The Buyer Margins of Firms' Exports

Abstract

We use detailed data on exporters from Costa Rica, Ecuador and Uruguay as well as on their buyers to show that aggregate exports are disproportionally driven by few multi-buyers exporters whose foreign sales of any product are in turn accounted for by few dominant buyers. We propose an analytically solvable multi-country model of endogenous selection in which dominant exporters, dominant products and dominant buyers emerge in parallel as multi-product sellers with heterogeneous technologies compete for buyers with heterogeneous needs. The model not only provides an explanation of the existence of dominant buyers but also makes specific predictions on how the relative importance of dominant buyers should vary across export destinations depending on their market size and accessibility. We show that these predictions are borne out by our data and discuss their welfare implications in terms of gains from trade.

Relational contracts and supplier turnover in the global economy

“…Moreover, cost dispersion is the theoretically ideal instrumental variable given the structure of our model, because the exclusion restriction is satisfied. 16 In the second stage, we include 15 The identifying assumption is that all parameters except the search costs F are industry-specific and common to all individual matches i within industry j and abstract from all cross-industry effects. This pins down the critical search cost levels F 1 j and F 2 j , and implies that all firms and suppliers in j encounter the same constellation as shown in Figure 2 above.…”

Institutions and export dynamics

“…Thus, in the first stages of a partnership relatively low quantities are exported; if the distributor proves to be reliable, the producer progressively improves his foreign sales. 21 In the limiting case where this process 21 Our model delivers this point too strongly, making no concession for sales to come down except when the producer exits Foreign's market. It also yields the counterfactual prediction that producers exit the market at their pinnacle (we thank Costas Arkolakis for pointing this out).…”

“…We do so by assuming that in each period while the partnership is active, the distributor privately finds an opportunity to default on the contract without incurring costs 12 Interestingly, Carballo et al (2013) show that, for Costa Rica, Ecuador and Uruguay, for which they have nearly the universe of transactions for 2005-2008, the median exporter has indeed a single buyer in each foreign destination it serves. However, the largest exporters deal with multiple buyers per destination.…”

“…26 A simple linear panel regression with firm-time fixed effects does the same for (21). 27 Finally, we report two-way clustered (country and firm) standard errors.…”

Customs

“…There are, however, reasons to believe that these effects may differ among groups of firms, products, and destinations, in which case such a restriction would not hold. Thus, for instance, impacts can be larger for time-sensitive products (e.g., Djankov et al, 2010) or in destinations with tougher competition (e.g., Mayer et al, 2013;and Carballo et al, 2013). Hence, we also generalize this equation to explore the existence of heterogeneous effects across those groups as follows:…”