Public investments in transport and economic growth the case of viet nam in the period 1996 2006

Of the three main components of publicinvestment, including transport, education and agriculture investments, empiricalevidences indicate that only public investment in transport has con

Nguyen Thi Que Huong

Supervisor: Dr Le Quoc Hoi - Journal of Economics and Development - National Economics University, Vietnam

October, 2011

First of all, I am grateful to my main supervisor, Dr Le Quoc Hoi who instilled on

me the importance of conducting meaningful research and thinking critically andthoroughly about the topic, economic theory and methods He has been extremely helpful

at all stages of my research, helping me to develop the research, enlightening me on manyaspects of doing good research, and providing invaluable comments and feedback Withouthis guidance and support this thesis would not have been completed Also, I would like tothank the assistance and guide by Dr Do Ngoc Huynh who enthusiastically supports meand spends hours with me discussing model techniques, and providing excellent commentsand suggestions I would like to thank Prof Vu Thieu for his professional guidance on theeconometric techniques over the past four years of my studies to date I can applied in thisthesis Along with my supervisors, I wish to sincerely thank my colleagues at Ministry ofFinance for their help in collecting data set My final gratitude is devoted my friends whoalways give me invaluable encouragement, support and tolerance during the thesis writing

This thesis analyzes the impact of public investment in transport on Vietnameseeconomic growth over the period 1996 - 2006 Relevant econometric models based onpanel data from 64 Vietnamese provinces are used to estimate the impact Empirical resultsobtained show that a change in public investment in transport has a positive and significantimpact on the country’s economic growth Of the three main components of publicinvestment, including transport, education and agriculture investments, empiricalevidences indicate that only public investment in transport has contributed to growth.Besides, a positive spillover effect of investment in transport on economic growth is alsoobserved from the empirical results

2.1.1 Exogenous growth theories 12

2.1.1.1 Harrod - Domar Model 12

2.1.1.2 Solow - Swan Model 13

2.1.3 Theoretical models on public expenditures 15

2.1.3.1 Barro Model 15

2.1.3.2 Devarajan, Swaroop and Zou Model 18

2.2 Empirical review 19

2.2.1 International studies 19

CHAPTER 3:OVERVIEW OF GOVERNMENT EXPENDITURE AND PUBLIC INVESTMENT IN TRANSPORT IN VIETNAM25

3.1 Overview of Vietnamese government expenditure 25

3.2 Transport system and public investment in transport in Vietnam 28

3.2.1 Overview of transport infrastructure system 28

3.2.2 Overview of public investment in transport in Vietnam 31 3.2.2.1 State and local government investment in transport 31

3.2.2.2 Transport infrastructure invested by the Ministry of Transport 34 3.2.2.3 Vietnam’s transport expenditure in comparison with neighboring countries 35

3.3 Transport investment and economic – social development in Vietnam 36

3.3.1 Contribution of transport investment to economic and social

development 36

3.3.3 Contribution of transport investment to GDP in Vietnam and some

4.2 Empirical results and analysis 45

4.2.1 Public investments and economic growth - baseline regression results and analysis: 45

4.2.2 Public investments and economic growth - extension regression results and analysis 50

4.2.2.1 The role of education in the impact of public investment in transport on

LIST OF TABLES AND FIGURES

Table 1: Public expenditure as ratio of GDP (%) 25

Table 2: State government investment by sectors 26

Table 3: Actual state of national road system in Vietnam 28

Table 4: Actual state of national road system in Vietnam 29

Table 5: Public investment in transport in 2000-2006 32

Table 6: State and local government’s investment in transport 2000-2006 34

Table 7: Kinds of transport infrastructure invested by MOT (1996-2007) 35

Table 8: Transport Expenditure Levels as Percentage of GDP of East Asian Countries in period 1999-2004 36

Table 9: Contribution of the transport sector to GDP in selected countries (%) 39

Table 10: Descriptive statistics 44

Table 11: Statistics for government investment shares, 64 provinces, 1996-2006 45

Table 12: Baseline regression estimates 46

Table 13: The effect of education level on growth elasticity of public investment in transport 50

Table 14: Spillover effect of public investment in transport 52

Table 15: Effect of a change in component’s shares on economic growth 54

Table 16: Public investments and economic growth – baseline regression results with lagged variables 55

Table 17: The role of education in the impact of public investment in transport on economic growth - extension regression results with lagged variables 56

Table 18: The spillover effect of public investments in transport of neighboring provinces -extension regression results with lagged variables 57

Figure 1: Trend of public investment in transport infrastructure in constant price 32

Figure 2: Public investment in transport infrastructure per GDP 33

Figure 3: Public investment in transport versus GDP 1996-2006 38

LIST OF ABBREVIATIONS

FDI – Foreign Direct Investment

GDP – Gross Domestic Product

GSO – General Statistics Office

IMF – International Monetary Fund

OECD - Organization for Economic Co-operation and Development MOF – Ministry of Finance

MOT – Ministry of Transport

CHAPER 1 INTRODUCTION

1.1 Problem statement

How to raise the rate of economic growth and improve human living standards isalways a great question by governments and policy makers In an economic perspective,the past decades have seen efforts of economists to clarify the source of economic growthand the root of increasing disparities between rich and poor countries

Exogenous growth models initially promoted by Solow (1956) recognized capitalaccumulation and technical progress as the determinants of economic growth According totheir studies, policy variables have no effect on long term economic growth Policyvariables such as investments in public infrastructure have only transitional growth effects

Contrast to the idea of the exogenous growth theory, a new stream of endogenousgrowth theory which is prompted by Lucas (1988) and Romer (1990) argued that the rate

of productivity growth is directly depended upon the level of some policy variables such asthe level of education of workforce, the amount of transport investment or level of publicinvestment This framework of endogenous growth theory has later stimulated the body forempirical analyses which attempt to examine the link between public investment andaggregate economic activities

Examining time series data from the United State in the post World War II,Aschauer (1989a) found a large and significantly positive elasticity of output with respect

to public capital stock Firming for his finding, Aschauer (1989b) argued that in area such

as public utilities, road and education, public expenditure may complement higher privatemarginal rate of return This finding is later supported by early study of Munnell (1990a)

By exploring the case of the US she indicated that at national level 1 percent increase instock of public capital would increase output by 0.34 percent

Also focusing on the relationship between public investment and economic growthbut based on panel data at state level, some other authors provided the result that publiccapital stock had a significant positive impact on output (Munnell, 1990b; Kelejian andRobinson, 1994) However, they show that the estimated magnitude of public capital stockelasticity generally gets smaller when moving from national to state data because ofinfrastructure spillover effects The explanation articulated the finding of Munnell (1992)when she argued the benefit of infrastructure spillover across state borders, so that state-

level studies do not measure the full effect of public investment This low correlation isalso recognized by Nourzad and Vrieze (1995) Based on econometric method of Aschauerand Munnell and taking into account random effect, their study on a panel data of OECDcountries over the period of 1963-1988 has shown a relatively low but significant outputelasticity of 0.05 with respect to public investment.

Recognizing the impact of public capital stock on economic productivity, othereconomists tend to focus on the relationship between economic growth and publicinvestment in one of core infrastructures - the provision of transport infrastructure such ashighway, road, port or airport - at national, regional and state levels Based on crosscountry data, a study of Antle in 1983 implies that transport and communication weakness

is the most constrain in less developing countries (LDCs) to raise their production Later in

1993, study of Easterly and Rebelo indicates that growth and public investment in transporthas a strong and robust association At the same time with study of Easterly and Rebelo,when exploring the case of US, Canning and Fay (1993) also found a large estimated rate

of return for transport and telephone systems at about 40 percent Based on this findingthey come to conclude that developed countries with higher income will have normal rate

of return for transport infrastructure while newly industrializing countries like South Koreaand Chile have a very high rate of this return This finding has been supported for arisingconcerns on identifying effects on economic growth of transport investment factor in thecase of industrializing and developing countries

Known as a developing country with impressive economic success of the grossdomestic product (GDP) growth rate at 7.5% over 2000-2006, identifying the determiningfactors of rising economic growth rate while maintaining macroeconomic stability is thuscrucial to the government and policy makers in Vietnam In process of finding a reasonableway to improve the economy, public investment is always considered as an engine ofeconomic improvement by the Vietnamese government Budget for transport infrastructureinvestment accounts about 45-50% of the country’s total non-military spending each year.Understanding the effects of public investments in transport on economic growth thus isessential to policy makers in the face of increasing globalization Upon the demand,economic studies on the subject have been widely discussed in recent years

Empirical study of Kokko and Tingvall (2005) has provided some policies that theVietnamese government would use to promote provincial growth and development, ofwhich more investments in transport infrastructure take an important part Exploring otherside of the subject, Lien (2004) examines reality of investments in transport system in

Vietnam and concluded that although Vietnam’s transport infrastructure system has beenimproved, it has not met demands of the country’s social and economic development.Other researches describe the impact of transport investment on economic developmentprocess in Vietnam and come to conclusion of the need for more investment in the field(Son, 2008 and Tin, 2008)

Although identifying impacts of investment in transport infrastructure is animportant issue for Vietnamese policy makers, existing researches on the subject are onlydescriptive It is easily observed that there have not been empirical studies usingeconometric method to examine the correlation in Vietnam These reasons require a studybased on a quantitative analysis to examine the contribution of investments in transport oneconomic growth and thus give a comparison between level of transport investment’scontribution with that of other public investment

In this study, based on frameworks of Aschauer (1989), Munnell (1990), Barro(1991) I focus on investigating the impact of public investment in transport on economicgrowth in Vietnam and making statistical comparison between the contribution of publicinvestment in transport and the contribution of other components of public spending Thestudy also gives policy implications which would be crucial for long-run development inVietnam

1.2 Objective of the thesis

 Review public investments in transport in Vietnam in recent years

 Investigate whether public investments in transport have a significant impact

on economic growth of Vietnam in the period 1996-2006

 Compare the contribution of public investments in transport with that of otherpublic investments to Vietnamese economic growth in the period 1996-2006

 Find out some factors which would affect level of the contribution of publicinvestments in transport to economic growth

 Recommend appropriate policies to raise the effectiveness of publicinvestments productivities in transport and government investments in general as well

1.3 Research questions

The thesis attempts to answer the following question:

“Do public investments in transport have a significant impact on economic growth

of Vietnam in the period 1996-2006?”

In order to answer this main research question, it is necessary to answer thefollowing sub-questions:

1 How is transport invested in Vietnam during the period 1996-2006?

2 What are the contributions of public investments in transport to economicgrowth during the period 1996-2006?

3 What are differences between the contributions of public investment intransport and that of other public investment to economic growth in Vietnam?

4 What are factors affect level of the contribution of public investment intransport to economic growth?

1.4 Methodology and Data

Data

The thesis employs the provincial data obtained from General Statistic Office ofVietnam (GSO) The data on public investments in transport infrastructure is collectedfrom the annual report on public expenditure by Ministry of Finance and statistics reported

by Ministry of Transport In the thesis, data on public investments in transportinfrastructure will be used to represent public investment in transport area because thecomponent takes a dominant part of total government investment in transport area

1.5 Expected outcome

The proposed topic is clear important for more understanding of the impacts ofpublic investments in transport on economic growth in Vietnam Findings from thisresearch are expected to add a new view to the development of empirical studies onrelationship between public investments in transport and economic growth in Vietnam Theresearch also provides appropriate recommendations to raise the effectiveness of publicinvestments in transport

1.6 Thesis structure

The thesis is organized as follow:

Chapter 1: Introduction

Chapter 2: Literature Review

Chapter 3: Overview of government expenditure and public investment in transport

in Vietnam

Chapter 4: Empirical analysis and discussion

Chapter 5: Conclusions and policy implications

CHAPTER 2 LITERATURE REVIEW

This chapter surveys the most theoretical and empirical studies on the relationbetween economic growth and public investment or public investment in transport inparticular Its purpose is to review theoretical models and technical methods which havebeen used to highlight the link The main conclusions come from literature review is thatdespite controversies remaining on level of the link, most of the studies show a significantcontribution of investment in transport to higher economic growth

2.1 Theoretical review

2.1.1 Exogenous growth theories

2.1.1.1 Harrod - Domar Model

Harrod (1939) and Domar (1946) attempted to investigate Keynesian analysis with elements of economic growth Their model regarded the role of aggregate demand as the engine

of growth and uses the incremental capital-output ratio (ICOR) to assess productivity or additional capital on the increase of output In Harrod - Domar Model, production function is form as follow:

Y = AK (1.1)Where: Y is total output, A is level of technology or productivity (1/A is sometimescalled incremental capital output ratio or ICOR in context of this model), and K is totalcapital The model yields a concept of capital accumulation K which means the change

over a short period of K.

K I

K    (1.2)

Where I is investment and is the depreciation rate (e.g  is the depreciation rate (e.g   is the depreciation rate (e.g =0.05 means that 5 percent

of capital stock depreciates in a year).

invested over that period, less the amount of depreciation that occurs From that, growthrate over that same short period of time will be displayed:

In equation (1.5) the main idea prompted by Harrod - Domar model is thateconomic growth is decided by saving rate and capital-output ratio The more the economy

is able to save and invest, the higher economic growth is

2.1.1.2 Solow - Swan Model

Extending the model of Harrod and Domar, Solow - Swan growth model focuses

on describing how the economy change and grow over time when saving, investment andlabor force grow, and there is a progress in advancing technology They take the form ofthe aggregate production function as follow:

(1.6)where Y(t), K(t), A(t) and L(t) are output, capital, knowledge and labor attime t respectively In this model, the level of knowledge or technology A(t) and laborforce grow, and there is a progress in advancing technology

The model develops the concept of efficiency of labor k=K/AL, under theassumption of competitive markets and there are no externalities and constant returns toscale in K and L:

Labor and technology grow at constant rates:

(1.7)National income identify for a close economy:

(1.8)(1.9) The law of motion of K:

(1.10)Using the chain rule Solow concludes that:

ksf(k)  ( ng)k



 (1.11)

From the analysis, they prove that the rate of change of the capital stock per unit ofeffective labor is the difference between actual investment per unit of effective labor and

K I K

K Y

(1.5)

)) ( ) ( ), ( ( ) (t F K t A t L t

gt nt

e A t A t

A g t A

e L t L t

L n t L

) 0 ( ) ( )

( )

(

) 0 ( ) ( )

( )

sY Y s S

C Y

K sY



break-even investment (the amount of investment that must be done just to keep k at itsexisting level) By examining the change in state of the economy to break-even point,Solow - Swan model states a conclusion that economic growth, defined as long run growth

in per capita income is independent of policy variables such as the savings rate or the level

or public sector investment An economy that has reached its optimum capital-labor ratiowill have a growth rate in GDP which equals the growth of labor force Real incomes canincrease only if there is an increase in labor productivity Policy variables can affecttransitional growth rates, but not long run growth rates The policy result is thatgovernment can have only short-term impact on real growth and should focus on policiesthat increase labor productivity

From his analysis, Solow (1956) argues that there is an only transitional growtheffect of public investment Based on the assumption of diminishing returns to capitalaccumulation, the Solow model gives a highlight that only technological change is source

of permanent growth Upon this viewpoint, Solow model is known as an exogenousgrowth model

2.1.2 Endogenous growth theories

Later, in the mid 1980s a new stream of growth study has been driven which isknown as the endogenous growth model The endogenous growth model avoids theimplication of the exogenous growth model that diminishing returns to capitalaccumulation where only technological change is source of long-run growth in income percapita They attempt to explain how private economic agents make decisions that drivelong-run growth through increasing returns, technology spillovers and other non-traditionaleffects Economists who are known placing the theoretical background for the stream areLucas, Romer and Barro

Based upon the Solow model, Romer (1986) launched a model of increasingreturns In the model, there is a stable positive equilibrium growth rate resulted fromendogenous accumulation of knowledge This is an important break with the Solow model

Under the Romer model, sustainable economic growth can be at least partlyendogenously explained by "learning by investing" which represented by a function ofaggregate capital where the firm’s production function is formed as follows:

Ytj= F(ktj,Zt,ltj) (1.12)Where: Zt is aggregate labor-augmenting technological progress

In the function, Romer assumes that knowledge creation is a side product ofinvestment The aggregate stock of knowledge in the economy is proportional to the

cumulative sum of past aggregate investment Romer also makes the crucial assumptionthat Zt represents overall learning in the economy and affects productivity By substitutingSolow type consumption function in place of the utility maximizing behavior of firms incompetitive market where firm is price takers, Romer explores that the economy will reachhigher return to capital accumulation at the social level than at the individual level

Lucas (1988) also partly explains endogenous economic growth He developed the

“AK” model which includes human capital as an independent determinant of economicgrowth differing from ordinary labor In the AK model all factors of production can beendogenously accumulated Unlike the Solow model, there are no constraints placed onaggregate growth by an independently determined supply of either labor input or naturalresources For example, his theory takes the view that the proper interpretation of the

"labor input" is as a flow of services generated by a stock of “human capital” The humancapital stock is endogenously determined over time by investments in training oreducation From the argument, he yields the conclusion that an increase in the savings rate

in public capital will increase the steady-state growth rate, or growth rate of permanentincome per capita in the economy, with elasticity approximately equals to the elasticity ofpublic capital in the aggregate production function

In conclusion, although Lucas, Romer and some others who focus on theendogenous growth theory have different approaches on factor of long - run growth, most

of them recognizes that the rate of productivity growth is directly dependent upon the level

of some policy variable, such as the level of education of workforce, the amount of R&Dinvestment, the amount of infrastructure investment, the savings rate, the tax rate, and thelevel of public sector investment This point of view later becomes the main theoreticalbase for many studiers who recognize the significant contribution of governmentinvestments to economic improvement

2.1.3 Theoretical models on public expenditures

2.1.3.1 Barro Model

While Romer (1986) and Lucas (1988) mentioned above emphasize the potentialfor externalities spillovers coming from the stock of knowledge and/or labor force skillswhich is considered as a key for government to prioritize improvements on education,Barro (1990) focuses the extension of the role of government expenditures in services Heargues that this government’s spending on service will enhance productivity in the privatesectors and may increase the growth rate of the economy

In an attempt to link the role of government services, as inputs to privateproduction, to the growth performance of a country, Barro (1990) formed the aggregateproduction function for the economy as follow:

yt = φ(kt, gt) = kt.φ(gt/kt) (1.13) where kt is aggregate capital per labor and gt is assumed to be measured by the percapita quantity of government purchases This production function presents constant return

to scale in kt and gt together, but decreasing returns to kt itself Thus, the source of thenon-rival input now is the presentation of gt, public goods

According to Barro, in an economy with many similar households, each householdwill try to maximize their utility Households’ utility will be assumed to form as follow:

U = 



 0

) ( (c t e t

(1.14)C(t) stands for consumption per capital ρ(>0) refers to time preference coefficient.From that, the utility function of household is in form as follow:

(t 1

c , σ > 0 and ≠ 1 (1.15)

= Lnc in case of σ = 1Each household producer has an access to production function:

y =  (k) Where k is denoted as capital per worker and y is output per worker

In a competitive economy, households will reach the utility maximization in theequilibrium condition as:

) )(

1 ( ) (

t c

;  is marginal product of capital (1.16)According to the presumption, the results to scale in production function arechangeless, then:

y = Ak ( '

 = A, A > 0) and (1)( )

) (

t c

(1.17)

 is the increase rate of growth or consumption

Based on a Cobb- Douglas production function with constant return to scale, heassumes that government will pay its different spends (current and capital spends), ofwhich each kinds of government spending has its’ different effect on total turnover of aneconomy Then production function will be formed as follows:

y =  (k, g) = Ak1  g with 0< α <1 (1.18)

In a study in 1990, he assumes that government will adopt a balance - budgetfinance policy by imposing a flat - rate income tax (τ) That means τ is constraint, then the) That means τ) That means τ is constraint, then the is constraint, then thebudget constraint is:

g = τ) That means τ is constraint, then the y = Ak1  g (1.19)Where g is quantity of government purchases on goods and services

The production function in equation (1.19) implied that the marginal product ofcapital is now:

fk= A (1-α)(g/k)α (1.20)Substitute (1.19) into production function (1.18):

y = k.A1/(1-α).τ) That means τ is constraint, then theα/(1-α) (1.21)From that he comes to conclude a link between the change of τ) That means τ is constraint, then the and y That is, anincrease in τ) That means τ is constraint, then the will lead to an increase in the relative amount of public input Moreover, byextract a ratio of y/k he yields the equation of the marginal product of capital is thefollowing:

fk = (1-α) A1/(1-α)

τ) That means τ is constraint, then theα/(1-α) (1.22)Based on the equation (1.22), Barro gives a highlight on a upward shift in increase

in the marginal product of capital to an increase in the government expenditure ratio τ) That means τ is constraint, then the

Turn to household’s utility optimization Households tend to satisfy the equation(1.16), so that:

 =

) (

)

t c

t

c

= (1/).[(1-α).A1/(1-α).(1-τ) That means τ is constraint, then the).τ) That means τ is constraint, then theα/(1-α) - ] (1.23)Equation (1.23) emphasizes a relationship between growth rate and governmentexpenditure and tax rate As long as tax rate is constant, the government sets g and tax atthe same rate as y At that point, the growth rate  is constant then the economy will reachits steady state

Parameter (1-τ) That means τ is constraint, then the) reflects the negative effect of tax to growth However, an increase

in tax is corresponded to an increase in government expenditures in public goods andservices which make an increase in marginal product and output in private production.Parameter τ) That means τ is constraint, then theα/(1-α) reflects the positive effect of public expenditures on growth

From τ) That means τ is constraint, then the deviation of equation (1.23), we can obtain the optimal tax rate to growth is:

τ) That means τ is constraint, then the* = 1- α (1.24)From his analysis, Barro concluded that the government has a role to play via theprovision of its expenditure in goods and public services which is used as productive inputs

in the private sector producers’ production functions However, although recognizing the

positive impact of government spending on long – run growth, the problem raised here isthe way by government to harmonize its expenditures and tax rate, which decides thesource of government revenue.

2.1.3.2 Devarajan, Swaroop and Zou Model

Based on the background of Barro (1990), Devarajan, Swaroop and Zou (1996)focus on other aspect of public expenditures to growth They build a model which reflectsthe impacts of the different components of public expenditures on growth The modelattempts to determine what components of public expenditures are efficient and how is theimpact of their transferences on the change in economic growth

In their model, Devarajan, Swaroop and Zou (1996) assume that productionfunction output is determined by private capital and two kinds of government expenditureswhich have the different impact on output

у = ƒ(k, g1, g2) = [αk -ζ +βgg 1 -ζ +γgg 2 -ζ]-1/ζ (1.25)

Where: α ≥0, β≥0, γ≥0; α+β+γ = 1; ς ≥-1

Follow ground by Barro, they assume that the government finances its consumption

by imposing a fix tax rate This means that government pursues a balance budget policy

τ) That means τ is constraint, then they = g1 + g2 (1.26)

g1=τ) That means τ is constraint, then they and g2=(1-)τ) That means τ is constraint, then they (1.27)

where  is the density of component g1 in the total government expenditure From(1.25)-(1.27) they can obtain an equation that reflect the relation between output andcomponents of government expenditures as following:

of household is in form as follow:

(t 1

c

, σ > 0 and ≠ 1 (1.29)Where c(t) stands for consumption per capital and ρ(>0) refers to time preferencecoefficient

Households will maximize their utility as follow:

U = 



 0

) ( (c t e t

(1.30)

With budget restriction: k = (1-τ) That means τ is constraint, then the) y – c (1.31)

From that they can obtain an equation reflecting economic growth rate asfollowing:

)]

)1(/(

)[

1(

c

In which 1/σ = -u’/u’’(c)c is known as consumption substitution parameter

From (1.32) they derive the relation between growth rate and the share of

government expenditure From derivation of γg( j ) we can clarify that whether an increase

in density of gi may increase economic growth rate:

1 ( ) 1 (

))1((

])1([

])[

1)(

1(

2.2 Empirical review

2.2.1 International studies

In an attempt to clarify growth effect of government expenditures, there are lot ofmethods used, however studies which based on aggregate linear production function seem

to be one of the most popular methods

Based on this line of framework, an early research of Aschauer (1989a) examinedtime series data of the US from 1949 to 1958 and reported the evidence that public capitalstock has a very important contribution to economic growth in the United States Hisempirical results have shown a large elasticity of lower growth in private productivity withrespect to lack of public investment In his study, Aschauer also disaggregated components

of public capital stock and found that the most important contribution come from the coreinfrastructure investments which consist street, highway, airport, mass transit system,electricity and system of water and sewerage This idea is later supported by studies ofBarro (1991); Munnell (1992); Easterly and Rebelo (1993) with cross countries data.However, many suspicions about its reliability are raised Criticisms release that timeseries and cross-sectional studies are not enough reliable because of their falses on

controlling for unobserved heterogeneity and non-stationary data (Holtz-Eakin, 1994;Tatom, 1991; Kelejian and Robinson, 1994)

To avoid the limitations of time series and cross-section methods, later researchestend to use panel data regressions to ascertain the subject In a study of Aschauer (1989c),

he specified in a Cobb-Douglas function to clarify the economic contribution of publicinvestment stock of the G7 countries using panel data for the period 1966-1985 The resultexposed the output elasticity of public investment from 0.34 to 0.73 Although the resultshave shown smaller elasticity that are estimated from his time series data but it clearlysupports the productive contribution of public investment to productivity and growth

Following the line of argument, Nourzad and Vrieze (1995) adopt Aschauerframework but control for energy input price and taking into account random effect Theiranalysis used the panel data for 7 OECD countries over the period 1963-88 and shown thatthe significant elasticity of output with respect to public investment is about 0.05.Although the elasticity found is much smaller than the finding of Aschauer but Nourzadand Vrieze believe that government investments enhanced the economic growth of OECDcountries in the condition of changing in energy price

In other attempts to clarify effects of public investment stock on private production

at state level, Munnell and Cook (1990) estimated the effects from pooled data of 48 states

in the United State and found that elasticity of private output with respect to state and localpublic capital is 0.15 At the same time, Merriman (1990) also highlights a large elasticity

of public capital expenditure Based on panel data in 1954-1963, the study on 9 Japaneseregions presents an elasticity of 0.43-0.58 of public investment stock with respect ofeconomic growth This result is consistent with the evidence of Munnell (1993) who findthe significant and positive elasticity of public investment at 0.14 to 0.17 However, thisline of studies faces some disputes from some economists who show the remarkably weak

or no link between public capital stock and growth The strongest evidence at state levelcomes from Holtz - Eakin (1994); Holz - Eakin and Schwartz (1995); Garcia-Mila,McGuire and Porter (1996) They argue that unobserved state specific characteristics arenot controlled for the study of Munnell and some other studiers

Also investigating the effect of public expenditures on growth, but some otherexperts tend to focus on analyzing how different components of public expenditure impacteconomic growth According to them, studies of economists like Aschauer or Munnell onlyutilized aggregate measures of public expenditures in form of either growth in governmentconsumption or government consumption as ratio of GDP In some extent, this kind of

study does not analyze how different components of government expenditures affecteconomic growth, while it’s seem to be that different components of governmentexpenditures may have different relations with growth (Devarajan, Swaroop, and Zou1996) As a consequence, the studies might not recommend an optimal structure ofgovernment expenditure Following the site of thinking, literatures record the hypothesis

on a relationship between economic growth and public investment in one of the coreinfrastructure (the provision of transport infrastructure such as highway, road, port orairport)

Based on cross country analysis, Antle (1983) determined the effects of transportand communication infrastructure improvement on aggregate agriculture productivity Sheexplored that transport and communication weakness is the most constrain in lessdeveloping countries (LDCs) to raise their production This idea later is corresponding tothe conclusion of Easterly and Rebelo (1993) when they found a strong and robustassociation between growth and public investment in transport At the same time, whenexploring the case of US, Canning and Fay (1993) also found a large estimated rate ofreturn for transport and telephone systems at about 40 percent Based on this finding theycome to conclude that developed countries with higher income will have normal rate ofreturn for transport infrastructure while newly industrializing countries like South Koreaand Chile have a very high rate of this return

This idea is consistent to a finding of Boarnet (1995) In a study on highway andstreet capital in California counties from 1969 to 1988, Boarnet argued for a significantimpact of highway and street capital to the county gross product which is around 0.19 to0.2 percent depending on common OLS regression or regression with partial and time lag

of capital Based on his figure, Boarnet highlights that infrastructure can be productive atsmall geographic scale of areas

Recently, a study by Boopen (2006) has proved for the argument Not focus ondeveloped countries as fore-authors, Boopen highlights the link between transport capital(government and local capital) and growth in developing and underdeveloped countries inSub Saharan Africa over the period 1980-2000 The study used both cross sectional andpanel data analysis In both cases, his empirical results show a contribution of transportcapital to the economic progress of those countries Depending on different methods, hefound a positive elasticity of transport capital to the countries’ total output from 0.019 withpanel data to 0.23 with cross section and pool data analysis

In other cited explanations relating to transport improvement and economic growth,the literature has been noticed other streams of hypothesis that reduced transportation costsand increased accessibility in firm’s production This argument refers to the importance oftransportation benefits which not only impacts directly on productivity and growth but alsowork through other important revenues such as higher level of private investment, widermarkets, specialization and economies of scale increasing (Seitz,1993; Nadiri andMamuneas, 1998; Piyapong, 2008)

Focusing on productivity effect of transport investment in firm’s production,Seitz(1993), Nardiri and Mamuneas(1998), and Cohen and Paul(2004) use cost function toexamine the effect In this line of studies, stock of highway investment is considered as afixed and free input that influences production technology For example, in a study in

1993, Seitz examined panel data of 31 German manufacturing industries from 1970 to

1989 and found that more capital stock in transport infrastructure capital could enhanceprivate production productivity This argument is consistent to a finding by Deno(1988)when he focuses on firm’s production to analyze effect of highway investment onmanufacturing production decision (i.e output produced or input employed) by usingmanufacturing data of US metropolitan areas from 1970-1978 His result shows that theelasticity between investment in highway and profit increased is 0.08

In general, literature shows several different results on significant level ofrelationship between investment in transport infrastructure and economic growth.Evidences from the literature reflect the complexity and difficulty in estimating the effect’smagnitude on an economy A link between investment in transport and economic growthcould be recognized but the direction of causation may not be unambiguous However,empirical studies present a domination of analysis which highlights a significant linkbetween public investment in transport and growth However, almost of those studies focus

on the case of developed countries, little empirical work has been done for the case ofunderdeveloped and developing countries

Statistics in recent years have shown the fact that in less wealthy countries, withstill relatively small stocks of both private and public capital, investment might have higherrates of return than in more developed economies Moreover, in public investmentallocation policy, there are a difference between developed and developing countries Indeveloping countries, a fact easily to see is that there is a dominant trend of publicinvestment in core infrastructures, of which public spending in transport improvement isalways a priority With this policy strategy, many developing countries have obtained

remarkable economic achievement Vietnam can be considered as an example Clarifyingcontribution of public investment policy to social and economic achievement of thecountry is more concerned by international and Vietnamese studiers.

2.2.2 Vietnam studies review

Catching up with the tendency of international development studies, in thecondition of one developing country with an impressive economic growth performance, thelink between government investment and economic growth in Vietnam case recently hasbeen widely analyzed

Shenggen Fan, Huong and Long (2004) examine the problem of governmentspends, agriculture production and poverty reduction in Vietnam and find that somespecific components of government investment have statistically significant impact on anincrease in agriculture production during the studied period One of the most importantcontributions to the country production comes from public investment in transportinfrastructure According to them, a capital increase in transport improvement will lead to

an increase in agriculture product and reduce poverty in Vietnam

Anh (2008) analyses the relation between components of government spending andeconomic development in process of provincial transformation in Vietnam By usingprovincial statistical figures in period 2001-2005 his purpose is to clarify optimalallocation of government spends in obtaining maximized economic growth Based on OLSmodel, similar to the finding by Devarajan, Swaroop and Zou (1996) and Ghosh andGregoriou (2008), Anh finds that government spends in some specific fields such astelecommunication and transport infrastructure have positive impacts on economicdevelopment However, level of that impact is different between spending in investmentand in government’s operation (capital and current spending) Transport investment haveboth short term and long term impact on growth but in long term transport investment havelarger significant impact in compared to other government spends such as education,healthcare, agriculture and forest

In other attempts to give an overview of transport infrastructure system in Vietnamand analyze its contribution to the country’s economic development, Alberto (2004)displays the status of all modes of transport system including road, inland waterway,railway, port and airport By giving a description on status of transport infrastructuresystem and investment by government and private in the field, Albertor finds thatVietnam’s transport infrastructure is still weak It needs a more efficient investmentmechanism by the government to meet the demand for process of economic development

This result is similar to findings by Tin (2008) In his study, Tin concentrated on reviewingsituation of transport infrastructure in Viet Nam, descriptively analyzing the contribution

of transport infrastructure to Vietnam economic development process He argues thatdeveloping infrastructure and transport infrastructure in particular need to be first step ineconomic development process and level of infrastructure development should be higherthan the rate of economic growth Weak transport infrastructure system is one of the threemain constraints to economic growth including labor force and administrator system Inorder to gain higher economic growth, it is essential to boost investment in transportsystem by using all government and private resources

From literature, it is very obvious that almost current studies on Vietnam still focus

on physical link between transport infrastructure and the country’s economic growth.Study by Anh (2008) is one of the very little empirical studies on the link However, hisresearch mainly focuses on examining components of Vietnamese government spend,include capital spending on transport infrastructure, and does not particularly analyze thelink between public investment in transport and economic growth Moreover he onlyexamines data collected in 5 year (2001-2005), while it is need longer time enough to give

a more confident conclusion about the impact of public investment in transport oneconomic growth

Moreover, existing researches on the subject are only descriptive It is easilyobserved that there is little empirical studies using econometric method to examine therelationship between public investment in transport and growth in Vietnam These reasonsrequire a study based on a quantitative analysis to examine the contribution of investments

in transport on economic growth and to give a comparison between contribution of publicinvestment in transport and that of public investment in other areas The novelty of thestudy is that it attempts to descriptively and quantitatively analysis the contribution ofone component of public capital, transport capital, in the Vietnam case Then the thesisproposes some policy implications on public capital allocation for the country’s policymakers

CHAPTER 3 OVERVIEW OF GOVERNMENT EXPENDITURE AND

PUBLIC INVESTMENT IN TRANSPORT IN VIETNAM

3.1 Overview of Vietnamese government expenditure

Government expenditures or public expenditures are government’s spending onproviding public good and services The government spends money on maintaining itsoperation, and through its expenditure the government regulates the economy In Vietnam,the impressive macroeconomic achievement of the country is a good evidence for thesuccess of government’s budget policy in recent years The adoption of the Budget Law in

1996 and later a new revision in 2005 has established an important framework for a moreorderly and effectively budget system

According to statistics, the ratio of government expenditure to GDP in period

2001-2005 was 31.8%, 6.4% higher than the ratio of that in period 1996-2000 This ratio is evenhigher in 2006 when government expenditure takes 33% of GDP Table 1 shows moredetails on the change in GDP’s ratio of government expenditure from 1986 to 2006

Table 1: Public expenditure as ratio of GDP (%)

1986-1990

1995

creditor 1.1 3.5 3.2 3.1 3.7 4.1 4.8 4.8 4.1 4.2

IV Financial reserves 0.2 0.1 0.0 0.0 0.0 0.1 0.0

Source: Ministry of Finance

As shown in table 1, in the period of 1986-2000 the ratio of public expenditure toGDP was relatively stable which is always less than 25% of GDP However, in period

from 2001 to 2006 there was a dramatically rise in the ratio Reason for this change can beexplained by a change of government strategy in expenditure In the period 2001-2006, theVietnamese government dramatically increases its expenditures to invest in infrastructureimprovement which is considered as the most important factor of economic development.

In the structure of public expenditure, it is obvious that current expenditure which is usedfor government’s operation still takes the largest part, while expenditure on investment isseen the fastest growth

Starting from a country with the primitive system of infrastructure includingtransport, telecommunication, electricity, in order to develop the economy the governmentsettles a target for the country’s infrastructure improvement as a matter of priority To dothis, the government has to pay huge amount of money for investment Evidence for thiscan be seen in table 1 that there is an increase of about 40% in the average ratio of publicinvestment to GDP from the period 1996-2000 to the period 2001-2006

Regarding the allocation of public investment, depending on the ten-year economicstrategy, the government allocates its spending Government’s investing in social andeconomic sectors are allocated in the general inter-relation between those sectors and withthe requirement of the country’s economic development Investment in transportinfrastructure and telecommunication is one of the maters of priority by the government.According to statistics, investment in transport and telecommunication accounts 27% oftotal public investment and 12% of the economy’s investment (not including publicinvestment by state bond and investment credit) Structure of investment by stategovernment is shown in table 2

Table 2: State government investment by sectors

Education, healthcare, society,

cultural and sport

Source: Social and economic development plan 2006-2010 – Ministry of Finance

Note: not include spending from state bond issued

As shown in Table 2, allocation of public investment has been positively changedand focuses on objectives of economic and social development Investment in economicactivities takes the dominant part at 70% of the total investment during five years from

2001 to 2005 Of which investment in agriculture, forestry and aquatic activities accountsfor 13.5%, investment in industry and construction is 44.3%, in transport andcommunication is 12.1% and social sectors is 27%

In sectors of economic activities, investment in transport and telecommunicationaccounts for 12% of total investments from government’s budget However, this figure isvery small compared to actual investment by government The huge money which is notcounted in government’s budget including state bond is also used to invest in transportinfrastructure improvement This investments focus on improving national road systemwhich take an important role in linking big economic centers, rural transport system inremote areas and frontiers

3.2 Transport system and public investment in transport in Vietnam3.2.1 Overview of transport infrastructure system

Vietnam has a transportation system with all modes of transport-roads: road,railway, waterway, seaway and airway Many transportation constructions have been built,repaired and upgraded by domestic investment and foreign aid In the period 1997-2002,

total capital spent on transportation infrastructure investment was 47,488 billions VND, ofwhich capital managed by the Ministry of Transport was 44,051 billions VND 8,924 km

of road was built and upgraded; 61,4 km of road bridge was newly built; 1,253 km ofrailway and 8 km of rail bridge was restored National waterway and seaport system wasupgraded and expanded as well (Report by Ministry of Transport) However, due toVietnam's geographical features and the concentration of population and economicactivity, the transport network is concentrated and traffic density is highest in and aroundthe centers in the northern and southern delta areas North-south connections are provided

by roads, the railway, coastal shipping and air

Roads are the most important modes of the country’s transport-roads to link all thedomestic economic and social activities and to access to the neighboring countries ofChina, Lao PDR and Cambodia The country has a well developed road network of over256,000 km, of which 6.8 percent are national roads, 8.9 percent are provincial roads andthe remainder is district, commune and village roads “Roads are by far the most importanttransport mode, accounting in 1998 for about 71 percent of the total land freight traffic(48 percent in ton) and 80 percent of passenger traffic (65 percent in passenger-km)”(World Bank, 2002)

Table 3: Actual state of road system in Vietnam

Source: Ministry of Transport, 2008

Actual state of national road system in Vietnam (%)

7% 9%

19%

59%

3% 3%

Source: Ministry of Transport, 2008

Also according to the reseach by World Bank, “in terms of coverage the networkcompares favorably with its neighbors and other ASEAN countries, with a road density of0.3km/sqkm, Vietnam equals that of the more developed ASEAN countries (Thailand, thePhilippines and Brunei) However, in terms of quality Vietnam's road network is amongthe worst, only 42 percent of the primary road network (national and provincial roads) andabout 4 percent of the district and commune roads are paved Surveys indicate that 20% ofroad network is in poor to very poor condition and one quarter of the bridges have reducedbearing capacity and need to be rehabilitated or replaced The situation of the provincial,district and commune networks is even worse More than 500 communes still have no roadaccess to district centers, 12 percent of the rural population has no access to motorizedtransport and more than 100,000 km of local roads can only be used during the dryseason” In term of national road, statistics from Ministry of Transport will give moredetails on the situation

Table 4: Actual state of national road system in Vietnam

The composition of road's surface

Actual state of national road system in Vietnam (%)

Source: The Ministry of Transport, 2008

As shown in table 4, more than a haft of total length of road has a width below 7m

In terms of quality, although most of road’s surface is asphalt and asphalt concrete, but still

20 percent of total road system is in bad and very bad condition The poor condition ofroads may be a subject for seasonal flooding This poor condition of transport system,especially road system is a disadvantage of Vietnam compared to neighboring countries Asurvey by the World Bank has shown that 21.6 percent of enterprises talk thattransportation is their major binding constraints on business investment The rate in EastAsia is 15.2 percent and the world is only 12.4 percent

Regarding other modes of transport system, there is about 41,000 km of inlandwaterways available in Vietnam Nealy all of those inland waterways are shallow andnarrow passages so that it is only used for informal and small scale commercial transport

A poor condition is also seen with railway system Built by French hundred years ago, thecountry’s railway system operates a network of 2,832 km from Lao Cai at the Chineseborder and Haiphong in the north to HCMC in the south Because of long time used,almost of railway and rolling stock have deteriorated and are in need of replacement/repair

About sea port, a report by Vietnam Maritime Administration, Vietnam has 49national sea ports, of which 17 sea ports is the first class 1 that take the most important role

in social – economic development of the country or regions In 2008 total good transported

by the country’s seaport system are 196.58 million tons and container is 5,023 TEU.However, because of the weakness of the country’s sea port system, capacity of goodcarried is still so small compared to the potentiality The weakness can be named as

43%

37%

shallow water depths, lack of transport network connected to industry zones and nationalroads, narrow port areas, outdated equipment and poor management are the mainlimitation

In summary, despite recent process of improvement, the productivity of the transportsystem is still low This hinders the government’s exertion to increase the country’seconomic growth rate According to a World Bank’s report, the poor state of the country'sroads imposes economic costs on users of about $160 million a year Problems with portsand railways exist as well This is the reason for expenditure priorities by the governmentwhich has focused on rehabilitation and modernization of transport infrastructure, especiallyroads

3.2.2 Overview of public investment in transport in Vietnam

3.2.2.1 State and local government investment in transport

With poor condition of the country’s transport system after long time of devastation

by wars, in a process of economic development, the rehabilitation and modernization of thecountry’s dilapidated transport infrastructure has been one of the main economic objectives

of the Vietnamese government Priority was assigned to the development of transportinfrastructure in three strategic economic zones i.e the Hanoi – Haiphong corridor in thenorth; Danang in the center and the Hochiminh city - Vungtau corridor in the south; thecreation of links with neighboring countries; the integration of transport systems, includingport development to facilitate international trade; and the development of urban and ruraltransport infrastructure To follow those objectives, the government has spent more andmore money for improving the country’s transport system In 1996, total expenditure of thegovernment including central and local government was about 4.470 billions dong In 2007the value was about 57.770 billions Vietnam dong, 14 times increase in concreted value Inorder to get such huge amount of money for transport investment, the government has tomobilize capital from all sources available including revenue from taxes, officialdevelopment assistance, loan favorite credit or state bond

In terms of government investment structure, investment for transport improvementalways takes an important part which change from 34 percent to 41 percent of total publicinvestment in period 2000-2006 (see table 5)

Table 5: Public investment in transport in 2000-2006

Billion VND

Government expenditure

(in-cluded state bond) 108,961 129,773 148,208 186,183 221,376 271,697 322,049

Capital expenditure

(Invest-ment) – Include state bond 29,624 40,236 45,218 64,629 73,315 88,199 102,332

Transport investment /Total

government investment

(per-cent) 41% 41% 46% 36% 36% 34% 40% Current expenditure 61,823 71,562 78,039 95,608 107,979 132,327 161,852

Financial reserve 846 849 535 111 78 69 135

Source: GSO and Ministry of Finance

It can be seen in table 5 is that government (local and state) has spent a largeamount of money, including public spending from state bond issue, on transport.Regarding state government investment alone, not including investment from state bondissue, the percentage of investment in transport which is shown in table 5 only account lessthan average 12 percent in period 2000-2006 However, total amount of money used byboth state and local government to invest in transport improvement is much higher whichhas an average amount of 24,300 billions Vietnamese dong each year The trend of publicinvestment in transport infrastructure is shown in Figure 1

Figure 1: Trend of public investment in transport infrastructure in constant price

Million VNĐ

Source: General Statistics Office

It can be seen from figure 1 that in overall value, total public investment intransport infrastructure system has continuously rising since 1996 The period of 1996 –

2005 has record a regular rise From 2005 to 2007 there was suddenly change in publicinvestment in the field because in this period the government used a huge amount ofcapital from government bond to finance transport projects This trend is also seen ininvestigating the change in a share of public investment in transport to GDP which will beshown by figure 2 below:

Figure 2: Public investment in transport infrastructure per GDP

(Source: Gross Statistics Office)

Source: General Statistics Office

In Vietnam, almost amount of capital invested in transport infrastructure comesfrom the government’s budget Private capital is only carried on building small transportconstruction for welfare of people in rural area or inside of industrial areas Governmentbudget for investment in transport infrastructure includes local and central governmentbudget There is a decentralized administration in transport infrastructure investment In-vestments by central and local governments are assigned by law Only national transportworks are invested by state budget Local government is responsible for investing otherkinds of transport work However, in the fact, this regulation sometimes will be flexiblyapplied In specific conditions, central government can spend its budget to assist local gov-ernment to perform important projects which beyond capacity of local government’s bud-get, especially for local governments in poor areas Investment in transport by Vietnamesestate and local government will be displayed in Table 6:

Table 6: State and local government’s investment in transport 2000-2006

Source: Ministry of Finance, Ministry of Transport and General Statistic Office

According to statistics, the rate of transportation investment in Vietnam in period2000-2006 as a percent of GDP has been remarkably stable which generally running be-tween 3% The amount of money invested in transport projects regularly increase eachyear As can be seen in table 6, investment by the central government always takes thelargest ratio in total government investment in transport Investment by the central govern-ment reaches the highest ratio at 67.1 percent of total government investment in 2006 Inthe period 2001-2006, state budget for transport investment was always rising Especially,the year 2006 notices a remarkable rise in state investment in the field Total capital dis-bursed of state government’s budget is 24,785 billion VND, account for 87% of total plantcapital In that year, it can be seen a fast disbursement of transport works using capitalfrom state bond issue Moreover, to finance a large demand for transport investment thegovernment also mobilized money from foreign official development aids Capital fromthe foreign official development aids has contributed a remarkable amount of money fortransport investment

3.2.2.2 Transport infrastructure invested by the Ministry of Transport

With its poor condition of transport infrastructure available, all modes of transport– roads are in a need for improvement in Vietnam However, because of the country’s lim-ited budget, the Vietnamese government has to give most priority on urgent projects which

is much important in social and economic development

Table 7: Kinds of transport infrastructure invested by MOT (1996-2007)