Báo cáo khoa học nông nghiệp " SMALL-MEDIUM ENTERPRISES IN THE LIVESTOCK FEED SECTOR IN VIETNAM ( VOLUME II)" ppt

INSTITUTE OF POLICY AND STRATEGY FOR AGRICULTURE AND RURAL DEVELOPMENT
CENTER FOR AGRICULTURAL POLICY

CARD Project 030/06 VIE: Developing a strategy for enhancing the
competitiveness of rural small and medium enterprises in the agro-food
chain: the case of animal feed

SMALL-MEDIUM ENTERPRISES IN THE
LIVESTOCK FEED SECTOR IN VIETNAM

VOLUME II: Feed use by pig and chicken livestock producers

Pham Thi Lien Phuong
1
, Nguyen Thi Thinh
1
, Donna Brennan
2
, Sally Marsh
2
, Bui
Hai Nguyen
1

1
Center for Agricultural Policy, Institute of Policy and Strategy for Agriculture and Rural
Development, Hanoi
2

School of Agricultural and Resource Economics, University of Western Australia









Hanoi, April 2010

i
EXECUTIVE SUMMARY

A survey of 300 pig and chicken producers was conducted as part of CARD Project
030/06 VIE: “Developing a strategy for enhancing the competitiveness of rural small and
medium enterprises in the agro-food chain: the case of animal feed”. This research
complemented an earlier survey of feed mills, and was aimed at obtaining an overview of
animal production patterns and use of feed. It was conducted during November and
December 2008 in six of the same seven provinces where the feed mill survey had been
conducted: Ha Noi and Hung Yen in the north, and Binh Duong, Dong Nai, Long An and
Tien Giang in the south.
This study looked at chicken and pig livestock production, and investigated costs, feed
use and productivity associated with each system. In this document, we report results
obtained on production and selling patterns, characteristics of livestock producing
households and, most importantly, the use of both industrial and raw feeds by production
scale and region. A major contribution of this research is the work done to closely
investigate feed use by producers. Feed conversion ratios (FCRs), and feed cost per kg
liveweight gain, have been calculated for on-farm use of different feeds by producers of

different production scale. This not only provides data on overall livestock production in
Vietnam and the efficiency of feed use, but also provides data linking feed providers
(mills) and feed users (producers).
The results from the data analysis indicate that in the following areas small scale
producers have production indicators that show they can be competitive with large scale
producers.
• The difference between average selling price/kg output and average cost/kg output
was positive at all production scales for both chicken and pig producers.
• Small producers tend to be more diversified in livestock production, whereas larger
producers tend to specialize in production of a main livestock product. Diversification
can be a risk reducing strategy for small producers.
• For chicken production, the survey found no difference in cost of purchased stock
between production scales. For pig production, the price of purchased stock was lower
for small-scale producers (as larger producers were more likely to raise exotic breeds
which are more expensive to purchase).
• For pig producers there was no difference between selling price of the livestock
product for different scales of production, although average batch length was longer
for smaller producers. The sale price of local chicken (more often raised by small
producers) was significantly higher than other breeds.
• The survey found no significant difference in prices paid by producers for raw feeds
by scale of production. For pig producers there was also no difference in the price
paid for industrial feed by scale. However, for chicken producers the price paid for
complete feed was significantly higher for small scale producers.
• Small producers make more use of mixed feed diets, and we found evidence to
suggest that for pig production, feed cost per kg of liveweight gain was significantly
lower for pigs fed mixed feed diets compared to complete feed only diets. However,
this was not the case for chicken production.
The survey also identified some issues associated with small scale livestock production.

ii

• Small producers were more likely to have poorer infrastructure facilities, and also
more likely to report incidences of disease outbreak in the previous 12 months. This
is likely to be associated with poorer infrastructure facilities and lack of adequate
veterinary/livestock health procedures.
• Small producers were also less likely to have a production loan, and more likely to
source loans from private sources rather than banks or other commercial lenders.
• Feed costs as a percentage of total production costs were higher for small scale
producers. This makes them particularly vulnerable to feed price increases.
• Small producers were less likely to be involved in contract arrangements for the
supply of feed and/or sale of livestock products. Exceptions to this generalization are
that small scale producers were found to have contracts for egg supply and porker
production.
Some findings from this producer survey have implications for domestic SMEs producing
livestock feed.
• Few surveyed producers were found to be using domestic brands of industrial feeds.
Foreign brands were clearly preferred by both pig and chicken producers. The main
reason given for this choice by producers was that foreign brands were considered to
be of higher quality and give better productivity. There was no significant difference
found between the price of foreign and domestic brand feeds.
• A comparison was made of FCRs achieved by producers using foreign and domestic
complete feed. For broiler producers, significantly lower FCRs were achieved by
producers using foreign brand complete feed. For pig producers there was no
significant difference in the FCR achieved using foreign or domestic brands. These
results indicate that there may be a quality difference between foreign and domestic
brands for chicken complete feed, but it was not evident from our results for pig
complete feed. However, producers perceive a quality difference, which is sufficient
to influence their choice of industrial feed in favor of foreign brands. This finding
supports results from the survey of feed mills (Vol I) which suggest that improving
quality control is a major issue for domestic mills.
• Larger scale producers are more likely to have contract arrangements for feed supply.

As most of the producers were using foreign brand industrial feeds, we can assume
that they have contracts for feed supply from foreign mills. As livestock production
scale increases in Vietnam, and with it the likelihood of contract feed supply
arrangements, SMEs producing livestock feed could find themselves increasingly
locked out of the feed supply markets.
• There is evidence from the survey of feed mills that SMEs obtain a greater share of
their revenue from production of concentrate, compared to large firms who focus
more on production of complete feed. Our results show that mixed feed diets (which
use concentrate and are more likely to be used by smaller producers) perform well in
terms of feed costs per kg of liveweight gain for pig production. The effective use of
concentrate feed in mixed diets for pig production needs to be further explored by
SME livestock feed producers, through possible direct links with small producers
and/or farmer cooperatives.




iii
ACKNOWLEDGEMENTS

The authors gratefully acknowledge funding for this research work from AusAID’s
CARD Program for Project 030/06 VIE: Developing a strategy for enhancing the
competitiveness of rural small and medium enterprises in the agro-food chain: the case of
animal feed.
The authors acknowledge contributions made to the research work from CAP staff other
than the listed authors, including: Pham Tuyet Mai, Tran Cong Thang, Nguyen Ngoc
Que, Nguyen Do Anh Tuan, Nguyen Anh Phong and Nguyen Le Hoa.
The authors also acknowledge helpful discussions with Mr Le Van Lich (VAFA), Mr
Tran Cong Xuan (VPA), Ms Bui Thi Oanh (MARD) and Mr La Van Kinh (IAS for South
Vietnam) as well as participants in the two stakeholder workshops held in December

2009 in Hanoi and January 2010 in HCMC. Advice on technical issues was received from
the Vietnamese experts mentioned above, and also from Dr. Johanna Pluske (livestock
economics consultant) and Professor John Pluske (animal nutritionist) from Murdoch
University in Western Australia. We also thank Dr Johanna Pluske for her helpful
comments on the draft report.
We also acknowledge and sincerely thank staff of six provincial Departments of
Agriculture and Rural Development who assisted our team to carry out the survey of
producers in their provinces. The effort of the team of enumerators is also highly
appreciated, and finally, we thank the 300 chicken and pig producers for their time and
willingness to support the survey work and give details associated with their businesses.

LIST OF ABBREVIATIONS

ANOVA Analysis of Variance
CAP Center for Agricultural Policy
CARD Collaboration for Agriculture and Rural Development
DARD Department of Agriculture and Rural Development
FCR Feed Conversion Ratio
IAE Institute of Agricultural Economics
IAS Institute of Agricultural Science
IFPRI International Food Policy Research Institute
MARD Ministry of Agriculture and Rural Development
NSD No Significant Difference
SD Standard Deviation
SME Small and Medium Enterprise
VAFA Vietnamese Animal Feed Association
VHLSS Vietnam Household Living Standard Survey
VPA Vietnamese Poultry Association
VND Vietnam Dong

iv

TABLE OF CONTENTS

EXECUTIVE SUMMARY i
LIST OF TABLES
vi
1 Introduction 2
1.1 Objectives and research questions 3
1.2 Survey implementation 3
1.2.1 Classification of chicken households 4
1.2.2 Classification of pig households 4
2 Demographic characteristics 6
3 Chicken production systems 8
3.1 Households hatching their stock 8
3.2 Broiler households with purchased stock 8
3.3 Layer households with purchased stock 10
3.4 Summary of chicken production systems 11
4 Pig production systems 13
4.1 Production activities 13
4.2 Production of porkers 14
4.3 Summary of pig production systems 15
5 Feed use on farms 17

5.1 Chicken feed use 17
5.2 Pig feed use 24
6 Marketing chains 31
6.1 Availability and use of contract arrangements 31
6.2 Suppliers of feed inputs 32
6.3 Market outlets 35
7 Infrastructure and other characteristics of production 37
7.1 Production infrastructure 37
7.2 Diseases and measures for prevention 42
7.3 Access to production inputs 44
8 Production cost and profit 49
8.1 Costs of production 49
8.2 Livestock and its contribution to total agricultural revenue 52
9 Summary of key findings and implications for small-scale livestock producers and
SMEs producing livestock feed 54
9.1 Production systems 54
9.1.1 Breed raised 54
9.1.2 Diversification 54
9.1.3 Production infrastructure 55
9.1.4 Prices paid and received for livestock 55
9.2 Production inputs, supply and distribution channels 55
9.2.1 Prices paid for livestock feeds 55
9.2.2 Choice of industrial feeds 55
9.2.3 Access to and use of credit 56
9.2.4 Supply of industrial feeds 56
9.2.5 Distribution of outputs 56
9.2.6 Use of contract arrangements 56
9.3 Costs of production and revenue 57
9.3.1 Feed costs as a percentage of total costs 57

v
9.3.2 Contribution of livestock revenue to total agricultural revenue 57
9.4 Feed use 58
9.4.1 Feed use for chicken production 58
9.4.2 Feed use for pig production 58
9.4.3 Feed conversion ratios (FCRs) 58
9.5 Competitiveness of smaller livestock producers 60
9.6 Implications for SMEs producing livestock feed 61
REFERENCES 63

vi

LIST OF TABLES

Table 1 Sample size by province and scale classification of broiler and layer chicken,
number of households 4
Table 2 Sample size by province and scale classification of piglet heads, number of
households 5

Table 3 General demographic characteristics of households by major livestock
product, region and scale 6
Table 4 Percentage of household heads with specified education level by livestock
production type, region and scale of production 7
Table 5 Chicken producers hatching their own stock
a
(may purchase as well) 8
Table 6 Breed types of broiler chickens and number of batches per year 9
Table 7 Characteristics of broiler production for those purchasing stock 10
Table 8 Breed types of layer chickens 10
Table 9 Characteristics of layer production for producers with purchased stock 11
Table 10 Structure of pig producing households, % in each category 13
Table 11 Sources of piglets for raising porkers (from own sow or purchased), by
region and scale and type of system, heads per farm per year 14
Table 12 Distribution of piglets purchased for porker production by breed type %14
Table 13 Production indicators for porker production (all sources of weaners) 15
Table 14 Percentage of households using complete feed in each production stage by
region, scale and breed type 17
Table 15 Percentage of households using concentrate feed in each production stage
by region, scale and breed type 18
Table 16 Feeding regimes on chicken farms, percentage of farms using different
feeds: complete only, complete and mixed feed, mixed feed only 19
Table 17 Average daily quantity fed to broiler chickens for each feeding regime: by
region, scale and breed type, grams per day 20
Table 18 Diet composition used by farmers feeding mixed and complete feed 20
Table 19 Feed conversion ratio for broilers by region, scale, breed and feed type 21
Table 20 Effect of brand on FCR from complete only diet, for medium and large
farms 22
Table 21 Feed cost per day and per kg liveweight gain: by region, scale, breed and
diet type 23

Table 22 Mean prices of industrial feed by type, region, scale and brand: ‘000 VND
per kg 24
Table 23 Sample size and percent of farms using complete feed for sow and porker
production: by production stage, region and scale 24
Table 24 Distribution of farms (%) and total kg fed per day, according to feeding
regime (complete only vs mixed), for region and scale by pig type 26
Table 25 Mean diet composition for porkers fed a mixed diet 27
Table 26 Feed conversion ratios for porkers, by feed type, region and scale 27
Table 27 Effect of brand on FCR from complete only diet 28
Table 28 Feed cost per day and per kg porker liveweight gain, by region, scale and
by diet type 28
Table 29 Mean prices of industrial feed by type, region, scale and brand (‘000 VND
per kg) 29
Table 30 Mean prices of some main raw feed by region and scale (‘000 VND per kg)
30

vii
Table 31 Percentage of chicken and pig producing households involved in contract
farming, and nature of the contract 31
Table 32 Percentage of animal feed bought from different suppliers 33
Table 33 Preference of producers for feed produced by foreign and domestic mills
(% households use) 34
Table 34: Percent of revenue from sales to each type of customer 35
Table 35 Percent of chicken and pig producing households with different types of
floor and roof materials for animal rearing infrastructure 38
Table 36 Percent of chicken and pig producing households with different materials
used for walls of rearing places 39
Table 37 Some indicators of production infrastructure (% of chicken and pig
producing households using) 40
Table 38 Types of cooling and warming systems (% of chicken and pig producing

households using) 41
Table 39 Experience of disease outbreak in the last 12 months and types of diseases
for chicken and pig producers (% of households having disease) 42
Table 40 Measures for disease prevention (% of chicken and pig producing
households using) 43
Table 41 Main provider of veterinary service in the last 12 months (% of chicken
and pig producing households) 44
Table 42 Main providers of agricultural extension service in the last 12 months (%
of chicken and pig producing household) 45
Table 43 Percentage of households having loans for livestock production, and main
loan sources in the last 12 months 46
Table 44 Main reasons for not having a loan in the last 12 months (% of chicken and
pig producing households) 47
Table 45 Proportion of total production cost of each cost item for chicken and pig
producers 50
Table 46 Cost of production per kg of livestock output, compared with average
selling price per kg 51
Table 47 Revenue from main livestock product and its importance in overall farm
revenue 52
Table 48. Summary of feed use by chicken households (broiler) 59
Table 49. Summary of feed use by pig households 60


2
1 Introduction
The livestock production sector in Vietnam has for a long time been considered as small-
scale production for the most part, mainly using residual feed products and domestic
labor. In 2006, the share of industrial feed used in the livestock sector was 41.6%
1
, which

is lower than the world average of 48.2% (Wild, 1994, cited in Department of Livestock
Production, 2007) and especially low compared to countries with advanced livestock
industries such as Sweden, Norway, the US, Japan, South Korea at over 80% (MARD,
2007).
Livestock feed has been shown to account for a considerable share of total livestock
production cost, ranging from 75-78.2%, depending on the household scale (IFPRI-
MARD, 2001). As a result, farmers get a limited profit from livestock production: about
1,000 VND per kg of pig output (IAE, 2005) and about 3,000 VND/kg of chicken output
(Dinh Xuan Tung, 2001). High production cost is the most important reason for the low
competitiveness of the Vietnamese livestock sector (IAE, 2005). In addition, livestock
households in the last few years have had to face a lot of risks related to production, often
deriving from disease outbreaks, and unstable market conditions.
There are mixed opinions on the use of residual products and low quality feeds associated
with the small-scale and dispersed production system. Some people support the use of
these feeds as it is believed to reduce the production cost and help make use of local feed
by-products and idle home labor. However larger farm-scale production is preferred by
many, who criticise the former production system for its low efficiency, low quality of
meat, and prevalence of diseases. There are concerns that the quality of feed used by
these small-scale producers is not stable, especially when there are constraints in
managing and controlling the quality of industrial feeds.
This study aimed to look at these two types of livestock production, and investigate costs
and productivity associated with each system. If small scale households prove to be
comparably productive to larger scale production, this supports the idea of maintaining
and promoting this system, as livestock production is often the main livelihood of small
rural households, along with support for larger scale production.
A survey of pig and chicken producers was conducted as part of the overall project with
the aim of obtaining an overview of animal production patterns and use of feed. It was
conducted during November and December 2008 in six of the same seven provinces
where the feed mill survey was conducted: Ha Noi and Hung Yen in the north, and Binh
Duong, Dong Nai, Long An and Tien Giang in the south. Information collected, together

with results from the survey of feed mills, will provide a basis for commenting on the
competitiveness of rural small and medium enterprises in the livestock sector in Vietnam.
The survey of feed mills looked at the supply of industrial livestock feeds, whereas this
survey looks at the demand for animal feed. Results obtained on production and selling
patterns, characteristics of livestock producing households and most importantly, their use
of both industrial and raw feeds are analysed separately by production scale and region.
This information not only provides data on overall livestock production in Vietnam, but
also further helps make the link between feed providers (mills) and feed users
(producers). Thus we have some basis for policy recommendations for small-medium


1
Vietnam’s animal feed processing industry is expected to raise the share of industrial feed in the livestock
sector to 55.5% in 2010, 67.3% in 2015 and 70.1% by 2020 (Strategy for Livestock Development to 2020,
MARD, 2007).


3
feed mill enterprises in their dealings with both small-medium and larger livestock
producers.

1.1 Objectives and research questions
The main objective of this producer survey was to get overall information on production
characteristics of livestock producing households, and more importantly, detailed
information on feed use. Pig and chicken households were selected for the survey as these
are the two most important livestock species in Vietnam in terms of meat production.
During the scoping activities carried out for the study, it became apparent that small scale
livestock producers were often regarded as those who usually use raw combined feeds,
low-protein materials (sometimes with contaminants), lacked adequate quality control and
had poor equipment. On the contrary, large scale livestock producers were thought to

mainly use industrial feed for their livestock. It is difficult to conclude which feeding
pattern is more economical for households, as the cost paid for more expensive industrial
feed may or may not be covered by reduction in the length of time the stock are raised
(that is, greater feed efficiency).
Therefore, in this study, the aim was to gain a deeper insight into feed use by different
types of households in terms of scale, region, and breed type raised. It was hoped that
these insights would provide some useful suggestions for SMEs in the feed production
sector to survive in the competitive livestock feed market.
The research therefore focused on answering the following questions:
• How different is the livestock production system and production infrastructure
between households by scale and region?
• Are the marketing chains used for input procurement and output distribution by
household types different?
• How do households differ with regard to feed use: industrial versus raw/combined
feed, Feed Conversion Ratios, feed cost per kg liveweight gain?
• How does production cost and profit vary between livestock household groups?
• Is there an opportunity for small and medium producers to directly access feed
mills, especially SMEs?

1.2 Survey implementation

The livestock producer survey was implemented in six of the seven provinces where the
feed mill survey had been done previously: Ha Noi and Hung Yen in the north, and Binh
Duong, Dong Nai, Long An and Tien Giang in the south. In total, 300 livestock
producing households were surveyed: 50 households were interviewed in each province
and these were almost equally divided between chicken and pig producing households.
The sampling design for the 300 livestock producing households was based on the
VHLSS 2006 framework, so that it would be representative of the overall production
situation. Within each province, the aim was to interview 25 households producing
chickens and 25 producing pigs. With consultation from the Livestock Division of

provincial Departments of Agriculture and Rural Development (DARDs), we selected
one of the biggest livestock producing districts in each province. From that district,

4
selection was narrowed to about four communes listed in the framework of VHLSS 2006
and having large numbers of households involved in chicken/pig production. However, in
many cases we did not find enough households on the VHLSS 2006 list, as some of them
at the survey time were no longer involved in livestock production, or were too distant to
be surveyed because of our time and budget constraints. Instead, additional households
were selected randomly for the survey, given their expected similarity to other producers
in the survey sites.
The report mainly focuses on the production differences between pig and chicken
producing households by scale and province, with more analysis of their use of industrial
feed versus raw feed, and feed conversion ratios (FCRs) for livestock production.
Evidence of different supply and marketing chains and production infrastructure is also of
interest.
1.2.1 Classification of chicken households
The total survey sample of chicken households was 153, and they were involved in
producing a variety of chicken types including broiler, layer, baby chicken, breed hen and
rooster. We found that the larger the households, the more likely they are to be raising
only one type of chicken output, while the smaller scale households tend to produce more
than one type. As there were multiple types of chicken production, we categorized the
farms into two main types, broiler and layer, depending on which of these produced the
greatest revenue on the farm. We further classified each of these farms into scale
categories, based on production over the 12 months between November 2007 and October
2008, guided by industry conventions and the need to distribute the sample fairly evenly
across the three size classes. For broiler production, our groups were: small size (<=500
heads); medium size (>500 and <=3000 heads); and large size (>3000 heads). For layer
production we used: small size (<=1000 heads); medium size (>1000 and <=4000 heads);
and large size (>4000 heads). The distribution of surveyed households by production type

and by scale is shown in Table 1, for the north, the south, and in total.
From Table 1, it can be seen that generally more households are engaged in producing
broiler than layer chickens. Broiler farms from the south were more likely to be in the
large category. There were no medium-scale layer farms surveyed in the north.

Table 1 Sample size by province and scale classification of broiler and layer chicken,
number of households


Total
Broiler
a
Layer
b



Small Medium Large Total Small Medium Large Total
North 51 21 19 2 42 4 0 5 9
South 102 17 18 31 66 14 13 9 36
Total
153
38 37 33 108 18 13 14 45
a For broiler production: small is less than 500 head, medium is between 500 and 3000 head, and large is
more than 3000 head
b For layer production: small is less than 1000 head, medium is 1000 to 4000 head, and large is more than
4000 head
1.2.2 Classification of pig households
A total sample of 149 pig producing households in the six provinces were interviewed.
Similar to the case of chicken, we classified production scale of pig households as small,

medium or large on the basis of their piglet heads raised during one year from November

5
2007 to October 2008. For pig producers small size was having raised less than 49 pigs,
medium size was from 49 to 120 pigs, and large more than 120 pigs raised during a year.
The number of households in each region and by production scale is shown in Table 2.
There were relatively more small and relatively less large firms in the north compared to
the south.

Table 2 Sample size by province and scale classification based on piglet heads raised,
number of households

Small
a
Medium
b
Large
c
Total
North
22 16 12 50
South
29 33 37 99
Total
51 49 49 149
a Small is <49 head
b Medium is from 49 to 120 head
c Large is >120 head

6

2 Demographic characteristics

The demographic characteristics of the surveyed households including average household
size, gender, and the allocation of labour are shown in Table 3. The data are grouped
according to livestock type, and by region and scale. Pig producers were more likely to
have a male household head, and farms in the north were more likely to have a male head.
All households surveyed are Kinh people.
The number of household members working on-farm is much higher than the number of
members working off-farm (nearly 3 times higher in general). Chicken producers tended
to have few people working on farm if farms were smaller, and this difference is
statistically significant at the 5% level between the small group and the large group.
There was no difference between production scale in the number of household members
working on pig farms.
We asked for information on the amount of time spent working off farm for the most
educated household member. On average, over 70% of the time of the most educated
person is spent working off-farm. The most educated household member tended to spend
more time working off farm if they were pig producers.
Table 3 General demographic characteristics of households by major livestock
product, region and scale

Sex HH head Average household size

(% male) Total
Working
on-farm
Working
off-farm
% of time worked
off-farm by the
most educated

household
member
Chicken producers

Overall 79.6 4.4 2.2 0.7 70.7
By region
North 84.0 4.3 2.2 0.6 74.0
South 77.5 4.5 2.1 0.7 69.0
By broiler scale

Small 81.6 4.0 1.8 0.7 77.5
Medium 83.8 4.5 2.1 0.8 65.4
Large 75.8 4.6 2.4 0.6 81.0
Pig producers

Overall
87.5 4.5 2.1 0.7 73.1
By region
North 100.0 4.4 2.2 0.6 80.8
South 80.9 4.5 2.1 0.7 69.6
By scale
Small 84.3 4.4 2.1 0.9 78.3
Medium 87.0 4.5 2.1 0.7 67.1
Large 91.5 4.6 2.2 0.5 74.6
Note: Only on-farm labor (for chicken) is significantly different at 5% level between small and large scale.

The education level of the household head is shown in Table 4. More than half the
household heads had a secondary school education. Household heads in the south were
more likely to have only completed primary school, and this was the case for both

7
chicken and pig households. For chicken households, the smallest scale group had a lower
share of household heads in the primary education category and more with secondary
education. This is consistent with larger scale households being in the south, and with the
lower education levels observed there for the sample farms. Only a small percentage of
the households had attained an education level such as vocational training or
bachelor/post graduate degree, and there was not much difference between pig and
chicken producers.
Table 4 Percentage of household heads with specified education level by livestock
production type, region and scale of production


None/ Primary
school
Secondary
school/ high
school
Vocational
training
Bachelor/ Post
graduate
Chicken producers
Overall 25.7 65.8 4.0 4.6
By region
North 14.0 74.0 8.0 4.0
South 31.4 61.8 2.0 4.9
By broiler scale
Small 15.8 76.3 5.3 2.6
Medium 32.4 59.5 5.4 2.7
Large 36.4 57.6 3.0 3.0

Pig producers

Overall
21.0 73.7 2.0 3.4
By region
North 6.0 86.0 4.0 4.0
South 28.6 67.4 1.0 3.1
By scale
Small 23.5 70.6 3.9 2.0
Medium 20.4 73.5 0.0 6.1
Large 18.8 77.1 2.1 2.1



8
3 Chicken production systems

This chapter describes two chicken production systems: broiler and layer. As it is difficult
to integrate those hatching and not-hatching their stock in one table and for analysis, the
first section only discusses information on the production system of households hatching
their stock, though there was only a very few number of households with this production
system. The second and third sections will consider households who only use purchased
stock, separately for broilers and layer chicken. Finally, a short summary of chicken
production systems is given.

3.1 Households hatching their stock
Only 13 out of 108 broiler farms hatched their own stock, and two layer farms out of 45
(these farms may purchase stock as well). Self hatching was more common in the south
for broilers. The two layer farms hatching their own stock were in the north. Only small
farms, and a few medium farms hatched their own stock. Broiler farms in the south

tended to also purchase stock, with the mean number of heads purchased being of similar
magnitude to mean number hatched. The selling weight of birds in the north was
significantly greater than in the south, at 2.3 kg versus 1.7 kg respectively. Chicken
selling prices seem to be a little higher in the south than in the north, but it is not
statistically significant.

Table 5 Chicken producers hatching their own stock
a
(may purchase as well)


No. of farms
raising
Broiler data

Broiler Layer
Average number
hatched (head)
(s.d in brackets)
Broilers
purchased over
year (head) (s.d in
brackets)
Broilers
sold
(head)
Selling
weight
kg
Selling

Price
(‘000d/kg)
By province

North 3 2 290.0
(185.2)
- 350 2.3* 65.0
South 10 181.3
(155.7)
217.0
(87.8)
153 1.7 66.4
Overall 13 2
203.1
(161.2)
217.0
(87.8)
196 1.8 66.1
By scale
Small 10 2 156.6
(117.2)
217.0
(87.8)
114 1.7 65.9
Medium 3 0 383.3
(202.1)
-
493 2.1 66.7
Large 0 0 -
-

- - - - -
Overall 13 2
208.9
(164.3)
217.0
(87.8)
196 1.8 66.1
a In addition to this data and not reported here, one farmer was a specialist hatching farm which produced
150,000 baby chickens for sale.
* Significantly greater at 5% level.

3.2 Broiler households with purchased stock
The majority of broiler households produced using purchased stock, and collection of
data from this subset of producers was simpler and included data on the main type of
breed used and on the number of batches per year. Results are shown in Table 6. In
overall terms, half the producers kept local chickens. Producers in the north tended to
keep crossbred chickens (46.2%), whereas in the south exotic breeds
2
were the main other


2
Exotic breeds are those breeds sourced from overseas.

9
type of bird kept (36.4%). The type of breed raised is also influenced by scale, with large
producers more likely to keep exotic birds (51.5%) while local breeds are mostly raised
by small households (66.7%).
The number of batches per year was considerably lower for local birds, both at the overall
sample level, and within categories for all cases except for small scale producers. There

was no significant difference in number of batches per year between breed types for small
scale broiler farms. Those large farms that produced crossbreds had significantly more
batches per year than producers of exotic breeds (nearly 6 batches versus 4 batches per
year).
Table 6 Breed types of broiler chickens and number of batches per year


No. Share of hh by main
breed type (%)
Number of batches per
year by main breed type

n
Local Exotic Crossbred Local Exotic Crossbred
Anova
results
Overall
94 51.1 23.4 25.5 2.5 4.0 4.1
L<E,C
1%
By region


North
39 48.7 5.1 46.2 2.4 5.0 4.3
L<E,C
10%
South
54 52.7 36.4 10.9 2.7 3.9 3.7
L<E,C

1%
By broiler scale


Small
27 66.7 7.4 25.9 2.4 3.5 2.7
n.s.d
Medium
34 55.9 8.8 35.3 2.4 4.3 4.3
L<E,C
1%
Large
33 33.3 51.5 15.2 3.1 4.0 5.8
L<E<C
5%

Production statistics are shown in Table 7 for the farms producing broilers from
purchased stock. Anova results for comparisons of reported means between groups are
shown in the table footnote. Average batch size for the overall sample group was 2,131
birds, with significant variation at the regional level, with producers in the south being
much larger scale. This is consistent with the result by breed, as farms with exotic breeds
tend to be of larger scale and these were more common in the south. There were no
significant differences between any of the group comparisons for the price paid per head
of stock. Mean batch length was 98 days, with no significant difference between regional
means. Large farms had a mean batch length of 77 days which was significantly shorter
than the other scale farms, and local chickens had a longer batch length cycle (125 days)
compared to the improved breeds (65 days for exotic and 73 for cross). Mean quantity of
broilers sold for the year was 4,511 for the sample overall, with a significantly larger
mean in the south. Mean quantity sold was significantly lower for farms with local
chickens, at 1,907 compared to more than 7,000 for the other breeds.

Mean selling prices are reported in thousands of VND per kg. Local chickens were sold at
a significantly higher price at 54.5 compared to the overall mean of 44.1, and the mean
price of improved chickens (i.e. exotic and crossbreds) of around 35. Prices received for
small scale farms were significantly higher, reflecting the dominance of local chickens in
this group, and prices were lower in the south (41.5) compared to the north (47.5).




10
Table 7 Characteristics of broiler production for those purchasing stock


Average
size per
batch
(heads)
a

Cost of
stock
(000VND/
head)
b

Average
length of a
batch
(days)
c

Average
selling
weight
(kg)
d

Quantity
sold
(heads)
e

Selling
price
(000VND/
kg)
f

Overall 2131 10.1 98 2.2 4511 44.1
By region
North 218 10.6 104 2.5 690 47.5
South 3488 9.6 93 1.9 7491 41.5
By broiler scale
Small
103
12.1
116
2.2
196
53.7

Medium
483
9.9
104
2.2
1060
44.9
Large
5489
8.1
77
2.2
12862
34.2
By breed type
Local
828
11.1
125
1.8
1907
54.5
Exotic
5052
9.7
65
2.5
7899
35.3
Cross

2060
8.1
73
2.7
7316
35.4
a. Region- N<S (15); Scale- S,M<L (1%); E > L,C (10%)
b. Means not significantly different between groups
c. Region- n.s.d; Scale- S,M>L (1%); Breed L>E,C (1%).
d. Region – N>S (1%); Scale- n.s.d; Breed- L<E,C (1%)
e. Region N<S (1%); Scale S,M<L (1%); Breed- nsd.
f. Region- N>S (10%) ; Scale - S>M>L (10%); Breed- L>E,C (1%)


3.3 Layer households with purchased stock
Breed types for layer chickens are shown in Table 8. Around 74% of all layer farms (with
purchased stock) in the sample were keeping exotic breeds. Farms in the north and
smaller farms were more likely to have crossbreds. Very interestingly, around 21% of
large scale farms had local chickens as the main layer breed, which is higher than the
small group (12.5%), and no producers from the medium group were involved in
producing local layer chicken.

Table 8 Breed types of layer chickens


Share of hh by main breed type (%)

Number of
households
involved (n)

Local Exotic Crossbred
Overall
43
11.6 74.4 14.0
By region


North
9
11.1 66.7 22.2
South
34
11.8 76.5 11.8
By layer scale


Small
16
12.5 62.5 25.0
Medium
13
0.0 84.6 15.4
Large
14
21.4 78.6 0.0


11
Characteristics of egg production are shown in Table 9. Average number of heads in a
layer batch was 3,323 in the overall sample, but significantly higher in the north. There

was no difference for batch size between breed types kept. Cost of stock was around 44
thousand VND per head, and was not significantly different for any of the categories
analyzed. Layer hens were generally kept for around 14 months, although batch length
was significantly shorter in the north, with a mean of around 12 months.
There were no other statistically significant differences between groups for length of
batch. The number of eggs produced per hen per year was 228 for the overall sample, and
production per hen was significantly lower for small farms compared to large scale farms.
Local hens produced fewer eggs per hen than exotic hens, with the means being 170
compared to 244, and these differences were significant at the 10% level. Farms in the
north tended to produce more eggs and have larger revenue per farm than in the south.
Differences in total egg production and revenue were not significantly different according
to type of breed on the farm.

Table 9 Characteristics of layer production for producers with purchased stock


Average
size per
batch
(heads)
a

Cost of
stock
(000VND/
head)
b

Average
length of

a batch
(days)
c

Number
of eggs
per hen
d

Quantity
sold
('000
eggs)
e

Egg
revenue
(m VND)
f

Overall 3323 43.9 442.0 228 709 792
By region
North 5556 55.0 351 214 1252 1412
South 2732 41.0 466 231 573 637
By layer scale
Small 667 34.2 406 197 121 126
Medium 2517 42.8 443 231 478 533
Large 7107 56.0 481 265 1680 1889
By breed type
Local 4500 39.4 419 170 1095 1234

Exotic 3566 46.7 456 244 777 865
Cross
breed
1044 33.0 387 217 263 296
a. Region S<N (5%); Scale S<M<L (10%); Breed: nsd
b. Means not significantly different between groups
c. Region N<S (5%); Scale nsd; Breed nsd
d. Region nsd; Scale S<L (5%); Breed L<E (10%)
e. Region N>S (5%); Scale L>S, M (1%); Breed nsd
f. Region N>S (5%); Scale L>S, M (1%); Breed nsd


3.4 Summary of chicken production systems

Southern broiler chicken households have a much larger production size than their
counterparts in the north. This is consistent with the result by breed, as farms raising
exotic/cross breeds tend to be of larger scale and these were more common in the south
(36.4% of households involved). However, for broiler production, local breeds are
generally still the most common type (over 50% of households) while exotic is the main
breed for layer households (74%). Breed is also influenced by scale, with large producers

12
more likely to keep exotic breeds for broiler production (51.5%) while local breeds are
mostly raised by small households (66.7%).
Generally, those keeping local birds have considerably fewer batches per year because of
much longer batch lengths than for the other two breed types. There were no significant
differences between any of the size groups regarding the price paid per head of stock, but
chicken selling prices were different. Local chickens were sold at a significantly higher
price of 54.5 thousand VND per kg compared to the overall mean of 44.1, and the mean
price of exotic and crossbred chickens of around 35. Prices received by small scale farms

were significantly higher, reflecting the dominance of local chickens in this group, and
prices were lower in the south (41.5) compared to the north (47.5).
Unlike broiler production, layer households had significantly larger scale in the north than
in the south, which is reflected in a larger egg quantity sold and revenue. Egg productivity
per hen is not statistically different by region, but it is significantly higher for large scale
and exotic breed-raising households. Average length of a batch is statistically longer in
the south (466 days versus 351 days), reflecting the regional dominance in the south of
exotic breeds which usually give higher productivity than local and crossbred birds.

13
4 Pig production systems

4.1 Production activities
The survey identified three distinct production patterns:
1. Raising piglets from farrow to weaning
2. Raising piglets from farrow to finishing
3. Buying in piglets for producing porkers (weaned piglets to finisher)
Pig households are involved in one type of production pattern or more. Households
having pattern (1) or (2) must have sows, but may also purchase extra piglets for
producing porkers. Those without sows can only produce system 3. The distribution of
households according to whether they produce piglets from sows, buy them in, or do both
is shown in Table 10. About 13% of households raising sows also purchase piglets,
however more households in the north are involved in this combined pattern than
southern ones. The majority of the pig farms surveyed (83%) had sows, although it was
more common to have sows in the south than in the north (85% vs 78%). The share of
households with sows increases by production scale, ranging from 67% in the small group
to 94% in the large group. Generally, about 29% of households purchase piglets to raise.
Buying-in piglets was more common in the north, with 48% of all farms practicing the

wean-to-finisher system, compared to only 19% in the south. Thirteen percent of farms
surveyed practiced both farrow to finish and wean to finish systems, supplementing their
own produced piglets with purchased piglets. This practice was most common in the
medium scale farms, and in the north.

Table 10 Structure of pig producing households, % in each category


Number of
households
(n)
Households
with sows %
Households
purchase
piglets to raise
%
Households
having both
sows and
purchase
piglets %
By region
North 49 78 48 26
South 99 85 19 6
By scale
Small 51 67 37 8
Medium 48 88 31 18
Large 49 94 18 12
Overall 148

83 29 13

In Table 11, sources of weaners for porker production are outlined for each of the three
combinations possible: self produced, purchased, or a combination of both sources. The
number of pigs born per farm in the south is significantly higher than in the north, and
this reflects larger scale farms (i.e. higher number of sows) in the south. The mean
number of piglets born on farms sourcing only from their own sows was 29 for small, 83
for medium and 504 for large farms. Those practicing a combination of raising self-
produced and purchased piglets tend to have smaller numbers born on farm, except in the
case of small farms where births are similar to numbers purchased. On average, these
farms tend to purchase fewer weaners than the farms that rely solely on purchasing
weaners.

14
Table 11 Sources of piglets for raising porkers (from own sow or purchased), by
region and scale and type of system, heads per farm per year


Sow only Sow + buy Buy only
n Born n Born Bought Born % n Bought
Overall 101 236.3 11 82.8 37.7 69% 24 97.7
By region
North 26 94.9 8 79.5 35.4 69% 12 151.0
South 75 285.3 3 91.7 44.0 68% 12 44.3
By scale
Small 25 28.6 3 28.3 11.3 71% 16 29.2
Medium 36 82.9 3 47.0 49.7 49% 6 59.5
Large 40 504.1 5 137.0 46.4 75% 2 760.0

For the subset of farmers who only purchased their piglets, we asked about the breed, and

the distribution of farms by main breed purchased is shown in Table 12. Overall, 55% of
households who purchased piglets raise crossbred pigs, followed by exotic breeds (31%)
and local breeds (15%). However, large enterprises are more likely to raise exotic breeds
(56%), while medium and small producers are more likely to raise crossbreds (60% and
71% respectively). Local breeds were more common in the south, and crossbreds more
common in the north (70%).

Table 12 Distribution of piglets purchased for porker production by breed type, %
of farms purchasing piglets

Local Exotic Crossbred
Overall
15 31 55
By region
North 5 25 70
South 19 33 48
By scale
Small 13 17 71
Medium 15 25 60
Large 17 56 28


4.2 Production of porkers
Data representing the production characteristics of porkers (for all weaners produced on-
farm and purchased) is shown in Table 13. The results of analysis of variance tests on the
mean values reported for each group are shown in the table footnote. The average batch
size was significantly smaller on the northern farms we surveyed, with 23 heads per batch
compared to 70. Small farms produced 9 heads per batch and large farms had more than
100 heads per batch. The mean size of weaners bought for porker production ranged
between 16 and 20 kg but the means were not significantly different between groups. The

cost of these weaners was 58,000VND per kg overall, but was significantly larger (77,000
VND/kg) for large farms, who were more likely to produce exotic pigs. The production
length on the surveyed farms was longer in the north (114 days) than in the south (89
days) and the final weight of the pigs was 5 kg heavier in the north. The selling price per
kg of the porkers was significantly higher in the south (34,000 VND/kg) compared to the

15
north (30,000 VND/kg). There was no significant difference in length of batch, final
weight or sale price between the different scales of production.

Table 13 Production indicators for porker production (for all weaners produced on-
farm and purchased)

Average
size per
batch
a
(heads)
Average
weight of
pig if
purchased
b

(kg)
Cost of
stock if
purchased
c

(000VND/kg)
Average
length of a
batch
d

(porker day)
(days)
Average
selling
weight
e

(kg)
Sale price
f

(000VND/kg)
Overall 51.8 19.1 57.8 97.9 93.2 32.5
By region
North 22.9 18.4 62.1 114.0 96.6 30.0
South 69.7 20.1 50.5 88.6 91.5 34.1
By scale
Small 9.0 20.5 51.5 95.0 91.4 31.6
Medium 32.5 19.6 52.7 102.0 93.7 32.2
Large 117.4 15.8 76.6 96.6 94.3 33.6
a Batch: Region S>N (1%), Scale S< L (1%);
b Purchased weight: nsd
c Purchased price: region nsd, S<L (1%)
d Days : N>S (1%), scale nsd

e Weight sold: region 5%, Scale nsd
f Sale price: N<S (1%) Scale nsd

4.3 Summary of pig production systems

The majority of the pig farms surveyed (83%) had sows, but it was more common in the
south and for larger producers, indicating their higher self-stock producing capacity.
However more households in the north tend to be involved in combined patterns (i.e.
owning sows plus purchasing piglets to raise) compared to southern ones (26% versus
only 6%).
There is more clear evidence of industrial production in southern households than
northern ones: they have larger average batch size (70 heads compared to 23 heads) and
also a larger number of pigs born per household (as they have more sows). The difference
is even bigger when comparing households by production scale, with only 9 heads per
batch in small households versus more than 100 in large farms.
Crossbreds are the most common type, which are raised by 55% of households
purchasing piglets, followed by exotic breeds (31%). Local breeds are no longer raised as
much as previously. However, large households are more likely to raise exotic breeds
(56%), while medium and small producers are more likely to raise crossbreds (60% and
71% respectively). Interestingly, local breeds were more common in the south where
industrial production is more common, while crossbreds are more common in the north
(70%).
The production length was relatively longer in the north (114 days) than in the south (89
days) and the final weight of the pigs was 5 kg heavier in the north. The shorter
production time may also be seen as another evidence of industrial production in southern
households.

16
The cost of stock was 58,000VND per kg overall, but was significantly higher (77,000
VND/kg) for large farms, who were more likely to produce exotic pigs. There was no

statistically significant difference between households by scale in the selling price per kg
for porkers, though it was higher in the south (34,000 VND/kg) compared to the north
(30,000 VND/kg).

17
5 Feed use on farms

5.1 Chicken feed use
We collected information on feed use at different stages of production, and report on the
prevalence of industrial feed use by stage in Table 14 and Table 15. Table 14 shows the
use of complete industrial feed for broiler and layer production. Farmers are generally
more likely to use complete feed for broiler production in the first stage of production
(which had an average length of 29 days). Over the entire sample, 90% of farmers used
complete feed in the initial stage, but the percentage of farms using complete feed is
lower in the second and third stage of production. In the north, the difference in the use of
complete feed between the stages is much more dramatic, with a halving of the
percentage of farms using complete feed in stage 2 compared to stage 1, and this number

again almost halved in stage 3 compared to stage 2. Farms in the south are more likely to
continue to use complete feed throughout the production cycle. Use of complete feed for
broiler production is affected by scale of operation, with fewer small farms using
complete feed, particularly in the later production stages. Broilers by all breed types are
more likely to be fed complete feed in the first stage of production. We also divided the
local chicken group into free range, and not free range, as the sample was split about
evenly between these two categories. Free range chickens were less likely to be fed
complete feed, particularly in later production stages.
While layer hens are less likely to be fed complete feed in the first stage compared to
broilers, they are more likely to be fed complete feed in later stages. In the north 100% of
layer farms fed complete feed at all stages.
Table 14 Percentage of households using complete feed in each production stage by
region, scale and breed type

Broiler

Layer


Stage 1
(29
days)
Stage 2
(48
days)
Stage 3
(40
days)
n
Stage 1

(94
days)
Stage 2
(172
days)
Stage 3
(250
days)
n
Overall 90 61 51 98 82 80 74 37
By region
North 83 39 26 35 100 100 100 9
South 94 77 78 63 77 75 70 28
By scale
Small 81 26 3 30 71 67 64 13
Medium 89 63 69 33 85 92 83 11
Large 100 97 95 33 93 85 83 13
By breed type
Local
85 61 46
42
100 75 50 5
Free range
75 52 19
19
100 100 0 1
Not free range
96 70 68
23
100 67 100 4

Exotic
100 90 91
21
81 86 71 27
Cross
100 61 63
24
67 80 100 4

18
Table 15 shows the use of concentrate industrial feed for broiler and layer production.
The patterns of use of concentrate feed by stage are consistent with observations of
complete feed use, given that they are substitutes in production. It is less likely that
broiler chickens will be fed concentrate in stage 1 when complete feed is predominately
used. In the later stages, and particularly in the north and for small scale producers where
complete feed is less commonly used, the incidence of concentrate feed use increases.

Table 15 Percentage of households using concentrate feed in each production stage
by region, scale and breed type

Broiler Layer

Stage 1
(29
days)
Stage 2
(48
days)
Stage 3
(40

days)
n
Stage 1
(94
days)
Stage 2
(172
days)
Stage 3
(250
days)
n
Overall
6 32 36 39 14 13 17 12
By region

North
10 54 54 27 0 0 0 0
South
3 17 17 12 17 16 20 12
By scale

Small
5 56 79 26 18 13 18 5
Medium
11 31 12 11 15 8 17 3
Large
0 6 5 2 7 15 17 4
By breed type

Local 9 35 37 21 0 25 0 1
Free range
17 43 69 14 0 0 0 0
Not free range
0 26 11 7 0 33 0 1
Exotic 0 10 9 2 16 14 29 10
Cross 0 35 32 10 17 0 0 1

For the remainder of this analysis we focus on feeding patterns over the entire production
cycle. In Table 16, the distribution of broiler and layer farms is shown according to the
feed combination that was used over the entire cycle. Those using complete and mixed
feed (which can include concentrate feed, premix, corn, rice/broken rice, bran, vegetable,
etc.) would usually have used complete for early stages and then switched to mixed feed,
although there were a few cases reported where the farmers actually put complete feed
into a mixed feed. At the aggregate level 9% of broiler farmers used only mixed feed, and
53% used only complete feed. Complete and mixed feed in combination was the most
common feed used for broiler farms in the north, and complete feed only was the most
common for farms in the south. Small broiler farms were most likely to use a combination
of mixed and complete feed, and large broiler farms were most likely to use complete
only. Exotic broilers were most commonly fed complete feed, and local chickens were
most commonly fed mixed feed. Within the local chicken category, the use of complete
feed only was more common for broilers that were not scavenging, around 50% of these
were fed complete only.
At the aggregate level, layer chickens were more likely to be fed complete feed only, with
64% in this category compared to 53% for broilers at the aggregate level. This reflects the
tendency to maintain complete feed in all production stages for layer chickens. However,