TABLE OF CONTENTS
1 - INTRODUCTION
1.2 OUTLINE OF THE PAPER
2 - DEFINITION AND MEASUREMENT
2.1 TRADE LIBERALIZATION
3 - REASONS FOR TRADE POLICY REFORMS
3. 1 ECONOMIC CRISIS
3. 2 STRUCTURAL ADJUSTMENT PROGRAMS
3. 3 THEORY AND RESEARCH
4 - EXPERIENCES WITH TRADE POLICY REFORMS
4.2 SOUTH AFRICA
5 - TRADE LIBERALIZATION AND POVERTY
5.1 ECONOMIC GROWTH AND POVERTY
5.1.1 ECONOMIC GROWTH
188.8.131.52 Theoretical Linkages
184.108.40.206 Empirical Evidence
5.2 PRODUCTIVITY AND VOLATILITY
220.127.116.11 Theoretical Linkages
18.104.22.168 Empirical Evidence
5.2.2 MACROECONOMIC VOLATILITY
22.214.171.124 Theoretical Linkages
126.96.36.199 Empirical Evidence
5.3 PRICES AND MARKETS
188.8.131.52 Theoretical Linkages
184.108.40.206 Empirical Evidence
5.4 WAGES AND EMPLOYMENT
5.4.1 WAGES AND EMPLOYMENT
220.127.116.11 Theoretical Linkages
18.104.22.168 Empirical Evidence
5.4.2 TRANSITIONAL UNEMPLOYMENT
5.5 GOVERNMENT REVENUE AND SOCIAL SPENDING
5.5.1 GOVERNMENT REVENUE
22.214.171.124 Theoretical Linkages
126.96.36.199 Empirical Evidence
5.5.2 SOCIAL SPENDING
6 - CONCLUSION
LIST OF LITERATURE
After the successful experience of newly industrializing countries in East Asia (e.g., the East Asian Tigers: Hong Kong, Singapore, South Korea, and Taiwan) in the 1960s and Southeast Asia (e.g., the Southeast Asian Little- Tigers: Indonesia, Malaysia, the Philippines, and Thailand) by the late 1970s, trade liberalization (TL) in less developed countries (LDCs) has been considered as a policy to achieve rapid economic development. The argu- ment, put forward for instance by Dollar et al. [2002, p.195], that “TL is good for [economic] growth” and that “[economic] growth, [in turn], is good for the poor” has since served as the departure point for the discussion of the link between TL and poverty among economists, researchers, and policymakers alike.
Spurred on by the dramatic failure of import substitution industrialization (ISI) strategies, and with the advice and support from international financial institutions (IFIs), such as the World Bank, the International Monetary Fund (IMF) and the World Trade Organization (WTO), Krueger [1998, p.1521], for instance, finds that the intervening period has seen a large wave of TL episodes in countries in Latin America, Middle- and North Africa, Sub- Saharan Africa, and South Asia respectively. Believing that TL is vital for economic growth and poverty alleviation, these nations have frequently and extensively used it as a centerpiece for their development strategy. Howev- er, the high expectations held at the times those countries embarked on their trade policy reforms (TPRs) have not always been fulfilled in retrospect.
Herod [2001, p.244] aptly remarks that the “free trade mantra”, as he calls it, which is “preached by developed countries and IFIs” has almost become “like a religion, holding out the promise that if developing countries (DCs) adopt the faith, they will soon be ‘saved’.” However, this view has proved to be inherently simplistic. Although there is a general consensus that in the long-run open economies fare better in aggregate than do closed ones and that relatively liberal trade policies contribute to medium- and long- term economic development, TL by its nature almost certainly involves short- and medium-term adjustment costs and so is likely to have distributional impacts.
Adjustment costs can be quite fatal to the economic welfare case for TL and Trebilcock et al. [1990, p.10], for instance, remark that they can provide an economic justification for a nation’s “retention of its barriers to international trade, even when free trade would be better for it in the long-run.” Although this statement is equally crude as the aforementioned, the reasons for such concerns are easy to comprehend, due to the fact that TL is inherently diffi- cult to justify from the standpoint of poverty reduction if it may adversely affect the poor. The remaining question therefore is, to what extent are the poor likely to suffer such adjustment cost in the aftermath of TL?
Various methods have been used to address the issue empirically, including cross-country comparisons; aggregate time series analyses at the country level, and various simulation methods using both partial and general equili- brium analyses. The ongoing methodological work to solve this nexus is generating ever more technically complex studies which are deemed to pro- vide a final answer. Unfortunately, there is still surprisingly little evidence. Goldberg et al. [2004, p.3], Ravallion [2003, p.3], and Winters et al. [2004, p.72], who all survey the recent literature, find that, although the relation- ship between TL and poverty has received considerable attention in recent years, there are almost no studies which test for a direct link to this date. Consequently, one can only review the indirect evidence, which is why Winters et al. [2004, p.74] developed an analytical framework that depicts the various transmission mechanisms through which TL might ultimately affect poor.
1.2 OUTLINE OF THE PAPER
Although “tracing the links between TL and poverty is going to be a de- tailed and frustrating task, for much of what one wishes to know is just un- known,” as Winters et al. [2004, p.74] note, this paper attempts to depict some of the channels that are likely to be important by using the previously mentioned framework. Such an approach has two advantages. First, it al- lows bringing to bear the knowledge about the way TL may affect an econ- omy and its comprising households, and second, it enables one to fit the fragmentary evidences into a single overall picture. Therefore, the structure of the paper is as follows.
The second chapter provides a brief overview of different definition and measurement concepts for TL and poverty that have been used in the recent literature in order to endow the reader with some basic tools. The third chapter considers several reasons and external pressures that led to wide- spread waves of trade policy reforms in DCs. The fourth chapter gives a concise historical background to the process of TL and its impact on eco- nomic development in three LDCs. The fifth chapter sets up the aforemen- tioned formalistic structure, in order to examine systematically the channels through which TL might affect the poor. The sixth chapter contains the con- clusion.
2 DEFINITION AND MEASUREMENT
Winters et al [2004, p.73] find that an important aspect of any analysis of the impacts of TL on poverty is the definition and measurement of both concepts. The reason for seeking an understanding of both concepts is of analytical as well as practical importance. Since ultimately the case for free trade is a theoretical hypothesis, sound but confined to highly restrictive assumptions, some definition and measurement approach is necessary if satisfactory research of the impact of TL on poverty is to be possible. Be- fore turning to discuss the links between TL and poverty it is therefore im- portant to state at the outset what is meant by each of the concepts.
2.1 TRADE LIBERALIZATION
Ravallion [2003, p.3] notes, and others broadly agree (e.g., Deaton2003; Pritchett1991), that researchers “are [typically] precise about these things.” Unfortunately, this is not so true in the popular debate on TL and poverty. Kirkpatrick et al. [2002, p.61] share this view and adds that “there is no concise and generally agreed definition and measurement approach of TL.” The absence of an agreed upon definition, however, makes it difficult to assess and measure the extent to which it has occurred in the past. None- theless, this section reviews several distinct concepts that have been used by researchers in the recent literature in order to determine the recent develop- ment in trade policy reform.
TL has been equated with becoming more “outward-oriented”. This term tends to be interpreted in three broad ways. Dean [1994, p.3] and Pritchett [1991, p.308] subsume that countries may be considered more outward- oriented if their trade policy reforms imply a move toward neutrality, libe- rality, or openness. A move toward neutrality involves equalizing incen- tives, on average, between the exporting and importing sectors. A more lib- eral trade regime is one where the level of intervention has been reduced below mere equalization. And an increase in trade openness is equated with an increase in the importance of trade in the economy. Both agree that due to the fact that in most LDCs one is starting from a position of relatively high levels of protection, any reform which reduces the level of protection (liberality) will also be a move toward neutrality.
Therefore, a broad and acceptable definition of TL refers to any change in trade policy that will reduce the distortion of trade flows caused by govern- ment intervention. Against this background, two types of policy changes are included: (i) price instruments, such as tariffs, duties, import surcharges, import deposits, subsidies to domestic producers of import competing goods, exchange rate manipulation, taxes etc. (ii) nontariff barriers (NTBs), such as import quotas (e.g. voluntary export restraints, orderly marketing agreements), prohibitions, import licenses, special labeling and packaging regulation, health and safety regulations, customs procedures and documen- tation requirements, etc.
In the face of an array of different trade restrictions, it is obvious that it is extremely difficult to measure the true trade orientation of a country’s overall trade policy stance. Anderson [2005, p.3] notes that traditional analysis provides little guidance on how to aggregate different trade restrictions. This impedes to assess the progress made in moving toward less restricted trade. Nonetheless, Pritchett [1991, p.309] and McCulloch [2001, p.14] provide an overview of different measures of TL, which can be classified into incidence-based and outcome-based measures.
Incidence-based measures count the level of tariffs or the frequency of NTBs. They include average tariff rates and nontariff restriction indices(table 1, p. 70). One common shortcoming of Average Tariff Rates, or on a
more general basis of all tariff-based measures, is that they do not reflect the effect of NTBs, which could make tariffs redundant for some products. Nontariff restriction indices, on the other hand, attempt to measure the ex- tent of NTBs. However, they do not pose reliable measures because the re- strictiveness of NTBs differs and their impacts vary across products and countries. Pritchett [1991, p.308] remarks that, although counting the fre- quency of NTBs is a relatively straightforward empirical exercise, mea- surement problems are created by the fact that LDCs apply multiple tariff and NTBS to imports at the same time. Therefore, the mere count of NTBs cannot assess the severity of the distortions as, unlike tariffs, where the bar- rier itself has a natural price metric.
Outcome-based measures, on the other hand, do not focus on the mere value and/or frequency of tariffs or NTBs, but assess the deviation of the observed outcome from the outcome without trade barriers (table 2, p. 70). Pritchett [1991, p.314] distinguishes between trade flow-based or price- based measures. Trade flow-based measures include trade intensities, struc- ture-adjusted trade intensities, Leamer’s openness index, and import pene- tration ratios. However, one shortcoming of all trade flow-based measures is that they are sensitive to assumptions and/or the construction of a counter- factual scenario of what would have happened without trade barriers. Price- based measures include implicit tariff rates, price distortion indices and the spread of the black market premium of exchange rates. The spread of the black market premium of exchange rates is a good proxy for exchange con- trol. However, the IMF [2004, p.559] finds that price-based measures are difficult to construct and limited by data availability.
The review of the wide range of measures of outward-orientation shows that given the complex nature of trade policy, and given that trade is being affected by an array of different trade restrictions it is extremely difficult to construct a single satisfactory indicator of trade orientation. Using crosscountry data, Pritchett [1991, p.308] examines the relationships among Average Tariff Rate, Nontariff Restriction Index, Structure-Adjusted Trade Intensity, Price Distortion Index, and Leamer’s Openness Index. The results (table 3, p.70) suggest disappointingly low correlations.
This raises concern as to whether any of these measures is able to reflect the trade orientation of a country. He concludes that there is no perfect trade policy measure to compare countries at a point in time and even to compare the trade policy of a given county at different points in time. Given the lack of consent on the most appropriate measure of trade orientation, it would be misleading to use one approach. Therefore, it is important to be clear which measure is being used when statements are made about the links between TL and poverty and whether the statements’ validity depends on the choice of measure.
In order to analyze how TL affects poverty, it is equally important to know the main characteristics of those in poverty. Therefore, a crucial aspect of any analysis of the effects of TL is the definition and measurement of poverty itself. Martin [1991, p15] and Deaton [2003, p.3] note that, although there “has been substantial written discussion and some controversy between economists over the most suitable approach,” the lack of agreement is evenly distinct as that concerning TL. However, this section reviews the different concepts that have been used in the literature in order to assess the extent of poverty in the aftermath of trade policy reforms.
Martin [1992, p.15] remarks that, in general, economists have defined po- verty as income insufficient to meet minimum consumption needs. Within this narrow definition, economic literature has mainly focused either on how to define income or on where to draw the line between the poor and non- poor. Despite this rather easy definition, the use of income poverty measures has some weakness.
McCulloch et al. [2001, p.45], for instance, note that people’s experience of poverty is multidimensional and may include such dimensions as: having little or no education, being susceptible to a wide range of shocks, such as price changes, harvest failure, having illness of various kinds, having little or poor quality land, housing, livestock or other productive assets etc. It would be possible to add many more dimensions to the concept of poverty. Each of these dimensions can be considered a form of poverty in its own right. The problem of defining poverty is therefore a problem of finding a suitable synthesis of the many different dimensions.
Although these dimensions are inherently important they are exceedingly hard to measure, which is why Winters [2002, p.1340] and Deaton [2003, p.7] find that the most common approach to assess the extent of poverty is to calculate a measure of household or individual income or consumption expenditure based on data from household surveys. Given such a narrow approach, it is still helpful to have some means of classifying different po- verty measures, which is why the literature often distinguishes between ab- solute and relative poverty measures.
Absolute poverty measures have thresholds, or poverty lines, that remain constant over time (table 4, p. 71). The assumption underlying most abso- lute measures is that there is a measureable subsistence level of income or consumption below which people are deemed poor. Absolute poverty meas- ures include the headcount index, the poverty gap index and Sen’s poverty index. The headcount index is the simplest and still most common poverty measure. It is easily understood and communicated. For certain sorts of po- verty comparisons, such as assessing overall progress in reducing poverty, it may be quite adequate. However, for some purposes, including analyses of the impacts on the poor of specific policies, the headcount index has serious drawbacks. If, for instance, an already poor person becomes even poorer, the headcount index will remain the same. Therefore, the index is insensi- tive to differences in the depth of poverty. From this point of view the po- verty gap index is preferable. In general, Iceland [2006, p.100] finds that the main advantage of absolute poverty measures is that they are conceptually easy to understand and intuitively appealing. However, the main disadvan- tage is that, as standards of living change, generally so do people’s percep- tion of what poverty means.
Relative poverty measures, against this background, attempt to circumvent this problem, because they are based on the notion that poverty is relative to a society’s existing level of economic, social, and cultural development. Therefore, Ravallion [2002, p.38] suggests that they can be defined as com- parative economic deprivation and notes they can take quite different forms.
The most common practice in setting relative poverty lines, however, is to use some proportion of the arithmetic mean or median of the distribution of income or consumption as the poverty line (e.g., many studies have used a poverty line which is set at about 50 percent of the national median). Citro [1995, p.31] points out that one criticism of relative poverty measures is that the choice income or consumption cutoff is arbitrary or subjective, rather than reflecting an objective standard of deprivation. Furthermore, relative poverty measures do not provide a stable target against which to measure the effects of TL because they change in response to increase and decreases in real consumptions levels instead of remaining fixed in real terms.
Which definition and measure of poverty is most appropriate remains the subject of much debate. The two concepts of absolute and relative poverty capture different, but equally important, dimensions of poverty. Macionis [2008, p.266] suggests that relative poverty lines make most sense in countries where absolute deprivation is not the social norm. Therefore, relative poverty lines are often used in rich countries, while research and policy in LDCs almost always uses an absolute poverty line.
Furthermore, McCulloch [2001, p.50] remarks that discussions of TL and poverty often tend to assume that poverty and inequality are the same. How- ever, poverty, as previously mentioned, refers to the failure by the individu- al or household to rise above a given minimally acceptable standard of liv- ing, while inequality refers to the distribution of wellbeing across house- holds. Thus, it is perfectly possible to have a highly unequal society in which no one is poor. Conversely, it is possible to have a very equal society in which a huge number of people are poor. Therefore, the concern in this thesis is with absolute poverty rather than with relative poverty or inequali- ty.
3 REASONS FOR TRADE POLICY REFORMS
There is probably no area in economics where professional opinion is so united. In general, free trade is seen as superior to protection by the vast majority of the economics profession. The attraction of free trade resides at one level in the theoretical elegance of the principle of comparative advan- tage -which, as Paul Samuelson once put it, is the only proposition in eco- nomics that is at once true and non-trivial. Ideas with regard to trade policy and economic development, however, are among those that have changed radically. In the early 1950’s and 1960’s, trade and development theories and policy prescription were based on some widely accepted stylized facts and premises which led to the broad consensus that trade policy for devel- opment should be based on inward-oriented ISI strategies. The contrast with contemporary policies, however, is striking. Recent studies (e.g., Sharer 1998, Subramanian2000; Rajapatirana2000) have shown that during the late 1980s and early 1990s many DCs shifted away from inward- oriented trade policies (based on ISI) to much more outward-oriented trade policies. Milner et al. [2005, p.109] point out that this rush to free trade is somewhat surprising due to the fact that over the years many scholars have emphasized the “durability of the status quo in trade policy1.”
3. 1 ECONOMIC CRISIS
Rajapatirana [2000, p.11] notes that some authors (e.g., Nelson1990; Pa- pageorgiou et al.1990; Rajapatirana2000) have emphasized economic crisis as the spur to TPRs. And indeed, the 1980’s were a decade of crisis, starting with the second oil shock of 1979-1981, a debt crisis that began when Mexico was unable to service its debt, the rise in interest rates as US monetary policy was targeted to reduce double digit inflation, and the en- suing worldwide recession.
Krueger [1997, p.2] remarks that early trade and development theories were based on some widely accepted stylized facts and premises. Among these premises was the belief (then certainly true) that the production structures in DCs’ were heavily oriented toward primary commodity production. It was reasoned that if DCs adopted policies of free trade, their comparative advan- tage would forever lie in primary commodity production and that industria- lization and, hence, development would not take place. This fear was further nourished by the conviction that both the global income and price elastici- ties of demand for primary commodities were low and that export earnings would not grow very rapidly, if they would grow at all. Based on these pre-mises policy prescriptions were derived. Since it was thought that industria- lization was necessary for development and that free trade would leave DCs specialized in primary commodity production, it followed that there had to be investment in new manufacturing industries. However, due to the general conviction that new industries in DCs could not possibly compete with their established counterparts in developed countries (a reasoning closely related to the infant industry argument the industry would have to be protected in its initial phase. This train of thought led to the widespread use of ISI strategies and inward-oriented trade policies with high levels of tariffs and NTBs among DCs.
Although trade policies in LDCs became increasingly restrictive, the 1950’s and 1960’s were a decade of unprecedented economic growth and initially the strategy seemed to work. Latin American countries and some Asian countries experienced a sudden spurt in their growth rates. However, Bordsgaard et al. [2000, p.38] remark that with the passing of time these countries began to stall. The negative examples were the countless cases of infant industries that refused to mature in old age, and spawned inefficien- cies throughout the economy. The increased import dependence accompany- ing ISI strategies caused in many cases severe balance of payments prob- lems during the 1970s. Until then, their rich natural resource endowments of these countries allowed them to keep trade deficits within reasonable limits through primary commodity export earnings that financed the increasing imports of capital equipment and intermediate goods necessary for contin- ued ISI. ISI policies were soon shown to discourage export growth, as pro- ducers were given little incentives to expand capacity more rapidly than domestic demand growth, as international prices were much lower than the prices of the protected domestic market. Krueger [1997, p.6] remarks that the drop in primary commodity prices accentuated the phenomenon. Period- ic balance of payments crisis arose in reaction to overvaluation of the real exchange rates, increased indebtedness, and the failure of export earnings to grow.
At the same time, starting with Taiwan, several East Asian economies began to grow rapidly under trade policies diametrically opposite to those preva- lent under ISI strategies, which became to be known as “export promotion”. Forsyth [2004, p.337] points out that these countries adopted more outward-oriented trade strategies and rigorously reduced the level of trade protection.
The results became to be known as the East Asian Miracle and culminated in spectacular economic growth and tremendous reductions in poverty. Their experiences clearly demonstrated that an outward-oriented trade strat- egy is not only viable, but essential for prospects of rapid growth. Unfortu- nately, it was not until the 1980’s that the importance of the difference be- came unarguable. After the second oil price increase, the worldwide reces- sion, and the accompanying debt crisis, the East Asian NICs rapidly re- sumed growth, whereas other heavily indebted countries were unable to service their debts and were hard hit by events in the international economy.
Krueger [1997, p. 7] notes that research undertaken in attempting to under- stand the impact of the debt crisis on the DCs made it abundantly evident that the debt-GDP ratios were not significantly different between inward and outward-oriented countries. What was significantly different was the debt-export ratios, as the East Asian countries were able to maintain debt servicing and resume growth because of the greater flexibility of their econ- omies. It was generally realized that the rapid growth in demand for foreign imports and the stagnating growth in foreign exchange earnings were not due to the vagaries of the international market, but rather due to the failure of the ISI policies. By the 1980s and early 1990s, most DCs began TL pro- grams and introduced policies to make incentives uniform for production across export and import competing goods.
3. 2 STRUCTURAL ADJUSTMENT PROGRAMS
Binns et al. [2000, p.281] notes that with the onset of the crisis in the early 1980’s, the World Bank and the IMF, respectively, began to stipulate basic policy reforms with the governments of DCs as a condition for granting them credits to overcome their economic crises. These policy reforms are known as structural adjustment policies (SAP) and were based on the per- ception that the crises in DCs were not a temporary phenomenon, but the result of accumulated government failures produced inevitably by their eco- nomic policies. To correct these failures, it was believed, government regu- lations must be reduced, including removal of trade restrictions.
Godo [2005, p.280] notes that the new pattern of though also required new lending instruments. In 1980, the World Bank launched structural adjust-ment lending (SAL), which was program, or non-project, lending for gener-al policy assistance rather than traditional lending for specific project im- plementations. The IMF’s structural adjustment policy began with the estab- lishment of a structural adjustment facility (SAF) and its role toward DCs was to advance stand-by credits in the event of a critical shortage of foreign exchange. In 1974, when non-oilproducing DCs suffered from major defi- cits in the balance of payments, the IMF established an extended fund facili- ty (EFF) to advance medium-term loans. SAF was a step beyond EFF and advanced medium-term loans on condition of policy reforms required for stable macroeconomic management.
SAP signaled support by the World Bank and the IMF for the TPR programs of recipient countries. As such it had the effect of enhancing the credibility of those governments, thereby reducing capital flight. Because IMF consent was indispensible to the rescheduling of external debts, IMF conditionality worked as a device by which foreign creditors could screen debtor countries. Therefore, the lending instruments of both organizations applied significant leverage to promote TL in LDCs.
3. 3 THEORY AND RESEARCH
In addition, theory and research strengthened the intellectual case for con- tinuing and consolidating trade policy reforms in the 1990s. By the late 1960’s and 1970’s, there were significant contributions which undermined some of the premises on which ISI strategies were based. At an analytical level, one line of research focused on whether the stylized facts of “market failure” in fact warranted the imposition of trade restrictions. Bhagwati et al. 1963and Johnson1965, among others, demonstrated that trade restric- tions were nether a first-best, nor often even second-best, instrument for achieving the objectives in the name of which protection had been granted.
As trade regimes became more disordered, empirical work began to docu- ment these problems supported by the development of appropriate mea- surement tools. The Organization for Economic Cooperation and Develop- ment, for instance, sponsored a series of country studies led by Little et al 1970, which provided estimates of effective rates of protection in a num- ber of DCs. These showed how high and indiscriminate protection levelswere and demonstrated the extent to which ISI had failed to achieve many of the objectives set for it.
A later series of country studies undertaken under the auspices of the Na- tional Bureau of Economic Research (NBER), synthesized in works by Bhagwati1978and Krueger1978, provided further systematic empirical evidence of the economic wastefulness and irrationality of inward-oriented trade regimes and empirical research associated with Dollar1992, Winters et al.2004, Sachs and Warner1995, and Edwards1997, among others, built on the work done by Little et al.1970, Bhagwati1978, and Krueger 1978. Furthermore, the creation of endogenous growth models by Romer 1986, Lucas1988, and Grossman and Helpman1991permitted an analysis of policy reforms that had not been possible within traditional trade theory.
4 EXPERIENCES WITH TRADE POLICY REFORMS
Since the mid 1980’s and early 1990’s it is widely accepted that growth prospects for developing countries are greatly enhanced through an out- ward-oriented trade regime. And indeed, some countries have dramatically reduced their trade protection and achieved high rates of growth with out- ward-oriented trade strategies as depicted in Figure 3.3. Collier and Dollar [2002, p.55], for instance, find that reductions in average tariff rates and NTBs have been particularly large in South Asia, Latin America and East Asia, whereas progress in Sub-Saharan Africa and in Middle East and North Africa has been more limited (figure 1, 2, 3, and 4; pp. 71 ). In addition, two recent World Bank [2004, p.3] and IMF [2004, p.48] publications re- veal that the worldwide incidence of poverty has declined. Although global trends in trade policy and economic performance provide useful information on the extent of both, they tell little about what will happen to poverty in any particular country as a result of TL. This section focuses on three LDCs where TL was expected to have had a significant impact on their economic performance, but where obtained results differed to a large extent.
The first country considered is Vietnam, which, as a result first of war and then of US sanctions and its membership in the socialist bloc, was largely isolated from the world economy. However, with the adoption of doi moi in 1986, Kinnvall and Jönsson [2002, p.117] point out that the Vietnamese economy underwent a gradual transition from being a centrally planned socialist economy to a much more market oriented system. As a result, the trade intensity of the economy more than doubled during the late 1980s, and almost doubled again between 1990 and 2002. However, although the Vietnamese government largely reduced restrictions on foreign trade in the mid1990s, the IMF [1999, p.59] still ranked Vietnam as one of the most restrictive trade regimes amidst all its members.
Nonetheless, at the same time, Vietnam’s real growth in GDP has been marked by one of the highest rates in the world since the early 1990s. Be- tween 1990 and 2002, the GDP grew at an average of 7.6 percent per an- num, while exports grew at almost 20 percent per annum. Jenkins [2005, p.8] points out that this rapid rate of economic growth has also been accom- panied by a substantial reduction in the proportion of the population below the national poverty line, which fell from 58 percent in 1992 to 37 percent in 1997 and 29 percent in 2002. This astonishing reduction in poverty represented a decrease of 17 million people in absolute numbers.
4.2 SOUTH AFRICA
In contrast to Vietnam, the experiences in South Africa were quite differ- ent. South Africa before 1994, although by no means a closed economy, could not be regarded as fully integrated with the global economy. The in- creasing isolation of the apartheid regime and the growing economic diffi- culties of the 1980s were reflected in falling trade intensity. In the 1990s, Nissanke [2007, p.167] finds that particularly after the African National Congress came to power in 1994, the country experienced a major shift. These changes partly reflected the ending of sanctions and the acceptance of the new South Africa within the international community. Nonetheless, they were also initiated by government policy. Even before the political chan- geover, the old regime had begun to liberalize trade in a piecemeal way in the early 1990s. Nonetheless, it was really from 1994 onwards that major TPRs were undertaken.
Jenkins [2005, p.5] points out that, although the trade intensity slightly be- gan to rise, South Africa’s growth performance since the early 1990s failed to live up to the high expectations. The real growth in GDP between 1990 and 2002 was just over 2 percent, and even when the very low growth rate during the apartheid regime is taken into account, the growth rate since 1995 has been lower than in the 1960s and 1970s. In addition, Nissanke [2007, p.169] finds that the proportion of the population below the national poverty line fell from 51 percent in 1995 to 48.5 percent in 2002 which, because of population growth, left the total amount of people living below the poverty line virtually unchanged.
A further example is Kenya. Despite having pursued import substitution during the first two decades after achieving independence, Kenya was the most open of the three countries in the 1980’s. However, Jenkins [2005, p.4] finds that in contrast to Vietnam and South Africa, where the trade intensity increased during the 1990s, the share of exports and imports in GDP in Kenya was virtually the same in 2002 as in 1990. This tendency, however, was not a result of the adoption of more restrictive trade policies. Nissanke [2007, p.164] remarks that the government introduced a phased program of tariff reductions and revoked most import licensing schedules from the early 1990s. As a result, almost all administrative controls hampering internation- al trade had been abolished, tariffs had been significantly reduced, export incentives put in place, and exchange rate controls removed.
However, although Kenya became more open in terms of trade policy dur- ing the 1990s, this was not reflected in trade flows. Jenkins [2005, p.5] notes that the lack of success is also reflected in the slow rate of GDP growth, which was less than 2 percent between 1990 and 2002 compared to over 4 percent per annum in the 1980s. Additionally, the overall rate of export growth has also been low at 3.3 percent per annum. Furthermore, data on poverty in Kenya during the 1990s indicate that the share of the population living in poverty increased sharply during the early 1990s, declined during the mid-1990s, and then rose steadily after 1997. In 2001 over 17 million people were living below the national poverty line, making up over 55 per- cent of the total population. This was a rise of more than six percent com- pared to 1990, which represented an increase of 6 million people in abso- lute numbers.
5 TRADE LIBERALIZATION AND POVERTY
If TL and poverty were concepts that could both be easily measured, and if there were many historical instances in which TL could be identified as the main economic shock, it would be easy to derive simple empirical regulari- ties linking both of them. Unfortunately, these conditions do not hold and the previous section has shown that the impact of TL on poverty varies widely from country to country. In addition, Edmonds [2003, p.85] remarks that, while there is extensive research on the impact of TL on income distri- bution, the direct links between poverty and TPRs are only beginning to be explored. As a consequence, there is little evidence on the question whether TL alleviates or exacerbates poverty. Winters et al. [2004, p, 72] point out that any analysis is therefore obliged to decompose the link into steps and to compile the evidence on each of them individually. As noted above, this paper approaches the question of TL and poverty by applying such an ana- lytical framework.
5.1 ECONOMIC GROWTH AND POVERTY
The first step in the framework, according to Winters et al. [2004, p. 74], is to analyze the effects of TL on economic growth. As is widely accepted by economists and policymakers, economic growth is one key factor to the re- duction of absolute poverty; at least in the long run. Although growth can be unequalizing, it has to be very strongly if it is to increase absolute poverty. Nonetheless, it creates the resources to raise incomes, and even in a case where growth is insufficient to alleviate poverty, governments will have scope for redistributive measures. This section identifies the growth path- way as the most critical asking whether TL aids growth and whether growth aids poverty reduction.
5.1.1 ECONOMIC GROWTH
The first link in the previously mentioned chain is from TL to economic growth. The idea that the trade policy of a country has an impact on the country’s economic growth is not new and theory offers many reasons to expect TL to stimulate economic growth. However, there is a difference between the short- and the long- run benefits. In the short-run, the static benefits of trade could look like growth.
1 A status quo bias exists if trade liberalization is welfare enhancing but policy change is not made because the political costs of liberalization outweigh the economic benefits to be gained from it. Particularly with respect to trade policy the status quo bias is often seen to be significant since the benefits of protectionism are highly concentrated while the costs are diffuse.
- Quote paper
- Christian Krüger (Author), 2010, Trade Liberalization and Poverty, Munich, GRIN Verlag, https://www.grin.com/document/153949