Vi Digital Library - Text Preview
Gains from Varieties: Analysis of the Benefits of Trade Agreements in South America
Working paper by Orlando Monteiro da Silva; Jacqueline Silva Batista, 2015
Gains from varieties: Analysis of the benefits of trade agreements in South America
Orlando Monteiro da Silva; Jacqueline Silva Batista1
Most researchers have emphasized the performance of exports and foreign investment, but neglect the investigation on the way in which countries have benefited from imports. This study seeks to contribute to this issue by analyzing the gains from imports and the contribution of Regional Trade Agreements to these gains. There are two main trading blocs in South America: MERCOSUR (Southern Common Market), which in its original form is composed of Argentina, Brazil, Paraguay and Uruguay, and the Andean Community or Andean Pact, whose current members are Bolivia, Peru, Ecuador and Colombia. Therefore, this study aimed to measure and analyzes the trade gains of the countries of South America due to the increased number of imported varieties, per sector and per country, from 1995 to 2011. The methodology proposed by Feesntra (1994) and adapted and expanded by Broda and Weinstein (2006) consists of constructing an exact price index that considers the effects of different varieties of imported products and the way they affect the welfare of consumers in each country. The results showed that, in that period, there was a substantial increase in varieties of imported products and that smaller countries had benefitted more from falling prices. The analysis of intra and extra-bloc imports showed that the declines in prices resulting from intra-bloc imports are relatively small compared to those of non-member countries. Openness to international trade and participation in bilateral and regional trade agreements are the major factors leading to gains from importing varieties.
Keywords: South America; trade benefits; gains from variety.
JEL: F14; F15
Regional Trade Agreements (RTA) are agreements between governments seeking to promote trade among the participating countries and are often defined according to regions and levels of economic integration. There are two integration blocs among the countries of South America: MERCOSUR, comprising Brazil, Argentina, Uruguay and Paraguay, and the Andean Pact, which includes Bolivia, Peru, Ecuador and Colombia. Currently, these blocs also interact with each other through a larger association called UNASUL (Union of South American Nations)2.
Since its inception, MERCOSUR has promoted reduced trade barriers between member countries, but failed to develop a more advanced institutional framework to deal with the tensions and conflicts among its original members (Brazil, Argentina, Uruguay and Paraguay) and among the associate members (Chile, Bolivia, Colombia and Ecuador) and the newest member (Venezuela). According to Serbin (2009), the fact that more than half of the Gross Domestic Product of MERCOSUR comes from Brazil causes a lot of friction within the block and has hampered the creation of a single currency. The existence of large market asymmetries in the bloc´s economy is often used to justify discontinuity in the process of economic liberalization among members.
The arguments for trade liberalization are based on the theories of comparative advantage, which highlight the importance of better allocation of resources, thus increasing technical efficiency and productivity generated by increased competition with international products. They are also based on the benefits of increased diversity or variety of products. In countries opened to trade, consumers and producers have access to a wider range of products, inputs and capital goods, which increases the well-being of consumers and enhances the efficiency of producers.
The process of trade liberalization in Latin America has progressed significantly in the last 20 years. Souza (2007), for example, shows that, according to the Warner-Sachs (SW) indicator, the 18 Latin American economies3 analyzed in the study could be regarded as opened in the 90s. Also, according to Burki and Perry (1997 apud Souza, 2007 p.23), levels of restrictions on international trade in Latin America are similar to those of the East Asian newly industrializing countries, such as Singapore, Taiwan and Thailand. According to Souza (2007), the intensity of the adjusted international trade index for Latin America indicates a positive influence of trade policies adopted and the trade flow in the 90s. This index was relatively low for Brazil and Argentina in the early 2000s, due to the energy crisis in Brazil and Argentina's lack of credibility. These facts discontinued the process of liberalization. There was, however, a growing share of manufactured goods in total imports, post-liberalization, which reached an annual average of 74.9% in Latin America as a whole, 75.2% in MERCOSUR and 67.0% in Brazil, in the 90s.
An important aspect of trade openness is the positive effect on consumer’s welfare, resulting from the greater number of varieties of imported goods. Although the gains of trade arising from importing new varieties are established in the theory of international trade (Dixit and Stiglitz, 1977, Krugman, 1979, 1980, 1981; Helpman, 1981), empirical estimates measuring the effect of the greater number of varieties on the aggregate consumer’s welfare have emerged only recently. The basic argument arose from the idea that the use of the conventional price index for a uniform set of products over time disregarded the benefits that consumers would obtain with an increased number of varieties.
The work of Feenstra (1994) is a pioneer study on this area. It recognized this bias and proposed an extension of price index by Diewert (1976), Sato (1976) and Vartia (1976), which took into account the emergence or extinction of varieties and allowed assessing gains from trade with the introduction of varieties of a particular product. The aggregation of all goods imported by a country, however, has brought problems related to the availability of a large number of data, which reduced the use of the measure of gains from trade proposed by Feenstra.
An aggregate measure of trade gains with varieties was not performed until Broda and Weinstein (2006) expanded the Feenstra’s methodology for the case of multiple goods, thus obtaining an accurate import price index that considers account changes in varieties. These authors analyzed the gains from trade with imports of new varieties in the United States between 1972 and 2001 and demonstrated that the bias on the import price index was 28% for that period, or 1.2% per year, and that the gain of consumers with increased number of varieties corresponded to 2.6% of GDP.
Other studies have emerged after Broda and Weinstein (2006), including those of Mohler (2009) and Mohler and Seitz (2010), who calculated the welfare gains generated by new varieties in Switzerland, between 1990 and 2006, and in 27 European Union countries between 1999 and 2008, respectively. The studies of Cabral and Manteu (2010), for Portugal, and Chen and Jacks (2012), for Canada, can also be mentioned. In Brazil, Barroso (2010) investigated the gains from imports of intermediate and final goods; and Silva and Batista (2013) studied the country gains from varieties, from 1995 to 2011, and assessed the contribution per country of origin and sector of the economy. However, no study has compared a large number of developing countries, for which the magnitude of the effect is expected to be different from that of developed countries.
It is known that trade liberalization leads to greater international integration of markets and creates more competition and new challenges for countries, such as increased trade flows. In the case of South America, besides the general openness of the markets, there are Regional Trade Agreements (MERCOSUR and the Andean Pact). It allows calculating the effects of imports that are external or internal to the agreements and the possible gains from trade among member countries.
Therefore, the present study aimed to analyze imports of the South American countries, from 1995 to 2011, with emphasis on the different varieties of products and the welfare gains they generated. Based on the calculation of an exact import price index, it is still possible to calculate gains and/or losses per sector and participation or not of countries in different Regional Trade Agreements.
The theoretical model is based on a model of monopolistic competition described by Dixit and Stiglitz (1977) and used by Krugman (1980) to explain, in the new international trade theory, how the decision of firms affects aggregate trade flows. The main idea is that trade growth can be decomposed into an "intensive" component referring to changes in the trading volume of existing firms and an "extensive" component, due to the entrance of new firms or exit of existing firms in the export market. The implications in terms of prices is that the entrance of new firms in the market leads to competitive effects and decreased aggregate prices of existing firms (intensive effect), and of the new firms (extensive effect), due to a self-selection caused by the effect of agglomeration.
The model described here agrees with the work of Manteu and Cabral (2010) and propositions of Feenstra (1994) and Broda and Weinstein (2006). Feenstra first proposed an exact import price index for a single good, allowing changes in variety and quality of that good. Broda and Weinstein (2006) expanded the methodology of Feenstra for all goods imported by a country.
The preferences of a representative consumer were described by Weinstein and Broda (2006) by a utility function with three levels. The first level aggregates the imported varieties of imported goods; on the second level, imported goods are aggregated; and in the third, these imported goods are combined with an aggregated domestic good to generate utility. The specification of the lower level of the utility function from consumption of an imported good in period, , is written as a utility function with constant elasticity of substitution (CES) on different varieties of this good, and a variety defined as a good imported from a country :
whereis the utility derived from the imported variety c of the product in the period t, >0 is the parameter that refers to taste or quality that describes consumer's preference for that differentiated variety c, and C refers to a set of available countries and therefore of potentially available varieties on period t. denotes the elasticity of substitution among varieties imported from various countries, differing among the goods. It is assumed that its value is greater than unity.
The minimum unit cost functions derived from the utility function may be used to obtain an exact price index for product, as shown by Diewert (1976). In the case of the CES function, Sato (1976) and Vartia (1976) demonstrated that the exact price index can be written as a geometric mean of the individual price changes using the ideal log-change weight:
where, is the price of the variety c of the goodin the period t, ⊂ C is a subset in the total of varieties of the good consumed in the period t . = ∩is the set of common varieties consumed in both periods t and t-1.is the ideal log-change weight calculated using the expenditure shares in the two periods4
2UNASUL is composed of 12 countries: Argentina, Bolivia, Brazil, Chile, Colombia, Ecuador, Guyana, Paraguay, Peru, Suriname, Uruguay and Venezuela.
3 The Countries are: Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Ecuador, Guatemala, Honduras, Mexico, Nicaragua, Panama, Peru, Paraguay, El Salvador, Uruguay e Venezuela.
4 , where = , and