Abstract

The thesis investigates the relationship between firm heterogeneity and international trade from the perspective of a developing country. The standard heterogeneous firm trade models (Melitz, 2003; Bernard et. al, 2003) have largely focused on differences among firms in terms of an exogenously given productivity which would explain why only some firms self-select into international markets while the others serve the domestic market. Empirical evidences, especially from developing countries, note that firms’ selection into export is also the result of conscious investment decisions by firms that aim to improve their productivity and product attributes with explicit purpose of becoming exporters (Lòpez, 2004). In order to further understand the selection mechanism of developing countries’ firms into international markets, this thesis explores the roles of prices, quality and access to finance differences across firms as additional sources of heterogeneity, as well as their interaction. The thesis is composed of three self-standing but related empirical papers that exploit a unique panel data set that come from the annual Ethiopian Large and Medium Scale Manufacturing Enterprise census, and one concluding chapter with policy implications. The census is run by the Central Statistical Agency of Ethiopia (CSA). Unusual in many firm-level surveys, this data set provides plant-product level information on quantity and value of production and sales in both the domestic and foreign markets. Among other information, the census also collects data on the obstacles that firms face in their activities including financial resources. The richness of the data set allows constructing plant-product-level price and quality index, and firm-level access to finance indicators. The first chapter investigates the role of price heterogeneity and demand factors in examining the link between export and productivity. One of the well-known empirical regularities in the literature is that, on average, exporters are more “productive” than non-exporters (ISGEP, 2008; Wagner, 2012). Despite this consensus, what the referred productivity really captures remains blurred: since quantity information is rarely available in most data sets, the standard approach is to proxy quantity by firm-level revenues deflated by industry-level price indices. However, this approach ignores within-sector price heterogeneity and, as a result, it confounds physical efficiency (output per unit input) and demand components into the revenue-based productivity estimates (De Loecker, 2011). This bias would be further exacerbated when firms operate in different markets, and thus have different pricing strategies. To the extent that the demand structure is different in domestic and foreign markets, it is reasonable to expect price differences between exporters and non-exporters, not to mention within-sector variations. Using a rich panel data from Ethiopian manufacturing firms, the first chapter characterizes firms by three distinctive margins that are confounded in revenue productivity: physical productivity, output prices and demand shocks, and examines the relationship between these different sources of competencies of firms, and their separate role in shaping export participation. The main results show that exporters are more productive than non-exporters in both revenue and quantity based productivity measures. However, the productivity gap is larger in revenue productivity. Regarding the decision to export, revenue productivity and price significantly explain probability to export, but physical productivity hardly plays a role. Further evidence shows that price is increasing in revenue productivity and decreasing in physical productivity. On average, exporters charge higher prices than non-exporters. The overall results suggest that revenue productivity overstate the relationship between export and productivity because exporters have favourable demand condition that allows them to charge higher prices than non-exporters. This finding provides a new insight in understanding how export contributes to aggregate productivity growth. To the extent that exporters are more productive than non-exporters, the findings about prices are at odds with the standard firm heterogeneity model of Melitz (2003), where more productive firms have lower marginal costs and thus charge lower prices. The second chapter addresses this puzzle, further investigating the determinants of firms’ decision to export. To this end, I refer to the analytical framework that extends the standard firm heterogeneity model by introducing quality (e.g., Hallak and Sivadsan, 2013). Recent empirical findings also confirm that exported products feature higher prices than domestic products, and this price difference is ascribed to quality differences (for example see, Iacovone and Javorcik, 2012). Nevertheless, even if prices do correlate with quality, a major limitation of using prices as a proxy for quality lies in the inability to distinguish between quality and cost factors. To address this issue, I follow the recent empirical strategy proposed by Khandelwal (2010) to estimate quality, thus relaxing the assumption product quality is fully captured by its price. The results show that high-price products are more likely to be exported. However, once price is adjusted for quality difference, products with higher quality-adjusted price are less likely to enter into foreign markets. Jointly these results suggest that the observed price-export relationship reflects quality differences. Furthermore, I find that quality is the most important factor in determining firm export decision; and the effect of firm efficiency on export mainly operates through the quality channel. Based on an analysis of the dynamics of quality and product entry, I find that high-quality products self-select into export. Specifically, the trajectories of exported products show that quality upgrading took place three years prior to export entry. In the run-up phases of export entry, firms also change the composition of their production in favor of future exported varieties. The third chapter, is the result of a joint work with my supervisor Stefano Schiavo, and investigates the mechanisms through which access to bank credit affects firms export. In particular it examines the interplay between financial constraints and product quality in explaining firms’ export behavior using information on a panel of Ethiopian manufacturing firms. Similar to many recent studies, the previous chapter suggests that quality matters a lot for export and, moreover, that firms need to make conscious investment decisions aimed at upgrading their product quality before entering foreign markets. The implication is that, in addition to its direct effect on firms’ ability to pay upfront entry costs, access to finance may affect export decisions through its effect on investment. Since firms in developing countries have typically limited internal revenue and operate in underdeveloped financial markets, financial resources are especially important in shaping the decision to export. The main results confirm the presence of substantial sunk costs associated with exporting. Despite this, bank finance does not appear to have a strong direct effect on export participation. On the other hand, both present and past product quality is robustly associated with export status, and quality upgrading requires substantial investment. Therefore, bank credit is relevant for export only insofar as it is channeled to the fixed investments required to enhance quality. An important implication of this chapter is that improving financial conditions and access to bank credit can help firms to move from low- to high-quality products, enhance their ability to access foreign markets and therefore improve the overall export performance of the economy. The common message of all the three chapters is that the success of Ethiopian firms in international market is mainly driven by demand factors in which only firms that able to attract demand for their products succeed in foreign markets. However, despite the presumed relevance of firm productivity efficiency to drive export, there is no strong evidence that this apply for Ethiopian firms. Further analyses of the demand factors unveil the crucial role of product quality upgrading in determining firms’ entry into export markets. These findings sheds some light on the sources of the productivity difference between exporters and non-exporters that have been found in the literature, including studies focusing on African firms (for example, Van Biesebroeck 2005, and Bigsten and Gebreeyesus 2009). Furthermore, it confirms the general importance of satisfying foreign market quality standards for firms in developing countries to succeed in international markets (Chen et.al., 2008). The evidence suggests that export promoting policies that exclusively focus on achieving quantitative targets, such as productivity, as a means to increase competitiveness in international markets should be revisited: especially for developing countries, a policy shift from quality to quantity is a right direction to go forward.