Prospects for the Cuban economy
An analysis for the case in which the United States’ trade embargo is fully lifted
(Image courtesy of: http://global.fncstat.com)
BSc Thesis Economics
Faculty Economics and Business University of Amsterdam
Author: Max ten Have
Student nr: 10261605
Supervisor: Dhr. prof. dr. Sweder van Wijnbergen June 2015
Statement of Originality
This document is written by student Max ten Have who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Abstract
Whilst the United States’ trade embargo on Cuba has been in place for over half a century, recent developments indicate an upcoming reconciliation. This thesis analyses the prospects of the Cuban economy in the event the embargo is fully lifted by estimating the potential trade creation between Cuba and the United States through applying the gravity model of trade and trade share analysis, mainly elaborating on Montenegro and Soto (2000). Subsequently, the effect of the newly created trade will be investigated, after which the plausibility of these theoretical results will be examined in a comparison with the historical development of Portugal from 1970 to 2000. The results show Cuban imports and exports creation opportunities that account for nearly 30% of the Cuba’s current GDP, which are expected to have a significant impact on the Cuban economy in the medium-long term. However this conclusion solely holds providing that Cuba is able to gradually implement reforms that address multiple restrictive domestic conditions in order to accommodate the new trade flows.
3
Table of Content
Page I) Introduction __________________________________________________________________ 4 II) Cuba: historical background and country profile _____________________________________ 5 III) Literature review ______________________________________________________________ 8 3.1 Gravity model of Trade __________________________________________________ 8 3.2 Research Montenegro & Soto (2000) _______________________________________ 10 3.3 Trade Creation vs. Trade Diversion ________________________________________ 11 3.4 Empirical framework: Portugal ____________________________________________ 12 IV) Methodology _________________________________________________________________ 14 4.1 Estimating potential future trade flows _____________________________________ 14 4.2 Potential trade creation _________________________________________________ 15 4.3 Impact on Cuban economy _______________________________________________ 16 4.4 Empirical framework: Portugal ___________________________________________ 17 V) Results _______________________________________________________________________ 18 5.1 Estimating potential future trade __________________________________________ 18 5.2 Potential trade creation _________________________________________________ 18 5.3 Impact on Cuban economy _______________________________________________ 19 5.4 Empirical framework: Portugal ___________________________________________ 21 VI) Conclusion and Discussion _______________________________________________________ 22 Bibliography _______________________________________________________________________ 24 Appendix 1 Graphs _________________________________________________________________ 27 Appendix 2 Results _________________________________________________________________ 29 Appendix 3 References ______________________________________________________________ 31
4
I)
Introduction
In a time where political conflicts remain to result in economic sanctions, such as the contemporary Russia vs. EU and United States case (The Economist, 2015), it might be interesting to take a closer look at one of the longest running trade embargos of all times, the U.S. trade embargo on Cuba. This U.S. embargo has been in place since 1961 (‘Cuba Overview’, 1999) in order to destabilize the Cuban regime under Fidel, and later Raúl Castro. The embargo has been subject to several adjustments, but remained in place up to the present time.
Although this embargo has been numerously analyzed, current affairs and the outdated relevance of former studies create an excellent opportunity to reinvestigate the effects of the embargo on the Cuban economy and to forecast its possible growth in case of a breakthrough in the distorted trade relation with the United States. This study has only become more relevant with the recent announcement of U.S. president Barack Obama to change the course of the U.S. foreign policy regarding Cuba and the first official meeting of the highest diplomatic level in more than half a century (Reuters UK, 2015) , which suggests that the U.S. trade embargo on Cuba could actually cease to exist in the near future.
In order to get a better view of the prospects of the Cuban economy once the embargo is lifted, the main focus of this thesis will be on predicting the possible future trade flows for Cuba, in particular the ones with the United States, and to estimate the effect on the economy. Therefore, the main question of this thesis will be:
To what extend could Cuba’s economy develop in the event the United States’ trade
embargo is lifted?
In order to answer this question, three sub questions will be covered.
- Firstly, what trade flows will arise between Cuba and the U.S. in the event the embargo is lifted?
- Secondly, how much new trade can be expected from this modification in policy and how does that impact the economy?
- And lastly, how does this theoretical estimation of economic development fit in an empirical framework?
Considering the substantial magnitude of the American market and moreover the extreme vicinity of Cuba and Florida, the lifting of the embargo is expected to significantly boost the Cuban economy. For predicting potential future trade flows, the Gravity model of trade will be applied,
5 whereby this part of the research will elaborate to a large extend on the Gravity model that Claudio Montenegro and Raimundo Soto have constructed in 2000. In order to estimate the impact on Cuba’s economy, the trade created is separated from trade diversion and a desk research will provide help with interpreting the potential created trade flows. Lastly, the development of the Portuguese economy after joining the European Community in 1986 will be analyzed in order create an empirical framework for comparison. This last step may be a valuable addition to the research as Portugal has experienced a similar dismantling of trade barriers and reopened access to a substantial and close neighborhood market . Apart from the necessity of renewed research on this topic with recent data, the comparison of Cuba’s prospects within the Portuguese empirical framework in particular is expected to increase the value that this thesis can add to the existing research.
Following this clarification of the main research focus, the next section will provide a country profile describing Cuba´s historical background with regards to the embargo and a brief analysis of the contemporary Cuban economy. Section 3 will discuss the literature that formed the theoretical foundation for this thesis. Elaborating on that provided expertise, section 4 will outline the
methodology used in this research, whereafter section 5 presents and provides interpretation of the subsequent results. Finally, section 6 will briefly summarize the research, answer the sub- and main questions to provide the conclusion and contain a discussion about the research conducted.
II)
Cuba: historical background and country profile
The historical background of the embargo links back to the Spanish colonial past as Spainrelinquished control over Cuba to the U.S. in 1898. The U.S. granted Cuba its independence in 1902, but remained the right to intervene in Cuban affairs until 1934. However, the U.S. remained closely involved with the Cuban political affairs during the 25 year-long reign of army sergeant Fulgencio Batista. In the 1950’s Cuba struggled with high rates of unemployment, starvation and homeless people, while American owned enterprises owned large parts of the land and primary industry. After an unsuccessful revolt in 1953, Fidel Castro, now aided by Ernesto ‘Che’ Guevara, did succeed to overthrow the Batista regime in 1959. In 1960, Castro nationalized all U.S. businesses in Cuba without compensation, which led to the U.S. imposing a trade and arms embargo on Cuba in response. As Castro proclaims Cuba a socialist state and pursued close relations with the Soviet Union, the U.S. retained their policy against Cuba in an attempt to destabilize the Castro regime (‘’Cuba Overview’’, 1999).
The Cuban economy grew steadily during the 1970s and 1980s (Graph 1), as the
6 the USSR (Union of Soviet Social Republics) (Feinberg, 2011, p. 8). Largely fueled by the continuing demand of primary goods and the willingness to pay subsidized prices from the USSR and other socialist state partners (CMEA), the Cuban exports concentrated heavily in sugar, minerals and tobacco (82%), while the main imports were fuels, machinery and other capital goods(65%) (Montenegro & Soto, 2000, p.4). As the East-European market collapsed in 1989, Russian support phased out and socialist barter-trade agreements were terminated, which caused the imbalanced Cuban economy to fall into a deep recession. Estimates of the period from 1989 to 1993 vary from a -31.6% to -48% decrease of Cuban GDP from 1989 to 1993 (Mesa-Lago, 1998, p. 858). During this period, often domestically referred to as the ‘Special Period’, the Cuban regime was forced to open up for market activities traditionally regarded as ‘counter-revolutionary’, mainly related to foreign direct investment (FDI), tourism and entrepreneurship (Laverty, 2011, p. 9). The most important reforms regarding facilitating FDI includes the implementing a dual currency system, thereby segregating the commodity and labour markets for interaction with the global market whilst
retaining a socialist domestic economy (Vos, Ganuza, Morley & Robinson, 2006, p.231). During these years, the composition of Cuban exports started changing fundamentally, as Cuba moved from an agriculture-led to a service-led economy. In 1989, 90% of total exports were primary good and the sugar industry itself accounted for 66%. However, due to inefficient central planning and lack of reinvesting profits along with decreasing world prices, the Cuban sugar industry collapsed
(McPherson & Trumbull, 2007, p. 18). Nowadays the sugar industry accounts for approximately 6.5% of all Cuban exports, whereas nickel, tobacco and coffee took over the agricultural exports and tourism develops into the new main export product1.
While Cuba explored liberalizing reforms, the U.S. tightened the embargo through the Helms-Burton Act of 1996 aiming to penalize every international financial transaction with Cuba worldwide in addition to the sanctions already imposed on domestic individuals and companies. Together with the emergence of Hugo Chávez as president of Venezuela in 1999 and the formation of an alliance between the two countries (Venezuela provides large amount of oil and Cuba in return exports medical personnel and educational services), Cuba started recovering from the ‘Special Period’ (Laverty, 2011).
In the aftermath of hurricane Michelle, the U.S. regulation allowed for specific food and medicines export to Cuba for humanitarian aid since 2001(Feinberg, 2011, p. 4). Raúl Castro succeeded his brother Fidel (known for his cry: ‘Socialism or death’) as Cuba’s president in 2008. Raúl carried out institutional reforms in order to improve government efficiency and accountability, and distributed over 180.000 new licenses for self-owned businesses (Sánchez Egozcue, 2004, pp.
7 23-26). Only recently under president Obama, the U.S. mitigated the restrictions on family travel and remittances, accompanied by the gradual closing of the Guantánamo Bay prison on Cuban territory.
Despite the recent reforms, the impact of decades of central planned economy remain visible. While nearly 20% of Cuba’s workforce still works in the agricultural sector, this sector only accounts for about 4% of the nation’s GDP2. According to Laverty (2011, p. 20) this discrepancy is due to shortages of equipment and technology, limited access to financial institutions and state interference in production decisions and pricing.
In terms of population (11.27 million in 20133) and estimated GDP ( 68.23 billion US$ in 20113), Cuba counts as the biggest market within the Caribbean, but remains a minor market on a global scale. Sánchez Egozcue (2004, p. 17) describes Cuba’s business environment as highly regulated by the government, points out the low availability and high cost of capital, but also emphasizes the high education level of the labor force and favorable natural resource endowments compared to other Caribbean countries. Furthermore, the colonial past provided Cuba with a sound basic infrastructure, although somewhat deteriorated due to lacking maintenance (Ritter, 2006, p. 13). Whereas Cuba remained a member of the United Nations, it resigned its membership as a member of the IMF and the World Bank under Fidel Castro.
Cuba’s unique historical background also has a considerable impact on the country’s trading structure. In 2011, Cuba’s main trading partner was Venezuela (42%, graph 2), whereas an
increasing shift towards the emerging markets can be observed. Feinberg (2011) points out the Cuban fear that they can not rely on Venezuela forever and therefor attempt to extend their partnerships with China and Brazil. However, given the extreme vicinity of Cuba and the U.S. (Havana and Florida are only 170km apart) and the size of the American market, the lifting of the trade embargo is expected to redirect a substantial part of Cuba’s future trade towards the United States (Montenegro & Soto, 2000).
2
According to data retrieved from la Oficina Nacional de Estadísticas (ONE) – Cuba 2011 – sections 5.8 & 7.3
3 Population and GDP according to the most recent data provided by the World Bank on
8
III) Literature review
3.1 Gravity model of Trade
The Gravity model of Trade and its development through time is thoroughly described by Shepherd (2013) and the UNCTAD & World Trade Organization (Bacchetta et al. 2012), who agree that the model is widely recognized as one of the key instruments in applied international trade literature. Both sources credit Jan Tinbergen for being the first to introduce the Gravity model to explain international trade patterns in 1962. The model is typically used to measure the impact of trade costs on bilateral trade flows. It owes its name to the key variables of ‘economic mass’ of the trading partners, which consists of their GDP, and the absolute distance in between the countries as an important proxy of the transportation costs. In reference to Newton’s law of gravity, the basic Gravity model of trade thus states that just like planets are mutually attracted in proportion to their size and proximity, larger countries are expected to trade more and that countries that are further apart from each other are likely to trade less. In this most basic ‘intuitive’ form, the Gravity model can be written as follows:
𝑀𝑗𝑖, 𝑋
𝑗𝑖 = 𝑓 (𝑇𝑃𝑖, 𝑇𝑃𝑗, 𝑇𝐶𝑗𝑖)
in which the import (M) and export (X) between country i and j is determined by a function of the trade potential (TP) of both countries, usually represented by the size of the economy (measured for example in GDP), and the trade costs (TC) between the two countries, which is usually represented by the transportation costs, traditionally measured by the absolute geographical distance between the countries. Empirical research displayed a strong positive correlation between trade and GDP, both for importer and exporter, whereas a strong negative correlation between trade and distance was found (Shepherd, 2013, p 10). Given the multiplicative nature of the Gravity equation, the equation is usually transformed into a log-linear equation when estimating the model by simply taking the natural logarithms of all variables (Bacchetta et al., 2012). This also allows for an easy interpretation of the estimated parameters as they can now be interpreted as elasticities. The basic Gravity equation can thereafter be estimated by:
ln 𝑀𝑗𝑖, ln 𝑋
𝑗𝑖= 𝑏0+ 𝑏1 𝑙𝑛 𝐺𝐷𝑃𝑖+ 𝑏2 𝑙𝑛 𝐺𝐷𝑃𝑗+ 𝑏3 ln 𝐷𝑖𝑠𝑡𝑎𝑛𝑐𝑒𝑗𝑖+ 𝑒𝑖,𝑗
in which 𝑏0 is a constant, 𝑏1, 𝑏2 𝑎𝑛𝑑 𝑏3 are coefficients estimating the effect of the GDP and
Distance variables on the trade flows and 𝑒𝑖,𝑗 is a random error term.
Despite the empirical success of the model, critics stated that the Gravity model lacked solid theoretical foundations. This motivated a search for theoretically justification, in which Linnemann
9 (1966), Anderson (1979), Bergstrand (1985 and 1989) and Trefler (1993) have among others
provided the classical Gravity model with a scientifically derived form and a framework that controls for nationally differentiated goods, technological differences and internationally dissimilar factor endowments. In addition, the basic equation has been expanded with several specific factor variables to acknowledge the effects of trade agreements, cultural and political factors, different factor endowments and colonial history. As a result the classical Gravity model of trade developed into a more theoretically justified version, that was described by Leamer and Levinson (1995, p. 1384) as ‘having provided some of the clearest and most robust empirical findings in economics’.
Even so, Anderson and Van Wincoop (2004) observed some significant shortcomings as the model did not account for relative trade costs. Their results show that bilateral trade is affected by relative trade costs regarding all possible trade partners and not just by the absolute trade costs between country i and j, which illustrates a clear case of omitted variable bias in the basic Gravity model. They therefore developed a so called ‘Gravity with gravitas’ model that incorporated ‘multilateral trade-resistance’ terms for the export and import flow. According Bacchetta et al. (2012) this omitted variable bias could otherwise be avoided by including a ‘remoteness variable’ or the application of country fixed effects for importers and exporters. This last option implies the use of dummy variables that take on unity value every time a unique country appears in the dataset in order to capture all country specific effects.
Another important issue in the gravity model is how to handle zero-trade flows within the examined trade matrix. The standard gravity model is estimated in a log-linear function and will therefore drop zero-trade flows, as the log of zero is undefined. The impact this has on the results depends on the origin of the zero-trade observation (Is it a reporting error, rounding error or useful information?), but Bacchetta et al. (2012, p. 112) stress that the OLS (Ordinary Least Squares) estimation method risks a significant sample selection bias. They point out that the Tobit estimation method has often been used to left-censor on the zero values in order to solve this matter, but note that this method can only be partially justified. Along with Bacchetta et al., Shepherd (2013, p. 55) therefore pleads strongly in favor of the Poisson Maximum Likelihood estimator to ensure consistent and robust results.
Furthermore Shepherd (2013, p. 15), emphasizes the importance of estimating the dependent variable as separate import and export trade flows and highlights the importance of expressing trade flows and GDP data in nominal terms. In addition, Bacchetta et al. (2012, p. 108) note that the use of panel data mitigates the bias generated by unobserved heterogeneity. Both researches underscore the strength of the gravity models to predict future trade flows, but stress that it is increasingly important that a gravity model be as theoretically grounded as possible.
10
3.2 Research Montenegro & Soto (2000)
A research that has previously applied the Gravity model of trade on Cuba is outlined by Montenegro and Soto (2000), on which this thesis will elaborate in the methodology section. Montenegro and Soto explore the degree of distortion of the Cuban trade structure and predicts its evolution in the absence of the embargo. Recognizing that Caribbean countries might share a particular trade structure as determined by location and endowment, they created a separate gravity model for 13 different countries in the Caribbean basin4 with each 100 trading partners, thus yielding 1300 observations. Although their database includes the period from 1980-1991,
Montenegro and Soto acknowledge the significant disparities affecting the economic environment and consequently performed the estimation in two different periods, 1980-1985 and 1986-1991. Montenegro and Soto define the gravity model as follows:
𝑀𝑗𝑖, 𝑋 𝑗𝑖 = 𝛽0+ 𝛽1 𝐺𝐷𝑃𝑖+ 𝛽2 𝐺𝐷𝑃 𝑁𝑖 𝑖 + 𝛽3 𝐴𝑟𝑒𝑎𝑖+ 𝛽 4 𝐼𝑠𝑙𝑎𝑛𝑑𝑖+ 𝛽5 𝐿𝑎𝑛𝑑𝑙𝑜𝑐𝑘𝑒𝑑𝑖+ 𝛽6 𝐺𝐷𝑃𝑖 + 𝛽7 𝐺𝐷𝑃 𝑁𝑖 𝑖 + 𝛽8 𝐴𝑟𝑒𝑎𝑖+ 𝛽 9 𝐼𝑠𝑙𝑎𝑛𝑑𝑖+ 𝛽10 𝐿𝑎𝑛𝑑𝑙𝑜𝑐𝑘𝑒𝑑𝑖+ 𝛽11𝐷𝑖𝑗+ 𝛽12𝐵𝑜𝑟𝑑𝑒𝑟𝑖𝑗,𝑘 + 𝛽13𝐿𝑖𝑛𝑑𝑒𝑟𝑖𝑗,𝑘+ 𝛽14∑δ𝑧𝐿𝑎𝑛𝑔𝑖𝑗,𝑘 5 𝑧=1 + 𝛽15𝑇𝐴𝑖𝑗,𝑘
in which 𝑀𝑗𝑖, and 𝑋𝑗𝑖 represent non-fuel imports and exports between the reporter and partner.
Furthermore, ‘Area is total area (in square miles), Island, Border and Landlocked are dummy variables taking value 1 if either the reporter or the partner is an island, share common boundaries or are landlocked. 𝐷𝑖𝑗 is the distance between countries in miles, which is used as a proxy of transportation costs. Linder corresponds to the absolute difference in per-capita GDP (expressed in US$) and is used to test the alternative hypotheses that countries trade more if their economies differ (when trade is determined by comparative advantages a positive sign is expected) or are similar (when they trade based on differentiated products a negative sign is expected). 𝐿𝑎𝑛𝑔𝑖𝑗,𝑘 represents cultural affinities as proxied by equal language among reporter and partner (languages, indexed by k, include English, Spanish, French, Arabic and Portuguese). Finally, 𝑇𝐴𝑖𝑗,𝑘 is a dummy for trade agreements, which is indexed i,{j,k} to reflect that a country may belong to more than one trade arrangement.’(2000, p.11).
As can be derived from this gravity specification, Montenegro and Soto do not include multilateral trade resistance terms or country specific fixed effect dummies, to account for the earlier discussed relative trading costs. They do however include several ‘remoteness’ dummies
4 Colombia, Costa Rica, Dominican Republic, Guatemala, Honduras, Haiti, Jamaica, Mexico, Nicaragua, Panama,
11 (such as Island and Landlocked) to at least partially mitigate this omitted variable bias. Furthermore, Montenegro and Soto chose to account for the zero-trade flow problem by using the Tobit
Maximum Likelihood estimation method. Their results appeared to be robust to specification, time period and trade determinants.
Assuming that the embargo is lifted, the model predicts that a substantial part (roughly 80%) of Cuban current exports and imports would be diverted from their current partners (at that time in particular Canada and Japan) towards the U.S., to reduce transportation and transaction costs. However, Montenegro and Soto do remark that this theoretical result might not be realized as trade barriers (such as quotas, tariffs, regulation) might never be fully absent and long-run economic ties might be hard to sever for political reasons. In addition, Montenegro and Soto acknowledge potential shortcomings due to the lack of available data and thus the inability to provide fully accurate results (2000, p. 14).
3.3 Trade Creation vs Trade Diversion
Jacob Viner (1950) was the first to make the distinction between trade creation and trade diversion in analyzing the effect of trade agreements. Newly created trade flows can emanate from developed trade opportunities through decreasing trade barriers (trade creation) or a shift in existing trade from one trading partner to another through shifts in relative trade barriers (trade diversion). In analyzing trade agreements, Viner associated trade creation with the positive effects a trade agreement had on global economic welfare due to the replacement of high-cost domestic
production by this new trade flow. On the other hand, Viner strictly associated trade diversion with the negative impact on global economic welfare due to shifting production away from the most cost-efficient trading partner. This traditional view was mainly based on the analysis of the effect of trade agreements on relative trade barriers and whether or not the most efficient producer was included or excluded in the agreement.
The strictly negative connotation of trade diversion was however contradicted by Gehrels (1956), Lipsey (1957) and Bhagwati (1971), who proved that both trade creation and trade diversion could yield positive economic effects for the countries involved as well as third party countries. It does however remain clear that the benefits created by a newly occurring trade flow depend strongly on the degree of trade creation relatively to trade diversion, the former providing the most significant economic benefits for the countries directly involved as this trade did not exist before.
12 According to Kendall and Gaisford (2007, pp. 127-129), there are three empirical
methodologies frequently used to determine the amount of trade creation and trade diversion through policy adjustments. The first method examines a country’s trade shares before and after a specific trade agreement comes in to effect. This method however implicitly assumes the trade pattern between the partners to have remained constant in the absence of the trade agreement and is therefore likely to provide inaccurate results. The second method involves an ex post
(retrospective) econometric analysis by adding specific dummy variables to the Gravity model of trade to observe the amount of trade creation versus trade diversion. The third widely used method involves an ex ante (prospective) analysis using computable general equilibrium (CGE) models.
As extensively outlined by Wing (2004, p4), the CGE-model finds its fundamental conceptual starting point in the circular flow of commodities and the Walrasian general equilibrium that occurs when demand and supply are equalized across all interconnected markets in the economy.
Consequently, the main actors in the model are households, who own the factors of production and simultaneously represent the final consumers of the produced commodities, and firms, who rent the provided factors of production in order to produce the goods and services that the households consume. Depending on the purpose of the model, government, trading partners and financial institutions can thereafter be added to the equation. Ivus and Strong (2007, p. 52) further outline how to construct a GCE model. Starting by specifying the functional forms of production and utility, a micro-consistent social accounting matrix is to be constructed or obtained in order to create a benchmark equilibrium to calibrate the model. To solve this framework of accounting relationships and support the general equilibrium in the conclusive GCE model, both Wing (2004) and Ivus and Strong (2007) highlight the use of three critical assumptions, namely the market clearing-, zero-profit- and income balance-conditions for all economic agents. The resulting CGE model should thereafter be capable of capturing the effects of policy changes regarding trade barriers, although Ivus and Strong (2007, p. 49) stress that the accuracy of the model will rely on the degree of aggregation of data.
3.4 Empirical framework: Portugal
As outlined by Lloyd-Jones (2001), Portugal has remained isolated from the European market for a long time, due to the fascist authoritarianism regimes of António Oliveira Salazar (1932–68) and Marcelo Caetano (1968–74). Whereas the European Community (EC) made it clear that they would only be open to democracies, the Portuguese dictatorial regime preferred their isolation and independence, whilst retaining power over their African colonies (Angola, Mozambique,
Guinea-13 Bissau and Cape Verde). However, along with the outbreak of the colonial wars in 1961 came
increasing political, social and economic costs for the Portuguese society. As the world oil crisis intensified the structural problems of the economy, the dictatorial regimes started a process of liberalizing economic reforms, providing a basis for an export-oriented strategy of industrialization and establishing better conditions for attracting FDI (Royo, 2004, p. 108).
However, these reforms did not satisfy the opposition, which resulted in a military coup in 1974, known as the Carnation Revolution. This initiated a period of political turmoil. Even after the first democratic regime was established in 1976, the economy and political structures proved to be highly unstable after years of colonial wars and revolutionary stir (Lloyd-Jones, 2001, pp. 5-6). Nevertheless, the first constitutional governments initiated a foreign policy shift aiming at European integration, according to Royo (2004) a decision based on the political incentive to consolidate the newly established democratic institutions. After the processes of democratization and
decolonization, Portugal officially entered the EC in 1986.
By entering the EC, Portugal experienced a substantial decrease in trade barriers to the European market, although it has to be stated that mutual trade were already decreased to some extent by the preferential trade agreement between Portugal and the EC signed in 1972, and an agreement between the EFTA (including Portugal) and EC in 1976. In addition, Portugal had gained access to IMF support contributing to economic export-led growth and took unilateral measures in preparation for the EC accession, including increasing economic flexibility, adopting a value added tax and intensifying trade liberalization. Nevertheless, Portugal’s EC-membership initiated an era of industrial restructuring, deregulation, increased FDI-inflow, decreased inflation, decreasing trade barriers and, most importantly, high EU subsidies for converging economic imbalance within the EU (Royo, 2004, pp. 109-110), which all contributed to Portugal’s strong economic growth.
The historical development of Portugal as a small, agricultural and industrial economy that was isolated from a large neighboring market due to political reasons (Royo, 2004) offers clear similarities with the Cuban case. Therefore, this relatively well documented process is expected to provide suitable grounds of comparison for the Cuban prospects that are rather scarce to find elsewhere in the last decades.
14
IV) Methodology
4.1 Estimating potential future trade flows
As outlined by Shepherd (2014) and Bacchetta et al. (2012), the construction of a Gravity model requires advanced econometric knowledge, an extensive database and numerous simulations in order to yield accurate and reliable results. This in combination with the fact that Cuba’s trade is severely distorted due to political factors, made it infeasible to construct a new reliable Gravity model during this thesis. Although not fully up to date, the research conducted by Montenegro & Soto (2000) implemented theoretical justification and modern estimation techniques, estimated a separate model specifically targeting Caribbean countries to proxy for Cuba, and provided direct access to their Tobit estimated coefficients. Therefore their model served as the main research instrument in predicting the potential trade flows between Cuba and the United States.
The coefficients from the Caribbean countries model for the period 1986-1991 (Montenegro & Soto, 2000, Table 5) will serve as the coefficients in this updated estimation. The Caribbean countries model is preferred over the general model in order to approach Cuba’s country specific characteristics as well as possible. As outlined in Montenegro and Soto, the period 1980-1986 captures the last period of a ‘pure’ global socialist trading block, before the first attempts of reform were introduced, whereas the period 1986-1991 attempts to depict the situation after the collapse of the socialist economic order. In addition, most Caribbean countries were severely affected by the Latin American Debt crisis during the period 1980-1985, whilst the period from 1986-1991 can be characterized by recovery and consolidation of the reforms. Therefore, the latter period is regarded as most appropriate for a contemporary research.
When filling in the gravity estimation for Cuba and the United States, it is clear that many variables can be dropped from the equation given that Cuba and the U.S. do not share a common land border, do not participate in similar trade agreements (which would however eventually be a natural continuation after terminating the economic sanctions), are not landlocked and do not share the same language (although the latter might be disputed when taking solely Florida and Cuba in to account). The remaining variables are all accompanied by parameters with the expected sign according to underlying economic theory, although it is worth mentioning that the positive sign accompanying the Linder hypothesis variable (the absolute difference in GDP per capita) implies that Caribbean countries trade with their partners on the basis of comparative advantage. Furthermore, the GDP per capita coefficients, addressing the degree of industrialization occur to be not
particularly significant. This is likely to be caused by multicollinearity issues as the variable GDP per capita tends to correlate strongly with the GDP variable (Stock & Watson, 2012, pp.241-244). Nonetheless Stock and Watson argue that this variable might still be holding valuable information
15 and should therefore be included in the estimation in order to avoid omitted variable bias.
Furthermore, the coefficients for the reporter’s ‘Island’ dummy and ‘Area’ variable are unobservable, but are assumed to be represented in the value of the constant.
The data on GDP and GDP per capita to perform the contemporary estimation are the most recent, nominal values to be obtained through the World Bank WDI database. The distance between the two countries is measured in miles between the most probable trade centers (the harbours of Havana and Miami), as opposed to the straight distance between the borders or country capitals. However, from looking at the estimation (table 1) it is clear that the U.S. GDP value will dominate the estimation and even yield highly unrealistic results by a factor thousand if GDP-values are to be entered in single U.S. Dollars. Therefore, although not clearly defined by Montenegro and Soto, the GDP values are assumed to be entered in thousands of U.S. Dollars, in the same way that
Montenegro and Soto did specify the Area variable in thousands of square miles.
4.2 Potential trade creation
In order to determine the amount of trade creation from the new potential trade flows, an ex ante counter-factual analysis is to be conducted. Ideally, the CGE-model would serve as the main research tool to analyze the amount of trade created as well as the effect of this additional trade, as the lifting of the embargo could be analyzed as a trade agreement that eliminates infinitely high trade barriers. However, the complexity of the Cuban economy and, again, the lack of available data make it difficult to construct a reliable social accounting matrix and a fitting equilibrium framework. For example, the Cuban economy differentiates itself from the common economic structures as it remains largely centralized. Apart from this distortion of natural equilibria through government intervention, the implemented dual structures with respect to commodities and labour markets as a consequence of the dual currency system, adds more complexity to establishing equilibria in all markets simultaneously. As this analysis deals with counter-factual scenarios, referring to predicting outcomes not by extrapolating current affairs but by estimating the impact of (multiple) policy interventions, the CGE-model would require even more sophisticated specifications, which makes the CGE-model an infeasible research tool in this thesis. Therefore, the method of market share analysis will be applied in order to determine potential trade creation, thereby accepting the loss of accuracy in the estimations to compensate for the simplicity.
This research will therefor elaborate on Montenegro and Soto’s claim that roughly 80 per cent of Cuba’s trade would divert from current trade partners towards the U.S. as a result of the negligible transportation costs. More specific, the Caribbean countries model (1986-1991) mentions trade diversion rates of 83.5 per cent for Cuba’s exports and 76.8 per cent for imports (2000, Table
16 6). Simple calculations derived by this claim could be applied in this trade share analysis to
determine trade diversion and thus trade creation. However it is not entirely clear whether
Montenegro and Soto refer to the amount of future trade share that the U.S. embodies consisting of both trade creation and diversion or whether it is purely the amount of current trade diverted from existing trade partners due to the trade policy change. Considering the expression of the terms ‘diversion’ and ‘switching away current trade’ in their abstract and conclusive section, it would be fair to assume the latter and thus allows for trade share analysis. Cuba’s recent total export and import are collected from the Cuba’s official statistical office (ONE), in which the minor share of current U.S. imports (approx. 0.5 billion US$ of food and medicines) is deducted from total imports in order to obtain the most accurate results.
4.3 Impact on the Cuban economy
In determining the effect of the potential trade creation as derived through the previous
methodology sections, an alternative method for constructing a CGE-model could be found in the economic theory of the Keynesian foreign trade multiplier. However, in order to do so, one would have to be able to determine the marginal propensity to consume (MPC) and the marginal
propensity to save (MPS). Again these economic indicators are not readily available for the Cuban economy. Moreover, if the possibility to investigate household income, savings and expenditure through surveys would occur, the theoretical value of these indicators would still be questionable considering the dual structures in Cuba’s economy. Therefore, a brief desk research will be conducted in order to discuss the potential impact of the potential trade creation for the Cuban economy. By elaborating on previous research and focusing on the relevant aspects for the Cuban case, at least a scientifically based indication of the final impact can be provided.
The research consulted in this section will contain the work of Vos, Ganuza, Morley and Robinson (2006, pp. 231-267) who investigated the gains of free trade in Latin-America and did manage to construct a customized CGE-model for Cuba through socio-economic household surveys. They acknowledged the more active and distorting role of the Cuban government and created a unique social accounting matrix by implementing dual commodity markets and according to the type of demanding agents (households or enterprises). Furthermore they implemented two financial accounts (an official exchange rate account for transactions in the business sector and an unofficial exchange rate account to create a money balance for households) to identify the current monetary duality. This allowed for differentiation between the two currency exchange markets between the economic agents, whilst preserving simultaneous equilibria in solving the CGE-model. Vos et al.
17 (2006) constructed the model with data from 1999 and subsequently tested various scenarios of external shocks and policy changes in estimating the impact on the Cuban economy compared to the base year 1996. They conducted the research with relatively inelastic relationships due to the lack of available econometric research providing estimates for substitution and transformation elasticities.
Whereas Fugarolas, Mañalich and Matesanz (2007) emphasize the historical importance of import-led growth for Cuba, several other papers will be touched upon to highlight possible limitations and specific conditions in order for Cuba to be able to take optimal advantage of the increased trade potential. Finally, Sánchez Egozcue (2004) will address the expected developments with respect to the time frame.
4.4 Empirical framework: Portugal
To put the theoretical predictions for the development of the Cuban economy in an empirical framework, the historical development of Portugal from 1970 to 2000 will be used as a comparison. As outlined earlier in the literature review this time frame includes several years prior to the military coup (1974) and consequently foreign policy shift (1976) as well as the long term effects after the actual accession to the EC (1986), whilst excluding the further reduction of trade barriers due to entering a currency union (EMU) as well as market distortions due to the European recession in the early 2000s.
The comparison will be conducted based on economic variables illustrating the size of the economy, GDP and GDP per capita values obtained through the World Bank and several other economic, political and institutional factors and circumstances provided by the OECD-database, the UNDP and several previously conducted researches. Furthermore, disparities in circumstances affecting the comparison of the two cases will be taken into consideration in an attempt to provide an accurate empirical framework for the Cuban case.
18
V)
Results
5.1 Estimated potential trade flows
Solving the estimation accordingly by taking the sum of the combined values of recent statistics and previously determined coefficients, the expected potential export and import trade flows between Cuba and the U.S. will come down to US$ 25.2 billion and US$ 29.5 billion respectively in the event of free trade (Table 1). Considering Cuba’s current total GDP of US$ 68.2 billion and compared to Cuba’s current total exports of approximately 65 billion US$ and imports of approximately 14 billion US$5, one might however interpret these substantial outcomes as overestimating results. In order to fairly assess the results a comparison with the Dominican Republic could offer support, as this country is quite similar to Cuba in various ways. The size of its economy approaches Cuba’s with a slightly lower GDP ($58 billion)6 and population (10 million)6, whereas its location in the Caribbean is still reasonably close to the U.S. market and both countries share a history of Spanish occupation, which may have left its mark on their (institutional) infrastructure. In 2014, the Dominican Republic imported for $7.27 billion from the U.S. and exported for $4.87 billion to the United States7. Bosco and Lionetti (2004, pp.14-15) also investigated Cuba’s loss of potential trade by conducting their research with a gravity model on proxy countries such as Jamaica, Mexico and Costa Rica from the period 1993-1998. They estimate the loss of potential trade for the Cuban economy varying from 31% (Jamaica) to 68% (Mexico), thus somewhat confirming our results.
Furthermore we should be aware of the fact that the Gravity model estimates trade flows in the event of complete free trade, which proved nearly unfeasibly in reality. For example, Grundke and Moser (2014) illustrate how the United States from 2002 to 2012 have increasingly blocked imports through product regulation, while decreasing traditional import tariffs. This form of ‘hidden protectionism’ illustrates a variety of trade barriers that could significantly deflate the actual potential trade flows. Hence the magnitude of the estimated potential trade flows can be at least partially explained.
5.2 Potential trade creation
Assuming that Cuba and the U.S. could in fact attain perfectly free trade, it would still be important to distinguish the trade created by this revived partnership from the trade that will simply be extracted from existing trading partners. Even though trade diversion would in this case arise from
5
Oficina Nacional de Estadísticas (ONE) – Cuba 2011, sections 8.5 & 8.6
6 World Bank 2011
19 efficiency gains, it is questionable if these minor effects would trickle down to improve domestic purchasing power. The amount of trade creation however is known for its beneficial effects on economies and therefor of much more value to determine. According to the calculations as outlined in table 2, the potential export creation between Cuba and the U.S. would come down to $20.2 billion and the potential import creation to $19.2 billion.
These potential new trade flows would be of substantial importance for the Cuban economy thereby confirming the expected impact of the proximity and size of the American market. In essence, this observation corresponds with the natural trading partner hypothesis, which argues that trade creation is likely to dominate trade diversion when two nearby countries re-connect due to a policy change. The natural trading relationship is characterized by relatively low transport and transaction costs. Furthermore, Kendall and Gaisford (2007, pp. 121-127) confirm that trade creation also tends to be higher between developed and developing economies, for increasing size of the new trade partner’s economy and when the initial trade tariff was prohibitive.
5.3 Impact on the Cuban economy
Vos et al. (2006) conducted their research by simulating various scenarios for the Cuban economy, of which the simulations concerning export demand shocks, export promotion policy, productivity shocks and the WTO (approaching free global trade) and FTAA (Free trade between the Americas) scenario are particularly valuable for this thesis subject. Vos et al. state that the Cuban economy is clearly restricted in the external sector, but show that a 10% increase in export demand alone would have a relatively small impact on the Cuban economy (GDP +3.63%). However, scenarios testing for a 9% increase in export demand combined with a 7% increase in generalized productivity show a far more substantial impact on the Cuban economy (+8% GDP) with major positive implications concerning household consumption and unemployment rates. Furthermore the WTO-scenarios do not show any substantial improvements for the Cuban economy, whereas the FTAA-scenarios (in particular the FTAA-scenario with a 9% increase in export demand and 8% increase in overall productivity) does illustrate the vast potential for the Cuban economy, yielding 10.5% increase in Cuban GDP and again numerous beneficial side effects( Vos et al., 2006, pp. 261-267). They explain this discrepancy by clarifying that Cuba would currently be unable to compete on the world market if it has to cut its agricultural subsidies. Hence, Vos et al. conclude that export demand, which is heavily impacted by the U.S. embargo, is a decisive element in Cuba’s economic development, whilst simultaneously noting that Cuba should improve its production efficiency, by increasing factor productivities and stimulate FDI, in order to take full advantage from any increased demand.
20 In support of these findings, Laverty (2011, pp. 64-67) confirms the need for increasing Cuba’s productivity as a condition for economic growth. He highlights the importance of liberalizing the economy with regard to entrepreneurship and self-employment to allow for job creation and increased productivity. Moreover, Feinberg (2011, pp. 96-99) argues that Cuba should improve its investment climate by collaborating with financial institutions such as the IMF and World Bank in order to use their accumulated expertise and financial resources. However, Feinberg acknowledges that Cuba already holds important strengths for a developing economy as it has relatively strong public institutions and technical expertise due to the high quality of the educational system.
McPherson and Trumbull (2007, p. 13), who analyzed Cuba’s trade patterns without focusing on the U.S. embargo and implemented a degree of freedom variable in their gravity model, add that addressing these domestic reforms would allow Cuba to utilize their unrealized trade potential with their current trade partners simultaneously.
Furthermore, Fugarolas et al. (2007, p. 31) argue not to focus solely on economic effects regarding export-led growth, by emphasizing the historical findings of the importance of import-led growth for the Cuban economy. Meanwhile, Margulies (2008, pp. 175-176) notes that despite the clear mutual benefits of enhanced trade relations between Cuba and the U.S., the lifting of the embargo might not have the estimated impact. He supports this statement by outlining that Cuba might not be eager to sacrifice their trade relationships with Venezuela and China, as they gave succour after the dissolution of the Soviet Union and still share similar political ideals. Therewithal, Margulies highlights the human capital stronghold within the Cuban economy with regard to health care services and expresses the expectations that the U.S. is not likely to displace the trade
agreements exchanging goods for services in the way Venezuela is currently trading oil for medical personnel. Finally, Sánchez Egozcue (2004, p. 24) summarizes the expectations expressed by nearly all researchers by focusing on the time frame in which to predict the development of the Cuban economy. He states that in case the embargo is lifted, trade diversion is likely to dominate in the short term, thereby yielding just minor efficiency gains. The majority of the research conducted expects the major trade creation benefits to occur in the medium-long term (3-5 years).
Hence, by consulting this previously conducted research, we can state that the estimated potential trade creation alone might not be sufficient for substantially boosting the Cuban economy after the trade embargo. However, if Cuba manages to improve domestic limitations regarding mainly the factor productivity and the investment climate, Cuba’s economy can be expected to experience a boost surpassing the earlier estimations made by Vos et al., considering the magnitude of the estimated potential trade creation.
21
5.4 Empirical framework: Portugal
At the time of accession to the EC (1986), Portugal accounted for a domestic market of
approximately 10 million inhabitants8 with a GDP8 of approximately 38.7 billion US$. Although the country shares a land border with the European market, the former dictatorial fascist regimes in both Portugal and Spain and conflicting EC principles isolated the Iberian peninsula from the European market. This created a similar domestic-market versus large nearby foreign-market situation as Cuba experiences with regard to the United States. It is however important to note that the trade barriers Cuba is facing are infinitely high, whereas Portugal experienced moderately high trade barriers to countries in the EC, which had already been gradually decreased prior to accession (as outlined earlier). The export- and consequently GDP boost that Portugal experienced is therefore distributed over 1972, 1976 and 1986 as we can see in graph 3, whereas the increased export demand in the Cuban case is expected to emerge more instantly.
Yet, Royo states that Portugal did experience a large immediate effect due to the integration in the EC, as exports to the EC increased 17.2% and imports 9.3% in 1986, followed by exports increasing another 19.5% and imports another 20.1% in 1987 (2004, p. 110). Royo continues by assigning Portugal’s export-led growth in the 1990s to the development of economies of scale and additional incentives for investment derived from entry to the European market (2004, p. 112). While Torres(1994, pp. 141-153) agrees that integration into the EC resulted in substantial trade creation, both emphasize the accompanying structural economic reforms. Portugal implemented multiple reforms that effectively increased productivity, improved the investment climate, decreased inflation, in which it was often advised and financially supported by the European Community.
Due to major liberalization and privatization programs, Portugal managed to significantly improve overall economic efficiency, whereas the desire to participate in the EMU (European Monetary Union) resulted in policies significantly decreasing the public deficit, inflation and real interest rates (Royo, 2004, p. 118). In this process, the EC acted as a monitoring agent guiding the necessary reforms, whereas Torres (1994) notes that its presence simultaneously enhanced credibility with regard to foreign investment. However, Royo (2004, p. 120) remarks that European subsidies aiming at rapid European economic convergence functioned as a significant direct catalyst as it accounted for 3-4% of the Portuguese GDP annually. As a result of these reforms, major
infrastructural shortcomings have been addressed, the competitiveness of the production sector has been improved and FDI inflow has vastly increased (Royo, 2004, p. 121) , although the Portuguese
22 levels of schooling remain under the OECD average9.
Cuba however is not likely to receive the extensive support that the EU provided to Portugal in dealing with domestic limitations to accommodate potential trade growth. In response, Feinberg (2011) suggested the close involvement of global institutions such as the IMF and the World Bank in order to assist Cuba in terms of expertise and loans. However, the fact that Portugal not only entered a large and nearby free trade union, but moreover a subsidizing community, did severely bias the short term impact of the political developments. Infrastructural and financial reforms are therefore likely to be implemented less rapidly in Cuba. On the other hand, Cuban education levels are considered well above the Latin American and Caribbean levels10, which can compensate for Cuban competitive advantage on the global market. In addition, Vos et al. (2006) emphasized Cuba’s vast potential for growth in the tourism industry, supporting Cuba’s comparative advantage for export-led growth.
IV) Conclusion & Discussion
Given the recent developments in the distorted trade relationship between Cuba and the U.S. and the outdated relevance of existing research, this thesis analyzed the prospects of the Cuban economy in case the U.S. trade embargo is lifted in the near future. By reinvestigating the potential trade creation for the Cuban economy through applying the gravity model of trade, the potential trade flows between Cuba and the U.S. could be estimated. Thereafter, trade creation was distinguished from trade diversion using trade share analysis. Subsequently, a desk research was conducted in order to provide a scientifically based indication of the impact the new trade flows could have on the Cuban economy, whilst simultaneously illustrating domestic limitations to accommodate this economic growth potential. In addition to these results, the economic
development of Portugal after joining the EC in 1986 was analyzed in order to provide an empirical framework for comparison.
The results show a potential increase of Cuban imports of 19.2 billion US$ and a potential increase of Cuban exports of 20.2 billion US$ on a current Cuban GDP of 68.2 billion US$. The impact of this major trade increase on the Cuban economy is estimated to be substantial in the medium-long term, providing that Cuba effectively addresses domestic restrictions. Although the Cuban case is not completely identical to that of Portugal, this comparison provided valuable insights in order to forecast Cuba’s economic development. Taking all results into account and providing that necessary
9 http://www.oecdbetterlifeindex.org/topics/education/, accessed 23-06-2015
23 domestic reforms are gradually implemented, the Cuban economic growth is likely to exceed that of Portugal after joining the EC, mainly due to the currently infinitely high trade barriers. However, the short term economic growth is expected to be less substantial due to the absence of significant EU subsidies and pressure to implement crucial reforms.
As illustrated by the widely varying results in predicting Cuba’s potential trade with the U.S. and its estimated economic impact in the existing literature, the research results depend heavily on the estimation method and the profoundness of the data used. This research elaborating on semi-dated research with basic calculations is presumably too simplistic to provide accurate estimation results, whereas a personally constructed Gravity- or CGE model could have enabled a more valuable research. Unfortunately, this proved to be an infeasible affair given the available data and
complexity of the subject on the one hand and deficiencies coexisting with a bachelor thesis constrained by skills sets and time.
Whereas this thesis attempted to estimate the effect of the lifting of the embargo, caution is required in analyzing trade-led growth with regard to correlation and causal relations. Therefore, it might be interesting to determine the actual effect of the eliminated trade barriers on Cuba’s economy through an ex post analysis once the U.S. trade embargo on Cuba is actually lifted. Until then, it would be fair to expect an increase in extensive prospective research on this matter given the seemingly forthcoming termination of an iconic foreign policy.
24
Bibliography
Anderson, J. E. (1979). A theoretical foundation for the gravity equation. The American Economic
Review, 106-116.
Anderson, J., & van Wincoop, E. (2004). Trade Costs. Journal Of Economic Literature, 42(3), 691-751. doi:10.1257/0022051042177649
Bacchetta, M., Beverelli, C., Cadot, O., Fugazza, M., Grether, J. M., Helble, M., Nicita, A. & Piermartini, R. (2012). A practical guide to trade policy analysis. World trade organization. Bhagwati, J. (1971). Trade-diverting customs unions and welfare-improvement: A clarification. The
Economic Journal, 580-587.
Bergstrand, J. H. (1985). The gravity equation in international trade: some microeconomic foundations and empirical evidence. The review of economics and statistics, 474-481.
Bergstrand, J. H. (1989). The generalized gravity equation, monopolistic competition, and the factor-proportions theory in international trade. The review of economics and statistics, 143-153. Cuba Overview. (1999). Congressional Digest, [online] 78(3), pp.68-71. Available at:
http://congressionaldigest.com/issue/u-s-cuba-relations/cuba-overview/ [Accessed 27 Dec. 2014].
Davis, J. M., Richardson, T. J., Ossowski, R., & Barnett, S. (2000). Fiscal and macroeconomic impact of
privatization (No. 194). International Monetary Fund.
Feinberg, R. (2011). Reaching out. Washington, DC: Brookings Institution.
Fugarolas, G., Mañalich, I., & Matesanz, D. (2007). Are exports Causing Growth? Evidence on International Trade Expansion in Cuba, 1960-2004.
Gehrels, F. (1956). Customs union from a single-country viewpoint. The Review of Economic Studies, 61-64.
Grundke, R. & Moser, C., (2014). Hidden Protectionism? Evidence from Non-tariff Barriers to Trade in
25 Ivus, O., & Strong, A. (2007). Modeling approaches to the analysis of trade policy: computable
general equilibrium and gravity models. In W. Kerr & J. Gaisford, Handbook on International
Trade Policy (1st ed., pp. 44-55). Cheltenham: Edward Elgar Publishing Limited.
Kendall, L., & Gaisford, J. (2007). Trade creation and trade diversion: analyzing the impact of regional trade agreements. In W. Kerr & J. Gaisford, Handbook on International Trade Policy (1st ed., pp. 120-130). Cheltenham: Edward Elgar Publishing Limited.
Laverty, C. (2011). Cuba's new resolve. Washington, D.C.: Center for Democracy in the Americas. Leamer, E. E., & Levinsohn, J. (1995). International Trade Theory: The Evidence. In: G. Grossman and
K. Rogoff (eds), Handbook of International Economics vol. III. Amsterdam: North-Holland. Linnemann, H. (1966). An econometric study of international trade flows. Amsterdam: North-Holland
Pub. Co.
Lipsey, R. G. (1957). The theory of customs unions: trade diversion and welfare. Economica, 40-46. Lloyd-Jones, S. (2001). Portugal’s History since 1974. ISCTE–Instituto Superior de Ciências do
Trabalho e da Empresa (Lisbon). http://www1.ci.uc.pt/cd25a/media/Textos/portugal-since-1974. pdf (accessed December 15, 2014).
Margulies, M. (2008). Stronger Trade or Stronger Embargo: What the Future Holds for United States-Cuba Relations. Asper Rev. Int'l Bus. & Trade L., 8, 147.
McPherson, M. Q., & Trumbull, W. N. (2007). What if Cuban Trade Was Based on Economic Fundamentals Instead of Political Policies? An Estimate of Cuba’s Trade Distortion. Cuban
Affairs Journal,2(3), 1-23.
Mesa-Lago, C. (1998). Assessing economic and social performance in the Cuban transition of the 1990s. World Development, 26(5), 857-876. doi:10.1016/s0305-750x(98)00020-5
Montenegro, C. E., & Soto, R. (2000). How distorted is Cuba's trade? Evidence and predictions from a gravity model. Journal of International Trade & Economic Development, 5(1), 45-68.
Reuters UK,. (2015, Apr 12). Obama meets Raul Castro in highest-level U.S.-Cuba talks in decades. Retrieved 11 June 2015, from http://uk.reuters.com/article/2015/04/12/uk-cuba-usa-idUKKBN0N20IW20150412
Royo, S. (2004). From authoritarianism to the European Union: the Europeanization of Portugal. Mediterranean Quarterly, 15(3), 95-129.
26 Sánchez Egozcue, J. M. (2004). Chapter 5: Challenges of Cuba´s Insertion in Caribbean-U.S. Trade. In:
J. I. Domínguez, O. E. P. Villanueva & L. Barberia (Eds.). (2004). The Cuban economy at the start
of the twenty-first century. Harvard University Press.
Stock, J., & Watson, M. (2012). Introduction to econometrics (pp. 216-245). Boston, Mass.: Pearson. Shepherd, B. (2013). The gravity model of international trade: A user guide.United Nations
publication. Bangkok.
The Economist,. (2015, Feb 12). Sanctions against Russia; Fancy Footwork. Retrieved from http://www.economist.com/news/business/21643122-how-businesses-linked-blacklisted-oligarchs-avoid-western-sanctions-fancy-footwork
Torres, F. (1994). Portugal. In J. Williamson (Ed.), The political economy of policy reform (pp. 141-153). Washington, DC: Peterson Institute.
Trefler, D. (1993). International factor price differences: Leontief was right!.Journal of political
Economy, 961-987.
Viner, J. (1950). The custom unions issue. Carnegie Endowment for World Peace. New York. Vos, R., Ganuza, E., Morley, S., & Robinson, S. (2006). Who gains from free trade?. London:
Routledge.
Wing, I. S. (2004). Computable general equilibrium models and their use in economy-wide policy analysis. Technical Note, Joint Program on the Science and Policy of Global Change, MIT.
27
Appendix 1 – Graphs
-20 -15 -10 -5 0 5 10 15 20 25 2011 2009 2007 2005 2003 2001 1999 1997 1995 1993 1991 1989 1987 1985 1983 1981 1979 1977 1975 1973 1971Cuba Economic Growth 1971-2011*
GDP per Capita annual growth % GDP annual growth %
*Data from World Bank
Graph 1
42% 19% 20% 10% 5% 3% 1%Cuba Trade partners 2011 *
Venezuela
Americas, excluding Venezuela Europe
China
Asia, excluding China Africa
Oceania
Graph 2
28 -10 -5 0 5 10 15 20
Portugal Economic Growth 1970-2000 *
GDP per Capita annual growth %
GDP annual growth %
*Data from World Bank
29
Appendix 2 - Results
Table 1
Potential Trade Flows Estimation Cuba – United States
Relevant coefficients from Appendix 3 Recent variablesImport
coefficients Output Import
Export
coefficients Output Export
Montenegro & Soto (2000) Table 5
constant 1 -9,61 -9,61 -8,68 -8,68
distance (in miles, Havana to Miami) 227,76 -1,99 -453,2424 -2,47 -562,5672
GDP Cuba (in thousands US Dollar) (2011) 68.233.900 0,91 62092849 1,41 96209799
GDP per capita Cuba (in US$) (2011) 6051 0,57 3449,07 0,5 3025,5
GDP U.S. (in thousands US$) (2011) 15.517.900.000 1,9 29.484.010.000 1,62 25.138.998.000
GDP per capita U.S. (in US Dollar) (2011) 49803 -0,12 -5976,36 0,12 5976,36
area U.S. (in thousands of square miles) 3717,792 -0,44 -1635,82848 -0,46 -1710,18432
Linder (absolute difference in GDP per capita) 43752 0,26 11375,52 0,25 10938
Total value of potential future trade flow in US$
29.546.109.598,55
25.235.225.457,43
30
Table 2
Potential Trade Creation Estimation Cuba – United States
*Data collected from Oficina Nacional de Estadísticos – Cuba 2011, Sections 8.5 and 8.6
11
Follows from Table 1
12
Montenegro & Soto (2000) Table 6 – Caribbean Countries model 1986-1991
Cuba - United States (US$)
Cuban imports
Cuban exports
Potential future imports Cuba - U.S.
1129.546.109.599
Potential future exports Cuba - U.S.
1125.235.225.457
Cuba total imports (2011) excl. U.S.*
13.525.578.413
Cuba total exports (2011)*
6.041.000.000
Estimated diversion rate
1276.8%
Estimated diversion rate
1283.5%
Total diverted imports
10.387.644.221
Total diverted exports
5.044.235.000
31
Appendix 3 - References
Claudio Montenegro & Raimundo Soto
How Distorted is Cuba’s Trade (2000)
TABLE 5
TOBIT ESTIMATES OF THE GRAVITY MODEL (1986-91)
General Model
Caribbean Countries Model
Variables Imports Exports Imports Exports
Coef. t-stat Coef. t-stat Coef. t-stat Coef. t-stat
