MINISTRY OF EDUCATION AND TRAINING
UNIVERSITY OF ECONOMICS HO CHI MINH CITY
NGUYEN THE KHANG
THE IMPACT OF INVESTMENTS
ON ECONOMIC GROWTH
AND INCOME CONVERGENCE IN VIET NAM
Major: Finance and Banking
Code: 62.31.12.01
SUMMARY OF PHD THESIS
Academic Advisors:
Assoc. Prof. Dr.Nguyen Ngoc Hung
HO CHI MINH CITY, 2016
Dissertation was finished at: University Of Economics Ho Chi Minh City
Academic Advisors: Assoc. Prof. Dr.Nguyen Ngoc Hung
Reveiwer 1 :…………………………………………………………………………….
Reveiwer 2 : ……………………………………………………………………………
Reveiwer 3 : ……………………………………………………………………………
The thesis will be defend in the council at:
…………………………………………………………………………………………..
At……….date …….month……..year………
Research at: Library of University Of Economics Ho Chi Minh City
[1]
SUMMARY
Thesis conduct quantitative analysis of the impact of investment on economic growth
by model of PMG (Pooled Mean Group), to consider the impact of the factors on economic
growth in the short and long term in Viet Nam. Study data includes 63 provinces from 2000
to 2014. Therein, the investments are classified into three categories: public investment,
domestic private investment and foreign direct investment. The results showed that: In short
term, the labor and trade openness negatively impacts on economic growth, other factors
have not statistically significant coefficients. In the long term, public investment negatively
impact on economic growth, factors such as domestic private investment, foreign direct
investment, labor and trade openness have a positive impact on economic growth .
The thesis also examines the issue of convergence of per capita income among the
provinces in Vietnam. The results showed that the phenomenon of convergence in per capita
income among the provinces in Vietnam. All categories of investments impact positively on
the speed of convergence, in which foreign direct investment is the strongest, followed by
public investment and domestic private investment.
[2]
CHAPTER 1
INTRODUCTION
1.1. Research proposal
The impact of investments on economic growth has been studied in worldwide with
lots of space, time and many different research methods. Therefore, there are several
contradictory statements about the impact of investment on economic growth, such as:
Aschauer (1989a, 1989b); Hadjimichael and Ghura (1995); Jwan and James (2014);
Blomström and Persson (1983); Aviral Kumar Tiwari and Mihai Mutascu (2011). In
addition, one important predictor of the growth model of the neoclassical Solow (1956) and
Cass (1965), it is the poorer countries or regions tend to faster economic growth than richer
countries or areas. However, there is very little research in Viet Nam to assess the
contribution of each type of specific investments to economic growth and the process of
convergence of per capita income in the economy. For this reason, the author select and
implement the topic: The impact of investments on economic growth and income
convergence in Vietnam, as his doctoral thesis.
1.2. Research objectives
The primary objective of the thesis is to evaluate the impact of investments on
economic growth and income convergence in Vietnam. To achieve this goal, the thesis
focuses find answers to the following research questions: (1) How is the degree of the
impact of investments on economic growth in Vietnam in the short term and long term? (2).
How is impact of investments on income convergence process in Vietnam?.
1.3. Object and scope of the study
1.3.1. Research subjects:
The mechanism of impact of public investment (si); domestic private investment (di)
and foreign direct investment (fdi) on economic growth (GDP) and income convergence in
Vietnam.
[3]
1.3.2. Research scope:
Research focused on economic growth and the key elements such as public
investment, private investment from domestic and foreign direct investment affecting on
economic growth and income convergence process on the overall scope of Vietnam includes
63 provinces from 2000 to 2014. In addition, model uses control variables belonging to
economic growth theory and former empirical studies.
1.4. Research Methods
Thesis uses quantitative research methods on the basis of Cobb-Douglas expanded
production function including the variables affecting economic growth according to
research by Wei (2008), Nguyen Minh Tien (2014). Since that would assess the impact of
investments on economic growth and income convergence process.
1.5. Thesis’ contributions
1.5.1. Academic contributions
Firstly, the thesis will complement the empirical evidence of the impact of public
investment, domestic private investment and foreign direct investment on economic growth.
Secondly, the thesis contributes more empirical evidence for cases identified in Vietnam by
the neoclassical growth theories of Solow (1956); Cass (1965), it is the poorer countries or
regions tend to faster economic growth than richer countries or areas.
1.5.2. Empirical contributions
It is an significant practical evidence to the policy makers in selecting resources for
economic growth, especially in the allocation of investment in the total investment structure
of the economy. It is the basis for the balance of resources towards ensuring sustainable
economic growth and poverty reduction in society. The study also gives some policy
suggestions and propose a number of specific recommendations to the Goverment in the
implementation of policies to attract investment and use for economic efficiency.
1.6. The structure of the thesis
The thesis consists of 145 pages, is structured into 05 chapters. Chapter 1
Introduction. Chapter 2 Overview of theories and related studies. Chapter 3 Research
Methodology. Chapter 4 Research Results. Chapter 5 Conclusions and Recommendations.
[4]
CHAPTER 2
OVERVIEW OF THEORIES AND RELATED RESEARCH
2.1. The concepts
2.1.1. Investments
According to Sachs and Larrain (1993), general definition of investments as follows:
"Investment is the accumulated output to increase production capacity in the later period of
the economy".
2.1.2. Economic growth
Economic growth was fairly uniformly understood as an increase in actual output of
an economy in a given time period. Common measure is the increase in total gross domestic
product (GDP) in a year or an increase of the per capita GDP in a year. Some countries use
other indicators to determine economic growth: GNP (gross national product); GNI (gross
national income); NNP (net national product) or NNI (net national income).
2.1.3. Income Convergence
Convergence of income (also sometimes called the effect "catch up") is the
hypothesis that the economists as Solow (1956) and Cass (1965) said that per capita income
of poorer countries or provinces will tend to grow faster than the richer countries or
provinces. As a result, all economies converge to same level of per capita income in long
term. The developing countries have the potential to grow at a faster rate than in developed
countries because the characteristics of declining marginal return on capital in the model the
neoclassical growth. Moreover, poor countries can copy the methods of production,
technology, and organizational activities of the developing country for a chance to "catch
up".
However, not all poor countries can achieve a high growth rate, if income is too low,
people will have consume everything they do and thus do not have savings to invest in order
to maintain the level of investment per employee while the population rise and fell into a
trap of poverty. At the same time, countries or richer areas, with conditions for development
of science and technology, from which the marginal return on capital will rise stronger and
[5]
faster the countries or poor areas. This resulted in income divergence across countries or
regions.
2.2. Investment theories
2.2.1. Investment multiplier theory
Theoretical models of Keynes's investment multiplier is stated in the work “The
General Theory of Employment, Interest and Money” in 1936. According to him, to
increase the national income (national output), it must first increase investment. Here, he
has studied the relationship between increasing investment and increasing national
production and introduced the concept of "investment multiplier." Investment multiplier (k)
represents the relationship between the increase in investment and income increased. It tells
us that when there is an additional amount of aggregate investment, the income will increase
by an amount equal to k times of the increase in investment.
2.2.2. Investment Accelerator theory
If the investment multiplier explains the relationship between the increase in
investment to increase production or to increase investments how that affect to production.
Thus, investment appear to be a factor of aggregate demand. According to Keynes (1936),
investments are also considered in view of the total supply, which means that every change
of the output making how the investment alter. It means that the implementation of
investment projects will increase a certain level of output and the output increases, which
increases the volume of capital and the promotion of increased investment.
2.2.3. Harrod – Domar investment theory
If call Y is total output, K is the scale of production capital. Output has relationships
with production capital: k = K / Y (k: ratio of capital - output). Currently, developing
countries still popularly apply this model of growth in planning and mobilizing investment
capital for current growth. Because these countries mainly base on investment in width to
exploit the resources that are not being fully used. Harrod - Dorma has pointed out the role
of capital and capital efficiency in economic growth.
2.2.4. Solow neoclassical theory of investment
According to this theory, the investment is equal to saving (at potential output).
Saving S = s * Y where 0 0 is the opposite. The value of λ is the speed of income
convergence (or divergence).
3.3. Research data
The data used is based on a survey of 63 provinces in terms of time from 2000 to
2014. The GDP data is real per capita GDP of each province (million/person), this value is
taken on the basis of GDP current exchange rates for the CPI to eliminate inflation. To
eliminate inflation of the variables in the research model, for the value of public investment,
domestic private investment, foreign direct investment, regular expenditures, trade
openness, thesis will be calculated by the ratio (%) current values of these variables on the
value of GDP at current prices. Labor variable is calculated on the basis of the percentage of
workers with the total population.
3.4. Estimation method
3.4.1. Inspecting the stationary of panel data (Panel unit root test)
The thesis use five different types of experimental tests. There are Levin, Lin and
Chu (2002) also referred to as LLC type; Breitung (2000); Im, Pesaran and Shin (2003),
also known as IPS; ADF-Fisher; Philips Perron (PP). LLC and Breitung test unit test
[12]
common assumptions for all provinces, ie ρi= ρ. Im also, Pesaran and Shin (2003), also
known as IPS; ADF-Fisher; Philips Perron (PP) are presented by Maddala and Wu (1999)
allowing the unit root tests for different province.
In PMG regression methods, to assess the impact of these variables, then the first
thing to do is unit root tests of variables, variables theories would not be in the same level
stationary I(0) or I (1), there are no variables getting stationary at I(2).
3.4.2. Model of experimental study
𝑛
𝑚
𝛥𝑔𝑑𝑝𝑖𝑡 =∝ + ∑ 𝛽𝑖0 𝑔𝑑𝑝𝑖𝑡−𝑘 + ∑ 𝛽𝑖1 𝑋𝑖𝑡−𝑗 + 𝛽2 𝑒𝑥𝑝𝑜𝑖𝑡 + 𝛽3 𝑠𝑒𝑖𝑡 + 𝜑𝑖 [𝑔𝑑𝑝𝑖𝑡−1 − {𝛾0𝑖 + 𝛾0𝑖 𝑋𝑖𝑡−1 }] + 𝑒𝑖𝑡
𝑘=1
𝑗=0
Where: 𝛥𝑔𝑑𝑝𝑖𝑡 is the dependent variable of the model
𝑔𝑑𝑝𝑖𝑡−𝑘 is k lagged variable of the dependent variable levels of GDP
𝑋𝑖𝑡−𝑗 is j lagged variable of the model including public investment (si), private
domestic investment (di), foreign direct investment (fdi) and labor (lb)
𝜑𝑖 component is adjusted to equilibrium balance
For convergence assessment model
Estimating equation 2 convergence models, thesis follow data format for crossVietnam's 63 provinces and cities as the way of Wei (2008). The aim is to find evidence of
convergence or dispersion of income among provinces in Vietnam in the research phase.
The estimate will follow step by step sequence, which takes each turn one of each type of
investment to the right of the model to assess whether certain types of investments have a
strong impact on the process of convergence of per capita income, since there are no
reviews to evaluate public investment efficiency in terms of both economic growth and
income convergence on the basis combined with the results of model 1. in order to ensure
the reliability of the findings, thesis conduct testing variance changes (heteroskedasticity
Test: Breusch-Pagan-Godfrey), test multicollinearity (variance Inflation Factor) to avoid
tampering regression results.
[13]
CHAPTER 4
RESEARCH RESULTS
4.1. The situation of economic growth and investment in Vietnam
4.1.1. The overall view of economic growth in Vietnam.
Opening, integrating, diversifying the types of investment for the economic growth
are strategies which Vietnam has been pursuing during this period. With programs, policies
and social - economic strategies, Vietnam has made great achievements in economic
modification, social changes and improve people's lives towards a modern economy,
efficiency, a developing society, justice and civilization. This result is reflected in the social
economic indicators, legal systems, communities cultural in fast progress.
4.1.2. Investment situation in Vietnam recently.
In recent years, our country's economy operates in the model of growth based mainly
on elements of capital. In the ten years 1991- 2000 total investment of 802.4 trillion,
accounting for 36.5% of GDP, but ten years from 2001 to 2010, the total investment has
amounted to 4336.6 trillion, accounting for 41 , 6% of GDP. However, the period decreased
to 34.6% in 2011-2014. In particular, the investment rate of the state sector fluctuated
around 37-38%, non-state sector in the region of 35% and direct investment abroad at
around 26% while cumulative internal rate of under 30%. During this period, the growth
rate of gross domestic product (GO) at around 11-13% and a growth rate of value added
(VA) ranged from 6-8%.
4.1.3. The relationship between investment and economic growth
Investment capital for the economy increased during the period 2000-2014, although
the crisis in this period also gave rise unstable. During this period, the average annual
growth rate of investment is higher than the GDP growth rate of 1.8 times. In addition, the
average growth rate of investment tends to increase over the years. Investment growth rate
on an annual average in 2001-2005 was 18.3% and 21.2% from 2006 to 2014 period. The
Developments of investment in economic sectors are very different.
[14]
4.1.4. Income convergence in Vietnam
Convergence sigma (σ)
Table 4.1. Vietnam Index CV
Year
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
CV
0.35
0.34
0.32
0.38
0.40
0.38
0.39
0.32
0.32
0.23
0.22
0.26
0.33
0.30
0.28
Overall, the index of Vietnam CV in recent years has seen decreasing per capita
income distance. Demonstrates the Government's efforts in reducing poverty and income
distribution policy, distribution of investment in the economy toward reducing the wealth
gap between regions is gradually effective. However, this trend did not bring stability,
suggesting the economy is subject to many factors that impact the objective, internal forces
are not strong economy, easy to "hurt" when there are external factors impact. At the same
time, Vietnam is a developing country, the Government has been implementing the policy
of key areas, key provinces in economic growth should generate instability in the coefficient
of variation of earnings is easy to understand.
4.2. Research results
4.2.1. The impact of investment on economic growth
4.2.1.1. Functional form of variables in model
Using Eviews 9.0 software to determines the distribution function of the variables.
From the distribution format, select the format of approximately normal distribution
function as a basis for selecting a function of the variable types. All variables expressed as
logarithms has distributed approximately normal distribution. Except variable "id" - the
domestic private investment and the variable "lb" - labor, which is already approximately
normal distribution format before moving into a logarithm.
4.2.1.2. Statistics describing the variables
Datasets used in the thesis is a full balance panel data with observers (provinces) for
15 years from 2000 to 2014.
[15]
Table 4.2. Statistics describing the variables
LNGDP
LNSI
DI
LNFDI
LNSE
LNOPEN
LB
Mean
2.337.462
2.837.439
2.192.208
-0.97057
2.503.980
3.534.071
5.357.029
Median
2.259.333
2.801.367
1.990.799
0.350657
2.473.244
3.545.646
5.368.571
Maximum
5.930.513
5.426.505
7.756.557
5.071.668
4.294.671
7.191.257
6.855.686
Minimum
0.556106
1.071.941
0.731309
-9.210340
-0.06656
-2.700949
3.578.148
Std. Dev.
0.908634
0.709386
1.095.112
3.851.602
0.634285
1.278.931
5.736.881
945
945
945
945
945
945
945
Observations
Source: Author calculations based on data from the General Statistics Office, with the assistance of Eviews 9.0 software.
The value of variables can differ quite large, which suggests resources for economic
growth is unevenly distributed between the provinces, especially in investment, which we
see resulting from the large difference in price of per capita income among regions.
4.2.1.3. The correlation between the variables
Table 4.3. Correlation coefficient of variation
LNGDP
LNSI
DI
LNFDI
LNSE
LNOPEN
LB
LNGDP
1.000000
-0.385956
0.004269
0.368899
-0.438036
0.478982
0.438945
LNSI
-0.385956
1.000000
0.150353
-0.142201
0.603052
-0.382823
-0.180485
DI
0.004269
0.150353
1.000000
0.071882
0.230296
-0.037018
0.209125
LNFDI
0.368899
-0.142201
0.071882
1.000000
-0.324508
0.491238
0.180378
LNSE
-0.438036
0.603052
0.230296
-0.324508
1.000000
-0.609974
0.174317
LNOPEN
0.478982
-0.382823
-0.037018
0.491238
-0.609974
1.000000
0.022387
LB
0.438945
-0.180485
0.209125
0.180378
0.174317
0.022387
1.000000
Source: Author calculations based on data from the General Statistics Office, with the assistance of Eviews 9.0 software.
4.2.1.4. Inspection of stationary (Panel unit root test)
Table 4.4. Results unit root tests in panel data
At Level
lngdp
lnsi
di
lnfdi
lnopen
lnse
lb
LLC
Breitung
IPS
ADF - Fisher
PP - Fisher
-10.1806
6.49571
-3.50669
185.627
286.799
(0.0000)
(1.0000)
( 0.0002)
(0.0004)
(0.0000)
-7.35587
0.79406
-3.19555
193.341
178.950
(0.0000)
( 0.7864)
( 0.0007)
(0.0001)
( 0.0014)
-6.06077
-0.54672
-1.81760
150.756
151.562
(0.0000)
(0.2923)
( 0.0346)
( 0.0656)
(0.0601)
-10.9674
-2.66343
-7.03636
261.496
233.000
(0.0000)
(0.0039)
(0.0000)
(0.0000)
(0.0000)
-6.48460
4.20426
-1.42456
165.206
192.921
(0.0000)
(1.0000)
( 0.0771)
(0.0109)
(0.0001)
-10.8075
-1.64987
-5.51927
225.542
226.006
(0.0000)
(0.0495)
(0.0000)
(0.0000)
(0.0000)
-5.96849
(0.0000)
1.18946
(0.8829)
-2.82891
(0.0023)
169.283
142.558
(0.0061)
(0.1487)
[16]
First difference
lngdp
lnsi
di
lnfdi
lnopen
lnse
lb
LLC
Breitung
IPS
ADF - Fisher
-15.7871
-4.18376
-8.19872
279.612
356.352
(0.0000)
(0.0000)
(0.0000)
(0.0000)
(0.0000)
-20.4342
-8.85122
-12.1502
364.112
502.754
(0.0000)
(0.0000)
(0.0000)
(0.0000)
(0.0000)
-22.2330
-13.9983
-14.5940
417.636
635.666
(0.0000)
(0.0000)
(0.0000)
(0.0000)
(0.0000)
-30.8449
-11.2584
-20.7588
542.147
650.407
(0.0000)
(0.0000)
(0.0000)
(0.0000)
(0.0000)
-23.5486
-7.85594
-15.2950
438.482
654.639
(0.0000)
(0.0000)
(0.0000)
(0.0000)
(0.0000)
-23.5341
-6.62774
-16.7263
473.959
751.826
(0.0000)
(0.0000)
(0.0000)
(0.0000)
(0.0000)
-19.3615
-9.73504
-12.5852
(0.0000)
(0.0000)
(0.0000)
369.679
518.803
(0.0000)
(0.0000)
Notes: Data are taken from the General Statistics Office, processed by Eviews 9.0. The value in () is the level of significance.
With a significance level of 5%, the variable "id", domestic private investment has no
stationary at level but at first differences I(1), the remaining variables have stationary at
level I (0). But it's important that all variables have stationary at first differences. That
means that the panel data is not the same level integration I(1) or I(0). According Pesaran et
al (1996), Hamuda et al (2013), the variables in the model are not the same stationary I(1) or
I(0), so applying the PMG is appropriate procedures for this study.
4.2.1.5. Results for the impact of various factors on economic growth
Table 4.5. Results for PMG, The dependent variable-GDP
Short-term effects
Variables
Coefficient
Public investment
Standard error
P-Value
-0.010755
0.015642
0.4921
Domestic private investment
0.001039
0.002237
0.6427
Foreign direct investment
0.002808
0.009587
0.7698
Trade openness
-0.079888
0.019827
0.0001
Labor
-0.012468
0.004893
0.0112
Current expenditure
-0.056258
0.038381
0.1436
The adjustment coefficient on the long-term balance 𝜑 𝑖
-0.434205
0.038679
0.0000
Long-term effects
Variables
Public investment
Coefficient
Standard error
P-Value
-0.017490
0.005841
0.0029
Domestic private investment
0.002412
0.000462
0.0000
Foreign direct investment
0.009672
0.000575
0.0000
Trade openness
0.019319
0.006265
0.0022
Labor
0.030277
0.000787
0.0000
Source: Author's calculations using Eviews 9.0 based on data from the General Statistics Office
[17]
With PMG methodology, processed by Eviews 9.0, we see the adjustment coefficient
on the long-term balance 𝜑𝑖 = -0.434205 statistically significant p-value = 0.0000 (<5%),
which means variables such as public investment, private domestic investment, foreign
direct investment, trade openness and labor tend to impact on economic growth in the long
term, we conclude as follows:
In the short term, factors such as trade openness and labor may negatively impact on
economic growth with 5% statistically significant. Other variables is not statistically
significant.
In the long term, factors such as public investment, private domestic investment,
foreign direct investment, trade openness and labor has an impact on economic growth with
statistical significance (p-value <5%). We see, in the long term, public investment impact
negatively on economic growth, while the variable domestic private investment, foreign
direct investment, trade openness and labor impact on economic growth positively. In which
the largest contribution is labor then the openness of trade, foreign direct investment and
domestic private investment.
4.2.2. Results of estimating convergence
Absolute β convergence
Based on the formula 2 models with gdpi0 in right to test absolute convergence, the
results table (4.6) as follows:
Table 4.6. Results of absolute convergence
Coefficient
Standard error
P-Value
Constant (𝛼)
2.502412
0.093230
0.0000
𝐿𝑛𝑔𝑑𝑝𝑖0 (𝛽)
-0.186983
0.066080
0.0063
Convergence rate ( 𝜆)
0.0138
Source: Data from the General Department of Statistics and calculations by author
The estimated coefficient β is negative and statistically significant, it means that there
is evidence of absolute convergence in per capita income, which means that in the period of
2000 to 2014 in Vietnam areas with low income in the period initial segment tends to
increase faster than the higher initial income, the rate of convergence is 1.38%.
The initial poor areas can enjoy more preferential policies of the government to have faster
growth rates. The results of this study contrast to Pham The Anh studies (2009), Hoang
[18]
Thuy Yen (2015). However the results were consistent with growth theory of Solow (1956)
is presented in detail in Chapter 2.
Conditional convergence 𝜷
The objective of thesis consider the type investments how to impact the income
convergence process in Vietnam. Thesis tested step by step way that Wei (2008) conducted
research in China. First will give each the right to invest in the pattern 2, and will put each
pair to invest in models and finally put a time on three types of investments. Aim to identify
the best value β from which to comment on the contribution of investment income
convergence process in Vietnam. The results are shown in Table 4.7.
Table 4.7. Results of estimating conditional convergence
Coefficient
Standard error
P-Value
Models with public investment
𝐿𝑛𝑔𝑑𝑝𝑖0 (𝛽)
-0.224480
0.077052
0.0050
Lnsi
-0.064081
0.067571
0.3468
-0.166204
0.070254
0.0212
di
0.003826
0.004331
0.3806
Convergence rate ( 𝜆)
0.012118
-0.275469
0.067059
0.0001
Lnfdi
0.038402
0.011730
0.0018
Convergence rate ( 𝜆)
0.021482
Convergence rate ( 𝜆)
0.016948
Models with domestic private investment
𝐿𝑛𝑔𝑑𝑝𝑖0 (𝛽)
Models with foreign direct investment
𝐿𝑛𝑔𝑑𝑝𝑖0 (𝛽)
Models with private and public investment
stment
𝐿𝑛𝑔𝑑𝑝𝑖0 (𝛽)
-0.205050
0.079604
0.0125
Lnsi
-0.070310
0.067896
0.3046
di
0.004249
0.004348
0.3324
Convergence rate ( 𝜆
0.015298
Models with public investment and FDI
𝐿𝑛𝑔𝑑𝑝𝑖0 (𝛽)
-0.304253
0.075891
0.0002
Lnsi
-0.051547
0.063000
0.4165
Lnfdi
0.037804
0.011785
0.0022
Convergence rate ( 𝜆)
0.024185
-0.267241
0.073178
0.0006
Lnfdi
0.037685
0.012071
0.0028
di
0.001211
0.004132
0.7706
Convergence rate ( 𝜆)
0.020729
Models with private investment and FDI
𝐿𝑛𝑔𝑑𝑝𝑖0 (𝛽)
- Xem thêm -
.PDF 
27 
264 
91 
