ESTIMATION RESULTS OF HOUSEHOLD WELFARE 22
I. DATA 22
II. DESCRIPTION OF VARIABLES 23
1. Dependent variable 23
2. Explanatory variables 23
III.ESTIMATION RESULTS: The household welfare generating potential of
land
25
SUMMARY OF FINDINGS
29
CHAPTER V
CONCLUSION AND POLICY IMPLICATIONS 30
I. CONCLUSION 30
II. POLICY IMPLICATIONS 30
III. SUGGESTION FOR FURTHER STUDY 31
REFERENCES 32
APPENDICES 35
iv
LIST OF TABLES
Table 1 - Major annual crops grown in Vietnam, 1995-2000 16
Table 2: Indicators of poverty in Vietnam 1990-2002 17
Table 3: Spatial dimensions of poverty and inequality in Vietnam 18
Table 4: Ethnic and gender dimensions of poverty in Vietnam 19
Table 5: Socio-characteristics of Vietnam’s poverty profile 20
Table 6: Number of household by region 22
Table 7 - Estimation of the welfare household index 27
v
LIST OF ABBREVIATIONS
CH Central Highland
CPI Consumers’ Price Index
GSO General Statistics Office of Vietnam
MKD Mekong Delta
NCC North Central Coast
NE North East
NW North West
OLS Ordinary Least Square
RRD Red River Delta
SCC South Central Coast
SE State-Owned Enterprises
VLSS 2002 Vietnam Living Standards Survey in 2002
VND Vietnamese Dong
WB World Bank
vi
CHAPTER I
INTRODUCTION
1.1 Research topic
The role of land in economic theory is extensively documented. Originally, land used by
agriculture was the main motivation for an economic treatment of land. Prominent effort of classical
economist was to established relation between land, labor and capital and its contribution to economic
growth. Later, land use did not occupy an important position in neoclassical core economic theory,
they considered it as a production factor of relatively little importance. Land as the principal source of
wealth in classical economics lost its central role in neoclassical economics (Klaus; Jeroen, 2002).
In reality, land in agricultural production still plays a key role both in term of household wealth
and mean of poverty reduction. Access to land has for many years advocated as one of the potentially
most effective approaches to rural poverty alleviation (Warriner, 1969; Thiesenheusen, 1989;
Bingswanger et al., 1995). Therefore land reforms in developing countries aimed to give the poor
opportunity to help them go out of poverty status. It was not far away from favoring low income
people, pro-poor policy in Vietnam has extensively and dramatically performed land reform over
renovation process (Doi Moi) since 1986. Vietnam started to transform its rural sector from
collectivized agricultural production to a system based on households’ initiative, a move that
culminated in the passage of the 1988 Land Law. A new law was enacted in July 1993 to deal with
factors such as threats of administrative reallocation, short duration of use rights, lack of
transferability and the inability to use land as collateral. Studies have shown that this led to significant
increases in overall rural productivity, although pre-existing differences between North and South
were not eliminated (Pingali and Xuan 1992, Tran 1998), other study suggest that land allocation
responded positively but slowly to the inefficiencies of the administrative allocation (Ravallion, M;
Walle, D.V.D). We see that there is, however, a remarkable absence of solid empirical evidence
measuring the potential of land in reducing poverty in Vietnam, particularly carefully taking into
account the specific conditions under which land is used by beneficiaries, when we well know that
these conditions are key determinants of success or failure. Aim to analyze further these effective
factors basically and systemically, the topic “Determinants of poverty reduction potential of land in
Vietnam rural” will be employed for my thesis.
1
1.2 Research objectives, scope and focus of the study
This thesis will examine determinants of poverty reduction potential of land in Vietnam rural
from Vietnam Living Standards Survey in 2002 (VLSS 2002). In the thesis, I will focus on exploring
the relationship between income and agricultural land endowment of household where household
mainly generate income from agricultural land and taking into account the specific conditions under
which land is used by beneficiaries.
The study focuses on four main works. Firstly, it is to introduce a general framework.
Secondly, it is to provide a overview of agricultural land and poverty in Vietnam. Thirdly, it is to
estimate household welfare equation, which include without and with correcting selectivity bias,.
Finally, the study provides policy implications to agricultural land.
1.3 Research questions
The thesis is to address main question:
Which determinants cause heterogeneity in marginal returns to land for households endowed with
agricultural land in Vietnam?
Sub-questions include:
Does marginal return to land endowed by household increase by area of agricututal land?
Are there any differences in welfare of household by regions?
Are there any differences in welfare of household by household characteristics, by which
ways?
Are there any differences in welfare of household by education?
1.4 Methodology
The study reviews theoretical frameworks and relevant researches of the research topic. The
research questions will be addressed by the quantitative method. Based on theoretical framework,
household welfare equation is estimated, including selectivity correction. In addition, statistical and
descriptive analyses are used as well.
2
1.5 Data
This study is based on the data from the Vietnam Living Standards Survey 2002 (VLSS 2002)
which conducted by World Bank (WB) and the General Statistic Office (GSO) of Vietnam. The
survey provides detailed information about employment, income, education, and demographic
characteristics of household members.
1.6 Structure of the thesis
The thesis is divided into five chapters:
Chapter 1: Introduction
Chapter 2: Introduces theoretical considerations, methodological framework, and a brief of relevant
literatures on returns to agricultural land.
Chapter 3: Provides an overview agricultural land and poverty in Vietnam
Chapter 4: Presents estimation results of household welfare equations.
Chapter 5: Summarizes main findings and provides policy implications based on the results
3
CHAPTER II
METHODOLOGICAL FRAMEWORK
This chapter introduces a economic framework for welfare equation. This chapter has three
sections. Section one gives a review of economic theory for welfare equation. Section two introduces
welfare equation. Section three is literature review.
I. Theoretical framework:
In this section we derive the specification of our income equation from a agrarian household
production model. We consider a situation of multiple market imperfections to investigate how these
distortions affect the economic return to land. The theoretical framework borrows from the work of
Carter and Mesbah (1993) and it assumes three important frictions: 1) Land transactions are ignored
and access to land is treated as exogenous, 2) households face the possibility of off-farm
unemployment, and 3) access to credit increases with land size.
Under these assumptions, consider a household that generates income by cultivating agricultural
land, in addition to possibly supplying labor at an exogenously determined market wage, w. The
household is endowed with T hectares of land and
L
hours of labor per year that are employed in on-
farm agricultural work (L
f
) and/or off-farm activities (Ls). The household cultivates a single crop
using X units of input purchased at a per unit market price of q. The crop can be sold at an exogenous
market price p. Let F(L
f
,X,T;z) be the production function, where z represents the set of household and
contextual characteristics that affect the return on productive assets. Let Ω(Ls) denotes the number of
days employed as a function of labor supplied, Ls, where Ω’>0, Ω”
0
≤
. Let Ґ(T) denotes the amount
of working capital available at an interest rate i to a household with land endowment, T. The cost of
production, qX, must be financed by the sum of initial wealth K, wage income wΩ(Ls), and available
capital Ґ (T).
Formally, the household chooses time allocation and purchased inputs to maximize its income:
max pF(L
f,
X, T;z) – qX + wΩ(Ls) – i(qX - K - wΩ(Ls))
Subject to
4
L
f
+ L
s
L
≤
qX
+≤
K
wΩ(Ls) + Ґ(T)
L
s
, L
f
0
≥
(1)
Assuming an interior solution for the labor allocation, the first-order conditions of this maximization
problem can be written:
pF
L
= w(1 + I + λ)Ω’
pF
x
= q(1 + i + λ)
λ(K + w Ω(Ls) + Ґ(T) – qX = 0; λ
≥
0; K + wΩ(Ls) + Ґ(T) – qX
≥
0 (2)
So that we have maximum in Y with optimal value of L
f
*
,
L
s
*
, X
*
Y = pF(L
f
*
,
X
*
, T;z) – q(1 + i)X
*
+ w(1 + i)Ω(Ls
*
) + iK
= Λ(p,q,i,w,
L
, K,T,z) (3)
The income equation is a function of prices, household’s endowment of productive assets, and
any characteristic that affects the return to these assets. We can differentiate Eq. (3) to see how an
increase in landholdings affects household income,
dT
dY
= pF
T
+ (pF
L
– w(1 + i) Ω’)
dT
dLf *
+ (pF
x
– q(1 + i))
dT
dX *
= pF
T
+ λw Ω’
dT
dLf *
+ λq
dT
dX *
(4)
If capital markets are perfect (λ =0), then the terms in parentheses are identically equal to zero
and the marginal value of land is simply the value of its marginal product, i.e., dY/dT = pFT.
Moreover, if households face the same opportunity costs of labor and inputs, and we assume constant
returns to scale, then the marginal return to land is constant for all land endowments.
Conversely, with imperfections in labor and credit markets, the terms in parentheses are positive.
An increase in landholding has both a direct and an indirect effect on income. In addition to
directly increasing production, more land affects the distortions in the allocation of production inputs.
As the land endowment increases, labor allocated to the farm increases, the under-employment
rate Ω decreases, and the marginal productivity of labor Ω’ increases. On the other hand, as the land
5
endowment increases, the credit constraint may either tighten or loosen (k can increase or decrease)
depending on the relative increase in credit availability Γ(T) and on the demand for inputs. For
instance, in poor rural areas characterized by thin labor markets, an increase in household land will
increase the marginal product value of household labor and reduce the difference between its
shadow wage and the market wage. If, as Eswaran and Kotwal (1986) suggest, larger farms have
better access to credit, then an increase in landholding will increase the use of variable inputs and
reduce the distortion in the input markets as well. With market distortions, we consequently expect the
marginal value of land to vary with the land endowment, and quite possibly in a nonlinear
manner.
II. Econometric specification of the welfare equation
This section outlines the semi-parametric procedure for estimating the relationship between
income and land endowments. Our production model implies that any characteristic that affects the
return to the productive assets of the household should influence the household’s income. This list
includes household demographics, constraints on factor use, as well as village and regional factors that
capture employment opportunities and market integration. Applied in this paper, semiparameter
procedure will be mentioned. According to this procedure, it still control for other factors that
determine household income one side, in another hand, it allows us relax the functional form on land.
Inheriting Thorsnes and McMillen, (1998), we estimate a model of the following form:
yi = β’xi + g(zi) + εi (5)
where x of dimension n × k is the set of control and z of dimension n × l is the hosehold’s land
endowment. The variable y proxies some measure od household welfare. Constant term is denote by
α, and β is a k×l vector of parameters of interest. i = 1, . . . , n. The function g(·) is assumed to be
smooth and continuous. The error terms, εi, are assumed to be independent, but they may be
heteroskedastic and do not have to have zero mean.
I use standard parametric approach, where g(zi) is the linear function δ’zi. In a parametric
approach, the vector of coefficients, δ, is most easily estimated by regressing y on x and z. Identical
estimates are obtained from a three-stage procedure. In the first stage, regress y and each of the K
variables in the vector x on z, and form the predicted values y^ and x^k. Next, regress y − y^ on the
vector of residuals x
*
k = xk − x^k to estimate β. In the third step, estimate δ by regressing y − β^’x
on z. Intuitively, the first step purges y and x of the effects of z, and the second stage estimates the
6
independent effect of x on y. The third stage estimates the effect of z on the dependent variable after
the effects of z have been removed.
III. Literature review
In recent years, programs of access to land have returned high on the agenda of poverty reducing
strategies pursued by governments, NGOs, and international development agencies (Deininger and
Feder, 2002). Measuring the poverty reducing value of land and the conditions F. Finan et al. / Journal
of Development Economics 77 (2005) 27–51 28 for successful use of this instrument has thus become
all the more urgent.
One of influential paper conducted by Frederico Finan, Elisabeth Sadoulet, Alain de Janvry (2002)
on measuring poverty reduction potential of land in rural Mexico. Results show that a small amount of
land can create large income gains as it permits to mobilize family labor affected by labor market
failures. In general, land has a high but decreasing marginal return for endowments of less than 3 ha
and a constant return for land sizes larger than 3 ha. Additionally, the marginal welfare value of land
depends crucially on both the complementary assets (such as education) and the contextual settings
(such as infrastructure) of the poor, prohibiting any absolute statement about the poverty reduction
value of land. They also found that ethnicity is an important negative social asset as the marginal value
of land for non-indigenous households is on average twice as high as it is for indigenous households.
When complementary assets and a favorable context are in place, access to even a small amount of
land can help households escape poverty. The paper thus helps establish the complex set of conditions
under which land can be a valuable poverty-reducing instrument.
Ramo´n Lo´pez and Alberto Valde´s (2002) in the paper of “Fighting Rural Poverty in Latin
America: New Evidence of the Effects of Education, Demographics, and Access to Land” in their
paper with data collected from 4 countries in Latin American including El Salvador, Paraguay and
Honduras have found that marginal productivity of land is higher for small farmers than for large
farmers. They found that the hypothesis that farm production elasticities are equal among small and
large farmers could not be rejected in any of the countries where this analysis was performed
(Paraguay, Honduras, and El Salvador). Combined with the fact that yields are larger for small farmers
than for large farmers, and they have conclusion that marginal productivity of land is higher for small
farmers than larger farmers.
7
Không có nhận xét nào:
Đăng nhận xét