Do South-South Trade Agreements Increase Trade? Commodity-Level Evidence from COMESA
  • 1 0000000404811396https://isni.org/isni/0000000404811396International Monetary Fund

Contributor Notes

Author(s) E-Mail Addresses: amm223@georgetown.edu; csteinberg@imf.org

South-South trade agreements are proliferating: Developing countries signed 70 new agreements between 1990 and 2003. Yet the impact of these agreements is largely unknown. This paper focuses on the static effects of South-South preferential trade agreements stemming from changes in trade patterns. Specifically, it estimates the impact of the Common Market for Eastern and Southern Africa (COMESA) on Uganda's imports between 1994 and 2003. Detailed import and tariff data at the 6-digit harmonized system level are used for more than 1,000 commodities. Based on a difference-in-difference estimation strategy, the paper finds that-in contrast to evidence from aggregate statistics-COMESA's preferential tariff liberalization has not considerably increased Uganda's trade with member countries, on average across sectors. The effect, however, is heterogeneous across sectors. Finally, the paper finds no evidence of trade-diversion effects.

Abstract

South-South trade agreements are proliferating: Developing countries signed 70 new agreements between 1990 and 2003. Yet the impact of these agreements is largely unknown. This paper focuses on the static effects of South-South preferential trade agreements stemming from changes in trade patterns. Specifically, it estimates the impact of the Common Market for Eastern and Southern Africa (COMESA) on Uganda's imports between 1994 and 2003. Detailed import and tariff data at the 6-digit harmonized system level are used for more than 1,000 commodities. Based on a difference-in-difference estimation strategy, the paper finds that-in contrast to evidence from aggregate statistics-COMESA's preferential tariff liberalization has not considerably increased Uganda's trade with member countries, on average across sectors. The effect, however, is heterogeneous across sectors. Finally, the paper finds no evidence of trade-diversion effects.

I. Introduction

The number of preferential trade agreements (PTAs) between low-income countries—so-called South-South trade agreements—, has increased dramatically in the last decade. Indeed, between 1990 and 2003, low-income countries signed 70 new PTAs (WTO, 2003). South-South arrangements account for more than 50 percent of all new trade agreements. Important examples of such arrangements include the Southern Cone Common Market (MERCOSUR) in South America and the Common Market for Eastern and Southern Africa (COMESA) in Africa. Countries that are both poor and small frequently enter into PTAs; Africa alone has 30 such arrangements (Yang and Gupta, 2005). Many PTA member countries belong to more than one agreement, resulting in competing demands.

While increasingly popular, South-South PTAs between small countries may not yield significant economic gains for their members. South-South PTAs are more likely to give rise to trade diversion rather than trade creation. In addition, pro-competitive effects for local firms arising from greater competition and dynamic efficiency gains linked to economies of scale are unlikely, as partner countries are usually both poor and small. Moreover, fiscal revenues in low-income countries are more vulnerable to trade reforms (see section III).

The empirical evidence of trade effects in PTAs is mixed (see section II). Papers in this literature, in general, use country-level data, and capture the impact of preferential trade agreements by introducing a PTA dummy variable in a gravity-model framework. The dummy variable, however, is endogenous, since the decision to create or join an agreement is not random. In addition, aggregate data masks commodity-level heterogeneity, which may also bias the estimates.

Clausing (2001) and Romalis (2005) eliminate some of these problems by using commodity-level data to analyze the effects of the North American Free Trade Agreement (NAFTA) and the Canada–United States Free Trade Agreement (CUSFTA). To the best of our knowledge, this paper is the first to apply their empirical strategy to a South-South trade agreement. Specifically, we focus on the static effects of COMESA resulting from changes in trade patterns. By exploiting the variation in the data across commodities, origin countries, and time, we estimate the impact of COMESA-related preferential trade liberalization on Uganda’s imports between 1994 and 2003. We also investigate whether these changes stem from trade creation or trade diversion.

The analysis here focuses on COMESA, as it is a good example of a South-South preferential trade agreement involving small economies. All member countries are truly small in the context of the world economy, and the agreement has been in effect since 1994. Within COMESA, we analyze the impact of preferential liberalization on Uganda’s trade patterns as Uganda represents a relatively stable economy during this time period.

Using a difference-in-difference estimation strategy, we show that reductions in the preferential tariff rate applied by Uganda to other COMESA member countries did not considerably increase Uganda’s imports from such countries. In other words, Ugandan consumers on average across the sectors examined have been reluctant to switch the origin of their purchases to COMESA countries following the advent of the COMESA agreement.

According to our findings, the elasticity of imports with respect to tariff rates is between 14 percent and 16 percent. In addition, the elasticity of substitution between varieties of the same good from different origin countries is approximately 1.7. The magnitude of these effects is relatively small, compared with the results from previous studies for the United States and Canada within CUSFTA and NAFTA (Clausing, 2001; Romalis, 2005). Romalis’s estimate for Mexican imports, however, is closer to our estimate for Uganda. This difference could mean that consumers in low-income countries, in general, have relatively inelastic demand curves and are thus less likely to benefit immediately from trade reform. Search costs may partly explain the reluctance of low-income consumers to switch the origin of their purchases from one country to another.

The results, however, are also consistent with the most important criticism of South-South PTAs—that is, because member countries are not natural trading partners, such agreements are unlikely to produce substantial increases in trade volumes. The finding that COMESA’s effect on Uganda’s imports is heterogeneous across sectors supports this interpretation. In particular, the industries that experienced larger and statistically significant increases in trade volume were those in which developing countries tend to have a comparative advantage.

The elasticity estimates withstand a number of robustness checks. One concern is that COMESA-related reductions in tariff rates might have been offset by an increase in nontariff barriers on the same commodities. For example, after COMESA’s initial implementation, Uganda imposed ad valorem excise taxes on selected goods that tended to be produced by COMESA countries. Such an offsetting effect is unlikely, given that we partially account for nontariff barriers by using data on import excise taxes. Political economy factors are also unlikely to affect the results because our main specification controls for both time-invariant political-economy factors and political-economy factors that change over time that are common across member and nonmember countries. In addition, the findings are not overturned by a triple-difference estimation strategy that controls for factors that change over time and are specific to each import country (a robustness check that follows Romalis, 2005). Lastly, the results grow more robust when we consider the possible impact of tax evasion on recorded imports, as documented by Fisman and Wei (2004).2

Finally, the empirical analysis investigates whether Uganda’s small increase in trade volumes following COMESA reflects trade-creation or trade-diversion effects. We find no evidence that Uganda’s imports from non-COMESA countries shrunk after the start of the agreement. Thus, COMESA’s small but positive effects on trade volumes appear to be associated with trade creation. Notice that this result is not consistent with the expectation in the literature that South-South PTAs imply trade diversion. As a final point, although we conclude that the trade effects are minimal, it is important to note that even small increases could represent a marked improvement for small, low-income countries in Sub-Saharan Africa.

II. Literature

Empirical work on preferential trade agreements is extensive. In general, these studies are either ex ante computable-general-equilibrium (CGE) studies (see Baldwin and Venables, 1995, for a survey of such work) or ex post empirical studies. The ex post analyses can be further divided into studies using aggregate-level data and those using either sector-level or commodity-level data.

The ex post studies drawing on aggregate-level data capture the impact of preferential trade agreements by introducing a PTA dummy variable in a gravity-model framework (e.g., Frankel and Wei, 1995).3 Although these papers generally find that PTAs boost trade volumes, the estimated effects are likely biased due to endogeneity and reverse causality concerns. Such bias mainly arises because the decision to create or join an agreement usually is not random. For example, high trade volumes increase the likelihood that countries will enter into an agreement. To address this concern, Magee (2003) models the PTA dummy variable as endogenous in a gravity-type equation. He finds that, once endogeneity is taken into account, the impact of PTAs on trade patterns is unstable and at times not positive across different specifications.

Many studies on South-South PTAs and on African PTAs, in particular, use the pre-Magee (2003) gravity-type approach and thus may be subject to endogeneity concerns (for example, Cernat 2001; and Subramanian and Tamirisa, 2001). Cernat (2001) finds that COMESA has produced net trade-creation effects with no evidence of trade diversion.4 Subramanian and Tamirisa (2001), however, find a negative block effect for COMESA countries before the formation of the agreement. In 1990, COMESA members traded significantly less goods with each other than did the average pair of countries in the sample. This finding suggests that COMESA countries are not natural trading partners and that the agreement is more likely to lead to trade diversion.5

The second subset of ex post studies employs sector-level and commodity-level trade data to help overcome some of the limitations of the gravity-type approach (Clausing 2001; Krueger 1999, 2000; Romalis, 2005; Yeats 1998a, 1998b). Clausing (2001) estimates the effect of CUSFTA on trade flows from Canada to the United States, and Romalis (2005) estimates the impact of NAFTA and CUSFTA on member countries’ imports using a triple-difference estimation technique.6 Clausing finds no evidence of trade diversion as a result of CUSFTA. Romalis, in contrast, finds evidence of trade-diversion effects on member countries’ imports. In addition, he finds that import demand in the United States and Canada—two large, developed countries—are highly sensitive to tariff movements. By contrast, he finds that import demand in Mexico—a poorer, less-developed nation—are fairly inelastic, consistent with our findings for Uganda. In addition, based on estimated elasticities of total export supply, Romalis finds evidence that NAFTA and CUSFTA had a modest effect on border prices and welfare.

From a methodological viewpoint, our paper is most closely related to Clausing (2001) and Romalis (2005). To the best of our knowledge, we are the first in the literature to apply their empirical strategy to a South-South trade agreement. Our paper is also closely related to recent works in the literature estimating import demand elasticities (Kee, Nicita, and Olarreaga, 2005) and elasticities of substitution (Broda and Weinstein, 2004).

III. Trade Creation and Trade Diversion

The welfare impact of PTAs is unclear. As first stated by Viner (1950), preferential trade liberalization can either result in inefficient, high-cost domestic production being supplanted by low-cost imports from member countries (i.e., trade creation) or in efficient, low-cost imports from nonmember countries being replaced with less-efficient imports from member countries (i.e., trade diversion). Consider the case of a small-open economy: If trade creation occurs as a result of a PTA, the agreement is welfare-improving. If trade diversion occurs, the effect on welfare through changes in trade patterns is unclear.7 In the case of large open economies, terms-of-trade changes make it harder to sign the net welfare effect of PTAs. However, our focus on COMESA, which involves small open economies,8 allows us to abstract from terms-of-trade changes.9

The difference between trade creation and trade diversion is also relevant from a political-economy point of view. Preferential trade agreements that result in trade creation are more likely to be building blocks for multilateral trade negotiations. Indeed, policymakers can build consensus around the visible gains of partial trade liberalization. By contrast, industries characterized by trade diversion—in which imports from PTA member countries replace imports from more efficient nonmember countries—could deter further multilateral free trade efforts. In such industries, the threat of direct competition with more efficient producers in nonmember countries could create greater resistance to global free trade (Krishna, 1998; Krueger, 1999).

The welfare effects occurring through trade creation and trade diversion—as well as through other channels—imply that South-South PTAs involving small countries are the least likely to produce gains for their members for several reasons. First, developing countries typically are not natural trading partners, as evidenced by the fact that they trade little with each other as a share of total imports.10 For example, the 2001 share of African imports from other African countries was approximately 9 percent (IMF, 2002). The reason is that low-income countries tend to have similar relative factors supplies, therefore the incentive to trade with each other is smaller than for dissimilar countries. In other words, developing countries tend to have a comparative advantage in the same sectors; therefore, they generally are not low-cost producers of goods imported by other developing countries. By this reasoning, South-South trade agreements are likely to lead to trade diversion as opposed to trade creation, if any increase in imports occurs at all. From a political-economy point of view, trade diversion in turn implies a stumbling-block effect of South-South trade agreements for multilateral trade liberalization.

Second, low-income and small PTA partner countries are less likely to produce efficiency gains linked to economies of scale and to trigger pro-competitive effects for local producers. The reason is that South-South PTAs offer their members access to smaller markets than do North-South agreements. In addition, firms in PTA member countries with developing economies may not be much more efficient than home firms. Therefore, competitive pressure on domestic producers may not be very strong. Finally, because trade taxes constitute a large proportion of developing countries’ domestic revenues, the loss of tariff revenue may hurt a developing country’s fiscal position more than a developed country’s. In Uganda, for example, tariff revenue declined significantly (by 8 percent of GDP) after the inception of COMESA (Figure 1). For these and other reasons, some researchers think developing countries gain more economically from North-South PTAs than from South-South PTAs (Schiff, 1997; Schiff and Winters, 2003).11

Figure 1.
Figure 1.

Uganda: Imports and Tariff Revenue, 1986–2003

(Percent of GDP)

Citation: IMF Working Papers 2007, 040; 10.5089/9781451866049.001.A001

Source: Ugandan Authorities, DOTS (IMF), and IFS (IMF).

IV. Data

We use commodity-level import and tariff data at the 6-digit Harmonized System level. Import statistics by origin country come from the COMTRADE database, developed by the United Nations Statistics Division. Data on preferential and most favored nation (MFN) tariff rates as well as import excise taxes are from TRAINS, developed by UNCTAD. We access both data sets through the World Integrated Trade Solution (WITS) system, designed by the World Bank.

COMESA is an example of a South-South PTA involving small economies. The treaty establishing COMESA as a preferential trade agreement of Eastern and Southern African states was ratified on December 8, 1994.12 At that date, some COMESA countries, including Uganda, were already part of a regional trade agreement called PTA.13 The data available for Uganda, used in this paper, cover the last year of the PTA agreement (1994) and four years of the COMESA agreement (2000 to 2003). For each of these five years, we merge data on the value of Uganda’s imports, at the commodity level and by country of origin, with data on Uganda’s PTA tariff rates (for 1994), COMESA tariff rates (for 2000 to 2003), and MFN tariff rates (for all years). We also use data on Uganda’s import excise taxes.

Uganda’s data for the five years examined is coded according to three different versions of the Harmonized System (HS) classification: H0 for 1994, H1 for 2000 and 2001, and H2 for 2002 and 2003. We use WITS’s concordance tables and recode all the data following the H0 classification.14 Tariff data is presented according to the HS classification up to the 8-digit level, but disaggregate import values only up to the 6-digit level. We use the simple average tariff rate for each 6-digit level code (averaged over the 7-digit and 8-digit level codes).15 Finally, the tariff rates used in the empirical analysis incorporate information on import excise taxes levied on each product.

Tables 1 through 4 present summary statistics of the main variables used in our analysis. The tables document the extent and patterns of preferential and MFN tariff liberalization in Uganda between 1994 and 2003. The tables also offer a preliminary view of the impact of trade liberalization (preferential and otherwise) on Uganda’s imports.

Table 1:

Summary Statistics for Uganda vis-à-vis COMESA Countries (1994-2003)

article image
Notes: The sample is resticted to commodities for which data on imports from COMESA countries is available for both 1994 and 2003. The tariff rates are adjusted for the existence of import excise taxes. The tariff rate for COMESA countries in 1994 is the average tariff rate faced by COMESA countries in 1994, which uses both PTA and MFN rates. Imports refer to a single 6-digit HS commodity. Preferential tariff rates are set equal to MFN tarfiff rates when no preferential rate is specified.
Table 2:

Summary Statistics for Uganda vis-à-vis non-COMESA Countries (1994-2003)

article image
Notes: The sample is resticted to commodities for which data on imports from non-COMESA countries is available for both 1994 and 2003. This is a subset of the dataset we use, which restricts product codes to commodities that in at least one of the years was imported from COMESA. Tarrif rates are adjusted for the existence of import excise taxes. Imports refer to a single 6-digit HS commodity.
Table 3:

Pattern of Protection in Uganda in 1994 and 2003 under the Preferential Trade Agreements

(PTA in 1994 and COMESA in 2003)

article image
Note: Preferential tariff rates are set equal to MFN tariff rates when no preferential rate is specified. Tariff rates are adjusted for the existence of import excise taxes.
Table 4:

Pattern of Protection in Uganda in 1994 and 2003 vis-à-vis the Rest of the World

(MFN Tariff Rates)

article image
Note: Tariff rates are adjusted for the existence of import excise taxes.

Table 1 shows that tariff rates faced by COMESA countries decreased substantially from 1994 to 2003, from an average preferential tariff rate (across tariff lines) of 11.3 percentage points to an average of 5.5 percentage points. At the same time, the average value of imports of a 6-digit HS commodity from these same countries increased substantially (from US$155,000 to US$289,000). Table 2 shows that MFN tariff rates decreased even more than preferential tariff rates (from 17.9 percentage points in 1994 to 10.2 percentage points in 2003), but were on average higher than preferential tariff rates in both 1994 and 2003. Imports from non-COMESA countries also increased during this time period. The overall evidence on changes in imports, from both COMESA and non-COMESA countries, is consistent with the pattern of total imports (as a percentage of GDP) shown in Figure 1.

Tables 3 and 4 track the distribution of tariff rates (both preferential and MFN rates) between 1994 and 2003. In 1994, the distribution mode of preferential tariff rates is between 5 and 10 percent. In 2003, that same distribution mode is less than 5 percent. However, during the period, the distribution mode of MFN tariff rates remained the same (between 5 and 10 percent).

Finally, Appendices I and II show preferential and MFN tariff rates in 1994 and 2003 by 2-digit 1996 HS codes. As shown in Appendix I, the sectors that experienced the greatest reduction in preferential tariff rates between 1994 and 2003 were “edible fruit and nuts…,” “vegetable plaiting materials…,” “essential oils, etc.; perfumery, cosmetic…,” “prep feathers, down etc..,” and “musical instruments…” Some sectors, including “tobacco and manufactured tobacco substitutes,” saw an increase in the preferential tariff rate due to import excise taxes that generally targeted COMESA-member goods. Appendix II shows that the sectors that experienced the greatest reduction in MFN tariff rates in the same period were “coffee, tea, mate and spices,” “vegetable plaiting materials,” and “raw hides and skins and leather.” Finally, appendix III shows Uganda’s total imports by country of origin in 1994 and 2003 (calculated based on data at the commodity level). Kenya is the largest exporter to Uganda in both years. Imports from other COMESA countries are substantially smaller.

In our empirical analysis, following the previous literature, we ask the following questions: To what extent did Uganda’s imports from COMESA countries increase as a result of COMESA’s preferential trade liberalization? And to the extent such imports did increase, how much of this increase was a result of trade diversion, as evidenced by a reduction in imports from non-COMESA countries? To fully explore both questions, we take a counterfactual approach, as we cannot simply consider the change in imports from COMESA and non-COMESA countries between 1994 and 2003. Instead, we estimate how much trade would have changed in the absence of the trade agreement and net this effect out from our measure, as described in the next section.

V. Empirical Strategy and Results

In this section, we exploit the time, commodity, and origin-country variation in imports and tariffs to identify COMESA’s impact on Uganda’s imports. We first develop a simple model that delivers the estimating equations of our empirical analysis using a methodology closely related to that used by Clausing (2001) and Romalis (2005). We proceed from the simplest to the most sophisticated estimation strategy, reflecting the successive advances in the literature.

We assume that each commodity i is differentiated by country of origin c (Armington assumption).16 Varieties from different origins of the same good are not perfect substitutes; the impact of preferential trade liberalization on trade patterns is captured by the elasticity of substitution between varieties of different origins. The representative consumer in Uganda maximizes the following Cobb-Douglas utility function (at time t) over aggregate consumption of each commodity i, Qit, subject to total expenditure being less or equal to total income Yt:

Ut=ibilgQit,whereibi=1.(1)

Consider a constant elasticity-of-substitution (CES) demand structure over varieties of commodity i coming from each country c at time t:

Qit=[cqictσi1σi]σiσi1,σi>1,(2)

where qict is the quantity demanded in Uganda of commodity i from country c at time t, and σi is the elasticity of substitution between different varieties of commodity i. The optimal demand for each variety is found through maximization of aggregate consumption Qit subject to the following budget constraint:

cqictpicttictgict=Eit,(3)

where pict =pict (aicti) equals the border price of variety c of commodity i at time t, aict equals the marginal cost to produce commodity i in country c at time t, tict is one plus the ad valorem tariff rate applied by Uganda at time t on variety c, and gict ≥ 1 represents iceberg transport costs (i.e., in order to have one unit of variety c of good i at time t, it is necessary to buy gict units), and Eit =bi ∙ Yt gives the total expenditure at time t on commodity i (this follows from (1)). In what follows, we will assume that the elasticity of substitution is equal across commodities (σi =σ, for every i).17 Maximization of (2) subject to (3) results in the following quantity demanded in Uganda of variety c relative to variety c′ of good i :

qictqict=(pictgicttictpictgicttict)σ.(4)

The quantity demanded of variety c is therefore equal to:

qict=(pictgicttict)σEit[c(pictgicttict)(1σ)],(5)

which gives a CIF value (cost including insurance and freight) of:

mictqictpictgict=(pictgictPit)(1σ)tictσEit,(6)

where Pit=[c(pictgicttict)(1σ)]1(1σ) is the price index of good i at time t. Taking logarithms of expression (6), we can derive the first specification of the empirical model:

lgmict=σlgtict+(1σ)lgpict+(1σ)lggict(1σ)lgPit+lgEit.(7)

Expression (7) is the starting point of our empirical analysis. Throughout our analysis, we use pooled yearly data for 1994 and 2000–2003, and measure the first term on the right hand side in expression (7) using two methods. In Table 5, we use the log of (one plus) the preferential tariff rate, as directly implied by (7).18 In Table 6, we use (tict -1), which is the ad valorem tariff rate applied by Uganda to commodity i from country c at time t (taking a first-order Taylor approximation, lgtict(tict1). While the coefficient on the first measure represents the impact of a percentage change of (one plus) the tariff rate, the coefficient on the second measure gives the impact of a percentage-point change. Each column in the two tables labeled by the same number corresponds to the same specification.

Table 5:

Estimates Based on Uganda Imports in 1994, 2000,2001, 2002, and 2003

article image
Notes: Robust standard errors in parentheses. + significant at 10%; * significant at 5%; ** significant at 1%The log of import equals the log of Uganda’s imports from COMESA countries (1994, 2000, 2001, 2002, 2003). The difference of log of imports: COMESA vs. non-COMESA equals the log of Uganda’s relative imports from COMESA vs. non-COMESA countries (1994, 2000, 2001, 2002, 2003). The log of preferential tariff equals the log of (100 plus) Uganda’s tariff rate (1994, 2002, 2001, 2002, 2003). The log preference margin equals the log of (100 plus) Uganda’s preferential tariff rate for COMESA countries (PTA tariff rate (for PTA countries) and customs-duty rate (for non-PTA countries) in 1994; and COMESA tariff rate in 2000, 2001, 2002, 2003) minus the log of (100 plus) Uganda’s customs-duty rate for non-COMESA countries. Commodity dummy variables are set at the 6-digit HS product-code level. Commodity-country dummy variables are for the pairwise combinations of commodities and import-origin countries.
Table 6:

Estimates Based on Uganda Imports in 1994, 2000, 2001, 2002, and 2003

article image
Notes: Robust standard errors in parentheses. + significant at 10%; * significant at 5%; ** significant at 1%The log of import equals the log of Uganda’s imports from COMESA countries (1994, 2000, 2001, 2002, 2003). The difference of log of imports: COMESA vs. non-COMESA equals the log of Uganda’s relative imports from COMESA vs. non-COMESA countries (1994, 2000). The log of preferential tariff equals the log of (100 plus) Uganda’s tariff rate (1994, 2002, 2001, 2002, 2003). The log preference margin equals the log of (100 plus) Uganda’s preferential tariff rate for COMESA countries (PTA tariff rate (for PTA count). Commodity dummy variables are set at the 6-digit HS product-code level. Commodity-country dummy variables are for the pairwise combinations of commodities and import-origin countries.

A. Benchmark Estimators

The first step in our empirical strategy is to estimate native benchmark regressions meant to demonstrate that omitted variables biases are important. In particular, in regression (1) of Table 5, we start by regressing the log of imports on the log of (one plus) the preferential tariff rate, the first term on the right-hand side of expression (7). The implicit assumption in this specification is that the remaining terms are orthogonal to the preferential tariff rate. Next, in regression (2), we augment the first regression with year dummies that capture the impact of time effects that are invariant across product codes (e.g., inflation, growth, etc.). Both estimates of trade liberalization (regressions (1) and (2)) are insignificant. We obtain the same insignificant results in Table 6.

Next, in regression (3) of Table 5, we add dummy variables for 6-digit HS product-codes. This specification assumes that the impact of varieties’ prices (pict) and transport costs (gict) in (7) is captured by commodity and time dummy variables (in addition to idiosyncratic shocks in the error term). It also posits that, controlling for goods’ dummy variables and time effects, the remaining variation in the price index Pit and expenditure Eit is orthogonal to tariff changes. The results of this regression show that the reduction of preferential tariff rates increases imports from COMESA countries. The effect is statistically significant at the 1 percent level. However, the size of the coefficient is not large relative to the coefficient estimated for some other countries in the existing literature (see below).19 Next, in regression (4) (Tables 5 and 6) we replace commodity dummy variables with commodity-by-country fixed effects. This technique allows us to control for time-invariant factors that affected demand for, say, Kenyan but not Malawian mangos, or vice versa. This specification controls for all time-invariant determinants of imports of commodity i from country c, resulting in a true fixed-effect estimation. Clausing (2001) uses a similar estimation strategy for imports by the United States from Canada. The estimates we find are now smaller in absolute value than in regression (3) but still significant at the 5 percent level. The elasticity of substitution (σ) is estimated to equal 1.7, while the elasticity of imports with respect to tariff rates is between 14 percent and 16 percent. In particular, if the ad valorem tariff rate decreases by 100 percent (for example, by 10 percentage points when the tariff rate equals 10 percent), then imports from COMESA countries increase by 16 percent (based on column (4), Table 5). Based on column (4), Table 6 if the ad valorem tariff rate decreases by 10 percentage points, imports increase by 14 percent. The magnitude of these effects is relatively small, compared with the results from previous studies for other countries and agreements. In her analysis of U.S. trade imports from Canada within the CUSFTA, Clausing (2001) finds that a 10 percentage point decrease in tariffs implies a 96 percent increase in imports from Canada. Our estimate of Uganda’s elasticity of substitution is also much smaller than the estimated elasticity for the United States computed by Romalis (2005), which ranges between 6.2 and 10.9.

For Mexican imports, however, Romalis’s estimate ranges between 0.6 and 2.5 and is close to our own for Uganda. Our estimate is also similar in magnitude to the elasticity of import demand for Uganda estimated by Kee, Nicita and Olarreaga (2005) (equal to 1.22).20 This similarity may suggest that consumers in low-income countries, in general, have more inelastic demand curves and are, therefore, less likely to immediately benefit from trade reform. Search costs may help explain the reluctance of low-income consumers to switch the origin of their purchases from one import country to another.

Another interpretation of our estimates is that the small effect on Uganda’s imports of COMESA’s preferential tariff liberalization is due to the South-South nature of the agreement. This reading of the results is consistent with what we find below in Section V.C when we investigate cross-sector heterogeneity.

B. Difference-in-Difference Estimator

The estimation strategy up to this point depends on several assumptions that may not hold. In particular, the price index Pit and expenditure Eit may not be orthogonal to preferential tariff rates, after controlling for commodity (or commodity-by-country) fixed effects and time effects. For example, if commodities with increased expenditure levels Eit (and thus high imports) are protected against preferential tariff reductions, then our coefficient estimate of - σ in regression (7) would be biased toward zero. Another concern is that Pit might be correlated with preferential tariff movements since, by construction, Pit is a function of all tariffs in the sector, including COMESA tariffs. In addition, in Uganda, COMESA and MFN tariff rates were liberalized simultaneously, resulting in a clear correlation between the regressor and Pit.

We next modify our empirical model to address these issues by constructing a difference-indifference estimator, in which the control group is imports from non-COMESA countries. Using expression (6) for CIF imports by Uganda of variety c and of variety c′ of good i at time t, we can calculate the following ratio:

mictmict=(pictgictpictgict)(1σ).(ticttict)σ.(8)

Let c represent COMESA countries and c′ represent non-COMESA countries. The advantage of considering expression (8), which represents Uganda’s relative imports from COMESA to non-COMESA countries, is that the terms Pit and Eit are canceled out from the estimating equation. Expression (8) suggests a new specification of the empirical model. The dependent variable now becomes the logarithm of the ratio of imports from COMESA countries to imports from non-COMESA countries. We regress it on the log of the preference margin afforded by Uganda to preferential trading partners. We calculate the log of the preference margin as the difference between the log of (one plus) the preferential tariff rate and the log of (one plus) the MFN tariff rate. In other words, we estimate the following model (regression (5), Table 5):

lgmictmict=σ(lgtictlgtict)+(1σ)lgpictpict+(1σ)lggictgict,(9)

where c and c′ represent, respectively, the varieties coming from each COMESA member country and from the rest of the world (as a whole). As in regressions (4), we introduce commodity-by-country fixed effects and time dummy variables. Therefore, in this last specification, we only need to assume that the time variation in relative prices and relative transportation costs of two varieties of the same commodity is orthogonal to tariff movements.

Equation (9) clarifies an important point. In the theoretical model, we assume that the elasticity of substitution between varieties of the same good is equal for any pair of origin countries of imports. In practice in the empirical analysis, as made clear by equation (9), the elasticity of substitution we estimate is between COMESA and non-COMESA origin varieties, since we exploit the differential variation in preferential versus MFN tariff rates.

This regression represents our difference-in-difference (and preferred) specification. As mentioned above, this strategy makes it possible to net out the impact of commodity-specific effects that are time-varying, such as Pit and Eit. Thus, our difference-in-difference estimator also allows us to net out the impact of changes in MFN tariff rates that take place over the same period.

Results in Table 5, regression (5), suggest that the biases due to Pit and Eit may not have been substantial, since our new estimate is very close to what we previously found: The coefficient on the log of the preference margin equals −1.9 (significant at the 10 percent level). In Table 6, regression (5), we also estimate this equation using, as an independent variable, (PTAtariffit -MFNtariffit), which is the preference margin afforded by Uganda to preferential trading partners, calculated as the difference between the preferential tariff rate and the MFN tariff rate (as before, we use a first-order Taylor approximation to approximate lgt). The results are similar.

C. Robustness Checks

We next test the robustness of these results (see Table 7). First, in regression (1) of Table 7, we expand the dataset. Some COMESA countries increased exports from zero to a positive value in a specific product code or vice versa. In the former case, by excluding this variation, our previous regression estimates would be biased toward zero. Therefore, whenever import data exists for at least a single year but not the other years, we add observations for the missing year(s), and assign them an import value of US$1. Results in column (1) suggest that the exclusion of these observations in Tables 5 and 6 did not bias our estimate toward zero.

Table 7:

Robustness Checks

article image
Notes: Robust standard errors in parentheses. + significant at 10%; * significant at 5%; ** significant at 1%The difference of log of imports: COMESA vs. non-COMESA equals the log of Uganda’s relative imports from COMESA vs. non-COMESA countries (1994, 2000, 2001, 2002, 2003). The diff- in - diff: COMESA vs. non-COMESA, Uganda vs. South Africa equals the log of relative imports from COMESA vs. non-COMESA countries in Uganda vs. South Africa (1994, 2001). The log preference margin equals the log of (100plus) Uganda’s preferential tariff rate for COMESA countries (PTA tariff rate (for PTA countries) and customs-duty rate (for non-PTA countries) in 1994; and COMESA tariff rate in 2000, 2001, 2002, 2003) minus the log of (100 plus) Uganda’s customs-duty rate for non-COMESA countries. The ratio of preference margin: Uganda vs. South Africa is the log difference between the preference margin in Uganda and the preference margin in South Africa. Commodity dummy variables are set at the 6-digit HS product-code level. Commodity-country dummy variables are for the pairwise combinations of commodities and import-origin countries. Broad HS codes are defined in Appendix I.

Second, we relax the assumption that the elasticity of substitution is constant across product codes and run regressions that are specific for each one-digit HS code (see appendices I and II for a list of two-digit codes included in each one-digit code). Estimates of the elasticity of substitution are insignificant for each one-digit sector except HS1, HS2, and HS3 (which include agricultural products and beverages). For these sectors, we estimate elasticities that are substantially higher than the average. We draw two conclusions from this exercise. First, our previous average estimates hide cross-sector heterogeneity. Second, and not surprisingly, the sectors where the impact is larger and significant are those where developing countries are more likely to have a comparative advantage. These results are presented in regressions (2) through (4) in Table 7.

Finally, we address the possibility that the relative price (pictpict) term in equation (9) might be correlated with the preference margin, even after controlling for commodity (or commodity-by-country) fixed effects and time effects, as done in regression (5), Table 5. Our third robustness check attempts to control for this bias, which is, for example, due to unobserved changes in the marginal cost of production of commodity i in country c (affecting the border price) that may be correlated with tariff movements. For example, production of beer in Kenya might have become more efficient relative to non-COMESA countries, and this increased efficiency might be negatively correlated with preferential concessions for political-economy reasons (e.g., the excise taxes on alcohol). This would bias our estimate towards zero.

Expression (8) above refers to Uganda’s relative imports (from COMESA versus non-COMESA countries). Based on the same model, we can derive a very similar expression for any other country’s relative imports from (the same) COMESA versus (the same) non-COMESA countries. In the following expression, we consider South African imports:21

mictSAmictSA=(pictgictSApictgictSA)(1σ).(tictSAtictSA)σ(10)

We then use this expression to construct our triple difference estimating equation, where the dependent variable is the logarithm of Uganda’s imports from COMESA countries relative to non-COMESA countries (expression (8)) divided by South Africa’s imports from COMESA countries relative to non-COMESA countries (expression (10)):

lgmictmictlgmictSAmictSA=(11)σ[(lgtictlgtict)(lgtictSAlgtictSA)]+(1σ)(lggictgictlggictSAgictSA)

This specification nets out the impact of the relative border-price term, which is independent of the identity of the importing country (pictpict appears in both equations (8) and (10) and is canceled out by taking their ratio). However, as in the previous specifications, we still need to assume that the relative transport-costs term is given by the sum of commodity-by-country fixed effects, time dummy variables, and a random component orthogonal to the preference margin.

This is the approach taken by Romalis (2005). The last column in Table 7 shows the results based on equation (11). The estimated elasticity of substitution is insignificantly different from zero. If this result is due to a true zero elasticity of substitution, then our previous estimates were not underestimating the impact of COMESA on imports; however, if the insignificance of the elasticity is caused by the imprecision of the estimate (e.g., due to few observations), then we cannot draw strong conclusions from this robustness check. In addition, because this robustness check is based on a much smaller number of observations than previous specifications, the estimate might be affected by a selection-bias problem. We check for this problem in regression (5), which delivers a coefficient estimate that is not statistically different from our original estimate of –1.93.

D. Trade Diversion

Our last test is for trade diversion. This test is important to make a welfare statement about the impact of the trade agreement. Our investigation is based on the fact that, if trade diversion resulted from the PTA agreement, holding all other factors constant, we would expect a decline in imports from non-COMESA countries in those sectors in which preferential tariff rates decline. Our empirical strategy relies on expression (7) above implemented for imports from non-COMESA countries. Results are presented in Table 8.

Table 8:

Estimates of Trade Diversion Based on Uganda Imports from Non-COMESA Countries in 1994, 2000, 2001, 2002, and 2003

article image
Notes: Robust standard errors in parentheses. + significant at 10%; * significant at 5%; ** significant at 1%The log of import equals the log of Uganda’s imports from non-COMESA countries (1994, 2000, 2001, 2002, 2003). The log of MFN tariff equals the log of (100 plus) Uganda’s tariff rate (1994, 2002, 2001, 2002, 2003). The log of COMESA imports equals the log of Uganda’s imports from COMESA countries (1994, 2002, 2001, 2002, 2003) in that commodity. The COMESA share in imports equals the share of Uganda’s imports from COMESA countries (1994, 2002, 2001, 2002, 2003) in that commodity. Commodity dummy variables are set at the 6-digit HS product-code level. Commodity-country dummy variables are for the pairwise combinations of commodities and import-origin countries.

The first column presents the results from the regression of the log of non-COMESA imports on the log of the MFN tariff rate. The equation includes commodity-by-country dummies and year effects; therefore, it is equivalent to the fourth regression in Table 5 for imports from COMESA countries. The number of observations is more than 62,000, accounting for the much higher share of non-COMESA imports in total imports to Uganda. The coefficient is also small, significant, and is consistent with the results in Table 5. That is, the estimates for the elasticity of substitution between COMESA and non-COMESA countries’ origin goods, measured using data for imports either from COMESA or non-COMESA countries, are similar.

To test for trade diversion effects, we include the log of the preferential tariff rate in regression (2) to capture the impact of COMESA trade liberalization on non-COMESA imports, which, according to the model, works through Pit. The coefficient on the latter variable is insignificantly different from zero, thus, giving no support to the trade-diversion hypothesis. Trade diversion, however, may occur only in sectors in which COMESA has a comparative advantage. In regressions (3) and (4), to control for this factor, we include as regressors the log of COMESA imports and the COMESA share in imports, respectively, and their interaction with preferential tariff rates. All trade diversion variables remain insignificant. Finally, we find additional evidence consistent with no trade diversion taking place in Figure 2, which shows that the ratio of imports from COMESA relative to non-COMESA (developing) countries decreased after 1994.

Figure 2.
Figure 2.

Uganda: Imports from Developing Countries, 1986–2003

(Percent of GDP)

Citation: IMF Working Papers 2007, 040; 10.5089/9781451866049.001.A001

Source: Ugandan Authorities, DOTS (IMF), and IFS (IMF); Non-COMESA countries are only non-COMESA developing countries. The ratio is COMESA imports to non-COMESA imports.

Therefore, although COMESA’s preferential tariff liberalization has not considerably increased Uganda’s trade with member countries, these small effects are likely to be associated with trade creation. This result is inconsistent with the expectations in the literature that South-South PTAs give rise to trade diversion.

VI. Conclusions

This paper presents evidence that South-South trade agreements create positive but minimal economic gains for member countries. In particular, using commodity-level data, it finds that Ugandan imports of goods from COMESA countries increased only slightly following the advent of COMESA. Notably, commodity-level data offer a different picture of the effect of COMESA than do aggregate-level data (see the summary statistics in Table 1 and Cernat, 2001, who uses a gravity-type analysis).22

Our estimates are similar to Romalis’s (2005) finding for Mexico within NAFTA. This similarity may indicate that low-income-country consumers generally have more inelastic demand curves than high-income-country consumers, and are thus less likely to immediately benefit from trade reform. Search costs may help explain low-income consumers’ reluctance to switch the origin of their imports. An alternative explanation, however, is that developing countries are not natural trading partners, owing to their size and similar resources.

Our elasticity estimates withstand a number of robustness checks. One concern is that COMESA-related reductions in tariff rates might have been offset by an increase in nontariff barriers. For example, after COMESA’s initial implementation, Uganda imposed ad valorem excise taxes on selected goods that tended to be imported from COMESA countries. Such an offsetting effect is unlikely, given that we partially account for nontariff barriers by using data on import excise taxes. Political economy factors are also unlikely to affect the results because our main specification controls for both time-invariant political-economy factors and political-economy factors that change over time that are common across member and nonmember countries. In addition, the findings are not overturned by a triple difference estimation strategy that controls for factors that change over time and are specific to each import country (a robustness check that follows Romalis, 2005). Lastly, the results grow more robust when we consider the possible impact of tax evasion on recorded imports, as documented by Fisman and Wei (2004).

The results of this paper suggest two important questions for future research. First, if economic gains are minimal, what other factors might explain the increased popularity of South-South PTAs? One explanation may be that such arrangements promote noneconomic benefits, such as peace and security within a region—a goal that is an official priority of COMESA. Indeed, Martin, Mayer, and Thoenig (2005) show that regional trade agreements can reduce the probability of war between liberalizing countries, while multilateral liberalization can potentially increase it. Second, from a normative point of view, given the limited capacity of institutions in the South, are resources efficiently spent in the negotiation and implementation of South-South trade agreements? Such an analysis would better inform efforts to promote trade in developing countries where institutions are weak and resources scarce.

Appendix I: Uganda’s Preferential Tariff Rates by 2-digit 1996 HS Codes, 1994 and 2003

article image
article image

Appendix II: Uganda’s MFN Tariff Rates by 2-digit 1996 HS Codes, 1994 and 2003

article image
article image

Appendix III: Uganda’s Imports by Country of Origin, 1994 and 2003

(Based on data at the commodity level)

article image
article image
article image
article image