April 28, 2005
Joint Document of the Socialist Republic of Vietnam and the World Bank
Prepared with the Support of the Like Minded Donor Group
Report No. 30035-VN
Vietnam
Managing Public Expenditure
for Poverty Reduction and Growth
(In Two Volumes) Volume II: Sectoral Issues
Public Expenditure Review and Integrated Fiduciary Assessment

Abbreviations
ADB
AFD
AFTA
BEC
CFAAs
CIDA
CIEM
CMS
CPAR
CPRGS
DAF
DANIDA
DFlD
DMFAS
DPP
EC
ECORY
EVN
FDI
m1c
GDP
GFS
GFSM
GI0
GSO
HCMC
HCF’Ps
HEPR-JC
Asian Development Bank
Agence Francaise de Developpement
Asean Free Trade Area
Bid Evaluation Committees
Country Financial Accountability
Assessment
Canadian International Development
Agency
Central Institute o f Economic
Management
Cooperative Medical System
Country Procurement Assessment
Report
Comprehensive Poverty Reduction and
Growth Strategy
Development Assistance Fund
Danish International Development
Agency
Department o f International
Development
Debt Management and Financial
Analysis System
Department for Public Procurement
European Commission
Research and Consulting
Electricity o f Vietnam
Foreign Direct Investment
Federal Deposit Insurance Corporation
Gross Domestic Product
Government Financial Statistics
Govemment Finance Statistics Manual
Govemment Inspection Office
General Statistics Office
H o Chi Minh City
Health Care Funds for the Poor
Hunger Eradication, Poverty Reduction
and Job Creation
HIVIAIDS Human Immunodeficiency
IMF
IPSAS
JBIC
KFW
LANS
LMDG
MARD
MIS
MOET
MOF
MOHA
M O H
MOLISA
VirusIAdquired Immunodeficiency
Syndrome
International Monetary Fund
International Public Sector Accounting
Standards
Japan Bank for International
Cooperation
Kreditanstalt fur Wiederaufbau
Local-Area Networks
Like Minded Group o f Donors
Ministry o f Agriculture and Rural
Development
Management Information System
Ministry o f Education and Training
Ministry o f Finance
Ministry o f Home Affairs
Ministry o f Health
Ministry o f Labor, Invalids and Social
Affairs
MPI
MPDF
M P I
MOT
MTEF
MTFF
NGO
NMDC
NORAD
NPL
NTP
ODA
OECD
O&M
PAR
PER-FA
PFMR.P
PIP
PMU
PRSC
SAV
SBD
SDC
SDU
SEDP
SIDA
S I 0
SOCBs
SOEs
TABMIS
UCOA
Ministry o f Planning and Investment
Mekong Project Development Facility
Ministry o f Planning and Investment
Ministry o f Transport
Medium-Term Expenditure Framework
Medium-Term fiscal Framework
Non-Governmental Organization
Netherlands Ministry Development
Corporation
Norwegian Royal Agency
Non-Performing Loans
National Target Programs
Official Development Assistant
Organization for Economic
Cooperation and Development
Operation and Maintenance
Public Administration Reform
Public Expenditure Review and
Integrated Fiduciary Assessment
Public Finance Management Reform
Project
Public Investment Program
Project Management Unit
Poverty Reduction Support Credit
State Audit o f Vietnam
Standard Bidding Document
Swiss Development Cooperation
Service Delivery Units
Socio-Economic Development Plan
Swedish International Development
Cooperation Agency
State Inspection Office
State-owned Commercial Banks
State Owned Enterprises
Treasury and Budget Management
Information System
United Charts o f Accountants
UNCTAD United Nations Commission on Trade
UNDP
USBTA
V A T
VDGs
VDR
VEA
VHLS S
VND
vss
WANS
and Development
United Nations Development Program
United States Bilateral Trade
Agreement
Valued Added Tax
Vietnam Development Goals
Vietnam Development Report
Vietnam Economics Association
Vietnam Household Living Standards
Survey
Vietnam Dong
Vietnam Social Security
Wide-area Networks

Table of Contents
Volume 2
10 . EDUCATION SECTOR EXPENDITURE ............................................................... 1
Introduction and Overview ............................................................................................. 1
Background and Policies Within The Education Sector ................................................. 1
Financing o f Education Spending ................................................................................... 2
Service Performance ....................................................................................................... 8
Planning and implementing o f sector priorities ............................................................ 24
Recommendations ......................................................................................................... 26
TRANSPORT SECTOR EXPENDITURE ............................................................. 29
Introduction and Overview ........................................................................................... 29
Sector Overview And Performance .............................................................................. 29
Overall Public Expenditure in Transport ...................................................................... 35
Future Expenditure Requirements ................................................................................ 47
Summary and Recommendations ................................................................................. 49
HEALTH SECTOR EXPENDITURE ..................................................................... 53
Introduction and Overview ........................................................................................... 53
Achievements and Challenges ...................................................................................... 53
Health Spending in the Period 199 1-2002 .................................................................... 57
Evolution o f Health Policy and Policy Instruments ...................................................... 60
Need for a Sharpened Paradigm ................................................................................... 70
Implementation ............................................................................................................. 71
Recommendations ......................................................................................................... 73
AGRICULTURE AND RURAL DEVELOPMENT SECTOR EXPENDITURE .. 79
Introduction and Overview ........................................................................................... 79
Sector Context ............................................................................................................... 79
Main Impacts and Scope for Improvement ................................................................... 87
Sub-sector Public Expenditure Issues .......................................................................... 91
Recommendations ......................................................................................................... 98
NATIONAL TARGET PROGRAM EXPENDITURE ......................................... 101
Introduction and Overview ......................................................................................... 101
Regulation and Management o f NTPs ........................................................................ 102
Financing o f National Target Programs ...................................................................... 105
Fiduciary Risk in HEPR-JC and Program 135 ........................................................... 107
Impact and Effectiveness o f NTPs .............................................................................. 110
Recommendations ....................................................................................................... 114
11 .
12 .
13 .
14 .
BOXES
Box 10.1 : Key Targets in the Education Sector ................................................................... 2
Box 10.2: Key Enabling Objectives for the Education Sector ............................................. 3
Box 12.1 : Early Experiences with Decree 10 in Public Hospitals ..................................... 69
Box 12.2: Budgeting for Central Hospitals ........................................................................ 73
Box 13.1 : Poverty Reduction and Growth Impact o f Rural Spending .............................. 88
Box 14.1 : Effectiveness of HEPR-HC Program .............................................................. 112
Box 14.2: Impact of Program 135 on Poverty ................................................................. 113
i
Box 14.3 : Effectiveness o f Health Targeted Programs .................................................... 113
FIGURES
Figure 1 1.1: Measures o f Rural Access ............................................................................. 31
Figure 11 :2: 40ft Container Transport Costs from Factory. via Nearest Port to Japan ..... 33
Figure 1 1.3: Comparison o f Network Deterioration Given Different Budget Scenarios .. 40
Figure 12.1 : Regional GDP per capita and Health Expenditure per capita as a
Percentage o f national Averages - 2000 ................................................................ 55
Figure 12:2: Lorenz Distributions o f Local-Level Government Health Spending and GDP
per capita among Provinces ................................................................................... 56
Figure 12.3 : Health Expenditure by Region and Source o f Funds ................................... -57
Figure 12.4: Central-local Shares o f Total Government Health Expenditure ................... 59
Figure 12.5: Economic Composition o f Government Health Expenditure, 1991-2002 .... 59
Figure 12.6: Central Health Spending by Type o f Care ................................................... 60
Figure 12.8: Household Out-of -Pocket Spending on Healthcare per capita during the
Previous 12 months in Constant (2002) VND by Standard o f Living Quintile ..... 61
Figure 12.9: Government Expenditure on National Health Programs .............................. 64
Figure 12.7: Provincial Health Spending by Type of Care ................................................ 60
Figure 14.1 : Shares o f Individual Programs in Total NTP Expenditure .......................... 105
TABLES
Table I O . 1 : Proportion o f Education Sector Expenditure in GDP and Total Public
Table 10.2: Education Expenditure in Several Countries 1999-200 1 Period ...................... 4
Table 10.3 : Budget Share for Education and Training ........................................................ 4
Table 10.4: Economic Composition o f Education and Training Expenditure by Education
Sub-sector ................................................................................................................ 5
Table 10.5 : Capital Expenditure for Education and Training.. ............................................ 5
Table 10.6: Annual per capita per Pupil Salary and Wage Expenditure for School
Education ................................................................................................................. 6
Table 10.7: Share o f Education and Training Expenditure by Budgetary Levels ............... 6
Table 10.8: Funding Sources o f Education Expenditures .................................................... 7
Table 10.10: Number and Condition o f Schools ................................................................. 8
Table 10.1 I: Teacher Numbers by Education Level and School Year ................................ 9
Table 10.12: N e t Participation Rate at Schools ................................................................. 10
Table 10.13 : Variations in Pupil Teacher Ratios in 2002 .................................................. 11
Table 10.14: Variation in Enrolment Rates by Socio-economic Category ........................ 12
Table 10.15: Share o f Pupils by Gender ............................................................................ 12
Table 10.16: Incidence o f Benefit from Education Expenditures ...................................... 13
Table 10.18: Tuition Fee Exemption and Reduction by Income Group in School
Education .............................................................................................................. 14
Table 10.19: Share of Budget-based National Target Programs by Region ...................... 15
Expenditure .............................................................................................................. 3
Table 10.9: Number o f Pupils at Sub-sectors ...................................................................... 8
Table 10.17: Make-up o f Private Costs ............................................................................. 13
11
Table 10.20: Cost per Student (thousand VND) ................................................................ 15
Table 10.2 1 : Pupil Teacher Ratios ..................................................................................... 16
Table 10.22: Unit Cost o f Higher Institutions ................................................................... 16
Table 10.23: Teachers Contact Hours at Grades ............................................................... 17
Table 10.24: Rate o f Literate Employees .......................................................................... 18
Table 10.25 : Promotion and Graduation Rate ................................................................... 19
Table 10.26: Rates o f Class Repetition and Drop-out by Region ...................................... 19
Table 10.27: Attainment o f Mathematics and Reading Skills at Grade 5 ......................... 20
Table 10.28: Years o f Study .............................................................................................. 20
Table 10.29: International Comparisons o f PTRs .............................................................. 21
Table 10.30:Qualification o f Teachers in Higher Education ............................................. 22
Table 10.3 1 : Ratio o f Teacher Salaries and Wages to GDP per capita ............................. 23
Table 10.32: Comparison o f Attainment in Full and Part Day Schooling ......................... 23
Table 1 1.1 : Domestic Transportation Volume 1999-2003 ................................................ 30
Table 1 1.2: Road Length in km by Category and Pavement Type .................................... 31
Table 11.3: Vehicle Fleet ................................................................................................... 32
Table 11.4: Vietnam Airlines Occupancy Rates in Percent o f Seats Offered ................... 34
Table 1 1.5 : International Comparisons .............................................................................. 35
Table 1 1.6: Overview o f Transport Sector Expenditures Based on MOF Data ............... 36
Table 11.7: Overview o f Transport Sector Expenditures .................................................. 37
Table 1 1.8: Central Level Expenditure and Funding Sources ........................................... 38
Table 1 1.9: International Comparison o f Transport Expenditure Levels ......................... -38
Table 11.10: Comparison o f VRA and MOF Recurrent Expenditure Data ....................... 40
Table 1 1 . 12. Expenditure Profile for Provincial Level Roads ........................................... 43
Table 1 1.13 : Regional Differences in Roads Expenditure ................................................. 43
Table 1 1.14: Public Expenditures by Transport Mode ...................................................... 44
Table 1 1.15: VR Revenues and Contributions from the State Budget .............................. 45
Table 11.16: Investment Demand for Transport Infrastructure ......................................... 48
Table 12.1 :Some Key Health Indicators for the Period of 1990-2003 and Targets
for 2010 ......................................................................................................... 54
Table 12.2: Population, Total State Budget Spending, State Budget Spending’ on Health
GDP, 199 1-2002 (current price) ................................................................... 58
Table 12.3: Public and Total Health Expenditure in Relation to GDP among Selected
Asian Developing Countries, 200 1 .............................................................. -58
Table 13.2: Budget Expenditure on Agriculture, Forestry and Fisheries 1997-2002 ........ 82
Table 13.3 : State Expenditure for Agriculture - 1997 - 2002 ........................................... 83
Table 13.4: Capital and Recurrent Budget Expenditure on Agriculture. ........................... 83
Table 13.6: Agriculture Budget by Sub-sector .................................................................. 84
Table 13.7: Financial Statistics o f Agriculture SOEs owned by the Centre ...................... 85
Table 13.9: Investment Capital by Ownership .................................................................. 90
Table 13.10: Vietnam: Public Expenditure on Irrigation ................................................... 91
Table 13.1 1 : Vietnam: MARD’s Debts to Irrigation Construction Companies ................ 92
Table 11.11: Revenue in 2001 by Road User Charge Item ............................................... 42
Table 13.1: Poverty, Growth and Economic Structure, 1998-2003 ................................... 80
Table 13.5: State Expepditure for Salaries and Wages and O&M in Agriculture ............. 84
Table 13.8: ODA to the Agriculture Sector ....................................................................... 87
... 111
Table 13.12: Public Expenditure on Agriculture Research from Central Sources ............ 95
Table 13.13 : Funding o f Agriculture Extension ................................................................ 96
Table 14.1 : Share o f NTP Spending in Total State Spending .......................................... 105
Table 13.14: Public Expenditure on Forestry and Related Services .................................. 97
Table 14.2: Relationship between HEPR Central Funding and Poverty Rates .............. 112
iv
10. EDUCATION SECTOR EXPENDITURE
INTRODUCTION AND OVERVIEW
10.1 Vietnam’s rapid progress with poverty reduction and economic growth was
triggered b y economic reform. At the beginning o f the reform period, however, Vietnam
had much higher levels o f literacy, and better developed middle and higher education,
than most other countries with similar levels o f per capita income. Further development
o f human capital w i l l be essential if the current rapid economic development i s to be
sustained.
10.2 This chapter reflects on progress made through education sector expenditure since
the 2000 PER. In general, substantial progress has been made in financing the sector,
participation rates have increased and physical infrastructure has improved. Gaps in those
participating and benefiting from education have narrowed, but remain a concern.
Nevertheless, important challenges are faced in improving the quality, equity and
effectiveness o f the service.
10.3 The first section o f the chapter reviews the background and policies within the
education sector. The second section reviews financing o f education spending. The third
section assesses the performance of the education sector. The fourth section considers the
approach to planning and implementing the sector’s priorities. A fifth section presents
conclusions and recommendations.
BACKGROUND AND POLICIES WITHIN THE EDUCATION SECTOR
10.4 There are four levels o f schooling in Vietnam-pre-primary, primary (5 years
from age 6), lower secondary (LS) (4 years) and upper secondary (US) (3 years). B y far
the greatest provision i s in public schools. A small element i s in semi-public schools
where Government provides the facilities, but parents meet the operating costs, including
salaries. In addition, Vocational Technical Education and Training (VTET) i s available
after primary and L S . Higher education i s available in colleges and universities. In
addition, there i s a continuing education system which provides a variety of facilities.
This whole system, consisting of education and training, w i l l be referred to in this PER-
IFA as “the education sector”.
10.5 The Ministry of Education and Training (MOET) has overall policy responsibility
for the sector. Broadly, the Ministry manages higher institutions. Provinces manage
upper-secondary schools; and districts and communes manage lower-secondary, primary
and pre-primary facilities. In practice this pattern varies around the country. Service
provision has become increasingly decentralized through removal o f controls from the
center and increased revenue raising powers and related incentives at the local level.
MOLISA has specific responsibilities with regards to VTET.
1
2 Public Expenditure Review and Integrated Fiduciary Assessment
10.6 The main objectives o f the Government's education policy are contained in the
Comprehensive Poverty Reduction and Growth Strategy (2002), the Education Strategic
Development Plan for 2001 - 2010 (2003), the Education for All Plan (2003) and the
Secondary Education Master Plan (2002). An overarching objective i s to increase
participation in all levels o f education, particularly o f disadvantaged communities; and in
primary education to achieve full participation b y the end o f the decade. A further
objective i s to increase the quality and relevance o f education to develop citizenship and
build foundations for strong growth in an open world economy. Another aim i s to
improve the management o f education at all levels, in particular b y decentralizing
management within a framework o f national standards.
10.7 Key targets and key enabling objectives are outlined in Boxes 10.1 and 10.2.
Box 10.1: Key Targets in the Education Sector
Pre-primary education:
Participation in crbches under age 3 - 15 percent by 2005 and 18 percent by 2010 and in
kindergartens between ages 3 and 5 years - 58 percent by 2005 and 67 percent by 2010.
Participation in kindergartens by 5-year-old children in preparation for the 1'' grade at school - 85
percent by 2005 and 95 percent by 2010.
General education:
B y 2005 participation of primary age children at 97 percent, and 99 percent by 2010.
Participation of L S school-age children - 80 percent by 2005, and 90 percent by 2010.
Participation of U S school-age children - 45 percent b y 2005, and 50 percent by 2010.
Continuing education:
B y 2005, 10 percent o f the vocational school age population participates in vocational and technical
schools, rising to 15 percent by 2010. Transition from L S to vocational schools 10 percent by 2005,
and 15 percent by 2010.
High-skill vocational programs shall be introduced, attracting 5 percent (by 2005) and 10 percent
College, under-graduate and post-graduate education:
B y 2010, there shall be 200 students per 10 000 population, 38 000 people shall participate in master
programs, another 15 000 shall follow PhD programs.
Education for children with disabilities:
B y 2005, 50 percent o f children with disabilities will be educated, rising to 70 percent by 2010.
(by 2010) of graduates from US and vocational schools.
FINANCING OF EDUCATION SPENDING
Share of Education in Total Public Spending
10.8 I t i s clear that Government has given priority to education expenditures in policy
and in practice. Table 10.1 indicates public education spending as a share o f total public
spending and as a percentage o f GDP.
Education Sector Expenditure 3
Box 10.2: Key Enabling Objectives for the Education Sector
Teachers:
By 2010, 40 percent o f teachers at primary schools shall graduate from colleges.
By 2005, 100 percent o f teachers at L S shall graduate from colleges or higher, heads and deputy head
of professional groups shall be university graduates.
By 20 10, 10 percent o f teachers at U S shall graduate from master programs.
Injwstructure:
For basic education in every commune (or group in sparse areas), there shall be at least one L S and
one U S school that reach national standards.
In every province there shall be at least one focal U S school.
The system o f ethnic boarding schools shall be strengthened and expanded.
Finance
State expenditure for education shall reach 6 percent o f GDP by 2005, and 7 percent by 2010.
Education share o f public expenditure shall grow from 15 percent in 2000 to 18 percent by 2005 and
20 percent by 2010. It will be increasingly focused on universal education; education in ethnic and
disadvantaged areas; fields where non-state funding i s rare; to ensure "averagely good" leaming
conditions for the whole country as well as provide leaming opportunities for disadvantaged children
(children in low-income families).
Table 10.1: Proportion of Education Sector Expenditure in GDP and Total Public
Expenditure
Total Price Base Trillion VND
1994 1998 2000 2002
GDP Current 170.3 361.5 441.6 536.1
1994 170.3 244.7 274.3 312.9
Public Expenditure Current 42.8 73.4 103.2 133.4
1994 42.8 49.7 64.1 78.0
Education Sector Current 6.0 12.8 15.6 22.6
1994 6.0 8.6 9.7 13.3
Education share o f GDP % 3.5 3.5 3.5 4.2
Education share o f PE % 14.0 17.4 15.1 16.9
Source : Department of Planning and Finance, MOET, Department ofstate Budget, MOF
10.9 In recent years, Vietnam has achieved a high and stable growth rate o f GDP. In
the period 2000-2003 the annual rate o f growth was 7 percent. Furthermore, education
sector spending in relation to GDP has risen from 3.5 percent in 1994 to 4.6 percent in
2004. Compared with other sectors, education has received priority, i t s share o f public
expenditure increasing from 14 percent to 18.6 percent. Compared with other countries in
the region, Vietnam has moved above some nations such as Indonesia, India and
Pakistan. However, in relation to GDP, public expenditure on education remains below
that o f Thailand and Malaysia. Progress towards fulfilling the planned strategic objectives
for education as a share o f public spending (20 percent in 2010) will allow increasing
focus on educational quality and effectiveness as a contribution to national objectives.
4 Public Expenditure Review and Integrated Fiduciary Assessment
Table 10.2: Education Expenditure in Several Countries 1999- 2001 Period
Nation Education as % of Education as % o f
GDP Total PE
Japan 3.6 10.5
Singapore 3.6 17.4
India 4.1 12.7
Indonesia 1.3 9.8
Pakistan 1.8 7.8
Thailand 5.0 31.0
Vietnam (1999-2001) 3.4 14.8
2004 4.6 18.6
Malaysia 7.9 20.0
Source: Other countries: UNDP. Human Development Report 2004, Vietnam: Dept. of State Budget, MOF
Composition of education spending by sub-sectors
10.10 There has been a shift o f sub-sector shares in education and training spending,
with spending on education increasing while spending on training has seen the reverse
trend. This reflects the priority accorded to basic education and education in the remote,
ethnic and disadvantaged regions in Vietnam. In 1998, education accounted for 73.3
percent while training accounted for 26.7 percent of total education and training
expenditure; b y 2002, as the result o f the faster “socialization” (user charging) process
taking place in the training sub-sector, the share o f education rose to 77.7 percent while
that o f training f e l l to 22.3 percent. Within school education, spending for primary
education declined (due to the annual reduction in the number o f pupils b y nearly half a
million) while spending for lower and upper secondary education increased (as a result o f
the increased number o f lower-secondary and upper-secondary pupils during this period).
In the near term, secondary education w i l l require additional resources to meet
universalization targets. The share of spending for higher education and training fell from
12.4 percent in 1998 to 9.7 percent in 2002, reflecting a reduction o f state subsidy to the
sub-sector that Government believes can rely most on tuition charging (Table 10.3).
Table 10.3: Budget Share for Education and Training
Total Budget Expenditure for Education and Training 100% 100% 100% 100%
1998 2000 2001 2002
For Education : 73.30 75.86 76.01 77.68
- Pre-primary 5.40 6.71 6.97 6.97
- Primary 35.27 32.17 32.71 31.61
- Lower Secondary 19.38 20.44 20.32 21.32
- Upper Secondary 8.33 10.02 11.02 10.40
- Others 4.92 6.52 4.99 7.39
For Training : 26.29 24.14 23.99 22.32
- Vocational 3.79 3.06 3.30 3.24
- Continuing education 4.80 3.54 3.22 2.86
- Higher education 12.43 9.27 9.58 9.71
- Post-graduate 0.81 0.45 0.48 0.46
- Others 4.86 7.82 7.40 6.05
Source: MOF
Education Sector Expenditure 5
Economic composition of education and training spending
10.11 Public spending on education and training covers both capital and recurrent
expenditures. During the 1999-2002 period, there has been little change in the economic
composition o f education and training spending, with capital expenditure fluctuating
around 73 percent and recurrent expenditure around 27 percent (Table 10.4). However, in
absolute term, capital spending nearly doubled during this period from 2,418 trillions
VND in 1999 to 4,375 trillion VND in 2002. Most of capital expenditure was spent on
construction o f facilities; purchase o f fixed assets and major repairs accounted for 10
percent. During the same period, pre-primary and lower secondary education saw the
fastest growth o f capital expenditure (about 2 times) followed b y upper secondary
education (1.8 times) and primary education (1.5 times) (Table 10.5).
Table 10.4: Economic Composition of Education and Training Expenditure by Education
Sub-sector (%)
1999 2000 2001 2002
Capital Recurrent Capital Recurrent Capital Recurrent Capital Recurrent
Total Expenditure 25.9 74.1 26.7 73.3 27.5 72.5 26.8 73.2
Education 23.9 76.1 23.5 76.5 23.0 77.0 24.9 75.1
- Pre-primary 22.1 77.9 19.7 80.3 22.7 77.3 27.2 72.8
- Primary 19.8 80.2 18.5 81.5 18.7 81.3 18.4 81.6
- Lower
Secondary 21.0 79.0 19.9 80.0 20.2 79.8 22.0 78.0
- Upper
Secondary 35.8 64.2 31.8 68.2 35.9 64.1 33.5 66.5
Training 32.3 67.7 36.6 63.4 41.6 58.4 33.2 66.9
- Continuing Ed. 35.1 64.9 27.4 72.6 38.0 62.0 37.0 63.0
- Higher Educ. 32.8 67.2 29.3 70.7 35.0 65 .O 31.4 68.6
Source: MOF
Table 10.5: Capital Expenditure for Education and Training
1999 2000 2001 2002
B y classification
Total expenditure (VND bil.) 2,418 2,918 3,409 4,375
- Share o f capital exp. (%) 66.9 80.3 89.8 62.0
- Share o f fixed assets purchase and major repairs (%) 33.1 19.7 10.2 38.0
1, Pre-primary 217 217 308 428
2. Primary 830 976 1,195 1,312
3. Lower Secondary 5 44 667 802 1,058
4. Upper secondary 438 520 77 1 786
5. Others' 389 538 334 79 1
By education levels (VND bil)
Source: MOF
10.12 There was a difference between education and training in the composition o f their
recurrent expenditure. In education, salary and wages accounted for 71.3 percent o f total
recurrent spending while in training this expenditure category made up only 27.4 percent.
From 1999 until now, per pupil teacher salary spending has increased in all sub-sectors.
~~ ~
Other education includes boarding schools, schools for talented students, etc.
6 Public Exuenditure Review and Integrated Fiduciary Assessment
The fastest growth was recorded in primary education where per pupil teacher salary
spending nearly doubled within four years (from VND 263,896 per pupil in 1999 to VND
516,023 in 2002), followed b y upper secondary education (1.62 times), lower secondary
education (1.6 times) and pre-primary (1.52 times). (See Table 10.6.)
Table 10.6: Annual per capita per Pupil Salary and Wage Expenditure for School
Education
1999 2000 2001 2002
Pre-primary 432,266 500,710 755,033 813,106
Primary 263,896 349,284 448,775 516,023
Lower Secondary 272,991 350,064 395,4 10 436,714
Upper Secondary 398,028 486,594 586,447 636,446
Source: Financial and Planning Dept., MOETand State Budget Dept., MOF
Decentralized funding
10.13 State budget management in education i s increasingly decentralized providing
greater autonomy to local government as well as educational and training institutions.
Local authorities are responsible for budget allocations to finance local educational and
training activities performed at basic educational sub-sectors and locally-managed
training institutions (Training).
Table 10.7: Share of Education and Training Expenditures by Budgetary Levels
1999 2000 2001 2002
Central Local Central Local Central Local Central Local
Total 19.34 80.67 17.13 82.87 16.2 83.8 14.65 85.35
Education 6.58 93.42 3.58 96.42 2.24 97.76 2.07 97.93
Training 60.46 39.54 59.70 40.30 60.42 39.58 58.45 41.55
Source: Department of State Budget, MOF
Private contributions
10.14 Since 1994, non-state education has increased. The most significant areas o f
provision are in U S where provision increased from 20 percent o f total provision in 1994
to 32 percent in 2004 and in pre-school from 30 percent to 58 percent over the same
period.
10.15 There are two types o f contribution made b y parents: compulsory and optional. In
accordance with the Government regulations, the main contributions from parents include
tuition fees, examination fees, and contributions for school construction. These
contributions, regarded as state budget revenues, are collected and retained b y
educational institutions to finance educational activities.
10.16 The 2002 Vietnam Household Living Standard Survey (VHLSS 2002) indicates
that the average parental expenditure on education was 627,000 VND per student, 14.6
percent higher than in 1997-1998. Educational spending has increased in all areas and
income groups. However, the level o f household spending on education varies
considerably. In urban areas, the annual household spending on education per student was
1,255,000 VND, 3 times above that in rural areas. Spending o f the wealthiest households
was 1,418,000 VND, 6 times higher than that o f the poorest. Household spending on
Education Sector Expenditure 7
education was highest in the South East (1,139,000 VND) and smallest in the North West
(278,000 VND). Table 10.8 indicates the share o f state subsidy and parents’ contributions
(on tuition fee and direct spending) in education. Although the absolute amount of private
expenditure has grown in all sub-sectors, the share o f state budget has accelerated
reflecting the Government’s view o f education as the top priority in public expenditure.
Nevertheless, the variations continue to contribute to significant variations in total
expenditure per student and the quality o f education received.
Table 10.8: Funding Sources of Education Expenditures (%)
Sub-sector 1993 1998 2002
Primary:
Contributions and direct expenditures by parents 55 45 27
Lower secondary:
Contributions and direct expenditures by parents 66 38 41
Upper secondary:
State subsidy 40 47 52
Contributions and direct expenditures by parents 60 53 48
Source: VHLSS, 2002
State subsidy 45 55 73
State subsidy 34 62 59
10.17 With a view to greater autonomy in revenue-earning state-owned administrative
units, the Government issued Decree 10/2002/ND-CP (see Chapter 7 for further details).
The main concept i s more decentralization to service delivery units such as schools which
now have more responsibly for effective and efficient budget management. So far, there
has not been a formal review o f Decree 10’s impact in education and training and limited
information was available during the course o f the PER-IFA. Nonetheless, feedback
collected from schools by MOET has indicated that:
a At present, in many areas financial self-autonomy has not been realized in schools
(pre-primary, primary and L S schools).
In areas where the Decree has been implemented, it has provided VTET schools,
colleges and universities located in economically developed regions with valuable
opportunities in terms o f service improvement resulting from financial autonomy.
In disadvantaged areas where incomes are low, the Decree has generated
significant challenges. Due to limited revenues, and the responsibility to finance
part o f teacher salaries and wages when the minimum wage i s raised, spending to
improve service quality has been squeezed in these areas. In particular, challenges
are faced b y mountainous areas, disadvantaged and poor regions because o f
limited contributions from parents and community.
10.18 Government’s Comprehensive Poverty Reduction and Growth Strategy and
Sector strategies indicate an aim o f eliminating charging in the primary and LSS sub-
sectors. On the other hand, the recent introduction o f Decree 10 creates added incentives
for charging at the operational level. There are now several policy statements that touch
upon the role of contributions as a component o f overall sector financing that potentially
pull in different directions. The point has now been reached when there i s perhaps a need
to step back and consider the role o f charges and contributions in the Sector in a more
strategic and consolidated way. Such a review would evaluate the resources yielded b y
0
a
8 Public Expenditure Review and Integrated Fiduciary Assessment
charging alongside other strategic variables such as investment in improved service
quality. For example, i s the ability to extend expenditures in some areas more important
to sector objectives than the possibly adverse affect o f charging to help finance it? Only
b y integrating revenue raising policy in this way, through the medium-term expenditure
framework (MTEF) being piloted in the sector, can a trade-off between the variables be
made with a view to optimizing the effectiveness o f the service as a whole.
SERVICE PERFORMANCE
Service access and infrastructure
10.19 Rapid progress has been made in achieving the Government’s aim o f increased
participation in the education system. This trend has been overlaid in more recent years
b y a significant demographic shift towards falling school age populations. These trends
are reflected in Table 10.9. The number o f primary pupils peaked in 1998 at a time when
net participation rates moved above 90 percent. Although the participation rate has
continued to edge up towards national targets in primary, the absolute number o f pupils
began to fall around 2000 and continues to fall rapidly. B y 2004, the number of pupils i s
down about 18 percent on the peak. Although the impact o f falling school age
populations has begun to edge into LS, pupil numbers continues to increase because o f
increased participation. But there w i l l be a point where the demographic shift must
ultimately cause falling numbers there too.
Table 10.9: Number of Pupils at Sub-sectors (thousands)
2000-01 2002-03 2003-04 1994-95 1998-99
Pre-primary 1777 2 171 2 114 2 144 2 173
Primary 10 048 10 250 9 751 8 841 8 350
Lower Secondary 3 679 5 565 5 918 6 497 6 612
Upper Secondary 863 1658 2 200 2 453 2 616
Vocational 146 217 255 310 3 60
College 57 158 186 214 23 1
University 260 602 689 747 801
Source: Department of Planning and Finance, MOET
10.20 In the year 2003/4 the building infrastructure had increased to the levels set out in
Table 10.10.
Table 10.10: Number and Condition of Schools
Sub-sector Number of Schools Meeting National
Standards
Pre-school 10,104 3%
Primary 14,346 20%
Lower Secondary 9,873 2%
Upper Secondary 2,140 1%
Source: Department of Planning and Finance, MOET
10.21 In mountainous provinces and districts, the network o f half-day and boarding
ethnic schools i s being strengthened and expanded. These schools are now moving to
train cadres from ethnic minority groups.
Education Sector Expenditure 9
10.22 The system of Continuing Training Centers and vocational schools, colleges and
universities has also expanded rapidly. By the 2003/4 school year, there were 546
Vocational Schools (including 226 Vocational and 320 Technical Training Centers), 286
combined Vocational and Continuing Training Centers, and 214 colleges and universities
(not including those managed b y the security and national defense sector). In total, the
country has 147 post-graduate education institutions (including 116 granting masters
degrees and 95 granting doctoral degrees).
10.23 A network o f non-formal education institutions has been growing rapidly.
Presently, there are 556 continuing training centers at the provincial and district levels
(57 provincial centers and 499 district ones), 86 continuing education schools and about
4,000 community learning centers.
10.24 In spite o f the expansion in infrastructure there are several challenges that the
Government recognizes the need to address:
0 The expansion o f crkches and kindergartens in mountainous villages;
0 The construction and development o f private-financed full day schools, and lower
secondary schools in mountainous, remote and offshore areas;
The establishment o f a Technical upper secondary school model;
The sustainable development o f Community and Remote Training Centers;
The expansion of vocational schools in Central Highlands and Mekong Delta;
The mobilization o f capital to invest in local, national, and multi-specialist
universities consistent with standardization, modernization and socialization.
The upgrading o f buildings to meet national standards while also constructing
buildings to reduce shift working requires careful planning, prioritization and
implementation management that i s a challenge for management at all levels.
10.25 The number of teachers has also expanded significantly. At the beginning o f
2003-2004 school year, the total number of teachers reached 950,725, o f which non-
public school teachers accounted for 13.6 percent (see Table 10.1 1).
Table 10.11: Teacher Numbers by Education Level and School Year
0
0
0
0
0
1998-99 2000-01 2003-04 2003-2004 in
comparison with
Total 772,960 836,136 950,725 177,765
Pre-primary 144,580 146,871 150,335 7,3 14
Primary 336,792 347,833 362,627 25,835
Lower Secondary 194,237 224,840 280,943 86,706
Upper Secondary 54,324 74,189 98,714 44,390
Vocational 5,849 5,500 7,056 1,207
Continuing education 9,732 10,189 11,121 1,389
University 2 1,229 24,362 28,434 7,205
Source: Department of Planning and Finance, MOET
1998-1999
Higher education 6,806 7,843 11,551 4,745
10 Public Expenditure Review and Integrated Fiduciary Assessment
10.26 I t i s notable that the teacher numbers continue to grow in Primary at a time when
student numbers are falling. The phenomenon i s also beginning to feed into LS. I t points
to both the opportunity and challenge that Vietnam faces to focus attention on the quality
and effectiveness o f the service provided rather than meeting the pressure o f rising
numbers in the past. Choices about priorities are opening up. Achieving the potential
gains, however, places new challenges on management and the need for active use and
redeployment o f resources to maximize their impact.
10.27 Since 1994, participation b y the relevant school-age population (Net Participation
Rate, NPR) has increased in all sub-sectors. B y 2003, the NPR reached 97.5 percent in
primary education, 80.6 percent in lower-secondary education and 36.6 percent in upper-
secondary education (Table 10.12). However, to make further progress i t i s clear that i t i s
important to pay attention and invest in poor, geographically-disadvantaged regions such
as the Central Highlands, the Northern Mountains and the Mekong River Delta (where
there i s a dense population o f ethnic groups). Also, attention i s required to attract more
children with disabilities to the educational system.
Table 10.12: Net Participation Rate at Schools
Sub-sector 1994 1998 2003
Primary 91.4 88.2 97.5
Lower Secondary 41.9 57.6 80.6
Upper Secondary 12.7 25.7 36.6
Source: Department of Planning and Finance, MOET
10.28 A poverty mapping exercise carried out in 2002 provided a more comprehensive
view o f spatial patterns o f relative wealth. Data from this exercise has been used to try
and draw some conclusions about the extent to which social and economic barriers to
educational performance are being overcome. Tables presented in this section use data
from the mapping exercise converted into an index o f relative levels o f poverty using the
Red River Region and Hanoi within it as a base o f 100.
10.29 A number o f measures are already being taken by Government to overcome the
impact o f social and economic disadvantage. In particular:
0 An adjustment i s made within resource distribution norms for education to reflect
special needs-for example, the need for lower pupil teacher ratios in sparsely
populated areas and the need to pay teachers more to go to remote areas.
Safety nets are applied to vulnerable provinces, for example, with regards to
maintaining a minimum level o f non-salary operational expenditure.
Targeted programs have been introduced to help tackle the specific needs o f
vulnerable areas (mountainous, ethnic and remote).
10.30 On the other hand, a number o f factors favor wealthier areas-in particular the
ability to raise and retain revenue both through the tax system and to make direct charges
to parents. Therefore, i t i s important to examine the reality in terms o f what matters most:
the level o f resources at the point o f service delivery to an individual pupil after all o f
these factors and the extent to which they help to overcome barriers to participation and
0
0
Education Sector Expenditure 11
attainment where they exist. Dealing firstly with variations in resource inputs, Table
10.13 indicates differences in pupil teacher ratios (PTRs) in 2002.
10.31 Table 10.13 indicates a much narrower range o f PTRs than would be expected
from national distributional norms which allow a multiplier o f up to 1.7 in the more
remote and vulnerable areas to reflect the difficult conditions. I t points to the power o f
other factors such as revenue raising capacity and calls into question the adequacy o f the
adjustment made for those disadvantaged areas. There may be a need to review the
adequacy o f those differentials in the distributional norms to ensure the capability o f the
more vulnerable areas to deliver at least minimum standards o f service provision.
Table 10.13: Variations in Pupil Teacher Ratios in 2002
Region Sub-sector Poverty
Red River 23.9 21.7 24.2 185
North East 19.9 22.2 28.1 293
North West 17.9 20.1 25.1 437
North Central 25.1 26.5 29.1 25 8
Central Coast 27.5 26.4 21.5 20 1
Central High 28.0 27.9 28.4 293
South East 28.4 27.1 28.2 95
Mekong Delta 25.6 27.5 31.1 209
AVERAGE 24.6 24.7 27.5
Source: Source: Department of Planning and Finance, MOET
Lower Secondary Upper Secondary Index Primary
10.32 Only partial data was available in the course o f the review about total (budget plus
parent financed) non-salary expenditures on school facilities, maintenance etc. Overall,
there has been a slight improvement in non-salary expenditures. However, the data
available indicated a pattern o f their continuing to be squeezed particularly in vulnerable
and less wealthy areas. In many such areas the level o f such expenditures continues to be
negligible.
10.33 The disparities indicated would matter less if it were possible to demonstrate that
all students in the system were receiving a standard o f service o f at least an adequate
minimum standard to ensure attainment. Under this scenario, parents in wealthier areas
contribute more to attract improvements above that minimum while the minimum i s
guaranteed to all. But a number o f studies continue to show that this i s not the case and
point to a continuing need for mechanisms to ensure that minimum levels o f resource (to
cover salary and non-salary costs as well as expenditure for maintenance of facilities)
arrive at the level o f each student.
10.34 Table 10.14 shows how net enrolment rates vary in accordance with wealth. There
are signs o f some narrowing o f the gap, particularly in Primary, but there are also
indications o f a continuing relationship between participation and wealth, most
significantly in L S and even more so in US. This may partly be due to a perception b y
parents o f a low economic return on the added years o f secondary education. O f course,
there are many factors that have an impact on participation and some cannot be overcome
by educational policies alone, including difficulties with transport, language problems
and cultural barriers in minority areas, early marriage for girls, and the perceived
irrelevance o f the school curriculum. For unregistered groups, administrative barriers
12 Public Expenditure Review and Integrated Fiduciary Assessment
complicate access to education. But also the rationing o f resources in those sub-sectors
and the impact o f formal and informal charging policies appears to play a significant part
and needs to be addressed by policies.
Table 10.14: Variation in Enrolment Rates by Socio-economic Category
Primarv Lower secondarv Umer secondarv
1993 1998 2002 1993 1998 2002 1993 1998 2002
Vietnam 86.7 91.4 90.1 30.1 61.7 72.1 7.2 28.6 41.8
Poorest 72.0 81.9 84.5 12.0 33.6 53.8 1.1 4.5 17.1
Near poorest 87.0 93.2 90.3 16.6 53.0 71.3 1.6 13.3 34.1
Middle 90.8 94.6 91.9 28.8 65.5 77.6 2.6 20.7 42.6
Near richest 93.5 96.0 93.7 38.4 71.8 78.8 7.7 36.4 53.0
Richest 95.9 96.4 95.3 55.0 91.0 85.8 20.9 64.3 67.2
Kinh and Chinese 90.6 93.3 92.1 33.6 66.2 75.9 7.9 31.9 45.2
Ethnic minorities 63.8 82.2 80.0 6.6 36.5 48.0 2.1 8.1 19.3
Urban 96.6 95.5 94.1 48.5 80.3 80.8 17.3 54.5 59.2
Rural 84.8 90.6 89.2 26.3 57.9 69.9 4.7 22.6 37.7
Source: Vietnam Living Standards Survey, 2002.
10.35 In terms o f gender participation, Table 10.15 shows the gender mix o f pupils in
schools and reflects a steady improvement in the participation o f females particularly in
U S education (the proportion o f females in the school population has actually fallen since
1994 due to a decline in females in the total school age population which has affected
Primary in particular). In recent years, the gender gap in adult literacy has also narrowed.
According to data collected from the Vietnam Living Standard Surveys in 1993 and
1998, and from the Vietnam Household Living Standard Survey in 2002, the share of
literate people in the total population increased from 86.6 percent in 1993 to 92.1 percent
in 2002. The increase in the female population was greatest from 82.4 percent in 1993 to
89.3 percent in 2002.
Table 10.15: Share of Pupils by Gender
Sub-sector 1994 1998 2000 2002 2003
Male 50.48 52.75 52.35 52.50 52.70
Female 49.52 47.25 47.65 47.50 47.30
Male 50.87 53.35 52.95 54.70 52.20
Female 49.13 46.65 47.05 47.30 47.80
Male 55.17 53.15 53.25 52.60 51.70
Female 44.83 46.85 46.75 47.40 48.30
Male 59.00 56.50 56.50 55.60 55.00
Female 41.00 43.50 43.70 44.40 45.00
Primary
Lower Secondary
Upper Secondary
Higher
Source: Department of Planning and Finance, MOET
10.36 An indicator o f social equity in the management o f education sector resources i s
the incidence of benefit derived from the service between the relatively wealthy and the
poor. In Vietnam, the differentials in participation tend to offset the added support given
to the poorest sections of the community when viewing the overall incidence o f the
Education Sector Expenditure 13
benefits of educational subsidies. As Table 10.16 illustrates, however, education
expenditures benefit the poor much more in 2002 than they did in 1998, even if the
distribution i s s t i l l skewed towards the rich, especially for upper secondary education.
Table 10.16: Incidence of Benefit from Education Expenditures
Poorest Near Poor Middle Near Rich Richest Total
2002 31 23 18 15 13 100
2002 20 23 23 20 14 100
Upper Secondary 1998 4 11 17 30 38 100
2002 9 16 22 28 26 100
Primary 1998 26 25 21 16 12 100
Lower Secondary 1998 13 19 23 24 21 100
Source: VLSS 1998 and VHLSS 2002
10.37 There has been a growing recognition over the last decade that, although tuition
fees and other contributions have become an important source o f funding for the
education sector, they at the same time have become a potential barrier for the very poor.
Both the CPRGS and the Education for All Strategy identify the elimination of charging
in primary and L S as a policy objective. Based on earlier studies o f the impact o f
charging, these seem to be important commitments. What i s less clear i s how charging i s
defined for this purpose. The use o f revenue-raising measures such as the levying o f
construction fees, for example, s t i l l appears to be widespread and such revenues are
substantial in some provinces. The make up o f private costs i s shown below.
Table 10.17: Make-up of Private Costs
In percent of
Primary Tuition Contri- Text- School Extra Household
Education Fee bution Uniform Books Tools Classes Others Total Expenditure
In thousand VND per year
Poorest 4.7 41.9 17.0 27.6 26.5 7.4 4.8 130.7 1.9
Near poorest 7.5 47.2 24.9 36.4 34.6 14.1 8.8 174.3 1.9
Middle 11.5 50.3 33.0 41.3 38.6 22.6 15.4 215.0 1.8
Near richest 26.4 59.8 44.9 44.9 43.8 44.7 22.0 290.8 1.8
Richest 131.1 102.5 73.9 58.8 62.6 218.2 89.3 756.7 2.4
Vietnam 27.8 56.0 34.4 39.5 38.6 47.2 22.3 270.3 1.9
Lower In thousand VND per year In percent of
Secondary Tuition Contri- Text- School Extra Household
Education Fee bution Uniform Books Tools Classes Others Total Expenditure
Poorest 30.7 51.3 28.3 49.0 40.4 15.5 9.1 225.7 2.9
Near poorest 45.9 56.4 39.1 56.3 49.3 28.9 16.0 293.2 2.9
Middle 55.0 60.5 44.5 62.7 54.7 45.6 18.0 343.1 2.7
Near richest 70.0 68.8 60.7 70.1 63.3 89.9 31.0 457.5 2.7
Richest 180.1 103.4 100.8 90.6 79.3 425.7 89.4 1076.0 3.1
Vietnam 72.2 66.7 53.1 65.0 56.8 107.5 30.3 454.8 2.9
Source : VHLSS 2002
10.38 A conclusion from Table 10.17 above i s that the private costs o f education remain
high and a significant element o f finance. Moreover, despite the existence of fee
exemptions the overall result does not appear to be progressive. Except for the richest
quintile, which pays slightly more, the burden measured in percent o f total household
expenditure i s roughly the same for all population groups. I f anything, i t i s slightly higher
14 Public Expenditure Review and lntegrated Fiduciary Assessment
for the two poorest quintiles. I t i s also clear from Table 10.17 that the poor spend less on
extra classes than the better off. I f these classes play an important role in the quality of
learning outcomes, then children from poor families are receiving less tuition than those
children from better off families.
10.39 Exemptions from charges were introduced in the last decade to relieve those who
cannot afford to pay. The pattern o f those exemptions in practice i s indicated below.
10.40 Table 10.18 suggests that while exemptions are, indeed, skewed towards those in
the bottom quintile o f wealth, i t i s not as great as might be expected. Substantial numbers
in the bottom quintile continue to pay full charges. Furthermore, the implication o f the
table i s that substantial costs fall directly on the parent and are not the subject o f
exemption or help. One explanation for the pattern o f exemptions i s that poorer provinces
are also those that face general resource pressures and where there i s likely to be a
reluctance to give up revenue where i t i s possible to generate it. I t i s not clear to what
extent measures introduced since the most recent VHLSS survey in 2002, including
Decree 10/2002, have changed these patterns. However, they suggest the potential for
charges remaining a barrier to education in poor areas and reinforce the suggestion made
earlier o f viewing charging as a strategic variable alongside other priorities such as
service quality and participation.
Table 10.18: Tuition Fee Exemption and Reduction by Income Group in School Education
~~
Income group Percentage of children between 6 and 14 years old
I I1 I11 I V V Nationwide
1998
Partial reduction 8.9 10.7 13.4 9.9 8.6 10.4
Full exemption 11.6 6.5 3.6 3.6 0.9 5.8
Total 20.5 17.2 17 13.6 9.5 16.2
Partial reduction 2.21 1.77 1.42 1.34 0.89 1.52
Full exemption 24.16 18.2 15.02 11.81 7.39 15.2
Total 26.37 19.97 16.44 13.15 8.28 16.72
2002
Source: Household Income Survey 1998,2002
10.41 A common technique used in Vietnam for tackling specific problems including
helping the more vulnerable areas i s through National Target Programs (NTPs). The
education NTP has 7 specific targets. Achieving these targets would bring clear benefit to
the sector, especially for disadvantageous areas. There i s evidence from international
experience that as the participation gap closes, achieving the last push towards national
targets requires action focused on very specific and concrete areas o f difficulty.
10.42 Table 10.19 indicates the distribution o f targeted expenditures. The pattern i s one
o f broad alignment with the poverty index, but perhaps not to the extent that might be
expected and a review o f targeting mechanisms may be appropriate. Also, as the
availability o f resources grows and the need to be focused on closing the final gap on
policy objectives increases i t may be appropriate to consider devoting a larger share o f
total resources to a broadened range o f targeted programs. In the light o f the analysis
carried out above, possible candidates for targeted action might be:
0 A program to support provinces who propose targeted action to tackle
participation problems through specific measures.
Education Sector Expenditure 15
a A fund to recompense those provinces in poorer areas that provide fee
exemptions.
Broader access to funds for providing text books and other learning materials free
o f charge.
a
Table 10.19: Share of Budget-based National Target Programs by Region (1997-2004 )
Share Illiteracy Mount- Teacher Infra- Textbook IT
Region ation and Education and enhance- ulum and
by elimin- ainous training structure Curric- training
lower Strength- enhance- ment at Renewal school
secondary ening ment for schools computer-
ization universal- Teacher
iztion Training
Colleges
Nation-wide 100 6.5 18.5 21.7 24.9 26.0 2.2
Red River
Delta 14.8 11.8 7.4 18.2 17.1 15.8 15.4
North East 21.7 21.3 22.2 18.0 21.7 24.7 19.2
North West 7.1 8.9 7.7 6.1 6.7 7.6 6.1
North Central 15.0 9.8 12.4 12.2 15.3 20.5 12.7
Coastal
South Central 8.7 7.2 9.4 9.3 9.4 7.2 8.9
Central
Highlands 7.3 9.7 9.2 6.3 5.9 7.4 8.7
South East 8.7 10.8 8.7 11.7 10.2 3.9 10.3
Mekong
River Delta 16.8 20.5 22.9 18.2 13.9 12.8 18.7
Source: Department of Planning and Finance, MOET
Service Efficiency
10.43 Efficiency in the education sector can be considered as the relationship between
the inputs made and the educational outputs achieved in terms o f both the number o f
students served and the quality o f the service that they receive. Table 10.20 shows how
the unit costs per student have moved between 1998 and 2002.
Table 10.20: Cost Per student (Thousand VND)
Sub-sector 1998 2002
Current Prices Current 1998
Prices Prices*
Primary 408 721 620
Lower Secondary 465 609 524
Upper Secondary 739 876 753
Source: Department of Planning and Finance, MOET and Department of State Budget, MOF
*Using the GDP deflator
10.44 There i s clearly a major increase in unit costs over this period. This i s not
necessarily a reduction in efficiency. The key question i s whether the increase in unit
costs has ‘bought’ an improvement in service standards appropriate to that value increase.
16 Public Expenditure Review and InteRrated Fiduciary Assessment
A key element in this respect i s the composition o f the cost increase. It i s clear that the
increase can largely be explained b y movements in the PTR as indicated below.
10.45 I t follows from Table 10.21 that a very considerable resource has been absorbed
b y lowering the number o f pupils per teacher. Since 1998 this has been happening in all
sub-sectors. In the case o f primary i t represents a large investment and helps to explain
why more resources are not available to invest in other elements o f the service. Smaller
class sizes may have a beneficial educational impact. However, there are several other
reasons why the PTR may be low such as the mechanism o f teacher resource
management (new recruitment and labor re-arrangement not connected to budgetary
considerations). At present the system of budget allocation i s not connected to the PTR
through allocation criteria. An increase in the ratio i s recognized as necessary to free up
budget resources to help fund moves towards full-day education and increased non-
personnel expenditures and thereby contribute to improved service quality. A solution to
this issue has not been found in the short- term due to the labor management mechanisms
regulated b y different levels o f authority and the current labor practices in the sector. The
implications for service quality, effectiveness and efficiency w i l l be considered in later
sections.
Table 10.21: Pupil Teacher Ratios
Sub-sector PupiYTeacher Ratio
1994 1998 2000 2002 2004
Primary 34.9 30.4 28.0 24.6 23.1
Lower Secondary 25.9 28.7 26.3 24.7 23.5
Upper Secondary 23.3 30.5 29.6 27.5 27.5
Source: Department of Planning and Finance, MOET
10.46 In the Higher sub-sector a feature o f the unit costs there i s a considerable variation
in unit costs in different establishments and in different parts o f the country. This i s
reflected in Table 10.22.
Table 10.22: Unit Costs of Higher Institutions
Unit Cost (million VND)
Financed Financed
National 7.60 2.36 9.96
Ministry 2.25 1.96 4.21
Other Central 3.22 1.85 5.07
Provincial 4.31 0.85 5.16
Source: Department of Planning and Finance, MOET
Institutions Budget Student Total
10.47 While part o f these variations in unit costs appear explainable in terms o f different
types o f courses and content, this i s not the whole story. The role o f charging i s also
changing the shape of unit costs with the advent o f Decree 10. I t i s suggested that the
time has come to have a fundamental look at budget contributions to these institutions
and to rationalize the contribution from public funds.
10.48 More generally, geographic variations in expenditure per student have major
implications for the efficiency o f education spending. If those variations lead to an
education service that i s sub-standard in some areas then the impact o f education
Education Sector ExDenditure 17
spending i s partly negated. I t also has a social impact in terms o f equity o f access to a
service o f at least an adequate minimum standard in all parts o f the country. Earlier
studies have pointed to the considerable variation in the final level o f expenditure
reaching the student in a school or institution. The studies have shown that there i s no
single reason for this variation and instead it arises from factors applied at each level o f
the cascade o f resources from top to bottom o f the budget system and in local capacity to
raise revenue directly in support o f the education service.
10.49 Studies demonstrated that disparities come about at all levels in what i s a multi-
stage process o f resource allocation. What seems to be needed i s a mechanism for
ensuring that an individual student in a school attracts sufficient resources to ensure the
capacity to deliver an efficient and effective level o f service. Although Government’s
socialization policies encourage the raising o f revenue to finance improvements above
minimum standards there i s also recognition that the achievement o f a minimum standard
o f provision should not rely upon that revenue. Mechanism were discussed during the
2000 PER to ensure that an adequate level o f funding reaches the point o f service
delivery. That discussion w i l l be deepened in the section o f this chapter on Planning and
Implementing of Sector Priorities.
10.50 Staff costs are a dominant element o f education spending. Therefore, the
efficiency o f staff deployment and utilization are critical concerns. A feature o f education
provision in Vietnam has been the relatively low contact hours required o f a teacher.
Table 10.23: Teachers Contact Hours at Grades (hours per week)
Country Teaching Hours
Examples from High Performing
Countries:
Japan 17
Republic of Korea
France
Germany
Australia
19
22.0
19.5
21.2
Vietnam: 16.7
Red River Delta 17.9
North East 16.3
North West 15.5
North Central 16.8
Central Coast 16.4
Central Highlands 16.3
South East 16.5
Mekong Delta 16.3
Source: UNESCO 2000
10.51 To some extent low teaching hours have given teachers the opportunity to earn
money through other routes such as private tuition to top up their low public salary. With
the recent and planned increases in teacher salaries there i s perhaps now an opportunity to
re-think the relationship between teachers and the state as employer.
10.52 The question i s what does the state seek to ‘buy’ in return for improving the
remuneration level o f teachers? Some possibilities might include:
18 Public Expenditure Review and Integrated Fiduciary Assessment
0 Extending the teaching hours to facilitate, in particular, moves towards full
daykeduced shift working in schools. I t may need to be recognized that changed
working patterns required by an extended school day would squeeze out hitherto
revenue raising opportunities for teachers. This would also help reduce the
tendency of teachers to charge for extra classes - a wide-spread phenomenon in
the urban areas.
Greater commitment on the part o f teachers to go to the more remote areas. The
question arises whether part o f the resources available for pay awards might be
used to add to the current differential for such postings and increase the incentive
to accept them
Introducing some element o f pay related to performance and broadening o f the
differential upon acquisition o f new skills.
10.53 A potential difficulty seems to be that measures to improve public sector pay have
already commenced, but have not been linked to a redefinition o f what i s expected o f
public servants. Nor i s any distinction being made between the interests o f particular
sectors such as education. I t i s suggested that there i s an important opportunity to give
consideration to these issues and to ‘what the pay increases are for’. The grasping o f
these opportunities w i l l have a fundamental impact on service efficiency and the delivery
o f policy objectives in the education sector.
Service quality and effectiveness
0
0
10.54 The earlier review o f resources deployed to the education sector concluded that
the fiscal effort has increased to the point where i t i s above the Asia mean and
comparable to fast developing countries, making room for increasing attention on the
quality and effectiveness of the service provided.
10.55 As a general indicator o f the impact o f education, the literacy rate among adults in
Vietnam i s improving and compares well with other countries in the region.
1999* 2002**
Table 10.24: Rate of Literate Employees (above 15 years old)
RegionProvinceICity
Rural Urban Rural Urban
Nationwide 90.3
Red River Delta 94.50
North East 89.3
North West 73.3
North Central 91.3
Coastal South Central 90.6
Central Highlands 83.0
South East 92.1
Mekong River Delta 88.1
96.26
99.32
92.57
81.91
96.9
97.21
90.24
97.94
96.64
99.03
99.83
99.69
99.76
99.23
99.25
98.53
97.15
97.35
95.39
99.2
9 1,02
79.4
96.55
96.43
87.20
96.62
96.49
Source: Human Development Report Vietnam 2001. MOLISA - Statistics on Labor and Employment 2002.
10.56 Another way o f considering service effectiveness i s to look at the record with
regards to promotion rates from one level o f education to another.
Education Sector Exuenditure 19
Table 10.25: Promotion and Graduation Rate
School year 1994 1998 2000 2002
Promotion rate (%)
Primary 87.97 91.73 94.04 95.63
Lower secondary 93.1 89.26 91.22 93.16
Upper secondary 92.7 93.13 92.47 91.56
Graduation rate (%)
Primary 95.66 98.81 99.32 99.58
Lower secondary 94.36 91.95 95.13 96.11
Upper secondary 90.94 94.98 93.35 92.14
Sub-sector completion rate (%)
Primary 57.38 69.64 74.42 80.5 1
Upper secondary 82.95 86.01 83.16 79.65
Lower secondary 59.12 69.36 70.01 73.44
Source: Department of Planning and Finance, MOET
10.57 The rates o f class repetition and drop out at primary and L S schools are also
relatively low. However, rates in primary schools are quite high in mountainous regions,
especially in Central Highlands and Mekong River Delta. In L S schools, the highest
drop-out rate i s in Mekong Delta River (1 1 percent) and South East (9 percent)
Table 10.26: Rates of Class Repetition and Drop-out by Region
Class-completion Class repetition Drop-out
Primary Lower Primary Lower Primary Lower
secondary secondary secondary
Nationwide 95.63 93.16 1.24 0.94 3.13 5.90
Red River Delta 99.31 97.11 0.17 0.53 0.52 2.37
North East 95.29 95.14 1.43 0.62 3.29 4.24
North West 89.90 93.39 3.42 0.97 6.68 5.64
North Central 97.91 94.99 0.47 0.41 1.62 4.59
Coastal South Central 97.77 93.42 1 .oo 0.99 1.23 5.58
Central Highlands 91.10 91.11 4.28 2.26 4.61 6.62
South East 95.82 90.02 1.37 1.84 2.81 8.14
Mekong River Delta 92.69 87.96 1.09 1.03 6.22 11.02
Source: Department of Planning and Finance, MOET
10.58 The rates appear good and improving. But they are based on internal
examinations and may be driven b y incentives which reduce objectivity. Another way o f
considering the quality of educational outcomes i s though external tests o f attainment in
key subjects. Table 10.27 indicates the results o f two such tests carried out in 1998 and
2003. They measured attainment at Grade 5 in reading and mathematics. The score i s the
percentage o f pupils with sufficient attainment to progress adequately.
10.59 Unfortunately, the tests were designed differently. They are therefore not directly
comparable. However, they do indicate a deficiency in attainment and wide variation in
different parts o f the country. The implication i s that considerable numbers o f pupils are
graduating who are not ready for i t and do not have a solid platform to make productive
use o f the further education they w i l l receive. This i s both a poor use o f resources and
20 Public Expenditure Review and Integrated Fiduciary Assessment
undermines the ability o f the system to develop pupils with the necessary abilities to
sustain the economy o f Vietnam.
Table 10.27: Attainment of Mathematics and Reading Skills at Grade 5
1998 2003
Sample
Provinces Region Reading Mathematics Reading Mathematics
Score Index Score Index Score Index Score Index
Red River Hanoi 79 100 73 100 74 100 95 100
North East Yen Bai 54 68 36 49 39 53 65 68
North West Son L a 46 62 75 79
North Central Thanh Hoa 28 35 17 23 51 69 84 88
Central Coast Quang Nam 20 25 20 27 50 68 83 87
Central Highlands Gia Lai 53 70 84 88
Mekong Delta Vinh Long 24 30 16 22 47 63 77 81
10.60 A concern i s that there are no consistently and objectively based tests that are
applied within the system b y which to gauge the reality o f attainment and how i t i s
moving over time and across provinces. Given the increased focus on service
effectiveness which now seems both desirable and possible, the development o f tests to
be applied systematically and objectively through suitable institutional arrangements
appears important
10.61 The impact o f these and other factors causes the number o f years’ education
completed b y an average student in Vietnam to be below regional comparisons.
Table 10.28: Years of Study
Country Years of Study
High Performing Countries
Japan 13.7
Republic of Korea 13.1
Taiwan 12.1
Other Asian Countries
China 10.3
India 7.8
Indonesia 9.9
Malaysia 12.8
Pakistan 3.6
Sri Lanka 11.5
Thailand 13.1
Asia Mean
Vietnam
Singapore 11.1
2003 Urban 8.4
Rural 6.2
Source: UNESCO 2002, MOET
10.62 Improved outcomes from the system depend critically on the quality o f the inputs
and mix between teachers, operational expenditure in the classroom, and spending on
infrastructure and equipment. They also depend on the intensity o f the education
experience for pupils in terms o f curriculum and the length o f the educational day.
10.63 The phenomenon o f rapidly falling pupil teacher ratios has already been noted.
The ratios in Vietnam have fallen to levels that are well below those in other countries in
Education Sector Expenditure 21
the region as reflected in Table 10.29 below. Many studies have been done to gauge the
impact o f lower PTRs and classes on educational attainment. Clearly, at very high PTRs
and class sizes the ability o f the teacher and the physical infrastructure to cope i s severely
stretched and w i l l undermine educational outcomes. Conclusions about the impact below
very high levels vary, but the general conclusion i s that until very low figures are
reached, the impact on attainment i s relatively slight.
Table 10.29: International Comparisons of PTRs
Country 1994 1998 2000 2002 2004
Primary
Korea
Malaysia
Philippines
China
India
Cambodia
Vietnam
Secondary
Korea
Malaysia
Philippines
China
India
Cambodia
Vietnam
32
19
35
25
47
48
35
25
19
35
17
32
18
24
32
22
35
19
43
50
30
22
19
34
17
34
20
29
22
19
35
20
40
53
28
21
18
36
19
34
20
28
25 23
26 25
Source: UNESCO
10.64 On the one hand there i s a need in Vietnam for the PTR to reduce due to
curriculum changes which require a degree o f specialization. There i s also a need to
develop teachers to overcome shortages in remote areas. But i t i s not clear that a general
reduction in PTRs to levels now being reached represents the optimum use o f resources
in achieving improved educational quality. The Education for All Strategy envisaged a
need for primary PTRs to fall to around 26, but current levels are already below this and
the signs are that they continue to fall. The scale o f the fall may be partly due to force o f
circumstances as the school age population has fallen rapidly and teacher training
institutions continue to produce new teachers. With a primary PTR o f currently 23 and
falling, each 1 pupil per teacher on average i s equivalent to some 228 billion VND
nationally.
10.65 I t i s important that options for the use o f these resources be evaluated and a
decision taken about what i s most likely to optimize educational outcomes. In a situation
o f rapidly changing demographics there i s a need to treat management o f the PTR as a
policy variable requiring decision and active management. To do this w i l l require the
deployment o f a number of measures which might include restricting recruitment and
letting natural attrition through retirement etc take i t s course, encouraging early
retirement where appropriate, replacing non-qualified teachers where appropriate and
introducing active policies and incentives for teachers to be redeployed to where they are
most needed.
10.66 The skills, experience and motivation o f the teaching force are at least as
important to educational outcomes as the quantity o f teachers. Currently, a considerable
22 Public Exuenditure Review and Intenrated Fiduciary Assessment
number o f teachers are unqualified or have only a basic qualification. A variety o f
programs are in hand to increase the level o f qualifications, but these need to be
accelerated as a matter o f priority. Qualifications at the higher level are also relatively
low (Table 10.30).
Table 10.30: Qualification of Teachers in Higher Education
Qualifications 1998 2000 2002 2004
Teachers with Advanced Percentage with Advanced Qualifications
National Universities 47 53 56 52
Regional Universities 39 50 57 56
University 31 41 45 48
College 16 20 23 22
Comprehensive 38 35 40 39
Technology 28 37 34 29
Agriculture etc 36 38 45 42
Economic and Law 26 35 34 33
Culture and A r t s 14 20 20 26
Teacher Training 17 23 25 26
Medicine and Sports 24 49 50 54
Total 23 30 32 33
Source: Department of Planning and Finance, MOET
10.67 Teacher motivation and the impact of low salary levels have been recognized as a
problem. Table 10.31 shows teacher salaries in relation to per capita GDP. In 1998,
Vietnam was well below both the Asia mean and the multiplier for most countries.
However, this has clearly improved with the introduction o f salary increases across the
public service as a whole. I t i s too soon to say what impact this i s having on the quality o f
education provided, but the need to change expectations upon teachers has been noted
above.
10.68 In terms o f non-salary operational expenditure in the classroom reference has
already been made to their inadequacy in many parts o f the country. Only when they are
topped up with parental contributions in the wealthier areas do they reach levels
consistent with an adequate school environment. What has been lacking in the past has
been a benchmark against which to judge an appropriate level o f non-salary expenditures
in a school. This problem has now been tackled in Vietnam with the definition o f a
Fundamental School Quality Level (FSQL). While the FSQL i s primarily being used as a
guide for a donor financed project in a limited number o f schools i t has potential as a
benchmark for all. As yet i t has not been turned into unit costs per student for funding
purposes. I t i s recommended that i t should be.
10.69 The level o f spending required to optimize the achievement of education
objectives and the length of school day are closely related. Multiple shift working i s s t i l l
the norm in many parts o f the country enforcing a shortened day for many pupils whose
parents are unable or unwilling to pay for additional tuition outside normal school hours,
To tackle it, however, requires a major expansion o f the stock o f classrooms through
investment and the upgrading o f the existing stock.
Education Sector Expenditure 23
Table 10.31: Ratio of Teacher Salaries and Wages to GDP per capita
Nation Primary Lower secondary
Japan
South Korea
Singapore
Taiwan
China
India
Indonesia
Malaysia
Pakistan
Sri Lanka
Thailand
Asia mean
2.1
3.2
1.9
2.8
1.5
2.9
2.7
2.5
3.4
1.3
2.2
2.4
2.1
3.3
2.0
3.0
2.7
3.1
3.1
3.2
3.4
1.6
2.3
2.7
Vietnam
1998* 1.7 2.4
2003** 2.4 2.2
Source: UNESCO 2000
10.70 There i s clear evidence from studies carried out in Vietnam that where an
extended day i s provided i t has a significant impact on educational outcomes. Table
10.32 shows the scale o f differential based on scores in tests o f attainment and was
broadly equivalent to the benefit of one full year o f schooling for an average pupil.
Table 10.32: Comparison of Attainment in Full and Part Day Schooling
Dispersion around Average"
Reading Maths
Form o f Schooling
Part day schooling
Average
Full dav schoolinp.
98
100
107
97
100
108
Source : MOET
10.71 To achieve the move towards less shift working requires considerable investment
and the alignment o f that investment with teacher and operating cost provision. A
problem in the past has been the separation of recurrent and capital budget planning at all
levels o f Government. I t i s important that these are brought together now if the planned
move towards full day working b y teachers i s to be effective. The proposed deployment
o f MTEF techniques for planning education expenditure can and should be used as a
vehicle for this improved integration. In the course of the review attempts were made to
assess the capital stock required to support a reduction in shift working together with the
improvements needed to achieve required standards. I t seemed that this information i s not
readily available. These calculations are not straightforward, but i t i s important that they
are made and brought together with other aspects o f education policy through the MTEF
techniques being piloted.
24 Public Expenditure Review and Intenrated Fiduciary Assessment
PLANNING AND IMPLEMENTING OF SECTOR PRIORITIES
10.72 A finding o f this education expenditure review has been the opportunity that
exists due to improvements in resources available to tackle important challenges for the
service including service quality. Talung advantage o f them, however, requires active
management:
0
e
To connect strategies to budgets within a longer term framework.
For national strategies to influence all key service providers in the administration
without the need to micro manage implementation.
To develop the s k i l l s to manage actively in an environment o f change. e
Bringing strategies together with budgets in a medium term framework
10.73 The education sector i s the first pilot for introducing a Medium Term Expenditure
Framework (MTEF) integrating planning and resource deployment across all agencies
and resource provision. Encouraging progress has been made in developing an MTEF
that i s consistent with the plans and strategies described earlier. At this stage, the plans
identify the implications o f on-going activities. They do not take a view on the further
measures needed to implement strategic plans. They do, however, identify the scope
available for investing in those further plans. The projections suggest that even after
allowing for developments in hand, scope for investing in improved service exists. This i s
due to the combined impact o f rapid economic growth and potential freeing up o f
resources as a result o f falling school age population. Using the opportunity effectively to
achieve an improved quality o f service i s now a key challenge that the PER-FA seeks to
inform. The MTEF provides a tool to take the suggestions made in the PER-FA and test
the potential impact on resource requirement in the medium term.
Influencing implementation strategies at all levels
10.74 Service implementation i s largely decentralized. This i s appropriate for a service
where contact at local level i s essential. But this i s not incompatible with having a strong
national policy framework. Indeed, such a framework i s vital to managing change in a
service that i s confronted b y many challenges including rapid demographic changes.
These changes create opportunities for focusing on the quality o f the education. But
experience suggests that opportunities may not be taken without a strong connection
between national policies and local implementation.
10.75 Mechanisms to communicate national policies so that they influence local
implementation more clearly should :
0 be simple, direct and convey clear messages about national priorities;
0 discourage micro-management b y the center;
0 leave room for local decision making about the detail of implementation;
0 create appropriate incentives for active management at the local level; and
0 be suitable for monitoring that i s realistic and avoids cumbersome data collection.
Education Sector Expenditure 25
10.76 Mechanism to achieve this were discussed in the 2000 PER through the concept
o f a ‘base’ level o f resource available for each student. Local bodies would be free to
provide more, but not expected to provide less. I f they do so i t would be clear. Recent
changes in national norms make such a mechanism easier to link to resource distribution
mechanisms. The calculation o f the base would be transparent and seek to convey signals
and incentives for local managers that would help them align with national policies.
MOET would update the calculations supporting the calculation o f the per capita per
student on an annual basis (there would be several per capita to target local needs).
MOET would signal the per capita in advance of local bodies completing their budgets
and would receive data in a standard format demonstrating the impact o f local allocations
in a form that can be compared to the per capita. There would be dialogue based on this
data and the results of that dialogue would be transparent and freely available locally.
E x a m p l e o f a M e c h a n l o m l o S e t a M i n l m u m P e r c a p i t a
S t a f f
c o s t s
N o n s t a f f
c OStS
T a r g e t e d
E x p e n d i t u r e s
C o m p o n e n t & O t h e r & T e a c h i n g M a i n t e n a n c e
a i d s i m a t e r i a l s
M e t r i c
P e r s t u d e n t
b a s e d o n
c o s t i n g o f
1
b a s e d o n c o s t
n e e d e d t o k e e p
t o s t a n d a r d
c o n v e r t e d t o
p e r s t u d e n t
1
a p p r o p r i a t e t o
p a r t i c u l a r
t a r g e t
c o v e r t e d t o p e r
c a p i t a f o r
1 1 1
M I N I M U M T A R G E T P E R C A P I T A P E R S T U D E N T P E R T Y P E O F A R E A I
10.77 In order to make such a mechanism effective, improved data flows between local
bodies and MOET would be necessary, but not burdensome. They should focus on
providing evidence on the extent to which the base i s being implemented and the nature
o f any variations to enable constructive dialogue between the local bodies and MOET.
They should not preclude those variations. A simple form would provide feedback on
how resources are being deployed at local level. The financial data would accompany
information about performance and measurement o f quality o f education outcomes. The
aim o f the mechanism described i s to help focus on active decisions influencing quality
rather than treating those outcomes as a passive result o f decisions taken at a lower level
that inevitably focus on short term pressures.
Developing dynamic management skills to implement change
10.78 The scale o f opportunity and challenge that faces sector management w i l l place
new demands on s k i l l s at all levels: in MOET in translating national strategies into
realistic resource plans and influencing managers at local level to implement
improvements within the national strategic framework; and at local level in thinking more
flexibly about how resources are deployed and in developing s k i l l s to design and
implement measures o f change and improvement.
26 Public Expenditure Review and lntegrated Fiduciary Assessment
10.79 In MOET, steps are already being taken to develop the capacity to manage
resources better through MTEF based techniques. Further upgrading o f skills to manage
strategically across the ministry as a whole i s likely to be necessary. Furthermore, i t may
be necessary to review MOET’s ability to exert influence over the sector as a whole
including establishments currently managed by other ministries.
10.80 At local level, some initial training was given to a sample o f provinces and district
as a prelude to MTEF piloting. This training needs to be broadened and deepened. There
are also problems with implementation capacity in areas such as the design and
management o f capital developments that need to be addressed.
RECOMMENDATIONS
10.8 1 Past and planned increases in education sector expenditure, combined with
demographic changes in the school age population, provide an opportunity for increased
focus on service quality, equity and effectiveness. I t i s important that options for the use
o f resources freed up as a result o f falling school age populations be evaluated and
decisions taken about deployment to optimize educational outcomes.
10.82
e
e
e
e
To this end, the PER-FA makes the following recommendations:
Rather than further improvements in pupil-teacher ratios, priority should be given
to introducing full day working more widely, reducing user charging for poor and
disadvantaged children, and improving teacher qualifications and classroom
infrastructure, particularly in the poorest localities.
Conduct an overall review o f the role o f charging in the education sector. This
review should be integrated with the planning priorities being developed as part o f
the education sector MTEF.
Review the adequacy o f differentials in the distribution norms for education to
ensure the capability o f more vulnerable areas to deliver to standards o f service
provision (e.g. Fundamental School Quality Level, FSQL).
Develop mechanisms to ensure that at least a minimum level o f budget funding
reaches the point o f service delivery for each student, sufficient to deliver to
standards o f service provision (e.g. FSQL). A mechanism should be developed to
enhance MOET’ s role in overall sector financial management including
developing strategies and plans as well as setting policies and norms related to the
use o f resources across the sector as a whole, without detracting from local
flexibility in determining the details o f that implementation.
Increase budget allocation further to enable the sector to achieve further quality
and access improvement objectives. As part o f this increase, consideration should
be given to devoting a larger share o f total education resources to a broadened
range o f target programs. Possible candidates for targeted action might include a
program to support provinces which proposed targeted action to tackle
participation problems through specific measures; a fund to recompense those
provinces in poorer areas which provide fee exemptions; and a fund to support the
provision o f text books and other learning materials free o f charge.
Education Sector Expenditure 27
e Consideration should be given to linking improvements in teachers’ pay to
changes in practice and performance.
The World Bank recommends that Decree 10 should not be implemented in the
education sector, at least at primary and secondary levels; instead, consideration
should be given to delegating flexibility to heads o f schools under Decision 192
or other powers.
Priority should be given to the development o f tests o f attainment at key stages to
be applied systematically and objectively.
The development o f improved resource management s k i l l s and information
systems at every level o f the education sector should be treated as a priority.
e
e
e

11. TRANSPORT SECTOR EXPENDITURE
INTRODUCTION AND OVERVIEW
11.1 Transport infrastructure makes an important contribution to both poverty
reduction and growth in Vietnam. Investments in rural roads so far have led to all but 220
out o f 10,602 communes being connected to all-weather roads. Research has shown that
these investments have had a significant impact on reducing rural poverty, and
particularly in the poorest areas. Growth in exports has been a key driver in the country’s
growing gross domestic product. Improved efficiency and lower cost of provision o f
logistics services have been key factors in facilitating this export led growth.
11.2 Vietnam’s transport sector has made much progress in the period covered b y this
review. The quality and extent o f transport infrastructure have improved, the size of the
vehicle fleets has increased and most importantly the efficiency and cost with which
transport services are delivered has improved. Nevertheless, serious concerns remain
regarding inadequate maintenance expenditure in the sector; and care needs to be taken to
ensure that investment in new roads does not reach unsustainable levels.
11.3 This chapter takes a broad overview o f the transport sector, but focuses in
particular on the roads sub-sector which accounts for over 90 percent o f total public
expenditure in the sector. The first section provides an overview o f each o f the transport
sub-sectors and information on the status o f infrastructure, efficiency o f the sub-sector,
international comparisons and progress made in the period o f this review. The second
section addresses overall expenditure in the sector, analyses trends and assesses issues
related to public expenditure management providing some recommendations. The third
section considers future financing requirements. The final section provides a summary o f
recommendations.
SECTOR OVERVIEW AND PERFORMANCE
11.4 The transport sector has made good progress in responding to the demands o f
rapid economic growth, an increasingly export orientated economy and the need to
connect the most remote communes. The achievements particularly for the roads,
maritime and aviation sub-sectors have been impressive. Despite low investment rates
rail has also shown improved performance but inland waterways has not seen the same
level o f improvement. The following are some o f the highlights during this period.
Increasing transport demand
11.5 Between 1999 and 2003, the demand for freight transport increased at slightly
above the rate of growth GDP, almost 9 percent per annum in terms o f ton-km. Passenger
transport increased at just under 8 percent per annum over the same period. Road i s the
dominant mode, accounting for 60 percent o f tons moved but coastal shipping accounts
29
30 Public Exuenditure Review and Intenrated Fiduciarv Assessment
for 76 percent o f all ton-km due to i t s dominance in long-distance movements (Table
11.1). The annual throughput o f the sea ports has increased rapidly, almost doubling over
the last five years, from 56 million tons in 1998 to 114 million tons in 2003. Cargo
through southern ports in 2000 exceeded the forecast made two years earlier in the
National Transport Development Strategy, or VITRANSS, by 50 percent.
Table 11.1: Domestic Transportation Volume 1999-2003
Mode 1999 2003 Annual
increase
Goods 1OOOtons % Milton- % 1OOOtons % Milton- % ton-
km km km% Transport
Railways 5,033 3.6 1,397 3.1 8,133 4.2 2,664 4.2 17.5
Roads 83,354 59.5 5,701 12.7 116,759 60.8 7,684 12.2 7.7
Waterways 35,826 25.6 3,589 8.0 46,056 24.0 4,590 7.3 6.3
Coastal ways 15,910 11.4 34,240 76 21,042 11.0 47,616 75.8 8.6
Aviation 42 0 102 0.2 90 0 257 0.4 26.0
Total 140,166 100 45,028 100 192,080 100 62,810 100 8.7
Passenger 1000 % Mil % 1OOOpass. % Mil % Annual
Transport pass. pass-km pass-km increase
pass-km
%
Railways 9,252 1.4 2,721 9.3 11,564 1.4 4,069 10.2 10.6
Road 554,523 84.4 19,671 67.2 727,286 85.1 25,827 64.9 7.0
Waterways 89,745 13.7 2,702 9.2 110,255 12.9 2,781 7.0 0.7
Aviation 2,731 0.4 4,116 14.1 4,503 0.5 7,021 17.7 14.3
Coastal ways 800 0.1 84 0.3 870 0.1 70 0.2 -4.5
Total 657,050 100 29,293 100 854,477 100 39,767 100 7.9
~~ ~~~
Source: Planning and Investment Department, MOT
11.6 Rail plays a less significant role, although in terms of ton-km i t s share increased
from 3 percent to 4 percent, taking market share from both road and waterways.
Waterways has seen i t s market share fall over the period, partly because o f improved
alternatives but also because o f lack o f investment in that sub-sector. Although from a
lower market share, air transport has also seen i t s share for both passengers and goods
increase over the period with impressive growth.
Sub-sector performance
Roads
11.7 Road transport was in poor shape after the war and subsequent years o f slow
economic growth and there was a large backlog in investment requirements. The situation
has now improved with a massive increase in investment in roads.
11.8 National roads - VITRANSS stated that in September 1997 there were 15,100
k m s o f national roads of which approximately 61 percent were paved. Since then the
network has expanded to 17,300 k m s today with 84 percent being paved (Table 11.2). A
large part of the increase has been the completion o f 1,400 k m s o f the Ho Chi Minh
Highway. The condition o f the national road network has also improved with the
Transport Sector Expenditure 3 1
percentage in good condition increasing from 36.6 percent in 1999 to 44.8 percent in
2002. The 4-lane national road network has increased from 2 percent to 3.9 percent and
the 2/3-lane network has increased from 36 percent to 66 percent. Bridges are s t i l l a weak
link in the system: 30 percent of the 4,100 bridges are restricted to relatively low loads
and 20 percent are narrow.
Table 11.2: Road Length in km by Category and Pavement type
Category Paved Gravel Earth Total % Paved
National 14,441 600 2,254 17,295 83.5
Provincial 11,657 533 9,552 21,762 53.6
District 9,106 2,077 33,830 45,013 20.2
Urban 4,04 1 68 2,543 6,654 60.7
Commune 2,922 52,446 76,087 13 1,455 2.2
Total 42,167 55,744 124,268 222,179 19.0
Source: VRA 2004.
11.9 Improvements in the network have led to truck speeds increasing from 4 0 M h to
5 0 M h on average and bus speeds increasing from 50km/hr to 6 0 W h and reaching 70-
8 0 M h on some routes. Trucking costs are comparable to those in Thailand and 40
percent lower than those in the Philippines. Similarly, intercity bus fares are 25 percent
and 37 percent lower than those in Thailand and the Philippines respectively. However,
there i s evidence from recent evaluations o f National Highway 1, and surveys o f bus
operators, that savings in operating costs are not being passed on to customers in the form
o f lower prices (GRIPS Development Forum, November 2003).
11.10 Provincial and Rural roads - The number o f communes s t i l l lacking access to
district centers was reduced by more than half, from over 600 in 1999 to 220 today. This
represents only 2.1 percent o f the 10,500 communes. Over the period of this review, the
national level o f rural access has improved from 73 percent o f the population being
connected by an all weather road to 76 percent, which i s much higher than for other
countries at similar income levels (Figure 11.1 - left panel). This represents improved
access for close to an additional 2.5 million people. Given the evidence o f the strong
relationship between poverty and lack o f access (Figure 11.1 - right panel), this increased
access i s likely to have contributed to the impressive poverty reduction.
Figure 11.1: Measures of Rural Access
Access to Rural Transport: aggregate average
i Vietnam 72%
100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
c x
=IDA-eligible Countries (31 countries)
l B l e n d countries (7 countries)
U I D A - o n l y Countrles (24 countries)
U I B R D countries (8 countries)
I
Poverty and Accessiblllty excluding
Mekong delta Provinces
50% 1 0 0 04 150%
A c c e s s i b i l i t y
32 Public Expenditure Review and Integrated Fiduciary Assessment
11.11 Both government and donor programs have focused on rural access and i t i s clear
that there have been significant achievements in this regard. However, provincial roads
have been relatively under invested and remain in poor condition with about 50 percent
having been paved.
11.12 The vehicle Jleet - The number o f 4-wheel vehicles and articulated trucks grew
slightly faster than GDP, increasing from 400,000 in 1997 to 600,000 in 2002, or 7.5
vehicles per 1000 population. Trucks make-up 30 percent o f the fleet, but 80 percent o f
them are small to medium size and only 10 percent are modern articulated trucks.
Motorbikes grew at over 20 percent per annum and totaled over 11 million in 2003 or 140
per 1000 people (Table 1 1.3). If motorcycles are included, this gives Vietnam a very high
rate o f motorization for i t s income level, more than double that o f the Philippines for
example.
Table 11.3: Vehicle Fleet
Vehicle 2000 2001 2002 2003
Cars and Utility 285,000 303,627 346,218
Truck 130,800 144,526 162,552
Articulated truck 12,500 13,317 15,185
Bus 65,000 69,250 78,962
Sub total 493,300 530,720 602,9 17 675,000
Motorcycles 6,478,000 8,395,835 10,273,000 1 1,379,000
675,000
Railways
11.13 The rail network consists o f 7 lines with a total length o f 3,134 km. All lines are
single track, mostly a meter gauge, with a few standard gauge and double gauge towards
the Chinese border. There are 1,767 bridges (52,162 m), 39 tunnels (11,468 m) and 265
stations. Communication equipment i s outdated and only 40 percent o f the railway
stations are supplied with semi-automatic signals.
11.14 The quality o f the rail and bridges are poor and many lines do not meet modern
technical standards. Most o f the investment in the sector goes in strengthening bridges
and keeping the network from deteriorating further. As a result train operating speeds are
low at 40 km/h for passenger trains and 22km/h for freight trains. However, operating
speeds have improved on some sections. For example, passenger train times from Hanoi
to HCMC have decreased from 34 to 30 hours.
11.15 Railway revenues have increased at 15 percent per annum between 1999 and 2003
and i t appears that VR basically covers i t s operating costs including a contribution o f 10
percent o f i t s revenues for the maintenance o f infrastructure. Despite the poor condition
o f i t s track and traction and rolling stock, the performance o f Vietnam Railways puts i t in
the middle o f a range o f eight countries in the region. Actually, i t s freight service
productivity ranks third behind only Korea and India. Staff productivity i s somewhat
lower, placing i t 6th out o f 8.
11.16 The rolling stock operated b y Vietnam Railways i s old. Only 120 o f the 397
locomotives are high capacity (over 1000 HP) and high quality. Over half o f the
passenger coaches are more than 20 years old and 90 percent o f the freight cars were built
Transuort Sector Exuenditure 33
before 1980. Despite i t s condition, Vietnam Railway traffic i s increasing both for freight
and passengers.
Coastal ports and shipping
11.17 Vietnam has over 80 sea ports, totaling 22,000 m o f wharfs with 2.2 million sq m
o f quays and 1 million sq m o f docks. Although s t i l l lower than in more modern ports of
the region, port efficiency has increased and port costs have come down (Figure 11.2).
Port productivity improved on average from 1,800 tordm in 1995 to 2,800 t o d m in 2000
although ports in Saigon and Hai Phong reach 3,500 ton/m. According to Vinalines,
throughput on container berths ranges from 20 to 25 units per hour in Saigon port and 30
units in the new port o f Cai Lan in Quang Ninh province and general cargo throughput i s
1,500 tons/gang/day. These compare very favorably with the performance o f other ports
in the region.
Figure 11.2: 40 ft Container Transport Costs from Factory, via Nearest Port, to Japan
(US$)
- 2 0 0 4 -Average 2 0 0 4 - 2 0 0 2
Source: 121h and 141h surveys of Investment Related Cost Comparison in Major Cities and Regions in Asia, JETRO,
2002 and 2004.
11.18 Costs have fallen at both Hai Phong and Saigon ports largely due to reduced port
charges, trucking costs and improvements in the custom operations. Generally the
Northern ports are not as competitive as those in the South and offer their customers
fewer discounts. The dominating factor in logistics costs for Vietnam i s now the cost o f
the ocean freight, which i s double that faced by Malaysia and Indonesia. These high
ocean freight rates result in part from the need to transship containers in and out o f
Vietnam through hub ports. This i s mainly due to limited accessibility o f larger container
vessels to existing shallow water ports in Vietnam.
11.19 The fleet o f vessels has also expanded from a total number o f 679 and capacity o f
1.6 million Dead Weight Tons (DWT) in 2000 to 928 vessels and capacity o f 1.79
million DWT in 2003. Annually, the number o f vessels increased b y 12 percent and the
total volume increased b y 4 percent over the period.
34 Public Expenditure Review and Integrated Fiduciary Assessment
Waterways Transport
11.20 Vietnam has 41,000 km of natural waterways, of which only 8,000 km are used
commercially. Vietnam Waterways Authority (VIWA) manages about 6000 km and the
rest i s under local government management. VIWA also manages the main river ports.
Expenditure in the sector allows for routine maintenance of the navigation system, small-
scale dredging and some upgrading of river ports. There i s very little capital improvement
works except for the improvement of the two main water channels of the Mekong Delta
financed by ODA.
11.21 Despite limited investment, the waterways remain attractive for the transport of
coal, rice, sand, stone, gravel, and other high-weight, low-value goods. River boats and
barges have rapidly developed. In 1999, there were 63,600 units with a capacity of 1.7
million DWT and 197,000 passenger seats. In 2003 this had increased to 83,000 boats
with a capacity of 3.7 million DWT, 280,000 passenger seats and 3.4 million horse
power. In addition there are tens of thousands of small “country” boats and ferry boats.
However, the utilization of these boats i s very low with each vessel only operating 223
km per month or 7.4 km per day. According to VIWA much of this low utilization can be
accounted for because of the poor loading and unloading facilities along the rivers and
canals and the subsequent slow turn around of vessels.
Aviation
11.22 There i s very little direct government expenditure in the aviation sector apart from
the upgrade of some airports and runways. By January 2004, there were 22 major airports
including, 3 international airports which are managed and operated by the Civil Aviation
Administration of Vietnam (CAAV). However, the airports are s t i l l of relatively small
size and their equipment i s not fully modernized.
11.23 Vietnam i s managing and executing flights over an airspace of about 1,200,000
km2 belonging to two regions (Hanoi Flight Information Region (FIR) and Ho Chi Minh
FIR). These are two regions with the highest density of flights, playing an important role
in flight activities in Pacific-Asia, with more than 600 daily flights, including arrival and
transit flights.
11.24 Air passenger and freight traffic i s growing rapidly. Between 1999 and 2003,
annual growth of passenger traffic, freight traffic and flights landing averaged 12.5, 17.7
and 10.8 percent respectively. International traffic nearly doubled from 2.3 million
passengers in 1998 to 4.2 million in 2002, and freight increased from 60,000 tons to
110,000 tons. The number of aircraft has increased from 20 in 1999 to 34 in 2003.
Occupancy levels were hit in 2003 because of SARS but they are recovering again in
2004 (Table 11.4).
Table 11.4: Vietnam Airlines Occupancy Rates in Percent o f Seats Offered
International National Weighted Average
200 1 72 83 74.5
2002
2003
72 82
64 81
76.4
67.7
Transport Sector Expenditure 35
11.25 Revenues from air services and non-air services have increased rapidly for the
past 6 years (1998-2003), averaging growth o f 17 percent per annum. Before tax profit
grew at 20 percent per annum during the same period. Therefore significant capital has
been accumulated to rehabilitate and improve infrastructure and build modern airports.
This i s required to keep up with rapidly increasing passenger and freight traffic.
Transport safety
11.26 The most serious implication from rapid growth in the transport sector has been a
large increase in the number o f accidents from all modes, but particularly road transport.
11.27 Road Accidents - Traffic accidents increased dramatically from 1999 to 2002, but
have decreased from 2002 to 2003. This reduction results from the implementation o f
Resolution 13/2002/NQ-CP dated November 19, 2002, adopted by the Government to
improve traffic safety through education, public campaigns, driving tests, vehicle
examination, monitoring and enforcement o f traffic laws, improvement o f black spots,
etc. However, the death toll o f 11.8/10,000 motorized vehicles in 2003 remains high. In
addition, the data collection o f traffic accidents i s not yet adequate and the real number o f
casualties i s likely to be much higher. Table 11.15 shows fatalities due to road accidents
for some countries in the region.
Table 11.5: International Comparisons
Fatalities per Fatalities per
100,000 people 10,000 vehicles
Vietnam 16.01 ++ 11.76 ++
Malaysia
Indonesia
Laos
Cambodia
26.01
4.49
6.92
3.73
5.69
4.22
13.98
11.95
1 -28 These igh accident rates are primarily due to poor driver bcLlavior. In 2003,
National Traffic Safety Committee (NTSC) data show that 80 percent o f all accidents
were caused for this reason including 33 percent for speeding, 17 percent for dangerous
overtaking, 16 percent for driver fatigue and 6.5 percent for drunken driving. The
condition o f roads and vehicles were a minor cause o f accidents. A large proportion o f
accidents, 48 percent, happened on national roads, particularly those passing through
populated areas. About a quarter o f the accidents, 24 percent, occurred on urban roads.
The “urbanization” of road space in Vietnam i s a major contributory factor to high
accident rates as many people live within a few meters o f the road sides.
11.29 Accidents on other modes - On the other modes o f transport the accident rates
have not increased so significantly. For railways, inland waterways and coastal shipping
there were 920 accidents in 2003 o f which 545 resulted in death. These figures were up
by about 10 percent from 2000, implying a reduction in the rate o f accidents per
kilometer travel.
OVERALL PUBLIC EXPENDITURE IN TRANSPORT
11.30 The review o f transport sector expenditure i s based principally on data on State
Budget Expenditures for the transport sector provided b y MOF for the period 1999-2002,
36 Public Expenditure Review and lntegrated Fiduciary Assessment
including a break-down into central and local budgets and data from M O T for the period
1999-2003. Provincial fieldwork was undertaken in Vinh Long and Phu Tho which
provide a sample for expenditure at the provincial level.
Overview of Public Expenditures in the Transport Sector
11.31 According to the data provided b y MOF, total nominal public expenditure on
transport increased at almost 21 percent per annum between 1999 and 2002 reflecting the
strong commitment o f the Government to improve the transport system to support
economic growth. As a result, transport expenditures reached 3.5 percent o f GDP in
2002, versus an average o f 2 percent in the late 1990s. Public expenditures for transport
exceeded US$1.2 billion in 2002, thus matching the goal set in earlier plans. The data in
Table 11.6 show that the proportion of the State Budget going directly to the provinces
has increased significantly from 44 percent in 1999 to 56 percent in 2002 and that
recurrent expenditure while having a large increase in 2000 has dropped back close to i t s
1999 level.
Table 11.6: Overview of Transport Sector Expenditures Based on MOF Data
(Billion VND unless otherwise indicated)
1999 2000 2001 2002 Gr0wthp.a
Total Transport Public Exp. 10,616 11,375 14,985 18,721 20.8
- Total Local Transport Exp. 4,7 15 4,984 8,403 10,416 30.2
%
- Total Central Transport Exp. 5,901 6,391 6,582 8,305 12.1
Transport Exp. As % of GDP 2.7 2.6 3.1 3.5
Transport Exp. As % o f Total Public Exp. 12.5 11 12.5 13.8
Local Exp. As % of Total Transport Exp. 44.4 43.8 56.1 55.6
Total Recurrent Expenditures 723 1,319 1,404 1,331 22.6
- Total Central Recurrent Exp. 33 1 792 799 580 20.6
- Total Local Recurrent Exp. 392 527 605 75 1 24.2
Recurrent Exp. As % of Total Exp. 6.8 11.6 9.4 7.1
11.32 However, as the 2000 PER also noted, major discrepancies s t i l l exist between the
MOF data and the data presented b y MOT2. Whereas MOF data i s on a pure cash basis,
the data from MOT presents i t s commitments on works carried out for approved projects.
There remains a disconnection between budget allocation and project approval in the
transport sector. As a result, not all o f these expenditures have been financed b y the State
Budget in advance, and there have subsequently been a number o f one-off payments from
the State Budget to settle certain outstanding balances3. For the purposes o f the analysis
below, and in order to better reflect the full expenditures committed in the transport
For example, the published State Budget final accounts for 2001 give an expenditure figure for MOT of 11.4 trillion.
T h i s i s close to the figure provided by MOT for 2001 but 73 percent higher than the GFS format final account data for
central transport spending provided by MOF for the PER-IFA.
Besides the reason mentioned above, there are other accounting differences. For example, MOT final accounts
include expenditures funded by advanced budget, expenditure funded by retained revenue and expenditure of MOT for
general administration, education and health etc.
Transport Sector Expenditure 37
sector, we have used the data provided by MOT for central expenditures, although MOF
data i s s t i l l used for local expenditures due to the unavailability o f a complete data set
with MOT.
11.33 Using this hybrid data set (Table 11.7), total nominal expenditure on transport
appears to have increased at over 23 percent between 1999 and 2002. As a result,
transport expenditures reached 4.5 percent o f GDP in 2002, versus 3.2 percent in 1999.
The figures also show that expenditure on transport i s increasing relative to total public
expenditure, that local expenditures on transport are increasing as a percentage o f the
total and that recurrent expenditure has increased. However, the figures suggest that
recurrent expenditure i s not keeping pace with investment expenditure and as a result the
percentage being spent on recurrent items has fallen from 12.3 percent o f total transport
expenditure in 1999 to 9.8 percent in 2002.
Table 11.7: Overview of Transport Sector Expenditures
(Billion VND unless otherwise indicated)
1999 2000 2001 2002 Growth
p.a. %
Total Transport Public Exp. 12,795 13,381 19,996 23,910 23.2
- Total Central Transport Exp. 8,080 8,397 1 1,593 13,494 18.6
- Total Local Transport Exp. 4,715 4,984 8,403 10,416 30.2
Transport Exp. as % of GDP 3.2 3 .O 4.2 4.5
Transport Exp. as % of Total Public Exp. 15.1 13.0 16.7 17.6
Local Exp. as % of Total Transport Exp. 36.9 37.2 42.0 43.6
Total Recurrent Expenditures 1,576 1,721 2,125 2,334 14.0
- Total Central Recurrent Exp. 1,183 1,194 1,520 1,600 10.6
- Total Local Recurrent Exp. 393 527 605 734 23.2
Recurrent Exp. as % of Total Exp. 12.3 12.9 10.6 9.8
Source: Central expenditure data MOT, other expenditure data MOF
11.34 I t i s possible that the MOF data for local expenditures under-reports provincial
level expenditure. Provincial public expenditure studies were carried out in Vinh Long
and Phu Tho and i t was found that the cash expenditures reported b y MOF for each
province were only 33 percent and 40 percent respectively o f the total accrued
expenditure reported b y the provinces. Some of the discrepancies are probably due to
double counting (from ministries and targeted programs) but i t i s likely that there i s also
significant expenditure from provincial own-revenues and also some provincial
commitments on construction projects.
11.35 This assessment seems to be substantiated by the figures in the 2001-2005 Public
Investment Program (MPI figures) which put total transport expenditure at 18.1 trillion
VND and 18.5 trillion VND for 1999 and 2000 respectively. If these figures are correct
transport expenditure both locally and centrally was running at 4.5 percent o f GDP back
in 1999. Assuming the 2000 level o f expenditure increased at the average rate of 23
percent per annum, estimated in Table 11.7, this would put transport expenditure at 5.3
percent o f GDP in 2002.
38 Public Expenditure Review and Integrated Fiduciary Assessment
Financing central expenditures in the transport sector
11.36 The high levels o f expenditure in the transport sector have not been fully funded
on an annual basis b y the state budget. Table 11.8 shows that, between 1999 and 2002 the
state budget has funded about 65 percent o f total commitments in the sector through the
annual budgeting process. A further 35 percent has been approved b y the Prime Minister
but not allocated funding in the annual budget process. These outstanding commitments
totaled VND 14.4 trillion over the period o f this review. The 2000 PER also identified
significantly higher MOT expenditure figures o f 40 percent and 65 percent higher
respectively for 1997 and 1998, than was allocated by MOF. Over this period the MOT
has relied on their contractors to undertake this work on the assumption that they would
receive the additional approved funds at a later date. This has placed a significant debt
burden on the contractors and the state banks that have provided them credit.
Table 11.8: Central Level Expenditure and Funding Sources
(Billion VND - unless otherwise indicated)
1999 2000 2001 2002 Total % oftotalexp.
Total Expenditure 8,080 8,397 11,593 13,494 41,564
Total budget from MOF: 5,901 6,391 6,582 8,305 27,179 65.4
State budget 2,373 2,797 2,293 4,504 11,967 28.8
ODA 3,528 3,594 4,289 3,801 15,212 36.6
Total outstanding commitments 2,179 2,006 501 1 5,189 14,385 34.6
11.37 The outstanding debt to the contractors i s being repaid b y a series o f installments
through the State Budget. For example, in August 2004, the PM allocated VND 1.762
trillion (Decision 910/2004/QD-TTg), and VND 6 Trillion has been allocated for
payment of arrears associated with previous construction projects.
11.38 These levels o f transport expenditure are generally much higher than for other
countries in the region (Table 11.9) although many o f these countries are at more
advanced stages o f their economic development and have reached a more developed state
of transport infrastructure. Some o f these economies such as Indonesia and the
Philippines, for example, have arguably been under-investing in their transport sectors.
Moreover, when considering Vietnam’s investment in the transport sector between 1994
and 2002, the figure comes out to about 3.2-3.5 percent, very much in line with Japan’s
investment between 1964 and 1973 and China over the past ten years.
Table 11.9: International Comparison of Transport Expenditure Levels
Country Transport Investment at Annual Economic Growth Rate
% of GDP % (2000-2002)
Malaysia 1.9 4.3
Korea 1.8 6.3
Thailand 1.7 4.1
Singapore 1.3 3.1
Japan (1964-73) 3.5-3.8
Vietnam 3.4-5.2 7.5
11.39 Vietnam faces an unique challenge because o f decades o f under investment due to
the war and trade limitations. However, this level of public expenditure must be close to
Source: IMF statistics and MOF/MOT
Transport Sector Expenditure 39
i t s upper limit considering other needs o f the economy. I f a higher level o f transport
investment i s required, private capital may need to be brought in to supplement public
resources.
Construction sector debt
11.40 The process o f spending beyond the amounts earmarked in the State Budget
described above has had a significant negative impact on the construction sector in
Vietnam. The contractors have been taking loans from the state banks to finance the
works in the expectation o f future payments from MOT. Now state banks are being forced
to grant loan rollovers as in many cases the interest payments due are in excess o f
enterprises capitalization. Data collected b y PricewaterhouseCoopers (200 1) for the years
1997-2000 shows that this problem has been building up over time. The 14 MOT SOEs
that they reviewed were all insolvent with substantial debt, accumulated deficits and
inadequate assets to liquidate liabilities and restore equity.
11.41 I t i s now very difficult for these enterprises to be equitized in their present
condition because the enterprises are not even able to pay salaries and social insurance.
Recently MOT reported that their construction SOEs are indebted to the tune of VND
6,223 trillion (Nguoi Lao Dong - Laborer 20/8/04). MOT s t i l l owes the CIENCOs VND 3
trillion (over VND 1,700 billion just paid) and the provinces owe VND 3.5 trillion.
11.42 I t i s essential that MOT, together with other affected ministries and provinces, take
the necessary decisions on restructuring the construction companies under their control.
Where possible, enterprises should be immediately equitized. In other cases i t may be
necessary to write o f f bad debt and in some cases to allow the enterprises to fail. A
vibrant, professional and financially secure contracting industry w i l l be essential to meet
the rapidly increasing demand for construction.
Recurrent expenditures
11.43 Based on the hybrid data, recurrent expenditures in the transport sector increased
on average b y 14 percent per annum over the period o f the review. However, this
expenditure did not keep pace with the level o f new investment in the sector and as a
percentage of total expenditure recurrent spending fell from 12.3 percent in 1999 to 9.8
percent in 2002.
11.44 As mentioned above, i t i s clear that recurrent expenditures were not keeping pace
with capital expenditures and that i f this were to continue, new projects would deteriorate
and the investment would be wasted. According to the 2001-2005 PIP, recurrent
expenditures for 2001 and 2002 in the sector should have been 2.8 and 3.0 trillion. The
actual numbers reported by MOF and M O T for these years, as shown in Table 11.7, were
VND 2.1 and VND 2.3 trillion, although these figures include expenditures on wages and
salaries, as well as on non-wage O&M. MOT has spent more on recurrent expenditure
than provided by MOF but again this has required going into debt.
40 Public ExDenditure Review and Integrated Fiducian, Assessment
Table 11.10: Comparison of VRA and MOF Recurrent Expenditure Data
VND Billion
47 1
60 1
829
Current Central Budget Road Expenditures provided by MOF
Maintenance budget received according to VRA
Actual Expenditures according to VRA
Sources: MOF, Expenditures of Central Budget for Transport, VRA data received at meeting
with VRA on April 24, 2004.
National road maintenance
11.45 One o f the most important issues for the transport sector i s to secure adequate and
reliable financing for road maintenance for all levels o f the road network. At the national
level the Vietnam Roads Administration (VRA) i s responsible for maintenance o f
approximately 56 percent o f the national roads network but i t also has a mandate to
monitor, provide advice and control standards for the rest o f the network.
11.46 A recent study for the World Bank funded Road Network Improvement Project
using models to predict network deterioration for given maintenance strategies
highlighted the need for significant increases in maintenance funding. Figure 11.3 shows
that if expenditures remain at their “current” levels over a 10 year period the condition o f
the network w i l l substantially deteriorate with 34 percent being in poor condition
including 55 percent o f the high traffic volume network. The percentage in good
condition would fall to just 10 percent o f the network. The study calculated that just to
maintain “constant” network condition an estimated VND 1,150 would be required
annually. To reach an economically optimum solution VND 1,500 billion would be
required annually and this does not include bridges.
Figure 11.3: Comparison of Network Deterioration Given Different Budget Scenarios
70
60
- 50 c
s 1 40
,z 30
C
I
s
0
20
10
Good 1 Fair 1 Poor Good ~ Falr 1 Poor Good I Fall 1 poor
current Constant optimum
Condition related to budget cateogry
O !
Source: data from 10 year strategic plan for national road network, prepared by Louis Berger Group for VRA,
December 2003
11.47 VRA have submitted a proposal o f capital requirement for road maintenance to
the Prime Minister that calls for financing annually of VND 2,060 billion rising to VND
Transport Sector Expenditure 41
2,960 billion b y 2005 to cover all maintenance needs on national highways. For the
period 2003 - 2005, VRA estimate the following breakdown o f expenditures:
> Routine maintenance - VND 692 Billion > Periodic maintenance - VND 1065 Billion > Bridge Maintenance - VND 75 Billion > Other maintenance (traffic signs, signals etc) - VND 190 Billion > Emergency repairs - VND 40 Billion
k Total - VND 2,062 Billion
11.48 In recent years the allocation for maintenance o f national roads has increased
substantially from 592 billion VND in 2002 to VND 946 billion in 2004. However, these
figures s t i l l only represent 50 percent o f VRA’s estimates. The situation i s made worse
b y the outstanding debt that has accumulated in VRA because they have been typically
spending more than i s allocated b y borrowing or incurring arrears to contractors to be
paid in subsequent years (Table 11.10). The arrears are presently at 800 billion VND
(US$51 million) and they plan to repay about VND 150 billion in 2004.
Provincial road maintenance
11.49 The situation i s worse for the network o f roads managed b y the provinces as they
only appear to be spending 4-5 percent o f their budgets on maintenance activities (in
Vinh Long maintenance accounts for 4 percent o f total transport expenditures compared
to 5 percent in Phu Tho). They rely heavily on community contributions and community
labor to undertake maintenance o f the commune and village road network. In the more
prosperous areas this strategy i s probably sustainable because the burden o f maintenance
per head o f the population i s low. However, in the poorer more remote communities this
i s not sustainable in the long run. On the higher provincial and district roads network
generally sufficient funds are being allocated for routine maintenance activities but not
for periodic repair.
11.50 I t i s also clear from provincial fieldwork, and evidence from on-going projects,
that shortages in maintenance funding are particularly acute at the provincial level. The
provincial maintenance strategy i s to pave roads in order to prevent the need for
maintenance. In the short run this strategy w i l l reduce the maintenance burden but in the
medium and long term provinces w i l l face a deteriorating network and significant costs
associated with periodic repair. Particularly at the provincial level i t i s imperative that
funds for maintenance are increased.
Financing options for road maintenance
11.5 1 The Government’s recent World Bank-supported Road Network Improvement
Project (RNIP) w i l l provide US$110 million over the next 5-6 years for periodic repair:
roughly US$20 million per annum or about 315 billion VND. This figure w i l l diminish
over time and s t i l l requires additional resources to meet all maintenance needs.
11.52 To address the sub-sector under-funding, a special transport fee o f VND 300 per
liter o f gasoline and diesel was introduced in 1994 with the aim to “generate funds for the
regular repair and maintenance o f the transport system.” The fee was increased in 1996 to
500 VND per liter for gasoline and remained unchanged for diesel but the fuel surcharge
42 Public Expenditure Review and Integrated Fiduciary Assessment
has become part o f the general budget revenue. The Prime Minister has approved the idea
o f a road fund in principle, but the concept o f a road fund managed b y a road user
association would be against the existing State Budget Law. The MOT has been tasked
with developing a specific proposal on the operation of the road fund. However, there are
two concerns: (i) this could set a precedent for creating other extra budgetary funds and
(ii) i f revenues collected are too much this w i l l reduce pressure on M O T to justify use o f
public resources.
Table 11.11: Revenue in 2001 by Road User Charge Item
VND Billion
Fuel Surcharge 2,537
Licenselinspection fees 2,583
Toll Charges 55 1
Total 5,671
Source: MOF
11.53 Ultimately if there i s a commitment to increase funding for road maintenance
there i s no particular need to set up special road funds. However, decision makers are
struggling to determine the true needs for road maintenance because the management and
information systems in place within MOT, VRA and the provinces are inadequate for
managing a network o f over 200,000 k m s . More effective monitoring and data collection
i s required if investment and maintenance decisions are to be prioritized in a systematic
manner using modem management tools.
Local budget expenditures
11.54 There has been an increase in the proportion o f the state budget going directly to
the provinces. Based on the hybrid data the share to local transport has increased from
about 37 percent o f the state budget for transport in 1999 to about 44 percent in 2002,
reflecting the commitment to decentralize decision making in the sector and emphasize
development of rural transport. Based on the MOF-only data these figures would be 44
percent and 56 percent respectively.
11.55 These figures only relate to allocations from the state budget and provinces are
able to mobilize their own funds for the transport sector. Although this PER-FA did not
collect detailed public expenditure information from all provinces, sample provincial
public expenditure studies were undertaken in Phu Tho, Vinh Long and Hanoi. In Vinh
Long for example, the state budget allocation only represents 33 percent o f their total
expenditure on transport and in Phu Tho i t i s 40 percent. People’s contributions are
approximately 7 percent of the total for Vinh Long and 20 percent for Phu Tho.
Regional trends in transport expenditure
11.56 An analysis of expenditure trends for provincial and regional transport suggest
that although there have been efforts to re-direct expenditure towards the poorest areas
there i s s t i l l a case for further redistribution. Table 11.12 shows that although expenditure
per capita i s 84 percent higher in the poorest provinces than in the richest, this s t i l l does
not compensate for the greater burden o f roads in the poorer areas. Because o f lower
population densities, per capita road length i s 3 times higher in the poorest provinces and
expenditure per km i s two thirds that for richer provinces. There i s also a significant
Transport Sector Expenditure 43
difference in terms o f need with the poorest provinces only having 61 percent o f their
populations within access o f all weather access compared with 94 percent in the richer
provinces.
Table 11.12: Expenditure Profile for Provincial Level Roads
10 richest provinces 10 poorest provinces
Av expenditurelkm 76.2 49
(VND million)
Road length/ 1000 people (km) 1.17 3.6
Expenditure/person (VND) 85,600 157,500
Av level o f rural access (%) 94.2 61.3
11.57 Table 11.13 shows that certain regions require specific attention to improved
access, particularly the North West. The Mekong Delta i s a unique case because although
levels o f rural access are low as defined by access to roads, most people do have access to
navigable canals. However, as 18 million people live in the Mekong Delta there appears a
very good case for extending the road network to a greater proportion o f the population.
Table 11.13: Regional Differences in Roads Expenditure
Region Rural Access Per capita Expenditure per km of
(%o) Expenditure Road (VND mil.)
North East 74 89,714 56
North West 52 1 15,245 56
Red River Delta 92 63,284 146
Excluding Hanoi
North Central 78 46,900 35
South Central 78 123,500 124
North-East South 93 176,522 250
Excluding HCMC
Mekong River Delta 57 72,463 89
Central Highlands 73 72,061 55
11.58 The province-specific data also show particular provinces need additional help in
other regions such as in Central Highlands and the North East. I t has been estimated that
annual road infrastructure contributions amount to 9.8 percent of the annual poverty line
income in the Northern Uplands versus 4.6 percent in the country as a whole (Foster and
Vo, 2003).
11.59 Overall there i s a lack of a defined method for determining expenditure
requirements in provinces. The provinces have a high degree o f autonomy over the level
of funding to transport but there i s a need for additional central support for road
development in certain regions andor provinces. Without the necessary expenditure and
road management data these decisions are very difficult to make.
Sub-sector Expenditures
11.60 The share o f the road sub-sector dominates all others and increased even further
in the review period, to average 87 percent o f total expenditures during the period 1999-
2002 (Table 11.14). Given the relative share o f goods and passengers carried by roads,
this level o f expenditure seems to give disproportionate weight to roads. This in part i s
because the maritime, air and rail modes have direct revenues and are therefore less
dependent on public expenditures for their development. I t i s possible that maritime and
44 Public Expenditure Review and Integrated Fiducialy Assessment
air transport could be self sufficient and finance their own development but this i s rarely
the case for railways and many governments support them on social and environmental
grounds. These figures also show that expenditures on waterways, which are dependent
on public expenditures, have fallen well behind the expenditure levels proposed in the
PIP and the levels that would be reasonable given the volumes carried on the waterways.
Table 11.14: Public Expenditures by Transport Mode
(VND billion unless otherwise indicated)
1999 2002 Av. Share 2002 PIP Proposed
% Plan Share %
Total 12,795 23,909 20,400
Railway Transport 762 1,393 6 2,000 9.8
Roadway Transport 10,784 20,982 87 14,700 72
1,700 8.3 Maritime Transport 539 489 3
Waterway Transport 215 386 2
Airway Transport 171 184 1 2,000 9.8
Pipeline Transport 0 8 0
Others 324 467 2
Sources: Plan from PIP 2001 -2005 (Table page 39), Actual from MOF Expenditures for local Budget and MOT
figures for central budget.
11.61 For the first two years o f the plan period (2001 and 2002), the actual expenditures
do not reflect the modal allocation of the PIP. The planned allocation to the road sector
was 72 percent while the actual expenditures have remained around 87 percent. The focus
on roads expenditure has had a positive impact on performance in that sub-sector.
However, progress in railways and inland waterways has been particularly hampered b y
under investment.
Inland waterways
11.62 The inland waterway sector generates two types of revenues: the fuel surcharge
on diesel o f VND300 per liter and a small one-time registration fee (US$2 per ship). A
previous annual fee o f VND800 per DWT i s no longer collected. A navigation fee i s
collected on a small short cut canal in the Red River delta. I t only brings 1 billion VND
per year, enough for a mechanism o f fee collection and managing station. The users o f
formal river ports pay a port fee based on tonnage. This fee brings about 13 billion VND
per annum, besides o f 20 percent for state payment, the remaining 80 percent i s not
sufficient to pay the managing mechanism o f domestic inland waterways.
11.63 VIWA therefore depends on the budget for the development and maintenance of
the waterways. Annual expenditures are only 2 percent o f the state expenditures on roads.
While most waterways are natural rivers and do not require much investments, current
maintenance expenditures are estimated to cover only 65 percent o f the needs. Cubic
meters dredged on Category A waterways has in fact decreased from 1.1 million cubic
meters in 1999-2001 to 700,000 in 2002 and 820,000 in 2003.
11.64 Expenditures for waterways maintenance should be increased from the current
VND15.5 million per km. In addition most ports under both central and local
management are small and ageing. They lack equipment and access roads and this affects
the productivity of the fleet. I t i s recommended that investments are increased for
modernizing river ports and developing specialist terminals. As with coastal ports there
Transport Sector Expenditure 45
are significant opportunities to increase productivity on inland waterway ports. The
private sector should be actively encouraged to manage these operations.
Rail transport
11.65 By international standards VR i s doing well to cover i t s operating costs plus 10
percent o f i t s revenues towards maintenance o f the infrastructure. However, except for
this marginal contribution, VR relies on the State Budget for the maintenance o f i t s
infrastructure and any capital investment. Budget support for railways seems to have
peaked in 2002, with State allocation amounting to 96 percent o f VR revenues (Table
11.15).
Table 11.15: VR Revenues and Contributions from the State Budget
(VND billions unless otherwise indicated)
Central Budget for Rail Transport
Year VR Revenues Capital Recurrent Total % of Revenues
1999 927 408 352 760 82
2000 1197 423 417 840 70
200 1 1282 490 609 1099 86
2002 1450 737 650 1387 96
2003 1600 497 678 1175 73
Source: JBIC for 1999-2001 revenues, VR for 2003 revenues and budget and MoTfor budget 1999-2002
11.66 Maintenance expenditures cover only 60 percent o f the need. I f the track i s
allowed to deteriorate further, increasing speed restrictions w i l l eventually affect traffic.
Commercial speeds are already low at about 40 km/h for passenger trains and 22km/h for
freight .
11.67 The role o f rail transport in the sector has not been defined analytically. Although
the shape o f the country would seem to favor rail transport over road, the combination o f
the relative economic autonomy o f the three Focal Economic Zones, requiring little inter-
regional transport, and the competition o f coastal shipping for these movements, pose a
serious question for the role o f the railways. I t seems that the approved Railway
Development Master Plan to 2020 i s highly optimistic, particularly in the outer years.
11.68 Although the railways are in need o f increased investment and maintenance
funding, in the short term VR should prepare a longer term economically justified
maintenance and development plan which i s based on the economic demand for railway
services. Should Government support be justified or deemed warranted on social
grounds, one alternative worth considering i s the use o f Public Service Obligations
(PSOs), negotiated with the railways on the basis o f agreed costs to provide non-
commercial services.
Project implementation problems
11.69 Basic construction management in general and the project management in
particular have made considerable progress over recent years and improved to be
increasingly in line with international laws and donor’s policy. The transport sector has
successfully implemented many big projects and implementation periods and quality
46 Public Exuenditure Review and Intearated Fiduciarv Assessment
have improved. However, there are a number o f problems that should be addressed to
maximize the effectiveness o f public expenditure in the transport sector.
Project planning, design, and preparation of bidding documents
11.70 Poor project planning, design and preparation o f bidding documents often leads to
final costs exceeding estimates. When insufficient funds are made available for a project,
this adversely impacts on implementation time, and ultimately the quality o f construction.
Under-estimation o f costs also distorts project appraisal, making investments appear
more economically and socially beneficial ex ante than they turn out to be ex post.
11.71 There i s an urgent need to improve the quality o f project planning, appraisal and
preparation o f bidding documents. More care needs to be taken over the selection of
consultants with relevant experience and to support the development o f the consulting
sector in Vietnam. Projects should be carefully appraised by competent and experienced
staff before being approved.
Bidding
11.72 There are a number o f issues in the procurement o f civil works contracts that need
to be addressed:
a Survival bids - for large and publicly competitive bidding packages, usually ODA
projects, there i s a tendency for contractors to put in unprofitable bids which only
serve to offset the debt problems o f large contractors. However, cash flow i s a
problem which results in slow project implementation, poor quality management
(because contractors try to reduce quantities to limit their losses) and minimum
working conditions for the workers.
Uncompetitive procurement - several domestic project procurements are carried
out with limited procurement due to the pressure o f implementation time, leading
to uncompetitive procurement. Uncompetitive procurement can lead to high costs
and poor quality construction particularly where supervision o f civil works i s
compromised. Evidence from World Bank ex-post procurement reviews
highlights a trend towards collusion o f bidders and the artificial restriction o f
competition. State control o f prices also limits truly competitive bidding. The
benefits from competitive procurement would clearly include lower prices, but i t
would also support the emergence o f a more professional contracting industry and
an improved quality o f construction. Since 2003, M O T has directed not to provide
permission o f limited procurement for all biddings.
11.73 MOT should work with MPI to strengthen procurement. Key steps include the
removal o f state control o f prices or the use o f “floor prices”, more involvement o f
communities in planning and bid evaluation and tougher regulations to minimize contract
nomination and direct contracting.
Implementation monitoring
a
11.74 For the ODA projects consultants are competitively selected to monitor
preparation, implementation and the completion o f construction. The cost o f these
consultants typically accounts for 5-6 percent o f the civil works contract value. In
Transport Sector Expenditure 47
general, ODA projects have relatively better quality than domestically funded projects.
Domestically funded projects do not invest as much in quality supervision o f the works
and they lack sufficiently experienced people adequately to monitor progress.
11.75 Funds invested in good supervision during preparation and construction w i l l be
more effective as the quality i s secured. I t i s recommended that, particularly for
domestically funded projects, greater priority be given to consultant services.
Land acquisition/Resettlement
11.76 Resettlement i s the cause for substantial delays and cost over-runs in the transport
sector. There are a number o f recommendations to improve the planning o f this critical
issue:
0 A comprehensive plan for land acquisition - this should include compensation
policies, definitions o f affected people, the compensation rate, resettlement plans,
sources o f resettlement funds, identification of implementation agencies and
implementation schedules so that land acquisition happens quickly and efficiently.
Agreed compensation rates - the compensation rate must be realistic, fair and
reasonable and publicly available to provide clarity among affected people.
0 Mobilizing funds for resettlement - financial sources must be mobilized
appropriately for land acquisition. This i s particularly the case in big cities where
the compensation rate i s high. I t may be necessary to establish a Resettlement
Fund which could be financed from the sale o f houses and land in the new urban
zones.
A professional resettlement process - i t may be necessary to create specific
organizations focusing on land acquisition activities at the central and local level.
These organizations could replace the current temporary land acquisition
committees. The benefits would be a quicker and more professional land
acquisition service.
0
0
FUTURE EXPENDITURE REQUIREMENTS
11.77 The Vietnam Transport Strategy up to 2020 was submitted b y MOT to the Prime
Minister on December 31, 2002. I t estimates the average investment demand between
2002 and 2020 at 117.7 trillion VND (US$7.5 billion) per year, with almost 60 percent o f
the total for rail and urban transport. Such an amount would be over 6 times the transport
expenditures in 2002 and would exceed 20 percent of GDP, though some types o f roads
(e.g. for highways) are not purely funded b y state budget but also b y other resources and
mechanisms. The details are listed in Table 11.16.
11.78 The figures submitted to the Prime Minister are close to 5 times the investment
requirements proposed by VITRANSS and include additional expenditure for urban
transport, local transport, highways, and railways.
48 Public Expenditure Review and Integrated Fiduciary Assessment
Table 11.16: Investment Demand for Transport Infrastructure (VND trillion)
2002-2010 2011-2020 Total period Annual
period period 2002-2020 average
Road: 245,990 328,530 574,520 31,918
In which: Expressway 56,570 158,530 215,100 11,950
National Highway 139,420 125,000 264,420 14,690
Provincial Road 50,000 45,000 95,000 5,278
Railway: 2 18,661 393,576 612,237 34,013
In which: Express Railway 204,000 361,500 565,500 31,417
Normal Railway 14,661 32,076 46,737 2,596
Maritime 20,387 65,000 85,387 4,744
Inland Waterways 4,673 4,507 9,180 510
Civil Aviation 17,880 36,330 54,210 3,012
Urban Transport (Hanoi & HCMC) 195,886 423,595 6 19,48 1 34,416
In which: Road 129,385 22 1,448 350,833 19,491
Railway 56,501 193,147 249,648 13,869
Supporting public trans. 10,000 9,000 19,000 1,056
Rural Transport 86,500 77,850 164,350 9,131
Total 789,977 1,329,388 2,119,364 117,744
Source: Submission by MOT to Prime Minister, December 2002.
Improved forward planning
11.79 The plan submitted to the Prime Minister i s unrealistic and should be reviewed to
take account o f the likely resources that w i l l be available for implementation. It i s also
important that the program distinguishes between the investments proposed for domestic,
donor and private finance so that clear priorities emerge. There i s a clear need within
MOT for a much stronger forward planning framework which deals with capital and
maintenance expenditure requirements within a sustainable budget constraint.
11.80 One problem i s that currently investment requirements are planned separately
from maintenance requirements. To achieve better forward planning would require giving
greater authority to the key sub-sectoral agencies to use the modern road planning and
management tools at i t s disposal to develop reasonable annual and multi-year investment
and maintenance programs.
11.81 Transport has been identified as one o f the pilot sectors to develop Medium Term
Expenditure Frameworks. A move in this direction would provide a discipline to fully
account for investment and recurrent expenditure, to link expenditure to performance
indicators and to plan within realistic sector budget constraints. This w i l l both improve
public expenditure management and increase the credibility o f future transport plans.
Financing sources
11.82 I f Vietnam i s to meet i t s future investment needs i t must make full use of not only
state budgets but also other funding sources. In future years, beyond ODA, i t w i l l be
necessary to mobilize private finance from both overseas and domestically.
Infrastructure bonds
11.83 The government has been issuing bonds with a 5-10 year maturity to partly
finance off-budget expenditure. These bonds are part o f the government’s intention to
TransDort Sector Expenditure 49
raise VND 63 trillion (US$ 4 billion) b y 2010 to be used to mainly finance infrastructure
projects including the north-south Ho Chi Minh Highway, roads along the borders with
China and Cambodia, and irrigation projects in the disaster-prone central provinces. The
formal reason for keeping these bonds off-budget i s to comply with the State Budget
Law’s limit on the budget deficit including amortization o f 5 percent o f GDP (or around
2-3 percent of GDP excluding amortization as per GFS standards). However, as argued in
earlier chapters o f this PER-FA, i t i s imperative for fiscal transparency that these bonds
are brought on-budget as soon as possible.
11.84 Although these bonds are off-budget their issuance has been subjected to the
scrutiny o f the National Assembly, The interest payment on these bonds w i l l be on-
budget. In 2004 the plan i s to sell about VND 8.2 trillion o f bonds for infrastructure
development. Bonds o f 2 years and 5 years maturity have carried coupons o f around 7.7
percent and 8.4 percent respectively. On maturity, the 5-year bonds w i l l be rolled over for
another period o f 5 years. The main buyers o f government bonds have been State-owned
commercial banks (SOCBs) and insurance companies. These interest rates have been
marginally higher than the deposit rates offered b y the banks for similar maturities.
Private participation in infrastructure
11.85 The Government has emphasized the mobilization o f the private sector for road
financing for a number o f years. However, no foreign investment and little domestic
funds have flowed to the sector. The lack o f financially sustainable projects in the short
run and the overall climate for FDI are the main causes for the lack o f interest. Since
BOT (Build-Operate-Transfer) regulations were enacted, foreign investors have
considered two operations only, but none materialized. MOT has entered in a few small
domestic BOTs, contracted out to SOEs under the ministry. The contracts have only a
short duration o f less than 10 years and are not full-fledged BOTs. Domestic BOTs also
benefit from a preferential treatment and various forms o f Government financial support.
11.86 Regarding the financing o f road development, the international experience i s that
full-fledged BOTs have not been very successful. They have worked better for localized
investments such as bridges and tunnels than for roads. Many countries have attempted
the approach without closing any deals (Turkey, Romania) and for the few that managed
to get some operations going, they often failed (Hungary, Mexico) and led to dispute with
the concessionaire (Thailand).
11.87 Japan and Korea followed a different approach, using a National Highway
Corporation, financed in the early stages b y bond issues, and later on b y revenues from
the early projects. China i s using the build and transfer method, concessioning the
completed expressways for O&M and toll collection to partially recover the investment
cost. Vietnam needs to decide which model i s the most appropriate for i t s conditions.
SUMMARY AND RECOMMENDATIONS
11.88 Performance in the transport sector has had strengths, with significant
improvements in the level and quality o f road access, improved efficiency for all
transport modes and more competitive pricing. Particular improvements have been made
regarding export related competitiveness and rural and mountain accessibility. There i s
evidence that improved transportation has encouraged investment in the country and
50 Public ExDenditure Review and Intenrated Fiducian, Assessment
supported production and the movement o f goods. Nevertheless challenges remain which
w i l l need to be addressed in the coming years.
11.89 Overall investment in the transport sector has been large, particularly when
expenditures accrued through the build up o f arrears are taken into account. Including
expenditure at the provincial level, total expenditure (including commitments and related
debt) in the transport sector i s estimated to be about 5 percent o f GDP. Existing estimates
o f future expenditure needs in the transport sector could be met only b y a combination o f
public and private finance. As part o f the planned transport sector Medium-Term
Expenditure Framework (MTEF), estimates o f future expenditure requirements should be
re-assessed, based on a stronger forward planning framework which addresses capital and
recurrent expenditure requirements in line with the socio-economic strategy and within a
sustainable budget constraint.
11.90 During the period o f this review, expenditure in the Inland Waterways and
Railways has been relatively small and this has affected the performance o f these sub-
sectors. These sub-sectors appear to offer good value for money. I t i s recommended that,
as part o f MTEF development, consideration be given to shifting resources to these sub-
sectors to allow them to develop further.
11.91 The problem o f expenditure arrears in SOE debt in the transport sector demands
urgent action. Much stronger commitment controls are required in the transport sector, to
ensure that no further arrears build up. There should be an immediate retroactive
assessment o f all projects approved in the transport sector to determine that the budgets
exist sufficiently for their implementation. I t i s recommended that M O T and MOF work
together to devise a strategy to clear arrears in the transport sector and accelerate
equitization o f construction sector SOEs.
11.92 Maintenance expenditure i s failing to keep pace with investments and the current
levels o f financing w i l l lead to a gradual deterioration o f the road network over the next
10 years. Recently maintenance expenditures have increased but there i s s t i l l a
requirement for a stable and secure financing mechanism. Maintenance requirements for
local roads are particularly acute, with contributions for road maintenance often a burden
on the poor, and should be considered in any discussion on road financing. VRA should
take a more pivotal role in developing maintenance strategies and identifying
maintenance requirements for all levels o f the road network. I t should be responsible for
developing annual and multi-year work programs which can be used to convince decision
makers o f the need for higher maintenance funding. I t i s recommended that funds should
be shifted from new investment expenditure to O&M expenditure in the transport sector,
particularly for local roads. The recently formed Advisory Group on Road Management
and Finance (AGRMM) should then make recommendations to the Minister o f Transport
for longer term solutions to road maintenance and management issues.
11.93 I t i s also clear that the transport sector needs to attract significant private sector
financing to support budget expenditures, ODA and bond issues. There i s also the need to
develop other funding modalities through sales or concession o f public infrastructure
assets. Progress to date has been limited, despite clear potential. I t i s recommended that
M O T and other related agencies take action to address some o f the key issues inhibiting
private sector investment such as legal and regulatory frameworks; the general business
Transport Sector Expenditure 5 1
environment; lack o f implementation capacity and identification o f projects open to
private participation; and insufficiently developed domestic funding and financing
instruments.
11.94 Institutional change within the transport sector i s required to respond to the
rapidly increasing demands on the transport system. I t i s necessary to identify the roles
and responsibilities o f the key management, spending and business units within the
transport sector with a view to achieve full cost recovery where possible. In the line with
the management o f investments, i t i s necessary to strengthen the capacity of sub-sector
agencies at the same time as reducing the number of ministerial PMus. There i s a
requirement for a study to assess how this may be best achieved and also to determine
possible advantages o f merging MOT’S Department o f Planning and Investment with the
Department o f Finance to bring together the planning o f investment and recurrent
expenditure.

12. HEALTH SECTOR EXPENDITURE
INTRODUCTION AND OVERVIEW
12.1 Vietnam’s poverty reduction and growth have gone hand-in-hand with significant
progress in health. Health indicators have continued to improve nationally since the early
1990s, while once high levels o f fertility appear to have been brought under control.
These and other achievements are all the more impressive in view o f Vietnam’s already
relatively strong position in the early 1990s. At the same time, some problems have
persisted. These include the limited gains for some population segments, leading to
marked differences among regions as well as income and ethnic groups in levels o f
healthcare utilization and in important indicators o f health status. New health challenges
have also appeared, including HIV/AIDS and more recently SARS and avian flu, while
Vietnam faces a rapidly changing epidemiological profile, due partly to earlier successes
in controlling many important communicable diseases.
12.2 The 2000 Public Expenditure Review proposed a “menu o f reform options” for
the health sector. During the past four years, the Government has moved forward with
several o f these options. Most dramatically, i t has issued and started implementing
Decision 139, whereby the healthcare costs o f the poor are reimbursed directly b y the
Health Care Fund for the Poor (HCFP) or covered b y health insurance, resulting in an
increased level o f health insurance coverage. Despite this success, however, there has
been more limited progress on several o f the other reform options presented in the 2000
PER, including revision o f the user fee schedule that dates back to 1995, development o f
a mechanism for ensuring consistency between patterns o f province health spending and
those advocated in national health strategies, and rethinking the role o f the Government
in health.
12.3 Achievements in health, along with current and emerging concerns are noted in
the first part o f this chapter. The second part presents an overview o f health spending
during the period o f 1991-2002. The third part reviews health sector policy
developments, with a particular focus on developments since the 2000 PER. The fourth
part suggests that further policy advance requires revisiting the rationale for government
health activities that has been consistently articulated in Vietnamese health policy
documents. Attention turns in the fifth part to the reassignment o f government activities
and reallocation o f expenditures that would follow the development o f a sharpened
paradigm for government involvement in the health sector. The sixth part provides an
updated “menu of reform options” aimed at capitalizing on these changes.
ACHIEVEMENTS AND CHALLENGES
12.4 Significant gains were achieved in Vietnam’s health sector during the 1990s and
beyond. Key health status indicators continued to improve nationally, while once high
53
54 Public Expenditure Review and Integrated Fiduciary Assessment
levels o f fertility appear to have been brought under control. Vietnam has made rapid
progress in controlling vaccine-preventable diseases such as measles, diphtheria,
pertussis and tetanus; new vaccines have also been deployed, to prevent hepatitis,
Japanese encephalitis and typhoid. Polio has been eliminated, as has infant tetanus. The
country also made significant gains in controlling micro nutrient deficiency diseases, e.g.,
goiter, while total fertility fell from 3.8 during 1988-1992 to 2.67 in the period 1992-1996
and to 1.87 in the period o f 1998-2002. Vietnam’s health indicators are generally better
than would be expected for a country at i t s level of income per capita. Recent gains in l i f e
expectancy are particularly impressive-from 65.2 years in 1989 to 68.3 years in 1999
and 71.3 years in 2003-and represent an important welfare gain for the population. In
fact, b y 2003 Vietnam has reached i t s 2010 target for life expectancy and other important
health indicators. These and other achievements were all the more impressive in view o f
Vietnam’s relatively strong position in the early 1990s following earlier progress.
Table 12.1: Some Key Health Indicators for the Period of 1990-2003 and Targets for 2010
Indicator 1990- 1992- 2002 2003 2010
1991 2000 Targets -
Life Expectancy (years) 65 68 71 71,3 71
IMR per 1 .OOO live birth
U5MR per 1.000 live birth
45 37 26 21 25
62 42 35 32.8 32
MMR per 100.000 live birth 160 100 91 85 70
Source: MOH-Health Statistics Books, 2003
12.5 Significant gains have been achieved in controlling communicable and social
diseases. However, the progress has not always been sustained. For example, as recently
as 1998, diarrhea, dengue fever, malaria and respiratory tuberculosis were among the ten
leading causes o f admission to public hospitals. Meanwhile, HIV/AIDS incidence
appears to be rising as the epidemic moves beyond most vulnerable groups into the wider
population, while protein-energy malnutrition continues to be a major health problem
among children under 5. These present serious challenges to the health sector and need to
be tackled as a matter of priority.
12.6 In addition, according to data from the 2001-2002 Vietnam National Health
Survey (VNHS) and the 2002 Vietnam Demographic and Health Survey (VDHS),
substantial disparities remain in most key health indicators by region, income and
ethnicity. For example, the VDHS data indicate a four-fold range in the infant mortality
rate between the Northern Upland (40.9) and Southeastern (11.3) regions and between
those with no education (58.6) and those who had completed secondary school (13.2).4
The VNHS data indicate that large disparities by household living standards also exist
with respect to child malnutrition and many other important health indicators. Benefit-
incidence estimates prepared with the 2001-2002 VNHS data indicate that the average
person in the richest standard o f living quintile received a per capita State Budget health
As recognized in the 2002 VDHS report, caution needs to guide interpretation o f these extremely low
estimated mortality levels because they have relatively large sampling errors, especially at the regional
level. Even at the national level, the 95 percent confidence interval for the infant mortality estimate o f 18
deaths per 1,000 live births ranges between 9 and 27.
Health Sector Exuenditure 55
subsidy more than twice as high as that received b y the average person in the poorest
quintile (i.e., VND 118.9 thousand versus VND 52.4 thousand).
12.7 New health challenges have also emerged, including SARS, avian flu and
resistant strains o f some diseases, while Vietnam faces a rapidly changing
epidemiological profile, due partly to earlier successes in controlling many important
communicable diseases. For example, the share o f deaths attributable to communicable
diseases has fallen from 53 percent in 1976 to 34 percent in 1999, while the share o f
deaths attributable to non-communicable diseases, accidents, injuries and poisoning has
increased from 47 percent in 1976 to 66 percent in 1999. Since the latter tend to be more
expensive both to prevent and to treat, Vietnam’s success in controlling many
communicable diseases as a low-income country means that i t now faces the difficult task
o f having to address a wide range of communicable and non-communicable diseases with
relatively limited resources.
12.8 Although many factors account for the continued disparities in key health
indicators, including income, education, health knowledge, and geographical, linguistic
and cultural barriers to healthcare utilization, disparities in health spending are probably
also an important contributing factor (Figure 12.1). Despite the generally equitable
procedures that are used in preparing the State Budget (for example, the use of per capita
noms for allocating government health funding among provinces), the distribution o f
local-level State Budget health spending s t i l l favors the relatively wealthy provinces
(although Government health spending i s clearly more equitably distributed than GDP).
There i s little evidence o f any change during the period 1997-2002 (Figure 12.2).
Figure 12.1: Regional GDP per capita and Health Expenditure per capita as a Percentage of
National Averages - 2000
8 250.0 2
5 150.0
s 200.0
,- 6 100.0
m
I
50.0
OGDP per capita WTdal heatlh spending per capib
Source: National Health Accounts
56 Public Exvenditure Review and Integrated Fiducian, Assessment
Figure 12.2: Lorenz Distributions of Local-Level Government Health Spending (1997 and
2002) and GDP per capita (2000) among Provinces
, , , , ,
0 20 40 60 80 100
Cumulative % o f population
Source: Ministry of Finance
12.9 There are several reasons why the distribution o f local-level government health
spending remains only weakly pro-poor (i.e., more equal than the distribution o f GDP per
capita but less than proportional to population size):
a Richer provinces generally collect more tax revenue than estimated and are free to
allocate some part o f the additional revenue to the health sector during the 3 to 5-
year “stable budget” period in which estimated revenues remain fixed.
Provinces that collect higher than estimated levels o f revenue from taxes whose
revenues are shared between the central and provincial governments may receive
additional budget allocations from the central government as a reward, and some
o f this budget windfall may be allocated to local health spending.
a Richer provinces may allocate a higher share o f their budgets to health.
12.10 As Chapter 6 discusses, knowledge remains limited about the methods used b y
provinces to allocate funds internally (for example, among districts and communes). The
Budget Law o f 2002 provides considerable latitude to provinces in allocating funding to
lower levels, including those provided through the National Target Programs (discussed
below and in Chapter 14). I t i s possible that inequitable allocation o f government health
funding at the district and commune levels may effectively offset any gains in equity at
the province level.
12.1 1 Disparities in government health spending among and within provinces
notwithstanding, the wide disparities in non-governmental sources o f financing, and
particularly in household out-of-pocket spending (OPS), are the main source o f
geographical inequality in total health expenditure-public and private (see Figure 12.3).
Differentials in household OPS have substantially widened over time, due to differential
rates o f economic growth among geographical areas (including those between urban and
rural areas). The widening geographical differentials in OPS have contributed to
a
Health Sector Exuenditure 57
0 Other
0 Household Budget
, Local Budget
widening quality differentials among public health facilities (since user fees are an
important source o f financing for quality improvements). These developments could have
been largely offset through a more pro-poor allocation of the State Budget, but as Figure
12.2 indicates, this did not occur during the period 1997-2002.5 However, the
implementation o f Decision 139 from 2003 on can be expected to make the distribution
of government health spending among provinces significantly more pro-poor in the
future.
Figure 12.3: Health Expenditure by Region and Source of Funds
450
400
350
300
250
200
150
100
50
0
Central Budget i
Source: Ministry of Health
HEALTH SPENDING IN THE PERIOD 1991-2002
12.12 State Budget spending on health has grown rapidly during the period 1991-2002,
increasing from about one percent of GDP and 6 percent o f total State Budget spending
in 1991 to about 1.6 percent o f GDP and 6.5 percent o f total State Budget spending in
2002. In constant (1994) terms, per capita State Budget health spending has grown from
VND 20,000 in 1991 to about VND 64,000 in 2002. However, much o f this growth as a
share o f spending was in the early 1990s. Between 1997 and 2002, the share o f Health
spending fluctuated without a clear direction, between 6.1 and 7.1 percent o f total
expenditure.
12.13 In 2001, the levels o f both public and total health expenditure in relation to GDP
in Vietnam were similar to those in most other low-income developing countries (Table
12.3). Nevertheless, Vietnam’s public health spending as a share o f GDP in 2001 could
be seen as rather on the low side b y international standards at 1.75 percent; and more so
at i t s lower level of 1.61 percent in 2002. This was below the East Asian average (1.86
percent) and just above half the average for the middle income countries which Vietnam
w i l l soon be joining (2.98 percent).
.
The 2000 PER recommended that “govemment spending be used as a policy tool to reduce inter-provincial disparities
in private healthcare spending and access to healthcare.” Importantly, however, it did not mention the problem o f intra-
provincial disparities in govemment health spending.
58 Public Expenditure Review and Integrated Fiduciary Assessment
Table 12.2: Population, Total State Budget Spending, State Budget Spending on Health,
GDP - 1991-2002 (current price)
Total State Share of
Health in Budget
total State spending on
Health as Budget
(%I share of GDP
(%)
Total State Total State
Year (thousand spending (billion spending on
VND) (billion VND)
1991 67,242 12,081 76,707 732 0.95 6.06
1992 68,450 23,711 110,532 1,179 1.07 4.97
1993 69,645 39,063 140,258 1,950 1.39 4.99
1994 70,825 44,655 178,534 2,499 1.40 5.60
1995 7 1,996 54,589 228,892 3,185 1.39 5.83
1996 73,157 62,889 272,623 3,59 1 1.32 5.71
1997 74,037 70.749 313,623 4,329 1.38 6.12
1998 75,456 73,419 361,016 5,207 1.44 7.09
1999 76,597 84,817 399,942 5,912 1.48 6.97
2000 77,635 103,151 441,646 6,549 1.48 6.35
200 1 78,686 119,430 484,523 8,475 1.75 7.10
2002 79,727 133,877 536,099 8,616 1.61 6.44
Average
Annual 1.6% 24.4% 19.3% 25.1%
Growth
Source: MOF.
Table 12.3: Public and Total Health Expenditure in Relation to GDP among Selected Asian
Developing Countries - 2001
Population Budget GDP Budget
person) (billion VND)
GDP per capita ppp
(current) Spending as % of
GDP
Total Health Spending as
% of GDP Country
Bangladesh 1,650 1.5 3.5
Cambodia 1,899 1.7 11.8
China 4,237 2.0 5.5
India 2,567 0.9 5.1
Indonesia 3,089 0.6 2.4
Lao PDR 1,679 1.7 3.1
Malaysia 8,918 2.0 3.8
Mongolia 1,674 4.6 6.4
Nepal 1,398 1.5 5.2
Philippines 4,098 1.5 3.3
Sri Lanka 3,441 1.8 3.6
Thailand 6,620 2.1 3.7
Vietnam 2,182 1.75 5.1
Low income 2,000 1.22 4.78
Low and middle
income 3,950 2.73 5.75
Middle income 5,450 2.98 5.91
East Asia & Pacific 4,060 1.86 4.92
Source: World Development Indicators
Health Sector Expenditure 59
12.14 Recently, there has been a strong trend toward decentralization in health
spending. Health spending at local level has increased faster than spending at the center.
The share o f local health spending rose significantly since 2000 (Figure 12.4). In 2002,
the center accounted only for 25 percent o f total state budget spending for health.
Figure 12.4: Central-Local Shares of Total Government Health Expenditure - 1991-2002
p 120.0
I s 1 I
1991 1992 1993 1 9 9 4 1995 1996 1997 1998 1999 2000 2001 2002
WLocal D C e n t r a l
Source: Ministly of Finance
12.15 In general, current expenditure accounts for about 70-75 percent o f total state
budget spending on health while
capital expenditure ranges from 25 to
30 percent (Figure 12.5). The
expenditure fluctuated sharply during
1998-2001 with the share o f capital
expenditure reaching i t s peak at 32
percent in 1999. However, i t dropped
significantly in the following years,
partly as the result o f reclassifying
some expenditure for purchase o f
equipment in the National Target
Program as recurrent expenditure.
Within recurrent expenditure, the
share o f salary and other
expenditures increased while the
share o f expenditures on goods and
services fell.
economic composition o f
Figure 12.5: Economic Composition of
Government Health Expenditure - 1991-2002 (%)
I____" ____I" ___I- I I 2 O 0 T---- I
Source: MOF
12.16 The share o f preventive care in central budget spending started to rise in 1993
(when Vietnam began implementation o f Health National Target Programs), but since
1996 i t saw a downward trend. The share of preventive care expenditure o f the center i s
much larger than that o f the local level (Figures 12.6 and 12.7). These data to some extent
validate a concern that the provinces tend to focus more on curative care. However, it i s
noteworthy that part of central spending on Health NTPs i s actually implemented at local
level. Therefore, actual resources spent on preventive care at the local level would be
higher than those shown in Figure 12.7.
60 Public Expenditure Review and Integrated Fiduciary Assessment
Figure 12.6: Central Health Spending by
Type of Care - 1991-2002 (%)
Figure 12.7: Provincial Health Spending by
Type of Care - 1991-2002 (%)
31 92 93 94 €5 Q6 87 38 03 00 01 C2
I I
PI tnatrrent preuembn Y *hers
91 92 93 94 95 96 97 98 99 00 01 02
p treatment prevention ofamliy pother
wlannina
Source: MOF
EVOLUTION OF HEALTH POLICY AND POLICY INSTRUMENTS
12.17 The sector’s objectives continue to be focused on protecting the population’s
health through the effective prevention and control o f both communicable and non-
communicable diseases while preserving equitable access to quality health care
(especially for vulnerable groups, such as children, the poor and ethnic minorities).6 Most
recently, the National Strategy for People’s Health Care and Protection 2001-2010 stated
the objectives of reducing morbidity, raising life expectancy, ensuring the quality and
effectiveness of the health care system to meet the rapidly growing demands of the
population, and other health improvements. These objectives have remained unchanged
during recent years. However, the policy instruments at the disposal o f the Ministry o f
Health (MOH) and other central agencies to guide and shape activities in the health sector
to achieve these objectives have diminished significantly, due to the rapid growth o f the
private sector, the greatly increased role o f private (household out-of-pocket) financing,
even among the poor (Figure 12.8), and the continuing process o f fiscal decentralization:
12.18 These changes appear to have had little impact on the thrust o f Vietnam’s health
sector strategy, which has consistently emphasized the importance of active prevention,
public service delivery at the grass roots level, the need to mobilize the entire society in
support o f improved health care, the expansion o f health insurance coverage, the value o f
traditional medicine, the active participation o f the private sector under the leadership o f
the government, and the imperative to increase the share o f the government budget
allocated to health.
Key health policy documents include: the 1992 Vietnamese Constitution, the 1989 Law on People’s Health
Protection, resolutions of Party Congresses and Conferences (for example, the January 1993 Resolution of the Central
Committee of the Communist Party on urban issues of health care for the people and the 2001 Resolution o f the gth
National Party Congress), Government resolutions (for example, Resolution No. 37/1996/NQ-CP providing strategies
for health care during the period 1996-2000 and up to 2020), and various national health strategies and related planning
documents (for example, the National Strategy for People’s Health Care and Protection 2001-2010 and the Health
Development Plan 2001-2005). Most recently, sector objectives and solutions have been described in the Prime
Minister’s Decision No. 35/2001/QD-Ttg “The Strategy for the Care and Protection o f Public Health.”
Health Sector Expenditure 61
Figure 12.8: Household Out-of-Pocket Spending on Healthcare per capita during the
Previous 12 Months in Constant (2002) VND by Standard of Living Quintile
4 00
3 50
c1 3 0 0
E 2 0 0
5 2 5 0
1 5 0 e 1 0 0
50
0
P 00 r N e a r p o o r A v e r a g e B e t t e r - o f f R i c h
1 1 9 9 3 1 1 9 9 8 0 2 0 02
Source: 1993 and 1998: Vietnam Living Standard Surveys; 2002: Vietnam National Health Survey.
12.19 Several new policy tools have been introduced during the past 15 years, including
decentralization, user fees and fee schedules, social health insurance, National Health
Programs to address critical health problems (for example, malaria, tuberculosis and
HIV/AIDS), the framework for regulating the private sector, drug registration and quality
inspection, targeted demand-side financing, and managerial autonomy. Most recently,
Decision 139 (2002) has further advanced the development o f targeted demand-side
financing by establishing province-level Health Care Funds for the Poor (HCFP) to
finance free healthcare for the poor, while Decree 10 (2002) has created a government-
wide framework for the managerial autonomy o f revenue-raising units (including public
hospitals). However, these and other existing policy tools require further development
before they can be used effectively to advance Vietnam’s unchanged sector objectives in
the context o f a rapidly developing decentralized market economy.
Decentralization
12.20 Decentralization of health care delivery has been an ongoing process in Vietnam
for some time. This has led to an increasing share o f government health spending at the
local level (Le. at the province level and below), as indicated in Figure 12.4. Most
recently, the new State Budget Law, approved b y the National Assembly in December
2002 and implemented from January 2004, gives increased budget authority to provinces
to decide on sectoral allocation o f their budget for all sectors other than education and
science. The 2002 State Budget Law also provides greater flexibility to provinces in
allocating funding among districts and communes. The ongoing process of
decentralization in the health sector has also led to the development o f mechanisms, such
as the National Target Programs, to ensure that national health priorities are respected b y
the provinces and to help the Ministry o f Health to provide a national leadership role in
the sector.
User Fees
12.21 User fees were introduced into public hospitals in 1989, allowing these facilities
to charge fees to cover parts o f their costs and to alleviate their financial difficulties. B y
2003, according to MOH’s review o f user fee collection, user fees accounted for 43
62 Public Expenditure Review and Integrated Fiduciary Assessment
percent o f hospital revenue nationwide. However, there i s some question about the
accuracy o f recorded user fee revenue since several household surveys have reported
household expenditures at public hospitals that are several times larger than those
reported b y hospital^.^ A user fee schedule, setting a range o f fees for different types of
services in various classes o f hospitals, was established under Decree 95/CP dated July
28, 1994 and has not been revised since. * At their current levels, user fees (which are
also the basis for social health insurance reimbursements) recover only about 30 percent
o f the cost o f services (excluding capital costs). Differences in fees b y level are also too
narrow to discourage people from bypassing the referral system so that many seek care
initially from the highest level o f hospital that i s conveniently accessible. The schedule o f
user fees i s currently under review.’ Meanwhile, the old fee schedule remains in effect.
12.22 User fees have provided an important additional source of financing for public
hospitals. The additional revenue has enabled many public hospitals to improve the
quality and scope o f their services. However, the quality improvements from user fees
have not been evenly distributed. Hospitals located in relatively prosperous areas have
been able to collect substantially more user fee revenue than those located in relatively
poor areas. User fee revenue as a share of total hospital revenue i s also considerably
higher in central and provincial general hospitals than in district hospitals or in some
types o f specialized hospitals (for example, pediatric hospitals or TB hospitals). The
result i s that the quality improvements made possible b y user fees have probably
accentuated previous disparities in access to good-quality health care.
Social health insurance
12.23 Social health insurance was introduced nationally in 1993. There are two separate
social health insurance schemes, a compulsory scheme for civil servants, pensioners and
employees o f enterprises (but not including their families), and a voluntary scheme with
mostly students as members (school health insurance). In 1998, the Government
centralized control over social health insurance (which had previously been implemented
b y the provinces) while introducing coinsurance (20 percent o f all covered expenditures,
up to an annual ceiling equal to six times the annual minimum wage) to control rapidly
rising health insurance expenditure.” In 2002, management o f the social health insurance
system was transferred from the Ministry o f Health to VSS, thereby separating the
functions o f purchase and provision. Most recently (2003), a regulation (Circular
77/2003/TTLB-BYT-BTC on voluntary health insurance) was issued to guide extension
o f voluntary family coverage to large segments o f the rural population and to informal
’ For example, in the 2001-2002 VNHS, households reported spending an average of VND 91,560 at public hospitals,
while hospitals reported average user fee revenue of only VND 19,000 per capita during approximately the same
reriod.
The 2000 PER recommended “that once the poor are protected, the Govemment should consider raising hospital fees
to reduce i t s subsidies to hospitals.”
The Ministry of Health has proposed that user fees be increased by 35 percent in large cities, by 30 percent in smaller
cities, and by 25 percent in remote and mountainous areas.
lo The social health insurance schemes had initially reimbursed hospitals for inpatient care at a fixed daily rate, but they
subsequently changed to fee-for-service reimbursement when hospitals complained that the method was unfair. Costs
escalated rapidly and by 1997-1998 several provincial funds were running substantial deficits (Dong et al. 2002).
Health Sector Expenditure 63
sector employees." The Government has set a goal to have universal health insurance
coverage by 2010.
12.24 Social health insurance benefits include comprehensive inpatient and outpatient
health care, including essential drugs. Some categories o f beneficiaries (but excluding
early-retired pensioners, social assistance beneficiaries and the poor) are required to pay
for 20 percent o f covered costs up to an annual maximum (currently equal to six times the
annual minimum wage). In principle, covered services can be provided b y both public
and private providers. However, in practice, the fact that VSS reimbursements are based
on official user fees (and hence do not cover full costs) means that participation in VSS i s
not financially attractive to most private providers.
12.25 There are also inequities in managing the social health insurance system. Health
insurance funds are managed through the component mechanisms, i.e., compulsory and
voluntary health insurance. Since reimbursements are based on current user fees, there
are differences in benefits. User fees are very low in mountainous and poor areas, so
hospitals and insured patients get low reimbursements although contributions from these
areas are a fixed percentage o f salaries. Another source o f inequity i s that benefits for
outpatient care are provided to facilities on a capitation basis, with the level o f the
capitation based on the annual contributions o f the members registered in a given facility,
affecting patients' benefits.
National Target Programs
12.26 Beginning with the 1991-1995 Development Plan, the Government has used
National Target Programs (NTPs) to seek to resolve the most critical and urgent problems
in prevention and treatment o f illness.12 At first, there were five NTP goals. This number
has increased to 10 currently under a single, consolidated health NTP, including:
prevention and treatment o f malaria, TB, dengue fever, leprosy, goiter, child
malnutrition, the expanded program o f immunization, prevention and treatment o f mental
disease in the community, food safety and prevention and treatment o f HIV/AIDS'3.
Spending on the NTPs remained broadly constant as a percentage o f government
recurrent health spending during the period 1993-1995, increasing sharply in 1996, and
then declining gradually but steadily after 1997 as the equipment upgrading target has
been moved to be funded through the mainstream recurrent budget (Figure 12.9).
12.27 Prior to 1997, funding for the province-level implementation o f NTPs was
allocated b y MOH. Over the period 1997-2000, NTP funding was transferred to
provinces in the form o f an authorized allocation o f central budget. However, beginning
in 2001, funding for local-level implementation o f the NTPs has been provided to the
provinces as a targeted block grant. With decentralization ongoing in the health sector,
National Target Programs have become important tools to ensure the implementation of
l1 MOF/MOH Inter-ministerial Circular No. 77/2003/TTLT-BTC-BYT on Voluntary Health Insurance, which i s
consistent with a recommendation in the 2000 PER "that health insurance coverage be expanded, especially the
voluntary scheme."
l2 These programs are targeted both in the sense of being targeted to specific diseases and also because resources are
provided with the expectation that specified targets w i l l be achieved.
l3 These programs are now collectively referred to as the National Program to Eliminate Social Diseases, Dangerous
Epidemics and HIV/AIDS. In addition, there are other public health programs such as prevention and treatment of
xerophthalmia, clean water and sanitation, safe use o f plant protection chemicals, prevention and treatment of acute
respiratory infection, diarrhea, tobacco control, and prevention o f accidents and injury.
64 Public Expenditure Review and Integrated Fiduciary Assessment
sector objectives and national leadership o f MOH. This block granting has advantages,
allowing more flexibility at local level and the opportunity to be more responsive to local
needs. At the same time, there i s growing concern within the MOH that i t no longer has
adequate instruments o f control over the implementation o f the various NTPs at the
province level. There i s also concern that at least some provinces may be giving
insufficient attention to preventive and public health services. Unfortunately, no new
mechanisms have been developed in recent years to protect national health interests in
these areas.14 A thorough evaluation of the efficiency and effectiveness o f health NTPs i s
needed.
Figure 12.9: Government Expenditure on National Health Programs as a Percentage of
Government Recurrent Health Spending - 1993-2002
12
10
8
$ 6
4
2
0
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
Source: Ministry of Finance
Private sector policy and regulation
12.28 The private provision o f health care has been officially authorized since 1993.15
At that time, Provincial Health Bureaus were assigned responsibility for monitoring
private practitioners, including the issuance of operating certificates or licenses.
However, insufficient resources have been allocated at the rovince level for the
monitoring o f private health providers and private pharmacies.“ A new Ordinance on
Private Medical and Pharmaceutical Practice was adopted in 2003. I t supports the further
development o f the private health sector, while strengthening the regulatory role of the
MOH over the private health sector.
12.29 Currently, the private sector accounts for about 50 percent o f outpatient visits (80
percent if visits to pharmacies for self treatment are included). Many private providers are
staff o f the public health system working outside o f their normal working hours. Retired
health providers are another important category o f private health providers. In addition to
licensed providers, many current and former nurses and medical assistants also have
unlicensed (informal) medical practices, particularly in the rural areas. Private providers
l4 The 2000 PER recommended “that a mechanism be developed to ensure that provinces make their health spending
pattems consistent with national health priorities and strategies (for example, allocate more resources for preventive
services).”
l5 The Standing Committee o f the National Assembly enacted the “Ordinance on Private Medical and Pharmaceutical
Practices” (Ordinance 07/2003/PL-UBTVQHll o f Feb. 25, 2003), which assigned responsibility to the Govemment to
regulate private medical and pharmaceutical practices.
l6 For example, in Ho Chi Minh City, only five staff o f the Municipal Health Bureau are currently charged with the task
o f supervising literally thousands o f private healthcare providers.
Health Sector Expenditure 65
are beginning to compete effectively with the public health system in the delivery o f
outpatient care, despite the fact that they do not receive any public subsidy. Household
surveys indicate that the cost o f outpatient care in the private sector i s roughly equal to
the out-of-pocket cost o f comparable public care. However, private care i s often available
during more convenient hours, and private providers are reportedly flexible in their
payment policies.
12.30 There are currently only a few private hospitals in Vietnam. In the larger cities,
the private hospitals serve mainly expatriates and employees o f some multinational
companies. In many regards, the Government enjoys an effective monopoly on the
provision o f hospital care to the Vietnamese population. Current Government policies o f
subsidizing public hospitals through block grants for recurrent and capital costs makes i t
difficult for private hospitals to compete effectively with public h0spita1s.l~ The absence
o f a dynamic private hospital sector in Vietnam contrasts with the situation in many
neighboring countries (for example, Malaysia, Thailand, the Philippines and Indonesia).
The substantial barriers to entry facing private hospitals have created opportunities for
public hospitals to enter markets for better-quality and more technologically advanced
care from the rapidly growing segment o f the population who are willing and able to pay
for such care (see discussion o f Decree 10 below).
Pharmaceutical policy and regulation
12.31 As recently as 10-15 years ago, pharmaceuticals were in short supply in Vietnam,
and the quality o f domestically produced pharmaceuticals was not assured. Opening up
retail trade in pharmaceuticals to the private sector and careful regulation o f the quality o f
both domestically manufactured and imported pharmaceuticals have resulted in the
widespread availability o f good quality pharmaceuticals throughout the health sector.'*
However, several new problems have arisen in pharmaceutical markets, including the
following:
0 The prices o f imported branded drugs, especially those without generic
substitutes, are not effectively controlled, and there have been frequent complaints
lately o f escalating prices.
The public s t i l l has a largely irrational preference for imported products over
locally produced products (for example, preferences based on the quality o f
packaging or influenced by marketing).
Retail pharmacies sell many prescription drugs directly to consumers without a
prescription, contributing to excessive use o f drugs, antibiotic resistance and other
health risks.
0
0
l7 Public hospitals also have important advantages in being located on prime govemment land for which they pay no
rent, in receiving foreign assistance, and in paying relatively low civil service salaries. In contrast, private hospitals
would have to charge fees that would cover their full costs.
l8 There are currently about 200 domestic producers o f pharmaceuticals, 41 of which have been confirmed to comply
with ASEAN Good Manufacturing Practices standards (Rietveld 2004). Imported drugs currently account for about 30
percent o f drug consumption by volume and perhaps 70-80 percent by value.
66 Public Expenditure Review and Integrated Fiduciary Assessment
e Private health providers sell expensive branded drugs to their patients instead o f
inexpensive generic substitutes, despite existing regulations that prohibit private
providers from dispensing drugs to their patients
12.32 These problems have reduced access o f the poor to needed pharmaceuticals and
have increased the cost o f healthcare for many more people. They have also contributed
to antibiotic resistance and other serious health risks.
12.33 Vietnam dispenses essential drugs in all government hospitals and commune
health stations. Currently, the Essential Drug List contains 600 generic drugs (based
largely on WHO lists) as well as several traditional medicines. Hospitals and district
health centers are responsible for procuring their own essential and non-essential drugs.
VSS reimburses providers only for essential drugs obtained at government health
facilities (or from private providers under contract with VSS and from pharmacies in rare
instances in which the drugs are not currently available at a public facility), while the
uninsured have to pay for both essential and non-essential drugs as well as user fees at
government health facilities. In addition, the drugs provided through the National Target
Programs (for example, the drugs used to treat TB) are procured centrally and distributed
to clients free o f charge, mostly in commune health stations.
Targeted Demand-side Financing (Decision 139)
12.34 The introduction o f user fees was widely perceived at the time to have caused a
sharp drop in the utilization o f public health services during the early 1990s. I t was feared
that user fees had become an effective barrier to the use o f public health services b y the
poor.” The Government responded in 1995 by exempting the poor from having to pay
user fees. However, i t proved to be operationally difficult to finance free health care for
the poor through the normal budget process. Fee exemptions were rarely granted at all.
The Government responded in 1999 b y establishing a program o f Free Health Cards for
the Poor to finance health care for the poorest four million Vietnamese. This program,
which purchased health insurance cards for the very poor from VSS at a cost o f VND
30,000 per beneficiary, was the first nationwide targeted demand-side financing
scheme.20 Unfortunately, i t relied heavily on provinces for funding, and many o f the
poorer provinces could not afford to purchase health insurance cards for the poor. B y
2002, fewer than two million o f the poor were covered under this scheme (as compared to
the original target of four million). Another problem was that the level o f funding
provided per beneficiary (VND 30,000) proved to be inadequate.
12.35 The Government responded in 2002 b y issuing Decision 139, creating Health
Care Funds for the Poor (HCFP) in each province.21 The HCFPs are authorized either to
purchase health insurance cards from VSS at a cost o f VND 50,000 per beneficiary or to
l9 Subsequently, i t appears that the decline in utilization may have been due instead to the declining quality of services
in public health facilities and to the rapid expansion of the private sector (including notably, the rapid expansion of
private pharmacies). The utilization of public health facilities, including by the poor, increased sharply during the
period 1993 to 1998, despite the absence during this period o f effective fee exemption policies for the poor (Bhushan et
al. 2001).
2o T h i s scheme built on several demand-side financing pilots in several provinces, for example, in Hanoi and Hai Phong
provinces (Dong et al. 2002).
21 T h i s important policy measure i s consistent with a recommendation in the 2000 PER “that a national strategy be
established for assisting the poor with their healthcare costs.” I t i s also consistent with the recommendations o f several
Govemment-sponsored evaluations of the earlier Health Care Funds for the Poor scheme.
Health Sector Expenditure 67
reimburse health care providers directly for care provided to Decision 139 beneficiaries.22
The remaining 25 percent i s expected to be mobilized from other sources (e.g.,
contribution of individuals and organizations, domestic and international).
12.36 In the initial year o f implementation o f Decision 139 (2003), the budget allocated
VND 520 billion o f a total o f VND 718 billion proposed for the Fund with the aim of
supporting about 14.3 million o f eligible beneficiaries. By December 2003, there have
been 3 1 provinces applying the direct reimbursement method, 24 provinces applying
health insurance card purchase for the poor and 9 provinces applying both mentioned
methods. In 2003, the Fund disbursed VND 302 billion o f VND 520 billion (58 percent),
of which VND 114 billion (38 percent o f the total) was used for purchase o f health
insurance cards and VND 188 billion (62 percent) paid health care facilities for the poor
through a direct reimbursement method. In this first year o f implementation,
disbursement was limited, mainly due to slow preparation. I t took time to identify
beneficiaries, implementation methods, and supply health insurance cards to the poor.
Moreover, beneficiaries did not become familiar with this policy due to limited
information. Poorer and mountainous provinces tend to prefer direct reimbursement
while richer, more densely populated provinces prefer to purchase health insurance
c a r d ~ . ~ ~ S t i l l , over 11 million poor people benefited from the program, about a third (3.6
million people) received health insurance cards and about two thirds (7.4 million people)
benefited from direct reimbursement. MOH w i l l encourage provinces to move gradually
to the purchase o f heath insurance cards for the poor to reach the goal o f universal heath
insurance coverage b y 2010, reducing fund managing burden at provincial health
departments, facilitating patients and health facilities as well as improving the operation
o f the Fund.
12.37 Although the preference o f many provinces i s for direct reimbursement, this
appears perhaps not to be the best option. There i s concern that many o f the provinces
that have opted for direct reimbursement have neither the capacity nor the resources to
act as effective purchasers o f health care and that efforts to develop such a capacity w i l l
wastefully duplicate existing VSS province-level capacity.24
12.38 Decision 139 has already significantly increased the flow o f governmental health
funding to the poor and to predominantly poor areas o f the country (for example, to
mountainous regions and to disadvantaged communes). However, a number o f problems
remain, including provinces’ preference for directly reimbursing providers instead o f
purchasing health insurance cards (as discussed above), s t i l l inadequate funding provided
per Decision 139 beneficiary, and the absence o f additional financial support to facilities
to cover the unreimbursed share o f the cost o f providing services to Decision 139
beneficiaries. As a result, the local burden (Le. both the burden o f financing the non-
22 Under a recent decision of the Minister of Health (Decision 2756/2003/QD-BYT of Jul. 11, 2003), the value of
health insurance cards for the poor was increased from VND 50,000 per beneficiary to VND 75,000. Under this same
decision, free health insurance coverage was also extended to all children under six years of age.
23 Provinces behave as i f they want to conserve their HCFP funds, even though MOH implementation guidelines
stipulate that any unspent funds must be retained within the HCFP for use in subsequent years, with the possibility that
subsequent levels of central funding may be reduced by the amount of any unspent funds.
24 There i s also a question o f whether the provinces can be expected to operate a direct reimbursement scheme
effectively, given the previous experience with the social health insurance system when it was managed by the
provinces, as discussed above.
68 Public Expenditure Review and Integrated Fiduciary Assessment
government funded 25 percent o f HCFP resources and that o f financing the unreimbursed
share o f the cost o f providing services to Decision 139 beneficiaries) falls
disproportionately on poor provinces and districts with relatively large numbers o f
Decision 139 beneficiaries (Le., the per capita Decision 139 financing burden for
provinces i s directly related to the percentage o f Decision 139 beneficiaries in their
population).
12.39 In the absence o f adequate funding, there i s also concern that the poor (especially
in the poorer provinces) w i l l be required to pay informal fees or w i l l receive sub-standard
care. This may lead to lower referral rates for the poor to central facilities. Mechanisms
are not yet in place to monitor the extent and quality of care received by Decision 139
beneficiaries. In addition, there may be a need for stronger demand-side measures to
ensure that the poor take full advantage o f Decision 139 (for example, effective social
marketing to inform eligible beneficiaries about the program. The government i s
considering ways o f rei,mbursing the opportunity costs o f Decision 139 patients such as
the cost of healthcare-related transportation and food.
Managerial autonomy (Decree 10)
12.40 Government revenue-producing units in all sectors have been granted substantial
managerial autonomy b y Decree 10 (2002). Given i t s high degree o f sensitivity and i t s
limited revenue resources (official hospital fees cover only partial costs), the Health
sector has started implementation o f Decree 10 later than other sectors. Within the health
sector, Decree 10 i s currently being implemented in some hospitals at the central,
provincial and district levels. However, the expectation i s that i t w i l l be extended
eventually to all public hospitals, including central hospitals. Decree 10 administrative
units receive recurrent budget funding in the form o f a block grant that i s fixed during a
three-year period. Decree 10 managers are permitted wide latitude in managing their
units, and cost savings can be reinvested or passed along to staff in the form o f increased
remuneration (within prescribed limits). Decree 10 units also have more autonomy in
setting fees for on-demand and non-core services (such as parking and laundry fees in
hospitals) and are also permitted to raise additional capital b y borrowing from banks or
from the Development Assistance Fund to expand the scope and to improve the quality o f
their services. They can also hire additional staff on a contractual basis.
12.41 Decree 10 provides strong incentives to public hospitals to expand services for
paying clients as well as to reduce their costs (Box 12.1). However, Decree 10 hospitals
also continue to receive government subsidies in terms o f capital budget and partial
recurrent budget for those that cannot fully cover their costs (60-80 percent o f which i s to
cover salary costs, 15-30 percent for operations and 5-10 percent for maintenance).
Although these are presumably intended to support key public health functions (such as
protecting the public’s health and ensuring health equity), there i s a concern that they w i l l
be used instead to enhance services for higher-paying clients. Currently, there are no
effective mechanisms in place to monitor the quality o f care received b y different
categories o f hospital patients (for example, the insured, the poor, other categories o f
exempted patients, and fee-paying patients). There i s a risk that the quality o f services for
non-fee-paying patients w i l l deteriorate as supply-subsidies (block grants to public
Health Sector Exuenditure 69
I
providers) intended to support their care are diverted to the care o f fee-paying patients.25
Another risk i s that non-fee-paying patients with relatively costly health problems w i l l be
unnecessarily referred to higher-level hospitals. Whether or not such behavior exists
currently, the incentives provided by Decree 10 would certainly reinforce such behavior.
Another concern i s whether i t i s sensible economic policy for the Government to
encourage public hospitals to enter markets that are usually served b y private hospitals in
other countries. There i s a risk that inefficient public hospitals w i l l become a drain on the
economy i f they are permitted to function essentially as state-owned enterprises, with
privileged access to public subsidies, subsidized public sources o f credit, and other
uncompetitive advantages.
I Box 12.1: Early Experiences with Decree 10 in Public Hospitals
Although Decree 10 has been implemented for only a little over one year in province-level hospitals
and some other fee-collecting province-level facilities, there i s already some information about its effects
on hospital operations (though not yet on its impact on patients). For example, Xanh Pon Municipal
Hospital in Hanoi (460 beds) has been implementing Decree 10 since late 2002. Although the level of its
state budget funding has been practically unchanged since 2001, Xanh Pon’s revenues have increased by
27 billion VND (86 percent) in only two years. Although some o f the increase (about 4 billion VND) i s
accounted for b y increases in revenue from normal user fees and health insurance reimbursements, the rest
i s due to a 350 percent increase in revenue from “special services,” which accounted for more than one-
half of total hospital revenue during 2003.
These special services, which are available on request, are of much higher quality than the hospital’s
normal services. With the approval of the Municipal Government, the hospital can set the fees for these
services much higher than those currently authorized for ordinary user fees. The investments necessary to
prepare the hospital to provide such special services have reportedly been mobilized from a variety o f
sources, including: hospital staff and other individuals, the hospital’s own revenues, ODA, and the
investment funds provided by the state budget. In 2003, for example, the hospital spent approximately 2.9
billion VND on equipment financed by the following sources: ODA (74 percent), hospital revenues (24
percent), and the state budget (2 percent). During the same year, the hospital provided fee exemptions and
reductions for children under 6 and the poor without health cards with a value of about 3.5 billion VND,
about the same level as in the year 2000 when total hospital revenues were only 37 percent of their 2003
level.
The Polyclinic o f Binh Duong province (600 beds) has also seen its revenues increase dramatically
during its initial year of Decree 10 implementation. Actual hospital revenue during 2003 was almost 25
percent higher than targeted revenue, with all of the increase obtained through increased fee revenue and
health insurance reimbursements (up 35 percent, compared to targeted levels). Inpatient admissions
increased by 015 percent compared to 2002, while average length of stay decreased by 6 percent and the
referral rate increased from 5.35 to 6.23 percent. During 2003, the average hospital salary increased by
about 485,000 VND per month (about US$31). The report from the hospital’s Director indicates that the
hospital i s operating more efficiently as the result o f Decree 10. However, i t complains o f the unchanged
level of user fees since 1995, o f the low subsidies received from the state budget (only 17 million VND
per bed, compared to 27-28 million per bed received by comparable hospitals in neighboring provinces),
and o f shortages in MOH-mandated equipment
Source: PER-IFA field visits.
25 Some hospitals claim that they “cross-subsidize” services for non-fee-paying patients by directing some percentage
of the fees collected to help pay for such services. However, the critical issue i s the share o f public subsidies (not
revenue) captured by different categories of patients.
70 Public Exuenditure Review and Integrated Fiduciaw Assessment
NEED FOR A SHARPENED PARADIGM
12.42 There i s an emerging consensus that a more focused approach i s needed to guide
Government actions in the health sector.26 This sharpened paradigm would concentrate
increasingly on two health objectives that have been clearly and consistently enunciated
in official Vietnamese health policy documents for many years: (1) protecting the
public’s health and (2) ensuring health equity.
12.43 Protecting the public’s health entails seeing to i t that the population i s effectively
protected against communicable diseases and i s well informed about health risks,
available programs and options, and how to make effective use o f health services and
inputs (including pharmaceuticals). This key health function also involves ensuring that
health services and inputs are both affordable and o f uniformly good quality.
12.44 Although Vietnamese are better informed about health matters than the
populations o f many low-income countries, data from the 2001-2002 VNHS reveal that
there are s t i l l some important gaps in health knowledge (for example, about
immunization, correct infant feeding practices, the benefits o f vitamin A and about
effective ways to prevent non-communicable diseases), and that these information gaps
are often related to socio-economic characteristics such as standard of living, education
and ethnicity. Surprisingly, in light o f the Government’s stated goal to achieve universal
health insurance coverage b y 2010,58 percent o f uninsured respondents in the 2001-2002
VNHS reported that they did not know what health insurance is, while another 25 percent
indicated that they did not know where to purchase it. Additionally, poor people typically
are unaware o f to how to make effective use o f Decision 139 and other health care
options.
12.45 The Vietnamese public also s t i l l prefers imported and branded drugs over locally
produced generic products, despite the fact that the latter are now considered to be o f
generally good quality. Retail pharmacies sell many drugs (and especially branded,
imported drugs) directly to consumers without a prescription, contributing to excessive
use of drugs, antibiotic resistance and other health risks. Despite regulations that prohibit
private providers from dispensing drugs to their patients, many sell expensive branded
drugs directly to their patients instead o f inexpensive generic substitutes, thereby
increasing the cost o f healthcare to many more people.
12.46 Ensuring health equity involves lessening disparities in basic healthcare
utilization and health status according to such characteristics as region, education,
income, ethnicity and gender; and providing mechanisms to spread the burden o f
healthcare financing equitably among the rich and poor and between the healthy and
unhealthy. Vietnam i s currently far from achieving these objectives. As previously noted,
marked disparities continue to exist with respect to many health indicators, while a very
high share of healthcare financing (about 60-65 percent) i s from a single inequitable
source, household out-of-pocket spending (OPS). At the same time, i t i s recognized that i t
26 The 2000 PER recommended that “over the longer term (next 5-10 years) the Government may want to rethink its
roles in the health sector (for example, to reduce its role in the direct provision o f curative hospital-based services and
focus instead on the provision of high-quality preventive services, on healthcare financing, especially for the poor, and
on improving regulation and monitoring of the rapidly growing private sector).”
Health Sector Expenditure 7 1
i s very difficult to prevent some disparities in access to healthcare from arising in a
market economy.
12.47 The main solution to existing health inequities i s an equitable expansion o f health
insurance coverage. For example, attaining the objective o f universal health insurance
coverage b y 2010 would ensure that the burden o f healthcare financing i s shared
equitably between the seriously sick and injured and the much larger group o f those in
good health. The recently issued Circular 77/2003 on expansion o f voluntary health
insurance coverage i s an important initial step toward reaching this objective.27 However,
linking healthcare financing to ability to pay can only be achieved i f health insurance
premiums and contributions are linked to income (as i s currently the case in VSS’s
compulsory scheme for civil servants and formal sector workers). Since i t i s difficult to
link voluntary health insurance premiums to income, equitable health financing may
ultimately require universal compulsory coverage. Even now, however, i t should be
possible to begin expanding compulsory coverage in relatively prosperous communes
with contributions linked to income.
IMPLEMENTATION
12.48 Development and implementation of a more sharply focused approach to
government’s role in health requires that first, the government action needed to achieve
these key health functions be identified clearly; second, responsibility for implementing
the required actions be assigned to the appropriate central andor sub-national
institutions; third, the resources needed to implement the required actions be carefully
estimated and secured in full; and fourth, the policy instruments needed to implement,
monitor and evaluate needed actions be operational.
Identification of the Government’s Critical Health Functions
12.49 Currently, the rationale for government action in the health sector i s not always
clear. Health policy makers and planners should take a careful look at the full range of
current government actions in health and identify those that are critical to support the
government’s two key health functions. For example, the scope o f the National Target
Programs may need to be broadened to include new diseases (for example, SARS and
avian flu) and some increasingly important non-communicable diseases and health risks,
such as injuries, tobacco and alcohol use, environmental pollution and obesity.
Additionally, government subsidies for curative care may need to be targeted more
effectively to the poor and other vulnerable groups than i s the case currently.
12.50 Such a sharpened policy and planning exercise w i l l necessarily lead to some
reallocation o f health resources at both central and local levels. For example, the
government subsidies currently provided to central hospitals may need to be re-focused
on activities that are more clearly related to protecting the public’s health (for example,
research, training and supervision o f the public health system). Government subsidies
provided to local-level hospitals (particularly those operating under Decree 10) may need
to be re-channeled through demand-side mechanisms to the poor and near poor to prevent
27 I t i s also consistent with one of the recommendations o f the 2000 PER, Le., that “health insurance coverage be
expanded, especially the voluntary scheme.”
72 Public Expenditure Review and Integrated Fiduciary Assessment
them from being diverted to care for higher-paying clients. A further option to be
analyzed and discussed i s an expanded role for private health providers in supplying
healthcare to higher-income groups.
Assigning Responsibilities for Required Government Actions
12.51 Government services should be provided at the lowest level o f government that
can do so efficiently (see Chapter 6). In deciding which level o f government should
provide a given service, several issues should be considered, including whether there are
significant economies o f scale involved in providing the service and at which level o f
government most such scale economies can be realized; whether the service has spillover
effects to other geographical regions and at which level o f government such spillover
effects are appropriately minimized (or effectively offset through targeted subsidies); and
whether the service i s related to national objectives and may therefore require direct
provision by the central government in order to minimize the risk that those objectives
w i l l not be achieved. In assigning responsibility for the provision of services to various
levels o f Government, Vietnam i s somewhat constrained b y the absence o f regional
administrative units between national and provincial levels. In some cases, the health
sector has established i t s own regional centers, for example facilities established for
specialized tertiary care, research, advanced specialty training, and specialized laboratory
diagnosis.
Resource Requirements
12.52 Once the necessary actions have been identified and assigned to an appropriate
level or levels o f government, the resources required need to be carefully estimated.
Currently, this aspect o f the planning process i s relatively weak in Vietnam. Overall
budget requests tend to be based on extrapolations o f previous budgets or notions o f an
appropriate target level o f health funding, such as 8 percent o f total recurrent expenditure
or 1.5 percent o f GDP (sometimes derived b y reference to spending levels in other
countries). Budgets for individual facilities and specific activities are often based on
norms, which are supposed to reflect actual costs but which often reflect the level o f
resources available. With the exception o f the National Target Programs, which have
clear results targets, the currently used norms also fail to identify the various purposes for
which budget support i s intended, making i t difficult to monitor whether the resources
have been used as intended and have obtained the desired results (see Box 12.2).
12.53 Some budget resources may be freed up through refocusing, reprioritization and
more careful costing o f needed actions. Still, i t i s possible that significant additional State
Budget funding for the health sector w i l l be required to perform adequately the key health
functions o f protecting the public’s health and ensuring health equity. For example,
resources allocated to ensure health equity may need to be increased substantially to
subsidize more adequate health insurance coverage for a larger number o f the poor and
near poor. The funding provided to National Target Programs may need to be increased
to cover a broader range o f communicable and non-communicable diseases.
Health Sector Expenditure 73
I Box 12.2: Budgeting for Central Hospitals
With implementation o f the Budget Law of 2002 (from January 2004), the MOH has full authority to
determine the norms it uses to allocate resources among central hospitals. The procedures i t uses are as
follows. All central hospitals are designated as (1) general hospitals, (2) specialty hospitals, or (3)
hospitals for social diseases such as tuberculosis, leprosy and mental disorders. Separate budget norms are
used for each category, Le., general hospitals (30-40 million VND per bed), specialty hospitals (25-35
million VND per bed), and hospitals for social diseases (25-30 million VND per bed). Budget allocations
to individual hospitals reflect: (1) reported expenditure in the previous year, (2) the number of staff per
bed, and (3) geographical location, and (4) projected revenues from user fees and health insurance
reimbursements. Unfortunately, the current budgeting procedures do not identify the purposes for which
state budget support to central hospitals i s intended, for example, teaching or research or to help pay for
services o f the poor and near poor. Accordingly, there i s no basis for monitoring whether the funds have
been used as intended, whether they are sufficient for the purposes intended, or whether the desired results
have been achieved.
I Source: Ministry of Health
Policy Instruments
12.54 Once revised responsibilities, actions, and funding levels have been determined,
existing policy instruments w i l l need to be strengthened and new ones may need to be
developed to supplement or replace those that are no longer as effective in a rapidly
developing decentralized market economy. For example, assigning responsibilities for
necessary actions to various levels o f Government may require new types o f
intergovernmental transfers, including unconditional (lump-sum), categorical or
conditional (closed- or open-ended matching grants, non-matching grants), or direct cost
reimbursement. Existing government procurement procedures, which were developed for
a relatively simple health sector, both technologically and institutionally, need to be
adapted to present circumstances in which a growing and diffuse group o f local,
provincial, and national authorities purchase an increasing number o f items from a
rapidly expanding group o f state and private f i r m s . Monitoring and evaluation functions
also need strengthening in many areas.
RECOMMENDATIONS
12.55 Substantial progress has been made b y the Government in moving forward on
several o f the reform options proposed in the 2000 PER. Nevertheless, faster progress i s
needed on several of the other reform items i t presented, including revision o f the user fee
schedule that dates back to 1995, development o f a mechanism for ensuring consistency
between patterns o f province health spending and those advocated in national health
strategies, and rethinking the role o f the Government in the health sector. The analysis
and recommendations presented below are in part designed to support progress in these
areas b y providing specific ideas about needed directions o f reform.
Ensure that the Level of Government Health Funding is Adequate to Perform Key
Health Functions
12.56 Limited government health funding i s currently spread very thinly among a very
wide range o f activities. Under these circumstances, the risk i s that nothing i s done very
effectively. The Government should estimate fully and on a medium-term basis the
74 Public ExDenditure Review and Intearated Fiduciarv Assessment
resources required to implement the critical actions needed to perform i t s key health
functions and proceed to secure those resources. M O H should take the initiative in
implementing i t s critical responsibilities o f coordination and direction in the health
sector: for example, gradually transferring common health provisions to private sector
and transferring currently direct funding for health facilities step b y step to funding health
beneficiaries. Doing so may well indicate that an increase in current levels o f government
health spending i s required. I f so, such an increase w i l l represent a sound investment in
Vietnam’s future, not only because there i s evidence that investments in health and
nutrition can accelerate the rate o f economic growth but also because improvements in
health status and longevity (similar to those recently obtained in Vietnam) are highly
valued by people.28 The next step, as part o f the planned development o f a Medium Term
Expenditure Framework (MTEF) for health, should be to develop tools and processes to
plan the budget on the basis o f a set o f clearly defined and carefully prioritized outputs,
instead o f on the basis o f the input-based norms that have been used in the past.
Make the Distribution of Health Spending More Equitable, both Geographically and
Among Income Groups
12.57 Both centrally allocated and local-level government health spending are already
weakly pro-poor at the province level (Figure 12.2). However, the much larger non-
governmental component o f total health spending (and particularly household out-of-
pocket spending) i s pro-rich. The allocation o f State Budget health funding needs to
become more strongly pro-poor at all levels (i.e., national, provincial, district and
commune) in order to offset the inequities caused b y the rapidly growing share of other
sources o f health funding (for example, user fees and health insurance reimbursements).
The most effective way to make State Budget health funding more pro-poor under the
current system o f fiscal decentralization may be a phased shift from direct budget support
(supply-side financing) for curative health care to targeted demand-side financing (for
example, by providing subsidized health insurance to the poor and near poor, as with
Decision 139). Expansion o f compulsory health insurance coverage, in which
contributions are linked to household income and revenues are managed in a single pool,
i s the most efficient and equitable strategy ultimately for replacing inequitable out-of-
pocket spending. The next step should be to develop (in the context o f the MTEF) a new
set o f budgeting norms that reflect not only population size but also local health needs
and capacit to mobilize local health resources (for example, through user fees and health
insurance) .
Strengthen the Implementation of Decision 139
2;
12.58 I t i s recommended that all provinces purchase health insurance cards for the poor
rather than reimburse healthcare providers directly. I t i s also recommended that the
** There i s macroeconomic evidence, for example, that increases in average life expectancy lead to higher household
savings, increased investment in children’s schooling, and more rapid economic growth (Commission on
Macroeconomics and Health 2001, Bloom et al. 2001). There i s also strong microeconomic evidence o f high economic
retums to investments in early childhood health and nutrition (Strauss and Thomas 1998, Hoddinott and Quisumbing
2003). The available evidence also suggests that the level o f GDP per capita i s a conservative estimate o f the value
people attach to an additional healthy year o f life (Belli et al. 1998, Ashenfelter and Greenstone 2004).
29 One current proposal i s that State Budget allocations to regions and localities should be revised b y increasing norms
per capita for the mountainous and remote areas from 1.6 times to 2.5 or 3.0 times that for delta areas.
Health Sector Expenditure 75
current level o f funding per Decision’139 beneficiary be carefully reviewed and that the
central contribution be raised if necessary to cover a larger share o f the costs o f health
care utilized by the poor and other vulnerable groups. Doing so w i l l make Decision 139
financing even more pro-poor b y reducing the financing burden on relatively poor
provinces. I t i s further recommended that the implementation o f Decision 139 (including
the methods used to identify the poor) should be carefully monitored and evaluated, with
appropriate steps taken to correct any problems encountered as implementation proceeds.
Proceed Cautiously with the Implementation of Decree 10
12.59 Although Decree 10 provides potential opportunities to improve the efficiency
and sustainability o f public hospitals, i t also creates a risk that public hospitals may
neglect their basic missions in an effort to improve their “bottom line.” In particular,
there i s an urgent need for careful monitoring o f the use o f state budget funding (both
recurrent and capital) provided directly to Decree 10 hospitals to ensure that i t does not
impact adversely the wider public missions o f certain hospitals (for example, training,
research and professional directions to lower levels); and i s not used to support the
expansion o f services for the rich. One way to do this i s to shift State Budget curative
health care subsidies to the demand side via subsidized health insurance, as discussed
above. The Government team felt that Decree 10 should continue to be implemented at
the central and provincial-level hospitals and be piloted at the district level hospital and
commune-level health facilities. The World Bank team recommended that Decree 10
implementation in the centrally managed hospitals be delayed until appropriate
monitoring mechanisms are in place to ensure that the larger public missions o f these
hospitals w i l l not be compromised; that an appropriate mechanism for Decree 10
implementation in district level hospitals should be developed and tested; and that Decree
10 should not be extended to commune-level health facilities. Shifting from input-based
budgeting (for example, bed-based norms) to output-based budgeting with carefully
designed performance indicators (again, in the context o f the MTEF) w i l l be an important
step toward putting an effective monitoring mechanism in place for Decree 10 hospitals.
Improve health information
12.60 Greater priority should be given to health information dissemination - which
evidence suggests would be a cost-effective investment. The fact that many effective
preventive measures are more commonly used and understood by better educated
Vietnamese (for example, quitting smoking and wearing helmets when driving or riding
on a motorbike) suggests that health education a i s necessary and effective. In addition to
active prevention, effective health education i s also needed to equip people better
knowledge o f drug use properly and safely and to inform the population more fully about
the advantages o f health insurance.
Develop effective quality assurance systems for both public and private providers
12.61 The Ministry o f Health should take the lead in further developing strong quality
assurance mechanisms (including peer review) in both public and private health facilities.
This should be a precondition for the use by VSS o f more flexible reimbursement
mechanisms, such as capitation, and i t i s probably also a necessary condition for effective
private sector participation in the social health insurance system. Effective quality
76 Public Exuenditure Review and Integrated Fiduciaw Assessment
monitoring i s also needed to ensure that the quality o f care received b y different
categories o f patients (for example, Decision 139 and insured patients versus fee-paying
patients). VSS can and should play an important role in developing strong quality
monitoring systems. VSS alone has access to the necessary data on hospital care
nationwide to provide a basis for monitoring the quality and cost effectiveness o f health
care at all levels. This information should be computerized as soon as possible so that i t i s
readily accessible to those monitoring trends in health care, including the prescription
practices o f health providers as part of efforts to control drug costs. Currently, i t i s very
difficult to obtain reliable information on the quality and cost effectiveness o f care
obtained from private providers. However, as health insurance coverage expands and as
more private providers are brought under the social health insurance program, VSS
should have access to the information needed to monitor the quality and cost
effectiveness o f services provided in both the public and private sectors. The next step
should be the development and implementation o f an accreditation mechanism covering
both public and private hospitals. Hospitals unable to achieve accreditation after a
suitable period o f time should not be eligible to provide services to the insured.
Increase the Role of the Private Sector, Particularly in the Provision of hospital
services
12.62 Vietnam i s out of step with most other countries in the region in having a very
small private hospital sector (and particularly with those countries such as Japan and
South Korea that have universal social health insurance coverage, as Vietnam aspires to
have by 2010). There i s already a substantial and rapidly growing demand in the upper-
income urban population for better-quality hospital care. In the absence o f a strong
private hospital sector, many o f the larger government hospitals are actively gearing up to
provide better quality “special services” to meet this demand. This trend has accelerated
with Decree 10. However, i t could be preferable to meet this demand in future through
private hospitals, with appropriate quality monitoring arrangements in place. In addition,
measures should be taken to avoid the risk that government hospitals would divert
resources from their basic missions in trying to meet the demand for higher-quality care.
An important barrier to the development o f a strong private hospital sector in Vietnam i s
the continuing practice o f subsidizing curative health care through direct budgetary
support (block grants) to public hospitals. A phased shift from direct budgetary support to
public hospitals to targeted demand-side financing for the poor and near poor (via
subsidized social health insurance) could facilitate the development o f a private hospital
sector by leveling the playing field between public and private hospitals. The next step
should be making all accredited private hospitals eligible to provide services to the
insured while simultaneously shifting government subsidies for curative health care from
the supply side (block grants to public hospitals) to the demand side (subsidized social
health insurance for vulnerable groups).
Strengthen Health Sector Public Procurement Practices
12.63 A two-stage process would likely be required in order to improve the
effectiveness o f procurement in the public health system. The first stage o f work would
focus on ensuring adherence to formal procurement rules across the entire system. During
this first phase o f work, local, provincial and national agencies should assemble complete
Health Sector Expenditure 77
registries o f the assets under their control/ownership to ensure that state-owned property
i s properly accounted for and maintained. The second phase o f work would focus on the
introduction o f practices that would lead to increased efficiency in health sector
procurement. As a first step, Government should review the enormous body of
international experience in the area o f health system purchasing (for example, the use o f
large, framework contracts, innovations in sourcing, and experience in purchasing
pharmaceuticals needed to treat HIV/AIDS).
Monitor and control, where necessary, key healthcare prices
12.64 The Government has a responsibility to monitor healthcare-related prices in
markets in which significant market failures are present and to intervene if necessary to
correct any serious problems. The two markets in which government actions to monitor
and control prices are probably most needed are markets for hospital services and
patented drugs. There i s an urgent need for the Government to adjust the current schedule
o f hospital user fees to reflect rising costs and the introduction o f more advanced
technology. The schedule o f hospital user fees i s currently under review, as previously
noted. This i s overdue-the fee schedule has not been revised since i t was first
introduced in 1995. The Government i s considering various reform options, whereby
hospital fees can be more effectively regulated in the future through the use o f a formula
that would permit regular adjustments to fees to reflect inflation while providing
sufficient flexibility to hospital directors to support efficient operation. Prices o f branded
drugs can be effectively controlled b y linhng local prices and price changes to similar
developments in a limited number o f reference markets in other countries. In addition,
more competitive mechanisms for procuring drugs should be introduced, and anti-
monopoly measures should be strengthened, and consideration should be given to
establishing a Reserve Stock for essential drugs. The next step should be for MOH
systematically to review international experience in regulating hospital and
pharmaceutical prices and propose effective mechanisms for regulating key health sector
prices.

13. AGRICULTURE AND RURAL DEVELOPMENT
SECTOR EXPENDITURE
INTRODUCTION AND OVERVIEW
13.1 Agriculture has an important role to play in the socio-economic development o f
Vietnam. During the review period, agriculture continued to record positive growth,
transforming the sector from a subsistence production sector to a commercial and export-
oriented one. This agricultural transformation has been a vital component in Vietnam’s
pro-poor growth. The use o f market mechanisms to provide incentives to farmers,
beginning with the doi moi reforms o f the 1980s, resulted in substantial agricultural
sector growth. In particular, much o f this growth was based on the household farm, a
critical factor in reducing rural poverty. The fast growth of small and medium-sized
enterprise generated widespread income earning opportunities. Rural poverty almost
halved between 1993 and 2002, falling from over 66 percent to below 36 percent in this
period.
13.2 I t i s important to recognize that economic growth and rural poverty alleviation in
the 1990s was not simply the result of economic liberalization in the agricultural sector.
Public expenditure, both directly and indirectly, has made an important contribution to
this. Increasingly competitive markets were supported b y public provision o f basic
infrastructure, b y substantial progress in human development and by publicly financed
research and extension services. Furthermore, i t i s important to recognize that some o f
the easiest gains from agricultural liberalization have already been achieved. In coming
years, the efficient, effective and equitable provision and management o f rural public
services and infrastructure w i l l play an even more important role in sustaining Vietnam’s
agricultural growth and rural development.
13.3 This chapter reviews public expenditure in the agriculture and rural development
sector3’ in Vietnam. The first section o f the chapter describes the sector context and
recent trends in sector expenditure. The second section assesses the main impacts of this
expenditure and the scope for improvement in i t s allocation and management. The third
section considers sub-sectoral expenditure issues. The fourth section presents
recommendations.
SECTOR CONTEXT
Sector Strategy and Priorities
13.4 Vietnam’s objectives for agriculture are given in the Government Ten Year
Strategy and Five Year Plan and consolidated in the 2002 Comprehensive Poverty
30 Including fisheries except where stated.
79
80 Public Expenditure Review and Integrated Fiduciary Assessment
Reduction and Growth Strategy (CPRGS). In summary, the general objective for the
sector i s to develop a commodity production agriculture which i s diverse, efficient,
competitive and sustainable. I t would meet domestic and export needs for food and
industrial inputs, provide jobs, and eliminate poverty and hunger. The strategic focus i s
on product quality, competitiveness, market linkages, and development o f human and
natural resources. Specific objectives, or targets, are to achieve a growth rate higher than
in the previous Plan period (4.5 percent), and reach commodity area and production
levels.
Growth and Sources of Growth
13.5 Agriculture’s long-run growth rate continues at about 3-4 percent which i s high
b y international standards. Nonetheless, i t i s much lower than that o f a decade ago and
only half that o f GDP, 7.3 percent in 2003 (Table 13.1), and thus w i l l not reach the
current Plan goal o f exceeding the previous rate (5 percent). Agriculture’s contribution i s
about one-seventh o f overall annual GDP growth o f 7 percent. Analyses o f sources o f
growth within agriculture (Prota and Smith 2004, Fan et al. 2004) suggest that past
growth was mostly due to factor accumulation-more land, irrigation, fertilizer,
improved varieties-and one-time institutional reform such as the return to family
farming, subsequent land allocation and titling, and agricultural commodity trade
liberalization.
Table 13.1: Poverty, Growth and Economic Structure - 1998-2003
(%)
1998 1999 2000 2001 2002 2003
Poverty Rate (IS)
National
Urban
Rural
Growth Rate
GDP
Agriculture
Agriculture Share of GDP Growth
Economic Structure
Share of Agriculture in total GDP
Share of Agriculture in total labor
37.4 n.a 32.0 n.a 28.9 n.a
9.2 n.a n.a n.a 6.6 n.a
45.5 n.a n.a n.a 35.6 n.a
5.8 4.8 6.8 6.9 7.1 7.3
3.5 5.2 4.6 3.0 4.2 3.2
0.9 1.2 1.1 0.7 0.9 0.7
25.8 25.4 24.5 23.2 23.0 21.8
69.6 68.9 66.3 65.1 60.9 59.6
Source: CIEM Vietnam’s Economy in 2001, 2003. Estimates for 2003
CPRGS Report of Steering Committee 2003
GSO, 2002-04. Statistical Yearbooks 2002-2004
13.6 This factor accumulation was largely due to state expenditure on irrigation while
the institutional reform was a direct result o f policy change. Productivity gains were a
secondary source o f growth (Barker et al. IFPRI 2002). This view was also supported by
Que and Goletti (200 1) investigating agricultural productivity changes in Vietnam
between 1985 and 1999, and b y Prota and Smith (2004). Their studies show that total
factor productivity (TFP) increased mainly from response to institutional reforms and that
research played only a small role since expenditure on research was relatively small. As
the scope for factor accumulation decreases, consistent productivity gains w i l l be needed
Agriculture and Rural Development Sector Expenditure 8 1
for domestic and international competitiveness and expenditure w i l l have to be more
selective. For example, while the total volume of public investment in irrigation was the
most important source of agricultural growth, the return per unit o f spending i s greatest
for research.
Expenditure and Sector Performance
13.7 There has been progress in implementing recommendations o f the 2000 public
expenditure review. According to the sector’s Fiv.e-Year Plan 2001-2005, the priority o f
public expenditure i s given to further developing and upgrading rural infrastructure
(irrigation, rural road and power supply), forest plantation and protection, agricultural
research, extension and human resource development, post-harvest processing and
marketing. In particular, since 2001 public expenditure on agricultural research and
extension in 1994 prices increased on average by 13.2 percent per annum. This slightly
increased their share in the total agricultural expenditure to 4.9 percent in 2003 - s t i l l low
compared with neighboring countries. Some important measures have been undertaken to
accelerate the reform o f the agricultural research system, decentralize extension services,
develop water user associations and improve the cost recovery rate o f irrigation services.
However, they are s t i l l to be implemented with an appropriate poverty focus.
Furthermore, despite reduced subsidies to sugar and fertilizer industries, intra-sectoral
resources reallocation remains in need o f improvement as SOEs, including some loss-
making SOEs, continue taking a lion share o f scarce resources and have even doubled
their debts to public sources.
13.8 There is, though, considerable scope for gains to be made from better use o f the
existing level o f public spending. Over the medium-term, public expenditure faces the
challenge o f finding the best path o f transition involving:
0 shifting relative roles o f public including SOE and private expenditure and
investment;
decentralization o f authority over expenditure decisions; and
gradual decline in the share o f irrigation and hydraulic infrastructure in the public
budget for agriculture.
Specific goals in public expenditure which are emerging that would have
Completing irrigation systems to improve the utilization rate o f existing schemes
above the current level estimated at 50-60 percent.
Rehabilitating and maintaining irrigation works. Due to the present system
inefficiency, farmers often cannot pay the high cost o f irrigation services,
especially in the North.
0 Rationalizing, modernizing and upgrading research to support productivity
increases. In the context o f available funds, there are too many research entities.
0 Providing professional extension services to the commune level. Extension
services below the district level currently are through generalist agricultural
employees of the communes and through multi-purpose mass organizations.
0
0
13.9
particularly high impacts include the following:
0
82 Public ExDenditure Review and Intenrated Fiduciarv Assessment
0 Providing market intelligence to increase responsiveness to market signals.
Present market information systems are too aggregate and unreliable.
0 Upgrading product quality standards. Upgrading the legal and regulatory
framework i s essential to enforce compliance and to protect incentives for quality.
Providing budget incentives for better than average performance in service
delivery b y provinces.
a
Level and Trends
13.10 The agricultural sector has maintained a relatively constant share o f overall
expenditure at about 5-6 percent over 1997-2002 (Table 13.2), which i s low in regional
and international context. Total budget expenditure for all sectors rose b y 91 percent
during the same period, while expenditure on agriculture, forestry, and fisheries
(collectively referred to as agriculture) rose 96 percent in nominal terms (88 percent in
constant 1997 terms). This maintained the sector’s ranking after education, transport and
health, and above industry (which has higher FDI). Spending on fisheries grew faster
than that on agriculture and forestry but comprised only 5 percent o f total spending in the
sector as a whole.
Table 13.2: Budget Expenditure on Agriculture, Forestry and Fisheries
1997 - 2002 (nominal VND)
Total Agriculture Agriculture Agriculture Budget
Budget Budget & Forestry Share of Total Budget
Year VND bn VND bn only %
1997 70,749 3,815 3,712
1998 73,419 5,075 4,591
1999 84,817 5,326 5,051
2000 103,151 5,804 5,647
2001 119,403 7,420 7,044
2002 135,490 7,47 1 6,993
5.4
6.9
6.3
5.6
6.2
5.5
Source: MOF (2004)
Composition
13.11 The major trend has been the faster growth in capital cost funding, 107 percent
since 1997 compared with 66 percent for recurrent expenditure (Table 13.3). Investment
expenditure has accounted for 75 to 80 percent o f the total over the review period, up
marginally from earlier, and spiking in response to unexpected revenue availability in
2001 (Table 13.4). The recurrent budget accounts for the balance o f 20 to 25 percent
which i s a decline from the long-run 25 percent between 1992-98. Within recurrent
expenditure, the wages and salaries share has risen from about 14 percent to 17 to 20
percent .
13.12 The major trend in investment expenditure has been in the proportion managed by
local government rising from 48 percent to 67 percent o f the total between 1997 and 2002
due to decentralization in investment management whereby central agencies concentrate
on major projects. Another significant trend has been the decline in the share of
AIzriculture and Rural DeveloDment Sector ExDenditure 83
construction o f irrigation works from 78 percent o f the total to 60 percent over the same
period. The balance o f sector public investment i s on support to programs in cultivation,
plant breeding, animal health, forestry, settlement (sedentary abode and farming), and
infrastructure programs such as rural water supply. O f these, forestry i s the largest
program.
Table 13.3: State Expenditure for Agriculture - 1997 - 2002 (current VND bn)
Year Total Capital Recurrent
1997 3,815 2,679 1,045
1998 5,075 3,946 1,129
1999 5,326 4,300 1,027
2000 5,804 4,542 1,263
200 1 7,420 5,953 1,466
2002 7,47 1 5,736 1,735
Change 96% 107% 66%
Source: MOF (2004)
Table 13.4: Capital and Recurrent Budget Expenditure on Agriculture - 1997 - 2002
~~ ~
Capital Recurrent Wages
% % %
1997 72.6 27.4 14.3
1998 77.8 22.2 15.4
1999 80.7 19.3 17.5
2000 78.3 21.7 20.7
200 1 80.2 19.8 20.9
2002 76.8 23.2 17.8
Source MOF (2004)
13.13 Under-funding of recurrent expenditure, and operations and maintenance
expenses in particular, was signaled as a problem in the 2000 PER and has continued to
worsen. There i s a persistent increase in imbalance between the sector’s spending on
operation and maintenance (O&M) and the rapidly accumulated stock o f capital assets as
a result o f the emphasis on the investment side (Table 13.5). For five years from 1997 to
2002, the sector’s spending on O&M increased only b y 31 percent as the larger part o f
the increase in i t s recurrent costs, 107 percent change, went to fixed items such as salaries
and wages. Local government i s playing an increasing role in management o f recurrent
spending. Total state recurrent expenditure in agriculture (fisheries excluded) was
VND1,641 billion in 2002, o f which 38 percent was managed b y MARD and the rest b y
local government. These costs are for research, extension, national and targeted
programs, and i t s own administration. MARD-managed recurrent expenditure has
doubled between 1999 (VND310 billion) and 2003 (VND748.3 billion).
Sub-sector Expenditure
13.14 Sub-sector expenditure and performance i s difficult to analyze because o f the
problems associated with availability o f data, i t s timeliness, and i t s reliability. Official
public expenditure data are presented in aggregated format that i s not suitable for detailed
performance analysis. For example, categorization o f public expenditure on agriculture i s
b y program type and not by function. The Budget category of “Cultivation” includes part
84 Public Expenditure Revieu! and Iiitegrated Fidiician Assessment
o f research, extension, and other items as do other categories. A more useful breakdown
would be as shown in Table 13.6, with “other” disaggregated. The numbers in this table
are not published but are obtained from MARD. There are some data discrepancies
between MARD data and MOF data, which are difficult to reconcile. One result o f the
relative inaccessibility i s the paucity o f published research, particularly into impact
evaluation o f expenditure programs in other than general terms. While there i s some
excellent survey work carried out, public expenditure research in the agricultural sector i s
limited in quantity and quality.
Table 13.5: State Expenditure for Salaries and Wages and O&M in Agriculture - 1997-
2002 (current VND bn)
Operation and
Year Salaries and Wages Maintenance
1997 149 727
1998 174 727
1999 180 610
2000 26 1 699
200 1 307 752
2002 308 952
Chanae 107% 31%
Source: MOF (2004)
Table 13.6: Agricultural Budget by Sub-sector (VND bn)
1999 2000 2001 2002 2003
Irrigation 3,241 3,620 4,678 4,211 na
Forestry 444 546 576 67 8 n.a
Research n.a. 150 162 168 197
Extension 85 108 155 196 185
Other n.a. 1,380 1,849 2,2 18 n.a.
Total 5,326 5,804 7,420 7,47 1
Source: MARD (2004)
13.15 According to MARD data, irrigation (inclusive o f drainage and flood control)
remains the largest sub-sector, averaging over 60 percent o f aggregate expenditure
although a downturn occurred in 2002. The “other” category has grown rapidly and i s the
second largest grouping, followed b y forestry, where the Five M i l l i o n Hectare
Afforestation Program (SMHAP) i s the largest program. Research and extension are the
remaining two major categories, though much smaller, and despite strong growth in
nominal terms, remain under 5 percent o f total expenditure.
Impact of Fiscal Autonomy on Agricultural Expenditure
13.16 The agricultural sector has experienced a rapid and radical decentralization o f
authority over public expenditure. The share o f the agricultural sector expenditure
implemented b y local levels has almost doubled, from 43 percent in 1997 to 79 percent in
2002. The implementation of the 2002 State Budget Law from 2004 further extends
budget delegation and autonomy to the provincial and commune levels. These changes
are implying a major change in the role o f the central MARD, one which i t i s only slowly
beginning to adjust to.
Agriculture and Rural Development Sector Expenditure 85
13.17 There has not yet been enough time to evaluate the impact on the structure o f
sector expenditure o f this decentralization. The implementation o f the new State Budget
Law and o f Decree 10 i s expected to increase provinces’ and spending units’ discretion
and this may have a positive impact on the spending structure. However, this w i l l be to a
large extent dependent on the priority that local governments attach to agriculture. Field
visits indicate that local governments seem to be reluctant to increase spending on
agriculture from revenues in excess o f those in the plan.
Public Expenditure on State Enterprises in Agriculture
13.18 There are 319 SOEs overseen b y the Center in the agriculture and related sector
(2003 estimate, Table 13.7). This number does not include the Irrigation Management
Companies (IMCs) and State Forestry Enterprises owned b y the provinces. Of this total,
28 are public utilities providing public goods and the rest offer services, and o f this latter,
171 are considered profitable, 43 just manage to cover their costs, and 105 are
unprofitable. The main sources o f capital are loans from State-owned commercial banks
(VND15,203 billion) and allocations from the State Budget (VND6,035 billion). Total
revenue for 2003 i s estimated as VND42,074 billion with profits o f VND 1,456 billion
and losses o f VND387 billion. Contributions to State Revenue are estimated as
VND1,266 billion for 2003. The SOEs owe VND 515 billion to the State, VND15,203
billion to the banks, and VND3,700 billion to other enterprises.
Table 13.7: Financial Statistics of Agriculture SOEs Owned by the Center (VND bn)
1999 2000 2001 2002 2003Est.
Number of SOEs
Capital
State Budget
Self generated
Loans
Other
Profit and Loss
Revenue
Profit
Loss
Contribution to Budget
Revenue Tax
Profit tax
Export/Imp. Tax
Consumption Tax
Capital Tax
Payable
To Budget
To Banks
494 462 448 387
5,120 5,161 5,463 6,028
2,211 2,459 2,575 2,583
6,951 9,515 11,832 14,714
200 206 265 241
25,377 25,917 31,869 34,773
360 617 609 831
462 555 681 356
388 476 401 425
100 172 170 232
134 177 150 155
12 10 10 11
47 57 62 n.a
605 760 1,039 317
6,951 9,515 11,832 14,714
319
6,035
2,590
15,203
250
42,074
1,456
3 87
65 2
407
195
12
n.a
5 15
15,203
To Other Firms 3,123 3,023 2,014 3,523 3,700
Source: MOF (2004)
13.19 In the four years between 1999 and 2003, the debts o f the SOEs to the budget and
state-owned banks have more than doubled from VND 7,556 billion to VND 15,718
86 Piiblic Expenditirre Review and Integrated Fidiiciaq Assessinent
billion or twice the agricultural budget in 2003. This administratively directed credit
exposes the State-owned commercial banks to high risk and represent potential fiscal
contingent liabilities. The debt figures probably underestimate the true cost o f those funds
because these are subsidized. Some SOEs provide important public services particularly
in remote areas but many SOEs also provide the marketing, processing, and other
services that the private sector can often do more efficiently. Thus, there exists scope for
budget savings that can be used to provide increased funding for those agricultural
services that cannot be provided by the private sector, such as rehabilitation o f
smallholder irrigation infrastructure and research. A small 2.5 percent decrease in the
debts o f the SOEs would free savings sufficient to double the annual budgets for research
and extension.
Impact of Other State Budget Measures on Agriculture
13.20 A review public expenditure on agriculture would be incomplete i f the effects o f
tax expenditures (tax reduction and exemption) on agriculture were not taken into
account. Starting from 2002, the State exempted 50 percent o f agriculture land use tax;
and since 2003 a 100 percent exemption o f this tax, equivalent to 1,600 billion VND in
revenue, i s granted. This could be considered as a direct investment b y the State in
agriculture, supporting the sector’s development and poverty reduction among rural
farmers which account for 75 percent o f Vietnam total population.
13.2 1 Similarly, several new target programs result in substantial additional funding for
agriculture and forestry and contribute to improved living conditions for rural people.
These include the program to support the most disadvantaged communes (Program 135),
the program to concretize irrigation canals, the program to support flood-prone areas in
Mekong Delta and most recently the program to support distribution o f production and
housing land to farmers. While exact data i s s t i l l not available, preliminary estimates
indicate that a combination o f indirect public expenditure on agricultural development
and tax exemption would amount to 20-30 percent o f direct public expenditure. As such,
public expenditure, both directly and indirectly, on agriculture might account for more
than 6 percent o f the total State Budget.
ODA Impact on Budget Resources
13.22 Actual ODA disbursements between 1997 and 2001 were stable at about VND
3,300 billion per year, but declined from 88 percent to 46 percent o f total expenditure for
agriculture which grew faster (Table 13.8). The two sets o f numbers are not strictly
comparable since the ODA disbursements partly refer to rural rather than agricultural
activities (for example, poverty mapping, rural roads and micro-enterprise development,
are included in the ODA figures, as well as project disbursements for rural credit, which
are not financed through the Budget). The majority o f the 564 rural projects in this ODA
database maintained b y the UNDP are for farming, forestry, and fisheries. O f these, 37
are classified as investment projects with the remainder being technical assistance and
small- grant activities. Nevertheless, they do give an indication o f the order o f magnitude
o f rural ODA in comparison with the sector budget and i t i s clear that ODA plays an
important but declining part in Vietnam’s public expenditure program for agriculture.
The forestry sub-sector i s particularly dependent on ODA finance: MARD reports
(Finance Department 2004) that from 1996-2003, ODA accounted for 61 percent o f the
Agriculture and Rural Development Sector Expenditure 87
investment resources (VND 2,043 billion) that forestry received from the State Budget.
Overall, since the largest ODA projects are for capital works, they may themselves
explain the increasing imbalance between capital and recurrent cost funding in the
Budget allocations.
Table 13.8: ODA to the Agriculture Sector - 1997-2002
ODA Share of Total ODA Total ODA Total
Disbursements Disbursements Agriculture Agriculture
for Agriculture for Agriculture Budget Budget
USD m. VND bn. VND bn. %
1997 215 3,345 3,815 87.7
1998 215 3,339 5,075 65.8
1999 192 2,982 5,326 56.0
2000 216 3,360 5,804 57.9
2001 218 3,397 7,420 45.8
Note: Including also spending on area-based development
Source: UNDP Vietnam Development Cooperation Report (2003)
MAIN IMPACTS AND SCOPE FOR IMPROVEMENT
Budget Adequacy
13.23 Even if agriculture’s share o f the total State Budget indeed makes up more than 6
percent o f State Budget spending, i t appears low relative to the policy emphasis given to
i t s role in the CPRGS and other Government documents. Since 90 percent of the poor
live in rural areas and 70 percent o f the income o f rural workers (13.9 million farm,
aquaculture and forestry households) comes from agricultural sector activities, a larger
share might have been anticipated for support to those activities. Moreover, agriculture’s
share o f the total State Budget i s low b y regional comparison with China, India and
Thailand which ranges from 8 percent to 16 percent (Kherallah and Golleti 2000). I t may
indicate possible under-funding o f sectoral services and maintenance o f rural
infrastructure.
13.24 A relatively low budget share for agriculture does not itself make a case for an
increase. An increase would be justified if the budget level for agriculture was found to
be insufficient to carry out the sector’s growth and equity obligations under the Plan, if
the budget level was found to be insufficient to maintain productive infrastructure, and i f
the returns from extra funding for agriculture were greater that those from allocating the
extra funds to other sectors, which i s the case with spending on agriculture research.
Further, the relatively high long-term growth rate in sector production does not suggest
that there has been long-term underinvestment. The sector’s contribution to GDP i s also a
result o f investment in transport, industry, education, health and other activities not
captured in the estimation o f investment bias (Box 13.1). While the emphasis on the
importance o f these sectors in the CPRGS w i l l help raise incomes from whatever source,
they may not necessarily solve the basic reduced rate in productivity growth in
agriculture. At present, the budget classifications are not detailed enough to support
monitoring and evaluation o f public expenditure at sub-sector level (for example,
extension or veterinary services); and “agriculture sector expenditure” data does not
88 Public Expenditure Review and Integrated Fiduciary Assessment
capture the significant contribution to public services in agriculture made by other
sectors, off-budget agencies and the private sector.
Box 13.1: Poverty Reduction and Growth Impact of Rural Spending
Government expenditure has been crucial to growth and poverty reduction in Vietnam. Two recent
studies have measured the impact of this spending. The first, b y Fan et al (2004), found that
government spending on irrigation, roads and agricultural research has contributed to both agriculture
growth and poverty reduction.
Among all types o f government spending, agricultural research had the largest return in agricultural
production. For every VND 1 spent, VND 12.22 of agricultural production value would be produced.
Road investment had the second largest return (VND 3.01 for every VND 1 invested), followed by
investment in education (VND 2.06 for every VND invested). In contrast to other types o f investments,
return to irrigation investment was less than 1, meaning that the economic benefit (excluding impacts
on food security, flood control and clean water supply) generated does not cover the cost
Interestingly, the ranking o f these investments by poverty effect i s exactly the same as the ranking
by agricultural growth effect. Among all types o f government spending, agricultural research has the
largest return in poverty reduction. For every billion dong spent on research, 339 poor people would be
lifted above the poverty line. The poverty impact o f roads ranks second, with every billion dong spent
on roads lifting 132 poor people above the poverty line. Investments in education also have very
favorable returns in poverty reduction (76 people escape poverty with every billion dong invested).
Among all types o f spending, irrigation had the smallest impact on poverty; only 13 people are lifted
above the poverty line for every one billion dong spent.
This implies that there i s no trade-off between agricultural growth and poverty reduction when
selecting investment priorities. However, great variations are observed among regions in the effects of
public investment on agriculture production and poverty reduction. For example, the largest poverty
reduction impact from road investment occurs in North Central Vietnam, while the return from
education i s the largest in the Southeast.
The second study, b y Baker et aZ(2002), estimated determinants o f agriculture growth for the same
period and found that public investment in irrigation was the most important source o f agriculture
growth (accounting for 28 percent of the growth), followed b y agriculture research (27 percent).
Investment in roads accounted for 11 percent of agricultural output growth and education 8 percent.
An important finding from these two studies i s that while irrigation has been the largest source of
growth, the aggregate irrigation investment programmed has become uneconomic. The conclusion i s
that the sector expenditure program should select only those irrigation investments with acceptable rates
o f return and that savings should be reallocated to high-return activities, particularly agricultural
research.
Source: Fan, Shenggen, Pham Lan Huong, and Trinh Quang Long, 2004. Government Spending and Poverty
Reduction in Vietnam Draft Report Prepared for the World Bank Funded Project “Pro-Poor Spending in
Vietnam”, by International Food Policy Research Institute, Washington D. C., and Central Institute for Economic
Management, Hanoi.
Barker, R., Ringler, C., Nguyen Minh Tien, and Rosegrant, M., 2002. VN-4: Macro Policies and Investmeni
Priorities for Irrigated Agriculture in Vietnam, National Component Paper for the Project on “lrrigatior;
Investment, Fiscal Policy, and Water Resource Allocation in Indonesia and Vietnam”, IFPRI Project No. 2635.
000, Country Report, Vietnam, Vol. 1, Asian Development Bank.
Budget Impact on Poverty and Food Security
13.25 Prior to the land reforms o f the 1980s, Vietnam imported some 500,000 tons o f
rice annually and seasonal hunger was widespread. Vietnam now exports 3 4 million
tons o f rice annually. While many factors contributed to this turnaround, i t would not
have occurred without the large investment program in irrigation. Food security at the
national level i s now largely secured and many households have moved out of food
Agriculture and Rural Development Sector Expenditure 89
poverty, although food security remains an issue at the household level. The general rural
development strategy focused on irrigation for rice production has now been
complemented by the targeted poverty programs for those “left behind”. Similarly, the
coffee development program in the Central Highlands has relieved some o f the pressure
on resources in the large river deltas and has lifted many out o f poverty through
resettlement. However, in some instances, expansion o f commodity production has led to
a fall in export prices and farm incomes; otherwise the commodity programs could have
had more positive impacts both on the rate o f agricultural growth and the rate o f poverty
reduction. In response, agricultural development and rural poverty alleviation strategies
are shifting to place more emphasis on product diversification and better targeting o f
interventions.
13.26 Three key issues in the sector’s expenditure policy yet to be resolved are: (i)
continued financial support to State-owned enterprises and commodity market
liberalization, (ii) the level o f funding to be provided to agricultural services, particularly
research and extension, and (iii) the direction, location, and content o f expenditure on
irrigation (as discussed below). Without resolution o f these issues, the burden o f the cost
o f the resulting distortions w i l l continue to fall heaviest on the poor.
Budget Impact on Private Investment
13.27 Agriculture i s heavily dependent on the State Budget which accounted for 57
percent o f the total investment in the sector in 1999-2002 because o f the relatively small
level of private enterprise (non-household) investment (estimated at VND 6,025 billion in
2002 or 40 percent o f the total investment in average). Although FDI in the sector has
contributed the equivalent o f some US$98 million (or less than 35 percent o f the total
commitment) between 1999 and 2002, FDI in agriculture accounted for less than 3
percent of investment capital in the period, much smaller than the 17.3 percent for the
whole economy. Non-state local enterprises similarly have a small presence.
13.28 Several possible reasons for this emerge. First i s the relatively low protection
level for agriculture, resulting in fewer opportunities for profit. Another i s the higher
transaction costs and s k i l l s shortage in rural areas. Another consideration may be that a
large part o f foreign investment in agriculture worldwide includes investment in purchase
or lease o f large tracts o f land for corporate farming associated with processing and
export. This i s not available in Vietnam. The presence o f State-Owned Enterprises in
marketing, processing, and production tends to crowd out private sector investment.
SOEs occupy the space normally filled b y the private sector and continued budget
support for SOEs i s likely to ensure that private investment w i l l go elsewhere. Examples
are the Rubber, Sugar and Coffee Companies. These deliver marketing and processing
services to growers and that could be provided by the private sector. The Government i s
implementing a reform program to equitize and rationalize the SOEs but in the meantime
they are a heavy drain on State Budget resources, directly and indirectly through funding
from State-owned commercial banks, such as the Vietnam Bank for Agricultural and
Rural Development (VBARD) and the Vietnam Policy Bank (VBP), and b y the
Development Assistance Fund (DAF).
90 Public Exvenditure Review and I n t e n r a t e d Fiduciarv Assessment
Table 13.9: Investment Capital by Ownership - 1999
Ownership types 1999 2000 2001 2002
Amounts in VND bill
State ’ 8,579 9,227 8,253 8,504
Non-state and households 6,437 7,146 5,266 5,508
FDI 570 41 1 226 214
Total 15,643 17,218 13,629 14,529
Percentage
State 55 54 60 58
Non-state and households 41 44 38 40
FDI 4 2 2 2
Total 100 100 100 100
Notes/sources:
1. Including loans from state-owned banks; GSO, 2003 and
2004. Statistical Yearbook 2002 (pp 323, 329) and 2003 (pp 289, 2931,
2. Actual disbursements based on MPI F D I Database as of December 31, 2002
Planning and Budget Processes and Institutional Roles
13.29 MARD participates in the preparation o f the 5-Year Plan by drafting the sector’s
development strategies, objectives, and targets and participating in appraisal o f concerned
plans, including those prepared b y i t s subordinated bodies and provinces. However,
MARD does not have allocative authority. Plans from these entities are consolidated and
sent to Government and the National Assembly for consideration and approval, after
which the Plan with financial earmarked figures i s disseminated MOF and MPI directly
to the bodies and provinces. An Annual Plan i s then prepared on the basis o f the 5-Year
Plan and the annual budget i s developed around the needs o f the annual plans and the
funds available. This i s largely replicated at the provincial level and so on, except local
authorities have to follow the guidelines from the top. Often annual plans lack the
assumed connection with the 5-year plan due to persistent budget constraints. While the
Plans do have some policy and strategy content (although not specific enough), they
appear to be more a l i s t o f commodity production targets similar to the commodity
balances o f central planning. Missing from the Plans i s a comprehensive analytical
framework that would link production plans to market analysis and incentives, clearly
defined roles o f the public and private sectors and appropriate medium-term resources
allocation. Further, authorities responsible for the agricultural sector commonly lack
regular access to data on the sector’s budget execution, especially b y local levels.
13.30 The result i s inconsistency and mismatch in both stages o f planning and
execution. For example, the Plan discusses the need for market responsiveness while not
discussing the significant distortions that impede markets from functioning properly.
Another example i s that the current 5-Year Plan, 2001-2005, emphasizes production o f
fruits and vegetables and the need to develop processing capacity. But there i s no
treatment of real constraints on that industry arising from the lack o f a regulatory
framework, weak standards establishment and enforcement, little if any public cold
storage facilities for products in transit, scarcity of credit and technical assistance for
Agricirltirre and Riiral Development Sector E.upenditiire 9 1
entrepreneurs, and absence o f a timely market information system. Not addressing these
basic issues (while overemphasizing particular fruits to be encouraged) puts the emphasis
for public policy and expenditure in the wrong place; product choices and the attendant
market risk belong as calculations by producers and processors rather than b y public
policies. A weakness with current planning i s that targets are determined and resources
allocated without adequate regard to other variables so important to modern agriculture,
such as the economics o f production. A Plan that takes into account these variables would
serve as a sound framework within which the public expenditure program for the sector i s
prepared.
SUB-SECTOR PUBLIC EXPENDITURE ISSUES
Irrigation
13.31 State expenditure on irrigation and related services i s the largest claimant on the
agricultural budget. In 2002, VND 4,211 billion o f the total sector allocation of VND
7,471 billion was spent on irrigation, a rise from VND 3,241 billion in 1999 (Table
13.10). Irrigatioddrainage and dyke constructiodmaintenance comprise 88 and 12
percent, respectively. About 94 percent o f this was for capital expenditure (investment) in
2002, the same as in 1999. The share implemented b y MARD declined from almost 50
percent in 1999 to only 21 percent in 2002, and the corresponding province share
increasing from 50 percent to 79 percent. Fiscal decentralization explains part o f this, but
the trend also reflects a shift in emphasis from building new structures, to upgrading and
completing existing works. This lowers the size o f the project to that implemented b y the
provinces. MARD reports that upgrading (including completion, rehabilitation and
improvement as distinct from construction o f new schemes) rose from 28 percent o f
irrigation investment in 1996, to 48 percent in 2000, and 64 percent in 2003. Priority has
also shifted from the big irrigation projects in two delta regions to small and medium-
sized irrigation systems in the Central Coast and Mountainous areas as well as flood
prevention dykes in the Mekong River Delta. This not only reduces the size o f schemes
being funded but also increases the poverty reduction impact.
Table 13.10: Vietnam - Public Expenditure on Irrigation (VND bn)
Share in the State
Budget
1999 2000 2001 2002 (irrigation)
Total State Budget (irrigation) 3,241 3,620 4,678 4,211 100%
Capital Expenditure 3,063 3,388 4,411 3,959
Recurrent Expenditure 178 232 267 252
Expenditure by MARD 1,612 1,364 1,273 920 21%
Recurrent Expenditure 12 47 46 49
Capital Expenditure 1,600 1,317 1,227 87 1
Expenditure by Province 1,628 2,255 3,404 3,291 79%
Capital Expenditure 1,463 2,070 3,184 3,087
Recurrent Expenditure 165 185 220 204
Source: MOF
92 Public Expenditure Review and Integrated Fiduciary Assessment
Irrigation - Funding of Capital Works
13.32 The national irrigation system covers some 80 percent o f the 7 million ha o f
cultivated land, and i s used mostly for rice culture. However, the area actually irrigated i s
reported to be only 50-60 percent o f designed capacity because schemes have not been
completed or maintained (MARD 2004). In large schemes, coverage i s higher but
supplementary pumping i s used to overcome water distribution problems with resulting
increase in costs. In smaller schemes, actual utilization i s as low as 25-30 percent. A
large majority o f irrigation systems remain incomplete. Partly, this i s because financing
and construction for head-works and main canals are responsibilities o f MARD while
financing and construction for distribution are responsibilities o f the provinces, with
poorer provinces often lacking funds to complete the systems started on MARD’s budget.
MARD and the provinces maintain design and construction companies having different
competencies and implementation capacities leading to an implementation imbalance.
Further, funding of the distribution system i s based on norms (e.g. per meter o f canal o f
standard specification) which may be insufficient to complete the works, or a mismatch
between spending and design norms. Government recognizes the importance o f this issue
and i s giving priority to system upgrading and completion.
13.33 Although state funding o f capital works i s claimed to be insufficient, as in the
transport sector (see Chapter ll), there i s evidence to suggest that public spending
accrued in the transport sector may have been significantly higher than the cash
expenditure figures from the Treasury might be interpreted to suggest.
13.34 The ineffective management of these works i s a serious problem both at the
central and provincial levels. Taking into account only works completed and transferred
to users, MARD’s accumulated debt to irrigation construction companies alone increased
from VND22 billion in 2000 to VND1,241 billion in June 2004-a quarter o f the annual
public investment in agriculture. The figure does not include more than VND 260 billion
o f debts resolved in early 2004 with the issuance o f Government Bonds. Clearly, this
badly affects not only the construction companies and their laborers but also related
sectors, e.g. banks and suppliers. One source o f the problem i s the practice o f making
investment decisions not in conformity with approved master plans and actual prioritized
needs and o f undertaking works without completing investment procedures, normal
documentation or securing necessary funding. The lack o f effective supervision also
represents a part of the problem. Commitment controls w i l l be greatly strengthened
across Government in the coming years, as part o f the implementation o f the new
Treasury and Budget Management Information System (TABMIS). In the meantime,
urgent steps need to be taken to eliminate the stock o f payment arrears and to eliminate
future over-commitment.
Table 13.11: Vietnam: MARD’s Debts to Irrigation Construction Companies (VND bn)’.
2000 2001 2002 2003 June 2004
Debts 22 229 484 282 223
Accumulated debts 23 252 736 1,018 1,241
Note: I. including projects and works completed and transferred to users
Source: MARD
Agriculture and Rural Development Sector Expenditure 93
Zrrigation - Funding of Operation and Maintenance
13.35 While cost recovery from users i s in theory expected fully to fund O&M, in
practice this i s not occurring: subsidies from Government’s budget are s t i l l a major
source of funds for O&M. These subsidies, however, are insufficient to cover the deficit
and MARD estimates that the State Budget allocation made to maintain existing
irrigation infrastructure i s only some 60 percent o f the amount needed. Funding o f O&M
of the distribution system from the headworks to the village i s a provincial responsibility
under Ordinance 32 (April 2001) and Decree 143-CP (November 2003) on Exploitation
and Protection o f Irrigation Schemes. Decree 143-CP obliges the provinces to set service
fees at a level to cover normal operating expenses o f their Irrigation Management
Companies (IMCs). The Decree allows for exemption or reduction o f fees in localities of
high poverty incidence or those affected b y disasters. However, i t does not specify a
mechanism b y which such exemptions should be financed. Consequently, these important
measures are s t i l l to be implemented by provincial People’s Councils and related central
agencies.
13.36 Two other problems are the low level of fees actually set and the low collection
rates. The fees are set b y the Provincial Peoples’ Committee with the approval o f the
Provincial Peoples’ Council and are usually set at a compromise level below the IMC’s
actual costs. Collection rates are 80-90 percent in low-cost gravity schemes (Mekong
Delta), and 50-70 percent in high-cost pump schemes (Red River Delta) and in high
poverty localities. As a result, available funds are meeting only about half o f the
estimated necessary annual maintenance, with budget focused on emergency repairs
rather than on a more cost-effective program o f routine maintenance.
13.37 Reform i s conceptually simple but difficult to implement. Raising fees to a level
that fully covers actual costs and ensuring collection o f 100 percent o f fees due would be
impractical, at least in the short to medium term on both poverty grounds and on grounds
that i t would force users to pay for major system inefficiency. Three steps towards a long-
term goal o f full cost recovery would be: (i) correct accounting o f real costs o f O&M to
quantify the deficit; (ii) full funding o f this deficit through increasing the O&M share of
public expenditure in irrigation, on an agreed schedule that would phase out the subsidy
as efficiency o f service delivery improves; and (iii) make payments from Government
directly to poor communes to pay irrigation fees to the IMCs rather than exempt poor
communes from fees. The present system o f paying the subsidy to the provinces does not
ensure the subsidy i s passed on to the IMC. This results in service deterioration and
impacts on I M C efficiency. A fourth step would be to introduce full cost recovery
packaged with rehabilitation and improved service delivery. Pilots in Vietnam have
shown that Participatory Irrigation Management (PIM) i s resisted b y farmers where the
section being turned over to them i s in poor condition (Tiep, Nguyen Xuan, 2002, and
Svendsen, 2003).
Zrrigation - Poverty Impact of Expenditure
13.38 Expenditure on irrigation has had the greatest impact on rural poverty where i t has
increased incomes and reduced risk, particularly in household food supply. Until recently,
Government has concentrated irrigation expenditures in the large river deltas and
94 Public Expenditure Review and Integrated Fiduciary Assessment
northern Central Vietnam where most o f the poor were, and s t i l l are, concentrated.
However, because o f very small farm sizes there i s a limit to the amount o f poverty
reduction that can be achieved by irrigation alone. There i s also a limit to poverty
reduction in producing rice, particularly where increased exports from Vietnam depress
the prices received. The next phase of poverty reduction in those areas, as given in the
Plan, w i l l be through expenditure on irrigation system efficiency and on crop
diversification which require modification o f the system to support higher value crops
(particularly perennials), as well as in promoting off-farm work opportunities.
13.39 At present, MARD and provincial authorities often do not carry out
comprehensive financial and economic analysis o f irrigation projects prior to and,
especially, after investment to guide decision making and lesson learning. Irrigation
investments are primarily designed for a single commodity-based cropping pattern,
usually rice. While donor-financed projects typically conduct the investment analyses,
they are made on the basis o f consultants’ works and remain external to MARD’s
investment choices and evaluation process. I f Vietnam i s to develop an internationally
competitive agriculture, investment in irrigation infrastructure w i l l need to be viable and
capable o f responding to changing cropping patterns. Scheme viability i s an important
determinant o f net farm income and poverty reduction.
Irrigation - Scheme Size
13.40 Government i s now prioritizing irrigation investment resources into the
mountainous and highland regions where poverty incidence and household food deficits
are highest. Schemes in these areas are much smaller. Scheme size i s an important
determinant o f investment viability and poverty impact since smaller schemes are more
likely to be completed b y poor provinces than larger ones. However, MARD reports
lower utilization o f design capacity in small schemes (25 - 30 percent), compared to
higher utilization rates in larger schemes. Very small schemes o f 100 ha or less are
promoted by NGOs for poverty reduction. Oxfam Hong Kong, for instance, evaluated the
impact o f six o f these in two communes in Ha Tinh Province o f Central Vietnam from
among 25 i t supports in the country (Oxfam, “Impact o f Small-scale Irrigation Systems
in Two Communes in H a Tinh Province”, 2002). These have significant potential but are
high-cost, partly due to their pilot nature. They are also risky because o f characteristics
accentuated in small schemes: vulnerability to local floods, poor design, and weak
support from local officials, high technical assistance and supervision costs, among
others. An important step forward would be to establish and maintain an integrated
technical and economic data base that would allow determinants o f performance to be
systematically monitored and evaluated, inclusive o f small schemes.
Research
13.41 Public expenditure on agricultural and related research i s difficult to track because
the budget classification does not show research and extension as separate items but
includes them within program categories such as cultivation and breeding. Funding i s
also through a large number of institutes and agencies and i s managed b y different parts
o f government, such as MOST, MARD, MONRE, and the provinces, making
consolidated tracking difficult. MARD data show that research funding through MARD
Agriculture and Rural Development Sector Expenditure 95
and MOST rose from VND 150.5 billion to VND 197.5 billion over 2000-2003 (Table
13.12), focusing on seed production (42 percent), husbandryheterinary services, forestry
and irrigation (about 4 percent each). However, the share o f research in total sector
expenditure has remained fairly constant at 2.0 - 2.5 percent, indicating no real change in
priority at the aggregate research level. This i s much lower than that for Thailand (10
percent) and China (6 percent) during the past decade (Fan and Pardey 1998). Funding
through MARD accounts for a consistent share o f about 85 percent, with MOST
providing the balance for related environmental research. The provinces also receive
funds for allocating to the research institutions for local priorities.
Table 13.12: Public Expenditure on Agricultural Research from Central Sources - 2000 -
2003 (VND bn)
Source of Funds 2000 2001 2002 2003
MUD 122.7 142.7 148.3 175.6
MOST 27.8 19.6 20.2 21.9
Total
Salaries %
150.5 162.3 168.5 197.5
31.5 35.8 37.5 31.0
Source: MOF (2004)
Research - Adequacy and Efficiency of Expenditures
13.42 This allocation i s unusually low and probably explains much o f the low
contribution o f productivity gains in agricultural growth. Until recently, the large number
of research organizations received budget support without much attention to output and
impact. At present, there are 25 national research institutes and 120 research centers.
Seventeen o f the institutes are in the Red River Delta and close to Hanoi, while only two
are in the Mekong River Delta; this geographic distribution o f public research resources
i s poorly aligned with the potential o f agriculture in both places. The research system has
been assessed b y joint national and international experts and Government i s assessing
measures aimed at rationalizing the system.
13.43 However, major reforms are already being introduced in research management
and expenditure with encouraging results. Among these are: (i) research allocations are
now made according to priority programmatic areas; (ii) research funding i s increasingly
based on a bid system whereby interested and qualified research entities submit
competitive bids for parts o f the research program; (iii) a declining share o f funds are
retained for entities unable to win bids while a strategy for their future i s developed, and
(iv) an increasing diversity in bid winners. This approach provides a suitable framework
for integrating private research suppliers into the national research system allowing wider
coverage and increasing research leverage o f the State Budget.
13.44 However, the research funding process appears to need attention. At present, there
i s generally no annual budget rollover resulting in difficulties in research implementation
where a longer time frame i s needed. MARD internal review processes including
reconciliation o f accounts has also taken time and resulted in delayed release o f approved
funds. Uncertainty o f timing of access to funds and no rollover discourages long term
research projects which may have a higher payoff.
96 Public Expenditure Review and Integrated Fiduciary Assessment
Extension
13.45 Most extension services are provided and funded b y the provinces. MARD’s
annual budget for extension i s about VND 68 billion (Table 13.13), part of which i s used
to fund additional services under national programs and part to support general extension
policy and training. Sixty-four provinces spend about VND 117 billion in extension
service delivery but in a wide range, mostly accounted for b y salaries and leaving little
for actual professional field work. In addition, beneficiaries contribute about VND 17
billion per year.
Table 13.13: Funding of Agricultural Extension (VND bn)
1999 2000 2001 2002 2003
Total State Budget 85 108 155 196 185
Centre 26 29 43 66 68
Local 59 79 112 130 117
Source: MARD
13.46 The National Agricultural Extension Center (NAEC) was established in 2002, as
part of the reorganization o f MARD under the Public Administration Reform (PAR)
program, to be the apex body in extension and an autonomous body reporting directly to
the MARD Minister. NAEC i s responsible for extension policy and support to the sub-
national level which implements extension. I t currently has only about 30 staff expected
to support provincial extension centers in 64 provinces.
13.47 Provinces average 15-20 full-time extension professionals supporting about 6-10
districts. Some 420 o f the country’s 600 districts have 5-10 extension agents who are
responsible for perhaps 14 communes o f 2,000 farmers each. Communes typically hire
part-time extension agents o f often uncertain qualification. Many extension workers are
generalists more involved with agricultural administration in the commune than in
extension and many are not qualified as professional extension specialists. This may also
partly explain the reluctance o f communes to allocate funds to more extension. Extension
services are also delivered b y SOEs, universities, domestic and foreign processors and
input suppliers, and NGOs.
13.48 Extension in Vietnam i s considered a part o f the general socialization and
community education program at the local level. At the commune level and below, the
socialization program i s carried out by the mass organizations, e.g. Women’s Union,
Farmers’ Association, and Youth Union. As a result, the professional extension service
does not maintain an apparatus in the commune. Districts deliver extension services to
the mass organizations and are supported b y the Provincial Extension Centers and b y
MARD’s NAEC. MARD reports that there are also 46,272 voluntary agricultural
extension clubs in the country. This i s a low-cost approach and may explain Vietnam’s
strong performance in agriculture in spite o f apparently incomplete extension service.
This approach was successful during the period o f factor accumulation and where top-
down promotion o f the commodity plan was the message. However, to raise Vietnam’s
agricultural economy to the level o f international competitiveness in an open market
economy, a better balance of professional support to the existing service providers in the
communes seems required.
Azriculture and Rural Development Sector Expenditure 97
13.49 The two key issues in upgrading extension w i l l be the level o f funding Vietnam
can afford for public extension provision and the attitude o f the provinces and the
communes towards spending more. To keep expenditure manageable, i t w i l l be important
to ensure that continued use i s made o f the many existing extension providers and further
encouragement i s given to emerging new providers-particularly from the private sector
and in high-return activities such as animal production where user fees are practicable.
Increased funding o f the operations o f the existing district professionals should receive
priority over increased staffing. Norms for operations, particularly travel, are too low to
enable the extension effort to be satisfactory. Vietnam i s fortunate in not having to
maintain an overstaffed, low-productivity extension service. Provinces and communes
are now able to increase their allocations to extension, but are apparently reluctant to do
so. In addition, there i s little flexibility in the communes’ budgets particularly in poor
regions. The Participatory Poverty Assessments carried out for the 2003 Poverty Report
indicated widespread dissatisfaction with extension provision. The causes should be
understood and addressed as an integral part o f extension reform.
Forestry
13.50 Spending on forestry and related services (VND 678 million, Table 13.14) i s
second to, but only 15 percent of, irrigation expenditure in terms o f share o f sector
budget. The largest part o f the forestry budget goes to that component o f the Five M i l l i o n
Hectares Afforestation Program (SMHAP) with the purpose o f providing jobs and
incomes while reforesting cleared lands o f low agricultural potential in hilly locations
throughout the country. I t i s categorized as a National Target Program (NTP) in which
MARD plays a major role. The program i s implemented b y the SFEs in partnership with
other organizations such as the VBARD, the Policy Bank and smallholders. The project
i s very large and diversified with a notional 12 year cost through 2010 shown in the
Project Document of VND 8,645 billion from the State Budget and VND 28,975 billion
from State Investment Fund (among other sources, not identified) as subsidized credit.
The program seeks to protect 2 million ha o f existing forests, plant 2 million ha o f
production forest and establish 1 million ha o f industrial and fruit trees.
Table 13.14: Public Expenditure on Forestry and Related Services (VND bn)
1999 2000 2001 2002
State Budget 444 546 576 678-
Central Budget 100 99 113 150
Local Budget 343 447 463 528
Source: MOF (2004)
13.51 The 5MHAP has contributed to the raising o f forestry coverage from 33.2 percent
in 1999 to 36.1 in 2002. Together with other production support programs, i t has
supported the establishment o f agriculture and wood product processing plants, sugar
mills, rubber and tea processing plants, as well as many other non-forestry operations,
including infrastructure. While this may be an effective way to create jobs and industries
in the countryside, i t i s not clear whether the economics o f the investments are sound.
The design o f the project lacks sound market analysis and consultation with farmers and
other local stakeholders. In addition, land titling and planning has not been effective. N o
analysis o f the costs and benefits o f the sub-projects i s required to obtain financing,
98 Public Expenditure Review and Integrated Fiduciary Assessment
Expected benefits are expressed as commodity targets. Further, the project documentation
identifies State enterprises as being the main beneficiaries o f the program. Given the
difficulties with the SOEs and uncertainties about their future, this seems contrary to the
SOE reform program. In addition, the Project i s a framework program setting out the
flexible principles under which hundreds o f institutions and locally identified sub-
projects operate. This makes project management, financial control, expenditure tracking
and impact evaluation unusually difficult.
RECOMMENDATIONS
Sector budget share
13.52 The above review suggests that although funding in the sector has been improved,
agriculture in Vietnam may s t i l l be under-funded with respect to achieving policy goals
set for the sector. Progress in poverty reduction and in delivery o f sector outcomes i s
indicative o f the important contribution that public expenditure has made to sector
development. However, the impact evaluation o f current sector spending, the sector’s
absorptive capacity and effectiveness o f resource use all need to be improved. Thus,
strengthening capacity in public expenditure management w i l l have to be addressed as a
matter o f high priority and urgency as fiscal decentralization pushes spending
management down to local levels where capacity i s an even greater problem.
Intra-sectoral allocation
13.53 There i s clearly a need for reprioritization o f spending within the sector before a
larger budget share for the sector could be considered. Most important are the following:
Balance between capital and recurrent cost expenditures, particularly in irrigation
needs to be further adjusted with more priority given to catching up with deferred
maintenance, system completion, rehabilitation, and adjusting irrigation systems
to support diversified crops.
Research is under-resourced to support a modern and diversified agriculture
sensitive to open markets and rapid technology change. The number o f research
institutions should be reviewed to concentrate resources and some expenditure
autonomy should be devolved to remove bias against long-term research projects
and avoid interruptions to work at the end on the financial year.
Extension is under-funded. Central support to provincial extension services i s
inadequate and needs to be increased if i t i s to be credible. Provincial extension
funding i s also insufficient but provinces need to be convinced that moving funds
out o f other activities w i l l have a higher payoff in extension. Also extension
services need to be organized on commodity-specific basis. Creation o f an
insupportably large extension service should be avoided.
Forestry funding seems adequate. However, this i s difficult to ascertain because
o f the activity under the Five M i l l i o n Hectare Afforestation Project.
Reduce the burden of SOEs on public expenditure. State enterprises are a net
recipient o f public funds yet many o f their activities can be provided by the
Agriculture and Rural Development Sector Expenditure 99
private sector. Improved SOE financial management i s likely to offer the largest
source o f Budget savings for reallocation within the sector. A reduction in the
state transfers to those SOEs making losses would provide more than sufficient
savings for use in research, extension, and irrigation maintenance.
Expenditure management in the sector
13.54 Public expenditure management in the sector needs to be strengthened before the
sector’s budget share should be increased:
Require stronger project appraisal including financial and economic analysis, as
a condition for funding o f major investments in the agriculture sector. This w i l l
not only improve design o f programs but w i l l also build up a basis for policy
formulation and claiming a greater share o f the Budget. At present, evaluation
documents are largely unbalanced between subjective narrative and in-depth and
evidence-based analysis.
Establish outcome and impact analysis o f public spending as a routine activity
and as a necessary condition for continued program funding. Current practice
measures program performance b y funds disbursed, processing plants built, and
inputs distributed. There i s little disaggregated information on impact o f programs
making i t hard to distinguish program impact from general economic
performance.
Establish much stronger commitment controls in agricultural sector and conduct a
rapid retroactive assessment o f all major projects approved in the agricultural
sector to determine that budgets exist for their implementation. Implement and
plan to clear the remaining stock o f payment arrears and take measures to prevent
them from reoccurring.
Adjust institutional roles and processes to support the PAR changes. For the
central agencies, less attention i s needed to preparation o f commodity targets and
more attention i s needed to regulation, quality standardization, market
information systems, monitoring and impact evaluation. For local agencies, urgent
attention i s needed to establish modern financial management and MIS, and
capacity building.
13.55 I t w i l l be valuable for MARD to maintain the momentum from participation in the
2004 PER-FA and start work on piloting a Medium-Term Expenditure Framework
(MTEF) in the sector. Given the sector’s complexity and existing capacity, the
agricultural/rural MTEF would evolve from the existing planning and budgeting system
but ensuring more transparent links between targets expressed in the Government
development strategies and CPRGS, performance indicators and available financial
resources in a rolling three-year period. This w i l l require even closer cooperation and
collaboration not only among MARD’s departments (particularly DOF, DOP and
NIAPP) but also among MARD, M O F and MPI and substantial amount o f technical
assistance (available through the Vietnam Public Financial Management Reform Project).

14. NATIONAL TARGET PROGRAM EXPENDITURE
INTRODUCTION AND OVERVIEW
14.1 National Target Programs (NTPs) play an important role in decentralized systems
o f Government. These conditional grants are a major tool b y which central government
can support and influence the delivery o f key national targets at local level where
mainstream programs have been delegated or devolved to lower tiers. Given that poverty
reduction i s a national target, they often also play a key role in tackling pockets o f
extreme poverty-supporting poverty reduction or the achievement o f other national
goals in deprived and remote areas that would otherwise be left behind. At the same time,
they pose major challenges, in terms o f management, coordination, fiduciary control and
monitoring; and their value for money needs to be weighed against the alternatives such
as channeling more funding through mainstream programs and creating stronger
incentives and requirements for local governments to allocate such funding for the
achievement o f national targets.
14.2 In the early 199Os, in addition to mainstream spending, Vietnam established a
number o f NTPs to support local governments and sector ministries to promote economic
and social development activities. Since 2001, these programs, including the Hunger
Eradication and Poverty Reduction and Job Creation Program (HEPR-JC), Program 135
(P135) and programs in the areas o f Health, Population and Family Planning, Water,
Education and Culture, have grown into an important instrument for economic
development and poverty reduction31. NTPs have contributed in important ways to the
achievement o f key development and poverty alleviation goals. Spending on NTPs has
grown rapidly. However, challenges remain including improving the focus and
management framework o f NTPs, improving procurement processes and practices and
tightening NTP monitoring and reporting.
14.3 This chapter assesses NTP expenditure. The chapter builds on Government
studies, as well as on an earlier UNDP-funded evaluation o f the HEPR-JC program and
Program 135. The first section presents a review o f the management and regulation o f
NTPs. The second section examines the financing o f NTPs and looks at the share o f
NTPs in total state expenditure and the share o f different programs in total NTP
spending. The third section examines fiduciary issues within HEPR and Program 135.
The fourth section reviews the impact and effectiveness o f NTPs. The final section
presents recommendations.
31 Program 135 i s not formally a National Target Program, but i s nonetheless discussed in this Chapter.
101
102 Public Eupendirirre Relieu. and Integrated Fidiiciaq Assessment
REGULATION AND MANAGEMENT OF NTPS
Regulatory framework
14.4 The current regulatory framework for NTPs i s contained primarily in two
decisions: Decision 42/2002/QD-Ttg, dated 19 March 2002, and Decision 7 1/2001/QD-
TTg o f the Prime Minister, dated 4 May 2001, concerning the period 2001-2005.
Decision 42 sets out how NTPs should be managed and executed.
14.5 The philosophy o f NTPs in Vietnam, and especially the two rural poverty
programs (HEPR and P135), i s that implementation should be done at the lowest level of
government possible. However, this has to be reconciled with the fact that the purpose o f
NTPs i s to ensure the local delivery o f national goals. National targets are established as
part o f the national planning process and determined b y sector ministries and provinces in
consultation with the Ministry o f Planning and Investment and the Ministry o f Finance.
Overall management arrangements
14.6 Decision 42 states that MPI i s the lead agency in charge o f overall management
and supervision o f NTPs, consolidating overall budget requests for these programs, while
MOF i s in charge o f consolidating recurrent budget funding requests, disbursing and
monitoring funds.
14.7 The primary role o f MPI i s to synthesize sector ministry submissions and take
responsibility for securing government approval for NTPs, based on 10 year socio-
economic development strategies, 5 year plans and proposals from ministries and local
governments. MPI i s in charge o f proposing overall funding levels for capital and
recurrent budgets o f NTPs. MPI also leads on drafting policies on management and
supervision o f NTPs after discussions with relevant ministries and local authorities. MF'I
i s also charged with evaluating the effectiveness o f NTPs.
14.8 MOF, in coordination with MPZ, proposes the total amount of recurrent budget to
be allocated to each program, as well as providing funds to relevant ministries, other
government agencies and provincial governments, once the decision has been ratified b y
the Prime Minister and the National Assembly. MOF i s also responsible for examining
and supervising the distribution o f funds.
14.9 The sector ministries are responsible for specifying the structure and content o f
NTPs and subsidiary projects under their NTPs; proposing goals, duties and funding for
the NTPs; preparing funding allocation plans for target projects; and managing the
programs, usually through a Management Board for the NTP to help the respective
minister to supervise the implementation o f the program.
14.10 Provincial authorities are responsible for managing resources as well as operating
and coordinating the implementation o f NTPs in the province/city, usually through a
provincial/city-level NTP Steering Committee. The provincial Department o f Planning
and Investment i s a permanent member o f the steering committee, with other members
coming from concerned departments and offices in the province/city. The provincial
People's Committee President i s responsible for the coordination and direction o f NTPs
National Target Program Expenditure 103
and other provincial projects in the province/city, and i s legally responsible for fiduciary
management, supervision and reporting on NTP projects.
14.11 Program management boards at province and district levels delegate various
organizations to manage the different components o f the NTPs. For example, in general,
the component o f training for the poor under HEPR i s delegated to DOLISA;
resettlement works are delegated to the Resettlement Department; and agricultural
extension i s delegated to the Department o f Agriculture and Rural Development
(DARD). Despite this allocation o f overall responsibility, many different organizations
and people w i l l s t i l l be involved in discussing and agreeing a project at the local (district
or commune) level. This poses challenges in terms o f coordination in cross-cutting areas.
14.12 To date, the decentralization o f responsibility for management and
implementation o f investment has been limited. Many provinces appear not to have
devolved implementation to communes despite official encouragement to do so and
hence districts s t i l l retain control. Most projects are deregulated from central level to
provincial and district level. For example, under Program 135 after five years in
operation, only 385 communes in 20 provinces are authorized to be owners o f projects,
out o f 2,374 communes in total targeted under the program32.
14.13 In theory, provinces are supposed to delegate responsibility to districts and
communes but, in practice, provinces retain control over large elements o f the supposedly
decentralized programs. Even in Program 135, which has a very explicit decentralization
philosophy, most provinces surveyed under the UNDP study on the HEPR-JC Program
and Program 135 (e.g. Son La, SOC Trang and Quang Ngai) have decentralized program
implementation only to the district level and had only a few pilot projects in some
communes where the province felt there was sufficient management capacity. Hence, the
role o f the commune in managing projects was found to be s t i l l very limited.
14.14 Another example i s with the NTP on clean water and rural sanitation where the
principle i s to construct plans at local level. In theory, what happens i s that people are
informed about the finance available from different sources and local contributions are
then determined. Proposals are then made to districtkommune authorities and a water
committee i s appointed at commune level, which draws up plans and devises technical
solutions (gravity-fed systems etc). Donors and state authorities w i l l revise the plan
(although i t i s risky to undermine and leave local supervisors disempowered). In theory
therefore the investment owner should be the beneficiary o f the project, but in practice
limited funding means that all commune level plans cannot be undertaken and provincial
authorities decide which communes w i l l benefit and how funding w i l l be allocated across
communes.
14.15 Participation b y local people in NTP decision-making and supervision sometimes
falls short o f the principles espoused in guiding documents. Although in some cases (e.g.
in Quang Ngai under Program 135), households do actively participate in deciding on the
contents of NTPs, types of projects etc, in most cases, these decisions are taken b y local
32 Report on Assessment of the Implementation o f Financial Regulations for NTPs and Large-Scale Projects and
Programs Funded by the Central Budget in the period of 2001-2004. Financial and Monetary Department, MPI-July
2004.
104 Public Expenditure Review and Integrated Fiduciary Assessment
officials with little direct consultation and involvement o f the beneficiaries of these
projects.
14.16 In theory, all provinces report to the center on the operation of the NTP within the
province. However, the management office for HEPR-JC in MOLISA, for example,
reports that they only get a fraction o f the reports required b y law from provinces and
these are often at a very high level o f generality and so they are not very useful in
monitoring the program. Ministries generally were not able to report exactly how much
money i s being spent by provinces on the objectives o f the NTP. In practice, very heavy
reliance i s placed on provinces to manage funds effectively and efficiently.
NTP expenditure management
14.17 Until 2000, the responsible agencies at the center (MPI, MOF and sector
ministries) held a key role in deciding expenditures on an NTP. Once the budget was
agreed, the nominated responsible ministry was given figures for total expenditures and
objectives, along with relevant monitoring requirements, and was assigned the budget for
projects that the agency directly administers. The ministry or agency was then
responsible for developing and passing the budget allocation plan for the NTP to central
and local agencies, including allocation for each o f the individual projects. The funding
allocation for constituent parts o f the NTP was based on the proposals o f the program
coordinating agency and provincial budget plans submitted to MPI and MOF.
14.18 After 2000, Decision 38/2000/QD-TTg-CP altered the way funds for NTPs are
managed. The most fundamental change i s that the authority over the planning, allocation
and management o f each project has been transferred from central agencies to provinces.
During the period o f 2001-2003, provinces were allocated a block o f funding for all
NTPs and were free to allocate funding between individual NTPs. The aim was to allow
provinces to respond to local needs, and to integrate activities o f various programs to
maximize their impacts. This mechanism, however, made i t difficult for the program
management line ministries to monitor provincial allocation o f funding to their programs
and the extent to which the program targets were being achieved. To remedy this, since
2004, the Government has allocated targets and funding for each NTP to provinces.
However, provinces s t i l l retain the authority to allocate funds for specific projects within
each NTP, under the guidance o f the relevant managing agency and in accordance with
M O F and MPI circulars and other regulations on spending. Under the new mechanism,
provinces are mandated to allocate for each NTP at minimum the funding level specified
by the Government but are free to provide additional funding from their own resources.
14.19 With regard to fund transfers, in the past NTP funds for implementation at the
local level were separately authorized b y NTP managers. However, from 2001 central
government funds for local implementation are part of the annual total budget transfer to
the provinces in the form o f targeted transfers.
14.20 Budget norms for allocating funding under NTPs are often unclear or out-of-date
(e.g. MOF Circular 33/1999/TT-BTC, dated 29 March 1999, outlining the guideline for
management and disbursement o f recurrent funds o f HEPR and MPI Circular
06/2000/TT-BKH, dated 27 June 2000).
National Target Program Expenditure 105
14.21 In theory, provinces are supposed to pool central, local, mass organization, state
enterprise and household-level contributions to achieve NTP goals. In practice, the extent
to which local agencies are able to raise funds varies greatly, with the richer provinces in
general raising more local contributions than poorer ones. The recent Tracking Study,
funded b y UNDP, clearly shows the extent o f variation in non-central contributions to
HEPR and Program 135 across provinces. The bulk o f mass organization and individual
contributions are provided for housing poor households and building infrastructure and
that other sub-programs receive very few individual or collective contributions.
FINANCING OF NATIONAL TARGET PROGRAMS
14.22 Table 14.1 and Figure 14.1 below respectively show the share o f NTPs in total
State Budget expenditure and the share o f individual NTPs out o f total NTP spending.
Table 14.1 shows that the share o f NTP spending has risen dramatically, albeit from a
low starting point: from 2.0 per cent to 3.4 per cent o f total State Budget between 1998
and 2002. This represents a 3-fold increase in funding in absolute terms, reflecting not
only the increasing commitment to target programs but also the very rapid growth o f the
total State Budget.
Table 14.1: Share of NTP Spending in Total State Spending - 1998-2002
Year Total NTP Total State NTP/Total
Expenditure Expenditure Expenditure
(VND bn) (VND bn) (%o)
1999 2,484 84,817 2.9
2000 2,988 103,151 2.9
200 1 3,405 119,403 2.9
2002 4,575 135,490 3.4
14.23 Figure 14.1, however, shows that within this overall increasing trend in financing
for NTPs, individual NTPs have fared differently over the PER-IFA period. Interestingly,
given their relatively high political profile, the shares o f both HEPR-JC and Education
have declined as a proportion o f total NTP expenditure. HEPR-JC has fallen from 16.2
percent to 13.7 percent over this period and Education has declined from 22.5 percent to
15 percent o f total NTP expenditure. However, in part this reflects the removal o f the
infrastructure component from these programs and i t s transfer to Program 135.
Figure 14.1: Shares of Individual Programs in Total NTP Expenditure - 1999-2003
35.0
30.0
25.0
20.0
15.0
10.0
5.0
0.0 I HEPR P135 Health Water Educ Culture
0 1999
I 2 0 0 0
0 2001
0 2002
2003
106 Public Exvenditure Review and Integrated Fiduciarv Assessment
14.24 Health, Water and Culture have had fairly constant shares at 10 percent, 5 percent
and 2-3 percent respectively between 1999 and 2003. The key gainer in terms of share
has been Program 135, whose share has nearly doubled from just over 15 percent to over
30 percent of the total in 2003. Preliminary data for 2004 suggest that education’s share
may have risen markedly in 2004, at the expense o f the other programs, with Program
135’s share dropping back to under 30 percent o f the total.
Funding of Individual Programs
14.25 VND 3,780 billion was allocated to HEPR-JC for the first four years o f the
current planning period, 2001-2005. In addition, many donors have financed projects on
hunger alleviation and poverty reduction in the northern mountainous provinces, central
provinces and elsewhere, providing additional resources to meet funding requirements.
Donor funds amount to around US$320 million b y 2005 (equivalent to VND 5,000
billion).
14.26 Under P135, around VND 1,200 billion was allocated in 2003 to 2,342 communes
and VND 4,910 billion in total over the 2001-04 period. Central funds are allocated
downwards to districts and on to the intended recipients-communes-on the basis of a
fixed sum allocated to provinces b y central government per qualifying commune (VND
500 million). The method o f allocating funds-a fixed sum per qualifying commune-
has the clear benefit o f clarity and transparency. During annual implementation o f P135,
provinces and districts may now allocate higher sums than VND 500 million to
communes and lower amounts to others, based on their assessment of the use that can be
made o f funds. However, they are required to ensure that, during the life of the program,
each commune w i l l receive an annual average o f VND 500 million. As with other NTPs,
provinces with larger central funding allocations tend to also make a larger amount o f
local (provincial) resources available.
14.27 The investment o f VND 1,650 billion from 2001-04 in the Health NTP compares
with a figure of VND 2,155 billion in state funds for the entire period 1990-2000. The
central government’s allocation for this program has risen by 46 per cent in nominal
terms and b y around a quarter in real terms, between 2004 and 2000 indicating the
priority the Government attaches to the program. Expenditure has risen faster due to
increased provincial contributions to the NTP. However, although health NTP
expenditure has risen in cash terms, it has declined as a percentage o f current government
health spending since 1999: total health spending increased b y 31.5 percent 2000-2002,
while NTP spending increased b y 22.1 percent. These figures exclude donor funding, for
example funding o f about VND 100 billion b y the World Bank for the malaria and
tuberculosis programs.
14.28 State Budget funding for the Water and Rural Sanitation NTP was VND 880
billion over the period 2001-04, with total funds from central government funds, local
budgets and community resources in 1999-2003 amounting to VND 4,795 billion. Over
VND 850 billion was invested in the period 1999-2003 compared to VND 318 billion
over the 1996-2000 period and, in addition, a very large sum (VND 3,000 billion) has
been committed b y international donors over this period.
National Target Program Expenditure 107
14.29 State Budget funding for the Education and Training NTP from 2001-04 was
VND 3,760 billion. This compares to VND 5,000 billion invested over the whole period
from 1991 to 2000. The budget for target education programs in 2004 (VND 1,250
billion) i s double that invested in 2000.
14.30 VND 540 billion o f state funding for the Culture NTP has been allocated between
2001-04, o f which a large proportion-VND230 billion-was allocated in 2004. Other
donors, local authorities and households were scheduled to contribute an additional
VND400 billion between 2001 and 2004. This level o f funding compares favorably to
state funds o f VND475 billion-a little over half the total-during 1994 and 2000, with
additional international funding over this period of VND300 billion (30 percent o f total)
and funding from the community o f VND215 billion (22 percent o f total).
FIDUCIARY RISK IN HEPR- JC AND PROGRAM 135
14.31 This section looks at the fiduciary risk in the Hunger Eradication and Poverty
Reduction and Job Creation Program and Program 135. However, it i s likely that some of
the lessons can be generalized to other NTPs. The discussion draws upon a UNDP-
financed Public Expenditure Tracking Study conducted in 2003-2004 which examined
fiduciary risk in these two programs across four provinces (Tuyen Quang, Som La,
Quang Ngai and SOC Trang) as well as an in-depth case study o f NTPs in SOC Trang
province. These studies, in turn, drew on an earlier Government study o f NTP budget
allocation.
Findings of Public Expenditure Tracking Study
14.32 The Tracking Study found that funds for both Program 135 and HEPR could be
traced from the center to provinces and then down from provinces to districts with
relatively small variations. Tracking Program 135 was particularly simple in this regard.
In terms o f capital program expenditure managed and disbursed through the Treasury
system, funding was satisfactorily based on recorded progress in project implementation
(signed certificates are a requirement to show that projects had been completed although,
in some circumstances, projects received acquittals from the financial agency based on
estimates and not on actual funding requirements).
14.33 However, the Study echoed the findings o f the earlier Budget Allocation study
that funds allocated from the center do not always bear a clear relationship to plans made
and funds requested at the level o f the province and lower levels. For example,
allocations for job creation and training do not always correspond to local needs. In
contrast, the allocation o f central Program 135 funds i s clearer and more transparent and
this has enhanced the visibility and ownership o f the program at local level. Under
Program 135, state central funds are allocated to districts and on to communes for local
infrastructure projects based on a norm (currently VND 500 million per commune).
Where there are capacity constraints, funds are temporarily re-directed to other
communes.
14.34 Budgetary allocations, funding flows and management/ implementation
responsibilities in the HEPR program vary widely across provinces as well as in a
particular province over time. The variation in HEPR allocations may reflect a judgment
at the center on local funding priorities from one year to the next. Rather than being
108 Public Expenditure Review and InteRrated Fiduciary Assessment
based on norms, the onward distribution o f central funding allocations to districts seems
to be determined by provinces themselves based on their own plans and funding
priorities. The study found that provinces use different physical norms to allocate
resources across sub-programs and these can be unnecessarily complicated and difficult
to understand.
14.35 While NTP capital investment i s managed and tracked clearly and transparently
b y the State Treasury system, recurrent funding cannot be separately monitored since i t i s
merged with mainstream budgets and not “tagged” with a program or project code. The
study recommended that spending budgets under the NTPs should be clearly identified
and given separate budget codes, so the funding here can be identified and tracked. Some
contributions from enterprises and foreign donors are not reflected in district and
commune books o f account and therefore their use i s not ‘integrated’ into state-funded
commune efforts. Integration o f donor funds into state budgets at provincial level would
have the benefit o f increasing the volume o f resources available for spending on local
priorities.
14.36 The Tracking Study could not accurately quantify fiduciary risk or fund leakage
from the HEPR program, given the time and resources available. Given that there has not
been an evaluation o f HEPR for several years prior to the 2004 review, there was little
evidence to draw on regarding fiduciary risk. Nevertheless, the Study found little
evidence of any substantial leakage o f funds in Program 135 and HEPR, though i t was
not possible to assess whether spending at the projects level actually reached poor
households. Overall, NTP management and planning, based on top-down planning and
bottom-up information provision, seems to be working reasonably well. The clarity o f the
management and funding structure and the formal province to district to commune
reporting lines has helped to attract increasing funding and support from international
donors to implement NTPs.
14.37 Although the financial control system i s relatively strong, the large number o f
separate NTP sub-programs militates against clear central oversight and control over
program funds. The quality of management of expenditure i s also constrained by: (i) the
frequent changing o f staff, especially at commune level; (ii) an excessive number o f
levels o f management; and (iii) a wide variety o f different sources o f program funding,
both in cash and in kind. The flow o f funds through the Treasury system from the central
level, province level and to the district means that i t i s relatively straightforward to
control and evaluate the use o f funds and prevent leakages in the program. However,
accounting, audit and review mechanisms are inadequate - the system relies heavily on
reports that are either at a high level o f generality or else rely on paper-based controls and
one-off audit type checks, which are useful but only provide a ‘snapshot’.
14.38 The study found that the main potential for leakage i s at commune level, where
records and accounts were incomplete or not kept at this level. In situations where
communes only have the role of passively receiving benefits or households largely
receive support in kind, there i s a higher potential for leakage o f funds due to limited
involvement of local people in monitoring activity and the difficulty in holding commune
contractors to account at higher levels. The study recommended that stronger efforts
should be made to decentralize the implementation o f Program 135 and the HEPR
program to the commune level, starting with communes with stronger management and
National Target Program Expenditure 109
accounting skills. However, delegation o f implementation to the communes should not
lead to a weakening o f effective control. Hence, i t was recommended that finance s k i l l s
should be strengthened at commune level and efforts made to ensure that all funding,
including household and mass organization contributions, applied to HEPR and Program
135 components are tracked and reported.
14.40
e
e
e
e
14.39 The study found that project implementation agencies have to prepare many types
o f reports (monthly, quarterly, annually, project completion, ad hoc etc) for higher level
management and financial agencies, Treasury, and other related agencies. Whilst there
was regular checking at many levels, the lack o f delegation o f responsibility means that
checking i s generally o f a formalistic nature, and few faults are generally reported.
Furthermore, reports provided to the center are usually in summary terms, making i t hard
to monitor detailed outcomes. There i s also insufficient exchange o f experiences and
lesson-learning that could improve the effectiveness o f service delivery. A review o f
existing reports and dissemination o f lessons learnt from these reports to localities would
have significant benefits.
A number o f other factors also weaken the effectiveness o f the programs:
The level o f central funding i s insufficient to achieve sustainable impact in
infrastructure, employment or productivity in the short or even the medium term
and funds sometimes need to be advanced from the next planning year.
There i s a bias towards capital spending in these programs which can lead to
initial investments being poorly maintained and hence depreciating in value and
usefulness quite quickly; and which can place burdens on the poor to contribute to
the maintenance o f the assets created.
Timing problems exist in relation both to disbursement and funding flows. For
example, often there are unspent in-year expenditures in the sub-program relating
to production support, due to slow disbursement o f funds.
Programs are not always well coordinated at a local level leading to management
difficulties and redundant spending.
The SOC Trang Case Study
14.41 An in-depth case study o f SOC Trang province broadly echoed the findings o f the
Tracking Study. The study confirmed that whilst leakage o f NTP funds i s low, the
efficiency and effectiveness of HEPR-JC and P135 funds can be improved further.
Wasteful spending was observed in the form o f un-audited, in-kind provision o f goods
and unused or under-utilized infrastructure facilities (e.g. roads, markets or irrigation
dams without water).
14.42 The field research in SOC Trang, undertaken as part o f the 2004 PER-FA, found
that the programs were implemented appropriately and in accordance with regulations.
The field visit, however, also confirmed the weaknesses noted in previously completed
NTP Public Expenditure Tracking Survey work, including: (a) lack o f connection
between funding allocations and planning requests; (b) mixed-up accounting for recurrent
expenditures of main programs and NTPs; (c) sluggish disbursement o f funds; and (d) the
lack o f clear norms and detailed monitoring o f expenditures at central and province
1 10 Public Expenditure Review and Intezrated Fiduciary Assessment
levels. This study confirmed the finding of the earlier Tracking Study with respect to
Program 135: funding was found to be well managed in the communes visited, spending
corresponded with local needs and the allocation o f capital funds and their use was public
and transparent. There i s no finding o f loss or misappropriation o f funds in these two
steps. The management o f the HEPR-JC program i s significantly less effective, leading to
poor results. HEPR-JC involves many management agencies, uses outdated norms and
has less involvement o f local people in monitoring o f spending. The main fiduciary risks
lie in the contracting process, the potential for diversion o f funds as they pass down
through the bureaucracy and the poor nature o f upward reporting.
14.43 Analysis o f allocation and spending data in SOC Trang showed that spending on
Program 135 was 97 percent or more o f provincial allocations in 2002 and 2003 and 109
percent o f allocation in 2001. HEPR-JC was 97 percent o f allocation in 2001, but only 68
percent and 62 percent o f allocations in 2002 and 2003. The fact that HEPR outturns
were below allocation in 2003 and 2004 i s due to a number o f factors including the fact
that the design work for infrastructure construction took longer than expected, leading to
a delay in work delivery and disbursement; delays in resettlement expenditure; and poor
local planning (especially at commune level), leading to slower disbursement and low
take up o f credit allocations
14.44 Procurement practice in SOC Trang was found to be relatively weak and needs to
be improved. Selection o f local enterprises to implement projects was found to be overall
in compliance with program objectives, although there i s often little or no competition in
the selection o f contractors and collusion between contractors and NTP managers i s
therefore a risk. In SOC Trang, design consultants and some contractors were appointed at
the provincial level, but contractors are designated and contracts monitored b y district
authorities, even where construction work was funded by local contribution. The fact that
commune authorities were not able to select contractors (and so procurement was out o f
their control) may lead to poorer VFM.
14.45 Fully competitive bidding i s generally used for those NTP capital works with a
value in excess o f VND 1 billion (US$65,000). Almost all the works undertaken through
HEPR-JC and P135 are either smaller than this ceiling, or broken down into projects that
are below the VND 1 billion threshold. The lack of competition-four enterprises in the
district studied in SOC Trang won all contracts-is likely to imply higher procurement
costs than under a competitive procedure, though the administrative costs o f enforcing
competition need also to be factored in. With few, well-known contractors, there i s a risk
of collusion between enterprises and investment holders, which could lead to excessive
costs being charged. Cost estimates are also often based on outdated norms and invoices
are not always matched with actual payments.
IMPACT AND EFFECTIVENESS OF "I'PS
14.46 Overall, i t appears that NTP achievements have facilitated important changes in
rural and mountainous areas, and have been an important component in Government
efforts to reduce hunger and poverty in the poorest regions. Significant improvements in
local circumstances have taken place through NTPs. NTPs have contributed to the
achievement o f national targets at local level.
National Target Program Expenditure 11 1
14.47 Official statistics suggest that individual plan targets supported b y NTPs are being
achieved:
Targets for Program 135 in terms of access to schools, water, roads etc are being
fulfilled.
Key indicators show that there has been progress against health targets in poor
and remote regions. Overall mortality has been cut in half and there has been a
marked improvement in the population’s health status with life expectancy
increasing from 65.2 to 71.3 years between 1989 and 2002.
The Clean Water NTP i s believed to be an effective way o f mobilizing resources
to reach specially disadvantaged communes. 14 million more rural households
have access to clean water in 2003 compared to in 1998-a 4.4 percent annual
increase. This represents an increase from 32 percent to 54 percent o f the rural
population who can access clean water.
All provinces/cities have met the national standards relating to primary education
and literacy. Primary education and illiteracy reduction programs have enabled
children to go to school at scheduled ages. Efforts to promote education in
mountainous areas have resulted in over 300 ethnic boarding schools being built
in remote areas, increasing literacy rates.
Over the period 2001-04, achievements o f the Culture NTP have included: the
preservation o f relics; restoration o f traditional villages, improvement in the
quality o f the film industry and the indirect benefits such as the creation o f jobs
and incomes for local people as a result o f restored siteshillages acting as tourist
attractions.
Challenges and Weaknesses inherent in NTPs
14.48
a
a
a
a
Despite these notable achievements, important challenges for NTPs remain:
Affordability i s a key concern. While NTPs are delivering real benefits, the level
o f central funding i s probably too small for genuine step changes in infrastructure,
employment or productivity to be achieved in the medium term. In addition, the
system i s biased towards capital spending: initial investments are often poorly
maintained and so they tend to depreciate in value and usefulness quite quickly.
Targeting the poorest: As Chapter 3 discussed, and as Boxes 14.1, 14.2 and 14.3
illustrate, there i s scope to improve the targeting o f the poorest through NTPs.
This could be achieved both through improving the norms through which funds
are allocated and through the use o f better poverty data.
Capacity building i s a key challenge: communes are rarely involved in project
management and handling funds at the local level. Involving poor households and
their representatives in implementing and supervising the program may help
NTPs to generate their full benefit.
Financial documentation and recording at the commune level i s often weak.
Many local program activities are not reflected in the accounts and commune
1 12 Public Expenditure Review and Integrated Fiduciary Assessment
finance officials often do not have knowledge and training in accounting and
financing.
The regulatory and financial management systems need to be updated. Adequate
accounting, audit, reporting and review mechanisms are not in place. The system
relies heavily on formal annual reports. Monitoring, supervising and evaluating o f
the programs are not well conducted, due to the absence o f clear and adequate
regulation regarding the co-ordination o f ministries and local authorities and other
departments o f the provinces.
There are timing problems with both disbursement and funding flows. Field visits
revealed unspent in-year expenditures due to slow disbursement o f funds and
slow uptake o f loans, etc. Conversely, there i s a shortage o f funding for
infrastructure projects with funds needing to be advanced from the next planning
year and a lack o f funds for many programs such as extension schemes.
The mobilization of contributions from organizations and individuals i s very
limited. Most mass organization/individual contributions are for housing for poor
households and building infrastructure for 135-communes while other sub-
programs receive very few individual or collective contributions, limiting their
impact.
There are s t i l l pressing social, health and educational problems, particularly in
the remoter communes: child malnutrition remains a serious problem in Vietnam
affecting around a third o f children, one o f the highest rates in the region; dengue
fever, TB and HIV/AIDS are resurging, driven b y various factors including drug
resistance, migration, overcrowding and air pollution; and nearly half o f rural
households s t i l l do not have access to clean water.
Box 14.1: Effectiveness of HEPR-JC Program
I t i s difficult to separate out the impact on poverty o f HEPR funds from that o f mainstream funding,
provincial equalization transfers and other funding sources. An examination of funding data shows that
HEPR provincial allocations are not very closely correlated with recorded poverty levels. The table below
shows that the relationship between HEPR central funding and poverty rates i s not direct-for example,
over the period 2001-03, Tuyen Quang received 80 percent of the funds received by Son L a and Quang
Ngai but had only a third o f the poverty rate in 2003.
Table 14.2: Relationship between HEPR Central Funding and Poverty Rates
2001 2002 2003
Province Poverty Allocation Poverty Allocation Poverty Allocation
rate (%)' (VND mn.) rate (%)' (VND mn.) rate (%)3 (VND mn.)
Tuyen Quang 12.4 38805 8.2 34259 6.5 1447 1
Son L a 27.1 37433 22.8 24247 18.3 38750
Quang Ngai 24.6 49988 22.0 27072 18.3 274 14
SOC Trang 34.3 16870 30.2 52594 26.9 98116
Vietnam 17.2 14.4 11.6
natural and climate conditions that affect the rate with which poverty i s being reduced.
Sources: Poverty data from MOLISA, budget Allocations from MOFI and DPls, DOLISAs in different provinces.
Notes: 1 At the beginning of 2001 (with the new poverty line); 2 on December, 2001, 3 on December 31, 2002.
On the other hand, i t should be recognized that there are other factors such as levels o f development,
National Target Program Expenditure 113
Box 14.2: Impact of Program 135 on Poverty
Targets for Program 135 (P135) -in terms of access to schools, water, roads etc-are being fulfilled.
As a result o f these P135 projects, and in combination with other programs and projects, after five years
most o f the communes targeted received investment in transport, schools, power and water projects,
markets, healthcare and other facilities. Specific achievements of the Program include:
Nearly three-quarters o f the communes received at least 5 substantial projects, such as roads, power,
schools, water supply and healthcare centers;
84 percent of communes and 65 percent of local households have access to electricity; 86 percent of
communes now have primary schools and 73 percent of the communes have brick-built secondary
schools and 96 percent of the communes have healthcare centers;
Many irrigation projects have been implemented, raising the average level o f food from 290kg to
384kglpersodyear and in some areas up to 500kglpersodyear.
Overall, i t i s clear that the achievements above have facilitated some important changes in rural and
mountainous areas, and have been a material component in Government efforts regarding hunger
alleviation and poverty reduction in the poorest regions.
The fixed amount allocated per commune allows funding to be directed to relatively poor
communes-those with a population where at least 40 percent o f the population fall beneath a nationally
defined poverty line. However, the allocation mechanism-a fixed amount per participating commune-
does not take into account differences in the number o f households in participating communes, physical
geography and other local factors and difficulties. Hence, funding i s probably best seen as going some
way to meeting basic (infrastructure-related) needs. Whilst it i s open to provinces and districts to
supplement central funding and this seems to be happening to a limited extent, i t i s not clear that
additional resources are necessarily being directed to communes with larger numbers o f people or poor
households.
Box 14.3: Effectiveness of Health Target Programs
Health standards o f the country have improved and overall mortality has been reduced by half. The
NTP and its subsidiary projects (addressing malaria, goiter, TB etc) have enabled Government and
external financial assistance to be effectively channeled, resulting in a marked improvement in the
population’s health status in recent years: life expectancy at birth has risen from 65.2 years in 1989 to
7 1.3 years in 2002. The effectiveness o f the health NTP i s evaluated on the basis of specific indicators for
each sub-program. Generally, all programs have been successful in terms o f planned indicators. Some
have increased or maintained their effectiveness:
the infant immunization rate has been maintained above 90 percent, leading to the ongoing success of
polio elimination;
the malnutrition program has reduced child malnutrition by 2.2 per cent a year compared with 0.6 per
cent a year over the period in 1985-1995, although arguably this rate o f change i s still too slow (and
i s well below GDP growth), given the rising general levels of living standards.
The NTP allows resources to be directed to people at commune level, e.g. through commune health
centers. The focus in 2006-10 w i l l be on sustainability o f the program and further reducing disease
incidence. I t i s important for Government to maintain and if possible increase the funding for prevention
as opposed to curative treatments, since the former i s much more cost effective in the long-run. However,
international experience shows that most health systems find i t difficult to do this. Tightening access to
prescription drugs, strengthening prevention programs (and giving doctors financial incentives to take part
in such programs) and investing in community-level health care w i l l all help to make health care both
more affordable and more effective.
continue.. .I
1 14 Public Expenditure Review and Integrated Fiduciary Assessment
I.. .contd. Box 14.3
sustainability. For example:
However certain program targets have not been reached or have been reached without stability and
not all leprosy cases are being detected;
the proportion of food poisoning cases and food-poisoned deaths per 100,000 people s t i l l remains
high;
although this i s a difficult area, the number o f schizophrenia patients found and administered b y
communities i s below the target set by the authorities.
There are clear regional differences in terms o f program effectiveness and efficiency. Substantial
differences are observed in terms o f health indicators related to NTPs among regions, with lower levels in
northern mountainous areas and central highlands. To achieve national targets, a clear priority should be
given within NTPs to projects implemented in disadvantaged provinces and areas. However, the
government w i l l probably need to go further still given the degree o f provincial autonomy and the
difficulty o f accessing the poorest households in the most remote areas. One method of ensuring greater
focus on the needs o f the poorest households who are most vulnerable to communicable diseases etc.
would be to specify the distribution of funds to such areas/communes in an agreement between M o H and
the PPC over the use o f funds.
RECOMMENDATIONS
14.49 National Target Programs play an important role in supporting and helping to
ensure the delivery o f national priorities at local level; and in addressing pockets o f
extreme poverty. At the same time, i t i s important to recognize the funding through these
programs i s very small in comparison with main funding, and as such their impact i s
necessarily limited. Compared with conditional grant programs in other highly
decentralized countries, NTPs in Vietnam are arguably rather small as a share o f public
spending, and could probably be grown over time. At the same time, the existence o f
NTPs does not substitute for a need to ensure that main programs contribute to the
delivery o f national targets including poverty reduction.
14.50 The PER-FA recommends that NTPs should continue to exist and could possibly
grow further as a share of spending, since they are playing an important role in
supporting the local-level achievement o f national targets such as poverty reduction. The
complementary approach o f ‘bending’ mainstream funding towards the poor may well be
a more powerful tool for reducing poverty and inequality in terms o f numbers o f people
lifted out o f poverty. However, i t w i l l always be a blunt instrument unable to reach the
very poorest people in the most remote areas. In other words, both the instruments o f
‘bending’ mainstream funding and NTPs are necessary but neither alone i s sufficient.
14.51 A further argument for supporting NTPs i s that there i s evidence (of which this
analysis forms part) that NTPs are not as weak in fiduciary terms as some have claimed
and that they represent reasonable value for money. Hence, there i s a probably a case for
both more targeting o f mainstream funding to the poorest as well as, within macro
resource constraints, placing more resources in NTPs, especially ones that are relatively
effective such as Program 135, Health and Clean Water NTPs.
14.52 Other recommendations include:
National Target Program Expenditure 115
NTPs should be rationalized and better coordinated both with each other and with
main programs. Some activities could be dropped, while others could be merged
into budgetdmainstream programs o f existing line ministries and agencies.
Stronger mechanisms should be put in place to ensure that NTPs support national
targets. An agreement should be made on an annual basis-based on a longer-
term agreed strategy-between each of the responsible central agencies/
ministries in charge o f NTPs and provincial PPC chairmen that sets in detail the
objectives o f the funding, budget allocations, relevant expenditure norms outputs
and the indicators through which these outputs and outcomes w i l l be measured.
Incentive mechanisms should be set up to improve the use o f funding and
sustainability o f investment. A particular problem i s the lack o f maintenance o f
capital equipment and infrastructure. Unless sufficient O&M spending i s being
planned and allocated, capital allocations should be cut accordingly.
There should be efforts to improve the s l u l l s and capacities o f commune level
personnel. Commune staff should receive training in accounting, financial and
project management skills.
Local consultation and participation in NTP management and planning should be
increased. There i s potential for communicating information much more clearly
and regularly down to the village level, and seeking the involvement o f poor
households in monitoring projects. This i s likely to yield benefits as projects are
more relevant to the needs o f the poor.
Eliminating weaknesses in the procurement system. Issues o f uncompetitive
bidding and goods being supplied in-kind rather than in cash need to be dealt with
through reforms to the process o f procurement itself.
Improving the monitoring and reporting o f NTPs. Provinces should be obliged to
report and record community contributions and there should be greater definition
o f how fund and expenditure flow i s going to be identified (using nationally
defined coding structures) as this w i l l certainly help to make expenditure more
transparent. The program/project and source o f funding codes in the new Treasury
and Budget Management Information System should help in this regard. NTP
indicators should be improved b y making them as specific as possible and better
project evaluation using standard project evaluation and economic evaluation
techniques should be instituted.
Regular evaluation of the NTPs should be undertaken. An audit and inspection
plan should be agreed with a clear remit and approach to sampling and reporting.
A reporting body should be designated within Government to review and report to
all stakeholders on progress in disbursing funds / tracking expenditure,
There i s a need to devise mechanisms to capture and disseminate lessons from
local innovations in NTP delivery and ensure mechanisms exist to analyze and
disseminate lessons at national level and across provinces. This i s related to the
points above about improving reporting and evaluation. At the moment, the
lessons (e.g. from Tuyen Quang in Program 135) are insufficiently applied in
1 16 Public Exvenditure Review and Inteerated Fiduciarv Assessment
other areas. There i s also the potential for lessons leant by NTPs to benefit
mainstream program funding over time.