کشش تقاضا در تجارت بین‌المللی: آیا این کشش واقعاً پایین است؟

At the intuitive level, trade economists generally believe that most developing and smaller developed countries do not have market power in the world market meaning that they face infinitely elastic demand for their goods. Yet, the estimates of import demand elasticities facing them rarely exceed 3. In this paper, we provide evidence supporting trade economists' intuition. Using highly disaggregated data on textiles and apparel and exploiting the fact that these products are subject to Multi-Fiber Arrangement quotas, we find the import demand elasticity facing Bangladesh to be 26. We also find high cross-price elasticity with respect to the competitor countries.

To-date, econometric estimates of demand elasticities in international trade have defied the intuition of trade economists. The consensus is that most developing countries and smaller developed countries have virtually no market power in the world markets. The estimates of demand elasticities in international trade, on the other hand, have rarely exceeded 3. For instance, in the widely cited survey by Goldstein and Khan (1985), the estimates of the elasticity of demand, facing the exports of such small countries as Austria, Belgium and Denmark, are uniformly less than 1.6. If we take these estimates seriously, the case for unilateral trade liberalization in small countries is seriously undermined. The estimates imply a considerable market power on the part of the countries and, beyond a point, make unilateral liberalization by them a welfare-reducing proposition. The estimates also raise doubts about exports serving as the engine of growth. For, even after we take into account the expansion of world demand due to growth in income, if price elasticities are as low as is suggested by the current estimates, a 20% per annum expansion of a country's exports is bound to worsen its terms of trade substantially. Likewise, the low elasticities cannot be reconciled with the rapid growth of East Asian exports that took place in recent decades at relatively stable terms of trade. The sole author, who has seriously challenged the validity of the low elasticity estimates, is James Riedel. In an influential paper, published more than 10 years ago, Riedel (1988) tests and accepts the null hypothesis that Hong Kong is a small country. Riedel begins by rejecting the common assumption in the literature that the elasticity of supply of exports is infinite and, therefore, price can be treated as an exogenous variable. Instead, he treats both quantity and price as endogenous and, in the demand equation, estimates the price as a function of quantity. He finds that the change in the quantity does not have a statistically significant effect on the price. Thus, the small-country model is validated. Athukorala and Riedel (1991) confirm this result for the Republic of Korea. The findings of Riedel (1988) and Athukorala and Riedel (1991) have led to a lively debate in the literature. Nguyen (1989), Muscatelli et al. (1992) and Muscatelli (1994) take issue with Riedel and defend the conventional, low elasticity estimates. While Riedel (1989) and Athukorala and Riedel (1994) stand their ground, a key weakness in the defense of the small-country assumption remains: their conclusion is based on statistically insignificant coefficients rather than an elasticity estimate that is large and statistically significant. In this paper, we offer strong evidence supporting the position argued by Riedel, 1988 and Riedel, 1989 and Athukorala and Riedel, 1991 and Athukorala and Riedel, 1994. We estimate the US demand for the products imported from Bangladesh under the Multi-Fibre Arrangement (MFA) and find the own-price elasticity of demand to be consistently high and statistically significant. In the case we report in the paper, our estimate of the elasticity is 26. In other cases, it is even higher. A key distinguishing feature of our estimates is their robustness. Authors estimating demand elasticities have had only limited success obtaining statistically significant coefficients. Using a Seemingly Unrelated Regression (SUR) model, we estimate a system of eight equations, one relating to each competitor of Bangladesh, with cross-equation restrictions. We estimate the model for some individual MFA products as well as for a larger sample obtained by pooling the data across several products. In the latter case, we estimate the system with as well as without product and time fixed effects. We find the estimated coefficients to be robust in terms of their sign and, for most part, statistical significance as well. All three estimation methods lead to very high own-price and cross-price elasticities. This consistency of results is rare in the empirical literature on import-demand elasticities. Our analysis departs from much of the literature on international trade elasticities in four important respects. First, contrary to the general practice of postulating an ad hoc equation, which violates theory, we derive a set of estimation equations from an explicit, utility-maximization model.1 We estimate these equations as a system and obtain the relevant parameters of the utility function. We then use the estimates to obtain the Marshallian own-price and cross-price elasticities as well as the income elasticity of demand. Thus, there is a tight link among our theoretical model, estimated equations and elasticities.2 Second, related to the first, our estimation exploits the fact that imports of MFA products are subject to country-specific quotas. Because the quotas are binding, we can treat the quantities as exogenous and prices as endogenous.3 Thus, we have a natural reason for treating prices as endogenous variables and quantities as exogenous variables. If quotas were unchanging over time, this will, of course, pose the problem of insufficient variation in the quantity variable. But at the time the quotas are introduced, the annual rates at which they are allowed to grow in the future are also specified. Thus, the quantities grow annually at predetermined rates, giving the necessary variation in data. Third, based on our theory, we take explicit account of imports from competitors of Bangladesh. The common practice in the literature is to estimate the demand for a country's (total) exports as a function of that country's price, relative to an index of the prices prevailing in the importing country. This approach misses the important feature of reality that, for products exported by developing countries, principal competitors are located outside the importing country. The latter is typically a developed country while the competitors are other developing countries. In our specific case, the competitors of MFA products exported by Bangladesh to the US are primarily exporters of similar products located in Asia and Latin America. It is critical to take into account the supplies of these countries in the course of estimating the demand for imports from Bangladesh. Finally, the bulk of the empirical literature on import-demand functions relies on highly aggregated data. We use disaggregated data by exploiting, for the first time, the information available on MFA imports into the US. A benefit of using the disaggregated data is that unit-value indices, which must inevitably be used to represent prices, are far more meaningful in these than aggregated data.4 Compositional changes are far less likely to pollute unit values when data are highly disaggregated. For example, as of 1994, there were as many as 148 MFA product categories in the US. Cotton shirts alone were divided into four separate categories: cotton-knit shirts for men and boys, cotton-knit shirts for women, girls and infants, cotton non-knit shirts for men and boys and cotton non-knit shirts for women, girls and infants. Additionally, since quotas are closely monitored, these data are likely to suffer less from measurement errors than the aggregate trade data used by most investigators. Having laid out our claims in strong terms, we may also note some of the limitations of our analysis. First, like other investigations that estimate the demand function in the log-linear form, we make use of separability in the utility function. Without this assumption, it is not possible to estimate a demand equation unless we have information on the entire company.5 It can be stated in our defense, however, that this is a common assumption in the literature on import-demand elasticities. Second, MFA products are rather special. For each MPA product, there is a detailed definition, which makes it relatively homogeneous. Due to the existence of quotas, we are also able to abstract from supply-side variables and treat prices as endogenous. There are few other products for which these assumptions are satisfied. For most products, it is necessary to address the issue of endogeneity of the quantity as well as price. Finally, as we will show, though the assumption of binding MFA quotas is broadly justified for our data, we cannot claim that it holds for all countries, products and time periods. Therefore, we cannot justifiably claim that simultaneity bias as altogether absent in our results. The paper is organized as follows. In Section 1, we outline the theoretical model from which our estimating equations are derived. Because the equations do not yield the conventional, Marshallian demand elasticities directly, we also explain in this section how these latter can be obtained from the parameters of the utility function we estimate. In Section 2, we defend the assumption of exogeneity of quantities in our regression equations. In Section 3, we estimate the system of equations derived in Section 1 and calculate the price and income elasticities facing Bangladesh. In Section 4, we conclude the paper. Appendix A and Appendix B offer details on the theoretical derivations of price and income elasticities and data sources, respectively.

In this paper, we have offered systematic evidence of very high demand elasticities for products coming from a small country, Bangladesh, into the US. Our analysis differs from the existing studies on the subject in four important ways. First, we use a new methodology, which exploits the fact that MFA exports are subject to binding quotas. Second, there is a tight connection between out theoretical model and econometric estimation. Third, we take explicit account of competitors of Bangladesh. Finally, we use highly disaggregated data which makes unit values a more reliable measure of prices than when aggregate data are used. The results of our estimation are relatively robust to the inclusion of commodity- and time-fixed effects. The most surprising finding is the consistently high value of the own-price elasticity. This high value accords with trade theorists' prior that small countries can essentially behave as price takers and vindicates Riedel (1988).