Determinants of non farm household in come in rural vietnam

7Table 2.2: The schemes for development of non-farm sector in India 14 Table 4.1: Participation of household into non-farm activities Table 4.2: Participation ofhousehold into non-farm a

UNIVERSITY OF ECONOMICS INSTITUTE OF SOCIAL STUDIES

-•

VIETNAM - NETHERLANDSPROGRAMME FOR M.A IN DEVELOPMENT ECONOMICS

DETERMINANTS OF NON-FARM HOUSEHOLD INCOME IN RURAL VIETNAM

A thesis submitted in partial fulfilment of the requirements for the degree of

MASTER OF ARTS IN DEVELOPMENT ECONOMICS

This is the very means that gives me a chance to formally express myappreciation to people who have really created the concepts and methodologyexpressed in this research

I always keep in my mind the gratefulness to professors ofMDE Programme,executive programme administrations in Vietnam Their enthusiasm about theexperience and what they were teaching, was the most meaningful encouragement

of mine

I would like to express my deep and sincere gratitude to my supervisor,Doctor Nguyen Ngoc Thuy His knowledge, logical way of thinking andencouragement has provided a good basic for the thesis presentation

I would like to express my great thanks to Professor Nguyen Trong Hoai,MDE programme who have eagerly supported during my study

BUI THANH GIANG

ii

Ho Chi Minh City, Septemper 2012

BUI THANH GIANG

Vietnam has made considerable progress in improving the living standards of

its people since the reforms implemented under doi moi However, the reform

progress has made problems of society such as the pressure in rural - urbanmigration, an increase in unequal income distribution between different regions andunstable economy Therefore, a development strategy should be more focused onpromotion of rural non-farm sector in order to create more employment and incomeopportunities for rural populations This paper estimated the non-farm incomefunction of households in rural of Vietnam and evaluated the importance of ruralnon-farm activities in generating employment and income opportunities Theeconometric analysis is based on the primary data from Vietnam Household LivingStandard Survey 2008 With 2117 observations (in rural area), the regression resultsshow that the factors such as household characteristics, regions, infrastructure,occupation of household head had strong impact to non-farm household income

in rural area From the regression results, this study gives some solutions to promoterural non-farm activities in Vietnam in the coming years

iv

2.2.3 The experience in the development of rural non-farm sector 92.3 Theoretical basis of the determinants of non-farm household income 13

3.4 Data Analysis 31

Chapter 4: DETERMINANTS OF NON-FARM HOUSEHOLD INCOME

4.1.1 The participation ofhousehold into non-farm activities 34 4.1.2 The contribution of non-farm activities in employment

4.2 Determinants of non-farm household income in rural Vietnam 46

LIST OF TABLES

PageTable 2.1: Classification of farm and non- farm activities 7Table 2.2: The schemes for development of non-farm sector in India 14

Table 4.1: Participation of household into non-farm activities

Table 4.2: Participation ofhousehold into non-farm activities

Table 4.3: The participation ofhousehold into non-farm activities

Table 4.4: Business sectors participation by areas 38Table 4.5: Business sectors participation by income groups 39

Table 4.6: Business sectors participation by ethnicity 40

Table 4.7: The distribution of sources of income to household income in 2008 41

Table 4.9: The result of income classified by ethnicity and gender 43

• Table 4.10: Labor market participation by age and schooling level, 2008 44

Table 4.11: Labor market participation by income groups 45Table 4.12: Definitions of variables used in the regression model 46

Figure 4.3: Labor shares by income sources in 2008 45

APPENDICES

Page

Appendix 1 : Model V- Regression without Age2 to remove Multicollinearity 72Appendix 2: Applying Robust Method to remove Het for Model V 74Appendix 3: Applying Robust Method with standardized Beta for Model V 7 5

viii

Chapter 1: INTRODUCTION

The determinants of rural non-farm household income in Vietnam will beinvestigated in this chapter In particular, this chapter includes five sections Thefirst section will raise the problem necessary to make clear for the case of Vietnam.The second section will present the research objectives of this thesis The nextsection will raise the questions to obtain the proposed objectives The fourth sectionwill briefly describe the research methodology And the structures of the thesis will

be described in this chapter

1.1 Problem Statement

For nearly 25 years, the government of Vietnam has been looking for thedirection of socio-economic development in order to escape the rural from deprivation,poverty and backwardness Government has many programs to promote the new ruralconstruction to improve people's living standards Economic reform in Vietnam in thelast two decades, characterized by the economic policy reform, opening and operatingeconomy towards a market economy According to Vietnam Economic Times (20 11 ),the average economic growth increased from 7.5

% in the period 1991 - 2002 to 8% in period 2002-2007 The economic structurehas changed the direction towards industrialization and modernization Theproportion of agriculture in total GDP of the country declining from 27.18% in

1995 to 24.53% in 2000 and to 21% in 2008 The poverty rate fell consistently from58% in 1993 to 29% in 2002 and to 14,5% in 2008 (Vietnam Academy of SocialSciences - 201 0) Economic growth and the stability of the macroeconomicenvironment has confirmed the correctness of innovation policy

In rural areas, the process of economic restructuring also has changedsignificantly Due to the success of diversified occupations strategy, the leadingposition of agriculture in the rural household economy have no meaning as before.The program of mechanization, modernization of irrigation has reduced labor time

in agriculture significantly Because of the longer leisure time, the rural householdcan participate in the non-farm activities on a regular basis Any economic activitywhich has effective may become the main activity of household economy in rural.The rural economy shifted towards increasing industries, services and otheractivities of non-farm sector The development of non-farm sector has contributed

to creating jobs, increasing incomes, improving living standards for a certain part ofresidents, and creating a new face for rural area This suggests that non-farmactivities play the important role in the living of households and development ofeconomy and society ofVietnam

The non-farm activities contributed to the improvement and diversification

of people's income This has been widely recognized Especially in rural Vietnam,the development of non-farm sector has great meaning in solving the poverty,underemployment and seasonal unemployment which are still the phenomenon ofthe rural area Vietnam has gained remarkable achievements in improving people'sliving standards since the reforms implemented under doi moi However, the reformprogress has created problems of society such as the pressure in rural - urbanmigration, an increase in unequal income distribution between different regions.Furthermore, the rural - urban gap has widened considerable Therefore, an effectivedevelopment strategy should be more focused on promotion of rural non-farmsector in order to create more employment and income opportunities for ruralpopulations and improve other problems of society

It is widely recognized that income from non- farm activities likely to playincreasingly significant role in the household income in rural This study investigatesthe participation of rural households into non-farm activities to have an overview ofnon-farm activities in rural Vietnam; identifies the contribution of non-farm activities

on household income; analyzes determinants of income of rural non-farm household toidentify the important determinants of rural non-farm household

2

income; and gives some solutions to promote rural non- farm activities in Vietnam

in the coming years

1.2 Research Objectives

The overall objective of the study is to assess the contribution of non-farmactivities in rural area, and then investigate the factors that may have influence onnon-farm household income and give some recommendations to promote non- farmactivities

Overall objective of the study are detailed m the following specificobjectives:

* Objective 1: To identify the contribution of the non-farm activities on the household income in rural Vietnam

* Objective 2: To analyze the determinants of rural non-farm household income in Vietnam

* Objective 3: To gtve some recommendations to promote non-farmactivities

1.3 Research questions

The thesis will focus on addressing four questions:

1 What are contributions of non-farm activities in rural Vietnam?

2 What are the major determinants of rural non-farm household income

in Vietnam?

3 What are recommendations to promote non - farm activities in rural Vietnam

non-+Regression analysis:

This study about determinants of non- farm household income can be carriedout by two approaches: macro and micro approach The data about non - farmhousehold income in rural of Vietnam at macro level is not enough Then, themicro approach is employed in this study The data from Vietnam HouseholdLiving Standard Survey (VHLSS) will be used in analysis To test the hypotheses,quantitative method applying ordinary least square (OLS) technique is employed

This study also employed the primary data and secondary data Theprimary data is mostly from VHLSS Secondary data come from other sources such

as previous empirical studies, World Bank, General Statistical Office (GSO)

1.5 Research Structures

The content of this thesis is divided into five chapters:

I Introduction: This chapter provides the problem statement for the

necessity of analyzing the determinants of non-farm household income in rural

Vietnam The research objectives, research questions and research methodology,research structures are presented in this chapter

4

II Literature review: This chapter presents some important concepts, varioustheories about determinants of rural household income and empirical studies m order tounderstand the theoretical review and to form an empirical framework.

III Methodology and Model Specification: The data and the methodologyare described in this chapter

IV Determinants of non-farm household income in rural Vietnam: Thischapter analyses the participation of rural household in non-farm activities fromdifferent aspects in order to have an overview of non-farm activities, identify thecontribution of non-farm activities on household income in rural and analysesregression results of rural non- farm income

V Conclusions and Recommendations: Based on the data findings andregression results, this chapter summarizes the main findings Then somerecommendations are suggested in this chapter

Chapter 2: LITERATURE REVIEW

To conduct the analysis of the factors affecting non-farm household income

in rural area and finding solutions to improve the efficiency of this activity, we need

a platform of theory as a basis for research Chapter 2 will provide some importantconcepts that will be used in the theoretical review and empirical study Afterunderstanding the main concepts, various theories about determinants of non- farmhousehold income are presented to draw a theoretical and conceptual framework forthe empirical analysis in the following chapters Next, the review of empiricalstudies as well as the related model is presented to show the impact of factors onnon-farm household income in reality

2.1 Concepts and definitions:

- Households: The Article 106, Civil Code of Vietnam 2005 stipulates that:

"Family households in which members have common property and jointly contributetheir efforts and labor to their common economic activities in agricultural, forestry orfishery production or in a number of other production and/or business domains defined

by law shall be subjects when participating in civil relations in such domains."

In addition, under Clause 1, Article 107, Civil Code 2005, "The head of afamily household shall be the representative of the household in civil transactionsfor the common interests of the household The father, mother or another adultmember may be the head of the household"

With these provisions, we can offer households defined as follow:households in which members have marital relationship, blood relation, commonproperty and jointly contribute their efforts and labor to their common economicactivities in agricultural, forestry or fishery production or in a number of other

6

production and/or business domains defined by law shall be subjects whenparticipating in civil relations in such domains.

-Non-farm activities and farm activities:

Farm activity means agricultural activity Farm activities is directly related

to the animal and the plant The non-farm activity is similar to nonagriculturalactivity The non-farm activities are related to construction, mining and quarrying,transport, manufacturing, handicrafts, repairs, trade, personal services etc

Classification of farm and non-farm activities in the rural areas as below:

Farm wage employment Hired farm worker m crop and livestock

farming

Non-farm wage employment Wage employment m manufacturing and/or

services; food processingNon-farm self-employment Own enterprise m manufacturing and/ or

non-•

is that an rural non-farm activity is performed in a location, which falls within adesignated rural area The second definition is based on the linkage approach Anindustrial enterprise generates significant developmental linkages with ruralresidents."

- The rural non- farm enterprises: In the report on "Rural non-farm

enterprises, investment climate constrains, infrastructure accessibility and

household's employment" of Central Statistical Agency of Ethiopia and the WorldBank, 2007, the term is defined : "An economic unit operating on its own accountoutside agriculture (this includes small-scale activities independent of the

technological sophistication It can be full/part time formal/informal, andseasonal/episodic It may take place at home, a specific business location, or beperformed by itinerant traders Processing farm products and then selling them isdefined as non-farm)"

- The rural non-farm economy: "The rural non-farm economy may bedefined as comprising all those non-agricultural activities which generate income torural households (including income in-kind and remittances), either through wagework or in self- employment" (Junior R.David, 2003) Definitions of the rural non-farm economy are problematic The rural non-farm economy might include trade andprocessing of primary agriculture, forestry, fisheries and other goods and services Therural non-farm sector is defined: "a set of economic activities carried out in the ruralareas that are not agricultural" (Sanchez, 2005)

- Income from rural non-farm activities is commonly defined as thatderivedfrom wage-employment activities and self-employment in the rural non-farm sector,and remittances However, this study will focus only on non-farm self-employmentactivities which are fully/partly owned or operated by households or householdmembers in rural area

According to VHLSS, there are four mam sources of income of the household.They include: income from employment, income from agricultural production activity,income from agricultural services and from non-farm self-employment activity In thisthesis, the rural non-farm income of the household is calculated by earning from non-farm self-employment activity of household or household members in rural area

8

2.2 Role of rural non-farm sector

2.2.1 The Push- Pull factor

There are many studies proposed the models to explain the reason of why ruralhousehold can engage in non-farm activities A basic household model provide aframework for analyzing household behavior (Hymer and Resnik, 1969; Bardhan andUdry, 1999; Singh, Squire and Strauss, 1986) It as a tool in developmentmicroeconomics to model the joint consumption and production decisions made by therural household An other simple explanation about the reason of participation intonon-farm activities of rural household is Push -Pull factor Rural household aremotivated to undertake non-farm activities by either "pull" or "push" factors (Reardon,1997; Islam, 1997)

The authors mentioned the "push" factors as follow: (1) population growth,(2) increasing scarcity of arable land and decreasing access to fertile land, (3)decreasing fertility and productivity of land, (4) decline of the natural resource basis,(5) declining return to farming, (6) increasing monetization of people's lives,

(7) temporary events and shocks, (8) lack of access to farm input markets, (9) absence

of rural financial markets The "pull" factors are mentioned: (1) higher return on labor

in rural non-farm sector, (2) higher return on investments in rural non-farm sector, (3)lower risk of rural non-farm compared to farm activities, (4) generation of cash inorder to meet family objectives, (5) investment opportunities In summary, "Push"factors relate to pressure or restriction of agriculture sector that force farmers to seekincome from other activities in order to improve their living conditions The "pull"factors create the attraction for farmers in order to join in non-farm activities

The non-farm activities can share the risks in agricultural activities and bringhigher income for rural families Besides, the non- farm activities also creates a higherpotential for diversification of income sources And the non-farm activities

also create higher productivity in agricultural production Through the impact of the

"pull" and "push" factors, we can see the importance of non-farm activities for ruralhouseholds

2.2.2 The importance of rural non-farm sector

There are a large number of studies showed that the non- farm sector can absorbthe surplus labor force in rural areas, narrow the gap of rural - urban, create more jobsand reduce inequality Junior R.Davis (2003) summarized the finding from more than

55 studies of rural economies and the rural non-farm economy and pressed that therural non-farm sector providing between 40% and 60% of income and employment inrural areas Reardon et al ( 1998) have found that the non-farm activities in rural areaplay an important role in generating income and employment About 32% of ruralincomes in Asia, 40% in Latin America and in Africa 42% came from non-farmincome source and about 44% of rural labor in Asia and 25% in Latin America join innon-farm sector, mainly from rural localities Along with the comparison betweenyears and with many other studies, Reardon showed a positive growth of the share ofnon-farm income over time in sub-Saharan African, Bangladesh, China, WesternKenya, Malaysia, Mexico, Northern Nigeria, Philippines, Taiwan, Korea, India Healso mentioned that small-scale non-farm activities offers the higher productivity inrural areas compared with large-scale enterprises because the cost per unit oflabor ofnon-farm is lower

The non-farm sector plays an important role in the rural economy In manydeveloping countries, agriculture is not the only source of employment and income forrural households The non-farm activities accounts for a third or more of ruralemployment in many Asian countries (Rashidpour, 2011 ) The larger share of ruralnon-farm economy is services, followed by trade and business, construction activities,and transport operations The non-farm sector provides employment to differentcategories of skilled and unskilled labor force And this sector also contributes in theincomes of both farm and non-farm households (Dave, D and

10

Dave, R 2012) According to Dave, the contribution in the creation of employmentand the generation of income of this sector has been consistently increasing Thedecrease of the ratio of land per capital and an overall decline in the net cultivatedland area are the reasons of this consistently increase.

The rural non-farm sector is playing an important role not only in the ruralbut also in the national economy (Islam and Jin, 1994 ) The activities of non-farmsector can create the positive impact on economic growth, rural development andincreasing potential sustainability of natural resources, food security and povertyalleviation (Pfluger, 2000)

In conclusion, the rural non-farm sector play a very importance role inproviding gainful employment and sustainable income to the rural households.Because of the increasing role of non-farm sector and the limited capability of theagricultural sector, the policy for the development of potential non-farm activities isvery necessary

2.2.3 The experience in the development of the rural non-farm sector

In fact, the government of many countries recognized the role of non-farmsector They offered many programs and plans to promote the development of non-farm activities And they get the encouraging results The thesis will discuss aboutthe experience of China and India in development of the rural non-farm sector inorder to confirm the important role of this sector

a China

As we know, following the reform decision in late 1978, China's economydeveloped rapidly From 1980 to early 1996, the economic growth of China wasalways around 10% (with the year up to 15%) China has made breakthroughs ineconomic development in rural areas In this period, the development of theTownship and Village Enterprises is most prominent

China is a big country of agriculture The rural population in Chinaaccounted for nearly 80% percent in 1980s There was the heavy populationpressure on the land And the agricultural labor productivity continued to stay at avery low level Because of stagnant and low rural incomes, the farmers werestrongly encouraged to abandon working the land However, the governmentcontrolled the rural-urban migration with several direct or indirect measures In theearly 1980s, the government's policy towards rural non-farm development changed.This policy is apart of new rural development strategy The new policy wasdescribed by the slogan "leave the land but not the countryside, enter the factory butnot the city" (Ho,l995) This slogan means non-farm participants work outsidetheir land plots but do not move out of their places of origin This policy allowed toabsorb local surplus labor force through the development of local non-farm sectors.

The rural non-farm sectors consist essentially of the Township and Village

• Enterprises The Township and Village Enterprises include all the medium and

small or very small enterprises in rural areas They basically constitute the ruralnon-farm sector in China Although operating in rural areas, but most Township andVillage Enterprises operating in the industrial sector, and also in agriculture,commerce, transportation and construction Most of the Township and VillageEnterprises are labor-intensive, absorbing large segments of surplus labor

Between 1981 and 1990, the total industrial output of Township and VillageEnterprises grew at an average annual rate of28.1%, The GDP share ofTownshipand Village Enterprises increased from 14.3% in 1980 to 37.5% in 1995(Chenggang Xu and Xiaobo Zhang, 2010) These enterprises contributedsignificantly to rural economy and infrastructure development in China

The rural non-farm sector is playing an important role both in the rural and thenational economy The development of the rural non-farm economy deeplytransformed the structure of rural household incomes in China Non-farm activities

12

gradually became an importance source of income for rural households, and served

as an engine of growth for rural areas

b India

Majority of population in India lives in rural areas and mostly depends onagriculture In the period 1951-1991, the share ofthe rural population in the nationaltotal has declined However, the rural population has more than doubled Thepressure of population and employment has forced the government to make thesuitable policies towards sustainable development of rural economy

The new economic policy of the government after 1991 has greater focus onmarket-oriented approach This created the positive impact on the growth of ruralnon-farm sector With the new policy, the rural economy is diversified into non-farm activities to provide employment to rural labor force Moreover, the rural

• people will be able to get employment in urban areas Rural non-farm economy is

assumed an effective tool for eradication of poverty in rural India This is also helpprevent rural-to-urban migration

According to Asian Productivity Organization (2004): "The rural non-farm

sector is classified into three major sub-sectors In the first sub-sector, theenterprises are run on more or less stable basis with target on the surplus generationand growth, employing labor with certain degree of technical sophistication Thesecond sub-sector consists of products or activities, which are often seasonal, runsolely with the help of unpaid family labor, using primitive technology and cateringmostly to the local market The third sub-sector consists of paid workerscharacterized by low earnings and disintegrated market with respect to laborsupply." The above classifications show that non-farm sector play an important role

in providing employment opportunity to different categories of skilled and unskilledlabor force

Government of India has launched many policies, plans and programs topromote non-farm production activities The various schemes oflndia's government fordevelopment of non-farm sector are highlighted below:

This scheme has been launched from 1999 Theobjective is to bring the assisted poor families above thepoverty line by providing them income-generating

1 Swarnajayanti Gram assets Their assets will be provided by a mix of bankSwarozgar Yojana credit and government subsidy The support in capacity

building, training, planning of activity clusters,infrastructure build-up, technology and marketing iscarried out

The main objective of this scheme is to create theadditional wage-employment opportunities during the

2 Employment Assurance period of acute shortage of wage-employment through

Scheme manual work The secondary objective is to create the

durable community assets for sustained employment anddevelopment

The program of this scheme was launched in 1993-94 in

3 District Rural Industries five selected districts The objective is to generate the

sustainable rural employment opportunities byProject

providingadequatecreditsupportforruralindustrialization together with promotional measures.This scheme focus on providing training and skilldevelopment of the upcommg entrepreneurs The

4 Rural Entrepreneurship contents of the training and skill development programDevelopment Program are project management, accounting practices,

marketing information, technical guidance, motivation,and institutional support mechanism

Source: Asian Productivity Organization (2004)

Table 2.2: The schemes for development ofnon-farm sector in India

The schemes provided comprehensive support for the development of non-farmactivities The program focuses on building infrastructure, credit support, skillstraining, ect The result of the development of non-farm sector in India is

14

remarkable According to the report of APO (2004 ), the number of workers in ruralnon-farm sector grew at an annual rate of 2.8% during the period 1981-1991 Theshare of the rural workforce of rural non- farm sector has been accounted for asteadily rising (from 15% in 1978 to 22% in 1987- 1988 and 23% in 1993-1994).And the rate of employment growth in this sector was 5% between 1987-1988 and1993-1994 The manufacturing accounted for highest share of rural non-farmemployment(31 %) in 1993-1994 Next, services 27%; trade 20%; and construction12%.

In India, rural non-farm sector is an effective tool helping to keep the poorfrom falling into deeper poverty, helping to lift the poor above the poverty line.This sector generates activities with a high employment potential for rural workersand becomes the important sectors to focus on in order to aileviate poverty

2.3 Theoretical basic of the determinants of non-farm household income

2.3.1 Theoretical framework

From the late 1980s to present, the poverty or non-farm empirical literatureshave been given by many researchers following many different ways Reardon(1997), Readom et al (1998) and Lanjouw and Lanjouw (2001) offered theliterature about the trends and analysis of the determinants of non-farmemployments Reardon et al (1994) and De Janvy et al (2005) offered the literaturethat has put on agriculture as a determinant of non-farm employments and hasexamined the effects of non-farm income on farm The literature focuses on theeffects of non-farm income on poverty is presented by Lanjouw (2001), Lajouw andMurgai ( 2009)

Oostendorp, R H et al (2008) studied the change of role of the non-farmhousehold enterprises (NFHEs) in Vietnam by using the VLSS 1993, VLSS1998and VHLSS 2002 household survey data They found that NFHEs increase income

and reduce between-household inequality They proposed general framework forestimating the contribution ofNFHEs to household income:

u/

,

NFJ NFJ

According to the model of Oostendorp, R H et al, the factors such as age,education, gender, household size, land size, local employment conditions, localinfrastructure, the migration status of relatives of the household have significant toincome household And through many studies show that they are also the factorsaffecting non- farm income of rural area

Excluding the impact of non-farm activities to household income, the factorseffect the household income include five factors as figure bellow:

16

The household and

farm characteristics

The local infrastructure

The local employment

conditions

The migration status

TOTAL HOUSEHOLD INCOME

Income from Income Income from Income

employment farm employment

Figure 2.1: The framework ofdeterminants ofhousehold income

There are four main sources of household income that include: income fromwage employment, income from farm, from self-employment and from transfers.Non-farm income as income from non-farm self-employment includes wages andprofits from business activities Existing studies also show that household wealthand composition, education level, infrastructure and regional differences areimportant determinants of non-farm business income

a Household Wealth and Composition

The wealth of a household is an important determinant of the level andnature of the participation of households in non-farm sector Because of theunderdeveloped of credit markets to fmance non-agricultural businesses, cash flowownership is very important to start non-farm businesses and to pay the transactioncosts of obtaining non-farm employment Many studies show that households withlarger land holdings to diversify often better paid non-farm sector in rural areas

Household composition also affects the kinds of jobs and businesses in farm sector Important variables include health status, family size, age, dependency

non-17

ratio and gender Gender also affects the non-farm businesses Gender is animportant factor influencing participation patterns and trends in the rural non-farmeconomy (Davis, 2003 ) Women are more likely to engage in small commercialenterprises, service, and textiles and clothing, while men focus more on production,construction, and transportation The role of gender in enabling or restricting access

to economic activity also have difference between countries (Davis, 2003)

b Education

Education is a determinant of business success in non- farm wage andproductivity that has been widely recognized Many studies showed that primaryeducation and secondary education contribute to the growth of non-agriculturalsector in rural areas It is associated with higher productivity in business,construction, service and manufacturing operations Literacy enhances theproductivity of the workforce and make it easier with the skills provided throughtraining on the job The high school education stimulates business capacity Thesuccess of rural non-farm business is often reported that have the correlation witheducation (Davis, 2003) "In developing countries, an entrepreneur with anelementary education can expect to earn an income 41 percent higher than one with

no education at all" ( Islam, 1997)

c Infrastructure

Infrastructure is also important for the development of enterprises of farm households Rural roads can allow motor vehicle to access reliable and regular.This serve both the farm and non-farm economy Electrification is also importantfor manufacturing activities of some scale in rural Abundant supplies of water canallow some rural non-farm activities to take place "The expansion of roads,transportation, and communication leads to specialization and division of labor ofthe household" (Islam, 1997) When the road, transportation and communication areexpanded, the development of commercial, marketing, and distribution network,

non-including subcontracting and associated farm and non-farm sector in local will beimproved Improved infrastructure means lower transaction costs for marketproducts (both agricultural and non agricultural) and availability of inputs(electricity, tractors, ) with more lower cost Therefore, the quality and quantity

of infrastructure is often identified as important determinants of investment in agricultural business

non-d Regions

Region also affects the farmers to diversify their income on non-farmactivities Households in an farm area with high risk will be pushed to diversify intonon-farm In contrast, households in areas where farm may be less risk involved innon-farm enterprises mainly for higher profits to reduce the limited cash and credit.Agricultural/farm sector tends to grow in popularity in areas where agro climaticzones are favorable

The thesis will research the determinants of non- farm household income inrelation to the socio-economic factors Then, the five factors in the model ofOostendorp will be used in analyzing the factors effecting the non-farm income inthis thesis

2.3.2 Empirical Studies

2.3.2.1 The Household Enterprises in Vietnam

Based on panel data from surveys of households living in Vietnam in 1993and 1998, Vijverberg and Haughton (2002) found some evidence that operating abusiness lead to greater prosperity The results of their research showed thatbusiness-farm households can be operated by urban households, those withmoderately good education

19

The authors have found that businesses in the <5outh of the country are lesslikely to survive, as well as smaller and younger businesses Some models are built andthe results showed that in areas with poverty, lack of education, credit, and effectivedemand limited the development of non-farm enterprises The rich areas had theattraction with wage labor.

The authors suggest that business-farm households is especially important role

in the transition period, and when agriculture is declining it's importance The authorsexpect these enterprises continue to play a supporting role in promoting economicgrowth in Vietnam

2.3.2.2 The model for determinants ofnon-farm income in Nigeria

In the study: "Determinant of Non farm income among farm households inSouth East Nigeria" (Ibekwe et al, 201 0), the authors offered the model to determinethe extent of relationship between socio-economic factors and the level of non-farmincome The regression function is of the form:

Y =f(Xl, X2, X3,X4,X5,X6,X7, e)

Y =Non-farm income

Xl =Age of household head

X2 = Education of Household head

X3 =Farm Size

X4 =Occupation

X5 = Household size

X6 =Farm Investment

X7 = Value of farm output

e = Stochastic error term

The most important result of this model for non-agricultural income is found to

be coefficient of education variable This coefficient is positive and significant at thelevel of significance (0.05) This suggests a fairly strong relationship of

education to non-agricultural income The results also showed that households with highereducation are more likely to seek off-farm employment in rural areas.

The author has reviewed the participation of households in non-agriculturalactivities by reference to the personal characteristics, household characteristics such asspecific occupations, education, land size of holding families as well as agriculturalproduction research results show that household size of land holding, number of years ofeducation of labor, per hectare of agricultural output value, occupation and age ofhousehold head are important factors for non-income farm household level This suggeststhat economic factors and social importance of non-agricultural policies of SoutheastNigeria

2.3.2.3 The mode/for determinants of income ofrural non-farm business

The model of Trung Tran Quoc (2002) for determinants of income of rural farm business activity:

non-Yi=F(Xi, Ei, Hi, Bi, li, Ri, u) (1)

Where Yi is income from the rural non-farm business activity i, Xi 1s a vector offamily-owned inputs, Ei is a set of the characteristics of the entrepreneur, Hi is a set of thecharacteristics of the household; Bi is characteristics of the business activity; Ii is a vector

of infrastructure variables, and Ri indicates seven regions of Vietnam Let u is a randomdisturbance, capturing unmeasured determinants of and stochastic influences on rural non-farm business income

The result showed that the returns to physical capital and labor are quite high andexhibit the diminishing returns to scale In addition, education is relatively high impact tonon-farm income Other findings are non-agricultural significantly related to age andgender of an entrepreneur This implies that there is gender discrimination in the businessactivities of non-agricultural rural

21

Model results also showed that households with larger size have higherincomes Credit rates also affect household income, especially the poor have theability to combine with higher borrowing rates So credit policy in rural areas isneeded to improve farm income for rural households.

2.3.2.4 The mode/for self-employment activities

Production function has been used as an important tool of economic analysis

in the neoclassical tradition The form of production function is most useful andpopular is Cobb-Douglas It was proposed by Knut Wicksell (1851 - 1926), andtested against statistical evidence by Charles Cobb and Paul Douglas during 1900-1947

Kurosaki (200 1) examined the self- employment activities by estimating theproduction function of value added (gross production value minus cost ofintermediate input) of self-employment activities, by applying a simple Cobb-Douglas production function

Where qhj is the value-added from self-employment activities j for household

h, Lhj is labor input by household h into activity j; Hhj is land input (farm) orcapital input (non-farm); Xhj is a vector of household h's characteristics such aseducation, age, and other characteristics of household head

This study examined the specific impact of household's education on themarginal returns of labor and on productivity of selt:.employment activities Theconsequence of the equation showed that the marginal return of educationcorresponds to a simultaneous increase in education of all household members

The results of this research show that: Only male middle and upper education levelhave significant positive effects on non-farm wage and only production factors

(such as labor, capital) and high education have an effect on non-farm enterpriseproductivity.

2.3.2.5 The model for determinants ofrural household income

Aikaeli (20 10) estimated the determinants of rural household income:

Y =f(Xl, X2, X3,X4,X5,X6, e)

Y = household per capital income

Xl =level of education of the household head

X2 = household labor force

X3 = acreage of land used by the household

X4 =dependency ratio

X5 =households owning rural non-farm economic activities

X6 = gender of household head

e = the error term

This study showed that education has appositive impact on per capital income.The investment in education will improve the income of household The householdwith larger the size of labor force and land will have higher per capita householdincome The household headed by men also found to have higher per capita incomethan those headed by women

2.4 Chapter remark

Many studies asserted that the non-farm activities play an important role in boththe rural economy and the national economy The activities of non-farm sector cancreate the positive impact on economic growth, rural development and increasingpotential sustainability of natural resources, food security and poverty alleviation(Pfluger, 2000) Specially, the non-farm activities play a very importance role inproviding gainful employment and sustainable income to the rural households Theexperience of some countries in the development of non-farm

23

activities also supported the role of this sector That is the reason why determinants ofrural non-farm household income is investigate by this thesis.

To provide the framework to determine the factors that impact to the non-farmhousehold income, the theoretical framework and empirical studies are presented Thefactors such as age, education, gender, household size, dependency ratio, land size,local infrastructure, the migration status, region are the factors that have the impact

to the household income The main purpose of this thesis is to find out the maindeterminants of non-farm household income in rural Vietnam

Chapter 3: METHODOLOGY AND MODEL SPECIFICATION

;

This chapter includes two main sections: model specification and datacollection The specification model about determinants of rural non-farm householdincome is proposed base on the theoretical review and empirical studies in previouschapters Besides, this chapter will define the variables of model and steps to dealwith data collection

3.1 Model Specification

Base on theoretical framework, the thesis chooses the factors that impact tothe non-farm income of rural household The income from non-farm activities ofrural households is affected by the factors belong to the household characteristicsand community characteristics The model specification includes five factors thatare presented in chapter 2 and two new factors

2 The local infrastructure

3 The local employment conditions

4 The miaration status

5 The region

6 Traditional handicraft village

RURAL NON-FARM HOUSEHOLD INCOME

Figure 3.1: The framework ofdeterminants ofrural non-farm household income

The integrated model can be expressed as follows:

25

In the linear model, Yi is the dependent variable is the income from farm activities of the household i Hi is the group of elements belonging to thedemographic characteristics of households Ei represent local employmentconditions Ii denotes the group of factors related to local infrastructure Mirepresents the migration status Ri represents the different regions of the country Tirepresents the presence of elements of traditional handicraft villages in thecommune Oi denotes the occupation ofhousehold head And finally, Ui isstochastic error term.

non-The dependent variable log(non-farm income) is generated because thedistribution of log (non-farm income) is look-alike normal distribution So, thedistribution of residual will also look like normal distribution The generaljustification is:

Log(non-farm income) = f (Household characteristics, The local infrastructure, The migration status, Occupation of household head, The local employment conditions, Traditional handicraft village, Region )

3.2 Data definition

The variables in the empirical model will be defined as below:

Non-farm income of household is calculated by earning from non-farm

self-employment activity of household in rural area Log (non-farm income) is Log ofvalue ofyearly non-farm income (1000 VND)

Household characteristics (Hi) include variables: age, number of years in

school, gender of household head, land size, the proportion of dependents andhousehold composition Age of the household head are counted in years Inquantitative analysis, only persons whose aged 16 and over are used in the analysis.According to previous studies, the age of the household head affect to non-farmincome of households but the effect will change with age According to data

analysis in Chapter 3, at the class of age 16 - 45, ifthe age increases, the labor willtend to join in non-farm activities Because the elements of experience in workingand health of labors shall be promoted However, the participation begins to fallafter age 55 In other words, the relationship between age and income of farmhouseholds may be a quadratic relationship So, when we estimate, the age variablewas divided into two variables: Agehead and Agehead 2 .

Years schooling variable is calculated by years of schooling of the household

head This variable represents the educational level of the head of household.According to several studies and in practice, education of the head of household usually

is highly crucial in the production and operation activities of the family In the case, ifthe household head had intermediate, college, university, masters or PhD degree, years

of schooling would be calculated the total number of school years + total time overhigher education Thus, the school year data ranged from 0 (illiterate) to 21 According

to data for analysis, data are limited in rural areas so that the school year ranged from 0

to 16 years

Genderhead is a dummy variable received variable value of 1 if the household

head is male and 0 if the household head is female Gender is taken into the equation todetermine that there is gender difference in creating non-farm income or not Dataanalysis in Chapter 3 showed that there are not different in gender of household headfor the participation in non-farm activities of households

Hhsize variable is the total number of household members, coupled with Dependency ratio variable - total number of household members divided by the total

number of members who are working of household Two variables are included toconsider the demographic characteristics of households that impact the non-farmincome of the household As in the theoretical analysis, demographic characteristics ofhouseholds play an important role in the decision to participate in non-farm activities.First, with large household size, the ability to specialize each labor will be higher, solabors will more likely to shift to non-farm activities

27

Similarly, the high dependents rate will create pressure on incomes, and wouldmake the labors to seek employment in non-farm activities And these householdsare more likely to engage in non-farm activities than other households (may be due

to agricultural work can be used the labor with outside working age)

Household composition represents the proportion of household members indifferent age groups: 16-25, 26-35, 36-45, 46-55, 56-65 and over 65 as well as theproportion of numbers of women in households aged 16 and over These variablesare included to consider in more detail about household characteristics It ispossible that non- farm income of household affected by the age of members in thehousehold rather than the age of household head

Landsize is the scale of agricultural land of household The hypothesis will

be tested: the wider the extent of land use by household, the higher the non-farmhousehold income There are two possibilities can occur: first, with large scale landhouseholds are more likely to make large capital for non- farm activities throughaccumulation from agricultural production or credit relationship with banks andtherefore a positive impact to non-agricultural income Second, with large-scaleagriculture land, households prefer to focus on agricultural production due to theeconomic benefits of scale in agricultural production can make agriculturalproduction more profitable or more easy restructuring plants That means thelandsize variable can have positive or negative sign

Traditional handicraft village (Ti) variable ts dummy variable This variableshows that there is the presence of traditional handicraft village in the commune ornot There are many analysis appreciated the role of the handicraft village inhelping families to restructure labor but also many ideas did not appreciate the role

of the handicraft village Because of the characteristics of trade secrets, market, ect,the scalability of handicraft villages is limited Therefore the promotion of laborparticipation of non-farm activities should be based heavily on the development of

Migration (Mi) variable represents migration status of relatives of thehousehold These members are doing away from home will be considered Theinclusion of this variable into the model to examine the impact of migration in ruralareas to non-farm income of the household There is the assumption that householdshaving members who work far from home will reduce the non-farm income ofhouseholds due to loss of labor However, the people working away from home cansend money that they earned to their household And their household can accumulatefor participation, maintain and develop non- farm self employment activities ofhousehold This variable is a dummy variable, receiving a value of 1 if householdreceived remittances of the members working away from home, whereas the value is 0.

Electricity, Road index1, Factory (Ii) present the condition of infrastructure inrural areas The inclusion of these variables into the model to assess the impact ofimproved rural infrastructure to non-farm income of the household Electricity variable

is dummy variable has the value is 1 if the commune with electricity and the value is 0otherwise The road index is calculated by the information on the survey of commune.Factory variable is a variable showing the presence of factory in the commune Thisvariable receives the value of 1 if the commune has a factory and 0 otherwise Theappearance of the factory may have two effect: increasing the number of wage laborand increasing numbers of non-farm self employment labor (through the provision ofservices for these factories)

1 According to the methodology developed by Vijverberg and Houghton (2002), a rural infrastructure indict-road index as follows:

Household income is also affected by local economic environment Theeconomic environment is the totality of economic factors (such as employment,income, inflation, interest rates, productivity ) that influence the buying behavior

of consumers and institutions So, Market lndex 2 (Ei) represents the economicenvironment of the local This variable is calculated from the information relating

to the daily market and period market in the commune The high index shows thelocal market will have more favorable conditions in the exchange, trading goodsand services, creating more employment opportunities for local labors This willstimulate the activities of non-farm production to meet demand for goods andservices The production of non-farm activities will have more advantages whenhouseholds want to sell their products

In summary, Market Index has a positive impact to non-farm income of the household

The Occupation of household bead (Oi) is considered in model Thevariables receive the value 1 if the household has a occupation and 0 otherwise.There are 10 occupations in the VHLSS 2008 The inclusion of these variables intothe model to consider occupation of the headed household that how to affect thenon-farm income of the household There is the assumption that occupations ofhousehold head have an important influence to the decision to participate in non-farm households

2

According to the methodology developed by Vijverberg and Houghton (2002), for market indexes, the truncation points are chosen according to values indicated in the survey The daily frequency is a proportion, so a market that operates once a week has a daily frequency of 1/7 Local market index is the combination of the daily market index and the periodic market index in the commune.

Daily Market Index= t

-Distance to nearest daily markej

72

t Distance to nearest period markej Period Market Index= 1- x Daily Frequency

50

Factor Ri includes the variables represent the different regions There are seven

variables present for seven regions to control geographic factor affecting non-farm income

of the household in rural And they are dummy variables

3.3 Data collection

This thesis uses the data from Vietnam Household Living Standard Survey 2008.The survey is completed by the General Statistical Office ofVietnam with the assistance ofthe UNDP and World Bank The survey was conducted nation-wide and included a samplescale of 45,945 households in 3,063 communes, 8 regions, urban and rural area andprovinces In that, there are 36,756 households for income survey, 9,189 households forincome and consumption survey Survey was implemented with 9,189 households in 64provinces (6,837 household living in rural and 2352 urban households) The survey forcommune included 2219 communes

This study focus on the households who have positive non-farm self-employmentincome in rural area Then, the volume of rural households which have self-employmentincome is collected Moreover, this thesis considers the impact of infrastructure on non-farm household income So, 2219 communes will be used Combine the volume ofcommunes survey and the rural households which have self-employment income, we have

2117 households without missing value and negative income The self-employment income

is transform into logarithmic form in order to remove possible volatility

3.4 Data Analysis

The independent variables without perfectly multicollinear is an importantassumption for the multiple regression model If the multicollinearity occurs, standarderrors may be inflated Stata will drop one of the variables to avoid a division by zero.Inflation factor method (VIF) is used to test the multicollinear

31