Vietnam Access to Resources Household Survey: Synthesis of findings 2006-2014

Workshop at the Central Institute for Economic Management (CIEM) 19 May 2015, Hanoi, Vietnam

Vietnam Access to Resources Household Survey: Synthesis of findings 2006-2014

• The origin of VARHS dates back to 2002• The results inspired CIEM, CAP/IPSARD, ILSSA and DERG, University of Copenhagen, to expand and

continue the survey in 2006, 2008, 2010, 2012 and 2014• Each survey round saw the publication of an extensive descriptive report and a number of in-depth

studies and policy briefs on issues of importance for rural society and economy in Vietnam• Now, a team of researchers led by Prof Finn Tarp have conducted a series of studies based on the

full panel dataset which synthesises our findings from VARHS to date• These studies, that will be presented at today’s workshop, including a consistent picture of how

Growth, Structural Transformation and Rural Change in Vietnam has impacted on rural households• Possible because following the same set of households over the years and collecting such detailed

data on access to resources, livelihoods and other aspects of rural society allows for an illuminating analysis of structural change in Vietnam and its impacts at the micro level

Vietnam Access to Resources Household Survey: Synthesis of findings 2006-2014Table of contents:Chapter 1: Introduction. Finn TarpPart I: A Rural Economy in TransformationChapter 2: Local transformation – a commune level analysis. Ulrik BeckChapter 3: Agriculture: diversification, commercialization and transformation. Andy McKayChapter 4: The non-farm rural economy. Carol Newman and Christina KinghanPart II: Key Production Factors and InstitutionsChapter 5: Land and land markets. Thomas MarkussenChapter 6: Labour and migration. Gaia NarcisoChapter 7: Technology and Innovation. Heidi KailaChapter 8: Social capital and political connections. Thomas MarkussenPart III: Welfare Outcomes and Distributional IssuesChapter 9: Household welfare. Andy McKayChapter 10: Gender. Carol NewmanChapter 11: Children and youth. Gaia Narciso and Carol NewmanChapter 12: Ethnicity. Saurabh Singhal and Ulrik BeckChapter 13: Conclusion. Finn Tarp

Chapter 2: Local Transformation in Rural Vietnam - A Commune Level Analysis

Ulrik BeckDERG – University of Copenhagen

The commune survey• Structural transformation takes place at many levels

– Household decisions– Government policies– Local level: Supply of basic infrastructure and public goods

• Since 2006, VARHS has included a commune questionnaire which is administered to commune administrators

• Chapter 2: What does the commune survey tells us about the development process at the commune level.

• Many topics: this presentation focused on:– Provision of public goods and infrastructure– Commune problems – past, present and future

Public facilities: Commune facilities• Mekong River Delta: Lowest rates of markets, sec. schools and

clinics – due to proximity to HCMC?• Red River Delta: Highest prevalence• Consistency between indicators – regions doing well in one

indicator is more likely to be doing well in others• Less consistency between relative income level and facilities rates

– Surprising but not necessarily bad: Poorer households in poorer provinces are more dependent on nearby facilities

Public facilities: Infrastructure• Street lighting and drinking water distribution

– Steady progression over time in most regions– The Delta regions are clear leaders in terms of street lighting– Other regions except North have been catching up over the period

• Internet access – at least one internet access point– Substantial progress in internet access over the period in all regions

• 2006: 33% • 2014: 87%

– Communes in the Red River Delta region were early adopters. 2006: 69 percent of communes had a connection to the internet

– North region lacks behind. 2014: 75% of communes had access

Public facilities: Roads, extension shops etc

• In general: Reduced distances to roads, extension shops etc. over the period

• Less improvement in the tail of the distribution: The share of communes which have very long distances to these facilities, is largely unchanged – Divergent experiences: Are some communes left behind?

• The two delta regions: Shorter distances• North region is worst off (mountainous and low population densities)

Commune problems: Past, present, future

• Quite positive: All problems (except climate change) are affecting fewer communes now than before

• Most important problem: natural disasters and diseases

– Has declined– Is expected to decline further

• Climate change issues on the rise: – 2012: 24% of communes reported

climate change as a problem– 2014: 27%– 2016 (expected): 36%

Commune problems by region, 2014• Notable differences between some

regions. • Natural disasters:

– <40% in Red River Delta– >75% in Central Coast

• Power, roads and water – 30% in Mekong River Delta – >55% in North and CH (low population

densities, complicated geography)• Irrigation problems

– 15% in RRD– >30% in North and CH

Conclusion• Signs of transformation at the local level over time

– Improvements in public facilities and infrastructure– Decline in reporting of some common problems

• Different experiences in different regions– Some catch-up is taking place – but not everywhere– Especially North region is lagging behind – and not catching up in all

indicators• Local administrators are increasingly worried about climate change

– Cannot be solved at the local or even national level – but adaptation can be facilitated

Chapter 3:Agriculture and commercialization in rural Vietnam, and related issues

Andy McKay, with Chiara Cazzuffi and Emilie PergeUniversity of Sussex

A first look at income data

• Consistently rising overall real income

• True in both more agriculturally focused and less agriculturally focused provinces

A first look at income data (2)

• High agriculture share in 6 agricultural provinces

• Wage share higher and rising in other provinces

Summary of agricultural activity • Numbers growing rice

falls slowly over time, but proportion of these selling and amount sold increases

• Cash crops fairly unchanged

Growing and selling rice

• More than half of panel households grow rice in all five waves

• Higher in lower quintiles, and in some provinces, including Northern Highlands

Growing and selling rice (2)

• Proportion sold highest in Long An by some way (also Dien Bien, Khanh Hoa)

• Many more consistent sellers in Long An

Growing and selling rice (3)

Cash crops

• Dominance of Central Highlands provinces and coffee; consistent growers

• Richer households more represented

Aquaculture

• Quite a few households undertake aquaculture activities in Dien Bien and Long An, also Lao Cai and Phu Tho

• Those doing aquaculture slightly better off

Common property resources

• Significant in most locations for sales or own consumption especially agricultural and poorer provinces

• Firewood and fishery products most important

• Agriculture still very important in rural Vietnam, even as non-agricultural livelihoods increase

• Rice cultivation falls slightly but commercialisation increases over period of panel

• Poorer households grow more and sell a bit less• Importance of cash crops in Central Highlands, more for richer households• Aquaculture important for some households; and most exploit common

property resources

Summary

Chapter 4: The rural non-farm economy

Christina Kinghan and Carol Newman, Trinity College Dublin

The rural non-farm economy• Paid employment and the operation of micro-enterprises represent increasingly

important sources of income generation in rural Vietnam• In Chapter 4 we:

– Examine the extent of diversification away from agriculture into these activities– Describe the nature of these activities– Examine the impact of diversification on welfare outcomes

• Important for policy to understand:– Impact of diversification on income inequality– Whether specialization in a particular activity, or multiple activities, is more

advantageous to households

Economic activities of households 2008-2014

• Share of households engaged only in agriculture has fallen steadily from 2008-2014 • Structural transformation taking place at the micro-level• Few households fully diversify away from agriculture• Small increase in the number of households specialising in labour or enterprises

Ag Only Labour Only Ent. Only Ag & Labour Ag & Ent. Ag & Labour & Ent. Labour & Ent. No Activity

2008 25.1 4.1 2.4 40.6 11.4 11.5 2.4 2.4

2010 22.4 4.5 3.0 41.9 12.1 10.0 2.9 3.2

2012 20.6 5.7 3.6 43.2 9.4 10.5 2.4 4.7

2014 19.5 5.6 3.6 45.6 6.8 10.4 3.4 4.9

Characteristics of wage employment

• Increasing proportion of households have a household member working• Few have formal labor contracts• Most are employed by other households

  2008 2010 2012 2014HH Members Working        Yes 1,275 (59%) 1,283 (59%) 1344 (62%) 1415 (65%)

Labor Contract        No 793 (62%) 830 (65%) 813 (60%) 837 (59%)Employed by other household        

Yes 911 (72%) 917 (71%) 908 (68%) 964 (68%)Employed by state        Yes 300 (24%) 303 (24%) 328 (24%) 346 (24%)Employed by formal private firm        

Yes 176 (14%) 172 (13%) 263 (20%) 321 (23%)

Who diversifies?• Higher income households are less likely to

transition out of agriculture• All types of income shocks (natural and

economic) are positively related to the probability of transitioning out of specialized agriculture

• This suggests that diversification into other activities might be a mechanism that households use to cope with shocks that affect agricultural production

• Even when income differences are controlled for, ethnic minority households are more likely to transition out of agriculture

  (1)Transitioned out of

agriculture

Lag log(income) -0.028***Lag asset index 0.001Lag hhsize -0.016**Lag ethnic minority 0.150***Natural shock 0.028***Economic shock 0.027**   Time dummies YesHousehold specific means Yes

   Number of households 2,150

Number of observations 6,174

*** p<0.01, ** p<0.05, * p<0.1Only statistically significant results shown

Chapter 4: Conclusions• Diversification from specialized agriculture is happening on a large scale in Vietnam,

although agriculture still remains important• An important role for government is to shape an environment that will help

cultivate enterprises both in terms of ease of starting a business and access to credit.

• Diversification into waged employment is also an important source of welfare gain• A key policy priority should be to enable job creation, particularly in rural areas, for

those leaving agricultural production• Similarly, increasing the quality and quantity of education is likely to have a large

effect on the ability of households to find suitable jobs.

Chapter 5: Land and Land Markets

Prepared by: Thomas Markussen, DERG – University of CopenhagenPresented by: Carol Newman, Trinity College Dublin

Introduction

• Land is an important issue because:– Agriculture still important– High population density land is a scarce resource– Economic development requires frequent changes

of land use and land users important that land markets function smoothly

Introduction

• Chapter 5 considers:– Landlessness– Farm size and land fragmentation– Land sales and rental markets– Land property rights (including effects of

Land Use Certificates on investment)

Landlessness by region

0

2

4

6

8

10

12

14

16

18

20

2006 2008 2010 2012 2014

Perc

ent l

andl

ess

Northern lowlands

Northern uplands

Southern lowlands

Central highlands

All

Farm size by region

0

2000

4000

6000

8000

10000

12000

14000

16000

2006 2008 2010 2012 2014

Med

ian

oper

ated

farm

are

a, sq

m.

Northern lowlands

Northern uplands

Southern lowlands

Central highlands

All

Land fragmentation by region

0

1

2

3

4

5

6

7

8

2006 2008 2010 2012 2014

Mea

n nu

mbe

r of p

lots

ope

rate

d pe

r fa

rmin

g hh

Northern lowlands

Northern uplands

Southern lowlands

Central highlands

All

Land sales by region

0

2

4

6

8

10

12

2006 2008 2010 2012 2014

Perc

ent o

f hh

who

sol

d la

nd in

last

two

year

s

Northern lowlands

Northern uplands

Southern lowlands

Central highlands

All

Renting land out by region

0

5

10

15

20

25

30

2006 2008 2010 2012 2014

Perc

ent o

f hh

renti

ng la

nd o

utNorthern lowlands

Northern uplands

Southern lowlands

Central highlands

All

Land Use Certificates (LUCs) by region

0

10

20

30

40

50

60

70

80

90

100

2006 2008 2010 2012 2014

Perc

ent o

f plo

ts w

ith

LUC

Northern lowlands

Northern uplands

Southern lowlands

Central highlands

All

Land expropriation by region

0

1

2

3

4

5

6

7

8

9

10

2006 2008 2010 2012 2014

Perc

ent o

f hh

expe

lled

from

land

in la

st

two

year

s

Northern lowlands

Northern uplands

Southern lowlands

Central highlands

All

Conclusions

• Landlessness is not increasing in general.• Farms are getting slightly smaller, but plots are being

consolidated.• Land sales market activity is stable. Sales markets much

more active in the Central Highlands than elsewhere.• Activity in land rental markets is increasing. • A significant number of plots remain without an LUC.

Chapter 6: Labor and Migration

Prepared by: Gaia Narciso Presented by: Carol NewmanTrinity College Dublin

Migration

• Chapter 6: Investigates the role of migration as a risk-sharing mechanism.

• VARHS 2012 and 2014:– About 20% of households have a migrant

Migrant HHs vs. Non-migrant HHs.Variable Migrant HH (1) Non-migrant HH (2) Difference (1)-(2)

2012

Age 41.O 43.7 -1.7**

Net income (000 VND) 92,768 84,880 7,888*

Kinh 87.8% 77.4% 10.0***

Economic shock 19.3% 18.0% 0.00

Natural shock 38.6% 31.1% 0.1***

2014

Age 40.7 44.7 -4.1***

Net income (000 VND) 116,009 87,962 28,047***

Kinh 82.1% 78.7% 0.03*

Economic shock 13.7% 13.1% 0.00

Natural shock 25.4% 22.8% 0.03

Who are the migrants?Migrants characteristics All migrants Working migrants t-test of difference

Variable Mean Mean 2012 Male (%) 51.1 58.0 ***Married (%) 30.5 36.7 ***Age at migration 22.5 25.4 ***No diploma (%) 62.4 40.5 ***Years since the migrant left 2.1 2.1 Permanent (%) 25.4 22.8

2014 Male (%) 52.8 57.3 ***Married (%) 27.9 32.2 ***Age at migration 22.6 24.5 ***No diploma (%) 63.7 47.8 ***Years since the migrant left 2.1 2.1 Permanent (%) 19.2 13.8 ***

Reasons for migrating All Migrants Temporary Migrants Permanent Migrants

2012 Work/Looking for work (%) 45.3 46.1 40

Education (%) 35.6 46.5 1.3

Marriage/Family Reunification (%) 13.6 1.1 52.3

Army service(%) 3.8 5.3 1.9 2014 Work/Looking for work (%) 45.5 47.2 24.8

Education (%) 36.6 44.6 1.9Marriage/Family Reunification (%) 10.7 2.3 60

Army service (%) 4.0 4.8 1.0

Remittances• Migrants may send remittances for

– altruistic motives, sense of social responsibility;– as a risk-sharing mechanism, to smooth consumption in the face of external

shocks;– or as a combination of these reasons.– (Maimbo and Ratha, 2005).

• Significant increase in the percentage of migrant households receiving remittances: – 25% in 2012– 45% in 2014

Migrant HHs: features of remittance recipient HHsVariable Remittance recipient HH

(1)Non-remittance recipient HH(2)

Difference(1)-(2)

2012 Age of the HH head 46.6 40.3 6.3***HH size 3.6 4.3 0.7***Net income (000 VND) 87,877 94,465 -6,587Kinh (%) 91.7 86.4 0.1Economic shock (%) 14.2 21.1 -0.1*Natural shock (%) 50.0 34.7 0.2***

2014 Age of the HH head 38.2 42.7 -4.5HH size 4.6 4.2 0.4***Net income (000 VND) 124,826 108,851 15,975*Kinh (%) 83.3 81.2 0.0Economic shock (%) 13.7 13.6 0.0Natural shock (%) 26.2 24.7 0.0

Conclusions

• Significant movements of household members, both intra-province and inter-province

• Main reasons for migrating: education and work related motives• Remittances act as a shock-coping mechanism, in particular in the presence

of natural shocks• Migrant households have better access to the market for credit• Remittance recipient households seem to react better to natural shocks, as

the remittances flows counterbalance for the need for formal borrowing.

Chapter 7: Technology and Innovation

Heidi Kaila,DERG – University of Copenhagen

Agricultural machinery• Agricultural machinery includes:

Pesticide sprayers, tractors, ploughs, feedgrinding machines, ricemilling machines, grainharvesting machines

• Decline in pesticide sprayer ownership in Mekong River Delta and Central Highlands (but large yearly variations)

– Due to overuse problem in pesticides being addressed?• Almost zero ownership of tractors and ploughs, except in Central Highlands. There levels have

stayed constant. • Feedgrinding machines, ricemilling machines, grainharvesting machines:

– The share of households owning at least one machine are at very low levels to begin with, not exceeding over 15% of households in any of the regions.

– The total amount of machinery has stayed at very similar levels over 2006-2014.

Phones

• Rapid increases in the number of phones:

– The median number of phones has increased from zero to 2 just over 2008-2014

– Nearly universal coverage of phones in 2014: 90% of households have at least one.

– The adoption in VARHS provinces has (possibly) been more rapid than in Vietnam on average

• The 10% of households that still did not have a phone in 2014 are:

– Smaller and over twice as often female-headed

– 39% are classified as poor– Dependent on transfers (almost half of the

income)– Have just 4.5 years of education per capita,

half of that of households with phones– Less likely to have electricity, toilet or good

water– Have also less of other technology per capita

(except tv’s)– NOT more remotely located

• Not having a phone strongly related to poverty

Distribution of number of phones per household

Internet & Computers• The share of households owning a computer has grown over five-fold from 2.4% in 2006 to 12.9% in 2014.

– Still a small increase compared to mobile phones• Access to internet (from work, home or internet café combined) has increased less: from 16.1% of households in 2006 to

28.4% in 2014.• The increase in internet subscriptions has been slower in VARHS provinces than in Vietnam on average.• In 2014 households without internet were likely to be:

– Less educated (7.4 years compared to 10.1 per capita)– Income per capita just 2/3’s of that of user households– More likely to be classified as poor, 16% of households– Depend more on transfers (over 20% of income) and income on crops, less on wage income – Generally own less technology per capita (except tv’s)– More remotely located than households with internet, less likely to have toilet and good water– Sligthly more often female-headed

Having internet related to a more urban lifestyle and wealth

Conclusions• Economic growth has not brought about mechanization of the agricultural sector

– Should be given attention in the policy making process, such that solutions are ecologically sustainable.

• The ownership of mobile phones has been catching up on the national average. Internet use is developing much more slowly (but our measure is probably biased downwards).

• Households that still do not own at least one phone in 2014 are often poor, female headed and have fewer members.

• Households that do not have internet are still a majority in 2014. They tend to be more rural: more remotely located, have a smaller income and especially smaller income from wage labour.

Conclusions• Phone and internet connection can be used to acquire and spread information.

Households without these technologies might have a risk of being excluded from economic activity.

• Since new technology creates growth, there is a risk that the income gap in the society widens if poor households and youngsters do not have the same opportunities with new technology– Access to technology and knowledge of its use could be promoted though the

education system– Poor households should have possibilities to use new technologies: Important

as internet cafés are losing popularity and using internet from a mobile phone is becoming more popular.

Chapter 8: Social and Political Capital

Prepared by: Thomas MarkussenPresented by: Ulrik Beck DERG – University of Copenhagen

Introduction

Chapter 8 investigates three forms of social capital:– Linking/Political capital: Communist party

membership, ties with local government officials– Bonding: Family ties– Bridging: Trust in strangers, membership of

organizations

Communist Party membership by region

0

2

4

6

8

10

12

14

2008 2010 2012 2014

Perc

ent o

f hh

wit

h at

leas

t one

mem

ber

Northern lowlands

Northern uplands

Southern lowlands

Central highlands

All

Non-Mass org. membership by income quintile

0

0,05

0,1

0,15

0,2

0,25

0,3

0,35

0,4

0,45

2008 2010 2012 2014

Mea

n nu

mbe

r of n

on-m

ass o

rg v

olun

tary

gr

oups

the

hh b

elon

gs to

Richest

Second-richest

Middle

Second-poorest

Poorest

Share of financial helpers who are relatives, by region(people you can borrow money from in an emergency)

0

10

20

30

40

50

60

70

80

90

100

2008 2010 2012 2014

Perc

ent o

f hel

pers

who

are

rela

tive

s

Northern lowlands

Northern uplands

Southern lowlands

Central highlands

All

Generalized Trust

0

10

20

30

40

50

60

70

80

90

100

2006 2008 2010 2012 2014

Perc

ent "

Agr

ee" "Most people are

basically honest andcan be trusted"

"In this commune onehas to be careful,there are people youcannot trust"

Conclusions

• Communist Party membership strongly correlated with income• Mass organizations dominate organizational life, but other

formal groups (e.g. groups for the elderly) are growing faster• Moderate increase in generalized trust• No signs of decreasing importance of family relations for

economic activities.• Positive effects of party membership, informal insurance

arrangements and trust on hh income

Chapter 9: Welfare dynamics in rural Vietnam: main lessons from the VARHS surveys

Prepared by: Andy McKay & Finn TarpPresented by: Andy McKay, University of Sussex

• 5 wave panel survey opportunity to look at dynamics of change in rural Vietnam and factors associated with this

• Three measures of household welfare can be constructed (comparable across 5 waves)• Food consumption• Income• Assets, summarised by an asset index

• Food consumption and income adjusted for price changes over time (CPI) and by location (VHLSS)

Introduction

Attrition in the panel - Low

• Attrition inevitably occurs in panel; in simple analysis (below) no obvious systematic pattern

Descriptive analysis of welfare• Look at levels of welfare measures in different waves, disaggregated

• Significant differences by province e.g. limited progress in Lao Cai, Lam Dong

• Education strong +ve correlation with welfare levels• Lower welfare levels for minorities and in more remote locations• Households that had migrants leave have higher welfare levels than those

without migrants• These patterns generally the case for income, food consumption and the asset

index• But our interest more in changes…

Average household income per capita, by province

Average food consumption per capita, by selected characteristics

Changes in welfare

• 17.6% of households have food consumption in 2014 more than 20% below it was in 2006

• Corresponding figure for income is 11.6%• Only 7.2% have fewer assets in 2014 than in 2006• Impressive improvements in food consumption and income on

average and overall … but a significant number became worse off over this period

• And there are some systematic patterns, e.g. some regions and minorities…

Average household income per capita, by province

Average household income per capita, by selected characteristics

Econometric analysis

• Model change in different welfare measures from beginning to end and from wave to wave within the panel, as a function of base period characteristics (controlling for province/district)

• Taking account of attrition .. because there may be some systematic patterns (there are)

• Education has a strong positive influence on changes in food consumption/income as do some assets and having migrants in household

• Being from a minority has a strong negative influence

Some lessons

• Considerable value in having a multi-wave panel in which we can measure household welfare

• Overall an impressive period of improvement in all welfare measures• But there is also diversity, not all have shared• And there are some systematic patterns to this

• Education and migration are positive factors• But some regions and ethnic minorities are over-represented among

the cases of decline

Chapter 10: Gender

Carol Newman, Trinity College Dublin

Gender

• Over the last two decades, a number of changes have been made to Vietnamese law to improve the rights and economic situation of women:– The 2003 Land Law – Gender Equality law 2007 to ensure equal rights of women in all

aspects of economic and political life• In Chapter 10 we explore the extent to which there are gender

differences in rural Vietnam and examine how gender relations have evolved over the period 2008 to 2014.

Female Headed Households 2008-2014

• One fifth of households are headed by a woman• These households are very different to male headed households in a

number of respects

– Older age– Less likely to be married - 68% are widowed– Less likely to have children– Have lower levels of education– More likely to suffer economic shocks

Female Headed Households 2008-2014

• They are also worse off in terms of durable goods (2012 and 2014)• While they have significantly smaller land holdings they are more likely than male headed households to have a red book for

their land.• Overall welfare has improved but not by as much as male-headed households• This makes them a vulnerable group particularly in the face of unexpected income shocks.

• They are also less well off in terms of income but not in terms of food expenditure:

  2008 2010 2012 2014

Head of household: Female Male Female Male Female Male Female Male

Income (000 VND) 59,393 71,384*** 69,878 84,695** 72,248 94,742*** 82,080 104,483***

Food exp p.c. (000 VND)326 314 371 342** 468 450 426 416

Durables (000 VND) 4,020 21,204 4,100 9,079 4,485 6,974*** 4,320 6,468***Land area (ha) 4,500 8,837*** 4,244 8,615*** 4,636 8,509*** 4,302 8,288***Red Book 0.85 0.86 0.85 0.80** 0.93 0.88*** 0.94 0.90***N 458 1,716 462 1,719 480 1,701 522 1,659

Cohort Analysis 2008-2014• Moving away from female-headed households to examine situation of women more

generally• Use individual level VARHS data to examine welfare outcomes for four cohorts: i)

18-30 year olds; ii) 31-45 year olds; iii) 46-60 year olds; and iv) individuals over 60.• Consider two measures of welfare:

– Education: i) whether the individual is literate; and ii) the years of education attained by the individual

– Economic activities: amount of time spent engaged in different types of economic activities. Focus on days spent working in household enterprises and waged employment as they are more likely to be associated with an independent source of income for women

Education Outcomes

• Significant improvements education across all age groups for both men and women. • Women outperform men on education outcome in almost all age cohorts in both 2008 and 2014

  18-30 year olds 31-45 year olds

  Female Male Female Male

Individual: 2008 2014 2008 2014 2008 2014 2008 2014

Literate 0.96 0.98 0.93*** 0.94*** 0.91 0.90 0.87** 0.84***Years of ed. 9.22 10.30 8.92** 10.11 7.12 7.85 6.43*** 6.96***n 1,121 1,099 987 946 923 730 1,009 740

  46-60 year olds 61+ year olds

  Female Male Female Male

Individual: 2008 2014 2008 2014 2008 2014 2008 2014

Literate 0.93 0.93 0.88*** 0.90** 0.89 0.92 0.63*** 0.76***Years of ed. 7.22 7.94 5.87*** 7.01*** 5.60 6.77 2.41*** 4.12***n 709 884 746 953 366 460 557 650

  18-30 year olds 31-45 year olds 46-60 year olds  Female Male Female Male Female Male

Individual: 2008 2014 2008 2014 2008 2014 2008 2014 2008 2014 2008 2014

Total days work 146 139 142 123*** 217 195 195*** 178*** 192 161 175*** 140***Days ag

49 26 52 26 90 54 107*** 64*** 101 62 112** 69**Days cpr 6 3 4** 3*** 8 6 6** 4** 6 5 4** 4**Days HH ent 13 12 15 10 33 35 36 41 31 27 39* 31Days wage

79 98 71** 86** 87 101 48*** 69*** 56 68 22*** 36***                         n 1,121 1,102 987 947 923 731 1,009 740 709 884 746 953

Economic activities and time use

• Declines in number of days spent working in agricultural activities in all cohorts for men and women• Increase in average number of days spent in waged employment• Women spend more days working than men in all cohorts mainly due to significantly more days spent in waged

employment• Among 18-30 year olds and 46-60 year olds this gap has widened

Empowerment and welfare• Women are more educated than men but this has always been the case and men’s education

outcomes are catching up with women’s• Women spend more time working than men, particularly in waged employment – women

face a greater burden of responsibility• Working for a wage, however, could empower women by increasing the resources under their

control potentially leading to better welfare outcomes for them and their families• Explore impact of female empowerment on welfare (food consumption per capita)• Measures of empowerment:

– Proportion of total days worked by women in waged employment– Indicator for whether women responsible for making decisions in relation to land– Indicator for whether woman’s name is in the red book

Empowerment and welfare• Households where women are

empowered have a higher level of welfare• Note that model controls for differences

in:– Income– Assets– Marital status, age of household

head, children– Time invariant household

characteristics – Exogenous shocks– General trends in household welfare

*** p<0.01, ** p<0.05, * p<0.1Only results for empowerment variables shown. Models include household fixed effects, year dummies and controls for household characteristics, income, wealth and shocks

Household consumption per capita

  (1) (2) (3) (4)Empowerment measures        

Prop wage work women 0.089***     0.056*

  (0.031)     (0.029)Female manager   0.041**   0.040**    (0.019)   (0.018)Joint property rights     0.050** 0.047**      (0.022) (0.023)

 Observations 6,230 6,630 6,630 6,841Number of HH 1,718 1,775 1,775 1,943

Chapter 10: Conclusions• Female-headed households are a distinct group within VARHS with very different characteristics to

other households– Their welfare has improved but not to the same extent as other households and they are more

vulnerable to income shocks• Focussing on women and men more generally we find:

– Education outcomes improved for both men and women – women outperform men on literacy and years of education but gap is closing over time

– Overall declines in the number of days spent working in agricultural activities and increase in days spent in waged employment consistent with structural transformation

– Women spend more days working than men in all cohorts mainly due to significantly more days spent in waged employment – for 18-30 year olds and 46-60 year olds this gap has widened

• Higher welfare levels in households where women are empowered

Chapter 11: Children and youth

Gaia Narciso and Carol Newman, Trinity College Dublin

Chapter 11: Children and youth

Chapter 11 examines how the lives of the children and youth living in rural Vietnam have been impacted by structural transformation

Cohort Analysis 2008-2014• The VARHS collects detailed information on all individuals in each household including

certain information on children• We examine how children’s welfare has evolved over the 2008-2014 period• We consider three different age cohorts: 5-9 year olds; 10-14 year olds; and 15-18 year

olds• Consider three measures of child welfare:

– Health: an indicator for whether the child suffered an illness or injury in the last 30 days

– Education: i) whether the child attends school; and ii) the years of education attained by the child

– Child labor: amount of time spent engaged in different types of economic activities

Cohort Analysis 2008-2014

• Decline in the proportion of children that experienced an illness• Children over 10 years of age are significantly more likely to attend school in 2014 and have on average more years of

schooling.• Children spend considerably fewer days working at all activities in 2014 in all age groups.• Decline in the number of days children spend doing agricultural activities particularly notable

Cohort: 5-9 year olds 10-14 year olds 15-18 year oldsYear 2008 2014 2008 2014 2008 2014Sick in last 2 weeks 0.10 0.07** 0.07 0.03*** 0.07 0.03***Attends school 0.57 0.59 0.91 0.97*** 0.64 0.75***Years of education 2.07 2.17 5.77 5.91* 8.93 9.58***Total days of work 5.17 1.44*** 21.34 6.70*** 64.64 34.40***Days work ag 3.53 0.99*** 17.23 5.16*** 38.55 15.23***Days work cpr 0.42 0.12** 1.63 0.53*** 3.81 1.92***Days work ent 0.00 0.00 1.04 0.49* 4.41 2.04**Days work wage 1.21 0.32 1.46 0.53 18.14 15.21n 680 778 1,028 836 1,071 738

Are improvements homogenous?

• Education outcomes for children aged 5-9 by 2008 food expenditure quintile

• In 2008 only 50 per cent of the children in the bottom quintile attended school, versus 61 per cent of the children in the top quintile

• While school attendance increases for all groups over time, the difference between the top and bottom quintile remains quite large

  2008 2010 2012 2014Quintile 2008

Attend School

Years of Education

Attend School

Years of Education

Attend School

Years of Education

Attend School

Years of Education

1 0.50 4.00 0.61 4.49 0.59 5.43 0.58 6.582 0.57 4.76 0.68 5.41 0.71 6.20 0.71 7.343 0.63 4.79 0.75 5.48 0.77 6.90 0.74 7.914 0.61 4.95 0.71 5.70 0.78 6.87 0.72 7.845 0.61 4.90 0.74 5.42 0.78 6.59 0.75 7.72

• Middle quintiles seem to catch up over time• While all groups improved their outcomes over time, the

bottom quintile, i.e. the children belonging to the poorest segment of the society in 2008, do not catch up with the other groups

• A divergence in human capital accumulation between the poorest group and the rest may in fact prolong welfare differences over time making it more difficult for them to catch up in the long run.

Comparing boys and girls

• Both girls and boys have experienced improvements in schooling outcomes but these gains have been particularly beneficial for boys.

– Boys under 15 have significantly more years of education than girls of the same age• Declines in the number of days children spend working evident across both girls and boys but

boys experience greater declines than girls– Girls aged 15-18 years spend more days working than boys in 2014

  5-9 year olds 10-14 year olds 15-18 year olds  2008 2014 2008 2014 2008 2014  Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys Girls Boys

Years of education 2.04 2.09 2.05

2.31** 5.79 5.75 5.78

6.04** 9.01 8.85 9.48 9.68

Total days of work 5.99 4.32 1.25 1.65 18.63

24.00* 6.24 7.19 62.96 66.58 36.05

32.75**

Household determinants of child welfare

*** p<0.01, ** p<0.05, * p<0.1Only significant results shown. Models include household fixed effects, year dummies and controls for household characteristics, income, wealth and shocks

Education outcomes for 5-18 year olds  (1) (2)  Attends

schoolYears of

education

Household characteristics:    Age 0.012*** -0.001Higher Ed 0.046* -0.050Income -0.023*** -0.026

Durables 0.004 0.054***Red Book 0.027 0.233***     Observations 9,882 8,784Number of HH 2,100 1,981

• Children with older heads of household are more likely to attend school

• Having a head of household with higher level education (more than second level education) is positively correlated with the child attending school

• A negative correlation is observed between household income and the probability that children attend school.

• Children attain more years of education in wealthier households

Household determinants of child welfare• Children with older heads of hh spend fewer days

working particularly in waged employment• Children in higher income households spend more

days working, particularly in waged work.• Suggests that in higher income households, children

play a role in supporting household income through working

• This comes at expense of school attendance• Number of days worked in agriculture increases as

the land size increases, but at a decreasing rate; opposite relationship emerges for the number of days worked for wage

• Natural shocks lead children to spend more days working in waged employment

• Economic shocks associated with children spending more days working in agricultural activities.

• Suggests that hhs use child labor as a shock-coping

*** p<0.01, ** p<0.05, * p<0.1Only significant results shown. Models include household fixed effects, year dummies and controls for household characteristics, income, wealth and shocks

Child labor outcomes for 5-18 year olds  (1) (2) (3)  Days

workedDays

worked agriculture

Days worked wage

Household characteristics:      

Age -0.463** -0.116 -0.347**Income 5.646*** 0.028 4.315***Land area -0.074 0.355* -0.395***Land area squared -0.002 -0.003*** 0.002***Natural Shock 1.985 0.347 1.925*Economic Shock 2.666 3.039** -0.647       Observations 9,889 9,889 9,889Number of HH 2,100 2,100 2,100

Female empowerment and child welfare

• In households where the woman has responsibility for managing the land children in the 10-15 age group are more likely to attend school

• Note that model controls for differences in:– Income– Assets– Marital status, age of household head,

children– Time invariant household

characteristics – Exogenous shocks– General trends in household welfare

*** p<0.01, ** p<0.05, * p<0.1Only results for empowerment variables shown. Models include household fixed effects, year dummies and controls for household characteristics, income, wealth and shocks

Education outcomes 10-15 year olds

  (1) (2)

  Attends school

Years of education

Empowerment Indicators    

Female Manager 0.033*** 0.087

Prop days female waged employment

0.005 -0.037

     Observations 3,427 3,427Number of HH 1,064 1,064

  (1) (2) (3)  Days

workedDays

worked agriculture

Days worked wage

Empowerment Indicators      

Female Manager 0.137 -1.523 1.077**Prop days female waged employment

-9.506** -9.483*** 1.267

       Observations 3,428 3,428 3,428Number of HH 1,064 1,064 1,064

Female empowerment and child welfare

• In households where woman spend the greater proportion of their working time in waged employment children spend significantly fewer days working in agriculture

• Note that model controls for differences in:

– Income– Assets– Marital status, age of household head,

children– Time invariant household

characteristics – Exogenous shocks– General trends in household welfare

*** p<0.01, ** p<0.05, * p<0.1Only results for empowerment variables shown. Models include household fixed effects, year dummies and controls for household characteristics, income, wealth and shocks

Child labour outcomes 10-15 year olds

Chapter 11: Conclusions• The analysis presented in Chapter 11 depicts a society that has made great progress towards improving child

welfare• Over the span of 6 years, the health of children and young people has improved, school attendance has

increased, in particular for children above the age of 10• There has been a decrease in child labor, which is most notable for the most vulnerable age group. • Many challenges, however, still lie ahead

– Both girls and boys have experienced improvements in health and schooling outcomes, but boys benefitted more than girls

– Well-being has increased over time for both minority and non-minority groups but our analysis highlights the fact that a substantial difference in the level of welfare of children in these groups still remains

– Of particular concern is the widening gap in educational outcomes for the poorest households and for ethnic minorities

– With slower rates of human capital accumulation for the poorest groups in society, convergence in living standards will be more difficult and will take a longer time to attain.

Chapter 12: Ethnic Disadvantage

Saurabh Singhal, UNU-WIDERUlrik Beck,DERG – University of Copenhagen

Ethnic Disadvantage

• Motivation

• 54 officially recognized ethnic groups in Vietnam• Kinh: about 86% of population• Increasing income gap between Kinh & non-Kinh after Doi Moi

reforms– Geographic isolation, poor land quality, education

• How has the gap evolved? • What are the factors that can explain the current gap?

Differences in welfare levels & trendsDaily per capita real food expenditures and income by ethnicity in 1,000 VND, 2006-2014

Constraints: (1) Credit

Constraints: (2) Land Quality

Constraints: (3) Distance

Differences within Non-Kinh

Daily per capita real food expenditures in 1,000 VND, 2008-2014

Summary• Persistent gap between Kinh & non-Kinh – but stable• Constraints:

– non-Kinh households have lower quality agricultural land & lower rates of ownership certificates

– non-Kinh households face more problems producing and selling their agricultural products

– non-Kinh households have worse access to formal and informal credit– non-Kinh households do not appear to be more remotely located than Kinh

households living in the same province– Some evidence of segregation along ethnic lines in social networks– Knowledge of Vietnamese matters

www.wider.unu.eduHelsinki, Finland