Bangladesh’s Trade Barriers in a Global Perspective: A Comparative Analysis by
SELIM RAIHAN ∗
This paper develops an index of trade barrier for 108 countries and makes a comparative analysis of Bangladesh’s trade barrier indices in a global context. Bi-variate as well as multivariate cross-country econometric models have been estimated to explain cross-country variations in trade barrier indices. The results show that cross-country variations in trade barrier indices are much influenced by variations in per capita income, population, the ratio of foreign direct investment to GDP and literacy rate. The findings for a much disaggregated level of commodity categories show that Bangladesh possesses high trade barrier indices. I. INTRODUCTION
How liberalised is the trade regime of Bangladesh in a crosscountry context? The answer to this question is not straightforward as there are many ways of defining trade liberalisation in a crosscountry context. 1 Three types of indicators of trade liberalisation have been used in the economic literature for cross-country comparisons of trade liberalisation. The first type of indicators includes the outcome variables, such as the export-GDP ratio, the import-GDP ratio – or the trade-GDP ratio (exports plus imports as a share in GDP). The second type of indicators includes the policy variables, such as tariffs and non-tariff barriers to trade. Finally, the third type of indicators refers to the indirect measures of trade liberalisation, which include non-trade variables or subjective ∗ 1
The author is an Assistant Professor at the Department of Economics, University of Dhaka. For a list of trade liberalisation indicators in a cross-country context see Appendix 1.
The Bangladesh Development Studies
considerations of a country’s trade regime. There is a considerable debate concerning which indicator best explains the extent of trade liberalisation in a cross-country context. Though an outcome variable, such as the trade-GDP ratio, provides a reasonable understanding of the changes in the overall trade regime in a single-country context, its use as a measure of trade liberalisation in a cross-country context has been criticised. There is a concern that the cross-country variation in the trade-GDP ratio may not be associated with differences in trade regimes. Rodriguez and Rodrik (2001) argued that cross-country variations in the trade-GDP ratio may be primarily influenced by structural factors, such as geographical location, rather than by differences in trade restrictions. On the other hand, the third type of indicators has been criticised because they rely on non-trade variables, such as the exchange rate and the black market premium. The second type of indicators has been preferred in many studies despite the difficulties in obtaining data on the policy variables, such as tariff and non-tariff barriers for cross-country comparisons. However, these indicators also have problems. Indicators such as the ‘‘simple average tariff’’ may understate the extent of tariff dispersion in many of the economies. The use of the ‘‘import-weighted tariff’’ has been considered better than the ‘‘simple average tariff’’, although the former may understate the extent of tariff protection as the import-weight does not capture the tariff figures for the zero-valued imports. Furthermore, the construction of the ‘‘import-weighted tariff’’ is usually performed at a relatively aggregated level, which may fail to take into account the variations in tariffs and import weights at a more disaggregated level. It is also the case that data on non-tariff barriers are blurred for many countries which makes it difficult to quantify them for cross-country comparisons. Against this backdrop, we have made an attempt to construct an index of trade barriers for a number of countries by using both the tariff and import data. The advantages...