Latin America regional integration can be characterized by the co-existence of multiple trade agreements: there are 33 of these in total, their membership is limited in scope, and they generally act as separate silos which lack important interconnections between them.
Common sense indicates that while this “spaghetti bowl” of agreements might encourage production linkages among the members of each agreement, it might also limit the emergence of supply chains between countries across agreements by making it too costly for producers to use foreign inputs from outside their respective blocs due to the combination of tariff rates and rules of origin (the provisions that determine the official origin of a good).
It is no wonder that Latin America compares unfavorably to other regions regarding participation in international supply chains in general and in the formation of regional supply chains in particular.
Given the state of integration in Latin America, the question is whether fixing this spaghetti bowl could lead to the emergence of more regional supply chains.
One way to answer this question is to look at one modest experience in our region in which the spaghetti bowl was eliminated. Let’s start by looking at how this spaghetti bowl came to life. Between 1995 and 2001, Mexico signed three separate free trade agreements (FTAs) with countries in Central America: one with Costa Rica (1995), one with Nicaragua (1998), and one with the “Northern Triangle” of El Salvador, Guatemala, and Honduras (2001).
Because these agreements had separate rules of origin, in practice they functioned as though there was one hub (Mexico) and three separate spokes (each FTA) that did not interact with each other. For example, chocolates from Costa Rica would not encounter tariff charges in Mexico as long as they were produced entirely in Costa Rica, but the same chocolates would pay a tariff duty in Mexico if they used cocoa paste from Honduras.
In 2011, however, all the countries signed a new agreement with a single set of rules of origin.
The original spaghetti bowl of three noninteractive agreements was effectively changed by a single agreement that gave firms much more flexibility regarding where they could source their inputs from within Central America.
Evidence using data from Costa Rica suggests that this increased flexibility when sourcing inputs might have triggered more production sharing across countries in the region. For instance, based on a comparison using transaction-level data before and after the agreement, it is revealing that between 2010 and 2013, the number of Costa Rican firms exporting to Mexico that used imported inputs from other Central American countries increased by 20%, while the number of Costa Rican firms exporting to Mexico that did not use imported inputs from Central America did not increase between these two years.
Likewise, a comparison of the sourcing patterns of Costa Rican exporters purchasing inputs from Guatemala, Honduras, Nicaragua, or El Salvador before and after the agreement reveals that the agreement almost doubled the percentage of inputs from these countries being used in exports that target the Mexican market.
Accordingly, this early evidence supports the notion that increasing the flexibility for exporters to source their inputs from other countries in the region might lead to more regional supply chains.
The Pacific Alliance countries provide another interesting example of a group of different, noninteracting FTAs being superseded by a single agreement. It would be interesting to analyze how this enlarged trade area will impact the formation of regional supply chains in the coming years.
Solving the spaghetti bowl in Latin America at the continental level may not be as hard as it appears, according to a recent IDB publication. This may greatly expand the set of potential countries within the region which Latin American firms can source inputs from, which could lead to major efficiency gains.