Chapter 5. Determinants of Export Diversification Using Disaggregate Level Data
5.2 Literature Review
Estimation of international trade flows is nothing new in the international economics research.
The pioneering work of Tinbergen in 1962 used gravity equations in empirical specifications of bilateral trade flows in which the volume of trade between two countries is proportional to the product of an index of their economic size, and the factor of proportionality depends on measures of trade resistance between them [Helpman, Melitz & Rubinstein, 2008]. The trade resistance measures include geographic distance, dummy for common borders, and dummies for Commonwealth and Benelux memberships, used by Tinbergen. His specification has been broadly used because it provides a good fit for most data sets of regional and international trade flows.
Over time, Tinbergen s approach has been accompanied by researchers with theoretical reinforcements and better estimation techniques. Moreover, it has dominated empirical international trade research and has been used to estimate the impact on trade flows of international borders, preferential trade blocs, currency unions, and membership in the WTO, along with the size of
home-market effects. Most of the studies estimate the gravity equation on large group of countries and on only positive trade flows. This study argues that, by disregarding the zero export flows, specifically for China, Japan and Korea, previous studies disregard important information contained in the data set.
Thus the existing studies suggest incomplete estimates of export diversification for these countries.
To correct these biases, this study follows the new-new trade theory proposed by Melitz (2003) that predicts positive as well as zero export flows between countries. Thus this study exploits the information contained in the data sets of Chinese, Japanese and Korean economies with the help of new-new trade theory.
According to Roberts and Tybout (1997), the start-up costs of becoming an exporter includes market research, product development, distribution, and learning. The World Bank and the Colombian government's export promotion agency PROEXPO (First Washington Associates 1991) interviewed a sample of 186 Colombian firms in 1990, to reveal managerial thinking on the decision to export.
Their views on the startup costs of breaking into foreign markets explain those start-up costs of becoming an exporter. The relevant costs are explained in detail, in the following paragraphs.
One of the major start-up costs is market research. The types of market research considered being most important among exporters and non-exporters for entry were: buyer identification and contact, foreign prices, market selection, and standards and testing requirements. Not only exporters but also non-exporters felt that legal advice and assistance were important. To overcome information obstacles, most firms envisioned using, or had actually used external services, both private (with payment) and public. Those outside assistance came from broker and distributors, chambers of commerce, associations of suppliers, and trading companies. Interestingly, "especially in the areas of foreign market selection, buyer identification and contact, as well as standards and testing requirements" many firms did their own research.
Another important cost includes product development. It was found that among firms that had already entered export markets, only one-tenth developed a new product to do so, two-thirds sold products that they already produced for domestic consumption, and another quarter adapted such products for export. In the same way, among non-exporters, "new product(s) would be developed by only four of the twenty-seven non-exporters interviewed". More than half of the non-exporters indicated that they would initiate exports by selling an existing product. About one third of the non-exporting companies stated that they would adapt an existing product for export sales. Those that did think it was necessary to change their product or develop a new one cited most frequently the need to improve product quality and to adapt its design to foreign markets. In short, product development is by no means a necessary precondition for exporting.
Next substantial start-up cost is related to distribution. The problem of establishing distribution channels is substantial, both domestically and internationally, for firms that sell directly to foreign buyers. But firms can hire third parties at some cost to handle distribution and contain this type of start-up cost. Consequently, indirect distribution channels were anticipated twice as frequently as direct channels among non-exporters. However, direct and indirect channels were used equally among firms already exporting. For indirect distribution, agents and distributors were the dominant mechanisms whereas trading companies were found unusual.
Finally, the efforts that firms invest in educating themselves and their buyers should also be viewed as part of the start-up costs of becoming an exporter. After transportation problems, firms viewed customs clearance, Colombia's international reputation, and documentation problems as the most important nonfinancial obstacles to exporting. Each of these problems persists with each shipment, but is moderated in some measure by learning. Therefore, learning of the exporting procedure is also considered as one of the start-up costs.
In order for firms to enter international markets, they have to pay a fixed entry cost which does not vary with export volume or per-unit cost. Kang and Kim (2010) explains with the help of Melitz (2003) model that firms export their products to countries with more than some cut-off level, since the profits from the foreign markets are non-negative. Countries with income higher than that cut-off level attract a large number of varieties as they provide the monopolistically competitive firms with positive profits. Therefore, firms should first export its varieties to large economies. Kang and Kim shows in their model, with increasing income levels of destination countries, profits of the domestic exporting firms also increase. As destination incomes rises, and trade and market entry costs decrease, firms export to more destinations because the required level of cutoff income decreases. Thus firms are prompted to extend their products geographically.
Table 5-1: Previous studies of export diversification, trade cost and findings Country and
time period Measurement of diversification and level of data disaggregation
trade cost Outcomes
Amurgo-Pacheco and Pierola, 2008
24 developed and developing countries (1990 to 2005)
Intensive and extensive margin;
Distance and FTA negative and positive
Amurgo-Pacheco, 2006 EU, the Mediterranean partners, African countries, EFTA countries, USA, Canada and Japan (1990-2004)
Intensive and extensive margin;
Distance and FTA negative and positive
Helpman, Melitz, &
(1970-1997) Intensive and
extensive margin Distance, FTA,
Currency Union negative, positive, positive
Shepherd, 2011 118 developing
countries (2005) Extensive margin,
Eurostat 6-digit Distance, Entry cost,
Export cost negative,
negative, negative Klinger and
Lederman, 2011 73 countries,
(1994-2003) Extensive margin,
HS 6-digit Barriers to entry positive
There are few studies that follow the assumptions of the new-new trade theory or the Melitz (2003) model and also relate the cost reducing tool for increasing export diversification. Among such studies, Amurgo-Pacheco and Pierola (2008) use Melitz (2003) model to explain the geographical and product diversification patterns across a group of developed and developing nations, using HS6 digit trade data. They used this internationally comparable disaggregated country level trade data from 1990 to 2005 and for 24 countries. The group of countries that were selected for their study accounts for a very significant share of world trade. They conclude that FTAs have positive impacts on export diversification for developing countries and distance has a negative relation with export diversification. Their finding indicates that reducing trade costs increases the chances of exporting a wider variety of goods.
Again the study of Amurgo-Pacheco (2006) is one where Melitz (2003) model is used to explain the effect of trade liberalization on the range of products. He uses HS6 digit data on Euro-Mediterranean trade and finds expansion in the range of products at the time of FTA. The set of countries encompasses the EU, the Mediterranean partners (Algeria, Egypt, Israel, Jordan, Lebanon, Morocco, Palestinian Authority, Syria, Tunisia, and Turkey), three other African countries (Mauritania, Libya, and Nigeria), the EFTA countries (Switzerland, Norway, and Island), USA,
Canada, and Japan. The time series inspection shows variation in export and that the number of zero trade values has been decreasing over time for all Euro-Med partners.
Helpman, Melitz and Rubinstein (2008) develop a theoretical model along with the lines suggested by Melitz (2003). Their model of internationel trade in differentiated products describes a situation in which firms face fixed and variable costs of exporting. Looking at the trade data for 158 countries from 1970 to 1997, they find that country i exports more to country j when. the two countries are closer to each other, they both belong to the same regional free trade agreement (FTA), they share a common language, they have a common land border, they are not islands, they share the same legal systems, they share the same currency, or one country has colonized the other.
Theoretical model of Helpman et al (2008) suggests that trade barriers that affect fixed trade costs but do not affect variable (per-unit) trade costs, satisfy the exclusion condition of less productive firms. They added that, these bilateral variables reflect regulation costs (by construction) should not depend on a firm s volume of exports to a particular country. They used country level data on regulation costs of firm entry, collected and analyzed by Djankov et al. (2002), for such fixed costs.
The entry costs for an entrepreneur to legally start operating a business are measured via their effects on the number of days, the number of legal procedures, and the relative cost as a percentage of GDP per capita. Helpman et al. empirically confirms that these costs also affect the costs faced by exporting firms from that country.
Dennis and Shepherd (2011) shows that trade facilitation can be highly effective in promoting export diversification in developing countries. They use the definition of trade facilitation as any policy that reduces the transaction costs of international trade. They use data from World Bank Doing Business that captures all official costs to export a standardized container-load of goods and also costs of starting business to proxy the cost of market entry. They find that reducing these costs by 10 percent of export costs, transport cost and market entry cost increase export diversification by 3,4 and 1 percent, respectively.
On the other hand, the study of Klinger and Lederman (2011) suggests a market failure hypothesis that discovering that a product can be profitably produced and exported by a particular country creates knowledge spillovers. Firms that imitate the first mover derive benefits from such discoveries of the first movers and can learn about production costs and foreign demand without any cost. Therefore, the first mover entrepreneurs cannot fully appropriate the gains from their investments. This is because entry by imitators either reduces the price of new export or raises the unit costs. Their study findings suggest that export growth is positively and significantly associated with the number of export varieties and also partially by barriers to entry. They explained that, when first
movers can internalize more of the benefits of discovering a new export activity, the frequency of discovery rises.
Similarly, other studies consider distance and FTAs as export cost factors. Therefore, existing literatures that consider various cost factors as barriers to trade do not specifically cover individual country cases of China, Japan and Korea. In addition, they do not cover time series or panel data analysis for the barriers to trade and export diversification. This study estimates the effect of export, import and starting business cost on extensive and intensive margin of export. These cost data are compiled from World Bank Doing Business database available from 2005 to 2011. Under the circumstances, this research is expecting to contribute to the international trade literature by including the study of export diversification and trade costs in China, Japan and Korea.