Trade Structure and Trade Flows in Cambodia: A Gravity Model

Article excerpt

I. Introduction

Cambodia's entry into the ASEAN Free Trade Area (AFTA) in 1999 offers substantial potential for its integration with international economies. Since international trade has been acknowledged as being important for developed economies as well as developing economies, it is widely expected that Cambodia's AFTA membership improves welfare through the creation of increased access to global markets and expanding trade flows) Indeed, Cambodia's trade volume has been growing steadily since its accession to AFTA.

Many international trade theories attempt to explain a country's trade flow, but the result is generally dependent on the model specification. Careful empirical work should be required to understand the determinants of trade flows. The gravity model, pioneered by Tinbergen (1962) and Linneman (1966), is renowned for the centre of empirical studies. Its theoretical foundations have also been proposed from different frameworks, such as a Heckscher-Ohlin model and an increasing returns to scale (IRS) model (see, for example, Helpman and Krugman (1985), Bergstrand (1990), Markusen and Wigle (1990), Eaton and Kortum (1997), and Deardorff (1998)). (2)

Examining which theory actually accounts for trade pattern in Cambodia would allow us to understand the causes of international trade and to seek for appropriate policies for continual increase in trade and sustainable economic growth. To address such an identification problem, this study empirically analyses the determinants of Cambodia's trade flows in a framework of the gravity model using trade flow data over the period from 2000 to 2004 after the entry into AFTA.

The pioneering work of Evenett and Keller (2002) examines such an identification issue by conditioning bilateral trade relations on factor endowment differences and on the share of intra-industry trade. Since policy implications are crucially connected with its underlying trade model, the model identification would be more important when we discuss trade issues in a specific country. Focusing on the problem of Korea's trade flows, Sohn (2005) identifies the underlying trade model in the gravity equation based on the argument that a Heckscher-Ohlin model explains bilateral trade flows between countries with large factor endowment differences and a high degree of inter-industry trade, while an IRS model explains bilateral trade flows with a high degree of intra-industry trade. His empirical finding is that South Korea's trade flows could follow a Heckscher-Ohlin model rather than an IRS model.

To discuss the model identification in Cambodia, we introduce a measure of trade structure in the gravity equation. One of the major characteristics in a standard gravity model is that it does not explicitly include a factor endowment variable, although income level differences reflect factor endowment differences. Following the concept in Drysdale (1969), Elliott and Ikemoto (2004), and Sohn (2005), our index measures the degree of trade complementarity between two countries and reflects factor endowment differences. Since a high degree of complementarity would be associated with a large difference in factor endowment, trade flows increase with rising the degree of complementarity in a Heckscher-Ohlin model. In contrast, they decrease with rising the degree of complementarity in an IRS model, since falling trade complementarity generally intensifies intra-industry competition in an IRS model with differentiated products. Thus, the coefficient on the measure of trade structure in the modified gravity equation provides crucial information for our identification issue, as discussed in Sohn (2005).

Our analysis is conducted from trade flow data of Cambodia and its twenty major trading countries. The empirical results show that a higher degree of trade complementarity is associated with a higher level of trade flows, i.e., a factor endowment difference between Cambodia and its trading partners is one of dominant driving forces to trade flows. …