Analyzing the trade potential of SIDs with a focus on CARICOM’s small resource exporters

Abstract

For developing economies, exporting broadens the horizon for facilitating effective integration into the global economy and improving overall competitiveness. While small developing states are particularly at high risk given their fundamental characteristics, the expected impact of these exogenous shocks on small resource dependent economies are amplified given the high possibility of lower levels of diversification. This paper therefore assesses the determinants of exports and exporting potential of SIDS (small island developing states) with a focus on Caribbean resource-based countries. Using non resource exports, in an aggregate and disaggregate form, as the dependent variable, the traditional gravity model was augmented to include a revealed comparative advantage (RCA) index, the real effective exchange rate (REER) of the exporter and the export sophistication (EXPY) index. The main findings indicate that an appreciation of the REER has an inverse impact on exports of non-resource commodities and that trade according to comparative advantage and differences in export sophistication occurs mainly in the agricultural sector.

KEYWORDS:

• Gravity model
• comparative advantage
• Dutch disease

JEL CLASSIFICATIONS:

• F14
• F10

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Dutch disease is a term used to describe the negative effect of a boom in a booming tradeable sector has on the non-booming tradeable sector.

2 It has been argued that Dutch disease does not necessarily lower economic growth, unless the manufacturing sector is assumed to be “special”. However, in relation to the export led growth hypothesis, the capacity of the manufacturing sector to generate sustainable growth, through various avenues including increased competition (based on lower concentration ratios) and technological and knowledge spillovers, cannot be overlooked.

3 The top exported commodities and top trade partners of Guyana, Suriname and Trinidad and Tobago are shown in Table A1 and Table A2 in Appendix, respectively.

4 T&T is primarily reliant on petroleum oils and gases for export revenue, while historically Guyana and Suriname are mineral exporters with gold being the main export of both countries (the top 10 exports according the HS6 classification of each country is presented in Table A1 in Appendix). In 2015, vast offshore reserves of oil were discovered in Guyana therefore changing the export dominance of the country in 2020 from mainly mineral exports to predominantly petroleum oils. These three countries have a relatively high comparative advantage in the resource sectors as compared to non-resource sectors, as shown in Table A3 in the Appendix and resource exports contribute a large share to total exports (Figure A1 in Appendix).

5 In handling the mass variable there are four alternatives according to Reinert (2013). The first alternative, with the most solid theoretical foundations, implies that mass is associated with the gross domestic product (GDP) of the countries. The second alternative associate mass with both GDP and population of both countries. The third alternative uses GDP per capita as a representation of mass, while the fourth uses both GDP per capita and GDP.

6 This model therefore infers that: $X_{\mathrm{ij}}=\frac{Y_i{E}_i}{d^2}$ Where X_ij: gives the predicted movement of goods or labor between i and j; Y_i is the mass of goods or labor or other factors of production at origin i; E_j is the mass of goods or labor at destination j; and d is the distance between i and j.

7 For instance, van Bergeijk and Brakman (2010, 5) described Tinbergen’s initial model as “providing only a common sense rationale for this quasi-postulated reduced form equation.”

8 Anderson’s (1979) model is built in three stages. Firstly, a pure expenditure model is specified under the assumptions of only two goods and two countries where no tariffs or transportation costs exists (frictionless trade) resulting in an over simplistic and unlikely equation. The second model is a trade share expenditure model, where the countries now produce both traded and non-traded goods. The demand for traded goods is therefore determined as if the homothetic utility function for traded goods alone was maximized subject to a budget constraint. This constraint is essentially the importing countries’ (j) share of expenditure dedicated to traded goods in total expenditure. Lastly, Anderson specified the most relevant model where there are many commodities with a full set of national tariffs available in each country case and the inclusion of transport cost, proxy being distance.

9 In terms of international trade, countries try to maximize trade at a minimum cost.

10 The simple intuition behind this assumption implies that if income increases, while relative prices remain the same then the consumption of the two differentiated goods increase at the same rate.

11 Also called the Armington assumption.

12 The literature suggests the use importer and exporter fixed effects to account for the multilateral resistance terms (Feenstra 2000; Silva and Tenreyro 2006).

13 While REER appreciation can result in cheaper imported machinery and intermediate goods available to the non-resource sector, it can also encourage production for domestic consumption given the uncompetitive exporting environment, which can further appreciate the REER. In the case of Dutch diseased economies, lack of diversification and shrinking non resource tradeable sectors indicate that the impact of REER is not positive.

14 Separate equations were estimated for each sector k as presented in the empirical section.

15 $B_{\mathrm{ij}}=\frac{X_{\mathrm{ij}}}{\overline{X_{\mathrm{ij}}}}$ where $\overline{X_{\mathrm{ij}}}$ = $\sum \overline{X_{\mathrm{ijk}}}$ and $\overline{X_{\mathrm{ijk}}}=\frac{X_{\mathrm{iwk}}\ast X_{\mathrm{wjk}}}{X_{\mathrm{wwk}}}$.

.

16 The traditional gravity model in multiplicative form can be shown as: $X_{\mathrm{ij}}=a_0{Y}_i^{a_1}{Y}_j^{a_2}{D}_{\mathrm{ij}}^{a_3}$, where a₀, a_1, a₂ and a₃ are parameters. It is obvious that logging the above equation results in indefinite values for the trade variable if no trade existed. Silva and Tenreyro (2006, 643) noted that “if economic theory suggests that y and x are linked by a constant-elasticity model of the form $y_i=\exp \left(x_i\text{β}\right)$, the function is interpreted as the conditional expectation of y_i given x.” The gravity model can therefore be written as an exponential function [$X_{\mathrm{ij}}=\text{exp}\left(\ln a_0+a_1\ln Y_i+a_2\ln Y_j+a_3\ln D_{\mathrm{ij}}\right)]$. Since the trade variable is not logged, a known value exists for each iteration therefore the model in this case is not truncated. Taking logs on both sides yield [$\ln X_{\mathrm{ij}}=\ln a_0+a_1\ln Y_i+a_2\ln Y_j+a_3\ln D_{\mathrm{ij}}],$ hence the interpretation remains the same as OLS estimated coefficients.

17 This equation differs from equation 1, as data for some variables were not available for all three resource dependent CARICOM economies. For example, data from World Development Indictors for Suriname’s REER was not available for any years in the dataset, and therefore this variable was dropped to allow for inclusion of Suriname in the estimation.

18 There were gaps in the dataset due to unavailability of data for some countries for some years. Therefore, the timeframe may differ for some countries. In those cases, the longest dataset available over the period 2000–2019 was used. At the time of the analysis, trade data, sourced from UNComtrade stopped at 2019.

19 This allows for an up-to-date assessment of each countries trade arrangements as compared to using a long average. Further, after 2010 commodity prices were on the downswing of a price super cycle.

20 This is relevant when goods are classified as being a part of the industrial sector but are used in the production process of other manufactured goods.

21 Guyana has the largest area of agriculture land when compared to other CARICOM countries. Therefore, the country has potential to encourage investment in this sector, especially with already established rice and sugarcane industries. Along with Suriname (which has the second largest area of agricultural land in the region), these two countries have the potential to reduce CARICOM’s extra-regional imports of many staple foods including corn and rice. T&T produces some of the highest grade cocoa in the world and can therefore further develop this industry. Overall, there are many industries ranging from coconuts to tubers which can be expanded to increase diversification of these SIDs.