporter-destination-product-importer level data is that the buyer margins can be a critical channel through which policies influence aggregate exports. Explicitly incorporating these margins into their assessment can therefore help us improve our understanding of how these policies work.

As suggested at the beginning of this note, this clearly concerns export promotion, but the same holds for a range of other policies relevant for international trade. Take, for example the case of customs. Volpe Martincus et al. (2013) show that delays associated with customs procedures have a significant negative effect on firms’ exports, which can be traced back to their negative impact on both the number of buyers and the average export values and quantities per buyer. Furthermore, effects seem to be more pronounced on sales to newer buyers who are most likely to be imperfectly informed of the reasons of such delays. Reducing delays can therefore act as a powerful trade promotion policy beyond obvious time saving.

Disclaimer: The views expressed here are those of the authors and do not necessarily represent those of the institutions with which they are affiliated.


Bernard, A, Moxnes, A and Ulltveit-Moe, K (2013), “Two-sided heterogeneity and trade”, Dartmouth College, mimeo.

Blum, B, Claro, C, and Horstmann, I (2010), “Facts and figures on intermediated trade”, The American Economic Review 100.

Blum, B, Claro, C, and Horstmann, I (2010), “Import intermediaries and trade: Theory and evidence”, University of Toronto, mimeo.

Carballo, J, Ottaviano, G, and Volpe Martincus, C (2013), “The buyer margins of firms’ exports”, CEPR Discussion Paper 9584.

Eaton, J, Eslava, M, Jinkins, D, Krizan, D, and Tybout, J (2013), “A search and learning model of export dynamics”, Penn State University, mimeo.

Egan, M and Moody, A (1992), “Buyer-seller links in export development”, World Development, 20(3).

Hallak, J (2006), “Product quality and the direction of trade”, Journal of International Economics 68(1).

Hummels, D and Klenow, P (2005), “The variety and quality of a nation’s exports”, The American Economic Review 95.

ITC (International Trade Center) (2011), “Strategic Plan 2012-2015”.

Kneller, R and Pisu, M (2011), “Barriers to exporting: What are they and who do they matter to?”, World Economy, 34(6).

Manova, K and Zhang, Z (2012), “Export prices across firms and destinations”, Quarterly Journal of Economics 127.

Volpe Martincus, C, Carballo, J, and Graziano, A (2013), “Customs as doorkeepers: What are their effects on international trade?”, Inter-American Development Bank, mimeo.

1 Egan and Mody (1992) focus on three sectorial case studies to examine the factors affecting the formation, maintenance, and implications of commercial relationships between developed countries’ buyers and developing countries’ exporters.

Topics: International trade
Tags: export buyers, exports

PhD student at the University of Maryland

Gianmarco I.P. Ottaviano

Professor of Economics, LSE; Non-Resident Senior Fellow, Bruegel; and CEPR Research Fellow

Lead Economist at the Integration and Trade Sector, Inter-American Development Bank