阿根廷的贸易改革、市场准入和贫困

WP5 ~3 hW POLItcY RESEARCH WORKING PAPER 3 135 Trade Reforms, Market Access, and Poverty in Argentina Guido G. Porto lThe World Bank Fi W Development Research Group 1Trade September 2003  POLICY RESEARCH WORKING PAPER 3135 Abs7tract Much of the literature that studies the relationship reforms, including tariff cuts on consumption goods and between trade and poverty in developing countries capital goods in Argentina. Foreign trade reforms include focuses on the effects of national trade reforms, such as the elimination, in industrial countries, of agricultural own tariff reductions. In contrast, the World Trade subsidies and trade barriers on agricultural manufactures Organization negotiations at the Doha Round were more and industrial manufactures. These policies enhance the concerned with the poverty effects on low-income market access of Argentine exports. Overall, a countries, and of foreign reforms, such as the elimination combination of own reforms and enhanced market access of agricultural subsidies in industrial cconomnies. would cause poverty to decline by betwveen 1.7 and 4.6 Porto empirically compares the relative poverty percentage points. This evidence suggests that trade impacts of national and foreign trade reforms in policies can be important poverty-reducing instruments Argentina. The author investigates national trade in Argentina. This paper-a product of Trade, Development Research Group-is part of a larger effort in the group to assess the poverty impacts of trade and trade policies. Copies of the paper are available frcc from the World Bank, 1818 H-. Street NW, Washington, DC 20433. Please contact Paulina Flewitt, room MC3-333, telephone 202-473.-2724, fax 202-522-1159, email address pflewitt@worldbank.org. Policy Research Working Papers ate also posted on the Web at http:// econ.worldbank.org. The author may be contacted at gportouworldbank.org. September 2003. (27 pages) The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and shoild be cited accordingly. The findings, interpretations, and conchlsions expressed in this paper are entirely those of the authors. They do not necessarily represent the view of the World Bank, its Executive Directors, or the countries they represent. Produced by the Research Support Team  TLrade Reforms, Market Access and Poverty in Argentina* Guido G. Portot Development Research Group The World Bank *1 wish to thank I. Brambilla, B. Hoekman, and M. Olarreaga for continuous support and very valuable comments. tCorrespondence: The World Bank 1818 H Street, Washington, DC, 20008. MC3-312. Email: gporto@worldbank.org   IL Introduction Much of the literature that studies the relationship between trade and poverty in developing countries focuses on the effects of national trade reforms, such as own tariff reductions (Porto, 2:002; Goldberg and Pavcnik, 2003). In contrast, the WTO negotiations at the Doha Round wvere more concerned with the poverty effects, on low-income countries, of foreign reforms, such as the elimination of agricultural subsidies in developed economies. The purpose of this paper is to empirically compare the relative poverty impacts of national and foreign trade reforms in Argentina. The national trade reforms investigated in this paper include tariff cuts on consumption goods (food, appliances) and capital goods (machines) in Argentina. These policies generate a decline in the domestic price of these goods. Foreign trade reforms include the elimination, in developed countries, of agricultural subsidies and trade barriers (tariff and non-tariff) on agricultural manufactures (dairy products, beef, oils) and industrial manufactures (textiles, transport materials, chemicals). These policies enhance the market access of Argentine Exports and cause an increase in the international (and hence in the domestic) prices of these goods. The theoretical links that form the basis of the methodology used in this paper are simple. The policy induced changes in the prices of traded goods that Argentine consumers and producers face generate a change in the relative demand of different factors of production, particularly labor. As a result, wages and household income react and poverty is affected. The estimation of the poverty impacts of trade liberalization comprises three steps. First, I assess the changes in the prices of imports and exports induced by national and foreign trade reforms. Second, I estimate wage price-elasticities that measure the responses of Argentine wages to the price changes. Finally, I use the simulated policy-induced price changes and the estimated wage price-elasticities to predict the labor income that would hypothetically be earned by each Argentine household after the reforms. To study the poverty impacts, I compute pre- and post-policy head-count ratios.' 'The head count ratio is the proportion of the population with an income lower than the poverty line (the amount of money needed to purchase a basic bundle of food and non-food products).  To compare the relative effects of national and foreign policies, I distinguish the marginal effects of the reforms, i.e. the marginal effect of a trade policy on household income, from the scope for reforms, i.e. the room for further policy changes. In Argentina, I find that national trade reforms have larger marginal effects than foreign trade reforms. However, since there is greater room for foreign reforms, policy changes in developed countries would have, in the end, larger poverty impacts.2 Specifically, the joint elimination of Argentine tariffs on consumption goods and capital goods would cause the head count ratio to decline by between 0.6 and 1.7 percentage points. Foreign policies on agriculture and industry would lower the poverty rate by between 1.4 and 2.9 percentage points. Overall, a combination of own reforms and enhanced market access would cause poverty to decline by between 1.7 and 4.6 percentage points. With an actual initial head count ratio of 25.7 percent, this evidence suggests that trade policies can be important poverty-reducing instrumerts in Argentina. The paper is organized as follows. In section 2, I define the poverty indicator, namely the head count ratio, and I use it to characterize the poverty situation in Argentina during recent years. Section 3 discusses the methodology used to estimate the changes in household income, the wage price-elasticities, and the poverty measures. Section 4 implements the methodology using Argentine data and assesses the poverty impacts of national and foreign trade reforms in the country. Section 5 provides a conclusion of findings. 2 Poverty 'an ArgeBnt'na The poverty measure used in this paper is the head count ratio, HC, defined as the fraction of the population with an income below the poverty line z. That is, (1) HC= Z l{Yi < z}, where N is the total population and 1{} is an indicator function that takes the value of one if the argument within brackets is true. The poverty line z is the level of income 2In other countries with higher national trade barriers (such as tariff or quantitative restrictions), own reform is likely to be more important. 2  needed to purchase the poverty consumption basket, which includes food items that satisfy a minimum caloric and energetic intake, and non-food essential items (clothing, housing, lhealth and education). The poverty line is measured so as to account for the different caloric requirements of individuals with different characteristics, such as sex and age.3 This means that z and individual income, y , are measured in per equivalent adult units (Deaton, 1997). ]:n Argentina, the National Institute of Statistics and Census (INDEC) estimates poverty lines and per equivalent adult scales (INDEC 2002). The poverty analysis focuses on the metropolitan area of Buenos Aires, the most populated urban area in Argentina (comprising around one third of the population of the country).4 After a peak of 47.3 percent in 1989, due to a hyperinflation episode, the head count ratio decreased until May 1994. Since then, it has steadily increased. In 1992, when a number of reforms began, the poverty rate fluctuated around 18-19 percent. Whereas in 1999, the baseline period that I use to explore the effects of trade policies, the head count ratio was 25.7 percent.5 Figure 1 plots the density of the logarithm of individual income per equivalent adult (lpai) in October 1999 (in monthly dollars) in Gran Buenos Aires (GBA). The density is estimated with standard kernel methods using the optimal bandwidth and a Gaussian Kernel (Pagan and Ullah, 1999; Silverman, 1986). There are two vertical lines in Figure 1. The rightmost vertical line at lpai=5.04 (so that income per equivalent adult is $155, around $5 per day) represents the official poverty line z for October 1999. The area below the density curve and to the left of this poverty line is the proportion of poor people in Buenos Aires, 25.7 percent. The leftmost vertical line at an income per equivalent adult of 65 monthly dollars represents the indigence line, the expenditure needed to purchase a minimal food bundle only. In October 1999, 11.3 percent of the population of GBA was living in extreme poverty. 3The reference group comprises adult males. The caloric requirements of other individuals are measured relative to this reference group (so that, for instance, an adult male requires more calories than an adult female, and children require less calories than adults). 4The reason why I look at Gran Buenos Aires (the metropolitan area) is that, before 2001, poverty lines were calculated only for this area. -II chose 1999 as the baseline year to isolate the impact of the recent recession and current crisis. In 2002, the head count ratio reached 49.7 percent, a figure that characterizes the impact of the crisis that is hitting the country. 3  The purpose of this paper is to assess how poverty, as measured by the head count ratio, is affected by national and foreign trade policies. I do this by predicting the new income that each Argentine household would earn as a result of a policy change and by comparing the pre and post policy head count ratios. 3 The MethdRogy Conceptually, there are three links in the methodology that I use in this paper. The initial step is a trade shock, i.e. a national or a foreign trade reform, which causes a change in the domestic prices of traded goods (exports and imports) in Argentina. The second step is the response of the labor income of Argentine households, which leads to the third step, the induced change in the head count ratio. Discussion of these three components follows. 3.1 Chtanges in th1e ILaIces oif flraded1 Goodls The transmission of trade policies to prices is different for different goods. In this paper, I work with four different products. There are two exportable goods: agnicultural manufactures (dairy products, beef, oils) and industrial manufactures (textiles, chemicals, transport material). There are two importable goods, too: consumption goods (food, appliances) and capital goods (machines). I assume that Argentina is a small open economy that faces exogenously given prices for these goods. In principle, national and foreign trade policies may affect the domestic prices of both exports and imports. Whereas the effects of foreign policies are revealed in changes in international prices, national policies introduce a wedge between international and domestic prices. Since Argentine trade policies mostly involve intervention on imports, I focus on national trade policies that affect the domestic prices of importable goods. In contrast, I focus on foreign trade reforms that affect the price of Argentine exports for these are the most important policies from the Argentine standpoint. Let Trg be the national tra-de policy parameter (a tariff). The price of an importable in 4  Argentina, pi, is (2) pi = pg*1(rg) where pt* is the international price and ip(.) is the function that characterizes the pass-through of national trade policies to domestic prices. To get the price change induced by a policy reform, I need estimates of this pass-through function, as I discuss in section 4. For exportable products, let rT be the foreign policy parameter (tariff protection, production support or export subsidies in large developed economies). The domestic price of an exportable, p', is (3) PC = p` (T*) wNhere p` is the international price, which depends upon the trade policy parameter. In section 4, I discuss how to get estimates of the price change induced by trade policy reforms. 3.2 Changes in Household Income: the Wage Price-Elasticities This section explains how to measure the changes in household income caused by trade policies. Let the income of household j, Yi, be (4) Yi = w + ki, m where wi is the wage earned by household member m (head and non-head), and ki is non-labor income, including profits, returns to specific factors and transfers. Due to data constraints, I am forced to focus on the labor income of all household members, thus neglecting non-labor income (ki).6 The change in household income caused "iThere might be concerns that leaving capital income and profits aside can produce biased results for the poverty analysis. However, such biases are unlikely to be relevant in Argentina because capital and land ownerships tend to be concentrated in the upper tail of the distribution of income, with no poverty impacts. 5  by a change in the policy parameter rT that affects the price of an exportable good pg is (5) dY3 = Z ape Pg dr + a m U 9 0T A similar expression can be obtained for the case of a trade policy that affects import prices. The proportional changes in the total (labor) income of household j is given by dY" elp (6) Yj =U4 "0mt