OBSERVARE
Universidade AutĂłnoma de Lisboa
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023)
76
ECONOMIC GROWTH IN LATIN AMERICA AND THE ROLE OF CHINA.
AN ANALYSIS OF LATIN AMERICAN NEOSTRUCTURALISM
RAFAEL GUSTAVO MIRANDA DELGADO
rafaelgustavomd@hotmail.com
Professor and co-founding director of the Research Group on Development Studies and
Democracy (GISEDD) at the Universidad de Los Andes Venezuela (Venezuela). Post-doctorate in
Social Sciences from the Universidad Central de Venezuela, PhD in Political Science and
International Relations from the Universidad Ricardo Palma de PerĂș, MA in International Relations
from the Universidad Andina SimĂłn BolĂ­var de Ecuador, Economist from the Universidad de los
Andes Venezuela. Author of more than thirty articles on Latin America published in Latin America,
Europe, Asia and North Africa.
ANTONIO AZUAJE
antonioaz28061@gmail.com
Guest researcher at the Research Group on Development Studies and Democracy (GISEDD) at
the Universidad de Los Andes Venezuela (Venezuela). Economist from the Universidad de Los
Andes Venezuela, Master in Economic Development from the Universidad Estadual de Campinas
Brazil and MBA in Data Science and Analytics from the University of SĂŁo Paulo.
Abstract
This research aims is to analyze the role that the Chinese economy has had in the economic
growth of Latin America. The methodology used consists of a multivariate approach for time
series, and causal relationships are obtained through the impulse response analysis functions.
It is stated that China's economic growth, in general, has been positive for the region but it
has also had negative effects, such as the re-primarization of Latin American economies. For
this reason, for Latin America to realize the full potential that Chinese economic growth has
for the region, it must apply policies that generate structural change.
Keywords
Economic growth; Neo–structuralism; Latin America; China; Global economy
Resumen
El objetivo de esta investigaciĂłn es analizar el rol que ha tenido la economĂ­a china en el
crecimiento económico de América Latina. La metodología utilizada consiste en un enfoque
multivariado para series de tiempo, y se obtienen relaciones de causalidades a través del
anĂĄlisis de las funciones impulso respuesta. Se afirma que el crecimiento econĂłmico de China
en general ha sido positivo para la región, pero este también ha tenido efectos negativos,
como la re – primarizacion de las economĂ­as latinoamericanas. Por ello, para que AmĂ©rica
Latina haga efectivo todo el potencial que tiene para la regiĂłn el crecimiento econĂłmico chino,
debe aplicar polĂ­ticas que generen un cambio estructural.
Palabras claves
Crecimiento econĂłmico; Neo–estructuralismo; AmĂ©rica Latina; China; EconomĂ­a global
How to cite this article
Delgado, Rafael Gustavo Miranda; Azuaje, Antonio (2022). Economic growth in Latin America and
the role of China. An analysis of Latin American Neostructuralism. Janus.net, e-journal of
international relations, Vol13 N2, Novembre 2022-April 2023. Consulted [on line] in date of last
view, https://doi.org/10.26619/1647-7251.13.2.3
Article received on 29 June 2022, accepted for publication on 19 de September 2022
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
77
ECONOMIC GROWTH IN LATIN AMERICA AND THE ROLE OF
CHINA. AN ANALYSIS OF LATIN AMERICAN NEOSTRUCTURALISM
RAFAEL GUSTAVO MIRANDA DELGADO
ANTONIO AZUAJE
Introduction
The analysis of economic growth consists of identifying, within the simultaneous evolution
of variables which are the ones that play the leading role and those that are modified
laggardly. The neo-structuralist school represents a reflexive effort that offers a
theoretical and analytical framework to understand economic growth, its explanatory
variables, and the policies that promote it.
Nowadays, one of the most interesting phenomena at a global level is the pluralization
of the centers of economic growth, with China being one of the most relevant emerging
economies. China has had significant economic growth in recent years, which has also
been able to improvement the entire global economy with its trade, investment, and
changes in the international political economy.
Therefore, the aim of this research is to analyze the role that the Chinese economy has
had in the economic growth of Latin America. The methodology used consists of a
multivariate approach for time series. It seeks to estimate cointegration vectors between
the variables studied that signify a stable relationship between them, interpreting such
relationships as long-term connections. Additionally, causal relationships are obtained
through the impulse response function analysis.
The research is presented in three parts. In analytical elements, it reflects Latin American
neo-structuralism on the influence of the Chinese economy in Latin America, emphasizing
the main channels of influence: investment, trade, and terms of trade. In the second
part, descriptive analysis, we seek to measure the associations between Latin America
economic growth and China economic growth, choosing as explanatory variables of the
Gross Domestic Product (GDP) of Latin America and the Caribbean, the Gross Domestic
Product (GDP) per capita of China, Foreign Direct Investment, Gross Fixed Capital
Formation and the economic openness index. Finally, in econometric analysis, the
multivariate analysis for time series is presented to estimate the degree of association
between the GDP of Latin America and the Caribbean with the GDP of China and the
other selected variables, and causal relationships are obtained through the impulse
response functions analysis.
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
78
Analytical elements
The Latin American structuralism and neo-structuralism schools represent a reflexive
effort that offers a theoretical, analytical, and categorical framework. From the Latin
American particularities look forward to the generation of knowledge that responds to
the main economic challenges of the region, is one of the contributions distinctive the
idea of structural change.
The fundamental argument that distinguishes the structuralism and neo-structuralism
schools is the warning that economic growth is not indifferent to the economic structure.
There are productive sectors that have a greater capacity to generate economic growth,
and structural change precedes economic growth.
For Ocampo (2005), economic growth in developing countries is intrinsically linked to the
dynamics of productive structures and the policies and institutions created specifically to
support them. The dynamics of production structures play a fundamental role in changing
the pace of economic growth. Economic growth is an essentially mesoeconomic process
determined by the dynamics of production structures. This dynamic of productive
structures depends on the interaction between innovation, understood as all kinds of new
activities and new ways of carrying out existing activities, and the complementarities,
linkages, and networks between companies and productive activities and their respective
learning processes. The State with the economic policy can affect the productive structure
and create comparative advantages. To generate growth must be constantly created
dynamic productive activities.
The productive structure is configured by market incentives and by the policies that
countries may adopt. The international economy plays a fundamental role due to the
forces of its incentives.
Nowadays, China rise.is one of the most significant variables to understand the economic
world. Since the end of the 1970s, began to implement strategies of economic reform
and opening, promoting an average annual growth of 9.9 percent and an annual increase
in international trade of 16.3 percent over the next thirty years. China overtook Japan in
2010 as the world's second-largest economy and replaced Germany as the world's largest
exporter. Since 2010, China has contributed nearly 1 percentage point per year to the
global GDP growth rate. In 2016, China contributed more than 15 percent of the world's
GDP, ranking as the second largest economy after the United States. It also has the
largest industrial GDP in the world with 22.5 percentage points; it is the largest
agricultural producer in the world with 30 percent of the value added of world agricultural
activity; it is the second economy in terms of final consumption of households with 9.6
percent. China is the world's leading exporter and the second largest importer. And in
2018, it contributed 15.8 percent of world GDP (World Bank, 2020).
Chinese economic growth has been especially resilient. For example, during the global
financial crisis of 2008, China had ample fiscal space and abundant foreign exchange
reserves. China's economy started to recover in the first quarter of 2009, its growth rate
for this year reaching 9.1 percent and 10.1 percent in 2010. China's solid growth during
the crisis was the strongest driving force for global recovery (Lin, 2010). At the time of
writing this article, China is the country that has presented the best recovery after the
COVID-19 crisis, and once again, it is emerging with the potential to drive world economic
growth.
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
79
China's economic growth since its reform and opening has had a significant impact at the
global level, benefiting economies with an abundance of natural resources, such as those
of Latin America, via Foreign Direct Investment, terms of trade, and international trade.
Currently, China is the second largest investing country in the region and Latin America
is the second region recipient of Chinese Foreign Direct Investment (FDI), receiving
around 15 percent of total FDI from this country. Of this FDI received, 80 percent is
concentrated in Brazil, Peru, and Argentina. China has also become the leading banker
in Latin America. The China Development Bank and the China Export-Import Bank have
overtaken the World Bank and the Inter-American Development Bank in the region. The
accumulated loans in the period 2005 - 2017 have reached 150 billion dollars (MMDD),
highlighting those received by Venezuela, Brazil, Ecuador, and Argentina. On the other
hand, a diversified set of cooperation mechanisms has been created, such as the
Infrastructure Fund, the Special Fund for Agriculture, and the Program for Scientific-
Technological Associations. It should be noted, that more than two thousand Chinese
companies have been founded in Latin America. (Berjano, 2019; Rios, 2018; Gallagher
and Myers, 2017; Chen and Li, 2017; Detsch, 2018).
FDI has been concentrated on projects and acquisitions in natural resource-intensive
sectors such as mining, oil, and gas. The composition of Chinese FDI in the region shows
that in the period 2010-2014, almost 90 percent was directed to natural resources, in oil
and gas extraction. China is among the most important foreign investors in Argentina,
Brazil, Colombia, Ecuador, Peru, and Venezuela. In mining, China has concentrated its
investments in Peru and a lesser extent in Brazil. Chinese FDI in the agricultural field
remains limited but shows a growing trend and the consolidation of large global players
operating in the region, such as the Chongqing group (GGG), is observed. Finally, in
recent years, Chinese investments have been increasingly directed towards
telecommunications, the automotive industry, and non-conventional energies (CEPAL,
2018).
The oil absorbs most of the Chinese financing in Latin America and its main instruments
to guarantee the supply of oil are direct investments by Chinese public companies and
the financing of Chinese public banks with a counterpart in oil. Next, copper and iron are
the sectors with the highest Chinese FDI in Latin America. In the case of soybeans, due
to legal difficulties in purchasing land, the strategy was to acquire two international
trading companies already present in the region and that control almost the entire value
chain, to seek to transform them into a large Latin American operator (Gallagher, Irwin,
and Koleski, 2013; Cheny Perez - Ludeña, 2014).
Additionally, the most significant project worldwide is the one promoted by China called
the Belt and Road initiative, which is of strategic importance to Latin America. This
initiative has important complementarities and spaces for cooperation. China has a
special interest in guaranteeing access to the region's natural resources and Latin
America in attracting FDI in strategic areas such as infrastructure and communication.
To realize the full Chinese FDI potential to generate growth in the region, the Latin
American economies must promote projects that allow the region's economies to join the
productive chains promoted by China; not only as suppliers of raw materials but also
identify activities with the highest added value, which promote investments and
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
80
technological alliances. Otherwise, the logic of the market and the reprimarization of
Latin American economies will continue to prevail.
The rise of China in the world economy has also affected the terms of trade, increasing
the prices of raw materials and energy resources, and reducing the prices of simple
manufacturers, which has had diametrically different effects in the Latin American sub
regions. South America has benefited from the high demand for raw materials, minerals,
and energy products, and the high supply of manufactured goods, which has improved
its terms of trade. For their part, the Central American countries have been affected by
the deterioration in the terms of trade due to this same situation; since they are net
importers of oil and net exporters of manufacturing. While Mexico was surpassed by
China as the second trading partner of the United States; especially due to the
displacement of its manufacturers. Central American manufacturers have also not been
able to compete with Chinese manufacturers in the US market (Berjano, 2019; Caputo,
2005; CEPAL, 2004).
Miranda (2021) warns that increases in the prices of natural resources produce incentives
for the re-primarization of productive structures; via international prices and the real
exchange rate. Economic growth is anchored on the volatility of international prices of
natural resources. Thus, the abundance of natural resources, as is the case in most of
Latin America, is an additional argument in favor of policies for structural change.
For his part, Palma (2005) points out that the process of early deindustrialization in Latin
America has been generated by changes in industrial policies and the drastic process of
trade and financial liberalization that brought economies back to their traditional
comparative advantages, to their natural Ricardian position. And Ocampo (2005)
highlights, that the Latin American economic growth of the 2000s was driven by the
increase in the prices of raw materials, especially hydrocarbons and mining products, and
by the massive inflow of capital during two periods of exuberance in international financial
markets, between mid-2004 and April 2006, and especially between mid-2006 and mid-
2007; which allowed rapid growth and simultaneously generated a current account
surplus, but without structural change.
This economic growth does not possess the qualities of growth during the period of state-
led development between the end of World War II and 1980, which was driven by
manufacturing and increased productivity.
For this reason, Ocampo (2005) distinguishes between deep structural change and
superficial structural change. The first is characterized by an intense learning process
and a high degree of development of complementarities, and therefore, by strong
dynamic economies of scale of a microeconomic and mesoeconomic nature. While the
second is characterized, by a low level of learning and a scarce development of productive
complementarities, such as the development of enclaves dedicated to natural resource
export activities.
For example, between the end of World War II and 1980, Latin America grew more than
the world average, achieving the highest growth in the entire history of the region: 5.5
percent per year and 2.7 percent per capita. The engine of economic growth was the
manufacturing industry sector; productivity also reached the highest levels in history,
estimated that the GDP per worker increased by 2.7 percent per year between 1950 and
1980, and it was a period of greater economic stability. The model of economic policies
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
81
for growth in this period was characterized by being a mixed model. Combined import
substitution with export promotion and regional integration, due to the growing attention
to industrialization, the significant expansion of the spheres of action of the State in
economic life; through the creation of public companies and in the development of some
industrial sectors (BĂ©rtola and Ocampo, 2014).
Thus, it can be affirmed that productive development policies for structural change have
been essential for economic growth. Incentives must be generated to channel
investments where the long-term benefits are strongest and change the structure of
relative returns in favor of more complex sectors. Productive diversification and
complexity allow the reallocation of productive factors to new activities, the addition of
value to production processes, and the promotion of productive chains, that allow the
diffusion of technology, the homogenization of productivity, and the reduction of
inequality and poverty.
Productive development policies consist of a particular way of affecting the economic
structure, a selective way, which intentionally seeks to favor, over others, a particular
industry. Successful productive development policies have included assistance in
capturing and adapting foreign technology, creating comparative advantages, protecting
international competition, promoting exports, coordinating and complementing financial
markets, promoting economies of scale, and regulating foreign direct investment (Chang,
2012; Rodrik, 2013).
In commercial matters, although South America presented a complementary productive
structure with China and have a trade surplus; Mexico and Central America presented a
competitive structure and maintained trade deficits.
Bilateral Chinese-Latin American trade in 2017 totaled 257.8 million dollars, exports to
China were 130.8 million dollars, and imports from China were 127 million dollars. China
is the first commercial partner of Brazil, Chile, and Peru; and the second of most of the
countries in the region. Exports to China are focused on raw materials, almost exclusively
energy and mining for manufacturers. China's degree of dependence on imports of
natural resources from Latin America, measured as a ratio between net imports and
consumption, already reaches 60 percent in the case of major commodities, such as oil,
copper, iron, and in the case of soy rising to 85 percent (Berjano, 2019; Rios, 2018;
Rosales and Kuwayana, 2012).
Latin America is relevant to China in the supply of various metallic minerals. Exports of
Latin American minerals to China increased from 1.7 million tons in 2000 to almost 220
million tons in 2015. Of this amount, Brazil's contribution was 192 million tons, Peru's 11
million tons, and Chile's 10 million tons. Additionally, it is estimated that the Chinese
demand for minerals will continue to increase since it is especially tied to the urbanization
process of the Asian country, which is estimated to increase by 70 percent by the year
2030. In terms of oil, exports to China reached 854,000 barrels per day in 2015, figures
corresponding to 13 percent of the continent's total exports to China and 8 percent of
Chinese oil consumption. About 91 percent of this amount originated in Venezuela
(CEPAL, 2018; World Bank, 2016).
Sustained economic growth requires structural change and a productive transformation
that incorporates and spreads technological progress. For this, the primary sector must
not only transfer income to other sectors but must also articulate productive linkages
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
82
with the industrial and service sectors. However, these synergies do not occur
spontaneously or due to market dynamics. So, specific policies are needed for the
generation, dissemination, and incorporation of knowledge into production. Policies for
the selective promotion of exports, granting of government incentives to those who
undertake innovative activities, support for the creation of high-tech companies, and
completion and adapt the technological infrastructure in the less advanced priority
sectors (CEPAL, 1995; CEPAL, 1996).
Trade with China is persistently in deficit for the majority of the region, which has
worsened since 2011, especially in Mexico and Central America. Only three countries in
the region register a surplus with China, Brazil, Venezuela, and Chile. It is an inter-
industrial exchange where the Latin American region practically exclusively exports
unprocessed natural resources and imports a wide range of manufactured goods. There
is a marked deficit with China in the manufacturing market since manufacturing exports
to China are very low, except in the case of Costa Rica and Mexico. The most exported
products to China from the region are oil, iron, copper in different forms, soybeans,
metals, wood, and sugar (CEPAL, 2018; CEPAL, 2016).
As the pioneering works of Prebisch (1973) already warned, in Latin America, there is an
asymmetry between the low dynamism of the demand for primary products that it
exports, compared to the wide demand for imported industrial products. This damages
the terms of trade and generates a structural imbalance in the balance of payment.
Additionally, China is substituting the import of processed goods with its capacity, which
has eroded the contribution of Latin American countries to the value chain. Economies
such as Bolivia, Ecuador, and Uruguay practically do not add value to their main products
exported to China: precious metals, fruits, and soybeans, respectively. In the case of
Argentine and Brazilian exports of soybeans and their derivatives, the percentage of
exported products with some level of processing fell significantly from 2004 to 2014. For
its part, the percentage of Venezuelan refined oil has decreased notably in total exports
of petroleum products. Only in the case of copper, the percentage of exported refined
material has remained stable for Chile and Peru. (CEPAL, 2016).
Sustained economic growth requires a structural change, a basket of varied exports that
are positioned in more complex sectors. Exports to China do not meet these conditions.
On the contrary, they accentuate the re-primarization of the region's economy. Processed
products have a minimal share in the current export basket of the region to China, and
the manufacturing sector has been reduced in the domestic market by international
competition, particularly Chinese competition. For Latin America to take advantage of the
potential of trade with China, it must apply productive development policies that do not
follow the short-term incentives of the market.
In agriculture, Latin America has significantly increased its weight as China's agricultural
supplier. The region's portion of Chinese imports of agricultural products went from 16
percent in 2000 to 27 percent in 2015. In this year, 2015, the region surpassed the joint
share of the United States as a supplier of Chinese imports. The United States and Canada
reached 26 percent which was much higher than the shares obtained by other relevant
competitors such as the Association of Southeast Asian nations with 15 percent, and
Austria and New Zealand with 11 percent. However, although the region as a whole has
increased its weight as an agricultural supplier to China, the growth of regional exports
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
83
has been represented almost entirely by a single basic product and a single supplier,
soybeans from Brazil (CEPAL, 2016).
Agricultural exports to China have significant potential to add value. For this, it is
necessary to know and satisfy the requirements of the Chinese consumer and the
regulatory requirements to access the market, such as sanitary and phytosanitary
requirements, and quality standards. This may represent a barrier to entry due to high
costs. However, States can offer industries and companies this information and support
as a public good.
It should be noted, that the technological and structural complexity gap between China
and Latin America has been widening. From 1995 - 2014, the complexity of China's
economic structure increased significantly. Industrial policy strategy focused on
investment in new technological sectors and areas of greater demand for knowledge such
as the digital economy. This has made it possible to reduce the productivity gap with the
most advanced economies and to develop new technological capabilities in areas such as
the Internet, data storage, metadata analysis, robotics, and artificial intelligence. While
in the principal economies of Latin America, such as Brazil, Mexico, and Argentina, the
complexity of the productive structure lagged, and this behavior has spread to the rest
of the countries of the region (CEPAL, 2018).
As Chang (2020) warns, economic theories and empirical evidence show that although
in the short term there is a certain probability that free trade allows all partners to
maximize their production and income, in the long term it harms the economic
development of the less developed partners, for whom it is impossible to establish
technologically advanced and highly productive industries if they have to compete with
the most advanced producers in the most economically developed countries. In the long
run, free trade between countries at different stages of economic development is
detrimental to less developed countries.
Thus, what intrinsically hinder sustained economic growth in the region is not the
incentives generated by FDI, the terms of trade or trade with China, but rather the lack
of productive development and foreign trade policies that generate structural change
with greater diversity and technological content.
Additionally, it is often sought to measure the associations between Latin America's
economic growth and China's economic growth.
Descriptive analysis
For the economic growth of Latin America and the Caribbean, the Gross Domestic Product
(GDP) per capita of all the countries that make up the region was considered as an
aggregate; this was obtained by dividing the GDP by the population at mid-year. GDP is
the sum of the gross value added of all resident producers in the economy plus all taxes
on products, less any subsidies not included in the value of products. The data was
obtained from the World Bank (2022) and is expressed in current dollars. In addition,
the natural logarithm was applied to statistically improve the model and observe the
convergence of growth between the countries of Latin America and the Caribbean as an
aggregate. The economic growth of Latin American countries is our variable to analyze
or the dependent variable.
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
84
Broadly, Latin America and the Caribbean countries present an average of 4,383 million
dollars in the selected period (1970-2020). This period is divided into half, where the
GDP registered its lowest values: 613 million dollars in 1970 to 3,760 billion in 2002. In
the second half (2002-2020), the region registered the highest levels of GDP; reaching
the figure of 10,433 billion dollars in 2014. The coefficient of variation is 70 percent,
indicating that there is a high degree of dispersion in what refers to the distribution of
GDP in the region in the period under study.
Table 1. Gross Domestic Product of Latin America and the Caribbean (GDP-LATAM)
Average
Deviation padron
CV%
Minimum
Quartil 1
Median
Maximum
4.383
3.065
70%
613
1.889
3.760
10.433
For Chinese economic growth, China's Gross Domestic Product (GDP) per capita was
considered. The data was obtained from the World Bank (2021) and is expressed in
current dollars. Additionally, the logarithm was applied to improve the model statistically.
Broadly, China presents an average of 3,026 trillion dollars in the selected period (1970-
2020). This period is divided into half where the GDP registered its lowest values of
92,602 billion dollars in 1970 to 734,547 billion in 1995. The second half (1995-2020)
registered the highest levels of China's GDP, reaching the figure of 14.279 trillion dollars
in the 2019. The coefficient of variation is 144 percent, indicating that there is a high
degree of dispersion in what refers to the distribution of GDP in China during the period
under study.
Table 2. Gross Domestic Product of China (GDP-CHINA)
Average
Deviation padron
CV%
Minimum
Quartil 1 1
Median
Quartil 3
Maximum
3.026.912.246.789
4.351.030.925.221
144%
92.602.973.434
17.888.223.506
734.547.898.221
4.072.324.884.836
14.279.937.467.431
Foreign Direct Investment (FDI) is the sum of equity capital, reinvested earnings, other
forms of long-term capital, and short-term capital as described in the balance of
payments. This series reflects the net total, that is, net FDI in the reporting economy
from foreign sources less net FDI of the reporting economy in the rest of the world. This
series reflects net inflows into the economy report and is divided by GDP. The data is
expressed in dollars at current prices. The variable was obtained from the World Bank
(2022) and is the aggregate of Latin America. Additionally, the logarithm was applied to
improve the model statistically.
Broadly, Latin America and the Caribbean countries present an average of 87.365 billion
dollars of FDI in the selected period (1970-2020). This period is divided into half where
FDI registered its lowest values, 918,606 million dollars in 1972, up to 30,168 billion in
1995. The second half (1995-2020) saw the region register its highest levels of FDI,
reaching the figure of 343.499 billion dollars in 2013. The coefficient of variation is 119
percent, indicating that there is a high degree of dispersion in what refers to the
distribution of FDI in the region in the period under study.
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
85
Table 3. Foreign Direct Investment in Latin America and the Caribbean (FDI)
Average
Deviation padron
CV%
Minimum
Quartil 1
Median
Quartil 3
Maximum
87.365.222.156
107.594.401.387
119%
918.606.901
5.439.428.290
30.168.276.207
160.087.884.767
343.499.295.040
Gross Fixed Capital Formation includes land improvements, the acquisition of facilities,
machinery, and equipment, and the construction of roads, railways, and related works,
including schools, offices, hospitals, private residences, and buildings of commercial and
industrial. Data is in US dollars at current prices. The variable was obtained from the
World Bank (2021) for Latin America as an aggregate. Additionally, the logarithm was
applied to improve the model statistically.
Broadly, Latin America and the Caribbean countries present an average of 469,220 billion
dollars, referring to FBC in the selected period (1970-2020). This period is divided into
half where the FBC registered its lowest value of 35,504 billion dollars in 1970 to 365,494
billion in 1995. In the second half (1995-2020), the region registered its highest FDI
levels, reaching the figure of 1,314 trillion dollars in the 2013. The coefficient of variation
is 85 percent, indicating that there is a high degree of dispersion in what refers to the
FBC distribution in the region in the period under study.
Table 4. Gross Capital Formation in Latin America and the Caribbean (FBC)
Average
Deviation padron
CV%
Minimum
Quartil 1
Median
Quartil 3
Maximum
469.220.807.100
400.501.398.655
85%
35.504.768.350
146.236.493.229
365.494.167.637
812.474.693.594
1.314.006.550.235
The economic openness index is an indicator that measures the degree of openness of a
country's economy, considering its foreign trade to its global economic activity as a
whole. It is the result of the sum of imports of goods and services plus exports of goods
and services, divided by GDP at buyer's prices, all at current prices in dollars. The variable
was obtained from the World Bank (2021) for Latin America as an aggregate.
Additionally, the logarithm was applied to improve the model statistically.
Broadly, Latin America and the Caribbean countries present an average of 0.38 points
regarding economic openness in the selected period (1970-2020). This period is divided
into half, where economic openness registered its lowest values, 0.27 points in 1970,
down to 0.34 points in 1988. The second half (1988-2020) saw the region register its
highest levels. of economic openness reaching the figure of 0.51 points in 2018. The
coefficient of variation is 21 percent, indicating that there is a relatively stable degree of
dispersion in what refers to the distribution of economic openness in the region during
the study period.
Table 5. Economic Opening of Latin America and the Caribbean (APERT-ECO)
Average
Deviation
padron
CV%
Minimum
Quartil 1
Median
Quartil 3
Maximum
0,38
0,08
21%
0,27
0,31
0,34
0,45
0,51
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
86
Econometric analysis
The methodology used consists of a multivariate approach for time series. It seeks to
estimate cointegration vectors between the variables studied that signify a stable
relationship, interpreting such relationships as long-term connections. Cointegration is a
multivariate procedure suitable for the time series treatment, considering the possibility
of stochastic trend's existence in the series because it results in a relational equation of
magnitudes in level.
With this approach, it will be possible to estimate the degree of association between the
GDP of Latin America and the Caribbean (our target variable) with the GDP of China,
Foreign Direct Investment in the region, Gross Capital Formation in the region, and finally
the Economic Opening in the region, during the period 1970-2020.
With this methodology, it is also possible to estimate the short-term adjustment
dynamics resulting from variations in the GDP of Latin America and the Caribbean and
the other variables under study, leaving visible the short- and long-term impacts on
variables of interest derived from standardized exogenous shocks, through the analysis
of impulse response functions.
Before applying the time series cointegration methods, it is essential to verify certain
characteristics, such as the homoscedasticity and stationarity of the variables and their
degree of integration. The verification of the heteroscedasticity pattern of the series was
carried out by graphical inspection of the series in second difference (See Appendix 1).
Additionally, cointegration methods can only be applied to stationary or first-order
integrated variables, so the unit root test was applied to each of the variables and the
complete model, to know their integration properties (See appendix 2). Finally, the
methodology of Johansen (1991) is used to perform a test to calculate the rank of matrix
II = 〖αÎČ〗^ÂŽ. Cointegration tests allow for verifying the long-term relationship between
economic variables. Johansen's (1991) cointegration test has been used because it
determines the number of cointegration vectors (See Appendix 3). By complying with
these elements; it was possible to move on to time series analysis.
To decide to work with the Vector Error Correction Model (VEC-M), we first analyzed the
behavior of the variables through the analysis of the Vector Autoregressive Model (VAR).
It should be noted that the VEC-M model is developed as an evolution of such a VAR
model. For this reason, it is essential to observe the behavior in the first instance through
the said model. The order of the VAR model was chosen by analyzing the Final Prediction
Error (FPE) information criterion, through which it was concluded that it is of order 2;
that is, this VAR model has 2 lags (see Appendix 3).
Having determined the number of unit roots and the order of the VAR model, Johansen's
(1991) methodology was applied to obtain the number of cointegration vectors.
Considering the behavior of the variables under study in the VAR model, and based on
it, the decision was made to work with the VEC-M model, through which a better
adaptation of the variables to the selection of the economic theory. We were able to
observe that using the VEC-M model, the result of the maximum eigenvalue was 3
cointegration vectors, with a significance of 5% for all models. The VEC-M model was
chosen, with equal linear trends, within the cointegration vector; since it is the most
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
87
parsimonious and includes less rigidity than the other candidate models. For this reason,
it was chosen.
Log(diff1(îŽČîŽ«îŽ€î˜ƒîŽźîŽŁîŽ¶îŽŁîŽŻî˜ƒîŽ»î˜ƒîŽ„îŽŁîŽŽîŽ«îŽ€îŽ§ï‡§ó°‡œ) = î·šï„Ž+î·šï„”+  + (îŽźî”‹î”ƒó°‡›îŽŠî”…î”‚î”‚î„Žó°‡›îŽČîŽȘîŽ«îŽ°îŽŁï‡§ó°‡œó°‡œ +
î·šï„žó°‡›îŽźî”‹î”ƒó°‡›îŽŠî”…î”‚î”‚î„Ž(îŽ«î”Šî”’î”î”Žî”î”…î›—î”Šî˜ƒîŽ§î””î”î”ŽîŽœî”Šî”†î”î”ŽîŽœî˜ƒîŽŠî”…î”Žî”îŽżî”îŽœï‡§ó°‡œó°‡œ + ï„čîŽźî”‹î”ƒó°‡›îŽŠî”…î”‚î”‚î„Žó°‡›(îŽšî”‹î”Žî”‰îŽœîŽżî”…î›—î”Šî˜ƒîŽ€î”Žî”‘î”îŽœî˜ƒî”€î”î˜ƒîŽ„îŽœî”Œî”…î”îŽœî”ˆï‡§ó°‡œó°‡œ +
ï„ș(îŽźî”‹î”ƒó°‡›îŽŠî”…î”‚î”‚î„Žó°‡›îŽŁî”Œî”î”Žî”î”‘î”ŽîŽœî˜ƒîŽ§îŽżî”‹î”Šî›—î”‰î”…îŽżîŽœï‡§ó°‡œó°‡œ =0
î„Čî„čî„șîŽČ = -0,55 + 0,02 + 0,11 - î„ČîŸĄî„łî„ŽîŽČîŽȘ + î„ČîŸĄî„¶î„»î˜ƒîŽ«î”Šî”’î”î”Žî”î”…î›—î”Šî˜ƒîŽ§î””î”î”ŽîŽœî”Šî”†î”î”ŽîŽœî˜ƒîŽŠî”…î”Žî”îŽżî”îŽœ +
î„ČîŸĄî„¶î„”î˜ƒîŽšî”‹î”Žî”‰îŽœîŽżî”…î›—î”Šî˜ƒîŽ€î”Žî”‘î”îŽœî˜ƒî”€î”î˜ƒîŽ„îŽœî”Œî”…î”îŽœî”ˆ +î„ČîŸĄî„łî„»îŽŁî”Œî”î”Žî”î”‘î”ŽîŽœî˜ƒîŽ§îŽżî”‹î”Šî›—î”‰î”…îŽżîŽœ =0
With the model offered, a stable relationship between the variables studied can be observed.
However, it is not yet possible to define causal relationships. These relations are obtained through
the analysis of the impulse response functions that will be developed later.
The VEC-M model complied with the stability condition. The roots of the characteristic polynomial
and its modules are all less than unity. The model is stationary; it does not have unit roots and
therefore the estimators are consistent (see Appendix 4).
The correlogram (see Appendix 5) shows good behavior of the residual correlations. The residuals
normality analysis the of the VEC-M model showed that with a significance level of 5%, the
residuals have a normal distribution. The normality of the residuals is verified in all the joint and
individual tests of asymmetry and kurtosis. The normality of the residuals for the VEC-M model
only gives more confidence for the hypothesis tests of the coefficients, but it is not fundamental.
The LM (see Appendix 6) and Portmanteau (see Appendix 7) tests corroborate the absence of
serial autocorrelation. Since this bloom is the best fit for these aspects, we will keep and assume
that the non-contemporary residual autocorrelation is resolved.
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
88
Finally, impulse response functions help to quantify how much an impact on one variable affects
the others over time. This impulse response analysis is used to investigate the dynamic
interactions between endogenous variables and is based on the Wold moving average
representation of a VAR(p) process. On the other hand, the impulse response function is used to
describe the reaction of the system of equations under study as a function of time or as a function
of some other independent variable that parameterizes the dynamic behavior of the system
(Pfaff, 2011).
Next, the impulse response function is presented graphically, through which we can
observe how the variations or shocks of each of the series under study have influenced
the GDP of Latin America and The Caribbean, where the red lines represent the
confidence interval estimated for each of them.
A shock on the GDP of Latin America and the Caribbean determines a behavior on the FDI series
in the region that begins at a high level; over the years, the influence grows positively until it
stabilizes in period 8. We can observe that the GDP of Latin America and the Caribbean influences
positively and in a large proportion on the growth of FDI in the region.
The impact of a shock on the GDP of Latin America and the Caribbean determines a behavior on
the economic opening series in the region that is practically non-existent, the confidence interval
is around zero, that is to say, an increase or a fall in the region GDP does not directly influence
the rise or fall of economic openness. On the other hand, a shock on the GDP of Latin America
and the Caribbean determines a behavior on the GDP series of China that at the beginning of our
series is at a low level; but with time, its influence increases, maintaining a growing trend,
indicating that an increase in the region's GDP contributes to an increase in China's GDP.
Graph 1. PIB-LATAM
Finally, a shock on the GDP of Latin America and the Caribbean determines a behavior on the
gross capital formation series of the region, that begins at a high level, tends to decrease over
time, and then tends to stabilize, we can highlight that throughout our series under study, a
positive but not directly proportional relationship has been established between both indicators,
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
89
that is, large variations in the region's GDP will not have the same proportionality in an increase
or decrease in the region's gross capital formation.
Graph 2. PIB-CHINA
A shock on theChina GDP determines a behavior on the GDP series of Latin America and
the Caribbean. Initially begins at a very low level, but over the years that influence takes
on a more relevant character; that is, as the years have passed in the period 1970-2020,
China's GDP growth has positively influenced the region's GDP growth..
Graph 3. IED
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
90
A shock on IED determines a behavior on the GDP of Latin America and the Caribbean
that begins at a low level. With the passing of the periods, its influence becomes
increasingly important, that is to say, as the years pass in the period under study, IED
has had a positive impact on the region's GDP growth.
Graph 4. FORMACIÓN BRUTA DE CAPITAL
A shock on the region's gross capital formation determines a behavior on its GDP series
that begins at a low level, as the periods go by it is influence begins to increase, and
from period 6 it tends to stabilize. In other words, over the years a positive and stable
influence between both indicators has been established..
Graph 5. APERT ECO
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
91
A shock on the economic opening of the region determines a negative behavior on its
GDP series, highlighting that it remains constant over the years in the same patamar. In
other words, throughout the period under study, the economic opening of the region has
not played a determining role in terms of its growth.
Next, the variance decomposition function is worked on. Pfaff (2011) explains the
decomposition function as a matrix-based method of the orthogonal impulse response
coefficient Κn, where the variance indicates in what proportion of the variability of each
of the variables that are part of the proposed model explains the variability of the variable
under study.
In this case, it is observed in what proportion the variation of the variables PIB LATAM,
PIB CHINA, IED, FBC, and APERT ECO explain the variability of PIB LATAM at the moment
in which our model tends to stabilize.
The variance decomposition function of the model is presented below:
At instant 1, the variation in LATAM PIB is fully explained by its variation. The model
tends to stabilize at instant 18 where the variation of the Latin America and the Caribbean
PIB is explained in 82.9% by its variation; 2% by the variation in the China PIB; 9.3%
by the IED variation; 4.2% due to the variation in economic openness and finally 1.7%
due to the variation in gross capital formation.
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
92
Conclusions
China's economic growth has had a positive impact on Latin America's economic growth.
Through the analysis of the impulse response functions, it can be stated that during the
period 1970-2020, China's GDP growth has positively influenced the region's GDP growth
and that this influence has been greater over the years. According to the results of the
variance decomposition function of the model, the variation of Latin America and the
Caribbean GDP is explained by 2 percent of the GDP China variation.
Broadly, China's economic growth has been positive for the region; but it has had adverse
effects, too. The main adverse effect is the re-primarization of Latin American economies.
Which inhibits sustained growth, and that has been a consequence of the concentration
of Chinese Foreign Direct Investment in natural resource-intensive sectors such as
mining, oil, and gas; of price increases of natural resources and their short-term
incentives; and of trade where the region practically exclusively exports unprocessed
natural resources.
For this reason, for Latin America to realize the full potential that Chinese economic
growth has for the region; it must apply policies that generate structural change with
greater diversity and technological content. Incentives must be generated to channel
investments where the long-term benefits are strongest, and that allow the productive
factors reallocation to more complex activities; and with better incorporation into the
global production chains promoted by China.
References
Akaike, Hirotugu. (1985). Prediction and entropy. En: A. C. Atkinson y S. E. Fienberg
(Eds.), A Celebration of Statistics. New York, Springer, 1–24.
Banco Mundial (2020). World Development Indicators. New York, Banco Mundial.
Banco Mundial (2016). Commodity Markets Outlook. Washington, Banco Mundial.
Berjano, Carola (2019). Globalización con “características chinas”. El creciente rol de
China en América Latina y el Caribe y sus principales desafíos. En: Pensamiento Propio,
49 - 50. 31 – 51.
Bértola, Luis y Ocampo, José (2014). Desarrollo, vaivenes y desigualdad. Una historia
económica de América Latina. Madrid. Secretaria general iberoamericana.
Caputo, Orlando (2005). Estados Unidos y China: Âżlocomotoras en la recuperaciĂłn y en
las crisis cĂ­clicas en la economĂ­a mundial? En: Jaime Estay (comp.) La economia mundial
y America Latina. Buenos Aires, CLACSO. 39 – 86-
CEPAL (2018). Explorando nuevos espacios de cooperación entre América Latina y el
Caribe y China. Santiago de Chile, Naciones Unidas.
CEPAL (2016). Relaciones económicas entre América Latina y el Caribe y China
oportunidades y desafĂ­os. Santiago de Chile, Naciones Unidas.
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
93
CEPAL (2004). Aspectos estratégicos de la relación entre China y América Latina y el
Caribe. En: Panorama de la insercion internacional de America Latina y el Caribe, 2004
tendencias 2005. Santiago de Chile, CEPAL. 151 – 188.
CEPAL (1996). TransformaciĂłn productiva con equidad. La terea prioritaria del desarrollo
de América Latina y el Caribe en los años noventa. Santiago de Chile, CEPAL.
CEPAL (1995). América Latina y el Caribe: políticas para mejorar la inserción en la
economĂ­a mundial. Santiago de Chile, Naciones Unidas.
Chang, Ha-Joon (2020). Construir un multilateralismo favorable al desarrollo hacia un
“nuevo” nuevo orden económico internacional. En: Revista CEPAL, 132 (diciembre): 67–
78.
Chang, Ha-Joon (2009). Industrial Policy: Can We Go Beyond an Unproductive
Confrontation? Seul, ABCDE (Annual World Bank Conference on Development Economics)
Chen, Yuanting y Li, Han (2017). La nueva Etapa del “Desarrollo Constructivo” de las
relaciones sino-latinoamericanas. En: Relaciones Internacionales, 53. 149 – 163.
Detsch, Claudia (2018). Escaramuzas geoestratégicas en el «patio trasero» China y Rusia
en AmĂ©rica Latina. En: Nueva Sociedad, 275 (mayo-junio). 74 – 91.
Dickey, David y Pantula, Sastry (1987). Determining the Order of Differencing in
Autoregressive Processes. In: Journal of Business & Economic Statistics, 5 (4). 455 –
461.
Dickey, David y Fuller, Wayne (1981). Likelihood Ratio Statistics for Autoregressive Time
Series with a Unit Root. En: Econometrica, 49 (4). 1057 173.
Gallagher, Kevin, Irwin, Amos y Koleski, Katherine (2012). The new banks in town:
Chinese finance in Latin America. Inter-American Dialogue Report, Washington, DC.
Gallagher, Kevin y Myers, Margaret (2017). China-Latin America Finance Database.
Washington: Inter-American Dialogue.
Johansen, SĂžren (1991). Estimation and Hypothesis Testing of Cointegration Vectors in
Gaussian Vector Autoregressive Models. En: Econometrica, 59 (6). 1551 – 1580.
Karl, Terry (1997). The paradox of the Plenty. Oil booms and Petro-States. Berkeley,
University of California Press.
Lin, Justin (2010). Demystifying the Chinese Economy. New York, Cambridge University
Press.
Miranda, Rafael (2021). Cambio estructural para la reducciĂłn de la pobreza. AnĂĄlisis
desde el neo – estructuralismo latinoamericano. En: Iberoamericana – Nordic Journal of
Latin American and Caribbean Studies, 50 (1). 19 – 27.
Miranda, Rafael (2020). Latin America looking for its autonomy: The role of extra-
hemispheric relations. En: Humania del Sur. Revista de Estudios Latinoamericanos,
Africanos y Asiáticos, 15 (29). 235 – 257.
Ocampo, José (2011). Macroeconomía para el desarrollo: políticas anticíclicas y
transformación productiva. En: Revista CEPAL, 104 (agosto): 7–35.
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
94
Ocampo, JosĂ© (2005). La bĂșsqueda de la eficiencia dinĂĄmica: dinĂĄmica estructural y
crecimiento económico en los países en desarrollo. En: José Ocampo (Ed.).Mås allå de
las reformas: dinĂĄmica estructural y vulnerabilidad macroeconĂłmica, BogotĂĄ: CEPAL. 3
– 50.
Palma, Gabriel (2005). Cuatro fuentes de “desindustrialización” y un nuevo concepto del
“sĂ­ndrome holandĂ©s. En: JosĂ© Ocampo (Ed.).MĂĄs allĂĄ de las reformas: dinĂĄmica
estructural y vulnerabilidad macroeconómica, Bogotá: CEPAL. 79 – 129.
Pfaff, Bernard (2011). Analysis of Integrated and Cointegrated Time Series with R. New
York, Springer.
Prebisch, RaĂșl (1973). Problemas teĂłricos y prĂĄcticos del crecimiento econĂłmico,
Santiago de Chile. CEPAL.
Ríos, Xulio (2018). Inflexiones y reflexiones a propósito de la relación China-América
Latina y el Caribe. En: Humania del Sur, 13 (25). 23 – 39.
Rodrik, Dani (2013). Structural change, fundamentals, and growth: an overview.
Washington. World Bank.
Rosales, Oswaldo y Kuwayama, Mikio (2012). China y América Latina y el Caribe: Hacia
una relación económica y comercial estratégica. En: Serie Libros de la CEPAL, 114
(LC/G.2519-P), Santiago de Chile, CEPAL.
Rosales, Oswaldo y Kuwayama, Mikio (2007). América Latina en el encuentro de China e
India: perspectivas y desafĂ­os en comercio e inversiĂłn. En: Revista de la CEPAL, 93. 85
– 108.
Ross, Michael (2001). Does oil hinder democracy? In: World Politics, 53 (3). 325-361.
Schwarz, Gideon (1978). Estimating the dimension of a model, En: Annals of Statistics,
6 (2): 461–464.
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
95
Appendix
1. Verification of the heteroscedasticity pattern of the series
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
96
The graphical analysis carried out on the variables under study determined the presence
of a trend in each of them. To continue with the analysis, it is necessary to eliminate its
presence. Hence, after several econometric devices application, the logarithm and second
difference were applied to the PIB-LATAM, PIB-CHINA, and APERTURA ECONOMICA
variables; for the variables INVERSION EXTRANGERA DIRECTA and APERTURA
ECONOMICA. It was not necessary to apply a second difference to control this problem.
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
97
After applying logarithms and differences to the series under study, we should not have
unit root problems, and we can also rule out the application of a model with a
deterministic trend.
2. Unit root test
A maximum of three-unit roots were assumed for each variable used in the model: PIB-
LATAM, PIB-CHINA, APERTURA ECONOMICA, INVERSION EXTRANGERA DIRECTA, and
FORMACÍON BRUTA DE CAPITAL. All variables have been linearly transformed by applying
logarithms and differences. It is highly unlikely that there are more than three roots in
the economic variables, hence the choice; also, because graphical inspection does not
show a stochastic trend of such magnitude.
Dickey & Pantula's (1987) tests for three-unit roots consist of three steps. First, the
existence of three-unit roots is analyzed. By rejecting this hypothesis, two-unit roots are
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
98
tested. Finally, a third step is to verify the presence of a single root, in case the hypothesis
of two-unit roots is rejected.
By using the LAPPLY function available in the R studio software, we can perform the unit
root test on all variables using a single algorithm.
Having ruled out the presence of 3-unit roots, we proceed directly to testing the presence
of 2 unit roots, where we will test the hypothesis of the existence of two unit roots under
the alternative existence hypothesis of only one. The lags of the dependent variable
should also be included to remove the correlation of the residuals; observed by the
correlogram analysis. The number of lags is defined by the Akaike (1985) and Schwarz
(1978) criteria; to guarantee the appropriate choice of the lag number.
The tested models have the following structural forms:
Dickey and Pantula's (1987) individual hypothesis test for the second is as follows:
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
99
The test statistics are Critical values
that were calculated by Dickey and Pantula (1987) and simulated using the R study
software. The decision criterion defines if the null hypothesis is
rejected.
Critical values are available in Dickey and Fuller (1981), according to the model
used, the level of significance, and the size of the sample. The decision criterion defines
that if: the null hypothesis that the model has a constant is rejected, if
the null hypothesis that the model has a trend is rejected.
The tests were initially executed for each of the variables and then for the complete
model, obtaining that the Tau1 statistic belongs to the critical region for each of the series
(the values of each of the roots are greater in modulus), which means that we do not
have unit roots for each of the series studied in log and with a first and second difference,
which is why we reject Ho. Our series is second (2) order integrated, that is, they have
two-unit roots and a constant.
3. Cointegration
The cointegration by the procedure of Johansen (1991) is based on the Grainger
Representation Theorem. This theorem states that if a group of variables cointegrates
with each other, then the cointegration vector has a representation in the form of an
error correction model. The error correction model is a variation of the VAR methodology.
By applying this methodology, we establish the order of the Var model.
Through the R studio software, we execute the 'both' command because a model with
constant and trend is being estimated. We apply the analysis to our database without
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
100
logarithms and differences and as a result, we can observe in the output that following
the decision criterion FPE we have a model of order 2.
Having determined the number of unit roots and the order of the VAR model, we can
apply the methodology of Johansen (1991) to obtain the number of cointegration vectors.
The calculation of the number of cointegration vectors is vital to apply the Vector Error
Correction model (VEC-M) presented in the article; such a procedure was carried out
through the R studio software.
The lambda values were 5.775979e-01, 4.949291e-01, 3.922071e-01, 1.397464e-01,
6.915117e-02, and 1.942890e-15 which represent the eigenvalues calculated for this
system. Critical values and test statistics are calculated based on these lambda values.
The values of r =0 to r <= 4 represent the null hypothesis, performing the test for r<=3,
we observe that the test statistic is 7.38, lower than the critical values 13.75, 15.67, and
20.20; therefore, we do not reject H_0, we are in the presence of 3 integration vectors.
4. Root stability analysis
The roots of the characteristic polynomial and its modules are all less than unity (1). The
model is stationary; it does not have unit roots and therefore the estimators are
consistent.
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
101
5. Model residuals
The serial correlation was not detected in the applied model.
6. LM test
We do not correlate the lags up to lag number 9, p-value=0.1717 >0.05.
JANUS.NET, e-journal of International Relations
e-ISSN: 1647-7251
Vol. 13, NÂș. 2 (November 2022-April 2023), pp. 76-102
Economic growth in Latin America and the role of China. An analysis of Latin American Neostructuralism
Rafael Gustavo Miranda Delgado; Antonio Azuaje
102
7. Portmanteau test
Through the Portmanteau test, we can ensure the absence of serial correlation up to lag
number 15, p-value= 0.5118 > 0.05.