Feed the Future Innovation Lab for Food Security Policy

Feed the Future Innovation Lab for Food Security Policy Research Paper 52

May 2017

Food Security Policy Project (FSPP)

AQUACULTURE IN MYANMAR: FISH FARM TECHNOLOGY, PRODUCTION ECONOMICS AND MANAGEMENT By Ben Belton, Mateusz Filipski and Chaoran Hu

Food Security Policy Research Papers This Research Paper series is designed to timely disseminate research and policy analytical outputs generated by the USAID funded Feed the Future Innovation Lab for Food Security Policy (FSP) and its Associate Awards. The FSP project is managed by the Food Security Group (FSG) of the Department of Agricultural, Food, and Resource Economics (AFRE) at Michigan State University (MSU), and implemented in partnership with the International Food Policy Research Institute (IFPRI) and the University of Pretoria (UP). Together, the MSU-IFPRI-UP consortium works with governments, researchers and private sector stakeholders in Feed the Future focus countries in Africa and Asia to increase agricultural productivity, improve dietary diversity and build greater resilience to challenges like climate change that affect livelihoods. The papers are aimed at researchers, policy makers, donor agencies, educators, and international development practitioners. Selected papers will be translated into French, Portuguese, or other languages. Copies of all FSP Research Papers and Policy Briefs are freely downloadable in pdf format from the following Web site: www.foodsecuritypolicy.msu.edu Copies of all FSP papers and briefs are also submitted to the USAID Development Experience Clearing House (DEC) at: http://dec.usaid.gov/

ii

AUTHORS Ben Belton is Assistant Professor, International Development in the Department of Agricultural, Food, and Resource Economics, Michigan State University Mateusz Filipski is Research Fellow at the International Food Policy Research Institute (IFPRI) Chaoran Hu is a graduate student in the Department of Agricultural, Food, and Resource Economics, Michigan State University Author’s Acknowledgment: This study was made possible by the generous support of the American people through the United States Agency for International Development (USAID). The contents are the responsibility of Michigan State University and the International Food Policy Research Institute, and do not necessarily reflect the views of USAID or the United States Government. This study is also supported with financial assistance from the Livelihoods and Food Security Trust Fund, supported by Australia, Denmark, the European Union, France, Ireland, Italy, Luxembourg, the Netherlands, New Zealand, Sweden, Switzerland, the United Kingdom, the United States of America, and the Mitsubishi Corporation. We thank these donors for their kind contributions to improving the livelihoods and food security of rural people in Myanmar. The views expressed herein should in no way be taken to reflect the official opinion of any of the LIFT donors.

This study is made possible by the generous support of the American people through the United States Agency for International Development (USAID) under the Feed the Future initiative. The contents are the responsibility of the study authors and do not necessarily reflect the views of USAID or the United States Government. Copyright © 2017, Michigan State University and IFPRI. All rights reserved. This material may be reproduced for personal and not-for-profit use without permission from but with acknowledgment to MSU and IFPRI. Published by the Department of Agricultural, Food, and Resource Economics, Michigan State University, Justin S. Morrill Hall of Agriculture, 446 West Circle Dr., Room 202, East Lansing, Michigan 48824, USA

iii

EXECUTIVE SUMMARY Fish farming (aquaculture) has grown rapidly in Myanmar over the last two decades and plays an increasingly important role in national fish supply, but its technical and economic characteristics have been poorly studied. This report addresses this knowledge gap by presenting data from the first statistically representative survey of fish farms conducted in Myanmar - the Myanmar Aquaculture-Agriculture Survey (MAAS). MAAS was implemented in May 2016. A total of 242 fish farming households (151 growout farms and 73 nurseries) were interviewed in a ‘cluster’ of 25 village tracts as part of a larger survey that covered 1102 households in 40 village tracts in the main fish growing areas of Myanmar in Ayeyarwady and Yangon regions. As estimated from satellite images, the village tracts surveyed contained 57% of the total area of inland fish ponds in the Ayeyarwady Delta. Surveyed farms represent the entire population of fish farming households resident in these village tracts. Survey results provide a comprehensive ‘benchmark’ of the characteristics of inland aquaculture in Myanmar. Features analyzed include: farm productivity and profitability; farm size; production cycle duration; use of feed, seed and other production inputs; demand for labor; harvesting and marketing behaviors; technological change; the economic and social characteristics of fish farming households; and land access and tenure. The following results stand out: 1. High returns Aquaculture generates much higher earnings per hectare than crop farming. The average gross margin earned by fish farmers with growout farms is nearly $650/acre ($1600/ha). Surveyed crop farming households in ‘aquaculture cluster’ village tracts make an average annual gross margin of just $150/acre ($380/ha) across all field crops. Gross margins for individual crops in these village tracts range from $85/acre ($210/ha) for monsoon paddy, to $175/acre ($430/ha) for black gram. Fish farming households are twice as well-off as the general population of village tracts in the aquaculture cluster. Fish farming households’ estimated consumption expenditure (a proxy for income) averages $1525 per capita per year, compared to $718 per capita for all households in the cluster. Households with large growout farms are 4.2 times wealthier than the cluster population average, with average consumption expenditures of $2980 per capita. 2. Job creation Aquaculture creates more on-farm employment opportunities than agriculture. Considering all inputs of family labor, hired casual labor and hired long-term labor, fish farms require almost four times more labor per acre than crop farms (94 person days/year, versus 24 person days/year). This difference reflects the constant nature demand for labor on fish farms, the strongly seasonal nature of demand for labor on crop farms, and the high degree of agricultural mechanization that has taken place in the areas surveyed.

iv

Small fish farms create much more employment than large fish farms. Small growout farms generate demand for 152 person days of labor per acre/year. Medium sized and large growout farms generate demand for 41 days and 17 days, respectively. These differences reflect economies of scale in the employment of certain types of labor. Fish farms pay higher wages than crop farms, especially for women. The daily wage for work on fish farms averages $4.22/day. Workers on crop farms earn 27% less on average ($3.32/day). Further, the gender gap in wages is smaller in aquaculture than in agriculture: Women employed in crop farming receive about 2/3 of the average male daily wage for their work, whereas women employed in aquaculture earn 3/4 of the male daily wage. Fish farming remains a male-dominated activity. Only 20% of individuals who reported aquaculture to be their primary occupation were women. In addition, women represent only 13% of the casual workforce on fish farms. These figures are lower than the corresponding shares in crop farming, where women represent 27% of farmers and 33% of wage workers. 3. Numerous nurseries and small farms Nurseries are an important but overlooked farm segment. Specialized nurseries growing juvenile fish (“seed”) for sale to growout farms account for 41% of all aqua-farms. Almost all nurseries (97%) are less than 10 acres in size. Nursery operators own approximately only half as much land as crop farmers: an average of 3.7 acres of land, with a median area of just 1.7 acres. The average gross margin earned by nurseries $680/acre ($1681/ha) is comparable to the average return from fish growout farms, and considerably higher than that from small and medium growout farms. Small farms dominate by number, large farms dominate by area. Contrary to the conventional wisdom that Myanmar’s fish farms are all very large, half (51%) of all growout farms are less than 10 acres in size. These farms account for only 9% of total surveyed pond area, however. Conversely, large farms (those over 40 acres), account for 21% of farms but 70% of surveyed pond area. Operators of growout farms own more than three times as much land as crop farming households on average (30.5 acres, versus 8.0 acres). 4. Specialized commercial activity Fish production is highly commercial. Although 93% of households reported consuming some of the fish they harvested, either directly or as gratuities to workers, fish consumed in these ways represented less than 1% of fish produced among farms in all size categories. Fish farming is a specialized activity. Households who farm fish operate little agricultural land on average (0.5 acres and 1.4 acres for households with specialized nurseries and growout farms respectively). 5. Variable yields Yields are modest. The average yield across all farms was 1.9 t/acre (4.8 t/ha). This level is comparable to yields from small well managed commercial carp farms in Bangladesh, but

v

approximately half as much as is typical in Andhra Pradesh, India, where carp-farming technologies are well advanced. Productivity is highly variable. The worst performing 20% of farms had yields 11 times lower than the best performing 20% of farms (0.2 t/acre or 0.6 t/ha, versus 2.8 t/acre or 6.9 t/ha). Much of this variability can be accounted for by differences in investment capacity - the best performing quintile of farms spend almost nine times more on operating costs than the least productive quintile. Larger farms have higher yields, contrary to expectations. Average productivity ranges from 1.5 t/acre (3.8 t/ha) on small farms, to 2.5 t/acre (6.1 t/ha) on large farms. This reflects longer average production cycles, better access to credit, and higher levels of feed use among larger farms. Large farms spent almost twice as per acre of pond on floating and sinking pelleted fish feeds than small farms, and 2.3 times as much on peanut oilcake. Expenditure on rice bran, a cheaper but less efficient feed, was similar across farms of different sizes. 6. Simple technologies Use of fertilizers is extremely limited. Pond fertilization is a simple, low cost technique that can significantly improve farm productivity by stimulating blooms of plankton that provide natural food for fish. However, only 25% of farms surveyed used any kind of fertilizer, and fertilizers accounted for less than 1% of total operating costs. This finding suggests significant scope to increase fertilizer application. Use of pelleted fish feeds is low. Pelleted feeds are formulated to ensure a complete diet for farmed fish, facilitating more efficient feed conversion and faster fish growth than other commonly used feeds such as rice bran. Only 15% of farms use any manufactured pelleted feeds. This is considerably lower than in other Asian countries (e.g. Bangladesh 38%, China 90%). The main feeds are byproducts from agricultural processing; most importantly rice bran (used by 86% of farms) and peanut oilcake (44% of farms). On average, more productive farms spent a smaller share of their feed budgets on rice bran than less productive farms, and a higher share on pelleted feeds. 7. Crop diversity The vast majority of farms are polycultures, meaning that they stock multiple fish species in the same pond. Eighty six percent of farms stock more than one species, and the average number of species stocked per farm is 3.3. Carp species dominate production. The three most commonly stocked fish - rohu (94% of farms), catla (74%), mrigal (60%) - are all carps. The three next most commonly farmed fish are non-carp species; pangasius (28% of farms), tilapia (11%), and pacu (8%). Almost half of farms stock only carp species, and fewer than 5% stock no carp at all. The range of species farmed is limited, but increasing gradually. Farmers were asked about the year in which they first produced the species they presently stocked on their farm. In 1995, 93% of farmers stocked one of four major species (rohu, catla, mrigal, and pangasius). By 2015, this share had fallen to 77%, indicating that some species diversification had taken place.

vi

Small farms specialize more in the production of non-carp species. Small farms are most likely to farm non-carp species, and obtain higher average yields of tilapia and pacu than medium or large farms growing these species. In contrast, yields of rohu and pangasius from large farms are almost double those of small farms. Nine percent of small farms stock freshwater prawn, a high value crop that is not produced by any medium or large farms. 8. Production costs Feed is by far the largest operating cost, accounting for 70% of variable costs. Feed costs average $1400/acre ($3300/ha), or a total of $16,800 for an average-sized farm. This is followed by seed (9% of operating costs), and non-feed inputs (7%), consisting primarily of fuel. Labor accounts for only 4% of operating costs. Price pressures may be eroding farm profitability. From 2006 to 2016, the nominal price of the most widely used fish feed, rice bran, rose by 38 percentage points more than the nominal farmgate price of rohu. Given the high share of feed in operating costs, this suggests that the average profitability of fish production probably fell over this period. 9. Finance Large farms have better access to credit. Twenty six percent of farms reported receiving credit from the main buyer of their fish. It is more common for large farms to receive credit (55% of farms) than small or medium (both 17%). Interest accounted for only 2% of operating costs, averaged across all farms, despite most loans being taken from informal lenders at high rates. This figure reflects relatively limited levels of credit utilization. Pelleted feeds are available on credit, other feeds are not. Suppliers of agricultural byproducts such as rice bran rarely provide these to farmers on credit. However, 28% of users of sinking pellets and 53% of users of floating pellets received part or all of these inputs on credit. Three quarters (75%) of farmers who obtained pelleted feeds on credit were required to sell harvested fish back to the credit provider. 10. Non-farm linkages Aquaculture creates demand for goods and services from off-farm enterprises. For example, 78% of farms utilize rented boats or motor vehicles to deliver harvested fish to market, and 93% of farms purchase ice for transporting fish. Two thirds of growout farms (66%) purchase fingerlings, primarily from nurseries. Small farms are more likely than medium or large farms to purchase seed from nurseries. 11. Market concentration The market for pelleted fish feeds is highly concentrated. Sixty percent of all pelleted feed used by surveyed farmers, including 65% of all sinking feed and 51% of floating feed, was sourced from a single company. Marketing channels are simple. Among farms of all sizes, the main buyer of harvested fish was overwhelmingly a fish trader (96%). Most farms (92%) sold their entire harvest to directly a single

vii

buyer. Three quarters of buyers (76%) were traders located at the main Yangon fish wholesale market, San Pya. Almost all farms (96%) paid a commission on sales, averaging 4.9%. Despite the high value of most sales, 98% of transactions were settled in cash, with just 2% of payment taking place via bank transfer. 12. Landholdings and tenure security Ownership of fish farms is becoming more concentrated. The mean size of growout farms rose from 14 to 22 acres from 2006-2016. As a result, the weighted Gini coefficient for land used for growout farming increased from 0.55 in 2001, to 0.73 in 2016. Small fish farms have less secure land tenure than large farms. Only 16% of ponds on small farms were reported to have La Na 39 (the land use title document that permits conversion of agricultural land to non-agricultural uses), compared to 60% of ponds on large farms. Implications for policy and programing These findings have the following implications for the design of policies and interventions aimed at promoting the growth of aquaculture. 1. Fish farming should be recognized and promoted as a mechanism for generating rural growth. Average returns from aquaculture are four times higher than those earned from crop farming. In addition, aquaculture creates numerous economic linkages within the rural economy where farms are located. These include demand for labor, intermediate inputs (e.g. fish seed from nurseries) and services (e.g. transport). However, yields and profitability are highly variable, and small farms in particular perform sub-optimally. Simple management improvements could enhance their performance. 2. Small farms (sized 10 acres or less) and nurseries should be the principal target of policy and technical interventions. Considered together, these account for 70% of all fish farms. Small farms create greater relative demand for labor and many goods and services than large farms. However, they remain disadvantaged in terms of tenure security, access to credit, and their capacity to invest in adequate levels of production inputs. Overcoming these constraints would help boost small farm productivity and profitability and bring them closer to that of larger farms. 3. Smaller farms have a competitive advantage in the production of non-carp species – especially tilapia, pacu and freshwater prawn, as indicated by small farms attaining higher yields of these species than medium or large. Investments in these species should be prioritized; especially by supporting the establishment of privately operated mono-sex tilapia hatcheries, and research to overcome bottlenecks in the hatchery production of freshwater prawn. 4. Conduct research and outreach on use of fertilizers. Pond fertilization is a simple low cost technique that can significantly improve production efficiency, and is widely adopted by fish farmers elsewhere in Asia. Research is needed to understand farmers’ attitudes toward and use of fertilizers, identify management protocols for optimal fertilizer use under Myanmar conditions through field trials, and disseminate results and recommendations to users.

viii

5. Identify mechanisms for providing commercial loans, tailored to the needs of actors in aquaculture value chains. Operating costs for these enterprises are high and access to even informal forms of credit is currently limited. 6. Encourage private investment in the feed sector to increase competitiveness, lower feed prices, and improve ease and terms of access to pelleted fish feeds.

ix

TABLE OF CONTENTS EXECUTIVE SUMMARY

iv

2.

1

3.

4.

5.

INTRODUCTION 2.1.

Study background

1

2.2.

Farm typology

2

2.3.

Nomenclature

3

2.4.

Units

3

FARM PRODUCTIVITY AND PROFITABILITY

4

3.1.

Growout farm productivity

4

3.2.

Growout farm budgets and gross margins

6

3.3.

Specialized nurseries

7

FARM CHARACTERISTICS AND MANAGEMENT

9

4.1.

Farm size

9

4.2.

Production cycle

10

4.3.

Seed

10

4.4.

Feed

14

4.5.

Non-feed inputs

17

4.6.

Labor

19

4.7.

Machinery and equipment

21

HARVESTING AND MARKETING

23

5.1.

Harvesting

23

5.2.

Marketing

25

6.

HISTORICAL TRENDS IN PRODUCTION PRACTICES

27

7.

FARM HOUSEHOLDS AND LANDHOLDINGS

29

8.

7.1.

Household characteristics

29

7.2.

Land acquisition, tenure, use and distribution

31

CONCLUSION

34

ANNEX 1: SURVEY METHODOLOGY

35

REFERENCES

37

x

1. INTRODUCTION 1.1.

Study background

Aquaculture (fish farming) has grown rapidly in Myanmar over the last two decades and plays an increasingly important role in national fish supply (Belton et al., 2015), but its technical and economic characteristics have been inadequately studied. This report addresses this knowledge gap, presenting benchmark data from the first statistically representative survey of fish farms conducted in Myanmar - the Myanmar Aquaculture-Agriculture Survey (MAAS). MAAS was implemented in May 2016. Data were collected from a total of 1102 rural households, including crop farmers, fish farmers, and the landless, located in 40 village tracts 1 in four townships (Twantay, Maubin, Nyaungdon, Kayan) in Ayeyarwady and Yangon regions. All the village tracts surveyed lie in a zone within a radius of 60 km from Myanmar’s largest city and main commercial center, Yangon. The households surveyed represent a total population of about 37,000 households. Details of the sampled households and their landholdings are provided in Table 1. Table 1: Summary statistics on households surveyed in MAAS In sample

In the represented Weighted farm size economy (weighted) (acres) N % N % Mean Median All households 1,102 100 37,390 100 Fish farming households* 224 20 2,450 7 14.1 3 Crop farming households 329 30 9,604 26 6.8 4 Non-farm households 549 50 25,336 68 0 0 *Note: Fish farming households here includes both growout farms and specialized nurseries

A subset of 242 fish farming households (151 growout farms and 73 nurseries) were interviewed in 25 village tracts (the ‘aquaculture cluster’ village tracts identified in Figure 1), representing a total of 2450 fish farming households. Analysis of satellite images conducted as part of the sampling process indicates that 57% of the total area of inland fish ponds in Myanmar lies within the village tracts surveyed. Farms surveyed were selected to represent the entire population of fish farming households resident in the 25 village tracts. 2 Given that 90% of Myanmar’s inland fish ponds are located in Ayeyarwady and Yangon regions (DOF, 2014), the sample can be considered to represent approximately half the area under freshwater aquaculture in Myanmar. A detailed summary of the survey methodology is provided in Annex 1. This report provides a comprehensive ‘benchmark’ of data on the technical, economic and social characteristics of inland aquaculture in Myanmar in 2016. It covers the following topics: farm productivity and profitability; farm size; the duration of the production cycle; use of seed, feed and other production inputs; demand for labor; harvesting and marketing practices; technological change; socioeconomic characteristics of fish farming households; land access and tenure. A village tract is the smallest administrative sub-unit in Myanmar, usually composed of 5-10 villages Fish farms operated by non-residents of these village tracts (i.e. absentee owners and companies) were not captured in the survey. The data presented here thus tend to underrepresent very large farms, sized 100 acres or more. 1 2

1

Figure 1. Location of Surveyed Village Tracts

1.2.

Farm typology

Two types of fish farms were surveyed: 1) specialized nurseries, comprising 41% of all aqua-farms, growing juvenile fish (“fingerlings”) for sale to growout farms (41% of aqua-farms); and 2) “growout” farms producing food fish for the market (59% of aqua-farms). All economic and technical analysis presented in this report pertains to growout farms, except where explicitly stated otherwise. In the analysis that follows, we divide growout farms into three size categories as follows: ‘small’ (