Trade Agreements Without Structural Transformation: Evidence on Export Upgrading in Latin America

Introduction

Over the past three decades, Latin American economies have increasingly relied on trade liberalization as a central strategy for economic development and international integration. Regional and bilateral trade agreements such as MERCOSUR, the Andean Community, and the Pacific Alliance have sought to reduce trade barriers, expand market access, and strengthen economic cooperation across the region. As a result, Latin America has experienced a significant expansion of trade agreements and export-oriented policies designed to enhance competitiveness in global markets.

Economic theory suggests that trade liberalization can promote productivity growth and technological upgrading by allowing firms to access larger markets and exploit comparative advantages more efficiently. Empirical evidence shows that trade agreements can significantly increase trade flows among member countries and facilitate economic integration (Florensa et al., 2015; Márquez-Ramos et al., 2017). However, the broader development effects of these agreements remain debated. While some studies emphasize their potential to stimulate export growth and structural transformation, others highlight risks such as trade diversion and uneven benefits across countries (Barboza & Trejos, 2010). Moreover, research suggests that the gains from trade liberalization depend critically on domestic economic conditions, including export diversification and institutional quality.

These structural factors have an important impact on Latin America. In fact, despite the proliferation of trade agreements, many economies in the region remain highly dependent on a limited number of export products, often concentrated in commodities or low-technology manufacturing sectors. Such specialization may limit the capacity of trade agreements to generate sustained productivity growth and export upgrading. To address these challenges, governments have implemented complementary policy instruments such as export processing zones (EPZs), which aim to attract export-oriented investment by offering fiscal incentives and simplified regulatory frameworks. These developments raise important questions about the mechanisms through which trade agreements influence export performance in Latin America. While trade agreements may expand market access, their effectiveness in promoting export upgrading may depend on domestic productive structures and institutional environments. Understanding these conditions is essential for assessing whether trade integration policies can effectively contribute to long-term economic development. This study therefore addresses two main research questions: Do trade agreements contribute to export upgrading in Latin American economies, and does this effect depend on the degree of export diversification? Do structural characteristics such as export diversification, institutional quality and informality influence the effectiveness of trade agreements in promoting export upgrading? By addressing these questions, this study contributes to the literature on trade integration and development by examining the structural conditions under which trade agreements may foster technological upgrading in Latin America.

Literature Review

Traditional customs union theory evaluates trade agreements through trade creation and diversion. However, recent trade and development literature emphasizes that their relevance depends on productive transformation rather than trade volume.

Moreover, firm-level evidence shows that liberalization can induce upgrading. Using Argentina manufacturing data during MERCOSUR tariff reductions, Bustos (2011) finds that improved market access increased export entry and technology adoption. This mechanism aligns with heterogeneous-firm trade models where openness reallocates resources toward more productive firms.

Yet macro evidence suggests this transformation is not automatic. Studies of Latin American export performance (Márquez-Ramos et al., 2017) and cross-country growth analyses (Barboza & Trejos, 2010) show that openness raises growth mainly when it changes export composition. Sectoral analyses also find specialization persistence: agricultural integration tends to reinforce comparative advantage rather than diversify (Josling et al., 2015), while OECD structural analyses attribute this to weak productive capabilities, low investment and informality. Thus, trade agreements create upgrading opportunities but do not necessarily generate structural change.

A large gravity-model literature evaluates trade agreements’ impact on trade flows. Surveys of Latin American agreements (Dahrawy & Timini, 2021) conclude agreements increase trade on average but with strong heterogeneity across countries. Evidence from MERCOSUR confirms this pattern. Using a first-difference gravity model, Ferreira (2019) finds initial trade creation followed by declining effects over time. Similar attenuation appears in Pacific Alliance analyses, where gains are concentrated in pre-existing sectors (Delgado-Martínez, 2024). Historical studies of the Andean Pact report comparable trajectories of early expansion without deep structural integration.

Political economy explanations reinforce these findings. Bohara, Gawande & Sanguinetti (2004) show tariff changes respond to lobbying pressures, implying liberalization may protect politically important sectors. Institutional and fiscal studies in Latin America further show policy credibility and tax capacity condition integration outcomes. Overall, the literature agrees that trade agreements increase trade volumes but produce heterogeneous development effects.

To explain these heterogeneous outcomes, several papers distinguish intensive and extensive margins of trade. Using bilateral panel data, Florensa et al. (2015) show integration mainly increases the intensive margin, while institutional quality drives the extensive margin, the entry of new products and markets.

MERCOSUR trade expansion largely occurs within existing industries (Nagarajan, 1998), while energy and environmental trade analyses reveal sector-specific expansion rather than broad diversification. Agricultural integration studies also document persistent specialization (Josling et al., 2015).

These findings complement firm-level evidence (Bustos, 2011): only some firms upgrade, so aggregate trade can grow without broad structural transformation. Trade agreements therefore often produce enclave-type integration — more exports, but not more diversification.

A growing literature links diversification directly to development outcomes. Cross-country panel studies (Barboza & Trejos, 2010) show export diversification mediates the relationship between openness and growth. Similarly, Florensa et al. (2015) and Márquez-Ramos et al. (2017) show institutional quality and logistics determine export variety, not only export volume. Combined with micro evidence (Bustos, 2011) and structural analyses (Daude et al., 2010) these studies suggest a causal chain: trade agreements → firm upgrading → diversification → development. However, most papers analyze only one link of this chain rather than the full mechanism.

Institutional quality consistently explains divergent outcomes across Latin America. Governance quality affects export entry (Florensa et al., 2015), infrastructure shapes trade performance (Márquez-Ramos et al., 2017), and fiscal capacity influences policy credibility. OECD analyses emphasize informality limits productivity upgrading and firm participation in export markets.

Political economy research (Bohara et al., 2004) and historical integration experiences in the Andean Pact further show agreements can preserve rents instead of transforming production structures. Trade agreements therefore function as conditional policies whose effects depend on domestic economic structures.

Despite a vast empirical literature on trade agreements in Latin America, existing studies mainly focus on average trade effects such as trade creation, export growth, or foreign direct investment attraction. Much less attention has been devoted to the conditions under which these agreements translate into structural transformation of the productive system. In particular, this study identifies the existence of a gap and points to the relative disorganization when it comes to studying at the intersection of trade liberalization, export diversification, and domestic institutional characteristics. While some contributions analyze diversification outcomes and others emphasize the role of institutions or informality, the literature rarely integrates these mechanisms within a unified empirical framework. As a result, it remains unclear why similar trade agreements generate heterogeneous development outcomes across countries in the region. Addressing this gap requires examining whether the developmental impact of trade agreements depends on productive structure and institutional environment rather than on liberalization alone. Therefore, this study seeks to provide new empirical evidence by testing the following hypotheses:

H1. The effect of trade depth on export upgrading is significantly stronger in countries exhibiting higher levels of ex-ante export diversification.

H2. The joint effect of trade agreements and diversification on upgrading materializes only when institutional quality exceeds a minimum threshold and labor informality remains relatively low.

H3. The effect of trade depth on export upgrading is jointly conditioned by domestic structural characteristics and by the degree of trade integration.

Data Collection

To evaluate the hypotheses presented above, this study employs a panel data framework combining country-level indicators of trade agreements, export diversification, institutional quality and productive structure across Latin American economies. The objective is to identify whether trade agreements generate export upgrading and productivity gains, and more importantly whether these effects depend on the structural characteristics of the economies involved. The dataset was constructed using secondary sources and focuses on identifying the most complete and recent data available for Latin American countries. Information was collected from several international databases, including the World Bank World Development Indicators, UN Comtrade, ILOSTAT, the World Governance Indicators and international trade datasets.

The analysis relies on a dataset consisting of a group of 17 Latin American countries over the period 2007–2022, therefore allowing the identification of both cross-country differences and temporal dynamics in trade performance

Following this framework, the baseline empirical specification evaluates the general relationship between trade agreements and export upgrading:

$$\small{Upgrade𝑖𝑡=α+β1TradeDepth𝑖𝑡+γX𝑖𝑡+μ𝑖+λ𝑡+ϵ𝑖𝑡}$$

where 𝑖 denotes the country and 𝑡 the year. Upgrade𝑖𝑡 measured as the share of high-technology exports in total manufactured exports. It represents the degree of export upgrading, TradeDepth𝑖𝑡 is measured as the number of policy provisions included in a trade agreement, capturing the extent to which agreements go beyond tariff reductions to cover a broader set of regulatory areas, including both WTO-related and extra-WTO provisions; and where X𝑖𝑡 include GDP per capita as an additional control variable to capture broader macroeconomic conditions affecting export performance. The model also includes country fixed effects (μ𝑖) to control for time-invariant structural differences across countries and time fixed effects (λ𝑡) capturing global shocks affecting all economies simultaneously.

While this baseline model measures the average effect of trade agreements, the hypotheses developed in the literature review emphasize that these effects are conditional on domestic productive structures. To capture these mechanisms, the analysis introduces interaction terms between trade agreements and key structural variables.

The first hypothesis evaluates whether the effect of trade depth depends on the degree of export diversification:

$$\small{\begin{aligned} Upgrade𝑖𝑡 &= α+β1TradeDepth𝑖𝑡+β2Diversification𝑖𝑡\\ & \quad +β3(TradeDepth𝑖𝑡×Diversification𝑖𝑡)\\ & \quad +γX𝑡+μ𝑖+λ𝑡+ϵ𝑖𝑡 \end{aligned}}$$

where Export diversification (Diversification𝑖𝑡) is measured using the inverse of the Herfindahl–Hirschman Index (HHI) which measures the export concentration. The diversification index is calculated as: Diversification𝑖𝑡=1−HHI𝑖𝑡, where higher values indicate a more diversified export structure.

In this specification, the coefficient β3 captures whether trade depth have a stronger upgrading effect in countries exhibiting higher levels of export diversification. A positive and statistically significant coefficient would support the hypothesis that diversified economies are better positioned to translate trade liberalization into productive transformation.

The second empirical specification incorporates institutional quality and labor informality as conditioning factors:

$$\small{\begin{aligned} Upgrade𝑖𝑡 &= α+β1TradeDepth𝑖𝑡+β2Diversification𝑖𝑡\\ & \quad +β3Institutions𝑖𝑡+β4Informality𝑖𝑡\\ & \quad +β5(TradeDepth𝑖𝑡×Diversification𝑖𝑡×Institutions𝑖𝑡)\\ & \quad +γX𝑖𝑡+μ𝑖+λt+ϵ𝑖𝑡 \end{aligned}}$$

where informality measures the share of economic activity occurring outside the formal sector, which may limit firms’ access to credit. Institutions, measured by Government Effectiveness, captures perceptions of the quality of public services, the quality of the civil service and the degree of its independence from political pressures, the quality of policy formulation and implementation, and the credibility of the government’s commitment to such policies. Estimate gives the country’s score on the aggregate indicator, in units of a standard normal distribution, i.e. ranging from approximately -2.5 to 2.5.

This model tests whether the joint effect of trade agreements and diversification depends on the institutional environment. Weak governance structures and high levels of informality may prevent firms from upgrading technologically and participating effectively in export markets, thereby limiting the developmental effects of trade liberalization.

Finally, the third specification adds Trade Openness to the precedent model:

$$\small{\begin{aligned} Upgrade𝑖𝑡 &= α+β1TradeDepth𝑖𝑡+β2Diversification𝑖𝑡\\ & \quad +β3Institutions𝑖𝑡+β4Informality𝑖𝑡\\ & \quad +β5TradeOpenness\\ & \quad +β6(TradeDepth𝑖𝑡×Diversification𝑖𝑡×Institutions𝑖𝑡)\\ & \quad +γX𝑖𝑡+μ𝑖+λt+ϵ𝑖𝑡 \end{aligned}}$$

Here, Trade openness, measured as the ratio of total trade to GDP, reflects the extent to which economies are embedded in global markets and therefore their capacity to leverage external demand and competitive pressures to support structural transformation.

This specification evaluates whether the effectiveness of trade agreements and domestic structural characteristics in promoting export upgrading depends on the degree of trade integration. Higher levels of trade openness may enhance the capacity of economies to translate trade agreements into upgrading outcomes, although this effect remains contingent on domestic structural constraints.

Empirical Analysis

This section presents the empirical estimation of the models described above and provides evidence to assess the hypotheses regarding the relationship between trade agreements, export diversification, and export upgrading in Latin America. The analysis relies on a panel dataset covering 17 Latin American countries over the period 2007–2022 and employs fixed-effects estimations to control for unobserved heterogeneity across countries and time.

All continuous variables are mean-centered prior to estimation in order to facilitate the interpretation of interaction terms. Standard errors are clustered at the country level to account for potential serial correlation and heteroskedasticity within panels.

Baseline Results

We begin by estimating the baseline specification, which evaluates the average effect of trade integration on export upgrading. The results indicate that trade depth does not exhibit a statistically significant effect on export upgrading. Similarly, GDP per capita, included as a control variable capturing macroeconomic conditions, does not appear to significantly influence upgrading outcomes.

These findings are consistent with the broader literature suggesting that trade agreements alone are insufficient to generate structural transformation. While trade integration may increase trade volumes, its impact on export composition and technological upgrading remains limited in the absence of supportive domestic conditions. This aligns with previous empirical evidence showing that trade openness does not systematically translate into productivity gains or structural change.

Diversification and Conditional Effects

The second specification introduces export diversification as a key conditioning variable. The results show that neither diversification nor trade depth individually have a statistically significant effect on export upgrading. Moreover, the interaction term between trade depth and diversification is also not statistically significant.

These findings suggest that higher levels of diversification do not automatically enhance the capacity of trade agreements to promote upgrading. In other words, even relatively diversified economies in Latin America do not appear to systematically convert trade integration into higher shares of high-technology exports.

This result provides empirical support for the argument that diversification alone is not sufficient to generate structural transformation. As highlighted in the literature, trade agreements tend to reinforce existing comparative advantages rather than induce diversification or technological upgrading (Florensa et al., 2015; Márquez-Ramos et al., 2017). Therefore, the hypothesis that diversification amplifies the effect of trade agreements on upgrading is not supported by the data.

Institutional constraints and Structural interactions

To further explore the mechanisms underlying these results, the third specification introduces institutional quality and informality as additional conditioning factors. The model includes a triple interaction between trade depth, diversification, and institutional quality in order to capture whether upgrading effects depend on the joint presence of favorable structural conditions.

The results indicate that institutional quality has a positive but statistically insignificant effect on export upgrading, while informality exhibits a negative effect that is marginally significant at the 10% level. This suggests that higher levels of informality may hinder the capacity of firms to engage in technological upgrading, likely due to limited access to finance, reduced productivity, and weaker integration into formal export markets.

Importantly, the triple interaction term between trade depth, diversification, and institutional quality is negative and marginally significant in this specification. This indicates that even when these structural dimensions are considered jointly, the effect of trade agreements does not become more favorable; if anything, it tends to move in the opposite direction.

These findings reinforce the view that structural constraints in Latin America, particularly informality and institutional weaknesses, limit the effectiveness of trade agreements. While trade integration may create opportunities, these opportunities are not automatically translated into productive transformation.

Extended Model: Trade openness and dynamics

The final specification extends the previous model by incorporating trade openness as an additional control variable. This allows us to distinguish between the effects of formal trade agreements (trade depth) and broader integration into international markets.

The results reveal several important patterns. First, trade openness does not exhibit a statistically significant direct effect on export upgrading. Second, the negative effect of informality becomes more pronounced, suggesting that structural rigidities continue to play a key role in limiting upgrading processes, although this relationship remains only marginally significant.

Most importantly, the triple interaction between trade depth, diversification, and institutional quality becomes negative and statistically significant at the 5% level. This result indicates that, even under more favorable structural conditions, trade agreements do not generate positive upgrading effects. On the contrary, the combined effect appears to be negative.

This finding represents a central contribution of this study. It suggests that trade integration in Latin America does not lead to structural transformation, even when economies exhibit higher levels of diversification and stronger institutional frameworks. Rather than fostering technological upgrading, trade agreements may reinforce existing production structures or lead to specialization patterns that are not conducive to upgrading.

Interpretation and discussion

Taken together, the empirical results provide consistent evidence that trade agreements in Latin America do not systematically promote export upgrading. Across all specifications, trade depth does not exhibit a significant positive effect on the share of high-technology exports.

Moreover, the results highlight the importance of domestic structural conditions. Informality appears as a relevant constraint, displaying a negative and marginally significant effect on upgrading outcomes. Institutional quality, while positively associated with upgrading, does not appear sufficient to generate a significant improvement in interaction with trade agreements when taken on its own.

The most striking result concerns the negative and significant triple interaction observed in the extended model. This suggests that even when economies combine higher levels of diversification and better institutional environments, trade agreements do not produce the expected upgrading effects. This finding challenges the conventional view that trade liberalization, combined with favorable domestic conditions, leads to structural transformation.

Instead, the results support the hypothesis that trade agreements in Latin America primarily increase trade volumes without altering the underlying production structure. This is consistent with the notion of “enclave-type integration” identified in the literature, where trade expansion occurs within existing sectors rather than through the development of new, higher-value activities.

Implication for hypotheses

The empirical evidence allows us to evaluate the hypotheses presented in Section II.

• H1 (Diversification hypothesis): Not supported. Export diversification does not significantly enhance the effect of trade depth on upgrading.

• H2 (Structural conditioning hypothesis – institutions and informality): Partially supported. Informality is negatively associated with upgrading, but only at the 10% significance level, while institutional quality alone does not generate a robust effect. The joint interaction between trade depth, diversification, and institutional quality is negative, suggesting that favorable structural conditions do not translate into stronger upgrading outcomes.

• H3 (Joint structural effect hypothesis): Not supported in the expected direction. Trade openness does not significantly improve export upgrading. However, the negative interaction becomes more robust when controlling for overall integration. The combined effect of trade depth, diversification, and institutional quality remains negative and statistically significant at the 5% level, indicating that broader integration does not reverse the limited developmental impact of trade agreements.

Summary of findings

Overall, the empirical analysis reveals that trade agreements in Latin America do not generate export upgrading, regardless of the level of diversification or institutional quality. While trade integration increases market access and trade flows, it does not appear to induce the structural transformation required for technological upgrading.

These findings contribute to the literature by providing evidence that the developmental impact of trade agreements is highly conditional and, in the case of Latin America, largely limited. Trade agreements alone are not sufficient to alter production structures, and their effectiveness depends critically on deeper structural reforms that go beyond trade policy.

Conclusions

This paper set out to examine whether trade agreements contribute to export upgrading in Latin America and whether their effects depend on domestic structural conditions. Using a panel dataset covering 17 countries over the period 2007–2022 and applying fixed-effects estimations, the analysis reveals a clear and consistent pattern: trade agreements have not delivered the expected boost in export upgrading.

Across all specifications, trade depth does not exhibit a statistically significant positive effect on the share of high-technology exports. This result remains robust when accounting for export diversification, institutional quality, and trade openness. More importantly, the most complete specification shows that the combined effect of trade depth, diversification, and institutional quality is negative and statistically significant. This finding indicates that even under relatively favorable structural conditions, trade agreements fail to generate technological upgrading.

These results challenge the common assumption that deeper trade integration, combined with sound domestic fundamentals, is sufficient to promote structural transformation. Instead, the evidence points toward a pattern of what can be described as “enclave integration,” where trade volumes expand while the underlying productive structure remains largely unchanged. In this context, trade agreements appear to reinforce existing specialization patterns rather than facilitate a transition toward higher value-added activities.

The analysis also highlights the central role of domestic constraints. Informality emerges as a key factor limiting upgrading, suggesting that structural weaknesses prevent firms from fully benefiting from new market access. While institutional quality is positively associated with upgrading, it does not appear sufficient on its own to unlock the transformative potential of trade agreements. Taken together, these findings suggest that the effectiveness of trade integration is highly conditional and depends critically on deeper domestic factors.

The analysis also highlights the role of domestic constraints. Informality appears as a structural limitation, suggesting that weaknesses in the productive and institutional environment may prevent firms from fully benefiting from new market access.

More broadly, these results are consistent with recent discussions emphasizing that external integration alone cannot drive development. Instead, long-term competitiveness depends on the ability of economies to strengthen institutions, improve productivity, and diversify their productive base. Without such internal capacities, increased exposure to international markets may lead to greater trade dependence rather than structural transformation.

From a policy perspective, the implications are clear. Trade liberalization should not be viewed as a standalone development strategy. Rather than relying primarily on the expansion of trade agreements, policymakers should prioritize complementary reforms aimed at reducing informality, strengthening institutional quality, fostering innovation, and supporting industrial upgrading. Without such measures, Latin American economies risk remaining locked into patterns of specialization based on primary goods, limiting their capacity to benefit from globalization.

Finally, this study is subject to several limitations. While it provides robust empirical evidence on the relationship between trade agreements and export upgrading, it does not directly identify the underlying mechanisms driving these results. Future research could explore these channels in greater detail, particularly by examining firm-level dynamics, participation in global value chains, and the role of targeted industrial policies.

Overall, this paper shows that trade agreements without structural transformation do not lead to export upgrading in Latin America. The findings underline the need to move beyond a purely trade-centered approach to development and to focus instead on the structural conditions necessary for sustained economic transformation.