Navigating Real Markets: The Economic Resilience of Food Wholesale Traders in Dhaka, Bangladesh 3515113797, 9783515113793

Citation preview

Markus Keck

Navigating Real Markets The Economic Resilience of Food Wholesale Traders in Dhaka, Bangladesh

Geographie

Megacities and Global Change Megastädte und globaler Wandel

Franz Steiner Verlag

Band 19

Markus Keck Navigating Real Markets

megacities and global change megastädte und globaler wandel herausgegeben von Frauke Kraas, Martin Coy, Peter Herrle und Volker Kreibich Band 19

Markus Keck

Navigating Real Markets The Economic Resilience of Food Wholesale Traders in Dhaka, Bangladesh

Franz Steiner Verlag

Gedruckt mit freundlicher Unterstützung der Deutschen Forschungsgemeinschaft

Umschlagabbildung: Wholesale traders’ rice warehouse under the 2nd Buriganga Bridge, 07.12.2007 © Markus Keck

Bibliografische Information der Deutschen Nationalbibliothek: Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliografie; detaillierte bibliografische Daten sind im Internet über abrufbar. Dieses Werk einschließlich aller seiner Teile ist urheberrechtlich geschützt. Jede Verwertung außerhalb der engen Grenzen des Urheberrechtsgesetzes ist unzulässig und strafbar. © Franz Steiner Verlag, Stuttgart 2016 Angenommen als Dissertation durch die Mathematisch-Naturwissenschaftliche Fakultät der Rheinischen Friedrich-Wilhelms-Universität Bonn unter dem Titel „Market Governance and Social Resilience. The Organization of Food Wholesaling in Dhaka, Bangladesh“. Druck: Hubert & Co, Göttingen Gedruckt auf säurefreiem, alterungsbeständigem Papier. Printed in Germany. ISBN 978-3-515-11379-3 (Print) ISBN 978-3-515-11380-9 (E-Book)

In memory of Korshed Alam (*25.01.1966, †17.11.2012)

TABLE OF CONTENT

LIST OF FIGURES ....................................................................................... x LIST OF TABLES ....................................................................................... xi LIST OF PHOTOGRAPHS ........................................................................xii LIST OF TEXT BOXES ............................................................................xiii PREFACE .............................................................................................................. xv 1.

INTRODUCTION ............................................................................................ 1 1.1 AIM OF THE STUDY ............................................................................ 1 1.2 OUTLINE OF THE STUDY .................................................................. 3

2.

NAVIGATING REAL MARKETS ................................................................. 7 2.1 GEOGRAPHY OF FOOD SYSTEMS ................................................... 8 2.1.1 Food Production ............................................................................... 8 2.1.2 Food Distribution ............................................................................. 9 2.1.3 Food Systems ................................................................................. 10 2.1.4 Resilience in Food Systems............................................................ 12 2.2 SOCIAL RESILIENCE ........................................................................ 13 2.2.1 What is Resilience? ........................................................................ 13 2.2.2 What is Social Resilience? ............................................................. 15 2.2.3 Three Capacities of Social Resilience ............................................ 18 2.2.4 Anchoring Resilience in Social Theory ......................................... 19 2.3 EMBEDDED AGENCY ....................................................................... 20 2.3.1 From Structural to Institutional Embeddedness ............................. 20 2.3.2 Institutions ...................................................................................... 22 2.3.3 Actors ............................................................................................. 24 2.3.4 Knowledge and Practice ................................................................. 25 2.4 REAL MARKETS ................................................................................ 28 2.4.1 What are Markets? ......................................................................... 28 2.4.2 Markets and Practice ...................................................................... 30 2.4.3 Markets and Networks ................................................................... 32

viii

Table Of Content

2.4.4 Markets and Arenas ........................................................................ 35 2.5 RÉSUMÉ............................................................................................... 37 3.

METHODOLOGY ......................................................................................... 40 3.1 TRIANGULATION .............................................................................. 40 3.2 QUANTITATIVE METHODS ............................................................. 43 3.2.1 Locating Food Markets .................................................................. 43 3.2.2 Recording Sales Volumes .............................................................. 44 3.2.3 Surveying Wholesale Traders ........................................................ 44 3.3 QUALITATIVE METHODS ................................................................ 46 3.3.1 Observing the Food System ........................................................... 46 3.3.2 Interviewing Wholesale Traders .................................................... 47 3.3.3 Using Participatory Tools............................................................... 49 3.4 LIMITATIONS OF THE STUDY ........................................................ 51

4.

THE FOOD SYSTEM OF DHAKA .............................................................. 52 4.1 THE MEGACITY OF DHAKA ........................................................... 52 4.1.1 History of Dhaka ............................................................................ 52 4.1.2 From City to Megacity ................................................................... 56 4.2 RICE FOR THE MEGACITY .............................................................. 61 4.2.1 Rice Production in Bangladesh ...................................................... 61 4.2.2 Availability of Rice in Dhaka......................................................... 66 4.2.3 Accessibility of Rice in Dhaka ....................................................... 71 4.3 FISH FOR THE MEGACITY .............................................................. 76 4.3.1 Fish Production in Bangladesh....................................................... 76 4.3.2 Availability of Fish in Dhaka ......................................................... 81 4.3.3 Accessibility of Fish in Dhaka ....................................................... 85 4.4 RÉSUMÉ............................................................................................... 90

5.

FOOD WHOLESALE MARKETS IN DHAKA ........................................... 93 5.1 DHAKA’S FOOD WHOLESALE MARKETS ................................... 93 5.1.1 The Emergence of Dhaka’s Food Wholesale Markets ................... 93 5.1.2 The Characteristics of Dhaka’s Food Wholesale Markets ............. 97 5.2 DHAKA’S VALUE CHAINS FOR FOOD ........................................ 102

Table Of Content

ix

5.2.1 The Value Chain for Rice............................................................. 102 5.2.2 The Value Chain for Fish ............................................................. 111 5.3 RICE AND FISH WHOLESALE TRADERS IN DHAKA ............... 119 5.3.1 The Traders .................................................................................. 119 5.3.2 The Stores ..................................................................................... 124 5.4 RÉSUMÉ............................................................................................. 130 6.

THE RESILIENCE OF FOOD WHOLESALE TRADERS IN DHAKA .. 134 6.1 THE EMBEDDEDNESS OF FOOD WHOLESALE TRADERS ..... 135 6.2 MARKETS AS PRACTICE: FOOD WHOLESALE TRADERS’ COPING CAPACITIES ...................................................................... 143 6.2.1 Delivery of Goods ........................................................................ 143 6.2.2 Sale and Price Building ................................................................ 146 6.2.3 Accountancy and Restocking ....................................................... 152 6.3 MARKETS AS NETWORKS: FOOD WHOLESALE TRADERS’ ADAPTIVE CAPACITIES ................................................................. 157 6.3.1 Exposure and Sensitivity to Business Risks................................. 157 6.3.2 Trust in Business Relations .......................................................... 163 6.3.3 Diversification of Networks ......................................................... 170 6.4 MARKETS AS ARENAS: FOOD WHOLESALE TRADERS’ TRANSFORMATIVE CAPACITIES ................................................ 179 6.4.1 Food Wholesale Traders in the Urban Arena ............................... 179 6.4.2 Urbanization and the Problem of Place ........................................ 183 6.4.3 Dealing with Informality .............................................................. 187 6.5 RÉSUMÉ............................................................................................. 199

7.

CONCLUSION ............................................................................................ 204

BIBLIOGRAPHY ................................................................................................ 211 APPENDIX .......................................................................................................... 227

x

Table Of Content

LIST OF FIGURES Figure 1: Research framework ................................................................................. 6 Figure 2: Actors-institutions loop .......................................................................... 27 Figure 3: Geography of markets – Conceptual framework ................................... 39 Figure 4: Population growth of Dhaka (1901–2010) ............................................. 58 Figure 5: City size ranking of municipalities and SMAs in Bangladesh ............... 60 Figure 6: Food grain production, import and requirement in Bangladesh a (1971–2011) ............................................................................................ 62 Figure 7: Cropped area and average annual yield of rice in Bangladesh g a a a (1971–2006) ............................................................................................ 63 Figure 8: Food consumption in urban areas and in Dhaka City LückeLü(from left to right).................................................................................... 67 Figure 9: Rice consumption in Dhaka City ........................................................... 70 Figure 10: Total catch and net availability of fish in Bangladesh (1989–2009).... 78 Figure 11: Average annual catch of fish in Bangladesh (1989–2009) .................. 79 Figure 12: National fish production and urban consumption in Dhaka City (from left to right) ................................................................................. 82 Figure 13: Fish consumption in Dhaka City .......................................................... 84 Figure 14: Proportion of total rice turnover in Dhaka City by market .................. 99 Figure 15: Distribution of rice wholesalers by market ........................................ 100 Figure 16: Proportion of total fish turnover in Dhaka City by market ................ 100 Figure 17: Distribution of fish wholesalers by market ........................................ 100 Figure 18: The value chain for coarse rice to Dhaka ........................................... 110 Figure 19: The value chain for carps to Dhaka .................................................... 118 Figure 20: Distribution of rice and fish wholesale traders by age ....................... 119 Figure 21: Distribution of rice and fish wholesale traders by educational level . 120 Figure 22: Wholesale traders’ residency in Dhaka .............................................. 122 Figure 23: Distribution of rice wholesale traders by region of origin ................. 123 Figure 24: Distribution of fish wholesale traders by region of origin ................. 123 Figure 25: The personal network of Bashir ......................................................... 137 Figure 26: The personal network of Gilgad ......................................................... 140 Figure 27: Rice wholesale traders’ exposure to business risks ............................ 159 Figure 28: Fish wholesale traders’ exposure to business risks ............................ 160 Figure 29: Rice and fish wholesale traders’ susceptibility to business risks ....... 161 Figure 30: Wholesale traders’ number of suppliers per year ............................... 171 Figure 31: Number of rice and fish types dealt by wholesale traders .................. 174 Figure 32: Wholesale traders’ number of customers ........................................... 176 Figure 33: The structure of wholesale traders’ business networks ...................... 178

Table Of Content

xi

LIST OF TABLES Table 1: Three capacities of social resilience ........................................................ 18 Table 2: Rice market survey 2009–2010 ............................................................... 44 Table 3: Fish market survey 2009–2010................................................................ 45 Table 4: List of qualitative interviews ................................................................... 48 Table 5: Codes for the analysis of qualitative data ................................................ 48 Table 6: List of participatory tools ........................................................................ 50 Table 7: Gross food availability in Dhaka City ..................................................... 66 Table 8: Average annual rice wholesale prices in Dhaka (2009–2010) ................ 69 Table 9: Seasonal changes in supply at Dhaka’s rice markets .............................. 72 Table 10: Seasonal price changes of major rice types at Dhaka’s LückeLü wholesale markets .................................................................................. 73 Table 11: Daily price changes of major rice types at Dhaka’s wholesale markets 74 Table 12: Recent price trends of major rice types at Dhaka’s wholesale markets. 75 Table 13: Average annual fish wholesale prices in Dhaka (2009–2010) .............. 83 Table 14: Seasonal changes in supply at Dhaka’s fish markets ............................ 86 Table 15: Seasonal price changes of major fish types at Dhaka’s wholesale markets .................................................................................. 87 Table 16: Daily price changes of major fish types at Dhaka’s wholesale markets 88 Table 17: Recent price trends of major fish types at Dhaka’s wholesale markets. 90 Table 18: Characteristics of food wholesale markets in Dhaka City ..................... 97 Table 19: Distribution of arising costs, net margin and consumers’ expenditure for domestically produced coarse rice (local variety) by value chain actors .......................................................................................... 109 Table 20: Distribution of arising costs, net margin and consumers’ expenditure for domestically raised Rui Carp by value chain actors....................... 117 Table 21: Workforce of rice and fish wholesale stores ....................................... 125 Table 22: Shop size of rice and fish wholesale stores ......................................... 126 Table 23: Daily turnover of rice and fish wholesale stores ................................. 127 Table 24: Capital invested by rice and fish wholesale stores .............................. 129 Table 25: Annual sales of rice and fish wholesale stores .................................... 129 Table 26: Assessment of rice wholesale traders’ business relations ................... 166 Table 27: Assessment of fish wholesale traders’ business relations ................... 167 Table 28: Total annual turnover of wholesale traders via regular suppliers ........ 168 Table 29: Total annual turnover of wholesale traders via regular customers ...... 169 Table 30: Supplier diversity of rice and fish wholesale traders ........................... 172 Table 31: Spatial diversity of supplies of rice and fish wholesale traders ........... 173 Table 32: Product diversity of rice and fish wholesale traders ............................ 175 Table 33: Customer diversity of rice and fish wholesale traders ......................... 177

xii

Table Of Content

LIST OF PHOTOGRAPHS Photo 1: Paddy nurturing fields, Kushtia Municipality, 17.12.2007 ................... 102 Photo 2: Irrigated paddy fields, Jessore, Boro Bazar, 12.02.2009 ....................... 102 Photo 3: Seasonal labor, paddy harvest, Kusthia Municipality, 17.12.2007 ....... 103 Photo 4: Paddy threshing cycle, Kushtia Municipality, 17.12.2007.................... 103 Photo 5: Paddy steamer, Kushtia Municipality, 17.12.2007 ............................... 104 Photo 6: Chatal, Kushtia Municipality, 17.12.2007 ............................................ 104 Photo 7: Rice milling machine, Kushtia Municipality, 17.12.2007 .................... 105 Photo 8: Tush and kura, Kushtia Municipality, 17.12.2007 ................................ 105 Photo 9: Rice sorting machine, Kushtia Municipality, 17.12.2007 ..................... 106 Photo 10: Mora chal (in the middle), Kushtia Municipality, 17.12.2007 ........... 106 Photo 11: Rice delivery by rickshaw, Dhaka City, Babubazar, 26.02.2008 ........ 107 Photo 12: Rice retail store, Dhaka, Tongi, 23.09.2007 ........................................ 107 Photo 13: Land port “Panama”, Dinajpur, Hili, 19.12.2007 ................................ 108 Photo 14: Transshipment of import rice, Sirajganj Municipality, 20.12.2007 .... 108 Photo 15: Carp breeding tank, Gov. Fish Seed Multiplication Farm, Khulna Municipality, 08.02.2009 .................................................................... 111 Photo 16: Incubator with carp eggs, Jagaroni Chakra Foundation, Jessore Municipality, 11.02.2009 .................................................................... 111 Photo 17: Carp post-larva, Jagaroni Chakra Foundation, Jessore Municipality, 11.02.2009 .................................................................... 112 Photo 18: Cultivating fry in private hatchery, Jessore Municipality, 12.02.2009 112 Photo 19: Netting of fingerlings, Jagaroni Chakra Foundation, Jessore Municipality, 11.02.2009 ..................................................................... 113 Photo 20: Harvesting of carps, Jagaroni Chakra Foundation, Jessore Municipality, 11.02.2009 ..................................................................... 113 Photo 21: Gher, Khulna, Botiaghata Upazilla, 08.02.2009 ................................. 114 Photo 22: Professional fish farm, Jessore Municipality, 12.02.2009 .................. 114 Photo 23: Transport of fish in drums, Savar, Banktown, 18.01.2009 ................. 114 Photo 24: Transport of fish in baskets, Jessore, Boro Bazar, 12.02.2009 ........... 114 Photo 25: Hawkers prepare for business, Dhaka City, Kawran Bazar, 31.03.2008 .......................................................................................... 116 Photo 26: Retailer on kitchen market, Dhaka City, Khilgaon, 24.09.2007 ......... 116 Photo 27: Rice wholesale store, Dhaka City, Badamtuli, 13.11.2007 ................. 126 Photo 28: Rice wholesale store, Dhaka City, Kochukhet, 06.02.2008 ................ 126 Photo 29: Fish vending sites, Dhaka City, Kawran Bazar, 25.01.2008 ............... 127 Photo 30: Fish vending sites, Dhaka City, Merul Badda, 08.03.2009 ................. 127 Photo 31: Laborer rest for night turn, Dhaka City, Badamtuli, 25.01.2008 ........ 144 Photo 32: Laborer replenish rice stocks, Dhaka City, Badamtuli, 25.01.2008 .... 144 Photo 33: Laborer unload frozen fish, Dhaka City, Kawran Bazar, 29.01.2009 . 145 Photo 34: Sleeping accommodation, Dhaka City, Kawran Bazar, 29.01.2009 ... 145 Photo 35: Displayed rice samples, Dhaka City, Badamtuli, 13.11.2007 ............. 147 Photo 36: Godi, Dhaka City, Badamtuli, 07.04.2009 .......................................... 147 Photo 37: Bonga, Dhaka City, Kawran Bazar, 04.02.2008 ................................. 147

Table Of Content

xiii

Photo 38: Hook, Dhaka City, Badamtuli, 13.11.2007 ......................................... 147 Photo 39: Laborer lifting bag of 85 kg, Dhaka City, Badamtuli, 25.01.2008 ..... 148 Photo 40: Delivery of rice bags, Dhaka City, Badamtuli, 26.02.2008 ................ 148 Photo 41: Displayed frozen fish, Dhaka City, Kawran Bazar, 31.03.2008 ......... 150 Photo 42: Displayed live fish, Dhaka City, Kawran Bazar, 31.03.2008 ............. 150 Photo 43: Weigher at work, Dhaka City, Kawran Bazar, 31.03.2008 ................. 151 Photo 44: Accountant at work, Savar, Bank Town, 18.01.2009.......................... 151 Photo 45: Helper amidst live fish, Dhaka City, Kawran Bazar, 31.03.2008 ....... 152 Photo 46: Auctioning of small fish, Dhaka City, Kawran Bazar, 31.03.2008 ..... 152 Photo 47: Manager at work, Dhaka City, Jatrabari, 06.02.2008.......................... 153 Photo 48: Account books, Dhaka City, Krishi Market, 10.03.2008 .................... 153 Photo 49: Krishi Market before demolition, 25.12.2006 ..................................... 189 Photo 50: Krishi Market after demolition, 27.02.2010 ........................................ 189 Photo 51: Structure of store before demolition, Dhaka City, Krishi Market, 10.03.2008 .............................................................................. 189 Photo 52: Store after demolition, Dhaka City, Krishi Market, 10.03.2008 ......... 189 Photo 53: Warehouse under bridge, Dhaka City, Badamtuli, 07.12.2007 ........... 191 Photo 54: Warehouse under bridge, Dhaka City, Badamtuli, 07.12.2007 ........... 191 Photo 55: Workers’ shelter at railway, Dhaka City, Kawran Bazar, 01.04.2008 193 Photo 56: Advertisement of hotel, Dhaka City, Kawran Bazar, 01.04.2008 ....... 193 Photo 57: Merul Badda, Wholesale Market, 17.01.2003..................................... 196 Photo 58: Merul Badda, Wholesale Market, 22.06.2006..................................... 196 Photo 59: Merul Badda, Wholesale Market, 12.07.2006..................................... 196 Photo 60: Merul Badda, Wholesale Market, 23.03.2010..................................... 196

LIST OF TEXT BOXES Box 1: A paddy farmer in Jessore ........................................................................ 103 Box 2: A rice miller in Kushtia ............................................................................ 105 Box 3: A gher owner in Jessore ........................................................................... 113 Box 4: Freight forwarding agents and truck drivers in Dhaka ............................ 115 Box 5: A rice wholesale trader in Dhaka and his business relations ................... 136 Box 6: A fish wholesale trader in Dhaka and his business relations ................... 139

PREFACE This study was not prepared in isolation. I was embedded in a network of people all along. I am deeply grateful for the active support that was shown to me by numerous people and for the chance to conduct this study in the first place. I would like to express my cordial gratitude to the following people: to Hans-Georg Bohle for being a thoughtful mentor and an inspiring academic in the field of critical human geography; to Wolfgang-Peter Zingel for making me understand that economics is basically about people – not about numbers and figures alone; to Benjamin Etzold for sharing with me a never fading interest in asking what geography is and should be about; to Shafique uz Zaman for showing me that strong arguments do not necessarily need loud voices to be heard; to Shahidul Haque for his constant will and readiness to give his level best; to Korshed Alam for proving me the possibility to be a critical thinker without losing the lust for life; to my dear fellows of “shatash number road” for company, food for thought and for conjointly creating a feeling for the place in Dhaka; to Snigdha Brian for teaching me “amar shonar Bangla”; to Patrick Sakdapolrak, Sebastian Jülich, Anna Zimmer, Thomas Schmitt, Jonas Hermsen, Johanna Kramm, Michael Eichholz and Sebastian Homm for being partners on the sometimes rough road toward becoming a researcher and for ongoiningly asking: “Why are we doing this?”; to all my interview partners for letting me partake in their lifes and businesses; and to Sheli for making me feel home in Dhaka. Furthermore, I am sincerely thankful to Frauke Kraas for courageously coordinating the priority program “Megacities – Megachallenge: Informal Dynamics of Global Change” without compromising scholars’ potentials to develop their own perspectives; to Harald Sterly for his strong commitment in bringing people together to share thoughts and learn from each other; to Gerd Storbeck, Martin Gref, Gabriele Bräuer-Jux, and Stefan Zöldi for making meaningful maps out of my ideas and sketches; to Irene Hillmer, Kristina Kümper-Schlake, Sonja Raupp, Miron Schmude and Gerdis Wischnath for giving me a helping hand to tackle all the small things that need to be done for a study of this size; and to the German Research Foundation (DFG) for funding this study within the project “The mega-urban food system of Dhaka/Bangladesh” (Grant No. BO 680/35–1/2). Last but not least, I would like to thank Daniela Geinitz and Felix Geinitz for never stopping to place their confidence on me. Markus Keck

1. INTRODUCTION

“Its trees were burned decades ago, its hills leveled and the fresh ponds drained and filled […]. Reaching out urbanizing tentacles from its island home, the city has become a megalopolis [...]: a slab of primordial granite and metamorphic rock bounded on all sides by water, squatting like a steel and stone spider in the midst of its web of bridges, tunnels, tubes, cables and ferries. [...T]here seems to be no limit to the people crowding here. They press in from the outside and raise their families, and their children and their children’s children raise families, until this city is populated as no other city has ever been in the history of the world”. Harry Harrison (1966: 11)

1.1 AIM OF THE STUDY Hardly any other city in the world grew as fast during the last decades as the capital of Bangladesh. Due to the political violence in the aftermath of the declared independence in 1971, Dhaka’s population growth reached a peak with annual rates of 8.9 percent. This meant nearly a doubling of the city’s population in not more than eight years. Although the urban growth rate decreased gradually to 3.5 percent (World Bank 2007), Dhaka’s pace of urbanization has remained high until today due to ongoing rural-urban migration (Siddiqui et al. 2000). Every year, the city becomes the new residence of approximately 450,000 people. In 1951, Dhaka had a population of 336,000 (BBS 1951 after Islam 2005). Today there are roughly 14.2 million1 (Islam 2010): 42 times more inhabitants than before. When I visited Dhaka for the first time, I was simply shocked by the mere number of people who moved back and forth through the labyrinth-like road networks. I remember very well one of my first CNG2 rides through the city. It was August and while the monsoon rainfalls got less and less, heat and dust reconquered mastery over the city. Out of my vehicle I saw an old looking woman who sat under a sunshade and mashed bricks in order to sell the obtained “soil” as filling material. Behind her two garbage collectors searched through mountains of foul-smelling waste for items still usable. Not two meters away, a white-collar professional sat in his air-conditioned four-wheel drive truck behind shadowed 1

2

While the first figure refers to the area of the Dhaka City Corporation (DCC) with an extension of 145 square kilometers, the latter number refers to Dhaka Metropolitan Development Planning Area (DMDP) with a size of 1,528 square kilometers. The autorickshaws in Dhaka are run by compressed natural gas, from which the name “CNG” stems from. The vehicles were converted at the behest of the government, mostly on private workshops, from the mid-1990s onwards in order to reduce air pollution and greenhouse gas emissions in Bangladesh’s cities (Rahman 2006a).

2

Introduction

window glass focussing on the screen of his laptop computer. In roughly one hundred meters height behind him I realized the vitreous façade of a skyscraper under construction. … I felt like someone had transferred me into the unsettling dystopia “Make Room! Make Room!”, in which Harry Harrison (1966) described the hardships of human life in a world driven by overpopulation and depleted resources, in which only a small elite can afford vital means while the masses of people suffer from the harsh consequences of environmental pollution, ongoing supply crises and extreme poverty.3 In subsequent years, I realized that I was not alone with this feeling. In the literature, megacities in the Global South are regarded as risk areas (Kraas 2008). Due to the concentration of people, their unprecedented growth rates, their complexities, and their dynamism, these agglomerations are seen as one of the major challenges of contemporary urbanization (Kraas 2010).4 From a development perspective, the concern is about the expected loss of urban governability, the uncontrolled spread of informality, crime and violence, and the inexorable increase of deprivation and vulnerability (Bronger 2004; Kraas & Mertins 2008; Satterthwaite 2005). Of central concern in this regard is the supply of megacities with resources essential for survival, such as water, energy or food. Megacities are seen to be particularly prone to supply crises (Kraas 2003). Against this background, I assume, one major proposition drives the entire international research initiative: Megacities are taken to be synonymous with human misery and hardship, whereas a seemingly “natural” link is established between the concentration of people and an expected urban crisis. I gathered my first impressions of the megacity of Dhaka in 2007, a time when a rapid rise in food prices pushed hundreds of millions of people around the globe into hunger and poverty – especially in cities (Cohen & Garrett 2010; FAO 2011; Ruel et al. 2010). In Bangladesh, this global food crisis was exacerbated due to the fact that it was accompanied by an ecological and a political crisis. The ecological crisis resulted from the combined effects of the 2007 monsoon floods and Cyclone Sidr (November 2007) and brought devastation, destroying much of the paddy crop in Bangladesh’s North and South (Webster et al. 2010). Consequently, the country had to import large quantities of food grain just when world market prices started to rise to unprecedented heights. In consequence, the “rice nation” of Bangladesh was particularly sensitive to the global food price hike of 2007/2008 (FAO 2011; World Bank 2010). This economic crisis most severely 3 4

Cf. the quote at the beginning of this chapter. This fact is reflected in a number of interdisciplinary research programs that work on the topic since 2005, among them “Risk Habitat Megacity” (German Helmholtz Association), “Emerging Megacities: Research for the Sustainable Development of the Megacities of Tomorrow” (German Ministry of Education and Research), “Megacities-Megachallenge: Informal Dynamics of Global Change” (German Research Foundation), “Urbanization and Global Environmental Change” (International Human Dimensions Program), “Urban Resilience” (Resilience Alliance), and most recently “North-South-Network on Urban SelfOrganization. Public Life in Europe, India and China” (European Union).

Introduction

3

affected the country’s urban poor, because they primarily rely on their cash income for accessing food, and thus face difficulties if prices rise rapidly while their income does not (Zingel et al. 2011). In addition, 2007 and 2008 were marked by a political crisis in Bangladesh, as the 2006 elections were suspended and a caretaker government took control. This military-backed interim government was supposed to remain in power for three months, until free and fair elections could take place. In fact, it ended up maintaining control of the country for two years in an effort to re-establish law and order and eradicate corruption throughout the political-economic system. Yet most visible among the interim government’s actions were large-scale slum clearance drives and evictions of informal markets, which again affected the livelihoods of the urban poor severely (IGS 2008; Siddiqui et al. 2010). This multidimensional crisis had devastating effects in the country’s capital, the megacity of Dhaka. Nevertheless, a large-scale catastrophe was prevented. Although Dhaka’s supplies were repeatedly disconnected as a result of production shortfalls and road blocks due to flooding, there was always enough food in the city (Keck et al. 2012). Even though the interim government evicted thousands of hawkers, street food was still available and the sale of prepared food was not substantially disturbed (Etzold 2013). And finally, even though the prices of staples rose by more than 100.0 percent within two years (MoA 2009), most of the urban poor found ways to endure the crisis (Zingel et al. 2011). Against this background, I want to raise the following questions: Who are the people who constitute Dhaka’s food system? How did they make it robust enough to avoid a catastrophe of even larger dimensions? And what hardships did they bear in building and maintaining this resilience? By answering these questions on the basis of empirical evidence, I attempt to contribute to the deconstruction of megacities as synonyms for human misery and hardship – not by overlooking poverty, exclusion and vulnerability that without doubt exist in Dhaka, but by highlighting people’s strengths, their confidence and their capacities to hang on despite all difficulties. 1.2 OUTLINE OF THE STUDY Geographers have contributed a lot to improve our common understanding of food systems and food security worldwide. With this study I want to add to this knowledge. I do this by diverging from other studies in three respects: 1) With a focus on the food system of Dhaka, I place attention on food security in cities in contrast to other studies that focus on rural areas (cf. Adhikari & Bohle 1999; Bohle et al. 1993; Devereux 2007; Dittrich 1995; Raithelhuber 2001; Rettberg 2009; Tröger 2004; Van Dillen 2004; Watts 1983). Even though in recent years, a number of studies on urban food systems have been put forward (Bohle & Adhikari 2002; Gertel 1995, 2010; Lohnert 1995), until today, food distribution in cities is least understood (Bohle et al. 2009). 2) While most of the above mentioned studies put the focus either on food production and food-short regions or on food consumption and food-insecure people, I concentrate on the locations where food

4

Introduction

enters the city, i.e. on wholesale markets, from where food flows into the various channels of the intra-urban food system. By doing so, I seek to meet Maureen Mackintosh’s (1990) call to study “real” food markets that she perceives as crucial link in the food system which determines who eats, how much and when. 3) Instead of asking for those factors and dynamics that make food systems vulnerable (cf. Blaikie et al. 1994; Bohle et al. 1994; Chambers 1989; Swift 1989; Watts & Bohle 1993), I shift attention to the resilience of Dhaka’s food system (cf. Bohle et al. 2009; Ericksen et al. 2010; Le Vallée 2008). From this vantage point, the focus of this study shall be directed toward people’s strengths to cope with and recover from adversities despite their limited resources and to their potentialities to develop institutions that improve their individual welfare and foster societal robustness toward future crises. I do this by emphasizing the particular role that wholesale traders play for feeding a megacity like Dhaka in the face of stress and crisis. By taking rice and fish as cases, I am in a position to compare the merchants’ situations in markets for perishable and non-perishable goods. This study exhibits the following structure: In the second chapter, I present key theoretical considerations and develop the conceptual framework that serves as guideline for the latter analysis. In the first part, I provide an outline to the contemporary geography of food systems and its current debate on vulnerability and resilience. In the second part, I compose a people-centered approach to social resilience which stands opposite to systems-oriented interpretations of resilience that are popular in the current debate on food systems and global environmental change. In the third part, I anchor the notion of social resilience in neoinstitutional organization theory and transfer it from development studies to economic geography in order to make it fruitful for the study of real markets. Real markets – the focus of part four – are understood as partly localized, partly virtual social spaces that comprise a set of institutions and a number of actors who are mutually interlinked by social relations, who possess divergent power resources, and who organize repeated exchange of goods while following the general aim to ensure the survival of their business organizations (Fligstein 1996, 2001; Fligstein & Dauter 2006). Three levels of analysis are distinguished for the study of food markets in Dhaka, i.e. practices, networks and arenas, which together reflect actors’ structural and their institutional embeddedness. In the fifth part, eventually, I summarize the findings of the theory chapter and specify the concern of this study: This is to study the economic resilience of food wholesale traders to business related risks and political uncertainties with the aim to understand the specific role they play for Dhaka’s food system. In the third chapter, I depict the methodology of the study. For data collection and analysis I resorted to the triangulation of qualitative and quantitative research methods. Mappings provided insights into the spatial organization, recent developments and present dynamics of the megacity’s food system. Observations, guided interviews and focus group discussions were of great use for getting insights into the everyday practices and personal perceptions of the wholesalers, and for revealing prevalent governance modes in Dhaka’s food markets. Venndiagrams, a participatory research tool, were of particular avail for understanding

Introduction

5

the meaning of the merchants’ business relations. Standardized surveys allowed for testing the representativeness of findings of the ethnographic field work, and proofed to be indispensable for capturing the different structures of the wholesale traders’ business networks. The empirical results of this study are presented in the chapters four, five and six. The fourth chapter (“The food system of Dhaka”) serves to equip the reader with the research context and illustrates the current situation and robustness of Dhaka’s food system. In the first part, I retrace Dhaka’s history and its development from city to megacity. In part two and three, I introduce Bangladesh’s rice and fishery economy, present recent production figures and assess Dhaka’s food system in terms of food availability and food accessibility. My principal research questions in this part are: How productive is Bangladesh’s rice and fishery economy? How is Dhaka’s food system spatially organized? And how can we evaluate it from a food security perspective? In chapter five (“Food wholesale markets in Dhaka”), I change the perspective and introduce Dhaka’s food markets and its actors to the reader. In the first part, I present how the numerous wholesale markets for food emerged in Dhaka in the past century. Subsequently, I outline Dhaka’s value chains for rice and fish in order to highlight the role that wholesale traders play for the urban food system. I complement this rather abstract description by more concrete figures on the very people who are involved in food wholesale trading in Dhaka and on their business organizations. My guiding questions in this section are: How did Dhaka’s food wholesale markets emerge historically? What are the basic characteristics of these markets? Who are Dhaka’s food wholesale traders? And what role do they play for the urban food system? After having answered these questions, the ground is prepared for discussing “The resilience of food wholesale traders in Dhaka” in chapter six. In a first step, I outline the food traders’ embeddedness in their business networks and depict basic organizational patterns of the food wholesale business that are fundamental for the traders’ capacities to cope with their everyday business. Afterwards, I ask for major threats to the more long-term survival of the business organizations in Dhaka’s wholesale trade and present evidence on what I call the adaptive capacities of the traders. Finally, I shift attention to the wholesalers’ position in the urban arena and to their potentialities for upgrading and development. Particular emphasis in this regard is put on the access to and appropriation of public places for marketing purposes, which turns out to provide key insights into underlying power structures of Dhaka’s food markets and into the traders’ so-called transformative capacities. My key questions in this part are: What are the major threats food traders have to deal with? How do the traders cope with and adapt to these threats? What can be said about the traders’ potentialities to transform present conditions? And, finally, who benefits from the current system and who bears its costs? The seventh chapter eventually provides space for a summary of findings and for some reflections on the meaning of this study for future research on megacities, urban food systems, real markets and social resilience. Figure 1 shows the research framework of this study.

6

Introduction

Figure 1: Research framework Normative perspective: Food security in megacities of the Global South Research focus: Rice and fish wholesale traders in Dhaka’s food markets Theoretical background: Geography of food systems Social and economic resilience Neoinstitutionalism and real markets Research methods: Triangulation of qualitative and quantitative methods Research questions: The food system of Dhaka How productive is Bangladesh’s rice and fishery economy? How is Dhaka’s food system spatially organized? How can we evaluate it from a food security perspective? Food wholesale markets in Dhaka How did Dhaka’s food wholesale markets emerge historically? What are the basic characterics of the markets? Who are Dhaka’s food wholesale traders? What role do they play for the urban food system? The resilience of food wholelsale traders in Dhaka What are the major threats food traders have to deal with? How do the traders cope with and adapt to these threats? What can be said about the traders’ potentialities to transform present conditions? Who bears the costs of Dhaka’s food markets? Central objectives: To study the role of wholesale traders in Dhaka’s food system To comprehend the factors that make Dhaka’s wholesale traders resilient against threats to the survival of their business organizations To understand the (individual) benefits and the (social) costs of contemporary food wholesaling in Dhaka

Source: Own draft

2. NAVIGATING REAL MARKETS

“Absorb what is useful, discard what is not, add what is uniquely your own” Bruce Lee (1940-1973)

In this chapter I am going to craft the conceptual framework of this study. The chapter is divided into four sections: The first one provides the reader with a short introduction to the geography of food security and retraces its trajectories from Malthus’s theory of famines to the actual call for resilience in globalized food systems. In the second section, the notion of resilience is re-framed from a social science perspective. Based on a critical review of recently published literature on the issue, social resilience is defined to comprise three dimensions: 1) Coping capacities: the ability of social actors to cope with and overcome all kinds of adversities; 2) adaptive capacities: their ability to learn from past experiences and adjust themselves to future challenges in their everyday lives; and 3) transformative capacities: their ability to craft sets of institutions that foster individual welfare and sustainable societal robustness towards future crises. Such a people-oriented understanding of resilience presumes a profound theoretical model of human agency, which is developed in the third section. Based on a critique of Marc Granovetter’s (1985) embeddedness approach, a call is made for an attention shift from actors’ structural to their institutional embeddedness. Programmatically positioned in the center of this chapter, an “actors-institutions loop” is developed that brings about the underlying theoretical key concepts of this study, i.e. institutions, actors, knowledge and practice. This heuristic allows to elaborate in the fourth section an explicit understanding of real markets, which are conceived to be constituted by the continuous interplay of powerful actors who create specific institutional arrangements in order to resolve fundamental coordination problems of interaction and exchange, and thus (re)produce – either implicitly or explicitly – the social order that constitutes markets as institutional facts. Section five provides a summary. I quote the famous martial artist Bruce Lee at the beginning of this chapter in order to express my personal attitude toward theory in general and the modus operandi of this chapter in particular. As Taoist and Buddhist philosopher, Bruce Lee argues for not following only one scholar in an orthodox manner, but for taking into account various schools of thought and different academic disciplines and for choosing one’s own perspective on the basis of a careful consideration of arguments.

8

Navigating Real Markets

2.1 GEOGRAPHY OF FOOD SYSTEMS 2.1.1 Food Production Thomas R. Malthus (2007[1798]) – well known as founding father of theoretical concern with food security – argued in his “Essay on the Principles of Population” that “[t]he power of population is indefinitely greater than the power in the earth to produce subsistence for man. Population, when unchecked, increases in a geometrical ratio. Subsistence increases only in an arithmetical ratio” (ibid.: 13f.). Malthus argued that the exponential population growth would sooner or later exceed available food supplies as the food production curve shows only a linear slope. He concluded: “The constant effort towards population […] increases the number of people before the means of subsistence are increased. The food therefore which before supported seven millions must now be divided among seven millions and a half or eight millions. The poor consequently must live much worse, and many of them be reduced to severe distress” (ibid.: 19). As such, Malthus continued, the limited carrying capacity of a particular region would play the role of a “natural” regulator of population growth in a world of generally finite available resources. In the absence of control mechanisms and “moral restraint”, hunger, famine and war would decrease population growth rates to finally balance food supply and demand in a drastic manner. Ever since the publication of Malthus’ study, the world has changed dramatically. One might think of the industrial revolution, the demographic transition in some regions of the world where populations have stabilized at zero growth, advances in transportation and communication technologies that have allowed food to be shipped around the globe to meet food deficit regions, and the significant increase of yields due to the widespread adoption of green revolution technologies. And yet, despite all these developments, Malthus’ explanations of hunger and famine have remained hegemonic for centuries – also in geography. WolfDieter Sick (1997) claimed even in the 1990s that food security could be understood in Malthusian terms. At that time, the topics of food insecurity and famine were dealt with in the subdiscipline of agricultural geography, which was understood as spatial science (Raumwissenschaft). The subdiscipline stood in line with Dietrich Bartels’ (1968) research agenda which implied a concept of absolute space in the tradition of Isaac Newton (Werlen 1999) and which sought to identify optimal agricultural regions on the basis of universal spatial laws (cf. Glückler 1999; Werlen 1997). Sick (1997) focused primarily on food production systems. As part of a similarly understood economic geography, the general aim was to elaborate localization theories for each economic sector that were coherent with models of classical and neo-classical economic theory. Only a short chapter in Sick’s (ibid.: 171-183) textbook was devoted to the food question in developing countries. Right at the beginning of this part, it was argued that this was essentially a question of the relation between food production and population figures. This clearly reflected a Malthusian perspective on the subject matter.

Navigating Real Markets

9

2.1.2 Food Distribution In the 1980s, however, another groundbreaking work caused this well-established explanation to hunger and famines to totter. In his famous study on “Poverty and Famines”, Amartya Sen (1981) criticized Malthusian explanations of food insecurity for their bare emphasis on food production and on questions of food availability. Instead, he called for the need to overcome production-centered approaches to hunger and famine and to shift attention to questions of distribution and people’s access to food. By means of empirical case studies on famines in Bengal (1943), Ethiopia (1972–74), the Sahel5 (1968–73) and Bangladesh (1974), Sen convincingly showed that famines had happened in times when there was not even a shortcoming of food supply. For the cases studied, he rather proved that rising prices next to unchanged income, associated with disastrous mismanagement of responsible governments, had barred large parts of societies from accessing enough food to survive. He concluded that “[s]tarvation is the characteristic of some people not having enough food to eat. It is not the characteristic of there being not enough food to eat. While the latter can be a cause of the former, it is but one of many possible causes” (ibid.: 1; emphasis by Sen). With his entitlement approach, Sen pioneered studies on the institutional frames that structure people’s capabilities to command over food, in times, when famines were still widely understood as “natural” disasters. Critique was mentioned that the entitlement approach, as developed by Sen, was biased by focussing legal rights while omiting prevalent local institutions (Devereux 2007). However, Sen (1981: 152ff.) himself noted that “[t]he analysis of [...] entitlements [...] can be extended in many ways by taking a more detailed view of the relationships that govern people’s ability to command food and other essential goods […] going from economic phenomena into social, political, and legal issues”. Against this background, large parts of the last decades’ studies on food insecurity can be said to be the analytic continuation of Amartya Sen’s ideas. In this spirit, geographers called for an alternative geography of food security in the 1980s and 1990s that would emphasize the very people who are suffering from malnutrition and food insecurity in order to better understand underlying causes of their situations (Atkins 1988; Bohle 1990; Watts 1983). Accordingly, a shift “from space to the human being” was called for, from geography as spatial science to geography as social science, with the aim to understand the vulnerability of people to hunger and famine (Bohle et al. 1993; Bohle & Krüger 1993). The vulnerability approach was elaborated that centered on Robert Chambers’ (1989) programmatic definition of the term: “Vulnerability here refers to exposure to contingencies and stress, and difficulties in coping with them. Vulnerability has thus two sides: an external side of risks, shocks and stress to which an individual or household is

5

Sen used a political definition of the sahel zone, covering Mauritania, Senegal, Mali, Upper Volta, Niger, and Chat (ibid.: 114).

10

Navigating Real Markets

subject; and an internal side which is defencelessness, meaning a lack of means to cope without damaging loss” (ibid.: 1; emphasis by author). In subsequent years, numerous authors contributed to the concept of vulnerability and widened its applicability to all kinds of situations of human misery, from land degradation, natural disasters and climate change (Birkmann 2006; Blaikie & Brookfield 1987; Bohle et al. 1994; Kasperson et al. 1995; O’brien et al. 2006) to deprivation, violence and health risks (Korf 2004; Moser 1998; Obrist 2006; Sakdapolrak 2010; Tröger 2004). For analyzing this wide range of empirical phenomena, structuralist approaches, such as theories of political economy, empowerment and human ecology, were seen to be necessary for the analysis of the “external” sphere of vulnerability (Watts & Bohle 1993). Actors-oriented approaches and theories of access, crises and conflict were deemed to be fruitful for the examination of the “internal” sphere of vulnerability (Bohle 2001). Among others, the vulnerability framework of Turner et al. (2003) became one of the most cited, as it identified crucial elements that guide social science oriented vulnerability analyses until today.6 In the 1990s, the concept of vulnerability found its way into the practical development cooperation with the creation of the sustainable livelihood approach. This approach aims at understanding poor people’s living conditions and their strategies to safeguard their family’s well-being in the face of stress and crisis in order to craft suitable support programs (Chambers & Conway 1992; DfID 1999; Krüger 2003; Moser 1998; Scoones 1998). 2.1.3 Food Systems Despite a steady growth of agricultural production around the world, and despite a tremendous increase in the international trade of edibles, global food security is far from being attained. Today, 15.0 percent of the population in the Global South are still undernourished (FAO 2011). Moreover, decades of deregulation and privatization policies together with innovations in the financial sector have led to an unprecedented volatilization of capital to which food prices are particularly sensitive (Altvater 2005; Ruel et al. 2010). In addition to that, global environmental change has unforeseeable impacts on food systems worldwide (Ingram et al. 2010; IPCC 2012). In this context, societies in general and urban food systems in partic6

In the last years, scholars have come to an understanding that vulnerability analyses need to consider the elements of hazards, exposure, sensitivity, coping strategies and adaptive capacities: Hazards can be said to be any event (natural, technical, man-made) that has the potential to cause the severe adverse effects that lie at the bottom of every emergency, disaster and catastrophe. Exposure is understood as the number of people and other elements at risk that can be affected by a particular incident. Sensitivity specifies people’s social status and their economic situation, which in turn determines the degree to which they will be affected by a hazard. Coping strategies encompass those measures that are pursued during the event with the aim to alleviate and contain damage. Adaptive capacities address the modification of behavior and activity in order to circumvent or avoid damaging events (Thywissen 2006).

Navigating Real Markets

11

ular have increasingly become subject to the “double exposures” (Leichenko & O’Brian 2008) of both an unpredictable nature due to global environmental change, and an unstable market economy due to globalization. These conditions have made scientists, policy-makers and non-governmental organizations call for systems-oriented approaches to food security in order to understand the global interconnectedness of today’s food production, trade and consumption activities. Scholars within the Global Environmental Change and Food Systems (GECAFS) project elaborated one such framework. They define food systems as the specific combination of social and natural components that leads to the potential satisfaction of nutrition of a given society through a set of spatially and functionally interlinked activities ranging from food production and import, over processing, packaging, exchange and distribution, to final consumption and disposal (Cannon 2002; Ericksen 2008). Four broad categories of food system activities are distinguished: Production includes all activities involved in producing raw food, from obtaining inputs such as land, labor, seeds, machinery, fertilizer and pestizides, over the planting of crops, the breeding of animals, and the care taking of both, to the final harvesting and slaughtering. Processing and packaging includes all transformations that raw food undergoes before it is sent to the markets for final consumption. The transformation of food can include the altering, combination or separation of food items, with consequences for their storage lifes, their nutritional and social value as well as their price. Exchange and distribution involves all activities that play a role for moving food from one place to another and marketing it. Eventually, consumption includes all activities from deciding upon and purchasing food, over its preparation, cooking and eating, to disposing the remains (Ericksen 2008). Food security is taken as the principal outcome of any food system, even though in many cases food security is not achieved. Global food security has been defined by the United Nations Food and Agriculture Organization (FAO) as “when all people, at all times, have physical, social and economic access to sufficient, safe and nutritious food that meets their dietary needs and food preferences for an active and healthy life” (FAO 2003). According to the GECAFS framework, food security involves three components that are availability, accessibility and utilization of food: Food availability refers to the amount, type and quality of food that is supplied to the end consumer and to the stability of these supplies. The major activities that contribute to availability are food production, exchange and distribution. Access to food is the ability to gain access to the type, quality and quantity of food required and can be analyzed in terms of the affordability of food that is available, how well allocation mechanisms such as markets or government policies work, and whether consumers can meet their social and cultural food preference. Utilization, finally, refers to the ability to consume and benefit from the nutritional and social values of food, and to the safety of available and affordable food (Ericksen et al. 2010). The value of food systems can thus be said to be a function of the stable provision of sufficient amounts (availability) of valued food (utilization) in the hand of the final consumer (accessibility) – predominantly the urban poor.

12

Navigating Real Markets

According to the GECAFS framework, a food system needs to be understood not in isolation, but as part of a wider social and ecological environment. The system is driven by socioeconomic factors such as demography, technology, political systems, cultural contexts and economic activity, and biogeophysical factors such as athmospheric components, climate variability, water availability, nutrients, biodiversity, or sea levels. At the same time, food system activities produce outcomes like employment, wealth, health, or social cohesion, but they also significantly impact on ecosystems through the entry of nitrate in soils, the spraying of crops with pesticides, the extension of areas under cultivation or the production of CO2 emissions in trade and transport (ibid.). 2.1.4 Resilience in Food Systems In the context of the present double exposures of global environmental change and globalization, today, the international debate on food systems revolves around the notion of resilience (Carr 2008; Gallopín 2006). In the face of current conditions scholars call for resilient food systems that are able to overcome future crises and enhance food security without further compromising environmental and social welfare outcomes (Ericksen et al. 2010). Like the notion of sustainability, resilience invokes a prospective and positive view, even though it does only exist in a context of potentially harmful change (Walker & Salt 2006). Folke et al. (2010) and others (Berkes et al. 2003; Holling 2001; Walker & Salt 2006) define resilience in general terms as comprising all those conditions and mechanisms that enable systems to maintain function under stress (persistability), to adapt to external change (adaptability) and to transform themselves in a way that enhances their future functionality (transformability). In this line of understanding, the resilience of a food system lies in its capacity to persist in its current state while facing disturbance, to adapt to economic, political, social and environmental changes, and to transform in ways that actually improves the food system outcomes in the future. Such a systemic approach to resilience has clear advantages as it enables researchers to capture the manifold interlinked relationships that comprise a food system, reaching from biophysical inputs over technological facilities to consumer practices (Ericksen et al. 2010). However, such an approach also has significant shortcomings, most prominently when it comes to questions of people’s agency, of underlying power relations and the role of the state (Bohle et al. 2009; Eakin et al. 2010; Schilpzand et al. 2010). From a social science perspective, Cannon & Müller-Mahn (2010: 623) argue that the concept of resilience is “inadequate and even false when it is being uncritically transferred to social phenomena”. Due to its empirical heritage rooted in ecosystem sciences, the concept is feared to “depoliticize” social structures and to unconsciously reinforce the status-quo of society by overlooking those mechanisms that put people at risk in the first place (Pelling & Manuel-Navarrete 2011). By advocating a positivistic, rationalistic and mecha-

Navigating Real Markets

13

nistic way of thinking, it would disguise the essence of the issue: power relations (Cannon & Müller-Mahn 2010). In my view the stated critical voices are correct and need to be acknowledged. However, they are not sufficient to dismiss the concept as a whole. I argue that resilience retains the potential to be crafted into a coherent analytic framework that, on the one hand, helps to disentangle the complexity of present-day food systems and, on the other hand, to carefully excavate the underlying webs of power that help stabilizing the current situation. Against this background, the aim of the next chapter is to reframe the notion of resilience from an actor-oriented perspective. In contrast to a systemic view, as mentioned above, such a perspective does not center on the resilience of food systems, but on people, e.g. producers, traders or consumers, and their resilience in food systems. Such an understanding of resilience provides space for analyzing the underlying dialectics of structure and agency at play: the economic, political, social and ecological conditions of food systems structure people’s resilience and their food security, while people make urban food security and the resilience of food systems through their everyday practices and personal interactions. 2.2 SOCIAL RESILIENCE7 2.2.1 What is Resilience? The concept of resilience has its roots in ecosystem sciences (besides psychology) and evolved stepwise from its initial emphasis on the general persistence of ecological system functions in a world that is subject to ongoing change, through an orientation towards coupled social-ecological systems and questions of the adaptation of humans in nature, to its most recent readjustment in taking up the more critical question of social transformation in the face of global change. Crawford Holling’s (1973) article on “Resilience and Stability of Ecological Systems” is referred to as groundbreaking work in the study of resilience (e.g. Walker & Salt 2006). By discussing examples such as spruce budworm outbreaks and their role for boreal forests in Canada (Holling 1973, 1986, 1996), Holling made the case that ecosystems would reveal nonlinear dynamics. With his paper, he radically called into question former static, equilibrium-based models of ecosystems. Instead he proposed to approach them as complex, adaptive systems that would retain cyclicity and exhibit a multitude of possible stable states, or “basins of attraction”. With the notion of resilience, he addressed “the persistence of [ecological] systems and their ability to absorb change and disturbance and still maintain the same relationships between populations and state variables” (Holling 1973: 14). This (ecological) resilience was measured by the magnitude of disturbance that a sys7

Parts of this section have meanwhile been published in Keck & Sakdapolrak (2013).

14

Navigating Real Markets

tem could tolerate and still persist (Carpenter et al. 2001). Such an understanding was fundamentally different from the meaning implied by the more established term “stability”, which described the ability of a system to return to an equilibrium state after a temporary disturbance (Holling 1996) and was also related to the time required for the system to return to this equilibrium (Pimm 1984). In shifting from the logic of stability to that of resilience, the emphasis was placed on those characteristics that enable systems to live with disturbance and instability and which promotes their inherent flexibility and strengths that would increase their chance of persistence. In subsequent years, scholars of the Resilience Alliance, an international and interdisciplinary research network, further substantiated the idea of resilience by conducting empirical case studies on coupled social-ecological systems (Carpenter et al. 1999, 2001; Walker 1993; Walker et al. 1981, 2002). Empirical findings and conceptual considerations merged into the elaboration of the meta-theoretical model of the “adaptive cycle” (Berkes et al. 2003; Gunderson & Holling 2002; Holling 2001). The adaptive cycle is a heuristic model that portrays an endogenously driven four-phase cyclicity of complex systems. The general phases that these systems pass through are periods of 1) accumulation and growth, 2) stagnation, rigidity and lock-in, 3) sudden collapse, and 4) re-organization and renewal. With the notion of “panarchy”, cross-scale dynamics and the interplay between nested adaptive cycles are addressed, in which the analyzed system is affected by both higher-ranked, slower cycles and subordinated, faster cycles. In this second phase of the concept’s lifespan, (social-ecological) resilience was defined as the “capacity of a [social-ecological] system to absorb disturbance and re-organize while undergoing change so as to still retain essentially the same function, structure, identity and feedbacks” (Folke 2006: 259). This definition served the aim of integrating the two ideas of ecological resilience and that of the adaptive cycle. As such, social-ecological resilience was defined as the magnitude of change the system could undergo and still remain within the same stable state (cf. ecological resilience), and the system’s degree of self-organization (Holling 2001), understood as its capacity to re-organize after perturbations in an emergent and path-dependent manner (cf. adaptive cycle). In order to make the concept applicable for sustainability studies, the system’s capacity for learning and adaptation was also included as third factor (Berkes et al. 2003). With the notion of adaptation or adaptability, proponents of resilience thinking positioned themselves within the climate change discourse by raising the question of whether “humans in nature” might be able to combine their experience and knowledge to successfully adapt to global environmental change. Resilience emerged as a “boundary object” (Star & Griesemer 1989; Star 2010) positioned between two communities of practice – i.e. natural and social sciences – and represented a means by which to allow for interdisciplinary collaboration and exchange. One of the fundamental ideas of resilience thinking was “that environmental problems cannot be addressed in isolation of the social context” (O’Brien et al. 2009: 5). As such, the concept of resilience could be seen as an invitation extended by natural scientists to social scientists to engage in integrative research under

Navigating Real Markets

15

the banner and normative goal of sustainability. As a response to the critique of conservatism implicit in the reading of the concept of resilience as applied to social systems, which has been raised by critical social scientists (Pelling & ManuelNavarrete 2011), most recently, resilience proponents have updated their concept by adding the notion of transformation or transformability. As stated above, a system is seen to possess multiple potential stable states, or basins of attraction, which together constitute its “stability landscape” (Gallopín 2006). In being exposed to a specific shock or stress, or through changes in internal structures and feedback loops, the system might move from one basin into another, and thus exhibit changes in its functionality. The notion of transformability, then, addresses a system’s capacity to transform the stability landscape and to create new system pathways when ecological, economic or social structures make the existing system untenable (Folke et al. 2010; Walker et al. 2004). The word “untenable” (Walker et al. 2004: 1) unmistakably addresses those issues that have so far been at the heart of the development discourse, i.e. equality, justice and human rights. This new focus on transformability can be said to have heralded the third phase of the concept’s lifespan. Against this background, the genealogy of the concept of resilience can be summarized as having evolved stepwise from its initial focus on the persistability of ecological system functions, through an emphasis on the adaptability of coupled social-ecological systems, to its most recent reorientation towards addressing the transformability of society in the face of global change. By taking these three genealogical steps as highlighting the underlying principles that constitute the concept, resilience can be defined in its most general sense as a system’s capacity to persist in its current state of functioning while facing disturbance and change, to adapt to future challenges, and to transform in ways that enhance its functioning. 2.2.2 What is Social Resilience? A first definition of social resilience was provided by Neil Adger (2000: 361) who considered it “as the ability of communities to withstand external shocks to their social infrastructure”. Rather like the aforementioned understanding of resilience as the ability to persist, the focus of this definition was on the capacities of social entities to protect themselves from all kinds of hazardous events. With a similar understanding in mind, Turner et al. (2003: 8075) incorporated the notion of resilience into their vulnerability concept and defined it as “system’s capacities to […] respond”: These responses, they write, “whether autonomous action or planned, public or private, individual or institutional, tactical or strategic, short- or longterm, anticipatory or reactive in kind, and their outcomes collectively determine the resilience of the coupled system” (ibid.: 8077). Against this background one could argue that resilience is a combination of those elements that have been addressed in former concepts with the terms “coping strategies” and “adaptive capacity”. However, the idea of resilience extends

16

Navigating Real Markets

beyond these two elements. The concept of resilience is intrinsically dynamic and relies on the Heraclitean notion of “everything changes, nothing remains still”. As such, it encompasses uncertainty, change and crisis as normal, rather than exceptional conditions. Therefore, the analysis of social resilience is geared toward understanding the mechanisms by which a system can adapt not only to the challenges that are directly at hand, but also to those that are unexpected and unknown (Kates & Clark 1996; Streets & Glantz 2000). Glavovic et al. (2003: 291) have made this explicit by defining social resilience as “the capacity to absorb […] change – the ability to deal with surprises or cope with disturbances”. In a second step the definition of social resilience was widened by including further skills and know-how that were deemed necessary for successfully dealing with uncertainty and change. Pelling (2003: 48), for instance, holds that social resilience is “a product of the degree of planned preparation undertaken in the light of a potential hazard, and of spontaneous or premeditated adjustments made in response to felt hazard, including relief and rescue”. Cutter et al. (2008: 599) defines social resilience as “the ability of a social system to respond and recover from disasters” and states that it “includes those inherent conditions that allow the system to absorb impacts and cope with an event, as well as post-event, adaptive processes that facilitate the ability of the social system to re-organize, change, and learn in response to a threat”. Both Pelling (2003) and Cutter et al. (2008) underline the anticipatory capacities and the pre-hazard preparedness of social actors and the capacity of a social system to learn from hazardous events how to better deal with it in future. This positive feedback of learning from past crises to better deal with uncertainties in the future has given direction to new composite definitions. Glavovic et al. (2003: 290f.) have written that social resilience is basically “influenced by […] institutions […] and networks that enable people to access resources, learn from experiences and develop constructive ways of dealing with common problems”. Based on these considerations, Obrist et al. (2010: 289) define social resilience “as the capacity of actors to access capitals in order to – not only cope with and adjust to adverse conditions (that is, reactive capacity) – but also search for and create options (that is, proactive capacity) and thus develop increased competence (that is, positive outcomes) in dealing with a threat”. From this perspective, social resilience not only addresses social actors’ or entities’ capacities to protect themselves from all kinds of threats. In addition to absorptive capacities in the face of perturbations and stress, the idea of social resilience also implies that catastrophes may be perceived as opportunities for doing new things, for innovation and development (Bohle et al. 2009). Being fully resilient, then, means coping with future crises by learning about which actions are more or less appropriate in the context of uncertainties through undergoing shocks and distress. Therefore the key question of social resilience is, as Obrist et al. (2010: 291) have pointed out, “what enhances capacities of individuals, groups and organizations to deal with threats more competently”. Even though this recent elaboration of the concept of social resilience may sound promising, many social researchers have deemed it too optimistic, since

Navigating Real Markets

17

social actors’ specific contexts are neglected. Lorenz (2013) has rightly pointed out that whether persons are able to cope with threats, learn from them, and adjust to future crises is not only decided by the persons themselves, or by their endowments and willingness to invest into mitigating and adaptive measures; most of all, it is a question involving all those societal factors that both facilitate and constrain people’s abilities to access assets, to gain capabilities for learning, and to become part of the decision-making process. Therefore, at its heart, social resilience has to provide a suitable answer to the question of the interplay between social structures and the agency of social actors (Bohle et al. 2009). Voss (2008: 45) has made clear that “the predominant opinion was that the pressure of an ‘objective’ problem was enough to initiate solution oriented processes. This was based on a fundamental trust that all problems today or in the future could be successfully dealt with through technology and science […] in a cloud of apoliticalness”. However, this opinion has turned out to be wrong; firstly, because the different perspectives and expectations of diverse stakeholders must be taken into account when coming to terms with today’s challenges, and secondly, because there is no Habermasian ideal speech situation (Lorenz 2013). What types of threats are perceived, how they are dealt with, and whether the poor and marginalized are heard or not – all of these are questions related to actors’ capacities to participate in governance processes (Voss 2008). If alternative or critical voices remain unheard for the sake of implementing standard solutions, the “participative capacity” of the system, and the unequal distribution of power and knowledge, become key issues of social resilience (e.g. Bohle et al. 2009; Davidson 2010; Glavovic et al. 2003; Lorenz 2013; O’Brien et al. 2009; Obrist et al. 2010). In reality, there are situations that make shifts in dealing with risks necessary which exceed established methods of coping and adapting. These shifts might include technological innovations and policy reforms. Such shifts are, however, strongly influenced by social factors (ethics, knowledge, attitudes to risk and culture) and social thresholds and start with people’s questioning of their everyday lives and routines, their norms, values and taken-for-granted assumptions about reality (Adger et al. 2009; Christensen & Krogman 2012; O’Brien 2011). Hence, Béné et al. (2012) have rightly underscored that there are considerable barriers to societal transformations that are rooted in systems of power and knowledge, which are often deeply entrenched in the social structures and protected by powerful interests.

18

Navigating Real Markets

2.2.3 Three Capacities of Social Resilience The state of the current debate over how to define social resilience has reached a point where several authors such as Voss (2008), Obrist et al. (2010), Béné et al. (2012) and Lorenz (2013), have suggested that three different types of capacities are necessary for understanding the notion in its full meaning. These are labelled coping capacities, adaptive capacities and transformative capacities. The differences between these three capacities of social resilience are shown in Table 1 along three dimensions. The first dimension addresses the different rationalities the respective capabilities are based on. In this case, a continuum is spanned from a person’s measures to immediately restore her present state of well-being, through various forms to secure her future well-being, to far-reaching steps to enhance her present and future well-being. The second dimension, the temporal scope, refers to the time horizon that is addressed. A continuum is spanned between agency based on immediacy and short-term thinking, and project-related, calculative agency based on a more long-term rationale. The third dimension points to the primary scale of analysis and ranges from specific forms of individual practice, over interpersonal relationships and network structures, to the wider socio-political arena a person is positioned in. Table 1: Three capacities of social resilience Coping Adaptive Transformative capacities capacities capacities Rationale Restoration of preSecurity of future Enhancement of sent level of wellwell-being present and future being well-being Temporal scope Short-term Long-term Long-term Scale of analysis Individual practices Social networks Political arenas Source: Own draft based on Béné et al. (2012), Lorenz (2013), Obrist et al. (2010), Voss (2008)

By means of these criteria, the three capacities of social resilience can be described as follows: 1) Coping capacities address “re-active” (ex-post) (Obrist 2010) and “absorptive” (Béné et al. 2012) measures of how people cope with and overcome immediate threats by the means of those resources that are directly available. The rationale behind coping is the restoration of the present level of well-being directly after a critical event. 2) Adaptive capacities refer to the “pro-active” (ex-ante) (Obrist 2010) or “preventive” measures (Béné et al. 2012) that people employ to learn from past experiences, anticipate future risks and adjust their livelihoods accordingly. Adaptation serves to secure the present status of people’s well-being in the face of future risks. The major difference between coping and adaptation is grounded in the temporal scope of the activities involved. While coping addresses tactical agency and short-term rationale, adaptation involves strategic agency and more long-term planning.

Navigating Real Markets

19

3) Finally, transformative capacities, or “participative capacities” in the words of Voss (2008) and Lorenz (2013), encompass people’s ability to access assets and assistance from the wider socio-political arena (i.e. from governmental organizations or civil society), to participate in decision-making processes, and to craft institutions that both improve their individual welfare and foster societal robustness toward future crises. The main difference between transformation and adaptation refers to the outcome it implies. Transformation is geared towards a shift in which the objective is not to secure, but to enhance people’s well-being in the face of present and/or future risks. As such it explicitly incorporates topics of progressive change and development. 2.2.4 Anchoring Resilience in Social Theory The concept of resilience was shown to have evolved stepwise from its initial focus on the persistence of system functions, through an emphasis on adaptation, to its most recent reorientation towards addressing the transformation of society in the face of global change. In loose correspondence with these genealogical steps, the idea of social resilience was developed from its initial meaning, referring simply to actors’ capacity to respond, and enlarged to encompass actors’ capacity to learn and adapt; now the concept also includes their capacity to participate in governance processes and to transform societal structures themselves. With the triad of coping capacities, adaptive capacities, and transformative capacities, the notion of social resilience thus shares the key principles of the general resilience concept. However, it departs from the general resilience discourse by adopting an actor-oriented perspective. Such an approach places the spotlight on social actors, rather than on systems, and focuses on capacities and practices instead of functionalities. This shift is necessary, as it brings important issues such as power, politics and participation back onto the agenda of the resilience debate. In my view, this mitigates much of the concern expressed by Cannon and Müller-Mahn (2010) with regard to the potentially depoliticizing effect of applying the concept of resilience to social contexts. The classical topics of resilience research are located in the areas of disaster control, resource management and climate change adaptation. I propose instead to transfer the concept of social resilience into the economic sphere in order to make it fruitful for the empirical study of markets. Before I can do so, the next step to take is to anchor the concept in social theory by elaborating a profound theoretical understanding of human agency. Thus, in the next section, the underlying social theory of this study is made explicit. At this point, it must be mentioned that none of the recently published works on markets get along without referring to the groundbreaking article of Granovetter (1985) on “Economic action and social structure: the problem of embeddedness”. This piece is similarly the founding manifesto of the new economic sociology (Beckert 2007) as it is the cornerstone of the relational economic geography (Bathelt & Glückler 2012). In this study, the notion of embeddedness serves

20

Navigating Real Markets

an important reference too. However, the term is not adopted without modifications that move it closer to its original meaning, how it has been formulated by Polanyi (2001[1944]). Thus, in the first part of the next section, Granovetter’s embeddedness approach is critically reexamined. Based on this critique, a distinction is made between actors’ structural and their institutional embeddedness. It is this distinction that allows to provide an elaborated understanding of human agency with its key elements of institutions, actors, knowledge and practice. These elements together converge in an „actors-institutions loop“ that serves as central heuristic figure of this study. 2.3 EMBEDDED AGENCY8 2.3.1 From Structural to Institutional Embeddedness With his seminal work on “Economic action and social structure: the problem of embeddedness”, Marc Granovetter (1985) made an attempt to settle down a decade long dispute between sociologists and economists on “undersocialized” and “oversocialized” conceptualizations of human agency by introducing the notion of people’s “embeddedness” that stresses “the role of concrete personal relations and structures (or “networks”) of such relations” (ibid.: 490). Due to cumulative interpretative misunderstandings, however, it is the irony of the career of this concept that it led to the reproduction of both former antagonistic positions (Beckert 2007). Granovetter begins his essay by recalling Dennis Wrong’s (1961) complaint about an “oversocialized” conception of human agency in modern sociology. He agrees with Wrong that in conceptualizations such as the structural functionalism of Talcott Parsons (1968[1937]), a picture is drawn of people who are “obedient to the dictates of consensually developed systems of norms and values, internalized through socialization” (Granovetter 1985: 483). From this perspective, the momentum of individual decision making is replaced by a pre-given “common sense” that is followed by actors in a blind and mechanical manner. Thus, people are seen as no more than robots, carrying out an underlying structural or cultural determinism. In a similar way, then, Granovetter turns his critique against the actor model of contemporary economics. While classical and neoclassical economic approaches exclude by definition any impact of social structures on economic outcomes, they reveal an “undersocialized” conception of human agency (ibid.). From such a vantage point, actors are treated as acting on the basis of independent, individualistic decisions by pursueing the universal “law” of self-interested utility maximization. With no societal factors intervening, such an actor model clearly reveals an extreme form of voluntarism (Bathelt & Glückler 2012).

8

Parts of this section have meanwhile been published in Etzold et al. (2012).

Navigating Real Markets

21

Despite these apparent contrasts between both views, Granovetter (1985: 485) highlights their implicit similarities: “[B]oth have in common a conception of action and decision carried out by atomized actors. In the undersocialized account, atomization results from narrow utilitarian pursuit of self-interest; in the oversocialized one, from the fact that behavioral patterns have been internalized [...]. Under- and oversocialized resolutions of the problem of order thus merge in their atomization of actors from immediate social context”. He thus concludes that “a fruitful analysis of human action requires us to avoid the atomization implicit in the theoretical extremes of under- and oversocialized conceptions. Actors do not behave or decide as atoms outside a social context, nor do they adhere slavishly to a script written for them by the particular intersection of social categories that they happen to occupy. Their attempts at purposive action are instead embedded in concrete, ongoing systems of social relations” (ibid.: 487). While such a mediatory position between under- and oversocialized conceptions can be appreciated in general, the way it has been worked out by Granovetter and others needs to be called into question: Granovetter adopted the notion of embeddedness from Karl Polanyi (2001[1944]). For Polanyi the meaning of embeddedness was twofold (Beckert 2007): First, he conceived economies and markets as necessarily being anchored in the institutional regulations of society which connect them also to concepts of morality. Second, the term of embeddedness served a deeply emancipatory and social reformist task: According to Polanyi, the economy had been historically associated with principles of reciprocity and redistribution. It was only capitalism, i.e. the discovery of self-regulating market mechanisms and its expansion to the “fictious commodities” of land, labor and money, which altered the system fundamentally and in a noxious manner, because “to include them [the factors of production, i.e. land, labor and money] in the market mechanism means to subordinate the substance of society itself to the laws of the market” (Polanyi 2001[1944]: 75). With the notion of embeddeness, Polanyi thus offered a general critique of capitalism. After Granovetter’s reinterpretation, though, the notion of “embeddedness” became successively reduced to the structural properties of social networks that was used as key variable for explaining human agency and economic outcomes of business organizations. As such it has provided vivid impetus to the so-called network approach (Burt 1992; Coleman 1990; Grabher 1993, 2006; Granovetter 1973; White 1992). This network approach, however, exhibits a limited focus on the structure of social relations while neglecting the social content underlying the observed structures (Beckert 2007). In this manner, the notion of embeddedness has been eagerly taken up by rational choice theorists who integrated it into their utilitarian actor models. Thus, the irony of the recent career of the embeddedness concept consists in the fact that it has failed to accomplish both aims of Polanyi: Neither his proposed emphasis on societies’ institutional frames for the explanation of economic outcomes was accounted for, nor his critical effort and his political intention to show ways of social reform. In fact, Granovetter’s concept was rather (mis)used to abolish the sociological grounding of human agency in economic sociology.

22

Navigating Real Markets

It is the aim of the subsequent pages to elaborate an adequate theoretical model of human agency that keep Granovetter’s promise to offer a balanced perspective on economic action that avoids both traps of under- and oversocialization and that comes closer to Polanyi’s intentions to provide a critical analysis of society. This is done by taking up the basal distinction between the structure and the content of social relations that shall serve as starting point for further considerations. As it should have become clear, Granovetter’s (1985) understanding of embeddedness puts emphasis on the structure of social networks, while Polanyi’s (2001[1944]) institutional embeddedness shifts the focus to intelligible aspects and, thus, to the meaning of social relationships. The main focus of the upcoming elaborations is on people’s institutional embeddedness and on social actors’ “embedded agency” (Walgenbach & Meyer 2008). As mentioned above, the next parts of this section are organized around four key terms, that are institutions, actors, knowledge and practice. These key terms are developed with reference to the philosopher John Searle (1995), the sociologists Anthony Giddens (1984) and Richard Scott (2008), and the economists Douglass North (1990) and Geoffrey Hodgson (1998). 2.3.2 Institutions What are institutions? John Searle (1995) provides an ingenious answer from the perspective of critical realism. He begins his work on “The Construction of Social Reality” by dividing the “real” world into two ontological boxes. On the one hand, there are “brute facts” whose existence is independent of any cognitive aspects of human mind: Mount Everest has snow and ice near the summit, hydrogen atoms have one electron (ibid.: 2). On the other hand, there are “institutional facts” like money, marriage, property, governments and markets that exist only because people believe them to exist. The first kind of facts are discribed as being “ontologically objective”, which identifies them as being “intrinsic to nature”. The second kind of facts are seen to be “ontologically subjective”, as they exist only in relation to the intentionality of people (ibid.: 8). In a second step, Searle subdivides ontologically subjective facts of reality into those that are “epistemically subjective” and those that are “epistemically objective”. The former facts represent individual’s attitudes toward reality like to be seen in the statement: “Apples are too expensive”. The latter facts instead are a matter of social consensus or of “collective intentionality” (ibid.: 23). Money, for instance, has a symbolic value that is not dependent on a single person’s attitudes and feelings about it. Nevertheless, it comes into existence only because people believe specific pieces of paper are Euros, US-Dollars or Bangladesh Taka. Epistemically objective facts, thus, exist independently of individuals’ representations, though not independently of all representations (ibid.: 190). Searle explicitly argues against the reduction of states of minds to individuals or to an opaque super-individual (ibid.: 25). Instead, “collective intentionality” refers to people’s ability to share beliefs, desires and intentions and to their com-

Navigating Real Markets

23

petence to direct their minds at something in mutual agreement (ibid.: 7). In cases such as people who are playing football together, “I intend only as part of our intending” (ibid.: 26). In general, collective intentionality leads to “social facts”, to be seen in events such as taking a walk or playing football together. However, specific cases of collective intentionality lead to “institutional facts”, namely those that are constituted by the ascription of a symbolic status to either brute or social facts. Searle summarizes this specific case of the creation of institutional facts by the logical formula of “X counts as Y in C” (ibid.: 43). As such the brute fact of a piece of paper (X) counts as money (Y) in the context of a monetized economy (C). Similarly, the social fact of a handshake (X) counts as relational contract (Y) in the context of Dhaka’s food markets (C). It goes without question that “the contrast between epistemic objectivity and epistemic subjectivity is a matter of degree” (ibid.: 8). This makes the creation of institutional facts tricky since it is subdued to constant negotiation processes within particular societies. Adelstein (2010) suggests in this regard to distinguish between a “we consciousness” and a “they consciousness” that accounts for divergent valuations of institutional facts to be real and relevant or not from the perspective of different groups of social actors. Searle highlights three basic features of institutions, that are the ordering of procedural steps of interaction, the deontic powers of creating rights and obligations, and the symbolic powers of creating meaning (Searle 1995). These features are congruent with Richard Scott’s (2008) three pillars of institutions. Scott (ibid.: 48) defines institutions as “comprised of regulative, normative and culturalcognitive elements that, together with associated activities and resources, provide stability and meaning to social life”. Thus, three layers or pillars of institutions can be distinguished: 1) rules; 2) norms and values; as well as 3) world views. 1) Rules are found in regulatory processes and can be understood after Douglass North (1990: 3) as formal “rules of the game in society or [...] the humanly devised constraints that shape human interaction”. As such they comprise sets of more or less codified regularizations and codes of conduct, decisionmaking procedures, as well as monitoring and sanctioning mechanisms that together guide human conduct. 2) “Values are conceptions of the preferred or the desirable [...]. Norms specify how things should be done; they define legitimate means to pursue valued ends” (Scott 2008: 54f.; emphasis by author). Together values and norms help to define the objectives of actors and the appropriate ways to achieve them. Moreover, norms decide upon actors’ social roles, their obligations and privileges, and their power position in society (Giddens 1984). 3) Worldviews work at the deepest level of society. While human cognition mediates between the external world and actors’ symbolic representations of and ascription to it, the very elements of representations – symbols such as words, signs or gestures – are socially constructed. As “constitutive rules” (Searle 1995), these symbolic systems help actors to break down the reality into comprehensible categories, which make sense and which can then be translated into appropriate social practices. These deeply embedded cultural “frames” (cf. Goffman 1974) are

24

Navigating Real Markets

collectively shared in a social group and most often simply taken for granted. Inadvertently, they shape the meaning that actors assign to objects, practices and places as well as to their own identities. Capturing this cultural-cognitive dimension of institutions requires the inclusion of emotional facts such as confusion and disorientation or confidence and certitude into institutional analysis (Scott 2008). By taking these considerations into account, institutions are defined as guidelines to human conduct that are permanently reproduced by means of social interaction, that comprise regulative, normative and cultural-cognitive elements, and that enable, constrain and give meaning to social life. 2.3.3 Actors The fact that institutions typically portray a high degree of stability and invariance over a long time and may even outlast the life span of individuals provided institutionalists a reason for choosing institutions rather than individuals as basic research units (Giddens 1984). Nevertheless, a pure focussing on institutions bears the risk of getting trapped in an extreme form of cultural or structural determinism (Hodgson 1998). In order to separate themselves from such conceptualizations, many scholars have thus applied the “methodological individualism” with a focus on atomized actors as research units that are treated as taken for granted. From such a perspective, actions are perceived as the result of “rational” decision making processes, an understanding that is reflected in James Duesenberry’s (1960: 233) famous quote: “Economics is all about how people make choices; sociology is all about how they don’t have any choices to make”. As mentioned above, a pure focussing on individuals however, bears the risk of getting trapped in an extreme form of voluntarism (Bathelt & Glückler 2012). Geoffrey Hodgson (1998) points out that the predominance of focussing on individuals as research units stems from assumptions on the nature of human actors as they derived from the 300-year old tradition of enlightenment. In fact, it was René Descartes (2004[1641]) who advocated for the individual as the first “source of truth” that he attempted to prove by his famous thought experiment. Against this background, Thomas Hobbes (2003[1651]) developed his idea of an institutions-free “state of nature”. Such a state of nature can still be found in contemporary works, such as in game theory or in neorealistic international relations. However, “[i]n any original, hypothetical “state of nature” from which institutions are seen to have emerged, a number of rules, and cultural and social norms are already presumed” (Hodgson 1998: 182). Descartes’ “Cogito me cogitare, ergo sum” and – related to that – any assumptions of an individual that is taken for granted and of a state of nature which exisits before institutions come into existence exhibits a logical error for the simple reason that this bare thought never appears at all unless it is made manifest by the means of the institution of language (Arendt 1978). Accordingly, Hodgson (1998) concludes: As institutions cannot be adequately explained in terms of individuals and their rational behavior, individuals and

Navigating Real Markets

25

their conduct cannot be explained in terms of bare institutions. Instead, both actors and institutions have to be conceptualized in an interrelated manner. Such an interrelated conceptualization of actors and institutions is reflected in Anthony Giddens’ (1984) call for a “decentered subject”. This does not imply the evaporation of subjectivity into an empty universe of signs but accounts for the fact that “[b]y institutions, individuals are not merely constrained and influenced. [...A]s social beings we are constituted by institutions. They are given by history and constitute our socio-economic flesh and blood” (Hodgson 1998: 189; emphasis by Hodgson). Accordingly, individuals, organizations and nation states cannot be seen as staying “outside” of an institutional context. They must rather be conceived as open social entities that are pervaded by their institutional environ (Walgenbach & Meyer 2008). Actors and societies, “do not have ‘structures’ but rather exhibit ‘structural properties’ and that structure exists, as time-space presence, only in its instantiations in [...] practices and as memory traces orienting the conduct of knowledgeable human agents” (Giddens 1984: 17; emphasis by Giddens). The constitution of actors and institutions are thus not two independently given sets of phenomena. They rather represent a dialectic interrelation. They seem to be contradictory but, at the same time, they are logically interdependent. As Berger & Luckmann (1967: 75) stated, institutions are dead “unless they are ongoingly ‘brought to life’ in actual human conduct”; they do not exisit as such unless they are realized by actors. Similarly, actors, their manifold social roles, and their practices in the course of everyday life are themselves only instantiations of broader social institutions (Giddens 1984). 2.3.4 Knowledge and Practice To elaborate on the interplay of actors and institutions, it is fruitful to refer to what Giddens calls the “knowledgeability” of actors. Giddens (1984: 375) defines the term as “[e]verything which actors know [...] about the circumstances of their action and that of others [...] including tacit as well as discursively available knowledge”. Giddens distinguishes between actors’ discursive and practical consciousness. “Discursive consciousness means being able to put things into words” (ibid.: 45). It represents actors’ abilities to give verbal expression about their actions, their interests, needs and objectives. Practical consciousness, on the contrary, “consists of all the things which actors know tacitly about how to ‘go on’ in the contexts of social life without being able to give them direct discursive expression” (ibid.: xxiii). Contrary to rational choice models – also those that assume bounded rationality – Giddens underlines that it is not the discursive but the tacit knowledge that is constitutive for most of our daily conduct. Conscious decisions do only play a marginal role in regard to human agency; in fact they are rather the exception than the normal case. In this understanding, Giddens comes close to what “old” institutionalists like Thorstein Veblen and others wrote about habits. Habits were understood as “largely non-deliberative and self-actuating propensity to engage in a previously

26

Navigating Real Markets

adopted pattern of behavior. A habit is a form of self-sustaining, non-reflective behavior that arises in repetitive situations” (Hodgson 1998: 178). It is these habitualized activities, that are grounded in the tacit knowledge of practical conscioussness, that bring about stability to human interaction by the means of institutionalized, historically sedimented, and thus often reified social practices (Berger & Luckmann 1967). These “routines” are “integral both to the continuity of the personality of the agent, as she or he moves along the paths of daily activities, and to the institutions of society, which are such only through their continued reproduction” (Giddens 1984: 60; emphasis by Giddens). Given the interrelation between institutions and actors and the significance of routines for the continuity of both, how then can human agency be conceptualized? With Meyer et al. (1987: 13), human conduct is nothing but actors’ “enactment of broad institutional scripts”. However, actors do not follow such scripts in a blind manner, but have a potential to choose between a multitude of such scripts. Jepperson (1991: 148), on the other hand, put it that way: “[O]ne enacts institutions; one takes action by departing from them, not by participating in them”. But from what has been written above, it should have become clear that actors can never reach a space that is free of institutions. In order to answer the above raised question, it is once again referred to Giddens. From Giddens’ (1984: 9) point of view, “[a]gency refers not to the intentions people have in doing things but to their capability of doing things in the first place. [...] Agency concerns events of which an individual is the perpetrator, in the sense that the individual could, at any phase in a given sequence of conduct, have acted differently”. Thus, agency “depends upon the capability of the individual to ‘make a difference’ to a pre-existing state of affairs or course of events” (ibid.: 14). This basic ability of actors to “make a difference” can be said to also stem from their discursive consciousness and their capacities for the reflexive monitoring of their activities. Reflexive monitoring means the purposive and intentional character of human behaviour (ibid.). There are two crucial points that need to be understood in regard to the discursive consciousness of institutions. First: Actors are not able to choose from alternatives and to intervene into the state of affairs despite the fact that they are institutionally embedded. In fact, actors are able to choose from alternatives and to intervene into the state of affairs because they are institutionally embedded. As their socioeconomic flesh and blood, institutions do not only constrain actors, they rather enable them in the first place. This can be seen from the institution of language as it has been formulated in the well-known Sapir-Whorf hypothesis (Sapir 1964[1929]): Language restricts the ways of thinking by the means of a limited amount of symbols and meanings that are available in a given cultural milieu while, at the same time, it enables human beings to logically reason. Thus, institutions “serve to open up certain possibilities of action at the same time as they restrict or deny others” (Giddens 1984: 173f.). Second: Institutions do not provide actors an explicit code of conduct for each and every specific and individual situation. Institutions are generalizations; they are only capable to provide broad guidelines that allow actors a minimum of meaning and orientation. Institutions cannot

Navigating Real Markets

27

be deterministic to human agency since concrete situtions of human interaction are never identical. Hence, institutions do only seem to stay static over time. In fact, they are ongoingly reproduced, and this reproduction process is open for the interpretation by actors, for human creativity and social innovation. With Giddens’ analytic term of social practice, always two aspects of human conduct are involved, i.e. habitualized routines and intelligible actions: Routines are crucial for the reproduction of both social roles and institutionalized practices that bring about stability and order to societal entities. Action refers to the fact that this reproduction is open for actors’ interpretations and for their capability to make a difference. In order to explain certain human activities by the means of the concept of social practice, both actors’ subconscious behavior (routines) and their conscious decision making (actions) must be accounted for. Figure 2: Actors-institutions loop INSTITUTIONS

Social Practice as Routines and Actions

Knowledge as Constraining and Enabling Orientation

ACTORS

Source: Own draft based on Giddens (1984) and Hodgson (1998)

Having said this, the interrelation of actors and institutions can be conceptualized as a dialectic “Actors-Institutions Loop” (s. Fig. 2): Institutions “feed” actors with a certain stock of rules, norms, values and meanings that together form actors’ knowledge. This knowledge serves as constraining and enabling orientation for actors who let it become manifest – be it in form of thoughts, words or conduct. This manifestation takes place as a continuous process of social practice that, in turn, leads to the ongoing reproduction and change of the prevalent institutional order of society. As such, institutions are medium and outcome of social practices while, at the same time, social practices are medium and outcome of institutions. With this heuristic figure in mind, we can now proceed to elaborate on the notion of “real markets”. In fact, markets are widely known as abstract mechanisms of demand and supply that lead to the formation of a specific price for a given quantity of commodities. What then are “real” markets? It was Maureen Mackintosh (1990) who coined this expression. As a researcher who dealt with food security she wondered why in so many studies on the issue concrete market places do not play any prominent role despite the fact that they are without doubt a “crucial link in the [food] system which determines who eats, how much and when” (Bernstein et al. 1990: 4). Against this background, she invited scholars to

28

Navigating Real Markets

place more emphasis on the very places where food is bought from producers, were it is exchanged and sold to rural or urban households and proposed to study them from an actors perspective which puts emphasis on the rationale of merchants, the institutional architecture of markets and their underlying power structures (Mackintosh 1990): “Perhaps the key point to emerge is that markets must not be conceptualized as the manifestation of impersonal laws of demand and supply [...They] are social institutions and processes that people enter from very different positions in structures of […] power. […H]ence no satisfactory economics of markets without a politics of markets” (Bernstein et al. 1990: 4). Invitation accepted. 2.4 REAL MARKETS 2.4.1 What are Markets? Markets are the central institutions of capitalist economies. The development of modern capitalism can be viewed as a process of the expansion of markets as mechanisms for the production and allocation of all kinds of consumer and investment goods and services (Beckert 2009). This makes it all the more surprising how restricted the attention has been to the study of markets in classical and neoclassical economic theory. Adam Smith (2005[1776]) used the metapher of an “invisible hand” to circumscribe the mostly unintended outcome of social welfare that he postulated to come about by the interaction of economic men who maximize their personal gain. Based on these considerations, economists developed an abstract understanding of markets; with Antoine Cournot (1897[1838]) for instance, they were seen as being constituted by price uniformity within a particular area. Real market places as localized social spaces that allow suppliers and customers to negotiate prices and that are shaped by particular institutions and power relations were only dealt with by a minority of institutional economists such as Thorsten Veblen, John Commons and Wesley Mitchell (Hodgson 1998, 2006). The overarching majority of economists was rather concerned with the much more general theory of rational choice. The focus was on the mathematical modelling of production and consumption functions that – under the precondition of perfect competition and complete information of “rational” self-interested actors – would lead to an equilibrium of market prices (Samuelson & Nordhaus 2005). In this context, economics became “the science which studies human behaviour as a relationship between ends and scarce means which have alternative uses” (Robbins 1945[1932]: 16) – an understanding that was claimed to be “applicable to all human behavior” (Becker 1976: 8). Consequently, as stated by John Lie (1997: 342), the market was left ignored as “the hollow core at the heart of economics”. In sociology, the founding fathers Max Weber, Èmile Durkheim and Georg Simmel showed a special interest in the exploration of the institutional preconditions of markets (Beckert 2007). For Weber (1978[1922]: 635) a market exisits “wherever there is competition, if only unilateral, for opportunities of exchange

Navigating Real Markets

29

among a plurality of potential parties [...that] ceases to exist with the act of exchanging the goods”. As the essential subject matter of a Sozialökonomik, Weber’s understanding was that of a “market struggle” in which prices turn out to be “the product of conflicts of interest and of compromises; they thus result from power constellations” (ibid.: 108). Such an understanding comes close to the contemporary perspective of economic sociology. However, it was especially Karl Marx who caused the subsequent concentration of sociologists on the organization of industrial work and labor markets. Marx was especially interested in elaborating the difference between barter and capitalist market exchange. For him, capitalism would have come into existence by the commodification of labor and the creation of labor markets in which “it is not the exchange of commodities which regulates the magnitude of their value; but, on the contrary, [...] it is the magnitude of their value which controls their exchange proportions” (Marx 2007[1867]: 73). Thus, in post-war sociology, priority was given to studying the “colonialization” of the lifeworld by the systems of the market economy and the bureaucratic state (Habermas 1981; 1985), while empirical research on real market places was widely left blank (Beckert 2007). In more recent years, though, new impetus have been given to the study of markets by economists and sociologists alike who both frame their new interest in the light of a general critique of neoclassical economics (Abolafia 1998; Callon 1998; Fligstein 1996, 2001; Fligstein & Dauter 2006; Lie 1997; Mackintosh 1990; Smelser & Swedberg 2005). In this new understanding, markets are no longer taken as abstract mechanisms of pricing, but as (partly localized, partially virtual) social fields that comprise a number of actors who organize the exchange of goods in repeating patterns and thereby attempt to secure the survival of their business organizations (Fligstein 1996, 2001; Fligstein & Dauter 2006). The analytical concept of social resilience can build on these considerations and be introduced for the empirical study of markets at the point where Fligstein (1996) yet speaks in a broad and rather undefined sense of the attempt of players to secure the “survival” of their companies. Given an unpredictable environment which is subject to constant change, the concept of social resilience offers to current economic sociology and relational economic geography the possibility to study this “survival” of business enterprises in a systematic manner. In this case, all those units of analysis come into play which have already been mentioned above, namely practices, networks, and arenas (cf. Tab. 1).

30

Navigating Real Markets

2.4.2 Markets and Practice Jens Beckert (2009) argues that the core issue of a social science of markets is to explain the order of markets. How is it possible that economic activities can be “coordinated” through markets despite the heterogeneous and partly antagonistic motives and interests of the participants? By coordination he means that actors succeed in aligning their actions in ways that allow for market exchange to take place. Such coordination is a precondition to the order of markets. Beckert’s point of departure is that markets are highly presuppositional arenas of social interaction in which actors are confronted with profound “coordination problems” (ibid.). He argues that one can distinguish three fundamental coordination problems which represent at the same time the central sources of uncertainty for market actors: 1) The value problem; 2) the problem of competition; and 3) the cooperation problem: 1) The value problem refers to the constitution of actor preferences. One crucial source of uncertainty confronting market actors derives from the difficulties of assessing the value of commodities given the multiplicity of goods and their complex quality properties. Only if sellers can reliably demonstrate the value of their goods and potential buyers are in a position to distinguish between the values of goods, will uncertainty be reduced and a disposition to buy arises. A theory explaining preferences for specific goods needs to take the role of producers and consumers into account. Producers attempt to create consumer attachment to their goods through their marketing strategies. At the same time, value attachments are created in the lifeworlds of consumers, and producers must react to new and often unpredictable trends that emerge. In this sense, the assignments of value are subject to a dynamic process of change and uncertainty is thus eliminated from the market only on a temporary basis (ibid.). 2) The problem of competition is related to one of the insights of neoclassical theory that while perfect markets are efficient, in market equilibrium no profit can be made. Suppliers therefore have an interest in establishing market structures that shield them from competitors, which allows them to reduce uncertainty with regard to their profit expectations. This, however, affects the interests of competitors, and the collective erection of market barriers can finally lead to prices that are higher than economically necessary. Firms alleviate some of the uncertainty created by competition themselves by product differentiation, first-mover advantages, reciprocal agreements, corruption, or the formation of cartels. In sum, however, the structuration of competition must be seen as a precarious compromise reflecting the inequalities of the power of actors in the market field (ibid.). 3) The cooperation problem arises from the business risks that market actors face because of their incomplete knowledge of the intentions of their exchange partners, the quality of the product they wish to purchase, and incalculable external factors of influence that might hinder the successful order or delivery of the product. Market relations are risky for instance when one exchange partner makes an advance payment without being sure whether the other party will actually fulfill the contractual obligations. The more difficult it is to specify the quality of a

Navigating Real Markets

31

product and the less able the buyer is to infer the seller’s actual intentions, the greater these risks are. Only if buyers are confident of not being exploited by their contract partners will they engage in market exchange. Creating this confidence is thus a fundamental precondition of stable markets (ibid.). While the notion of the order of markets refers for Beckert (ibid.) to the macrolevel result of the solution of the three identified coordination problems, the expectations and concrete institutional arrangements formed by actors through social practices of interaction constitute the building blocks of this order on the actor level. If the order of markets depends on the resolution of the three identified coordination problems, the task of market sociology is to study both the emergence (and change) of the social macrostructures through market actors’ interactions and the structuring of market actions by these macrostructures (ibid.). All three coordination problems mentioned by Beckert (ibid.) are vital for an understanding of the order of markets. Nonetheless, in this study the main emphasis is given to the problem of cooperation as it is most important when dealing with food security issues. In addition to that, a fourth coordination problem is discussed, which has been neglected in Beckert’s (ibid.) sociology of markets, but plays a special role for a geography of markets: The problem of place. 4) The problem of place refers to the fact that markets – also virtual ones – need some kind of physical backbone for their functioning – storage facilities, sales stalls, offices, or computers, which all need to be located somewhere. Markets sit in places. Similarly to markets as institutional facts (Searle 1995), so are market places as socio-spatial entities highly presuppositional. Specific locations must be designated, the access of traders to these places must be formalized, the maintenance of market facilities must be organized, and illegitimate access and usage must be hindered. As it will be shown in this study, markets in Dhaka are highly politicized, especially those markets that are located on public land. In Dhaka, none of these markets have been handed over to traders to fully selforganize the above mentioned issues. There is rather a mighty alliance of land brokers, ward-level bureaucrats, political activists and local criminals, who seek to control Dhaka’s public market places and to conserve the rents brought in from them. Such a politicized environment creates its own uncertainties that market actors need to deal with. The traders pay stand fees, maintain relationships to political activists, or unionize to express their interests. All these practices help reducing these spatialized kind of uncertainties. However, just as it is the case with the other coordination problems mentioned above, the reduction of place-related uncertainty is subject to a dynamic process of political change and can thus be eliminated only temporarily too. This understanding allows to define the notion of market governance in its broadest sense as the “processual ordering” (Strauss 1993) of markets. In other words, market governance is constituted by the continuous interplay of powerful social actors who create certain expectations and specific institutional arrangements in order to resolve the fundamental coordination problems of interaction and exchange, and thus (re)produce – either implicitly or explicitly – the social

32

Navigating Real Markets

order that constitutes markets as localized institutional facts (Beckert 2009; Searle 1995; Walgenbach & Meyer 2008). 2.4.3 Markets and Networks The study of networks puts emphasis on the cooperation problem that has been identified by Beckert (2009) and on the question of how market actors deal with particular business risks through the production of social capital. In the context of this study, risks are taken as all those actual and/or potential trends, shocks or seasonal developments that have a negative impact on market actors’ businesses and livelihoods, and which jeopardize their contribution to the functioning of the food system (Béné et al. 2012; Glavovic et al. 2003). Social capital is defined as actors’ legitimate access to others’ resources (Li 2007) and “describes relations of trust [...] and the role of networks (Adger 2003: 389). Social capital can be analyzed in regard to two different aspects, i.e. the quality of business relations that provides a measurement for the legitimacy of the access to others’ resources, and the overall quantity of business relations that provides a measurement for the potentially accessed resources. Accordingly, in the subsequent part it is distinguished between the content (Beckert 2007) of actors’ business relations and the structure (Granovetter 1985) of their business networks: Content refers to the underlying institutional arrangements that make up the network relations, i.e. the underlying expectations of the involved actors, their mutual knowledge of one another, and the meaning and feelings that make up the “glue” of these relations. From this perspective, social capital can be measured as tie strength, understood as preferably high degree of “trust that each participant has in the continuing contribution of others” (Portes 1998: 13). Structure, on the contrary, refers in this context to quantifiable aspects of networks, most prominently to the number of interconnected people and to the amount of goods and finances that are exchanged. In this sense, social capital can be measured as network diversity, understood as preferably dispersed access to numerous partners and various resources (Li 2007). Both aspects of social capital, trust and network diversity, are discussed in more detail below. With Susan Boon and John Holmes (1991), trust can be defined as an actor’s state of mind involving confident positive expectations about another’s motives and actions with respect to herself in situations entailing risk and uncertainty. Trust in persons is not simply a given institutional resource that can be easily mined by the members of a society. In a context of essential mutual uncertainty about the exchange partners’ intentions and actions, it needs to be actively produced. As Giddens (1990: 121; emphasis by Giddens) highlights, “personal trust becomes a project, to be ‘worked at’ by the parties involved, and demands the opening out of the individual to the other [...], trust has to be won”. In this sense, both trustors and trustees have to actively engage in a process of trust building (Möllering 2005).

Navigating Real Markets

33

This trust building can be described as a step-wise process: Initially, every form of trust is “in a certain sense blind trust” (Giddens 1990: 33), as it needs an initial “leap of faith” (Möllering 2005) in order to come into existence in the first place. This leap of faith enables actors to engage with one another despite the fact that institutional arrangements, that would guarantee both actors’ commitment and constrain their potential opportunistic behavior, are still there to be established in future. Once this initial step is taken, there awakes the potential for genuin trust to emerge as a continuous process of reciprocal reflexive monitoring. Thus “actors do not need to trust each other fully right from the beginning of a relationship, because they can engage experimentally in a kind of as-if trust which may gradually produce genuine trust” (ibid.: 19). Roy Lewicki and Barbara Bunker (1996) distinguish three types or “levels” of genuin trust: 1) The first type is calculus-based trust that rests on calculative reasoning about the other’s incentives to maintain the relationship and on the deterrents preventing her from breaking trust (ibid.). The calculation and the role of deterrence and effective punishment relates this kind of trust to the regulative pillar of institutions as formulated by Scott (2008). 2) The second type of Lewicki & Bunker (1996) is knowledge-based trust that is grounded in the parties’ sufficient knowing of one another which makes their actions predictable for one another. “Knowledge-based trust relies on information rather than deterrence. It develops over time, largely as a function of the parties having a history of interaction that allows them to develop a generalized expectancy that the other’s behaviour is predictable” (ibid.: 121). This “generalized” expectancy is grounded in Scott’s (2008) second pillar of institutions, since the involved persons usually align their action with norms and values rather than with explicit rules and regulations. 3) The third type in Lewicki & Bunker’s (ibid.: 122) typology is identification-based trust in which “the parties effectively understand and appreciate the other’s wants; this mutual understanding is developed to the point that each can effectively act for the other”. Such a deep mutual understanding refers to the sharing of worldviews, meanings and feelings in regard to particular situations that make them an aspect of the third pillar of institutions after Scott (2008). Trust building relates to those processes of human interaction that result in a general strengthening of relationships. This strength of social ties is most prominently expressed in terms of a growing felt social closeness or familiarity of both partners. Thus, at the beginning of trust building, one can speak of weak ties (Granovetter 1973). These weak ties can be characterized to be, in essence, unidimensional with rather clear and fixed role perceptions and with a general deficit of meaning, since mutual knowledge and an original relationship quality is still there to emerge. They can be distinguished from strong ties (Portes 1998) that come into existence not before the relationship is hardened over time by reciprocal activities, iterative interaction. These strong ties are “multiplex” (Boissevain 1974), since they offer the involved parties a surplus of joint experiences and in combination with that a multitude of social roles and available institutional scripts

34

Navigating Real Markets

to draw on. Strong ties are thus characterized by a genuin versatility that allows actors to pursue their interaction in various ways (Li 2007). Both strong and weak ties have certain advantages and disadvantages for the involved actors. The specific “strengh of weak ties” (Granovetter 1973) lies in constellations that reveal “structural holes” (Burt 1992) and allow those people advantages who act as brokers between otherwise unconnected persons or networks. By means of their position these brokers have access to information and resources that would otherwise be out of reach. Such assets play a crucial role for innovation processes and have the potential to foster the adaptability of business organizations. However, due to their impersonal character, weak relations offer market actors only limited scope for mutual commitment and support, that is their main disadvantage. Strong relations, in contrast, have the potential for market actors to result in mutual commitments to prolong their business relations (Coleman 1990; Li 2007). Alejandro Portes (1998) has called this fact “the strength of strong ties”. This mutual commitment allows exchange parties to achieve “economies of time” (Uzzi 1996, 1997), i.e. the transfer of fine-grained information that allows for joint problem-solving activities and that can help to foster business organizations’ capacities to cope with and adjust to particularly pressing adversities. At the same time, strong ties do bear the risk of negative consequences too, such as the active exclusion of outsiders, excess claims on group members, and tight restrictions on individuals’ freedoms (Bohle 2005). Based on these considerations, actors have basically two options for creating social capital. They establish strong ties to a selected number of well-known, trustworthy partners. Or they maintain weak ties to a favorably high number of network partners in order to achieve network diversity. Both strategies are reflected in particular network structures. Advocates of the “network approach” (cf. 2.3.1) have taken these structures as explanans for social practice, for the performance of business organizations and their probability to fail (cf. Uzzi 1996, 1997). Such an analytic alignment, however, neglects the underlying factors that made actors build these networks in the first place. The theoretical advantages and disadvantages of certain network constellations are meaningless unless one understands the reason why businessmen build their personal networks in the specific way they do. Furthermore, as Beckert rightly mentions, the pure focus on network structures “does away with a grounding of economic sociology [and economic geography] in action theory altogether [...]. Network structures, rather than social action, become the explanatory variable” (Beckert 2007: 9). By shifting the explanans from actors’ structural embeddedness back to actors’ institutional embeddedness, it is the aim of this study, to acknowledge the first one of Polanyi’s two original objectives to put emphasis on societies’ institutional frames for the explanation of economic outcomes (cf. 2.3.1).

Navigating Real Markets

35

2.4.4 Markets and Arenas The study of the urban arena puts the focus on the problem of place that has been identified above and on the question of how actors achieve and maintain legitimacy to access and use urban space for marketing. The notion of the arena serves to analyze the “positioning” (Giddens 1984) of Dhaka’s wholesalers in the urban society as a whole, which is strongly related to the role of the state in Bangladesh. Ortwin Renn (1992) introduced the notion of arena with reference to Herbert Kitschelt (1980), who originally coined the term. Renn (1992) stated that “arenas are neither geographical entities nor organizational systems. They describe the political actions of all social actors involved in a specific issue. [...] The arena concept attempts to explain the process of policy formulation and enforcement in a specific policy field” (ibid.: 181). In this study, arenas are understood as (partly localized, partially virtual) centers of debate in which competing and sometimes conflicting interest groups negotiate over issues and their interpretation (Hoffman 1999). In the context of this study, the center of debate is the legitimacy of actors to access and use public places for trading food. A standard definition of legitimacy has been formulated by Mark Suchman (1995). Accordingly, legitimacy is conceived as a generalized perception that the actions of a social entity are desirable, proper and appropriate within a given socially reproduced system of rules, norms, values and worldviews. The emphasis on the “generalized perception” (ibid.) highlights the fact that legitimacy is not understood as a possession of a single organization, but rather as a statement that is awarded a number organizations from a broader public audience. With Scott (2008) three types of legitimacy can be distinguished that are related to the above discussed three pillars of institutions: 1) The regulative pillar of legitimacy highlights the conformity or nonconformity of organizations with given sets of formal rules, such as governmental regulations and ordinances. Those organizations are perceived as legitimate that are legally established and operate in compliance with existing law. 2) The second pillar underlines the normative basis of legitimacy. Its focus is on the question whether organizations act in accordance with given moral concepts and value systems. Legal organizations can be denied legitimacy in the case that the public does not consider their activities as proper or desirable. 3) Finally, there is also a cultural-cognitive pillar of legitimacy. This mode of legitimacy mostly rests on routines and subconscious taken-for-granted understandings of the world. Thus, organizations’ legitimacy can also stem from conforming to a common definition of a situation or a shared frame of reference. Scott (ibid.) points to the fact that, in most cases, legitimacy is a matter of conflict and negotiation. More often than not these conflicts arise from a discrepancy between the “we consciousness” of the focal actors and the “they consciousness” which is ascribed to them on part of state authorities (Adelstein 2010). A business organization can be perceived as legitimate by reference to its legal character while, at the same time, it can be framed as inappropriate and illegitimate due to its operations that are identified as incompatible with moral standards. Sim-

36

Navigating Real Markets

ilarly, governmental regulations can be seen by some as legitimate due to their legal character but, at the same time, as illegitimate by others due to their underlying assumptions on reality or their defective implementation. In this study, the focus is on those struggles for legitimacy which ignite in regard to the appropriation and utilization of market places that are located in the highly politicized public space of the megacity. In order to understand these conflicts, it is helpful to refer to the informality of current urbanization processes in the Global South. Since its “discovery” by Keith Hart and his colleague within the frame of the employment mission of the International Labour Organization in the 1970s (ILO 1972), the notion of informality has encompassed an increasing number of meanings and referred to highly heterogeneous phenomena and measurement methods (McFarlane & Waibel 2012). Traditionally, the term was used in a categorical way in order to identify the informal sector or the informal economy (Castells & Portes 1989; de Soto 1989, 2000; Hart 1973; ILO 1972; Moser 1978). The notion enabled scholars and practitioners to subsume the highly heterogeneous urban economies of day laborer, petty hawkers, waste-pickers, backyard artisans, or domestic service providers – just to name a view – under one single category that was defined by an ever growing list of charactersitics, ranging from selfemployment and easy market entry to lack of authorization, tax evasion and illegality. Despite all differences between the scholars, in all these studies a rather clear-cut division line was drawn between the formal and the informal. From the mid-2000s onwards, informality has been re-framed by Nezar AlSayyad and Ananya Roy (2004) as a particular “organizing urban logic”, as a system of rules, norms, values and worldviews that governs the process of urbanization itself. Roy (2005) points to the fact that, in recent years, informal housing and land markets have remained not just the domain of the poor but have become important for the middle class and the elite of cities in the Global South as well. She provides examples where squatter settlements formed through land invasion and self-help housing exist alongside high-level residents formed through legal ownership and market transaction but in violation of land use regulations. She (ibid.) holds that the divide here is not between formality and informality, since both forms of housing are informal. The differentiation is rather within informality, which ranges from the urban poor to the urban elite and embodies very different concretizations of legitimacy. In many parts of the world, informal urbanization is found in city centers, where the relocation of squatters to statesponsored resettlement sites on the urban periphery has become a widespread phenomenon (ibid.). But it is also found in the metropolitan fringe, where there are gated communities. Unlike squatter settlements, such forms of high-end informality usually enjoy premium infrastructure and guaranteed security of tenure, and in many cases they are even promoted and encouraged by the state. In this context, AlSayyad & Roy (2004: 5) argue that informality should not be used as a fixed labelling of social entities or economic sectors, but rather as a marker for the specific mode of urban governance which has become prevalent in the Global South. This mode of governance is made possible through the particular regulatory logics of land that exist in and around the cities. “This in turn means that informality must be understood

Navigating Real Markets

37

not as the object of state regulation but rather as produced by the state itself” (Roy 2005: 149). The planning and legal apparatus of the state has the power to determine what is informal and what is not, and which forms of informality will thrive and which will be destroyed. State power in this sense means the capacity to define and cancel categories of legitimacy and illegitimacy (ibid.). From this perspective, it becomes apparent that the legalization of informal property systems is not simply a bureaucratic or technical act, but rather a complex political struggle. In this study, informality refers to food wholesale traders’ ongoing negotiations with the state apparatus to appropriate and utilize public places for their marketing activities and – arising from that – to the highly volatile tenure security that is brought about by local elites’ “politics of the back room”. With this term I refer to the work of Erwing Goffman (1959) who introduced the general distinction between society’s “front” and “back regions”. Being inspired by this idea, I take the front region of the state to contain all those regulatory bodies from parlaments, courts and the bureaucracy to acts, ordinances and guidelines that together constitute the visible and accessable sphere of politics. In contrast, the back region of the state refers to arcane governance modes and to those decisions that are made “behind the scene”. Even though it is extremely difficult – if not impossible – to directly study this politics of the back region, at least its outcomes can be observed, be it in the form of providing informal markets a stay order or in the form of demolishing them. By expanding the concept of embeddedness through the dimensions of place, legitimacy, the urban arena and informality, it is the aim of this study, to also acknowledge the second one of Polanyi’s two original objectives to critically analyze the political dimension of economies and to show ways of social reform (cf. 2.3.1). 2.5 RÉSUMÉ In the context of the present double exposures of global environmental change and globalization (Leichenko & O’Brian 2008), scientists, policy-makers and nongovernmental organizations call for resilient food systems that are able to overcome future crises and enhance global food security without further compromising environmental and social welfare outcomes (Ericksen et al. 2010). Like the notion of sustainability, resilience invokes a prospective and positive view, even though it does only exist in a context of potentially harmful change (Walker & Salt 2006). But what does it exactly mean for a food system to be resilient? In this chapter, I have developed a social science based understanding of resilience. Social resilience shares the key principles of the general resilience concept, which is rooted in social-ecological systems thinking, specifically in its focus on systems’ persistability, adaptability, and transformability. However, it departs from the general resilience discourse by adopting an actor-oriented perspective. Based on a review of relevant literature, I have identified three important dimensions of social resilience, which together take into account social actors’ capacities to cope with and to overcome all kinds of immediate adversities (coping capacities), their ca-

38

Navigating Real Markets

pacities to learn from past experiences and adjust themselves to pressing new challenges in the future (adaptive capacities), and their capacities to craft institutions that foster individual welfare and sustainable societal robustness in the event of present and future crises (transformative capacities). While the general discourse speaks of the resilience of food systems, with my approach I place emphasis on the people who together make up the food system through their daily activities, and on their resilience in food systems. Such an understanding offers the opportunity for analyzing the underlying dialectics of structure and agency at play in today’s food systems: the economic, political, social and ecological conditions of food systems structure people’s resilience and their food security, while people make urban food security and the resilience of food systems through their everyday practices and interactions. The classical topics of resilience research are located in the areas of disaster control, resource management and climate change adaptation. I have proposed instead to transfer the concept of social resilience into the economic sphere and to make it fruitful for the empirical study of markets. In fact, markets are classically understood as abstract mechanisms of demand and supply that lead to the formation of a specific price for a given quantity of commodities. However, in more recent years, a new understanding of markets has been developed (Abolafia 1998; Callon 1998; Fligstein 1996, 2001; Fligstein & Dauter 2006; Lie 1997; Mackintosh 1990; Smelser & Swedberg 2005). In this new perspective, markets are no longer taken as abstract mechanisms of pricing, but as (partly localized, partially virtual) social fields that comprise a number of actors who organize the exchange of goods in repeating patterns and thereby attempt to secure the survival of their business organizations (Fligstein 1996, 2001; Fligstein & Dauter 2006). I have argued that the analytical concept of social resilience can be used fruitfully for the empirical study of markets at the point where Fligstein (1996) yet speaks in a broad and rather undefined sense of the attempt of players to secure the “survival” of their companies. Given an unpredictable environment which is subject to constant change, the concept of social resilience offers to current economic sociology and economic geography the possibility to study the survival of business enterprises in a systematic manner. As the focus of this study is on markets, I will speak in the rest of this study not anymore of the social resilience, but of the economic resilience of market actors. When studying markets, one needs to understand that these institutions are highly presuppositional arenas of social interaction in which actors are confronted with profound coordination problems. According to Beckert (2009), market actors need to reduce at least three types of uncertainty in order to succeed in aligning their actions in ways that allow for market exchange to take place: The value problem, which refers to difficulties of assessing the value of commodities given the multiplicity of goods and their complex quality properties; the problem of competition, which is related to the fact that suppliers need to establish structures that shield them from competitors in order to make profit; and the cooperation problem, that arises from the business risks that market actors face because of their incomplete knowledge of the intentions of their exchange partners and incal-

Navigating Real Markets

39

culable external factors of influence that might hinder the successful order or delivery of the product. When transforming this sociology of markets into a geography of markets, a fourth coordination problem needs to be taken into account: The problem of place, which refers to the fact that markets – also virtual ones – need some sort of physical backbone for their functioning, such as storage facilities, sales stalls, offices, or computers, which all need to be located somewhere. Similarly to markets as institutional facts (Searle 1995), so are market places as sociospatial entities highly presuppositional. Specific locations must be designated, the access of traders to these places must be formalized, the maintenance of market facilities must be organized, and illegitime access and usage must be hindered. If the order of markets depends on the resolution of the four identified coordination problems, then the task of market geography, I suggest, is to study both the emergence (and change) of the social macrostructures through market actors’ interactions and the structuring of market actions by these macrostructures (cf. Beckert 2009). The main emphasis in this study will be given to the problem of cooperation and to the problem of place, which will be analysed be referring to the embeddedness of market actors’ in networks and in the urban arena (Fig. 3). Figure 3: Geography of markets – Conceptual framework

Source: Own draft based on Etzold et al. (2009)

3. METHODOLOGY

“[I]n using deduction the conclusion must be constantly checked with facts by observation and premises revised accordingly, while empirical laws [regularities] resulting from induction must in turn be shown to follow from the general principles of the science before they can be credited with much significance [...W]e see that there is little divergence left between the two methods.” Frank Knight (2006[1921]: 7)

3.1 TRIANGULATION For this study I applied a combination of both qualitative and quantitative research methods. I deployed them in the course of an overall eleven-month fieldwork which I undertook from 6 August to 26 September 2007, from 25 October 2007 to 2 April 2008, and from 9 January to 24 April 2009. In this chapter, all these research methods are explained. Still, before I do so, I want to highlight the underlying methodological principle that drives this study: the triangulation. Originally, the term of triangulation is borrowed from navigation and military strategy where it addresses the technique of determining the location of an object’s exact position by measuring angles to it from at least two perspectives (Blaikie 1991). In social science, Norman Denzin (1978: 291) defined the notion of triangulation as “the combination of methodologies in the study of the same phenomenon”. In this latter perspective, triangulation means to take up at least two perspectives for answering one specific research question. Denzin (ibid.) distinguished four main types of triangulation: 1) The data triangulation rests on the idea that the robustness of data can vary based on the time in which the data was collected, the people involved in the data collection process, and the specific place from which the data was collected. Against this background, data triangulation refers to the use of multiple data sources for validation purposes (Hussein 2009). 2) The investigator triangulation is described as the use of more than one researcher in the process of data collection, analysis and interpretation. It involves multiple surveyers, interviewers, or data analysts in the same study for confirmation purposes (ibid.). 3) The theoretical triangulation can be defined as the use of multiple competing theories in the same study for the purpose of developing various perspectives on a research problem, for formulating hypotheses and for supporting or refuting them in order to get a better understanding of the phenomenon (ibid.).

Methodology

41

4) The methodological triangulation, eventually, is understood as the use of more than one method in studying one particular phenomenon. This type of triangulation is the most commonly used one in contempory social science (Flick 2011). Usually two types are distinguished, i.e. the within-method triangulation and the between-method triangulation (ibid.). The first one implies that multiple complementary methods are used in data collection and analysis which all belong to one single research paradim (quantitative or qualitative research). The second one involves the combination of both quantitative and qualitative methods in studying one single phenomenon. In this study, I follow the idea of the betweenmethod triangulation. In fact, using both quantitative and qualitative methods in the same study has resulted in critique on part of researchers arguing that the two paradigms differ not only epistemologically, but also in terms of their ontological grounding (Hossein 2009). With the quote of Frank Knight (2006[1921]) above, I would like to express my doubts about the widespread conviction that deductive and inductive research designs as well as quantitative and qualitative approaches cannot be combined. From my personal point of view, two lines of argumentation are pivotal in this regard. One is born out of a pragmatic understanding of science and is related to the epistemological dimension of triangulation; the other one is related to the ontological dimension and results from the perspective of critical realism which informs this study. The first argumentation goes like this: Generally, both paradigms have their strengths and weaknesses. Semi-structured interviews, for instance, are suitable to capture a person’s individual motives and situational interpretations without necessarily reducing the complexity and messiness of everyday life (Mattissek et al. 2013). At the same time, however, they are limited when it comes to terms with the representativity of findings in a numerical sense. A standardized questionnaire, in contrast, offers the opportunity to survey the societal circumstances – e.g. income, marital status, level of education etc. – of large quantities of people and to provide a picture representative of the total population (ibid.). And yet, there is only limited scope to gain information about a layperson’s theories on a specific subject matter or about the rationale behind her daily practices. Since all methods have got their particular strengths and blind spots, the point of methodological triangulation is to treat both methods as being of equal avail for gaining insights into a specific subject matter of interest. Taken together they offer the chance to minimize the drawbacks of each individual method and to get deeper insights into a phenomenon by applying multiple perspectives. From this vantage point, quantitative and qualitative methods are seen as complementary rather than as mutualy exclusive (Flick 2011; Hossein 2009). The second argument is related to critical realism, which has been developed as an ontological middle ground between the two extremes of, on the one hand, naïve realism, and, on the other hand, radical constructivism. In opposition to radical constructivism, Searle (1995) defends the idea that there is a reality independent of us and that statements can be claimed to be true if they correspond to (either natural or social) facts. As such, one cannot argue that one representation of a

42

Methodology

phenomenon is as good as the other one. A true statement can be distinguished from a false one based on its inherent coherence to certain aspects of reality, since “statements are true in virtue of conditions in the world that are not part of the statement” (ibid.: 219). At the same time, Searle (ibid.) takes a stand against naïve realism in the sense that human representation of reality can never be the same as reality itself. “Every representation has an aspectual shape. It represents its target under certain aspects and not others. In short, it is only from a point of view that we represent reality. […] Strictly speaking, there is an indefinitely large number of different points of view, different aspects, and different conceptual systems under which anything can be represented. If that is right […], then it will be impossible to get the coincidence between truth and reality after which so many traditional philosophers seem to hanker” (ibid.: 176). From this argument follows that any representation is a social construct and it remains to be so, whether this representation is true or false. While one sort of (natural or social) science puts emphasis on facts, another one focusses on the representation of facts as social constructs. Both alignments are legitimate and necessary. Searle (ibid.) underlines that the opposition made between realism and constructivism, and between quantitative and qualitative research, stems from the traditional distinction of body and mind, and between nature and culture (cf. 2.3.3). Against such forms of dichotomy he claimes that both are related, since culture is only a form that nature takes. There is no radical break that departs the two, but an interconnecting sphere which he addresses by the terms of consciousness and intentionality. From such a standpoint, one needs to acknowledge that both physical and institutional facts can principally be studied with the same methods, since both are part of a continuum of real phenomena. Quantitative methods are therefore suitable for the study of both nature and society. Nevertheless, even though there is a continuum from the chemistry of our brains to the content of our thoughts, mental states are distinguished from physical states in that they involve consciousness (if only potentially, as it is the case with our daily routines). Similarly, even though there is a continuum of collective behavior from animal herds to national governments, institutional facts have a special feature which is its symbolism (cf. 2.3.2). In other words, institutional facts exhibit peculiarities that distinguish them from physical facts and make them special. Qualitative research methods have been developed in order to account for this speciality. As such, they are necessary for both the study of the social representation of physical reality and for the study of the construction of social reality. From an ontological perspective there is no dispute between the two research paradigms.

Methodology

43

3.2 QUANTITATIVE METHODS Three types of quantitative methods were conducted. First, the entire area of the Dhaka City Corporation (DCC) was studied for locating all food wholesale markets of the city. Second, a panel survey was organized in order to achieve a more exact picture of the sales volumes of rice and fish wholesale traders in Dhaka. Third, a survey was undertaken with the aim to test the representativeness of selected findings of the ethnographic field work, and to reveal the structure of the wholesale traders’ business networks. 3.2.1 Locating Food Markets As a prerequisite of this study I undertook a survey in order get an overall picture of the number and location of all food wholesale markets which are located within the boundaries of the DCC. Despite the use of a geographical information system (GIS) and global positioning system (GPS) devices, the hardly planned and rapidly growing city made the locating of these markets a challenging task. At the beginning, I collected available data about markets in Dhaka from DCC (visit on 28 March 2008) and the Department of Agricultural Marketing, Ministry of Agriculture (visit on 5 February 2008). The list of the ministry, unfortunately out of date, contained plenty of information about markets but none about their locations. By contrast, the list of the DCC provided locations of markets, but no information beyond their bare names. Back in Germany I forwarded all information to my student assistants of the Universities of Bonn and Heidelberg who entered them into an ArcGIS database. They read the locations of markets and mosques, the road network, waterbodies and the city ward boundaries into a satellite image and prepared a single map for each and every one of the 90 city wards. Afterwards, they sent these ward maps to Dhaka, where they were used by my student assistants of Dhaka University and Bangladesh University of Engineering and Technology to adjust the preliminary data to reality. Some of the landmarks that the city administration had put into the market list turned out to indicate shopping centers, retail kitchen markets, or conglomerates of several markets at one location. Therefore the students had to check every single location in order to distinguish wholesale from retail markets as well as food from non-food markets. In addition to that, they asked at each location for further food wholesale markets in order to extend our list and update our data base. At each food wholesale market, the students took a GPS waypoint and conducted an interview with one of the representatives of the local market association wherever applicable or with an experienced businessman at place (N=87). The interview was conducted by means of a widely standardized two pages questionnaire that comprised a number of 42 questions. After digitalization, I analyzed the data by means of the computer program Microsoft Excel and sent my results to the technical cartography of the University of Bonn. The cartographers and I prepared in close collaboration four

44

Methodology

maps (Map 1, 2, 5, 8) out of the most important findings of this survey by means of the computer program Adobe Illustrator. 3.2.2 Recording Sales Volumes A preliminary survey in 2008 and 2009 served the general aim to gain more precise information about the daily quantities of rice and fish that are sold in Dhaka’s food markets. For this purpose, 12 rice markets (out of a total number of 24) and 9 fish markets (out of a total number of 13) were selected for a panel survey. This panel survey comprised two countings at each market. For rice, the first counting was conducted in February and March 2008, and the second one exactly one year later in February and March 2009 in order to see the difference between a crisis (2008) and a post-crisis year (2009). For fish, the countings were done in February and March 2009 and half a year later in July and August 2009 in order to see saisonal differences. At each market, the number and names of all available wholesale traders were listed. In addition to that, each wholesaler was asked to estimate his average daily turnover and to provide basic information about the organization of his supplies. All gained data was compiled into a Microsoft Excel datasheet and served to countercheck the results of the above mentioned food market mapping. The number and names of the wholesalers served as prerequiste for the large-scale survey that was conducted afterwards. 3.2.3 Surveying Wholesale Traders A comprehensive standardized survey was conducted to complete the collection of quantitative data in Dhaka. At that time, a total sum of 775 rice wholesalers ran their business at one of the 24 markets of the city. From them, 206 merchants were interviewed at 12 selected markets. From the total sum of 898 wholesalers who ran their business at one of the 13 markets for fresh fish at that time, a number of 242 merchants were interviewed at 9 different markets (s. Tab. 2 and 3). Table 2: Rice market survey 2009–2010 Market size

Tot. no. of markets

No. of markets surveyed

Tot. no. of wholesale traders

Large markets* Medium markets** Small markets*** Total

2 6 16 24

2 (100%) 3 (50.0%) 7 (43.8%) 12 (50.0%)

354 255 166 775

No. of wholesale traders interviewed 102 (28.8%) 56 (22.0%) 48 (28.9%) 206 (26.6%) Source: Own draft

Notes: * Large markets are Badamtuli and Krishi Market ** Medium markets are Kochukhet, BDR Wholesale Market, Mirpur 1, Jatrabari, Chowdhuripara and Malibag

45

Methodology *** Small markets are Govt. Hawkers Market Tejgaon, Uttar Badda, Lalbag Fort Mor Bazar, Kawran Bazar, Fakirbari Bazar, Mugdapara Boro Bazar, Mirpur 10, Sipahibag Hotel Bazar, Madertek Bazar, Nawabganj, West Nakhal Para, Boro Bazar, Mirpur 2, Mohakali Puran Bazar, Modda Badda Bazar and Narda Bazar

The previously organized mapping of Dhaka’s food markets and the conducted panel surveys provided important information that was used to classify Dhaka’s rice and fish markets according to size. Based on the number of wholesale traders per market, three categories were built, i.e. large, medium, and small markets. A significant number of markets and merchants were selected from each category. The final ratio of interviewed persons ranged from 22.0 to 28.9 percent of all traders of the respective market size category. In total, 26.6 percent of all rice wholesalers and 27.0 percent of all fish wholesalers within the area of Dhaka City were surveyed (s. Tab. 2 and 3). Table 3: Fish market survey 2009–2010 Market size

Tot. no. of markets

No. of markets surveyed

Tot. no. of wholesale traders

Large markets* Medium markets** Small markets*** Total

2 5 6 13

2 (100%) 4 (80.0%) 3 (50.0%) 9 (69.2%)

523 283 92 898

No. of wholesale traders interviewed 147 (28.1%) 75 (26.5%) 20 (21.7%) 242 (27.0%) Source: Own draft

Notes: * Large markets are Kawran Bazar and Jatrabari ** Medium markets are Merul Badda, Sultan Shah, Swarighat, Mugdapara and New Market *** Small markets are Tarun Mahmud Shah Market, Kochukhet, Kaptan Bazar, Azampur Rail Gate Market, Dipjol Bazar and Mirpur 1

At each market, a random number of wholesalers was selected for the survey. For rice, the markets were inspected and an exact route was defined that the group of interviewers would take. A list of random numbers was computed that would guarantee a non-biased selection of wholesalers. However, in the case of large scale markets, this approach was not practical due to the bare size of the market. Thus, prior to the walks through Badamtuli and Krishi Market, a dice was thrown in order to select the first, second or third wholesale store. Afterwards, every third wholesaler was selected for the survey. For fish, the survey was more difficult to realize. A single market day usually lasts for no longer than four hours. During that time, the markets are frequented by hundreds of people. Due to this heavy rush, the merchants cannot offer a single minute for being interviewed at that time. For this reason, the survey was organized for the time after the markets had closed. In between 10 and 11 a.m., fish wholesalers usually go back to their offices for taking rest and for doing the accountancy. These offices are scattered around the market place and difficult to be located. For this reason, we used the cell phone numbers of the wholesalers which

46

Methodology

had already been recorded in the course of the panel survey. These numbers enabled us to randomly select every third available number and to arrange a meeting for the interview. As the resulting ratio of interviewed fish wholesalers shows, this procedure worked quite well. Nonetheless, I do not want to keep quiet about that fact that this procedure was highly time consuming. While the entire procedure of sampling was tested during fieldwork in 2009, the survey was realized from November 2009 until January 2010 under the supervision of my permanently employed research assistant Shahidul Haque. I prepared an eleven-pages questionnaire and forwarded it to him. Prior to conducting the interviews, the entire questionnaire was gone through with all interviewers. By means of test interviewing, it was assured that the surveyors had sufficient knowledge of the questions to be asked and of the underlying research objectives. At that, the supervisor was present during the entire process of data collection. Finally, two assistants bore responsibility for data entry in Bangladesh in order to reduce transcription errors. The data was compiled in an SPSS database and sent to me. The entire analysis of the data was done by myself. 3.3 QUALITATIVE METHODS Three types of qualitative methods were conducted. First, selected locations important for Dhaka’s food supply were visited and observed in order to understand the basic functioning mechanisms of the food system. Second, guided interviews were used to study the daily business practices of Dhaka’s food wholesale traders. Third, participatory techniques were applied in order to delve into particular aspects of the traders’ daily business which came up in the course of the prior qualitative research. 3.3.1 Observing the Food System Observations were most decisive for understanding the extent and functioning of Dhaka’s food system. Of special value in this regard were the numerous field trips that I undertook. Excursions beyond the boundaries of Dhaka City helped me to draw a sort of mental map of the megacity and its environ. Food enters Dhaka through a small number of physical entrypoints on the arterial roads. I visited all of them, namely Tongi Bridge in the North, Gabtoli Bridge in the Northwest, 2nd Buriganga Bridge and Bangladesh-China Friendship Bridge in the South, and Kanchpur Bridge in the Southeast. On the outer fringe, the Jamuna Bridge and Archia Ferry Ghat in the Northwest, Mawa Ferry Ghat and the two Meghna Bridges in the South and Southeast, as well as the Bairabh Bridge crossing Meghna in the Northeast are the main entry points that I visited as well. I investigated the city’s railway system at Kamalapur train station and its river ports at Sadarghat and Narayanganj. For understanding the physical handling of food imports, I visited the landport of Hili, the freight depot of Sirajganj, and the seaports of Chittagong and Mongla (Khulna). At each of these loading points, responsible officers

Methodology

47

provided me helpful explanations. Staying in major rice and fish supplying regions, i.e. Rajshahi, Bogra, Dinajpur, Kushtia, Jessore, Khulna and Chittagong (s. Map 3 and 4), provided me additional, much needed information. In Dhaka City I undertook a detailed mapping of a selected number of wholesale markets with the aim to later assist the reader in comprehending the physical shape and the functioning of these structures. In the course of my fieldwork, I selected four market for this mapping, i.e. the rice markets of Badamtuli and Krishi Market, and the fish markets of Kawran Bazar Five-Star Market and Merul Badda Market (cf. Map 1). These markets were selected for two reasons: Badamtuli and Kawran Bazar Five-Star Market are the city’s largest markets for rice and fish in terms of daily turnover and the number of involved wholesalers. At that, both markets show a long tradition in wholesaling which had made them appropriate to be intensely studied by means of observations and qualitative interviews. Krishi Market and Merul Badda Market are markets of considerable size as well. However, the major criterion for selecting these two markets was the fact that both cases provide illustrative examples of the role that informality plays for food traders in Dhaka (cf. 2.4.4). I organized all mappings in collaboration with my research assistant Shahidul Haque. The first step of the mapping was to inspect the market and to draw its exact boundaries. The second step was to enter building structures, and the number and locations of wholesale stores. The third and final step was to consult representatives of the local market association in order to elucidate the functional structures of the market and – in the case of Krishi Market and Merul Badda Market – those areas which are subject to particular conflicts. Back in Germany, the handwritten sketches were adjusted to satellite images and served the cartographers of the University of Bonn to prepare four maps (Map 9, 10, 11, 12) by means of the computer program Adobe Illustrator. 3.3.2 Interviewing Wholesale Traders Guided interviews were most decisive for understanding the general functioning of Dhaka’s food system and the organization of the wholesalers’ daily business. I interviewed three different groups of persons, i.e. experts, wholesale traders and other food system actors (s. Tab. 4). In sum, I conducted 28 expert interviews in the course of my fieldwork. During my first field stay I consulted experts with various academic backgrounds and from different organizations to adjust my research focus and discuss questions of feasibility. One of the results of these consultations was my decision to choose rice and fish as commodities to study, as it created the opportunity to compare non-perishable and perishable goods and the distinct organizational set-up of respective markets. Furthermore, I interviewed a number of public officers in order to access official data on the subject matter and to reveal how governmental bodies perceive the food wholesale business. In order to broaden my perspective, I also consulted experts who belong to the so-called civil society, i.e. executive

48

Methodology

employees of local non-governmental organizations who are concerned with the city of Dhaka and the issues of food supply and food security. Table 4: List of qualitative interviews Type of interview

Type of interview partner

Expert interviews

Academic experts Governmental experts Civil society experts Rice and fish wholesale traders Other food system actors

Guided interviews Total

No. of interview partners 7 12 9 42 18 88 Source: Own draf

I interviewed a total number of 42 rice and fish wholesale traders and 18 other food system actors. Given the focus of this study, I treated wholesalers as a single group of interview partners, while I defined the group of food system actors to consist of all other actors relevant for Dhaka’s supply of rice and fish, ranging from peasants and pond owners, over millers and rural intermediaries, to urban retailers. The interviews with this group of persons served me to draw a complete picture of the entire food system “from field to fork” and “from pond to plate”. Table 5: Codes for the analysis of qualitative data Wholesale markets - business relations - composition - customer - daily routine - expenses (fish) - expenses (rice) - finance - fish types Food system - actors (fish) - actors (rice) - export - food prices - import - processing (fish) - processing (rice) - production (fish) - production (rice) - risk

- income & profit - land tenure - negotiations - official documents - personal background - prices (fish) - prices (rice) - rice types

- risk - storage - supplier - supply regions - sweeping rice - transportation - turnover - worker

Megacity - consumption culture - demand - distribution - food markets - risk - supply - urbanization

Urban governance - caretaker government - market association - market organization - public distribution system - regularization - syndication - case studies

Source: Own draft

The interviews with wholesalers allowed me to gather information on the merchants’ personal backgrounds, their business organizations, and their daily busi-

Methodology

49

ness practices which all together helped me to understand how food wholesaling is currently organized in Dhaka. As a prerequisite of this method I identified areas of interest and compiled a list of questions that served me as general guideline for the interview’s direction. The duration of each interview showed to be highly variable, lasting from about twenty minutes up to several hours, depending on the willingness and time of the respondents. Some of the interviewed persons showed to be especially supportive and became to me what ethnographers call key informants. They were interviewed several times. Due to the wide range of topics that I adressed, the interviews differed also in terms of content. Some of them were conducted with a clear focus on specific questions, such as the organization of sales or restocking, while others were rather narrative in their nature and provided information on the traders’ personal backgrounds. Due to the sheer number of interview partners I decided to record all conversations by means of handwritten notes. Of course, I was aware of the fact that this method involves a higher degree of fuzziness in the data, as there is no chance to rewind the answers given by the interview partners. Nonetheless, I deemed this technique suitable as the process of simultaneous interpretation allowed me to note down central sentences word by word. In case of longer passages, I tried to note the central statements of my interviewees and rearranged them afterwards in direct speech as close as possible to the originally used words. This was seen to be legitimate, since all my interview partners provided their answers in Bangla, which was afterwards translated into English, anyways. Back in Germany, I digitalized all my notes of the interviews, sorted, indexed and commented them, and analyzed them by means of the computer program Atlas.ti. Table 5 shows the list of codes that I used for analyzing the qualitative data. 3.3.3 Using Participatory Tools In the course of the fieldwork, two issues turned out to be of special interest for this study. The first one emerged as the question of how the personal networks of Dhaka’s rice and fish wholesalers are structured (cf. 2.4.3). This question involved to figure out the number and type of persons to whom wholesalers are connected, the merchant’s underlying motivations for maintaining these relations, and the way of how these relations are governed. For answering these questions from a qualitative point of view, I studied the networks of four businessmen by jointly drawing so-called Venn-diagrams (Kumar 2002). This participatory technique involved the naming of relevant persons, the placing of these persons on a scale of social distance, and the evaluating of them in regard to their relevance for the interviewee’s business performance.9 The entire procedure of drawing a Venndiagram was a collaborative act that took in between two and four hours. It provided both the interviewer and the interviewee plenty of time for explanations, 9

More details can be found below (cf. 6.1).

50

Methodology

inquiries and discussions. The major aim of this technique was to understand the institutional embeddedness of Dhaka’s wholesale traders (cf. 2.4.1). The second field of interest emerged around the question of business risks. What kind of risks are Dhaka’s wholesalers exposed to and what efforts do they make to cope with and adjust to these adversities? These questions involved to reveal the types of risk that the merchants are exposed to, the frequency and severity of these risks, and the merchants strategies and tactics to deal with them. The underlying aim was to understand those factors that are most pressing for the wholesalers from their own perspective and that most severely threaten the survival of their business organizations (cf. 2.4.1). For debating these questions in a rather open and self-determined way, I organized focus group discussions with market associations of the above mentioned four markets of Badamtuli, Krishi Market, Kawran Bazar Five-Star Fish Market and Merul Badda Market (cf. Map 1). Each of these discussions took in between two to four hours. Table 6 provides a summary of the participatory tools that I applied in this study. Map 1 in the annex gives a summary of all my research sites in Dhaka City as well as of the most relevant methods that I applied in the course of this study. Table 6: List of participatory tools Type of interview

Type of interview partner

Focus group discussion Venn-diagram Total

Rice and Fish Wholesalers Rice and Fish Wholesalers

No. of interview partners 4 4 8 Source: Own draft

3.4 LIMITATIONS OF THE STUDY Since every representation of reality necessarily has an aspectual shape, there are always certain features of a phenomenon that are left aside. Similarly, in this study, some of those issues are left unconsidered that others might deem fundamental. This is not what I consider a limitation. This rather comes about by the very nature of science as it always involves the need to decide where to put the focus and where not. Under the headline of limitations, I would rather like to address the topic of language and its role in this study. To begin with I shall mention that I completed two Bangla courses at a private language school during my field stay, one from November 2007 to January 2008 and one from January to March 2009. The learnt language skills helped me a lot as they enabled me to move freely through the city and to get to know at least pieces of the rich Bangladeshi culture. They were also helpful for observing Dhaka’s food markets, for gaining my interview partners’ confidence, and for conducting countless informal interviews in the course of this study. Nonetheless, it would be false to keep quiet about the fact that my knowledge of Bangla remained limited. In fact, my skills never allowed me to conduct in-depth interviews on my own. This is the reason

Methodology

51

why I was almost all the time during my data collection accompanied by my research assistant and interpreter. My language skills allowed me to reduce the social distance between myself and my interviewees to some extent. I was able to intervene in particular interview situations, to point out a word or a sentence that my interpreter had kept quiet about, or to make myself comprehensible by using both English and Bangla terms. Nevertheless, all the information that I gained are basically the result of a “double translation” (Bertuzzo 2009) from my ideas into the words of my interpreter, and from my interpreter’s sentences into the concepts of the interviewee. There was no scope for me to finally clear this condition.

4. THE FOOD SYSTEM OF DHAKA

“Machhe Bhate Bengali” – “People of Bengal are made of fish and rice” Bengali proverb

In this section I am going to discuss the current performance of Dhaka’s food system from a food security perspective. In the first part, I will retrace Dhaka’s history and its development from city to megacity, which will be done in some detail, since exact figures on the city’s demography and extension are necessary for later calculations. In the second and third part, I will take the examples of rice and fish provision in order to discuss present-day functionalities and challenges of the megaurban food system. I will present national production figures, assess available amounts of food in Dhaka, show where these amounts come from, and discuss the access to food on the basis of seasonally fluctuating prices. My principal research questions in this section are: How productive is Bangladesh’s rice and fishery economy? How is Dhaka’s food system spatially organized? And how can we evaluate the urban food system from a food security perspective? 4.1 THE MEGACITY OF DHAKA 4.1.1 History of Dhaka Dhaka’s historical origin is unknown, but due to its strategic location it was a place of some importance already during the period of the Delhi Sultanate that lasted from the early 13th until the middle of the 16th century. Officially, Dhaka was founded during the Mughal period, when the governor (subahdar) Islam Khan Chishti transferred the capital of the province of Bangalah (subah-ibangalah) from Rajmahal to Dhaka in 1610. Until 1639 and from 1659, Dhaka served as the provincial capital until 1717, when the subahdar Murshid Quli Khan shifted it to Murshidabad (in western Bengal). In this time period of Mughal rule, Dhaka prospered as an important hub of inter-provincial and foreign trade. Being famous for its high quality cotton textiles (muslin), Dhaka attracted merchants from all over the world (Karim 2006b). From the middle of the 17th century onwards also European merchants appeared in the city and established their trading stations (called factories) at the urban fringe around the area of today’s Tejgaon,

The Food System Of Dhaka

53

i.e. the Dutch in 1663, the English in 1667, and the French in 1682 (Karim 2006a).10 In this period, the East India Company step by step became the dominant economic and political power in the region. The Mughal emperor Shahjahan permitted the company to pursue trade and commerce in Bengal in 1633, and the subahdar Shah Shuja allowed the traders to buy and sell goods in Bengal without having to pay customs duties in 1651. In 1696, the subahdar Ibrahim Khan gave the company the permission to establish a fort at Calcutta, and in 1717 the Mughal emperor Farrukh Siyar issued an order (farman) that allowed the British traders to purchase land titles (zamindari) in this area. This farman also allowed them to move their goods without being stopped at the custom houses (chowkis), a fact that was not tolerated, however, by the Bengal governors and that finally provoked the subahdar Sirajuddaula to attack the British at Calcutta in 1756. Even though the subahdar accomplished the British to flee and to leave their fort, this was only a Pyrrhic victory. Due to a conspiracy of the company and Mir Jafar, the head of the army (bakhshi) of the Bengal province, later in 1757, the British defeated the subahdar in the sham battle of Plassey (Palashi). Finally, in 1765, the East India Company was assigned the right to control the revenue administration (diwani) of Bengal by the Mughal emperor on payment of a fixed amount of money. This made them the de facto rulers of the province (Islam 2006). Calcutta became the seat of colonial rule. The entire muslin industry of Dhaka literally vanished (Ahmed 2006) and the collection of arbitrarily high and unsteady taxes made agriculture unprofitable that let the peasants to stop tilling their fields. This severe mismanagement culminated in the most severe famine of that time that – triggered by natural calamities – caused in between 1769 and 1770 the deaths of an estimated number of 10 million people or one third of the total population of Bengal (Rahman 2006). The famine aroused the British parlament and forced it to take action against the exploitative practices of the company and its servants and to gradually establish governmental control over the company’s affairs and over Bengal, a process that started with the Regulating Act of 1773 and that was completed with the issuing of the India Act in 1858. After the mutiny of 1857, the East India Company went bankrupt and had to be taken over by the British government. From that time onwards, the Company was abolished and Bengal together with India was officially brought under the rule of the British parlament (Islam 2006). Dhaka became the headquarters of the large Dhaka Division in 1829 and in 1864, the Dhaka Municipality was established. Around that time, the city also began to regain economic and cultural importance. In the 1820s and 1830s, the trade in indigo, that was used to color textiles, heralded a new time of prosperity. From the 1850s onwards, it was especially the trade in jute that became the main 10 The Portuguese established a port for trading purposes in Hughli in western Bengal already in 1579. However, their influence on Dhaka remained small as their trade declined by the beginning of the 17th century when Dhaka became the capital of the province (Akhtaruzzaman 2006; Karim 2006a).

54

The Food System Of Dhaka

engine of the Bengal economy. A railway line connecting Dhaka with Mymensingh in the North and with Narayanganj in the Southeast was built in 1885 and 1886 (Khatun 2003). Contemporaneously, Dhaka experienced an important impetus by the foundation of educational centers, e.g. the Dhaka College (1835), Dhaka Madrasa (1874), Dhaka Medical School (1875), and Dhaka Survey School (1876) that is known today as Bangladesh University of Engineering and Technology (Ahmed 2006). The 20th century again entailed severe political and administrative changes in the region, with Dhaka playing a central role in this process. After an official visit of Bengal, which included at that time also Bihar and Orissa, the general governor and viceroy Lord Georg Curzon called for a redrawing of the administrative boundries since the region was seen to be too large to be governed effectively. Thus, the new province of “Eastern Bengal and Assam” was created in 1905 with Dhaka as its capital. Opponents heavily criticized the partition of Bengal that was said to be a measure against the nation of all Bengali speaking people. Antipartitionist agitation manifested itself most prominently in the form of the Swadeshi Movement (1905 to 1908) to boycott the import of British manufactured goods – a measure that should later inspire the disobedience movements of Mohandas Karamchand (Mahatma) Gandhi. It was this movement that provoked the famous poet and philosopher of Bengal Rabindranath Tagore to write in 1906 the song “Amar Sonar Bangla” (“My golden Bengal”) that constitutes the national anthem of Bangladesh since independence. But the partition did not only provoke solidarity, but also antipathy among the people of Bengal, since it successively led to the organisation and solidification of a mutual antagonism between Hindus and Muslims. Due to an increasingly felt influence of Hindus in the All India National Congress (that was founded in 1885), in 1906, the All-India-Muslim-League was founded that used the purported agrarian conflicts between Hindu landlords (zamindars) and Muslim tenants (raiyats) to call for a separate representation of Muslims and, later, explicitly supported the partition. This led in 1909 to the passing of the Morley-Minto Reforms that allowed the Bengal Muslims a separate electorate. The conflict culminated in 1910. The partition of Bengal was revocated in 1911 and the capital of British India was moved from Calcutta to Delhi (Ahmed 2006). In 1921, the Dhaka University was founded as a compensation for the annulment of the partitition that involved Dhaka to loose its function as the administrative center of Eastern Bengal (Miah 2006). British rule over Bengal ended like it had started, that is by a severe famine in 1943, in which 1.5 to 3 million people lost their lives in times when there was not even an absolute shortage of food (Sen 1981). Only four years later, the war-torn British empire released the Indian subcontinent into independence. With the passing of the Indian Independence Act of 1947 British India was partitioned in two sovereign states, the “Union of India” (that became the Republic of India in 1950) and the “Dominion of Pakistan” (that became the Islamic Republic of Pakistan in 1956). In this course, the former province of Bengal was again partitioned into West Bengal that – with a majority of Hindus – became part of the Indian dominion, and East Bengal that – with a Muslim majority – became part of the Pakistan

The Food System Of Dhaka

55

dominion. In a separate referendum, the electorate of Sylhet, then still part of Assam province, voted for joining Pakistan. In 1956, East Bengal became the new province of East Pakistan (Rashid 2006). Immediatly after its creation, Pakistan faced a new conflict that emerged over the question of the national language. Even though more than half of the total population of Pakistan spoke Bangla as their mother tongue, the political elites in West Pakistan demanded Urdu to become the one and only national language. Since the prime minister of Pakistan, Liaquat Ali Khan, and the chief minister of East Pakistan, Khwaja Nazimuddin, opposed the demand to declare Bangla as one of the state languages and as the official language of East Pakistan, in 1948, the Language Movement started and rapidly gained public support. In 1949, the Awami Muslim League (that became the Awami League in 1955) was founded against the background of a growing antipathy against the Muslim League and the new form of colonialism that Pakistan was seen to pursue under its rule. The Language Movement culminated in 1952, after Khwaja Nazimuddin – then Prime Minister of Pakistan – visited East Pakistan and, once again, declared Urdu the national language. When the students of Dhaka University called for strikes (hartals) and assembled on the campus on 21 and 22 February of 1952, armed police forces opened fire against them and killed five persons (Van Schendel 2009). In the face of this tragedy, the Language Movement went on until Bangla became one of the two national languages in the constitution of the Islamic Republic of Pakistan in 1956 (Helal 2006).11 In the following years, the conflict over language, religion and national identity aggravated between East and West Pakistan and culminated in 1970, when Sheikh Mujibur Rahman, the leader of the Awami League, came off as winner of the general elections. After the political and military elite of West Pakistan refused to yield power to him, Sheikh Mujibur Rahman – as the main representative of the people of East Pakistan and the majority leader in the Pakistan Assembly – gave his historic adress on 7 March 1971 at the Ramna Racecourse that constituted the starting point of the civil disobedience movement in the eastern part of the country: “The struggle this time is for emancipation! The struggle this time is for independence!“, he declared. In the morning hours of 26 March, the Pakistan army launched the military operation “Search Light” and attacked the people of Bengal. Shortly after, on the evening of 26 March, the independence of Bangladesh was declared12. On the same day, Sheikh Mujibur Rahman was taken into custody by the Pakistan military. An exile government of the “People’s Republic of Bangla Desh” was formed on 10 April and the freedom fighters (mukti bahini) – mostly former soldiers of the Pakistan Army who had deserted directly after the announcement of civil disobedience – organized a guerilla warfare with the active 11 In 1963, the memorial of Shaheed Minar was established on the campus of Dhaka University to commemorate the struggle of the people of Bengal for their language. In 1999, the United Nations Educational, Scientific and Cultural Organization (UNESCO) adoped a resolution proclaiming 21 February as the International Language Day (Helal 2006). 12 As Independence Day, 26 March is today a national holiday in Bangladesh.

56

The Food System Of Dhaka

support of the Indian army. In the United Nations security council, the United States of America and the People’s Republic of China considered the civil war to be an internal affair of Pakistan, while the Soviet Union backed the Bengal struggle for independence. On 3 December, the joined command of the mukti bahini and the Indian Army launched their attack and the Pakistan-India war started. 13 days later the Pakistani troops in East Pakistan surrendered to the Indian army (Islam 2006).13 Since then, Bangladesh is a sovereign state with Dhaka as its capital. Only three years later, the dream of a “golden Bangladesh” came to an end. The material damage caused by the war was extensive and despite international relief and reconstruction projects that were coordinated by the UN, the young country soon found itself in an economic and political crises. In 1974, the adverse situation culminated in the most severe famine of recent times that caused the death of 1.5 million people (Van Schendel 2009). Prices rose dramatically, a fact that was mainly attributed to the hoarding and speculating activities of large-scale brokers that were spured by previous floodings, and to an embargo on food aid that was errected by the USA to force Bangladesh to abdicate its export activities to Cuba (Sen 1981). And yet, these facts were outweighed by the dubious course of the Awami League government that promised to take action against the hoarders, but backed those brokers who exhibited party affiliations. This behavior led to a tremendous tremor of the people’s trust in their government, who in turn started to help the famine victims by providing free cooked food and relief on the basis of private initiatives and thus helped to save the lifes of millions. Clientelism and corruption on part of the government – whether led by Awami League or by the Bangladesh Nationalist Party – intensified in Bangladesh in subsequent years so that today, mistrust in the government has become a part of daily life. 4.1.2 From City to Megacity The fluctuation of the political, administrative, economic and cultural position of Dhaka in the regional context involved severe changes of the city’s population size and spatial extent over time. In the pre-Mughal period, the city is said to have comprised “fifty two bazars and fifty three lanes” (Islam 2005: 6) and was located at the bank of Buriganga river which also marked its southern border. Around 1600, the city might have occupied a total area of three km2 (Khatun 2003). Its most probable spatial extent can be described as a semi circle with today’s Sadar Ghat as its center point and the Dulai canal (Dulai khal) as its circumference. If this is correct, the city of that time extended from Badamtuli and today’s 2nd Buriganga Bridge in the West, to Narinda in the North with the Dulai canal as its demarkation line, and to Farashganj in the East where, today, remains of the for-

13 As Victory Day, 16 December is today a national holiday in Bangladesh.

The Food System Of Dhaka

57

mer canal can still be visited (s. Map 2 in the annex). There are no available estimations on Dhaka’s population size for that time. During Mughal rule, Dhaka faced considerable population increase and physical expansion due to its status as provincial capital and its prospering muslin industries. The french traveller and trade pioneer Jean-Baptiste Tavernier (2004[1676]), who visited Dhaka in 1666, estimated the city to extend to a length of two leagues (9.6 km), mostly along the bank of Buriganga. At that time, the city is said to have comprised approximately 200,000 people (Islam 2005).14 The East India Company gave in 1786 and 1800 the same figures. Remarkably, the company recorded the city’s boundaries to be the Buriganga river in the South, Mirpur in the Northwest, and Tongi in the North, an area that equals the territory of today’s municipality of Dhaka. In this context, however, it should not be overseen that, at that time, the area up to Tongi was only scarcely populated due to its interspersion of waterways, marshes and swamps (Chowdhury 2006). The builtup area covered not more than 30 to 40 km2 (Ahmed 2006; Siddiqui et al. 2000) and might have been located between Rayer Bazar and Hazaribag in the Northwest, Postogola and Sutrapur in the East, and Ramna Park in the North (s. Map 2). Within the period of the rule of the East India Company and later the British, Dhaka did not only loose its importance in the region, but also shrinked in terms of population and extent. The commercial resident James Taylor (2011(1840]) described Dhaka as being decayed to large extents with most of the former city area filled with jungle and swamps. In 1859, its built-up area is said to have covered an area of ten to 14 km2 (Chowdhury 2006; Siddiqui et al. 2000) and might have been located in between Lalbag in the West and Postgola in the East, and from Buriganga river in the South to Shahbag in the North (s. Map 2). In 1872, the first census recorded for Dhaka a total population of 69,212 people (Ahmed 2006; Siddiqui et al. 2000). In the 20th century, Dhaka again experienced a constant growth in both physical and demographic terms. From 1901 to 1911 Dhaka’s population size increased from 104,000 to 154,000 people at an average annual growth rate15 of 4.0 percent due to Dhaka’s new status as provincial capital of Eastern Bengal and Assam. After the annulation of the partition of Bengal in 1911, however, Dhaka’s growth rate decelerated to 0.9 percent (1911–1921), but slowly regained momentum with 1.5 (1921–1931) and 2.0 percent (1931–1941) (Islam 2005). Massive migration inflows came along with the partition of the subcontinent and led to a population increase from 239,000 to 336,000 people in between 1941 and 1951 (Ahmed 2006; Islam 2005) at an average annual growth rate of 3.5 percent and to 557,000 14 Some historians even speak of a population figure of 900,000 for Dhaka of the 1700s. However, in this regard, I follow the Nazrul Islam (2005) who criticizes this figure to be unrealistic. 15 Assuming an exponential population growth, the average annual growth rate is calculated by the formula: Growth rate (t0, t) = [A(t)/A(t0)]1/N – 1, with A being the population figure at a given time (t) and N being the years in between the two reference years t0 and t. Accordingly, the growth rate (1901, 1911) = [154,000/104,000]1/10 – 1 = 0.04 = 4.0 percent.

58

The Food System Of Dhaka

in 1961 at a rate of 5.2 percent. Until 1961, Dhaka is said to have successively expanded to 73 km2 (Khatun 2003), a figure that is taken as the main built-up area for the time of Pakistan rule (s. Map 2 and Fig. 4). After independence, Dhaka became one of the fastest growing cities worldwide. In order to produce a picture as exact as possible, special caution must be paid on the city’s constant spatial expansion. Hence, from 197416 onwards, it is distinguished between the population of Dhaka City and the population of the Metropolitan Area of Dhaka. Dhaka City refers to the area of the Dhaka City Corporation (DCC) that comprises 145 km2 and extends from Buriganga river in the South, to Mirpur in the Northwest, and Tongi in the North (Islam 2005). While the DCC area stays constant, the Metropolitan Area of Dhaka changes over time. In 1974, it refers to the Dhaka Metropolitan Area (DMA) that covers a territory of 306 km2 including DCC (ibid.). In contrast, from 1981 to 2001, it refers to the Dhaka Statistical Metropolitan Area (DSMA). In 1981, the DSMA covered an area of 797 km2, while in 1991 and 2001, it was extended to an area of 1,353 km2. The DSMA includes the DMA, the DCC, and the municipalities of Narayanganj and Munshiganj in the Southeast, Savar in the West, as well as Tongi and Gazipur in the North (ibid.) (cf. side map in Map 2). In order to avoid mistakes that stem from this constant change of reference areas, all figures on growth rates displayed in Figure 4 refer to the DCC area only. Figure 4: Population growth of Dhaka (1901–2010) 1.047

Total P opulation of D haka (in mio)

8 12 6.039

10

6

8 4.339

4

6 3.005

2

4

7.144 0.094

0.964

2

5.334

0

3.839 2.476 0.557 0.104 0.154 0.169 0.196 0.239 0.336 1.68

Megacity of Dhaka, Additional Population (in mio) Metropolitan Area of Dhaka, Additional Population (in mio) Dhaka City, Total Population (in mio) Average Annual Population Growth Rate, Dhaka City (in %)

.) (f c st

01

91

81

74

61

51

41

20

10

20

19

19

19

19

19

19

21

31 19

19

19

19

11

-2 01

0

A verage A nnual P opulation Growth R ate (in %)

10 14

Source: Own draft based on Ahmed (2006), BBS (2007), Islam (2005, 2010), Khatun (2003)

16 The census of 1971 could not be held in time due to war reasons. It was conducted in 1974.

The Food System Of Dhaka

59

Notes: 1. Population figures of “Dhaka City, Total Population (in mio)” refer to Dhaka Municipal resp. Dhaka City Corporation (DCC) area. 2. Population figures of „Metropolitan Area of Dhaka, Additional Population (in mio)” are calculated by taking the specific reference area and subtracting the resp. figures that refer to the DCC area. The 1974 figure refers to Dhaka Metropolitan Area (DMA). The 1981, 1991 and 2001 figures as well as the 2010 forecast refer to Dhaka Statistical Metropolitan Area (DSMA) in the size of the respective year. 3. The 2010 population forecast of “Megacity of Dhaka, Additional Population (in mio)” refers to the Dhaka Metropolitan Development Plan (DMDP) area.

By taking these figures, the population growth of Dhaka City reached a peak from 1961 to 1974 with an average annual rate of 8.9 percent. Its population jumped in this time from 557,000 to 1.680 million people. Afterwards, the growth rate slowly decelerated from 5.6 percent (1974–1981) to 4.5 percent (1981–1991) and in 1991, the city reached a population size of 3.839 million. Witin the same time period, the Metropolitan Area of Dhaka grew from 1.774 million in 1974 (Islam 2005) to 6.844 million people in 1991 (Islam 2005; Khatun 2003). The census of 2001 estimated the population of Dhaka City to be 5.334 million and that of the Metropolitan Area of Dhaka as 9.673 million (BBS 2007a). The average annual growth rate of Dhaka City from 1991 to 2001 was 3.5 percent (s. Fig. 4). In order to provide a figure for 2010, Islam proposes for the Megacity of Dhaka to take into consideration the entire territory of the Dhaka Metropolitan Development Plan (DMDP) that equals an area of 1,528 km2 (cf. side map in Map 2). For this reference area, he estimates a population of 14.230 million people in 2010. Assuming a static population increase at an average annual population growth rate of 3.5 percent (which equals the rate in between 1991 and 2001), the respective population figures for Dhaka City and for the Metropolitan Area of Dhaka would be 7.144 million and 13.183 million people respectivley (s. Fig. 4). If these figures are correct, 50.2 percent of the total population of Dhaka live today within the DCC area, while 49.8 percent live in the peri-urban adjacencies around the megacity. These figures let Dhaka become a primate city in both demographic and functional terms. Figure 5 shows the country’s city size ranking that exhibits the capital’s exeptional position within the urban system of Bangladesh. While the horizontal axis represents 220 listed municipalities and the Statistical Metropolitan Areas (SMAs) of the four largest cities Dhaka, Chittagong, Khulna and Rajshahi ranked after population size, on the vertical axis the absolute population figures of each city are charted. Dhaka’s demographic primacy is revealed by the pointed curve progression. While the grey line represents the balanced ideal (Heineberg 2006), in Bangladesh, Dhaka is clearly outstanding.

60

The Food System Of Dhaka

Figure 5: City size ranking of municipalities and SMAs in Bangladesh

Source: Own draft based on BBS (2007) Note: The population figures of the highest ranked cities, i.e. Dhaka, Chittagong, Khulna and Rajshahi, refer to Statistical Metropolitan Areas (SMAs). All other figures refer to Municipalities.

According to the 2001 census, the population size of Chittagong – as the second largest city of Bangladesh – is with a population of 3.266 million people only onethird of Dhaka. Khulna and Rajshahi follow with 1.173 and 0.651 million people (BBS 2007a). Dhaka’s functional primacy is reflected in the fact that the city is not only the seat of the national government, the center of political power, but with most of the garment industry concentrated in and around the metropolis also the workshop of Bangladesh’s export-oriented economy. In addition to that, Dhaka is also the cultural heart of the country with its vast university campus and numerous private universities, as well as with its museums, exhibitions and art performances that are unique in the country. How can such a city that has grown so fast in the recent past be constantly supplied with sufficient quantities of food? In the subsequent two sections I am going to focus on Dhaka’s supply with rice and fish and discuss the current performance of the urban food system from a food security perspective.

The Food System Of Dhaka

61

4.2 RICE FOR THE MEGACITY17 4.2.1 Rice Production in Bangladesh Agriculture (including horticulture, livestock and forestry) is one of the most important sectors of the Bangladesh economy contributing (in 2006) 15.9 percent to the national gross domestic product (GDP) and providing (in 2000) employment for 62.9 percent of all employed persons (MoA 2007). Accordingly, today, roughly 40 million people make a living from agriculture (cf. BBS 2011). Rice is the most important food grain in Bangladesh today, even though, at times, a slight proportional decline could be observed. In the 1970s (1971–1979), rice made on average 98.2 percent of the total annual food grain production, while in the 2000s (2000–2006) this ratio decreased to an average of 93.9 percent (cf. MoA 2007). Today, rice contributes 68.1 percent to the nation’s daily calorie intake, and is the nation’s largest single protein provider with 51.6 percent of the daily intake (BBS 2007c). The confluence of three major rivers, i.e. Padma (Ganges), Jamuna (Brahmaputra) and Meghana (Meghna), led to the emergence of a gigantic delta. The constant deposition of alluvial soils has created some of the most fertile plains in the world. Due to its tropical climate18 with its high monsoon rainfalls in summer19 and its mild and dry winter seasons, more than 4,000 varieties of rice exist in Bangladesh, as it is documented in the National Museum in Dhaka (visit on 10 August 2007). Natural disasters occur frequently, such as floods and river erosion (mainly in the north-eastern and central low lands), droughts (mainly in the western hill areas), cyclones and salt water intrusion (in the coastal regions) (Ahmed 2001). Nevertheless, the general topology of vast low lands20, plenty of soil moisture and high temperatures21 throughout the year are well suited for the cultivation of rice and allow for up to three rice harvests per year, namely aman, boro and aus. The seedlings of aman, the flood season rice, are broadcasted in the month of Caitra (March/April), while they are transplanted during monsoon in Asar (June/July) and harvested in Agrahayan (November/ December). Boro, the dry season rice, is planted during Paus (December/January) and harvested in Baisakh (April/May) and Jyaistha (May/June), just before the beginning of monsoon rainfalls. The inter-period season of aus lasts from the broadcasting and transplanting of paddy in Baisakh to its harvest in Sraban (July/August) (BBS 2007d: 29). 17 Parts of this section have meanwhile been published in Keck et al. (2014). 18 Bangladesh’s climate is classified as an Aslh climate after Lauer and Frankenberg (cf. Lauer 1999: fold-up map) with a growing period of 12 months per annum and seven to nine humid months a year. 19 The mean annual precipitation in Bangladesh ranges from 1,400 millimeter in the West to 4,300 millimeter in the Northeast (BBS 2011). 20 About half of Bangladesh’s territory is less than 12.5 meters above sea level (Braun & Shoeb 2007). 21 January is with a mean temperature of 25°C the most temperate month, while April is with 35°C the hottest (BSS 2011).

62

The Food System Of Dhaka

30,000

Import/Donation (Rice and Wheat) Net Maize Production

25,000

Net Wheat Production

20,000

Net Boro Rice Production

15,000

Net Aman Rice Production

10,000

Net Aus Rice Production Projected Food Grain Requirement

5,000

Food Grain Requirement

Year

st .)

-0 8 20

10

-1 1

(f c

-0 5

07 20

-0 2

04 20

01 20

98

-9 9

-9 6

19

19

95

-9 3

-9 0

92 19

19

89

-8 7

-8 4

86 19

-8 1

83 19

80 19

-7 5

77 19

71

74 19

19

-7 8

0 -7 2

Production and Requirement (in thousand metric tons)

Figure 6: Food grain production, import and requirement in Bangladesh (1971– 2011)

Source: Own draft based on BBS (2011), MoA (2007) Note: The national requirement is calculated at 16 ounce (453.66 grams) per day and person (MoA 2007).

As it is depicted in Figure 6, Bangladesh’s rice production22 shows a steady increase over the period of the last three and a half decades. In the 1970s, the average annual rice production amounted to 10.2 million metric tons (mt). Until the 2000s, the average rice production had more than doubled to an amount of 22.2 million mt per year. For the year 2010, an even larger amount of 25.8 million mt is forecasted. According to data that is provided by the Ministry of Agriculture (MoA 2007), in the year 2000, the total net food grain production in Bangladesh reached the national requirement. Imports are still necessary to bridge seasonal supply shortfalls. Nevertheless, by taking the aggregate figures, Bangladesh has become self-reliant in terms of food grain production from the year 2000 onwards. Today, the major part of Bangladesh’s rice production stems from boro season. But this was not always the case. In fact, the past decades saw a dramatic change in this regard (s. Fig. 6). In the 1970s, the major share of Bangladesh’s annual production was grown in aman season that amounted to 57.3 percent, and also the share of aus exceeded that of boro with 24.8 percent compared to 17.9 percent. In the subsequent decades, the share of aus declined from 21.0 percent in the 1980s (1980–1989) to only 7.1 in the 2000s and also aman declined, even though only 22 Statistics on rice production do usually refer to quantities of either paddy or rice, that is milled paddy. Due to wastage that incurs during the milling process, rice production figures are usually calculated at two-thirds of paddy production. In this study, the production figures explicitly refer to rice or milled paddy.

The Food System Of Dhaka

63

slightly, from 53.5 (1980s) to 43.1 (2000s). Boro, on the contrary, became the rice season of capital importance. Within the same period, its share increased from 25.5 percent in the 1980s to 49.9 percent in the 2000s. The Ministry of Agriculture forecastes the present ratio in 2010 to be 6.6 percent aus, 40.7 percent aman, and 52.7 percent boro (MoA 2007). Given Bangladesh’s natural suitability for producing rice, it would be highly misleading to attribute the recent production growth to ecological factors. In fact, two developments must be seen as the main drivers of recent production increases, namely the expansion of the total cropped area and the intensification of agriculture by means of so called green revolution technologies, i.e. the use of high yielding varieties (HYV), irrigation, fertilizer and pesticides (Ahmed 2001). Figure 7: Cropped area and average annual yield of rice in Bangladesh (1971– 2006)

Source: Own draft based on MoA (2007)

As it can be seen in Figure 7, the cropped area under rice cultivation increased from an average of 8.1 million hectare (ha) in the 1970s to an average of 10.0 million ha in the 2000s. Accompanied with that was an ever increasing proportion of the area under boro rice. While the area cropped with aman rice stayed more or less the same in this period (5.7 million ha in the 1970s and 5.6 million ha in the 2000s), the area under aus rice decreased significantly from an average of 3.2 million ha in the 1970s to an average of 1.2 million ha in the 2000s. Within the same period, the area under boro rice increased from 1.1 million ha to 3.9 million ha (MoA 2007). Nevertheless, the increase in productivity was mainly due to the adoption of cultivating HYVs. In the 1970s, an average of only 16.7 percent of the total area was cropped with HYVs, while local varieties made the bulk of 83.3 percent. In subsequent years, this ratio successively reversed. In the 1990s (1990–

64

The Food System Of Dhaka

1999) the area under HYVs outnumbered that of local varieties for the first time. Currently (2000–2006), more than two-thirds (70.5 percent) of the total area is cropped with HYVs on average. In consequence of that, the average yields increased significantly. The yields of aus rice increased from an average of 0.9 mt per ha in the 1970s to 1.5 mt per ha in the 2000s. The yields of aman rice increased in the same time from 1.2 mt per ha to 1.9 mt per ha. The yields of boro rice, eventually, increased from 2.0 mt per ha to 3.2 mt per ha (ibid.). These new rice varieties, that are developed by the Bangladesh Rice Research Institute23 (BRRI), have turned out to be a blessing for the people of Bangladesh, since they are the backbone of the country’s food provision. Without the increase in productivity resulting from them, the 142 million people of Bangladesh (BBS 2011) could not be fed. Nonetheless, the new rice varieties are a curse at the same time. The fact that most rice is produced in the dry winter season makes irrigation indispensible. The irrigated area under boro rice increased from 1.0 million ha in 1979 to nearly 3.7 million ha in 2005 (MoA 2007). Of all listed techniques, the area under tube well irrigation – by means of low lift pumps, shallow tube wells and deep tube wells – showed the most remarkable increase from 0.2 million ha in 1979 to 3.7 million ha in 2005 (ibid.). Furthermore, with 276 kilogram (kg) per ha, HYV boro rice requires the highest dosage of fertilizer – mainly nitrogen (urea), phosphorus (trisodium phosphate, TSP), and potassium (muriate of potash, MP) (ibid.). With an average amount of 159.7 kg per ha, the rice economy of Bangladesh consumes more fertilizer than those of its South Asian neightbours, e.g. Pakistan (131.9 kg per ha), Sri Lanka (122.7 kg per ha), India (102.2 kg per ha), or Nepal (22.0 kg per ha) (ibid.). The national consumption of pesticides – mainly insecticides, fungicides and herbicides – increased from 7,300 mt in the early 1990s (1990–1992) to 16,100 mt in the early 2000s (2000–2002) (ibid.). The extensive use of these inputs is not without side effects. Large-scale irrigation leads to the sinking of ground water levels. The extensive use of artificial chemicals exposes the country to wildly fluctuating world market prices and leads to contaminations of the ground and surface water and the soils (World Bank 2006a). Eventually, this agricultural practice has socio-economic effects, since market-oriented rice production makes cost-intensive inputs as necessary as never before in Bangladesh. As one can see in Map 3 in the annex, today (2001), the country’s rice granaries are located in the Northwest and North of the country. Before independence and long before the advent of the green revolution in Bangladesh, today’s re-

23 The Bangladesh Rice Research Institute (BRRI) – established in 1970 as East Pakistan Rice Research Institute (EPRRI) and located in Gazipur in the North of Dhaka – is an independent research organization that deals with research and development in relation to rice production. Its main mission is to breed ecologically fit rice types at low costs through genetic improvement and to train farmers in the use of new technologies (Miah 2006). Its activities are based on research of the International Rice Research Institute (IRRI) in Los Baños, Philippines.

The Food System Of Dhaka

65

gions24 of Barisal, Patuakhali and Sylhet were considered the granaries of Bengal (Ahmed 2001). Today, Sylhet is still ranked fourth of all regions in terms of absolute productivity, but Barisal and Patuakhali have been outpaced by many, being ranked 11th and 18th out of a total of 21 regions. As an effect of the fundamental change that Bangladesh’s rice economy has undergone in the last decades, the most productive regions (with a total net production of more than one million mt per year) are at present: Mymensingh (2,242,000 mt), Rangpur (1,819,000 mt), Rajshahi (1,726,000 mt), Sylhet (1,630,000 mt), Comilla (1,570,000 mt), Jessore (1,413,000 mt), Dhaka (1,133,000 mt), Dinajpur (1,120,000 mt) and Bogra (1,101,000 mt) (BBS 2007d). By taking population figures (BBS 2007b) into account and the average amount of 458.5 gram (g) rice that is consumed in Bangladesh per person and day (BBS 2007c), we can distinguish in between surplus and deficit regions in terms of net rice availability. As can be seen by the color of the surfaces in Map 3, there are considerable differences in the regions’ population density. Clearly outstanding is the region of Dhaka with a total population of 17.3 million and a population density of 2,326 people per km2 followed by Comilla (population: 9.2 million, population density: 1,371 people per km2) and Cittagong (population: 8.3 million, population density: 1,068 people per km2). The other extreme are the two hill tract regions. Khagrachari is populated by around one million people and Bandarban by slightly more than 300,000. Their population densities are 118 and 67 people per km2, respectively. Given these population differences, considerable surplus is produced (in 2001) in the regions of Mymensingh (738,000 mt), Jessore (484,000 mt), Rajshahi (454,000 mt), Bogra (453,000 mt), Dinajpur (334,000 mt), Sylhet (309,000 mt) and Rangpur (298,000 mt). Regional production deficits prevail in Bandarban (-12,000 mt), Noakhali (-12,000 mt), Kushtia (-47,000 mt), Pabna (62,000 mt), Khagrachari (-78,000 mt), Khulna (-133,000 mt) and Faridpur (161,000 mt). It is not surprising that the two major rice deficit regions are those where the two largest urban agglomerations are located, namely Chittagong (439,000 mt) and Dhaka (-1,763,000 mt). As it can be seen in Map 3, Dhaka’s rice supplies have their origin almost entirely in the north-western part of the country that once were the main growing areas of jute (Van Schendel 2009). According to the Rice Market Survey 2009– 2010, the lion’s share of 42.8 percent stems from the Rajshahi region, followed by 17.3 percent that comes from Kushtia. 9.5 percent are brought in from Bogra, 8.3 percent from Dinajpur and 5.2 percent from Jamalpur. Remarkably, only 6.9 percent of Dhaka’s supplies are produced in the region. Amounts, ranging from 1.0 to 5.0 percent each, are from Mymensingh (4.0 percent), Papna (1.6 percent), Jessore (1.1 percent) and Faridpur (1.1 percent), while amounts smaller than 1.0 percent come from Tangail, Comilla, Sylhet, and Barisal. With Kushtia, Pabna and Farid24 The term “region” is adopted from the Bangladesh Bureau of Statistics (BBS 2007d). The regions correspond in their extent to the country’s old districts (from 1971 to 1984), but are not entirely congruent.

66

The Food System Of Dhaka

pur, Dhaka receives rice supplies also from rice deficit regions. This is due to the fact that wholesalers in Dhaka do not get their supplies directly from producers, but from intermediaries who are located in these regions, as the value chain analysis further below will show (s. 5.2.1). 4.2.2 Availability of Rice in Dhaka At all food wholesale markets in Dhaka City together, every single day, the substantial quantity of 9,026 mt is realized. In terms of weight, rice and vegetables exhibit the highest quantities with amounts of approximately 3,000 mt per day each. Around 1,000 mt pulses and spices and nearly 700 mt fresh fish is sold at Dhaka’s wholesale markets. Fresh fruits, edible oil and meat follow in some distance with more than 300 mt each. The amounts of eggs, wheat and dried fish are comparatively low with quantities below 100 mt (s. Tab. 7). By taking the total amount of 3,010 mt of rice that is sold in Dhaka City every day and by dividing this amount by the city’s estimated total population of 7,144 million people in 2010, it turns out that there are 421.3 g of rice available per capita and day. Table 7: Gross food availability in Dhaka City Food type

Total turnover (in mt)

Gross food availability (in g per capita and day)

National urban Food balance in average of food Dhaka City intake in 2010 (in % of the (fcst) (in g per national urban capita and day) average) 379.3 111.1 58.3 169.8 88.9 165.9 30.4 168.4 260.7 164.6 27.5 166.9 23.5 162.6 6.9 140.6 19.9 42.7 895.4 Source: Own draft based on surveys 2008 –2009

Rice Fresh Fish Pulses/Spices Fruits Vegetables Edible Oil Meat Eggs Wheat Total

3,010 707 1,054 366 3,065 353 341 69 61 9,026

421.3 99.0 147.5 51.2 429.0 45.9 38.2 9.7 8.5 1250.3

Note: 1. In regard to the total daily turnover, the figures of rice and fish are the result of a triangulation of data of three surveys, i.e. the food wholesale market survey 2009, the rice and fish panel survey 2008–2009, and the rice and fish market survey 2009–2010. All other figures are based on the food wholesale market survey 2009 only. In order to provide maximal accuracy in terms of the total daily turnover per food type, figures of considerabel size (pulses/spices, fruits, vegetables and edible oil) were reduced by an average error ratio of 20.0 percent. This ratio resulted from the above mentioned triangulation. 2. For calculating the quantity of meat supply to Dhaka the absolute figure of live cattle was taken, while the figure of meat was left out in order to avoid double counting. The selection of live cattle was done in cognizance of the fact that not all animals are sold to meat wholesalers. A considerable amount of animals is directly sold alive or it goes directly to retail butchers. As such, the meat wholesale markets do represent only a part of Dhaka’s meat supply. The calculation of the meat

67

The Food System Of Dhaka

equivalent of the live cattle was done as follows: From the absolute figure, a weight loss of 50.0 percent was assumed due to slaughtering. This loss mainly stems from bones and skin. From the result another 25.0 percent was subtracted due to sinews and (parts of the) innards that are usually not eaten. These percentages are based on local information and mainly apply to cattle and goats. For the case of chicken they are much less. Nevertheless, since cattle and goats make by far the largest share of meat supplies in terms of absolute weight, respective percentages for these animals were taken.

According to official data, the average quantity of rice intake per person and day for urban areas in Bangladesh in 2010 is likely to be 379.3 g – a figure that is the result of extrapolating the statistics available for the years of 2000 and 2005 by assuming linear growth (cf. BBS 2007c). If these figures are correct, the availability of rice in Dhaka equals 111.1 percent of the national urban average. If we proceed likewise with other available figures, it turns out that Dhaka is well supplied by all food types exept wheat. In fact, for the case of fresh fish, pulses and spices, fruits, vegetables, edible oil, and meat it is found that Dhaka’s supply in terms of quantity is with an average of 166.4 percent well above the national urban average. The supply of eggs is a bit lower but still amounts to 140.6 percent of the national urban average (s. Tab. 7). Wheat is the only food item that is below average, a figure that might be explained by the fact that the large wheat mills and storages of Dhaka are traditionally located in Narayanganj, which is just outside of the DCC area and not far from the city center at the bank of Buriganga in the Southeast. Figure 8: Food consumption in urban areas and in Dhaka City (from left to right) 4% 1%

3% 1% 7%

Rice

8%

Wheat

3%

3%

3%

33%

4%

Pulses and Spices

42%

10%

Vegetables

Edible Oil

12%

Meat Fish 1%

Fruits 29% 2%

34%

Source: Own draft, calculation for 2010, based on BBS 2007c

Eggs

Source: Own draft, estimation based on own survey 2009–2010

Further information is provided by considering the percentaged share of each food type. As can be seen in Figure 8, rice consumption in Dhaka is lower than in other cities. Accordingly, in Dhaka, only 33.7 percent of the total quantity of food is rice while the national urban average is forecasted to be 42.4 percent in 2010 (cf. BBS 2007c). These figures can be read as an indicator for the existence of numerous middle and upper class families in Dhaka whose higher incomes allow them

68

The Food System Of Dhaka

to substitute staples like rice by superior and more expensive food. Accordingly, fish and meat are in Dhaka constitutive for 7.9 and 3.1 percent, while they make only 6.5 and 2.6 percent on national urban average. Similarly, the consumption of edible oil, vegetables, spices and pulses, and fruits is relatively higher in Dhaka than in other cities in Bangladesh with amounts of 3.7, 34.3, 11.8 and 4.1 percent respectively compared to 3.1 (edible oil), 29.1 (vegetables), 9.9 (spices and pulses), and 3.4 percent (fruits). Going by their price levels, the various rice types that are sold in Dhaka were classified into four categories, i.e. low value, medium value, high value and special value rice (s. Tab. 8). Low value rice ranges from 23.10 to 25.40 BDT per kg and comprises mainly coarse rice (Mota Chal), but also the variety of Guti. Coarse rice is again subdivided by local varieties and HYVs that stem from the Bangladesh Rice Research Institute (BRRI). The average of this price category is 24.30 BDT per kg. Medium value rice range from 27.60 to 31.90 BDT per kg. With an average of 29.60 BDT per kg it comprises the varities of Shorna, BR 29, Parija, BR 28, Atob, Paijam, Kajal Lata, and Haski. High value rice is Miniket and Nazirchal whose average annual prices are 35.30 and 39.40 BDT per kg. The average of this price category is 37.30 BDT per kg. Special value rice are those types that are mainly served during festival times, public holidays and private ceremonies. This category exceeds others by far with an average of 69.10 BDT per kg and comprises mainly the two varieties of Chinigura (66.60 BDT per kg) and Polau Chal (71.70 BDT per kg).25

25 It is not possible to get an accurate figure of the share of Dhaka’s rice supplies that stem from HYV and from local varieties. Information on the names of both types of varieties is available from the Ministry of Agriculture, but these names do only in some rare cases match the rice types that are offered by wholesalers in Dhaka. While some types can be found in the official rice list, e.g. BR 28, BR 29, Haski (BR 17) or Paijam, other types such as Miniket, Parija, Atob or Guti do not corresponded to any of the listed varieties. They are the outcome of local classification and processing practices of several rice varieties; they are no varieties in a biological or bio-technological sense. Thus, by taking into account that 70.5 percent of the total cropped area in Bangladesh is under HYV and by acknowledging the statement of peasants on the role of HYV as cash crops that are mainly produced for the urban markets (s. 5.2.1), one can reckon that at least 80.0 percent of Dhaka’s supplies consist of HYV rice.

69

The Food System Of Dhaka

Table 8: Average annual rice wholesale prices in Dhaka (2009–2010) Low value rice Mota Chal (BRRI) Mota Chal (Local) Guti

Average price (in BDT/kg) 23.10

n

Medium value rice Shorna

Average price (in BDT/kg) 27.60

8

24.50 25.40

n

8

High and special value rice Miniket

Average price (in BDT/kg) 35.30

n

49

44

BR 29

28.60

2

Nazirchal

39.40

3

18

Parija BR 28 Atob Paijam Kajal Lata Haski

28.90 29 Chinigura 66.60 3 29.00 6 Polau 71.70 2 29.20 5 29.70 1 31.70 14 31.90 5 Source: Own draft based on survey 2009–2010

Note: 1. Average annual prices are taken to even the price variability of the individual rice types throughout the year in order to get more general figures. The average annual wholesale price is calculated by the formula: PØ = [P(th)*th]+[P(tl)*tl]/12. P(th) is the price during high season, that is multiplied by the number of months in which this season is prevalent (th). P(tl) is the price during low season, that is similarly multiplied by the number of months in which this season lasts (tl). The result is then divided by 12 month. The price dynamics at Dhaka’s rice and fish markets are discussed in full detail in the next part (s. section 4.2.3). 2. In the course of the survey, the curreny rate for 100.00 BDT changed from 0.973 EUR (15.12.2009) to 1.011 EUR (01.01.2010) to 0.984 EUR (15.01.2010) (cf. http://www. bankenverband.de/service/waehrungsrechner). By taking the mean of these three, the average rate in the time of the survey was 0.989 EUR.

Figure 9 shows that the rice types, that are mostly traded at Dhaka’s wholesale markets, are Miniket with 16.6 percent, Mota Chal (Local) with 12.6 percent, and Parija with 11.7 percent. Kajal Lata (7.4 percent), Nazirchal (7.0 percent), Shorna (6.6 percent), and Guti (6.5 percent) range in the middle while BR 28 (4.4 percent), Haski (3.6 percent), Mota Chal (BRRI) (2.3 percent), and Atob (1.9 percent) follow behind. Smaller proportions of below one percent are Paijam (0.8 percent), BR 29 (0.8 percent), Polau (0.7 percent), and Chinigura (0.7 percent). Miscellaneous rice types with individual proportions of below 0.5 percent are Balam Chikan, Katari Vogh, Akash, Kaloira, BR 10, BR 11, Berui, Sona Masuri, Ratna and others26. The distribution of these rice categories shows that – by leaving out the 15.0 percent of miscellaneous rice types – 70.1 percent of the total rice turnover in Dhaka City is low and medium value rice. Out of that, 25.6 percent belongs to the low and 44.5 percent to medium value rice. 28.2 percent is high value rice, while the remaining 1.7 percent is special value rice. While Table 7 above showed

26 In the questionnaire, it was only asked for the price of the most important rice type (in terms of its respective share of the total turnover) of each individual wholesaler. Since none of the miscellansous rice types was of main priority for any of the wholesalers, it was not possible to assign them to any of the above mentioned price categories.

70

The Food System Of Dhaka

that Dhaka’s total rice supply is well secured, these figures now provide a more detailed overview. Figure 9: Rice consumption in Dhaka City 13%

15%

Mota Chal (LV) Guti Mota Chal (BRRI)

1%

7%

Parija

1%

Kajal Lata 2%

7%

Shorna BR 28 Haski

12%

Atob Paijam BR 29

16%

Miniket 7%

Nazirchal Polau Chal

1% 1% 2%

4%

4%

7%

Chinigura Miscellaneous

Source: Own draft based on survey 2009–2010 (n=197)

By taking the different types of rice and their respective prices, one gets into the position to get a sort of negative image of Dhaka’s society. Generally speaking, it can be assumed that low value rice is mainly consumed by the urban poor, middle value rice is mainly consumed by the middle class, and high value rice is mainly consumed by the upper class, while special value rice is left aside due to its ceremonial character. Based on this assumption it is possible to calculate the rice availability individually for each of these income groups: According to the latest World Bank (2007) study on poverty in Bangladesh’s capital, 28.0 percent of Dhaka’s total population can be classified as poor as they are living below poverty line. These 28.0 percent are fed by 26.0 percent of Dhaka’s rice supplies (that is the proportion of low value rice calculated without taking special value rice into account). In other words, a number of roughly 2.0 million people (that equals 28.0 percent of the total population of 7.144 million people in Dhaka City) is fed by 782,600 kg of rice (that equals 26.0 percent of Dhaka’s total rice supplies). By assuming further that the rice availability of the middle class matches the average of 421 g per capita and day, 3.2 million people are fed by medium value rice that amounts to 1,363,500 kg (that equals 45.3 percent of Dhaka’s total rice supplies without taking special value rice into account). As a result it turns out that 863,900 kg of high value rice (that equals 28.7 percent of Dhaka’s total rice supplies without taking special value rice into account) is provided to 1.9 million upper class people. As such the average rice availability of the lower class in Dhaka is 391 g per head and day, while it is 455 g in case of the upper class. This dis-

The Food System Of Dhaka

71

crepancy provides a first hint on the socio-economic inequalities that currently prevail in the megacity of Dhaka (ibid.). 4.2.3 Accessibility of Rice in Dhaka In terms of supply, Dhaka’s rice and fish markets follow the country’s general seasons. These are the above mentioned three harvest times of boro, aus and aman rice (s. Tab. 9). Most of the rice types that are sold at Dhaka’s markets stem from the boro season. Seven varieties, i.e. Mota Chal (Local), Parija, Atob, Haski, BR 28, Paijam, and Kajal Lata, can be said to be pure boro crops, with high season being at the beginning of the Bengladeshi year in the months of Baisakh (April/May) and Jyaistha (May/June). Mota Chal (BRRI), BR 29 and Miniket are boro and aus crops (whose harvest starts in the month of Sraban), and Guti, Shorna, and Polau Chal are boro and aman crops (whose harvest time is in the month of Agrahayan). Nazirchal and Chinigura, eventually, are exclusively cultivated in the aman season. As Table 9 shows, there are basically two seasonal supply bottlenecks, one in Kartik (October/November) and one in Caitra (March/April). Such shortcomings are not confined to Dhaka, but are prevalent throughout the entire country. In rural areas of Bangladesh – mainly in the Northwest of the country – people call this annual rice supply bottlenecks monga. With this term they address a season of scarcity and hunger, which arises from an employment and income deficit that is most severe before aman rice is harvested (mainly in the months of Asvin and Kartik), but also occurs in the time before boro rice is harvested (mainly in the month of Caitra) (Zug 2006). The monga season is said to mainly affect Bangladesh’s peasants and sharecroppers whose income is directly or indirectly based on agriculture, as the stocks that stem from the previous harvest become depleted while, at the same time, it takes at least two further months for the next harvest to come. Nevertheless, seasonal bottlenecks in supply do also affect Bangladesh’s cities. The most obvious effect is that prices rise as the general supply decreases. In Dhaka, though, this effect is even aggravated due to the seasonal inmigration of rural poor who come to the city to find a job, to be fed by governmental pro-poor programs, or simply to bag in order to secure at least a minimum of food and income. Besides the general lack of income opportunities during that time, it is river erosion and floods that destroy the belongings and livelihoods of people who are then forced to follow the general flows of migrants to the cities. In between 2007 and 2011, an additional factor must be added, that is Ramadan which took place in the very time of monga. The month of fasting as prescribed by Islam is determined by the moon calendar and moves backward by about ten days every year. While in 2007, it started in mid September, in 2011, it started at the first of August. For believers, it is prohibited to eat or drink from dawn to dusk, which they make good for during night. Thus, Ramadan has become the month of feasting as well. Food prices usually rise and worsen the situation for the urban poor. So, in the particular years of this research, the seasonal price variabil-

72

The Food System Of Dhaka

ity in Dhaka was influenced by the general decline in supplies at that time, the increase in demand due to temporal in-migration of people from rural areas, and rising prices due to Ramadan.

Boro harvest

Aus harvest

Caitra (March/April)

Falgun (February/March)

Magh (January/February)

Paus (December/January)

Agrahayan (November/December)

Kartik (October/November)

Bhadra (August/September)

Sraban (July/August)

Asarh (June/July)

Jyaistha (May/June)

Baisakh (April/May)

Rice Type

Asvin (September/October)

Table 9: Seasonal changes in supply at Dhaka’s rice markets

Aman harvest

Low value rice Mota Chal (BRRI) Mota Chal (Local) Guti Medium value rice Shorna BR 29 Parija BR 28 Atob Paijam Kajal Lata Haski High value rice Miniket Nazirchal Chinigura Polau Chal

Source: Own draft based on survey 2009–2010 (n=199) Note: Squares in dark gray color indicate high season, squares in light gray color indicate mid-season, and squares in white color indicate low season.

Table 10 shows the dimensions of seasonal price changes. Notably, the seasonal price variability of low value rice types exceed that one of high value rice types. While the average price increase of high value rice (Miniket, Nazirchal, Chinigura, Polau Chal) during off-season is 18.2 percent, the similar figure for low value rice types (BRRI, Mota Chal, Guti) is 20.0 percent (medium value rice types increase on average by 22.9 percent). Furthermore, in the case of high value rice, there are two distinct high seasons that reduce the time in which prices are high to four months, from Bhadra to Kartik and in Caitra. In the case of low value rice, however, high prices basically last for six months, from Asvin to Kartik and from Paus to Caitra. In the case of Mota Chal (Local), the type of rice that is mostly consumed by low income groups, the period of high prices even last for 11

73

The Food System Of Dhaka

months per year. Thus, as the rice prices in Dhaka show, the phenomenon of monga is also prevalent in urban areas in Bangladesh. In cities, monga can be said to be a seasonally re-occuring phase of food insecurity that arises from an increase in food prices by simultaneously invariable earnings and income that most severely threatens the low income groups and the urban poor. Table 10: Seasonal price changes of major rice types at Dhaka’s wholesale markets Average price, past high season (in BDT/kg) Low value rice Mota Chal (BRRI) Mota Chal (local) Guti Medium value rice Shorna BR 29 Parija BR 28 Atob Paijam Kajal Lata Haski High value rice Miniket Nazirchal Chinigura Polau

n

Standard deviation (in BDT/kg)

Average price, past low season (in BDT/kg)

n

Standard deviation (in BDT/kg)

20.40

8

2.30

25.00

8

2.10

21.00

46

3.20

24.80

44

2.70

22.20

18

2.50

26.50

18

2.00

23.20 24.50 24.60 22.70 24.80 26.00 27.00 26.30

8 2 30 6 5 1 14 5

3.40 3.50 2.80 3.80 3.60 3.80 4.50

29.10 31.50 29.30 30.20 29.60 30.00 32.10 32.40

8 2 29 6 5 1 14 5

4.60 3.50 2.60 4.90 4.20 3.40 3.40

30.50 34.70 59.80 65.00

49 3 3 2

3.60 0.60 8.50 7.10

37.70 49 3.20 41.70 3 3.50 67.90 3 6.30 75.00 2 7.10 Source: Own draft based on survey 2009–2010

Of some importance are also the daily dynamics of Dhaka’s rice wholesale markets as they are shown in Table 11. In the course of a day, the prices of most rice types (Mota Chal (BRRI), Mota Chal (Local), Guti, BR 29, BR 28, Atob, Paijam, Kajal Lata, Chinigura, Polau) stay unchanged. If changes occur, they happen in a range from 0.10 (Haski, Miniket, Nazirchal) to maximally 0.30 BDT per kg (Parija). While the prices for Shorna, Parija, Haski, and Miniket slightly decreased, Nazirchal showed a slight increase in the course of the day. The strongest variability was measured for the cases of Parija and Shorna that decreased by 1.0 and 0.6 percent. The variability of Haski, Miniket and Nazirchal amounts to not more than 0.3 percent each. As such, the daily price dynamics of rice is not neglegible, but very low. Eventually, the longer-term price trend from 2009 to 2010 is presented in Table 12. Even though the individual answers are not completely without ambiguity, one can speak of a general upward trend of prices. In absolute

74

The Food System Of Dhaka

amounts, this trend ranges from 0.30 BDT per kg (Chinigura) up to 20.00 BDT per kg (Polau Chal) and in percentaged figures from 0.5 (Chinigura) to 17.6 percent (Polau Chal). Just like in the cases before, price changes of low value rice are with an average of 11.8 percent slightly more intense than those of high value rice with an average of 11.5 percent (medium value rice increased on average by 13.0 percent).27 Table 11: Daily price changes of major rice types at Dhaka’s wholesale markets Average opening price (in BDT/kg) Low value rice Mota Chal (BRRI) Mota Chal (Local) Guti Medium value rice Shorna BR 29 Parija BR 28 Atob Paijam Kajal Lata Haski High value rice Miniket Nazirchal Chinigura Polau

n

Standard deviation (in BDT/kg)

Average closing price (in BDT/kg)

n

Standard deviation (in BDT/kg)

24.70

8

2.20

24.70

8

1.90

24.60

46

1.90

24.60

45

1.90

25.70

18

1.50

25.70

18

1.50

31.10 31.20 29.60 31.70 31.00 32.00 32.60 31.20

8 2 30 6 5 1 14 5

2.70 3.10 2.80 4.60 4.80 2.70 1.50

30.90 31.20 29.30 31.70 31.00 32.00 32.60 31.10

8 2 30 6 5 1 14 5

2.60 3.10 3.60 4.60 4.80 2.70 1.40

37.30 37.30 64.60 72.50

50 4 3 2

2.00 0.90 1.40 3.50

37.20 49 2.40 37.40 4 0.60 64.60 3 1.40 72.50 2 3.50 Source: Own draft based on survey 2009–2010

27 Apart from seasonal, daily and annual variations, Dhaka’s rice markets do also reveal price changes within the course of a month. Governmental employees and many other businessmen in better-paid positions in private banks or in larger industries usually get paid their salary at the beginning of each (western) month (that is in the middle of the Bangladeshi month). At that time, many of them come personally to Dhaka’s wholesale markets to purchase whole bags of rice that is enough for the whole month to feed themselves and their families. At that time, prices slightly increase due to the higher demand. Furthermore, there are market dynamics within the course of a week, since governmental offices are usually closed during the weekend, that is on Fridays and Saturdays, and also Dhaka’s rice markets are closed on one day per week. As a result, the demand usually shows some slight increase on the day before the markets close. At Badamtuli, for instance, this day is Thursday, since the market is closed on Fridays.

75

The Food System Of Dhaka

Table 12: Recent price trends of major rice types at Dhaka’s wholesale markets Average price (in BDT/kg) Low value rice Mota Chal (BRRI) Mota Chal (Local) Guti Medium value rice Shorna BR 29 Parija BR 28 Atob Paijam Kajal Lata Haski High value rice Miniket Nazirchal Chinigura Polau

n

Standard deviation (in BDT/kg)

Average price, 1 Year Ago (in BDT/kg)

n

Standard deviation (in BDT/kg)

24.80

8

2.20

20.10

8

2.60

24.70

46

1.70

23.20

45

4.50

25.90

18

1.40

23.20

17

4.70

30.80 31.20 29.60 31.70 31.00 32.00 32.60 31.20

8 2 30 6 5 1 14 5

2.50 3.10 2.90 4.60 4.80 2.70 1.50

23.10 29.00 26.00 27.00 25.80 29.00 27.70 30.00

8 2 29 6 5 1 14 5

1.60 5.70 4.30 5.40 3.30 6.00 5.50

37.50 37.50 63.60 72.50

50 4 3 2

1.20 0.40 3.10 3.50

32.30 48 4.40 36.00 3 5.20 63.30 3 2.90 52.50 2 3.50 Source: Own draft based on survey 2009–2010

According to one of my key informants, the price differences between Dhaka’s markets decreased notably in the last fifteen to twenty years. This is mainly due to innovations in and the widely use of mobile telecommunication technologies that came up in the 1990s and count today to the most important tools of any food merchant – not only in Dhaka, but in whole Bangladesh. Before 1990, he stated, the price difference between the markets for one particular rice type and quality was 50 to 75 paisha [0.50–0.75 BDT] per kg. Today, it is usually not higher than 25 to 37,5 paisha [0,25–0,375 BDT] per kg (Own interview, 06.12.2007). This statement can be checked by means of the standard deviation28 of rice prices that was included in the Tables 10, 11, and 12. As Table 10 shows, the variability of actual rice prices ranges from 0.60 BDT per kg in the case of Nazirchal to 4.80 BDT per kg in the case of Atob. In other words, individual prices deviate on average by 1.7 percent (Nazirchal) to maximally 15.5 percent (Atob) from the mean value. These differences can be seen as the natural variability that stems from the different quality of rice within the same typification, and as the market variability that stems from divergent offers of the same rice type by different wholesalers. Even though it cannot be said which factor weighs more, it can be shown that 28 The coefficient of variation could have been used instead.

76

The Food System Of Dhaka

there are again differences between the individual rice price categories. Low value rice types deviate on average by 7.1 percent from the mean, while high value rice types deviate only by 3.5 percent (medium value rice types deviate on average by 10.1 percent). This pattern also holds true for the other figures.29 4.3 FISH FOR THE MEGACITY30 4.3.1 Fish Production in Bangladesh In terms of its economic weight, fisheries is a much less significant sector compared to agriculture, contributing (in 2006) 4.2 percent to the national GDP (MoA 2007) and providing (in 1999) employment to 3.6 percent of economically active persons in agriculture (FAO 1999). Today, just about 1.5 million people are directly involved in fishery. Nevertheless, fishing plays a significant role in Bangladesh’s food system by contributing 62.4 percent of the nation’s animal and 13.6 percent of the total protein intake (cf. BBS 2007c). The vast and diverse inland waters of Bangladesh are inhabitated by 260 species of finfish belonging to 55 families, and by 25 species of prawn. Another 475 species of finfish and 38 shrimp species occur in the marine and brackish waters of the Bay of Bengal (Rahman 1997). The total inland water area is estimated to be roughly 4.6 million ha. This area comprises, on the one hand, man-made formations of ponds and ditches with a total area of 305,025 ha, coastal shrimp, prawn and fish farms with a total size of 217,877 ha, and Kaptai Lake, a reservoir that was dammed up for hydropower purposes in the Hill Tracts in Khagrachari (s. Map 4 in the annex) with a size of 68,800 ha. On the other hand, it comprises natural formations such as the Sundarbans, a huge mangrove forest with an area of 177,700 ha, ox-bow lakes (baors) with a total area of 5,488 ha, depressions which retain water throughout the year (beels) with a total area of 114,161 ha, innumerable rivers and estuaries with a total area of 853,863 ha, and flood lands (flood plains and seasonally flooded depressions (haors)) in a total size of 2,832,792 ha (Kabir & Amin 2007; DoF 2009). Additionally, Bangladesh commands offshore an exclusive economic zone of 7,853,800 ha (SAU 2011). Bangladesh’s fisheries basically follow the country’s annual flood cycle. As it was already indicated in the previous part on rice, the monsoon rainfalls and the subsequent floods are those factors that basically define seasons in Bangladesh. The dry season lasts from Paus (December/January) until Falgun (Febru29 For instance, if prices of the last high season are taken (Tab. 10), it turns out that those for low value rice deviate from the mean value by 12.6 percent on average, while those of high value rice deviate only by 9.7 percent on average (medium value rice types deviate on average by 14.7 percent). In case of the opening prices (Tab. 11), prices for low value rice deviate from the mean value by 7.5 percent on average, while prices of high value rice deviate only by 3.7 percent on average (medium value rice types deviate on average by 10.2 percent). 30 Parts of this section have meanwhile been published in Keck (2015).

The Food System Of Dhaka

77

ary/March). During this period, water remains only in perennial rivers, canals and beels. The pre-monsoon season, which starts in Caitra (April/March) and ends in Jyaistha (May/June), is the time of widespread breeding migration of fishes and prawn. Especially major carps migrate over long distances upstream to reach their breeding grounds. The flood season from Asarh (June/July) to Bhadra (August/ September) is the time when the fish is usually dispersed to floodplains for feeding and growing. During the post-monsoon season, that lasts from Asvin (September/October) to Agrahayan (November/December), the flooded areas start shrinking and the fish and prawn move with the water flows back into deepwater areas of rivers and canals. By following this cycle, the main season of capture fisheries in the giant floodlands starts shortly before the monsoon period in Jyaistha (May/June) and lasts more or less until Asvin (September/October). The main season of pisciculture in ponds, ditches, and baors directly follows up, starting in Kartik (October/ November) and lasting to Falgun (February/March). During the post-monsoon phase, capture fisheries in perennial rivers, beels and in Lake Kaptai come along. Marine capture fisheries have their high season mostly in between Asarh (June/July) and Kartik (October/November). Similar to the case of rice, Bangladesh’s total catch of fish shows a significant increase over the period of the last 20 years, as it is shown in Figure 10. In the early 1990s (1990–91 to 1994–95), the average total annual catch amounted to 1.0 million mt (DoF 1993; cited after Rahmen 1997; DoF 2002). Until the late 2000s (2004–05 to 2008–09) this figure had increased to an amount of roughly 2.5 million mt. For the year 2008–09, a total catch of 2.7 million mt is estimated (DoF 2007; DoF 2009). Over this whole period of time, Bangladesh exported fish that amounted to a more or less constant portion of 2.9 percent. In real terms, though, this meant an increase from an average amount of 29,700 mt (early 1990s) to an average of 70,800 mt per year (late 2000s). At that, the net availability of fish (total production minus exports) almost doubled within the same period from an average of 8.5 kg to an average of 17.3 kg per person and year. Due to additional import of frozen fish, mainly from Myanmar and India, the factual availability of fish is even higher. In the early 1990s, the largest proportion of Bangladesh’s annual catch stemmed with 31.5 percent from captures in flood lands, followed by captures in the Bay of Bengal with 24.7 percent, and from pisciculture in ponds, ditches and baors with 20.2 percent. A significant share of 14.4 percent came from capture fisheries in rivers and estuaries (incl. Sundarbans), while the share of coastal shrimp and fish farms was with a portion of 3.5 percent ranked last, even behind captures in beels and Lake Kaptai that made 5.7 percent. Within 20 years, all types of capture fisheries showed a downward trend in respect to their individual shares of the total catch. Rivers and estuaries (incl. Sundarbans) lost significance most remarkably, falling from 14.4 to 6.3 percent until the late 2000s. The already small percentage of beels and Lake Kaptai decreased from 5.7 to 3.5 percent in the same time period, and also the share of marine fish lost weight, shrinking from 24.7 to 20.0 percent. Flood land capture fisheries decreased only slightly from 31.5 to 31.0 percent. At the same time, the role of inland and coastal pisciculture

78

The Food System Of Dhaka

increased significantly. The proportion of fish that stems from ponds, ditches and baors increased from 20.2 to 33.7 percent and that of coastal shrimp and fish farms from 3.5 to 5.4 percent (cf. Fig. 10). Figure 10: Total catch and net availability of fish in Bangladesh (1989–2009) 3,000 Marine Fisheries

16

2,500

14 2,000

12 10

1,500

8 1,000

6 4

500 2

Inland Fisheries (Culture): Ponds, Ditches and Baors Inland Fisheries (Capture): Flood Lands Inland Fisheries (Capture): Beels and Lakes Inland Fisheries (Capture): Rivers and Estuaries Net Availability of Fish (in kg per capita)

-0 8

-0 6 05

07 20

20

03

-0 4

-0 2 20

01

-0 0 20

99 19

97

-9 8

-9 6 19

-9 4 93

95 19

-9 2 91

19

19

89 19

Inland Fisheries (Culture): Coastal Shrimp and Fish Farms

0

-9 0

0

N et A vailability (in kg per capita)

Total C atch (in thousand metric tons)

18

Year

Source: Own draft based on BBS (2011), DoF (1993, 2002, 2007, 2009), MoA (2007)

A more detailed view shows that, today, the dominant share of fish that is consumed in Bangladesh stems from inland pisciculture in ponds and ditches (without boars). In absolute terms, the annual catch more than quadruplicated from an average of 203,600 mt in the early 1990s to an average of 821,400 mt in the late 2000s. The increase in the total catch resulted in part from an expansion of the total water area from 146,890 ha in 1992–93 to 305,025 ha in 2008–09, but also from an intensification of production that increased from an average of 1.4 mt per ha in 1992–93 to an average of 3.0 mt per ha in 2008–09 (DoF 1993, cited after Rahman 1997; DoF 2009) (s. Fig. 11). The second largest share of Bangladesh’s fish consumption stems from inland capture fisheries in flood plains that comprise the largest water area of all sectors. Its average annual catch more than doubled in the same time from 320,400 mt to 761,500 mt, or from 113 to 269 kg per ha, as it is shown in Fig. 11. Today, catch from marine waters are ranked third with an annual amount of 514,600 mt in 2008–09. Given the vast water area of the Bay of Bengal, its increase in productivity from 32 to 62 kg per ha must be seen to be remarkable. Until today, the bulk of marine catch (456,200 mt or 92.9 percent) stems from artisanal fisheries while trawler fisheries are almost negligible.

79

The Food System Of Dhaka

Figure 11: Average annual catch of fish in Bangladesh (1989–2009) 1,000

3,500 Flood Lands

0,800

Productivity (in mt/ha)

2,500

0,600

2,000

1,500

0,400

1,000 0,200 0,500

0,000

Productivity, Ponds and Ditches (in mt/ha)

3,000 Marine Fisheries Coastal Shrimp and Fish Farms Rivers and Estuaries Beels Sundarbans Kaptai Lake Baors Ponds and Ditches

0,000

Year

Source: Own draft based on DoF (1993, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009)

Even though the share of coastal shrimp and fish farms in total production is still small, its production increase from 35,800 mt in the early 1990s to 131,600 mt in the late 2000s is striking. Similar to the case of ponds and ditches, this increase was the result of an expansion of the water area from 108,280 ha in 1992–1993 (DoF 1993, cited after Rahman 1997) to 217,877 ha in 2008–09 (DoF 2009) and an intensification from an annual catch of 312 kg per ha in 1992–93 to 668 kg per ha in 2008–09. With a quantity of 145,600 mt in 2008–09, it overtook already the amount provided by rivers and estuaries (138,200 mt), that is the only sector that shows a decline of productivity from the mid 1990s onwards (s. Fig. 11). However, the setback in 2002–03 and 2003–04 might be an early indication of the limits of coastal shrimp and fish farming that has to be seen. Today, fish from beels amount to at least 79,200 mt, while the Sundabans (18,500 mt), Kaptai Lake (8,600 mt) and boars (5,000 mt) are minor fishing grounds (DoF 2009), even though fishing was intensified. In beels and in Lake Kaptai, captures could be increased from the early 1990s to the late 2000s from 463 to 671 kg per ha and from 73 to 116 kg per ha, respectively. Fish farming in baors yielded a doubling of productivity from 353 to 853 kg per year. Also capture fisheries in the Sundarbans increased substantially from an average of 38 to 97 kg per ha (cf. Fig. 11), despite the fact that they were declared a Reserve Forest already in 1875 and became an UNESCO world heritage site in 1999 (Siddiqui 2006). From these figures, it becomes clear that – similar to the rice economy – also fishery in Bangladesh unterwent a substantial change in the last 20 years and must be seen as a success story. Nonetheless, fishery has its dark side as well. Especially the downward trend in capture fisheries in rivers and estuaries is alarming. This negative development is attributed to many factors, such as pollution through the

80

The Food System Of Dhaka

use of fertilizer and pesticides in agriculture, and through industrial effluents especially in large cities, to flood control measures, the erection of large-scale drainage and irrigation systems, and the encroachment of wetlands (Belton et al. 2011; Sultana 2012). All these features led to the destruction of sanctuaries and hinder migration of fish to their breeding grounds. At the same time, fishery itself contributes to this list of negative factors, as the conversion of wetlands into agricultural land (most notably in the Sundarbans) led to a reduction of biodiversity. The growing use of fertilizer, pesticides and vaccines in pond and shrimp farm fisheries led to residua of heavy metals in the soil and groundwater and that has given reason for public health (Ali 1997; World Bank 2006a). As opposed to the rice granaries, Bangladesh’s fishing grounds are located mainly in the southern and south-eastern part of the country (s. Map 4). Today (2001), the most productive regions (with a total annual fish catch of more than 100,000 mt) are: Chittagong (491,000 mt31), Khulna (175,000 mt), Comilla (144,000 mt), Mymensingh (139,000 mt), Barisal (110,000 mt) and Noakhali (101,000 mt). Close behind are Rajshahi (98,000 mt) and Sylhet (98,000 mt). Due to their different physical environments, each region has its particular strength in one or more fishery sectors. Thus, Chittagong as the country’s largest sea port is Bangladesh’s major source of marine fish. It also houses large ponds for aquaculture, while Khulna is considered to be the center of the country’s export-oriented prawn and shrimp industry. In Comilla the overarching share of catch stems from ponds, while in Mymensingh it stems from flood lands. Rivers and estuaries are the dominant fishing grounds in Barisal, ponds in Noakhali and Rajshahi and beels in Silhet (cf. DoF 2002). By taking the population figures per region (BBS 2007b) and the average amount of 38.5 g fish that is consumed in Bangladesh per person and day (BBS 2007c) as a basis, we can distinguish between fish surplus and deficit regions. Accordingly, surplus regions are Chittagong (374,000 mt), Khulna (95,000 mt), Barisal (43,000 mt), Noakhali (27,000 mt), Comilla (14,000 mt) and Mymensingh (13,000 mt). In all other regions, there is a notable deficit of fish. This underlines the importance of marine catch for Bangladesh to meet its national demand. By inland fisheries alone, Bangladesh could not be fed. 27.2 percent of Dhaka’s fish supplies are from Khulna, followed by 17.3 percent that are from Mymensingh. 11.6 percent are brought in from Chittagong, 10.6 percent from Barisal, 6.0 percent from Comilla and Sylhet each. Notably, fish is transported to Dhaka also from deficit regions. Almost 5.0 percent are supplied from Faridpur and Rajshahi each. Amounts of more than 1.0 percent stem from Noakhali (2.0 percent), Dhaka (1.8 percent) and Patuakhali (1.5 percent), while amounts of less than 1.0 percent 31 For the lack of available data on the annual catch of individual sea ports and landing stages, the entire catch of marine fin fish was assigned to the major harbor of the country, that is Chittagong Port. For a similar reason, the entire catch of marine shrimp catch was assigned to Mongla Port in Khulna, that is the second largest port. Landing stages of smaller size exist in Cox’s Bazar (Chittagong region), in Barisal (Barisal region), in Khepupara and in Patharghata (Patuakhali region) (Chakraborty 2006) (s. Map 4).

The Food System Of Dhaka

81

come from Jessore (0.9 percent), Bogra (0.4 percent), Jamalpur (0.4 percent), Kushtia (0.1 percent), Rangpur (0.1 percent), Pabna (0.04 percent) and Dinajpur (0.03 percent). 4.3.2 Availability of Fish in Dhaka As it was shown in Tab. 7, Dhaka City is well supplied with fish. In fact, the fish supplies equate to 169.8 percent of the national urban average, that is the highest score of all measured food types. Similar to the case of rice, the total daily fish supply to Dhaka City can be analyzed by distinguishing fish types and prices. For fish types, the categorization of the Department of Fisheries (DoF) can be used. Accordingly, from an economic point of view, the most significant inland water species are major carp (Rui, Catla, Mrigal, Kalbasu), exotic carp (Silver Carp, Grass Carp, Mirror Carp), other carp (Ghania), cat fish (Rita, Boal, Pangas, Silon, Aor, Bacha), snake head (Shol, Gazar, Taki), live fish (Koi, Singhi, Magur), inland water prawn (Golda), and Hilsha (Ilish; river catch). All other species are assigned to the miscellaneous category. Most significant marine species are Hilsha (Ilish; sea catch), marine and brackish water shrimp (Bagda), Bombay Duck, Indian Salmon, Pomfret (Rup Chanda, Hail Chanda, Foli Chanda), Jew Fish (Poa, Lambu, Kaladatina), Sea Cat Fish, Sharks, Skates and Rays (DoF 2009). On the basis of this categorization, it is possible to compare the figures of the total national fish catch with consumption figures in Dhaka (s. Fig. 12). Hilsha is commonly perceived as the national fish of Bangladesh. Nevertheless, in terms of quantity, it has been outstripped by another species in the meantime. On the national level, the bulk of catch stems from carps (major carps, exotic carps, and other carps) that are cultivated in ponds and ditches and amount together to 34.6 percent of the total national production. Hilsha is with a proportion of 11.1 percent not far ahead of marine fish (9.6 percent). The export-oriented shrimp industry experienced considerable growth rates in recent decades and ranges today third with a share of 9.1 percent. Snake head (4.5 percent), cat-fish (4.4 percent) and live-fish (2.9 percent) are less significant but still play an important role compared to the small amounts of numerous species that are concealed behind the miscellaneous category. When turning to Dhaka it can be seen that carps and Hilsha do also play the major role in terms of fish consumption there. From the total fish demand, 26.3 percent is carp fish and 15.9 percent is Hilsha. While the share of carps is below the national urban average, that of Hilsha is clearly above. These two varieties make together 42.2 percent of the total fish consumption in Dhaka, which is slightly less than the regarding proportion of the national production (45.7 percent). The demand for cat-fish is with 10.8 percent disproportionately high compared to its share of the national figure. Similarly, the consumption of prawns and shrimps (chingri) as well as live fish is above average with portions of 10.5 and 4.2 percent respectively. According to wholesalers at Dhaka’s fish markets, it is mainly the growing number of Chinese restaurants that demand for high quality chingri. The rest is said to be mainly sub-standard shrimp that could not be

82

The Food System Of Dhaka

exported. Interestingly, the demand for marine fish (2.9 percent) and snake-heads (2.0 percent) is disproportionally small. This seems odd since marine fish gains importance on the national level, but is most probably explained by the deficiency of cold stores and difficulties in maintaining cold chains that together make the long journey from the Bay of Bengal to Dhaka a hindrance. Figure 12: National fish production and urban consumption in Dhaka City (from left to right) 24%

26%

27%

Carps Hilsha

35%

Cat-Fish Prawn/Shrimps Live Fish

5%

2%

Marine Fish

3% 10%

Snake-Head 16%

4% 3%

11% 4% 9%

Source: Own draft, calculation for 2010, based on DoF (2009)

Miscellaneous

11% 11%

Source: Own draft, estimation based on survey 2009–2010

Fish can be categorized according to price in order to reveal a more detailed picture on the present situation of food consumption of different socio-economic sections in Dhaka. Going by their price levels, three categories of fish are distinguished, that are low value, medium value and high value fish (s. Tab. 13). Low value fish comprise Silver Carp, Chanda, Chapila, Magur, Poa, Rita, Pangas, Kaski and Taki. If any, it is these fishes that are most affordable for the urban poor with prices that range from 39.60 to 70.00 BDT per kg in case of small fishes (at an average of 56.70 BDT). Medium value fish types are Tilapia, Mola, Punti, Rui, Tengra and Koi with prices that range from 84.90 to 130.20 BDT per kg (at an average of 102.40 BDT). Finally, Chingri, Shol, Koral, Hilsha, Boal and Shing can be subsumed under the high value category with prices that range from 149.70 to 346.70 BDT per kg (at an average of 213.60 BDT).32 These latter fish types are far beyond the scope of the urban poor. This becomes obvious by recalling the fact that rickshaw pullers – one of the main jobs of urban poor male workers in Dhaka (World Bank 2007) – hardly earn 100.00 BDT per day when daily expenses like the fee for renting the vehicle are subducted. This situation applies not only to jobs in the so-called informal sector, but also to formal jobs in export-oriented 32 For this study, it was consciously decided to take the native fish names as they are prevalent in Bangladesh. Anyone who is interested in the biological names of these fishes, is referred to information available online (http://en.wikipedia.org/wiki/List_of_fishes_in_Bangladesh).

83

The Food System Of Dhaka

industries like the garment industry in which the majority of urban poor female workers is employed. From 2007 until 2009, when the fieldwork was conducted, the minimum wage of a female garment worker was officially fixed to the amount of 1,662.00 BDT per month. These few examples show that the majority of fish is too costly for the urban poor. Table 13: Average annual fish wholesale prices in Dhaka (2009–2010) Low value fish Silver Carp Chanda Chapila Magur Poa Rita Pangas Kaski Taki

Average price (in BDT/kg) 39.60

n

Medium value fish Tilapia

Average price (in BDT/kg) 84.90

1

45.00 45.00 52.50 58.70 63.30 67.10 68.90 70.00

1 2 1 6 1 19 3 3

Mola Punti Rui Tengra Koi

84.90 90.00 110.70 113.80 130.20

n

15

High and special value fish Chingri

Average price (in BDT/kg) 149.70

n

28

17 1 64 4 9

Shol Koral Hilsha Boal Shing

165.00 182.50 201.30 236.70 346.70

3 1 50 1 3

Source: Own draft based on survey 2009–2010 Note: 1. Average annual prices are taken to even the price variability of the individual fish types throughout the year in order to get more general figures. The average annual wholesale price is calculated by the formula: PØ = [P(th)*th]+[P(tl)*tl]/12. P(th) is the price during high season, that is multiplied by the number of months in which this season is prevalent (th). P(tl) is the price during low season, that is similarly multiplied by the number of months in which this season lasts (tl). The result is then divided by 12 month. The price dynamics at Dhaka’s rice and fish markets are discussed in full detail in the next part (s. section 4.3.3). 2. In the course of the survey, the curreny rate for 100.00 BDT changed from 0.973 EUR (15.12.2009) to 1.011 EUR (01.01.2010) to 0.984 EUR (15.01.2010) (cf. http://www. bankenverband.de/service/waehrungsrechner). By taking the mean of these three, the average rate in the time of the survey was 0.989 EUR.

As it can be comprehended by Figure 13, those types of fish that are mostly consumed in Dhaka are Rui with 18.5 percent, Hilsha with 15.9 percent and Chingri with 10.5 percent. Also Pangas (8.7 percent), Catla (5.7 percent), Tilapia (5.2 percent) and Mola (4.0 percent) take considerable proportions. Fish types with individual shares of above 1.0 percent are further Koi (2.3 percent), Poa (2.1 percent), Mrigal (1.6 percent), Tengra (1.4 percent), Shing (1.3 percent), Shol (1.2 percent) and Punti (1.1 percent). Specified fish types with proportions of below 1.0 percent are Koral, Boal, Bele and Taki (0.8 percent each), Rita (0.7 percent), as well as Air, Magur and Silver Carp (0.6 percent each). Miscellaneous fish types with individual proportions of below 0.5 percent are Chanda (mainly Rubchanda),

84

The Food System Of Dhaka

Biam, Bata, Meni, Kholisa, Pabda, Tuna, Salmon, Chewa, Kalibaus, Dhela, Gazar and others33. Figure 13: Fish consumption in Dhaka City Hilsha

Chingri

Catol/Catla

Shing

Shol

Koral

Boal

Air

Rui

Tilapia

Mola

Koi

Mrigal/Mirigel

Tengra

6%

Punti

Bele

1%

Pangas

Poa

Taki

Rita

Magur

Silver Carp

Chapila

Miscellaneous

1% 1% 14%

1%

16%

1% 1% 2% 10% 9% 1% 1% 1% 2% 2%

1% 4%

1% 5%

1% 18%

1%

Source: Own draft based on survey 2009–2010 (n=233)

By leaving the 14.4 percent of miscellaneous fish aside for which no price data is available, it can be seen from Figure 13 that the bulk of Dhaka’s fish supplies (42.9 percent) belongs to the high value category. 40.8 percent have to be addressed as medium value fish, while the share of low value fish amounts to only 16.4 percent. By performing the same sort of rule-of-the-three calculation as it was done for the case of rice, it is revealed that upon a number of 2.9 million middle class people an average of 99 g of fish per person and day was bestowed. In contrast to this figure, 2.2 million people of the upper class consume on average an amount of 135 g of fish per head per day, while the 28.0 percent of the urban poor – 2.0 million people, a figure that is worth to be repeated at this place – consume on average only 58 g per capita and day. These figures now quarry the full dimension of the prevalent socio-economic disparities in Dhaka City. They show how poverty hinders a large portion of Dhaka’s population to access Bangladesh’s main source of animal protein provision. Due to the fact that similar consumption patterns can be assumed for fresh vegetables and fruits, the case of fish 33 In the questionnaire, it was only asked for the price of the most important fish type (in terms of its respective share of the total turnover) of each individual wholesaler. Since none of the miscellaneous fish types was of main priority for any of the wholesalers, it was not possible to assign them to any of the above mentioned price categories.

The Food System Of Dhaka

85

is just one example that highlights the fact that almost one third of Dhaka’s population – mostly that part that is responsible for the so-called 3-D jobs (that stands for dirty, dangerous and demanding) – is systematically excluded from affording socially valued and physically healthy food. As such, poverty is not just an abstract idea that comes about by categorizing people. It is real conditions that factually pierce down to the very bodies of individual human beings. Even though the last famine dates back to 1974, it is this silent violence (Watts 1983) that is still widepread in Bangladesh in general and in Dhaka in particular. 4.3.3 Accessibility of Fish in Dhaka From the main fish varieties that are sold at Dhaka’s markets, seven have their high season in the first half of the (Bangladeshi) year. The remaining 14 varieties have their high season in the second half of the (Bangladehsi) year. Two temporal bottlenecks of fish supply can be identified, one from Caitra to Baisakh, and one during Asvin. As such, the monga period is not confined to rice alone, but also affects the people’s fish consumption. Likewise to the case of rice, this phenomenon manifests itself in rising prices that come along with a general lack of income opportunities and earnings that are not adjusted to these seasonal price changes. In regard to capture fisheries in rivers, these bottlenacks stem from the natural fish behavior and their cycles of migration, breeding and growing. In regard to fisheries in floodplains, they are the outcome of high precipitation during summer that let the rivers, ditches and canals overflow their banks onto low-lying areas. The lean season of pisciculture in ponds and canals during summer is related to heavy rainfalls that can lead to the overflowing of ponds and to fish diseases that are generally more prevalent at that time. During the dry winter season, it is the moderate temperatures and extreme weather phenomena like cold waves and hail storms that have negative impacts on the reproduction behavior of fish. The dimensions of the seasonal prices changes are shown in Table 15. Pangas, a low value fish that is consumed a lot, has an average wholesale price of 58.20 BDT per kg for small fishes during high season that lasts for three months from Jyaistha to Sraban. During the rest of the year, its price rises to an average of 72.50 BDT per kg, which means an increase by 24.6 percent. The seasonal price variation of other fish that is consumed in Dhaka ranges from 7.50 BDT in case of Chapila to 106.70 BDT in case of Shing or from 18.8 percent (Chapila) to 150.0 percent (Chanda). As such, Pangas is an example of medium seasonal price changes. By calculating the mean figures, it turns out that prices for low value fish during off-season are on average 54.0 percent higher than during high season. Prices for high value fish vary from low to high seasons only by 47.7 percent, and those of medium value fish by 42.6 percent.

86

The Food System Of Dhaka

Caitra (March/April)

Falgun (February/March)

Magh (January/February)

Paus (December/January)

Agrahayan (November/December)

Kartik (October/November)

Bhadra (August/September)

Sraban (July/August)

Asarh (June/July)

Jyaistha (May/June)

Baisakh (April/May)

Low value fish Silver Carp Chanda Chapila Magur Poa Rita Pangas Kaski Taki Medium value fish Tilapia Mola Punti Rui Tengra Koi High value fish Chingri Shol Koral Hilsha Boal Shing

Asvin (September/October)

Table 14: Seasonal changes in supply at Dhaka’s fish markets

Fish type

Source: Own draft based on survey 2009–2010 (n=241) Note: Squares in dark gray color indicate high season, squares in light gray color indicate mid-season, and squares in white color indicate low season.

87

The Food System Of Dhaka

Thus, it turns out that it is again the low value food that shows the most extreme seasonal price changes. In general, fish shows a seasonal price variability that is almost twice as high as that of rice.34 Table 15: Seasonal price changes of major fish types at Dhaka’s wholesale markets Average price, past high season (in BDT/kg) Low value fish Silver Carp Chanda Chapila Magur Poa Rita Pangas Kaski Taki Medium value fish Tilapia Mola Punti Rui Tengra Koi High value fish Chingri Shol Koral Hilsha Boal Shing

n

Standard deviation (in BDT/kg)

Average price, past low season (in BDT/kg)

n

Standard deviation (in BDT/kg)

25.00 20.00 40.00 45.00 44.70 50.00 58.20 56.70 53.30

1 1 2 1 6 1 19 3 3

14.10 15.40 25.60 11.50 16.10

50.00 50.00 47.50 60.00 65.00 70.00 72.50 70.00 80.00

1 1 2 1 6 1 19 3 3

10.60 30.20 30.60 17.30 0.00

66.70 68.50 60.00 93.30 83.80 106.70

15 17 1 64 4 9

16.80 17.70 24.30 35.40 23.30

91.00 90.60 100.00 118.50 128.80 148.90

15 17 1 64 4 9

19.80 23.10 28.90 54.50 45.90

122.90 130.00 130.00 140.30 170.00 270.00

28 3 1 50 1 3

86.80 65.60 74.50 166.40

163.40 28 105.90 190.00 3 55.70 200.00 1 233.80 50 136.70 250.00 1 376.70 3 92.90 Source: Own draft based on survey 2009–2010

34 The monthly price variation of fish markets is generally less distinct in case of fish as compared to rice. There is a need for fresh fish everyday that keeps people from purchasing large quantities of fish at the beginning of the month. This is the reason why fish wholesale markets run usually for seven days a week. It is only due to irregular supply cuts that individual wholesalers are forced to leave their vending sites closed for one day or another. Throughout Dhaka City, the sales volume of fish is highest on Fridays. It is the first day of the weekend and a time when mostly the male heads of housholds purchase fresh fish on the kitchen markets for dinner (that is usually prepared by the women of the household, by the wife and/or her maids). Reportedly, at that day of the week, wholesalers’ daily sales amounts increase by 25.0 to 100.0 percent as compared to the other days of the week. Prices are consequently higher due to the higher demand.

88

The Food System Of Dhaka

Table 16 presents the price variability over the course of a single day, which shows the most obvious difference of rice and fish. As a non-perishable, the opening and closing prices of rice were found to be almost identical. Fish as a perishable good, in contrast, shows distinct changes in the course of one day. Early in the morning when the markets open, the price of Pangas (small fishes) was found to be on average 72.10 BDT per kg. Not more than four hours later, the average price for the same fish went down to an average of 66.30 BDT per kg. That equals a decrease of 8.0 percent. Again, in the case of low value fish, the mean price variability in the course of the day was found to be highest. On average, the difference between opening and closing prices was found to be 15.1 percent in the case of low value fish, 8.3 percent in the case of medium value fish, and 7.7 percent in the case of high value fish. As such, daily price changes of low value fish are three quarters of the seasonal price changes of low value rice. Table 16: Daily price changes of major fish types at Dhaka’s wholesale markets Average opening price (in BDT/kg) Low value fish Silver Carp Chanda Chapila Magur Poa Rita Pangas Kaski Taki Medium value fish Tilapia Mola Punti Rui Tengra Koi High value fish Chingri Shol Koral Hilsha Boal Shing

n

Standard deviation (in BDT/kg)

Average closing price (in BDT/kg)

n

Standard deviation (in BDT/kg)

40.00 40.00 42.50 60.00 65.00 70.00 72.10 73.30 61.70

1 1 2 1 5 1 19 3 3

10.60 25.00 31.50 23.10 20.20

32.00 35.00 37.50 60.00 51.00 35.00 66.30 70.00 57.30

1 1 2 1 5 1 19 3 3

10.60 23.00 30.60 26.50 15.50

74.60 83.40 60.00 107.50 126.70 124.50

13 18 1 64 3 10

12.80 19.20 26.20 46.20 38.80

70.80 73.10 60.00 101.30 108.30 109.50

13 18 1 64 3 10

14.10 16.00 25.90 49.10 35.20

139.90 142.50 130.00 178.20 210.00 283.30

28 4 1 51 1 3

85.40 123.10 28 82.20 38.60 127.50 4 35.90 130.00 1 93.00 152.50 51 74.20 210.00 1 144.30 256.70 3 106.90 Source: Own draft based on survey 2009–2010

Changes in prices over the course of the last year (2009–2010) were again found to be ambiguous (s. Tab 18). Prices for Chapila, Pangas, Chingri, and Koral were found to decrease by an amount of 2.5 (Chingri) to 56.3 percent (Chapila). Be-

The Food System Of Dhaka

89

sides, the price for Chanda stayed unchanged. Nonetheless, the majority of fish types showed a price increase that ranges from 3.4 (Tilapia) to 34.2 percent (Poa). By taking again the mean figures, it turns out that the average price increase of the low value category is 9.3 percent, that of medium value fish is 14.3 percent, while that of high value fish is 7.1 percent. The price variability of fish in between the particular fish wholesalers is much higher than in case of rice. As the figures of the standard deviation in Table 17 show, the variability of actual prices ranges from 7.10 BDT per kg in the case of Chapila to 144.30 BDT per kg in the case of Shing. In percentaged figures, this means a standard deviation in between 17.8 (Chapila) and 62.6 percent (Chingri) of the average sales price. This variability seems to be predominantly the result of differences of fish in quality and size. This is a general problem that was attempted to be minimized in the survey questionaire by asking separately for the average price for small (up to a weight of one kg) and big examples (that exceed the weight of one kg) of the same type of fish. All prices presented so far refer to small examples of the regarding fish type. Even though this measure helped to minimize the price variability, the categorization of “small” and “big” fishes is a matter of estimation, which leaves plenty of space for interpretation. As such, it can be argued that in case of fish, the price differences and the standard deviations provide much less information about the integration of Dhaka’s fish market system, but more about the quality of fish on offer that varies from one wholesaler to another. These quality differences are greatest in the category of high value fish, where the actual price difference amounts to an average of 49.5 percent of the mean value. From all types in this group, Chingri show the highest standard deviation as this term subsumes all different types of shrimps from small prawn up to internationally traded Tiger Shrimps. But also Hilsha, Bangladesh’s national fish, and other varieties show huge price differences despite the fact that all are of the same type. In the category of low value fish, the individual prices deviate from the mean value by an average of 32.8 percent, while medium value fish deviates on average by 25.7 percent. This pattern could be observed in all cases.35

35 If prices of the last high season (Tab. 15) are taken, it turns out that those for high value fish deviate from the mean value by 59.0 percent on average, those of low value fish by 32.8 percent, and those of medium value fish by 28.2 percent. In case of the opening prices, Table 16 shows that prices for high value fish deviate from the mean value by 47.8 percent on average, while prices of low value fish deviate by 34.3 percent, and that of medium value fish by 26.4 percent.

90

The Food System Of Dhaka

Table 17: Recent price trends of major fish types at Dhaka’s wholesale markets Current average price (in BDT/kg) Low value fish Silver Carp Chanda Chapila Magur Poa Rita Pangas Kaski Taki Medium value fish Tilapia Mola Punti Rui Tengra Koi High value fish Chingri Shol Koral Hilsha Boal Shing

n

Standard deviation (in BDT/kg)

Average price 1 Year ago (in BDT/kg)

n

Standard deviation (in BDT/kg)

35.00 30.00 40.00 60.00 57.00 60.00 68.00 73.30 61.70

1 1 2 1 5 1 19 3 3

7.10 21.70 29.90 23.10 20.20

30.00 30.00 62.50 50.00 37.50 40.00 71.00 53.30 50.00

1 1 2 1 6 1 20 3 3

17.70 14.70 36.20 25.20 10.00

74.20 81.40 60.00 107.70 130.00 132.50

13 18 1 64 3 10

14.00 18.30 27.60 43.60 37.40

71.70 75.60 40.00 96.50 102.50 119.00

15 18 1 65 4 10

12.20 19.20 23.50 41.70 24.70

141.30 145.00 130.00 183.80 200.00 283.30

28 4 1 51 1 3

88.40 144.80 28 101.50 37.90 115.00 4 28.90 150.00 1 107.00 161.40 50 107.80 180.00 1 144.30 233.30 3 104.10 Source: Own draft based on survey 2009–2010

4.4 RÉSUMÉ Five findings can be drawn from this section. First: Ever since its meanwhile fourcenturies’ history, Dhaka played a prominent role for the Bengal province and is playing even more so for the young state of Bangladesh. First of all, without question, Dhaka is the political center of Bangladesh, not so much because it is the seat of government, but more so as it was the main stage of the war of independence. Second, the city is one of the major economic engines of Bangladesh. Since independence, the country opened up for the global market with garment industries playing the most important role – most prominently located in and around the metropolis. Third, the large university campus and the number of private universities, the museums, exhibitions and art performances, the mosques and the outdoor sports and recreational facilities prove Dhaka to be also the cultural heart of the country. These facts together bring to mind the city’s extraordinary functional primacy that it holds in Bangladesh today.

The Food System Of Dhaka

91

Second: Given widespread poverty in Bangladesh, the lack of employment opportunities in the countryside and the seasonal loss of arable land due to heavy monsoon rains, subsequent floods and severe river erosion, the megacity has become the only viable destination for many Bangladeshis. Thus, vast numbers of people migrate to Dhaka every year in order to find a job and to get at least minimum access to basic public services, especially to education and health. These masses of migrants made Dhaka over the last decades to become a primate city also in demographic terms, a fact that challenges not only urban planners, the municipality and the national government, but first and foremost the people who need to live with these conditions. Due to its exceptional growth rates in recent decades, Dhaka needs to be attached great importance that reaches even beyond the case of Bangladesh but holds significance for urban planners of the entire Global South. Third: Bangladesh is a natural habitat for rice and fish due to its location in the world’s largest river delta with highly fertile alluvial soils, its thousands of ponds, canals, rivers, haors and beels, and its vicinity to the ocean. Nevertheless, as the figures above showed, in recent years, Bangladesh’s natural suitability successively made room for another driving force of food production that is green revolution technologies. The impressive increase in rice yields, that can be noticed over the course of the last decades, could not have been achieved without extensive irrigation and the widespread adoption of fertilizer and insecticides. Likewise, with the breeding and rearing of carps, it is the resource and technology intensive inland water fisheries in ponds and ditches that need to be addressed as today’s major source of fish in Bangladesh. These factors have enabled Bangladesh to meet its rice and fish demand by own production. Due to frequently occuring natural disasters like floods, cyclones, river erosion, droughts, cold waves and hail storms, Bangladesh has not yet become fully self-sufficient. The country is and will depend on imports and the international market. But its overall situation in terms of availability of rice and fish has never been better – albeit with all negative consequences for ground water, soils and biodiversity. Fourth: Dhaka is well supplied with food. The megacity must be seen as an island of plenitude. Nonetheless, food security in Dhaka is far from being guaranteed. Still, more than a quarter (28.0 percent) of Dhaka’s population lives in a situation of severe poverty and undernourishment while, at the same time, considerable wealth and opulence exists. The consumption figures reflect such socioeconomic disparities in Bangladesh’s capital. Only 391 g rice is consumed by lower income groups on average per person and day. Even though this figure is above the national urban average, it is much less than the 455 g rice that are available for higher income groups, especially when taking into account that food grain is often the main (if not the only) diet of the poor, while the better-off are in the position to supplement rice with superior valued and more expensive items. This contradictory situation becomes even more manifest when the case of fish is taken. In fact, the average amount of fish that the urban poor consume makes up less than half (58 g per person and day) the amount that bestows to upper class people

92

The Food System Of Dhaka

(135 g per person and day). As the list of individual prices showed, the majority of fish is simply out of reach for the urban poor. Fifth: The rice and fish price dynamics indicated that monga is not a phenomenon of rural areas in Bangladesh alone. The urban monga in Dhaka can be understood as a seasonally re-occuring period of food insecurity that arises from an increase in food prices by simultaneously invariable earnings and income that most severely threatens the low income groups and the urban poor. For rice, it could be confirmed that the periods of high season are generally shorter in case of low value rice than in case of high value rice. In other words, sufficient supplies of low value rice and prevalent low prices are more seldom than in case of high value rice. Furthermore, prices for low value rice change more intensely in between high and low season, and they faced a stronger general increase in the years from 2009 to 2010 as compared to high value rice. Likewise, in case of fish, prices for low value fish do change most intensely in between high and low season, and during the course of a day. Low value fish did increase from 2009 to 2010 more intensely than high value fish and showed a larger degree of price variability than medium value fish. Thus, it can be stated that of all people who live in Dhaka, it is the urban poor who are mostly threatened by high prices and a comparatively high temporal variability.

5. FOOD WHOLESALE MARKETS IN DHAKA

“[T]o make available a more balanced and nutritious diet largely depends on how markets function.” NFPCSP (2007: 57)

In this section I am going to change the perspective and introduce Dhaka’s food markets and its actors to the reader. In the first part, I will show how the numerous wholesale markets for food have emerged in Dhaka in the past century and present some figures on basic characteristics of these markets. Subsequently, I will outline Dhaka’s value chains36 for rice and fish in order to identify the position that wholesale traders assume in the networks of the urban food system. After this rather abstract description I will close this section by giving more concrete information on the very people who are involved in food wholesale trading in Dhaka and on their business organizations. My guiding questions in this section are: How have Dhaka’s food wholesale markets emerge over time? What are the basic characteristics of these markets today? Who are Dhaka’s food wholesale traders? And what role do they play for the urban food system? 5.1 DHAKA’S FOOD WHOLESALE MARKETS 5.1.1 The Emergence of Dhaka’s Food Wholesale Markets Two factors are decisive for understanding the development of Dhaka’s food wholesale markets from the city’s century-old past to the present. The first one is changes in Bangladesh’s freight transportation system, that can be divided into three development stages, while the second one is the ever more noticable intervention of the municipality in urban development affairs. The first stage of Bangladesh’s transportation system was predominant more or less from the early times of the city until the end of the 19th century. In this period, Dhaka’s food provision relied almost entirely on inland water transport. This explains not only the location of ancient Dhaka, but also the places of the city’s first wholesale markets. One of Dhaka’s rice importers, owner of several wholesale stores and executive member of the market association of Badamtuli explained:

36 In this study I use the terms „value chain“ and „supply chain“ synonymously for variety.

94

Food Wholesale Markets “Noone really knows the age of Badamtuli. But it is said that there was a river port already during Mughal times. At that time, it was mainly aman rice from Barisal that was transported along Buriganga river and came to Dhaka by boat. The rice was directly sold at the river port. This is how the rice market of Badamtuli came into existence.” Source: Own interview (06.12.2007)

In line with this statement, all but one of the wholesale markets that were established before 1900 are located in Old Dhaka at the bank of Buriganga. The markets with the oldest tradition are Shyam Bazar, Kaptan Bazar, Swarighat Fish Market, Chowk Bazar, Moulavi Bazar, Badamtuli, Begum Bazar, and Kawran Bazar. At least four of them could be verified: Moulavi Bazar was established in commemoration of Moulavi Abdul Ali between the end of the 18th and the beginning of the 19th century in direct vicinity of Chowk Bazar (Khan 2006). Swarighat – the oldest fish wholesale market of Dhaka – was officially founded in 1850 (Rahman 2006d). And the rice wholesale market of Badamtuli was official inaugurated in 1925 (Rahman 2006b). Kawran Bazar, finally, seems to be an exception in this context, as it was established in the late 18th century by the Marwari tradesman Kawran (Karwan) Singh far away from Buriganga river on the outskirts of the main settlement area of Dhaka (Rahman 2006c). However, one needs to know that, at that time, the city was crisscrossed by navigable waterways linking the Buriganga with the Balu River so that also in this case most goods were transported by boat to the market. One of these waterways still begins right at Kawran Bazar, which is the area that has recently been restructured under the title “The Hatirjheel Project”. In 1862, the second development stage started when the Eastern Bengal Railway inaugurated the Calcutta-Ranaghat section and its continuation, the Darsana-Jagati section, which meant the beginning of the history of railway in British-Bengal. In 1885, the Narayanganj-Dhaka-Mymensingh railway line was opened under the title of Dhaka State Railway (Fida 2006). Until 1916 the railway network covered substantial shares of the north-western part of Bangladesh mainly for transporting raw jute from Dhaka, Mymensingh and Rangpur to the inland port of Narayanganj and then onwards by river to Calcutta (Fida 2006; Van Schendel 2009). As a side effect, new regions – such as Dinajpur, Rangpur and Bogra (cf. Map 3) – became accessible to supply Dhaka with rice. As it was shown above (cf. 4.2.1), it is the former jute production areas that are today’s rice granaries of Bangladesh. In the second half of the 20th century, Kawran Bazar was relocated to an area along the railway line, where it continues to be located to this day (Rahman 2006c).37 This relocation must be seen as an expression of this second development stage. However, it must be mentioned that the railway was never able to replace the country’s inland water transport system. Dhaka never became a junction of the national railroad network, which is the reason why rail-bound transport affected the capital’s food system only marginaly.

37 The former location of Kawran Bazar is not indicated in Map 2. This is the reason why in the map Kawran Bazar appears to be founded not before 1960.

Food Wholesale Markets

95

The third development stage cannot be precisely fixed in time, but started around the second half of the 20th century when the government took initiative to develop the road network with support of international donor agencies. The Jamuna Multi-Purpose Bridge serves as the most prominent example of that phase that was completed in 1998 to connect the districts of Tangail and Sirajganj. Today, it is this bridge that allows the transport of boro rice from north-western regions to Dhaka and that serves as the infrastructural backbone of the city’s rice supply. This third phase made it possible for food wholesalers in Dhaka to establish markets all over the city since it was not necessary anymore to be located at the bank of a river or at the railway line. Even though inland waterways and the railroad are still in use for transport, today, at least 80.0 percent of food reaches the megacity by truck. Against the background of these long-term trends in transportation, Dhaka’s food wholesale markets evolved step-wise by keeping pace with the municipality’s efforts to develop certain areas, and with the general spatial expansion of the city. The DCC completed the construction of New Market in 1954. Around the same time, Motijheel and Tejgaon were successively developed to become commercial centers of the city and until the mid-1960s, not only Mohammadpur, Dhanmondi and Mirpur, but also Gulshan, Banani (west of Gulshan), Baridhara (north-east of Gulshan) and Uttara were developed into planned residential areas (Chowdhury 2006). During all of Dhaka’s physical expansion, Old Dhaka remained as the center of food wholesale trade. In fact, at this time, a densification process of Dhaka’s markets took place: While Shyam Bazar (1950), Kaptan Bazar (1960, 1968) and Badamtuli (1967) were extended for the sale of additional products, new wholesale markets were established like Nazira Bazar (1950) and Lalbag Fort Mor Bazar (1964). Only Kawran Bazar Fish Market, that was established in 1960, was located outside of the old part of the city (s. Map 2). After independence, Dhaka’s food wholesale markets successively became more dispersed and differentiated. In the 1970s, the densification process in Old Dhaka continued with market extensions at Shyam Bazar (1976) and Kaptan Bazar (1978) and with the founding of a new market at Sadar Ghat (1976). But simultaneously, the municipality allowed for the rapid physical expansion of the city and took initiative by establishing markets at the fringes of the city of that time, i.e. the Governmental Hawkers’ Market in Tejgaon (1972) and Krishi Market in Mohammadpur (1978). New Market, that had been completed in 1954 (Chowdhury 2006), got a wholesale market for fish in 1979. With Merul Badda Fruits Market (1972), the second private wholesale market was established outside of Old Dhaka (s. Map 2). From the early 1980s onwards, the earlier planned areas of Gulshan, Banani, Baridhara and Uttara became inhabited and the airport was shifted further north. Low-lying areas between Khilgaon and Badda in the East, at Kamrangir Char in the South and around Adabar in the West of the DCC area were filled with soil to make the area inhabitable (Chowdhury 2006) – a process that is still going on. In that time, the municipality focused on the development of Kawran Bazar. In addition to that, further markets were established on a private basis, notably at the out-

96

Food Wholesale Markets

skirts of the city, such as Mohakali Puran Bazar (1980), Merul Badda Fish Market (established as a vegetables market in 1980), Fakirbari Bazar (1985), Mugdapara Boro Bazar (1987), and Narda Bazar Rice Market (1988) (s. Map 2). After Kawran Bazar, the DCC turned its focus towards Gabtoli, Mirpur, and Kochukhet. In 1990, a huge area was assigned at Gabtoli to become the city’s major market for live cattle. Every year, before Eid ul Azha, thousands of he-goats and bulls are brought in to this place to be sold to those families who celebrate the festival of sacrifice in Dhaka. At Kochukhet, a market complex for selling rice, edible oil and live cattle was completed in 1996 in order to provide proper food supply for the cantonment. Finally, food wholesale markets were established in Mirpur (Mirpur 1 in 1992, Mirpur 2 in 1997, Mirpur 10 in 1997, and Mirpur 11 in 1997). And also private markets emerged (Chowdhuripara Rice Market in 1990, West Nakhal Para Bazar in 1994, Mugdapara Fish Market in 1996, Boro Bazar in 1997, Malibag in 1998 and Jatrabari in 1990) (s. Map 2). Apart from a fish market that was established at Kochukhet in 2004, five areas can be identified that represent the latest spots of market development in Dhaka: 1) In Hazaribagh in the south-western part of the city, the three new markets Hazi Sayed Hossain Fruits Market (2000), Nawabganj Rice Market (2000) and Khalil Sarder Krishi Market (2003) were established. 2) The area between Badda and Khilgaon in the East saw the establishment of Moddo Badda Bazar (2000), Rampura Bazar (2000) and Sipahibag Bazar (2005). 3) At Jatrabari in the Southeast, markets for vegetables (2000), for fish (2001) and for live cattle (2006) emerged. 4) In the western part of the city, in the South of Gabtoli, the DCC established Dipjol Bazar (2003) that accommodates markets for vegetables, fish and fruits. 5) And in Uttara in the North of Dhaka, the DCC established Azampur Rail Gate Fish Market (2003) and a Bangladesh Rifles38 (BDR) market for rice (2007), while Sultan Shah Fish Market (2006) and Tarun Mahmud Shah Fish Market (2008) were established privately (s. Map 2).

38 The Bangladesh Rifles (BDR) – like its predecessors, the East Pakistan Rifles (established in 1947) and the Eastern Frontier Rifles (established in 1920) – is the frontier protection force of Bangladesh designed to be vigilant against illegal intruders and to support the government in maintaining law and order in the country (Uddin 2006). In recent times, the para-military organization became active in establishing food markets in Dhaka and elsewhere in order to support the government to care for providing affordable food to people in need.

Food Wholesale Markets

97

5.1.2 The Characteristics of Dhaka’s Food Wholesale Markets Today, there are 87 food wholesale markets within the DCC area. Out of them, 74 are pure wholesalers’ markets, while 13 are mixed markets with wholesalers and retailers at one place or traders who run wholesale and retail business at the same time (s. Map 5). The markets for rice and fresh fish are the most numerous with 24 and 13 markets, respectively. In accordance to the fact that rice is usually served with various types of lentils (dal), and due to the importance of meshed fresh spices (bata moshla) for the Bengal kitchen, markets for pulses and spices are there third most. The nine fruits markets follow close behind out of which some are specialized on bananas only. There are seven markets for vegetables, six for edible oil and for live cattle, five for eggs and three for meat (mainly beef, mutton, and chicken). The two wheat markets, that share the last rank together with the two markets for dried fish, give a hint to the marginal role that this cereal plays for the people of Dhaka, despite the long history of food aid. According to the survey, there is an estimated number of 4,410 traders who are active in the wholesale business in Dhaka City at present. Most numerous are business organizations for selling fresh fish (898) and vegetables (834), followed by those for rice (775), live cattle (570), and fruits (566). Pulses and spices (345), edible oil (202) and eggs (129) are traded by a much smaller number of wholesalers. The numbers of wheat (33), meat (31) and dried fish (27) merchants are smallest (s. Tab. 18). Table 18: Characteristics of food wholesale markets in Dhaka City Food type

Total no. of markets

Total no. of wholesalers

Rice Fresh Fish Pulses/Spices Fruits Vegetables Edible Oil Live Cattle Eggs Meat Dried Fish Wheat Total

24 13 10 9 7 6 6 5 3 2 2 87

775 898 345 566 834 202 570 129 31 27 33 4,410

Average Total no. of Average daily market days daily turnover business (in mt) hours 12 7,979 3,010 4 5,110 707 11 3,234 1,054 14 3,181 366 18 2,920 3,065 13 2,086 353 14 2,190 910 13 1,773 69 12 678 190 12 730 22 14 678 61 12 30,559 9,807 Source: Own draft based on surveys 2008–2010

Note: Regarding the total number of wholesalers and the total daily turnover, the figures of rice and fish are the result of a triangulation of data of three surveys, i.e. the food wholesale market survey 2009, the rice and fish panel survey 2008–2009, and the rice and fish market survey 2009–2010. All other figures are based on the food wholesale market survey 2009 only. In order to provide maximal accuracy in terms of the total daily turnover per food type, figures of considerabel size (pulses/spices, fruits, vegetables and edible oil) were reduced by an average error ratio of 20.0 percent. This ratio resulted from the above mentioned triangulation.

98

Food Wholesale Markets

The daily opening hours of markets for selling rice, wheat, pulses and spices, edible oil, eggs, meat and dried fish range from 8 to 17 hours a day. The time range of vegetables, fruits and live cattle markets is even larger. While some individual markets are open only in the morning for two to four hours, other markets run for 24 hours everyday. Those “twenty-four-seven” markets are the vegetable and fruit markets at Kawran Bazar, the vegetable market at Shyam Bazar, the market for spices at Kaptan Bazar, and the markets for vegetables, fruits and live cattle at Jatrabari. With an average of only four hours per day, fish markets are the exception. The main reason for that is the lack of adequate cold storage facilities at those markets that force the traders to sell their loads within the shortest possible time. Nevertheless, in general, one can speak of comparatively long market days in Dhaka, a fact that entails a high amount of hours of work for the involved people and that demands extreme fitness especially from the countless laborer without whom those markets would not function. Altogether, Dhaka’s wholesalers provide food to the city at more than 30,000 market days per year (s. Tab. 18). At all these markets together, every single day, the substantial quantity of 9,807 metric tons (mt) is realized (s. Tab. 18).39 Map 5 shows the food wholesale markets of Dhaka in a spatial perspective and makes it possible to identify the major food hubs of Dhaka. Accordingly, Chowk Bazar is with a number of 150 merchants and a daily turnover of more than 250 mt the largest market for pulses and spices. The major single markets for fruits are Jatrabari, Khalil Sarder Krishi Market and Sadarghat with a daily turnover of 50 to 100 mt and approximately 100 wholesalers each. Nevertheless, it must be kept in mind that there are two fruits markets at Kawran Bazar, one for various types of fruits and one that trades only bananas. These two markets have a daily turnover of more than 100 mt together that are marketed by 120 wholesalers. This makes Kawran Bazar the major hub for fruits in Dhaka City. In terms of vegetables, there are four major spots, namely Kawran Bazar, Jatrabari, Shyam Bazar and Notun Bazar. At each of these markets, amounts of more than 250 mt of vegetables are sold every day. While there are more than 250 wholesalers at the first three of these markets, only 60 merchants run their business at Notun Bazar. The major wholesale market for edible oil is Shyam Bazar in Old Dhaka, where about 100 merchants sell slightly more than 250 mt oil (mainly vegetable oil) per day. Gabtoli is the major market for live cattle with more than 250 wholesalers selling several thousand animals per day, ranging from poultry over goats to buffalos and cattle. The major hub for eggs is Kawran Bazar with a daily turnover of 25 to 50 mt. Out of the three meat wholesale markets in Dhaka, Kaptan Bazar is with a daily turnover of 100 to 250 mt clearly in the leading position. The main wheat wholesale market is Moulavi Bazar with a daily turnover of slightly more than 50 39 This calculation includes both live cattle and meat, which means in fact a double counting, since many animals are sold from the cattle market at Gabtoli to the meat markets in southern Dhaka. In this table, however, both figures are included as they better reflect the real workload of the traders in the markets. A more precise calculation in terms of available food in Dhaka can be found in 4.2.2.

99

Food Wholesale Markets

mt and a number of 30 merchants. Eventually, the major market for dried fish is located at Jatrabari. Given the fact that consumption of dried fish is generally low in Dhaka, the 20 wholesalers reach considerable turnovers that together amount to 20 mt per day. For rice and fish, even more precise information can be provided. Two rice markets are outstanding in terms of size, that are Badamtuli/Babubazar and Krishi Market. Badamtuli is the largest rice market with an average turnover of 880 mt per day. Krishi Market is with 700 mt per day the second largest market. At both markets together, 52.3 percent of the total rice turnover of Dhaka City is traded – 29.2 percent at Badamtuli and 23.1 percent at Krishi Market. Of considerable size are further the BDR wholesale market in Uttara (7.5 percent), and the rice markets in Kochukhet (5.7 percent), Mirpur 1 (4.4 percent), Uttar Badda (4.1 percent), Chowdhuripara (3.8 percent) and Malibag (3.8 percent). Jatrabari (2.6 percent), the Governmental Hawkers Market in Tejgaon (2.5 percent), Lalbag Fort Mor Bazar (1.9 percent), Fakirbari Bazar (1.8 percent), Mugdapara Boro Bazar (1.5 percent), and the rice market in Mirpur 10 (1.4 percent) follow, while all other markets are small in size with shares of below 1.0 percent (s. Fig. 14). 45.7 precent of all rice wholesalers in Dhaka run their businesses at the two largest markets of Badamtuli and Krishi Market. There are approximatly 200 wholesalers (25.9 percent) at Badamtuli and 150 at Krishi Markets (19.7 percent). 32.9 percent of Dhaka’s rice wholesalers are located at medium size markets (Kochukhet 8.4 percent, BDR Wholesale Market 7.7 percent, Mirpur 1 4.5 percent, Jatrabari 4.5 percent, Chowdhuripara Market 3.9 percent and Malibag 3.9 percent). The remaining 21.4 percent of wholesalers have their stalls at one of the smaller markets (s. Fig. 15).

Badamtuli/Babubazar Kochukhet Chowdhuripara Market Govt. Hawkers Market Tejgaon Mugdapara Boro Bazar Madertek Bazar Mirpur 2 Boro Bazar

Krishi Market Mirpur 1 Malibag Lalbag Fort Mor Bazar Mirpur 10 Nawabganj West Nakhal Para Mohakali Puran Bazar

10

0%

% 90

% 80

% 70

% 60

% 50

% 40

% 30

% 20

% 10

0%

Figure 14: Proportion of total rice turnover in Dhaka City by market

BDR Wholesale Market Uttar Badda Market Jatrabari Fakirbari Bazar Sipahibag Hotel Bazar Kawran Bazar Railway Market Modda Badda Bazar Narda Bazar

Source: Own draft based on survey 2009–2010

100

Food Wholesale Markets

Badamtuli/Babubazar BDR Wholesale Market Chowdhuripara Market Uttar Badda Market Fakirbari Bazar Sipahibag Hotel Bazar West Nakhal Para Mohakali Puran Bazar

Krishi Market Mirpur 1 Malibag Lalbag Fort Mor Bazar Mugdapara Boro Bazar Madertek Bazar Boro Bazar Modda Badda Bazar

10 0%

90

80

%

%

% 70

% 60

50

40

%

%

% 30

20

%

% 10

0%

Figure 15: Distribution of rice wholesalers by market

Kochukhet Jatrabari Govt. Hawkers Market Tejgaon Kawran Bazar Railway Market Mirpur 10 Nawabganj Mirpur 2 Narda Bazar

Source: Own draft based on survey 2009–2010

0% 10

% 90

% 80

% 70

% 60

% 50

% 40

% 30

% 20

% 10

0%

Figure 16: Proportion of total fish turnover in Dhaka City by market

Kawran Bazar

Jatrabari

Merul Badda

Swarighat

Sultan Shah

Mugdapara

New Market

Tarun Mahmud Shah Market

Kochukhet

Azampur Rail Gate Market

Mirpur 1

Dipjol Bazar

Kaptan Bazar

Source: Own draft based on survey 2009–2010

0% 10

% 90

% 80

% 70

% 60

% 50

% 40

% 30

% 20

% 10

0%

Figure 17: Distribution of fish wholesalers by market

Kawran Bazar

Jatrabari

Merul Badda

Sultan Shah

Swarighat

Mugdapara

New Market

Tarun Mahmud Shah Market

Kochukhet

Kaptan Bazar

Azampur Rail Gate Market

Dipjol Bazar

Mirpur 1

Source: Own draft based on survey 2009–2010

Food Wholesale Markets

101

For fresh fish, the largest markets are Kawran Bazar with a daily turnover of 220 mt (30.4 percent), Jatrabari with 180 mt (24.7 percent) and Merul Badda with 110 mt (15.1 percent). At these three markets together, wholesalers manage 70.2 percent of Dhaka’s total fish supply. Of considerable size are further Swarighat (6.9 percent), Sultan Shah (6.5 percent) and Mugdapara (5.9 percent), New Market (2.8 percent), Tarun Mahmud Shah Fish Market (2.2 percent), Kochukhet (1.8 percent), and Azampur Rail Gate Market (1.8 percent). Mirpur 1, Dipjol Bazar, and Kaptan Bazar play only marginal roles (s. Fig. 16). If the number of wholesalers per market is taken as criterion, the situation looks slightly different, as there are only two outstanding markets, i.e. Kawran Bazar (29.5 percent) and Jatrabari (28.7 percent). 58.2 percent of all fish wholesalers run their business at one of these two markets. With 8.8 percent of all wholesalers, Merul Badda is one of the medium size markets together with Sultan Shah (8.6 percent), Swarighat (7.0 percent), Mugdapara (3.8 percent) and New Market (3.3 percent). These markets together comprise 31.5 percent of all fish wholesalers. 10.2 percent have their stalls at the remaining six markets (s. Fig. 17). Map 6 and Map 7 in the appendix depict the dense webs of channels through which loads of rice and fish flow from the markets into the intra-urban distribution system. The sum of these radial resource streams allows to understand the physical extent of the catchment areas of the mapped markets, that are confined to circles or semi-circles with a radius of between one and five kilometers, with differences that can be explained by the different sizes of the markets. In the case of rice, it can be seen that the resource streams that originate from the markets follow more or less the mosaic pattern of the population density in Dhaka. As such, the largest portions of rice are delivered to the most densely populated areas that together form a sort of spiral from Khilgaon, Sabujbag and Motijheel in the East, down to Jatrabari, Old Dhaka, Sutrapur and Lalbag in the South, over New Market and Hazaribag as well as Mohammadpur and Hazaribag in the West, up to Mirpur, Karful and Pallabi in the North. The distribution pattern of fish does also follow the spatial distribution of Dhaka’s population. Nevertheless, when comparing both maps, there is an obvious difference in the supply of northern and northeastern parts of Dhaka with rice and fish. In fact, Mirpur and Pallabi in the Northwest, the area in between Gulshan and Khilket in the Northeast, as well as Uttara in the North of the city are well supplied with fish but much less with rice. This difference can only be explained by the purchasing power of the local population. In fact, all of the mentioned quarters are residential areas of the upper and middle class who can afford to buy fish.

102

Food Wholesale Markets

5.2 DHAKA’S VALUE CHAINS FOR FOOD40 5.2.1 The Value Chain for Rice In this study, I define a value chain as a “network of organizations that are involved, through upstream and downstream linkages, in the different processes and activities that produce value in the form of products and services in the hands of the ultimate consumer” (Christopher 1998: 15). Basically, value chains consist of resource flows from the producer down to the consumer, and of cash flows that are conversely directed from the consumer to the producer. The value that is produced through involved activities can be measured as the economic rent that is earned at each step of the chain (Porter 1985). However, the analysis of value chains does not only focus on the flow of products and services along the chain, but also incorporates the study of the flow of information and knowledge and of power relations within the chain (Gereffi 1994; Sheffi 2005). Against his background, the aim of this section is to portray Dhaka’s supply chains for rice and fish in order to identify the most powerful actor groups within the chain in regard to price determination processes.

Photo 1: Paddy nurturing fields Kushtia Municipality, 17.12.2007 Source: Photo by author

Photo 2: Irrigated paddy fields Jessore, Boro Bazar, 12.02.2009 Source: Photo by author

Farmers constitute the primary link in the value chain for rice. According to official data, 79.4 percent of all farm holdings in Bangladesh are smaller than one hectare (2.5 acres), a ratio that also holds true for the granaries of the country. 18.0 percent have a farm size of one to three ha, and only 2.6 percent are larger than three ha (MoA 2007). Given the absence of large producer associations, these figures prove Bangladesh’s agriculture to be clearly dominated by peasants (krishok), who usually produce in part for their own consumption (subsistence) and in part for the market (s. Box 1). Until today, there is only little contract farming in Bangladesh.

40 Parts of this section have meanwhile been published in Keck et al. (2014) and Keck (2015).

Food Wholesale Markets

103

Box 1: A paddy farmer in Jessore Jalal41 owns 20 bikha (one bikha equals 0.3 acre or 0.125 ha) of land in a small hamlet not far from the Indian border in the district of Jessore in western Bangladesh (s. Map 3). Half of this holding is currently cultivated by himself and his family. The other half is leased out to sharecroppers. Jalal provides me an overview of current paddy production patterns in Bangladesh and its cost factors. He explains that there are basically two types of paddy available, namely hybrid paddy (HYV) and domestic (deshi) paddy (local varieties). For cropping one bikha of land for one season with domestic paddy, Jalal calculates the total production costs at 3,000.00 Bangladesh Taka42 (BDT) for the land lease, 300.00 BDT for seeds, 2,500.00 BDT for fertilizer (mainly TSP, MP and Uria, but also ammonium sulfate and zinc) and insecticides, 2,000.00 BDT for irrigation (mostly for diesel for the pumps), and 2,800.00 BDT for seasonal laborer (who help carrying out the preparing and ploughing of the field, the implanting of the seedlings and the harvesting of the paddy, s. Photo 3). He estimates the harvest to be roughly 850 kg paddy. Accordingly, this meant total production costs of 10,600.00 BDT per bikha or 12.50 BDT per kg paddy. In the case of hybrid paddy, production costs add to at least 12,000.00 BDT per bikha, he says. But due to higher output rates of about 1,100 kg paddy per bikha, the final production costs of hybrid paddy at 10.90 BDT per kg are lower.

Photo 3: Seasonal labor, paddy harvest Kusthia Municipality, 17.12.2007 Source: Photo by author

Photo 4: Paddy threshing cycle Kushtia Municipality, 17.12.2007 Source: Photo by author

Despite the fact that hybrid paddy is more profitable, most of Jalal’s tenants grow domestic paddy. The reason behind is the higher input costs. First, purchasing costs of hybrid seeds are higher: While current prices for domestic seeds are 660.00 to 670.00 BDT per maund (37.3 kg), hybrid seeds are with 750.00 to 800.00 BDT per maund much more expensive. Beyond that, domestic seeds can 41 The use of the first name of my interview partners is not meant to be disrespectful, but serves the purpose of anonymization. 42 At the time of the interview, 100.00 BDT equated 1.102 EUR (cf. http://www. bankenverband.de/service/waehrungsrechner).

104

Food Wholesale Markets

be reproduced for up to five years, while hybrid seeds need to be purchased for each season anew. Eventually, according to his calculation, hybrid paddy needs about 40.0 percent more fertilizer and insecticides. That is the reason why Jalal plants not only hybrid seeds from companies such as East West, Namdhari, BRAC seeds and others that are available at the local market. Even though he has got the financial resources, he grows both hybrid and domestic paddy. “Domestic rice tastes much better than hybrid rice”, he explains. “That’s why I sell only hybrid rice. For my own consumption, I still grow domestic rice.” Source: Own interview (12.02.2009)

Photo 5: Paddy steamer Kushtia Municipality, 17.12.2007 Source: Photo by author

Photo 6: Chatal Kushtia Municipality, 17.12.2007 Source: Photo by author

In the harvest time, the peasants thresh the sheaves on the farm by means of manually driven or motorized threshing cycles (s. Photo 4). While the remaining spires are used as straw and fodder for livestock, the paddy (dhan) is packed into sacks and sold on weekly markets (hats). Since the individual amounts of the peasants are small in size, an intermediary, the so called faria, purchases paddy from numerous peasants and transports the bags to the miller. In some cases, there are additional intermediaries, so called beparis, who are located in between farias and millers. Beparis are professional traders who own storage capacities and do business by purchasing paddy from producer markets in district towns (mokam) and by selling it to millers. Farias do usually not command own godowns and are thus not in the position to pursue trading business at a larger scale. They are independent businessmen, but work as commission agents of either millers or beparis. The milling process comprises basically three steps: 1) the parboiling of paddy; 2) the drying; and 3) the milling itself. First, the grain is filled in round metal drums (so called steamers) in which steam is piped into, whose pressure can be adjusted according to needs and wishes. In these steamers (s. Photo 5) the parboiling takes place. Afterwards, the paddy is put into small basins aside to cool down. It is siphoned and dispersed on large concrete yards (chatals) (s. Photo 6) for drying. Then the paddy is ready for milling, a process that again comprises two operations. In a first step, the husk (tush) is separated from the grain, and in the second step, the brown membranes (kura) enclosing the grain are peeled off (s. Photo 7,

Food Wholesale Markets

105

8). After these two steps, the paddy (dhan) has turned into milled rice (chal). Since the demand for white and short grain rice is high in Dhaka, in some cases, the rice is polished and broken in order to improve its marketing potentials. Furthermore, discolored grains are sorted out (s. Box 2 and the Photos 9).

Photo 7: Rice milling machine Kushtia Municipality, 17.12.2007 Source: Photo by author

Photo 8: Tush and kura Kushtia Municipality, 17.12.2007 Source: Photo by author

Box 2: A rice miller in Kushtia Bilal, the oldest son of the owner of a large rice mill in Kusthia (s. Map 3), is in charge of the daily business of the mill and monitors every step of his 16 employees who currently work for him. With a maximum capacity of 40 mt per day, the fully automatized mill is one of the larger ones in rural Bangladesh. Around the year, the business volume amounts to approximatly 15 truckloads or 300 mt per month. Generally, there are two types of paddy. The first one is boiled before the milling process, that is called Sidho. The second type, called Atop, is only sun dried before processing. Today, most of the paddy is pre-boiled. Usually, the rice mill produces the varieties Mota Chal, Pariji, Paijam and Katari Vogh that are brought in from the northern regions of Rajshahi and Dinajpur, but at the moment Miniket and BR 29 is milled, which stem from Jessore (s. Map 3). All milled rice is brand-named and sent to Dhaka, where it is sold by a selected number of commission agents at Jatrabari, Babubazar and Malibag (s. Map 5). In times of shortages, the millers do also trade with imported rice. “In the rice business”, Bilal explains, “there is no waste. Every single byproduct has its specific purpose”. Accordingly, the millers sell tush as fuel for 65.00–80.00 BDT43 per maund and kura as fish and cow fodder at a rate of 200.00 43 At the time of the interview, 100.00 BDT equated 0.990 EUR (cf. http://www. bankenverband.de/service/waehrungsrechner).

106

Food Wholesale Markets

BDT per maund. Furthermore they sell the by-products that are produced by their grain polishing and sorting machine, that is discolored rice (mora chal) and split rice grain parts (kuth), for 14.00 BDT per kg (s. Photo 10).

Photo 9: Rice sorting machine Kushtia Municipality, 17.12.2007 Source: Photo by author

Photo 10: Mora chal (in the middle) Kushtia Municipality, 17.12.2007 Source: Photo by author

Their most prominent expenses are the monthly electricity bills. Due to the large machines they amount to 1.5 lakh (150,000.00) BDT per month. Further costs arise from land lease and debt servicing for the machinery (85,000.00 BDT per month), the salary of their employees (64,000.00 BDT per month), and the transportation costs from the mill to Dhaka (almost 9,000.00 BDT for one way). Nevertheless, these costs are more than covered by a considerable trade margin. In the case of coarse rice (mota chal), they purchase paddy for around 14.00 to 17.00 BDT per kg and sell the milled rice for 21.00 to 23.00 BDT per kg. In the case of fine rice (cikon chal), they get paddy for 17.00 to 20.00 BDT and sell rice for 30.00 to 33.00 BDT. At present, they sell BR-29 for 27.00 BDT. Source: Own interview (17.12.2007) After milling, the raw rice (chal) is filled into 50–85 kg jute or plastic sacks and is transported to Dhaka. Dhaka has never been a junction of the national railroad network. The city is sited at the end of a meter gauge railway from Mymensingh in the North to Narayanganj in the Southeast, and at a broad gauge that connects the city with the western grid. Transshipment makes the train inappropriate for the transportation of rice. Inland water transportation has become less popular in the last decades, a process that is associated with a decline of financial resources that are available for the maintenance of the waterways which, in turn, become more and more sedimented. Thus, almost all of Dhaka’s rice is supplied by trucks that are allowed to enter the city only from 8 p.m. to 6 a.m in order to prevent the already heavily burdened traffic from complete collapse. During daytime, the goods are distributed throughout the city by means of pushcarts, rickshaws (s. Photo 11), CNGs or pickups to be get to the numerous retailers. Most rice is directly transported from millers to urban wholesalers, but in some cases there is again a bepari who acts as intermediary. Beparis, as in the case above, are businessmen who

Food Wholesale Markets

107

command own storage capacities that allow them to perform professional rice trading business. In contrast, the urban wholesalers (arotdars) are independent business organizations who have their own establishments at their markets, but are – similar to the farias in rural areas – commission agents of either millers or beparis. Retailers are mainly owners of grocery shops or small vending stalls at one of the countless “kitchen markets” (katcha bazar) in Dhaka (s. Photo 12), but some of them do also operate restaurants, canteens, backyard cookshops or street food stalls.

Photo 11: Rice delivery by rickshaw Dhaka City, Babubazar, 26.02.2008 Source: Photo by author

Photo 12: Rice retail store Dhaka, Tongi, 23.09.2007 Source: Photo by author

After the value chain for domestically produced rice, the chain for imported rice needs to be explained. In fact, this chain is much shorter, being constituted by only three actors (excluding producers and exporters abroad), i.e. the importer, the arotdar and the retailer. Out of these three actors, only the importer’s business is outstanding. Like the bepari, the importer is a professional trader who is responsible for bringing rice44 from abroad to Bangladesh. Today, the majority of rice imports stem from India, mostly from Uttar Pradesh, Bihar, and West Bengal. While imports of governmental agencies mainly come by ship, private imports are brought in by train or by truck. As it is depicted in Map 3, there are currently 12 land ports in Bangladesh. Half of them can be reached by road network, i.e. in Khulna (Bhomra), Rajshahi (Sonamosjid), Dinajpur (Hili), Sylhet (Tamabill and Shwola) and Chittagong (Teknaf), while the other half comes by railway, i.e. in Jessore (Benapole), Kusthia (Darshana), Rajshahi (Rahanpur45), Dinajpur (Birol), Rangpur (Burimari) and Sylhet (Jokiganj). According to official information provided by the Directorate General of Food, Ministry of Food and Disaster Management (visit on 14 January 2008), most prominent land ports for rice imports (2006–2007) are Benapole, followed by Sonamosjid, Hili, Bhomra and Teknaf. At 44 There is almost no import of paddy in Bangladesh. 45 Due to their spatial proximity to each other, the land ports Sonamosjid and Rahanpur are illustrated by only one symbol in Map 3.

108

Food Wholesale Markets

all these inland landports, truckloads are transshipped directly at the border (s. Photo 13). The lorries are inspected, their load is weighted and the drivers’ documents are checked before goods are repacked on domestic trucks. Due to same gauges in parts of India and Bangladesh, trains can be routed further inland. Thus, a prominent freight depot for imported rice is Sirajganj (s. Photo 14) in the Pabna region (s. Map 3). Nonetheless, in the end, the freight that is brought to Bangladesh by train must also be transshipped onto trucks before it can be distributed throughout the country.

Photo 13: Land port “Panama” Dinajpur, Hili, 19.12.2007 Source: Photo by author

Photo 14: Transshipment of import rice Sirajganj Municipality, 20.12.2007 Source: Photo by author

The description above shows that two kinds of value chains are most common in the rice business in Bangladesh. The first chain consists of seven nodal points, from producing paddy (dhan), over processing raw rice (chal), to providing cooked rice (bhat) to the people, with a distinct actor operating at each of these nodes. As it is illustrated by thin arrows in Fig. 18, this chain comprises the following actors: producer faria bepari miller bepari arotdar retailer final consumer. The second chain is shorter and comprises only five actors: producer faria miller arotdar retailer final consumer. Having presented this, it must be mentioned that reality is much more complex. Usually, a single actor (apart from producers) acquires paddy or rice through various channels. Furthermore, these channels are not static but vary by season, location and rice type. Eventually, some actors play the roles of more than one value chain agent. Likewise, a miller might get aman paddy from farias in Jessore and boro paddy from beparis in Dinajpur (s. Map 3), and he might additionally perform import and trading business in profitable times. Yet, this simplification makes it possible to draw an estimation of the value that is produced in the supply chain and of the relative economic rent that is earned by each chain actor. For doing this, the example of domestically produced coarse rice is taken, which is the rice type that is predominantly consumed by poor people in Bangladesh.

109

Food Wholesale Markets

Table 19: Distribution of arising costs, net margin and consumers’ expenditure for domestically produced coarse rice (local variety) by value chain actors Actors

Activities

Arising costs (in BDT/kg)

Producer Faria Miller

Production Marketing Processing Transportation Marketing Marketing

12.50 (82.5%) 3.80 (32.8%) 0.15 (1.0%) 0.15 (1.3%) 1.50 (9.9%) 5.50 (47.4%) 0.35 (2.3%) 0.15 (1.0%) 0.15 (1.3%) 0.30 (1.1%) 0.50 (3.3%) 2.00 (17.2%) 2.50 (9.4%) 15.15 (100.0%) 11.60 (100.0%) 26.75 (100.0%) Source: Own draft based on surveys 2008–2010

Arotdar Retailer Total

Net margin (in BDT/kg)

Consumers’ expenditure (in BDT/kg) 16.30 (60.9%) 0.30 (1.1%) 7.35 (27.5%)

Note: 1. Arising costs: Producers’ production costs comprise expenses for land lease, seeds, fertilizer, plant protection, irrigation, and seasonal labor. Millers’ processing costs comprise expenses for electricity and labor, and costs that accrue to loss (rice output is usually calculated at two-thirds of paddy inputs). Millers’ transportation costs comprise expenses for up- and downloading of trucks, fees for toll collection points (e.g. Jamuna Bridge, Faricha Ferry Ghat, etc.), labor, and fuel. The marketing costs of faria and arotdar comprise business lease and labor, while the marketing costs of retailers comprise stall fees and transportation. 2. The net margin is calculated as the difference between purchase and sales price of respective actors less arising costs. The consumers’ expenditure is the sum of arising costs and net margins.

Accordingly, it was found that 82.5 percent of all arising costs are generated at the farm gate, 12.2 percent accrue to the miller through processing and transportation, and 3.3 percent are generated by the retailer. It was further found that 47.4 percent of the total net margin is generated by millers, 32.8 percent is earned by the producer and 17.2 percent by the retailer (s. Tab. 19). Apart from the share attributed to production (46.7 percent), to the milling and transportation cost of millers (6.9 percent), and to the marketing cost of retailers (1.9 percent), also the largest proportion of the consumers’ expenditure accrues to the millers with 20.6 percent, followed by producers with 14.2 percent, and retailers with 7.5 percent. Remarkably, the arotdar (together with the faria) is responsible for only 1.1 percent of the consumers’ expenditure while, out of that, just 0.5 percent is his net margin (s. Fig. 18). Altogether, arising costs make up 56.6 percent (15.15 BDT per kg) and the net margin of all involved value chain actors amount to 43.4 percent (11.60 BDT per kg) of the final consumer price (s. Tab. 19).

110

Food Wholesale Markets

Figure 18: The value chain for coarse rice to Dhaka

Source: Own draft based on survey 2007–2009

Thus, in relative terms, it is the millers and the producers who wield significant control over the market and its prices, while commission agents (arotdars accompanied by faria) earn the smallest share of the produced value in the chain. However, this picture is not complete for two reasons. First, the particular context of each actor must be considered. In regard to Bangladesh’s peasants, for instance, it must be added that they are those chain actors most at risk, since floods, cyclones, drougths and cold waves regularly lead to crop loss that too often cause human calamities in the absence of sufficient financial compensation measures. Millers, in contrast, are affected by such shocks in a different way. By means of their connections to producers from many different places, they are able to run their mills throughout the whole year despite local harvest losses. By commanding godowns, they are in the position to store paddy during harvest times and to process it according to market demand. Against this background, the millers are in a postion much more comfortable as compared to the situation of peasants. The second reasons for the incompleteness of the picture shown above is the fact that all findings presented so far are based upon calculations on the share of the consumer price only. Yet, in order to complete the picture, we need to also consider the absolute amounts that are earned by each individual actor. This will be done in the next

Food Wholesale Markets

111

section (5.3) – not for all value chain actors, but at least for Dhaka’s wholesale traders. 5.2.2 The Value Chain for Fish The major share of fish that is consumed in Bangladesh and in Dhaka is carps that are cultivated in ponds and ditches. Accordingly, the focus of this part is on pisciculture fisheries. Within the value chain for cultivated carps (mainly Rui, Catla, and Mrigal, but also Silver Carp, Grass Carp, Japanese Carp) hatcheries constitute the primary link. In hatcheries, fish reproduction is organized in a controlled environment. Hatchery owners usually command a number of ponds for keeping broodfish, facilities for fish spawning, egg incubation and the rearing of hatchlings to post-larva stage, ponds for nursing post-larvae to fry, and ponds for rearing fry to fingerlings.

Photo 15: Carp breeding tank Gov. Fish Seed Multiplication Farm Khulna Municipality, 08.02.2009 Source: Photo by author

Photo 16: Incubator with carp eggs Jagaroni Chakra Foundation Jessore Municipality, 11.02.2009 Source: Photo by author

In the broodfish ponds, male and female fish is kept separate in order to hinder uncontrolled reproduction. When the fish grew mature and it is the season, the best examples in terms of size, weight and shape are selected for reproduction. Female carps are transfered to “mother ponds”, where they get hormonal treatment by which the fish is artificially brought about to spawn. For the purpose of spawning and fertilizing the eggs (ova), male and female fish is kept together for 10 to 12 hours in shallow breeding ponds where there are devices to pump out the fish eggs after fertilization (s. Photo 15). There are usually three carp reproduction seasons per year, from February to March, during monsoon months, and from October to November. Female carps are kept for around five years in hacheries for

112

Food Wholesale Markets

reproduction purposes. Afterwards they get sold to the market in order to keep up the quality of eggs of the hatchery, to keep down diseases, and to prevent inbreeding. If allowed to hatch in ponds, fertilized carp eggs are normally subject to heavy predation and exposed to diseases due to infections or chemical hazards of the environment. Thus, the fertilized fish eggs are channeled into special incubators (s. Photo 16). Under the constant addition of oxygenated and well tempered water, up to three kg of eggs per incubator are brought to eclosion, that takes up to a period of four days.

Photo 17: Carp post-larva Jagaroni Chakra Foundation Jessore Municipality, 11.02.2009 Source: Photo by author

Photo 18: Cultivating fry in private hatchery Jessore Municipality, 12.02.2009 Source: Photo by author

Three days after hatching, the larva starts to require exoganeous food (prior to that they subsist on their own yolksacs), that constitute their transition to the post-larva stage (Jhingran & Pullin 1985: 58ff.). Rearing post-larvae to fry and fry to fingerlings usually involves two ponds, a nursery and a rearing pond. These ponds need special preparations, i.e. the liming of the ground (that acts as a general pond disinfectant, effectively stops fluctuations of pH, and promotes mineralization), the eradication of unwanted fish, predatory insects and harmful vegetation, as well as the fertilization of the pond (mainly nitrogen, phosphorus, and potassium that enables the suitable production of zooplankton which forms the main food of postlarvae, fry and hatchlings). After these preparations, the pond is stocked with post-larvae (s. Photo 17). Manure and supplement feed is constantly added until the post-larvae grow to fry of a length of 2 to 2.5 cm (s. Photo 18). The fry are netted from the nursery ponds and released in rearing ponds which need the same preparations as the nursery ponds. After three to four months in the rearing ponds, fry have grown to fingerling stage that equals a length of 2.5 to 13 cm (s. Photo 19). Then they are caught and sold to aquaculturist customers. The pond culture for raising fingerlings is similarly organized as the hatchery pond culture (s. Box 3). It relys on natural feed but is supplemented with feed and fertilizer that makes

Food Wholesale Markets

113

it a semi-intensive enterprise46 (Alam 2010). Professional aquaculturists raise most of their fish until an age of nine month to one year when one exemplar weighs approximately one kg (s. Photo 20). Then they are sold to the market.47

Photo 19: Netting of fingerlings Jagaroni Chakra Foundation Jessore Municipality, 11.02.2009 Source: Photo by author

Photo 20: Harvesting of carps Jagaroni Chakra Foundation Jessore Municipality, 11.02.2009 Source: Photo by author

Box 3: A gher owner in Jessore Since 1998, Shahidul leases eight ponds that together occupy ten bikha (1.25 ha) of land and that extend alongside the Jessore-Benapole highway at the outer skirts of the municipality of Jessore (s. Map 4). These eight ponds are the basis of his sweet water fish cultivation business – his gher business. Gher is the Bangla term for any physical construction that serves the cultivation of fish or prawn. Traditionally, ghers are half ponds and half paddy fields so that paddy and fish farming can be done at the same time (s. Photo 21). Today, professional aquaculturists resign from this double function and specialize on fish only (s. Photo 22). Shahidul explains the main cost factors of his business. For a pond of the size of one bikha (0.125 ha), he calculates the production costs for one fish season (that usually takes three to four months) at 14,300.00 BDT48 for the land lease, 10,000.00 BDT for lime and pesticides, 2,500.00 BDT for seasonal laborer, 21,000.00 BDT for his two permanent employees, 30,000.00 BDT for fertilizer and fodder, and 2,000.00 BDT for water. For a pond of that size, he usually purchases about 2,700 small fry for a total amount of 15,000.00 BDT (inclusively transportation costs). He gets these small fishes either from governmental fish 46 Extensive pond cultures entirely rely on natural feed, while intensive pond cultures entirely rely on externally induced feed and fertilizer (Lewis et al. 1996). 47 Nonetheless, also bigger fishes are available in the market. These are usually older fishes that were used for reproduction purposes. Carps become sexually mature not before the age of three to four years. These examples are usually not only larger and heavier but also more expensive. 48 At the time of the interview, 100.00 BDT equated 1.102 EUR (cf. http://www. bankenverband.de/service/waehrungsrechner).

114

Food Wholesale Markets

seed multiplication farms or from private hatcheries. This makes his total investment to be 94,800.00 BDT per season. If no severe incidents happen, his total catch accounts for 2,500 fishes, which means an average production cost of 37.90 BDT per fish.

Photo 21: Gher Khulna, Botiaghata Upazilla, 08.02.2009 Source: Photo by author

Photo 22: Professional fish farm Jessore Municipality, 12.02.2009 Source: Photo by author

The sales price per fish ranges on local markets in between 55.00 BDT and 70.00 BDT for one fish of a weight of one kg. By taking the average price of 62.50 BDT, this means a total profit of 61,450.00 BDT per season, or of roughly 15,400.00 BDT per month. However, this income is not always secure as the fish business is generally subject to some risks. “In case of a flood, the catch of an entire season is lost”, Shahidul mentions. In fact, in 2007, Jessore was hit by the cyclone Sidr that made him loose the catch of two entire ponds. Since then, he is heavily indebted. “This fish business is a type of lottery”. This is why he plans to readjust his livelihood: “In future, I will do only hatchery business, because this is more profitable and less risky than gher business”. Source: Own interview (12.02.2009)

Photo 23: Transport of fish in drums Savar, Banktown, 18.01.2009 Source: Photo by author

Photo 24: Transport of fish in baskets Jessore, Boro Bazar, 12.02.2009 Source: Photo by author

Food Wholesale Markets

115

The supply of carps from rural areas to Dhaka is comparable to the supply of rice. At the first stage after cultivation, the fish is sold to the dalal. Similar to faria, dalal are independent players who do their business on commission on behalf of a bepari. In some cases, a local money lender (mohajan) is in between both, but usually the fish is directly collected by the bepari who is also called “collector” and who organizes the transportation to Dhaka. That means he either sends his own trucks or he instructs a freight forwarding agent (s. Box 4) to deliver his goods. These forwarding agents are usually specialized in the transportation of specific food types. The fish is either transported alive in water-filled drums (s. Photo 23), or frozen in ice-filled metal boxes or baskets (s. Photo 24). Similar to the case of rice, most fish reaches Dhaka by truck, where it arrives after a journey of maximally 12 hours. At these markets, the fresh fish is sold via commission agents (arotdars). In the early morning hours, fresh fish is transported by rickshaws or CNGs to the customers, i.e. hawkers who perform door-to-door sale (Photo 25), retailers at kitchen markets (s. Photo 26), or operators of restaurants and canteens. Box 4: Freight forwarding agents and truck drivers in Dhaka A couple of truck drivers and helpers at the truck stop of Gabtoli at the western fringe of Dhaka (cf. Map 2 and 5) explain their business. It is the time of Ramadan (Ramsan) and we sit in a small tea stall that is covered with curtains in order to hide us from the sightings of devout passers-by. All of the men are drivers of 25 mt trucks that are usually used for inter-district transportation. It is afternoon and they wait for the evening to come until they are allowed to enter Dhaka with their vehicles. Around Dhaka there are basically three such truck stops. All of them are located at one of the city’s arterial roads. Here at Gabtoli trucks arrive mainly from north-western districts. In Tongi, in the North of Uttara, the trucks mainly come via the Dhaka-Mymensingh Highway from northern regions. Saidabath, which is close to Jatrabari, is the place where trucks arrive via Dhaka-Chittagong Highway from the south-eastern part of the country. A less important forth truck stop is located on the southern side of the 2nd Buriganga bridge, which is used by trucks that come from southern regions crossing the Mawa Ferry Ghat. Truck terminals are in Tejgaon, Tikatuli and Amanitula, but these places are for reparation purposes and the parking of trucks, only. “Only 1 out of 1,000 drivers owns his truck” one of them tells me. “All of us are just employed”. Usually, there is one driver and one or two helpers on board of a truck during the time of the journey. All of them are paid per tip. One tip means the transportation of goods from its place of origin to its final destination. The drivers earn about 1,000.00 BDT49 per tip. Usually they do 6 to 7 tips per month. The payment is based on an agreement between the drivers and the truck 49 At the time of the interview, 100.00 BDT equated 1.032 EUR (cf. http://www. bankenverband.de/service/waehrungsrechner).

116

Food Wholesale Markets

owners without any written contracts. Whatsoever the place of origin is, a journey would usually take not more than 12 hours. But there are often long waiting hours at the ferries. One man mentions that he just came from Pabna region. His plan was to cross Jamuna at Aricha Ferry Ghat, but due to low tide it was closed. So he had to take a detour to Paturia Ferry Ghat where he had to wait a full day to cross the Padma. He could have also taken the way via Jamuna Bridge, but this is more expensive. While it takes 800.00 BDT for a truck to cross Padma at Paturia, it costs 2,000.00 BDT to cross Jamuna Bridge. The waiting period at Jamuna Bridge is much less, but for the truck owner it makes often more sense to let the drivers wait than to pay the bridge toll. Only in case of perishables, the loads must reach the markets as fast as possible. The current payment pattern did not change over the past 10 years, the men complain. That is why some drivers transport extra loads that are beyond the trucks offical carrying capacity in order to earn extra money. This business is not without risk, as there are police checkposts on the way where the truck license (blue book), the fitness of the vehicle and the load is inspected. But this risk is calculable: “Whether we have the correct weight or not, policemen always find something to complain and to charge money for.” Source: Own interview (18.09.2007)

Photo 25: Hawkers prepare for business Dhaka City, Kawran Bazar, 31.03.2008 Source: Photo by author

Photo 26: Retailer on kitchen market Dhaka City, Khilgaon, 24.09.2007 Source: Photo by author

One major kind of value chain exisits in the Bangladesh’s carp business that involves six nodal points where distinct agents operate before the goods reach the final consumer. Being illustrated in Fig. 19, this chain comprises the following actors: hatchery owner pond owner dalal bepari arotdar retailer consumer. In some cases, a money lender (mohajan) serves as additional intermediary in between dalal and bepari, that makes the value chain to comprise in total seven nodal points: hatchery owner pond owner dalal mohajan bepari arotdar retailer consumer. The value chain analysis is performed by tak-

117

Food Wholesale Markets

ing the example of rui carp, which is the fish type that is mostly consumed in Dhaka (s. Fig. 12). It is found that 67.3 percent of all arising costs are generated at the producers level, i.e. 7.4 percent for nurturing fish in hatcheries and 59.9 percent for raising fish in ponds. Due to the small sales amounts of retailers and their comparably high expenses for the rent of their stalls (venders at kitchen markets) and bribes (hawkers), the marketing costs of retailers are second amounting to 11.3 percent of the total arising costs. The costs of commission agents are higher for fish than for rice, with 6.9 and 3.8 percent that accrue to arotdars and dalals respectively. With 10.8 percent, transportation costs of beparis are considerable as well. The reason is large quantities of water and ice that are needed in transport in order to keep the fish fresh. It is estimated that drums – used for the transportation of fish – are filled with water to an amount of three fourth to four fifth. Likewise, about one third of the total weight of baskets stems from the ice. It was further found that 41.0 percent of the total net margin goes into the beparis’ pocket that makes them the players who gain most of the current system. Nevertheless, in case of successful management, also aquaculturist are in a position to make considerable net margins that amount to 32.2 percent. Just like in the case of rice, the retailers follow with 12.0 percent of the total net margin (s. Tab. 20). Notably, the total arising costs (66.60 BDT per kg) make up 51.6 percent of the final consumer price. In other words, the earnings of the actors involved in the value chain sum up to almost half (48.4 percent or 62.40 BDT per kg) of the final consumer price. Table 20: Distribution of arising costs, net margin and consumers’ expenditure for domestically raised Rui Carp by value chain actors Actors

Activities

Arising costs (in BDT/kg)

Net margin (in BDT/kg)

Hatchery Owner Gher Owner Dalal Bepari Arotdar Retailer Total

Production

4.90 (7.4%)

2.10 (3.4%)

Marketing

39.90 (59.9%) 20.10 (32.2%) 60.00 (46.5%) 2.50 (3.8%) 2.50 (4.0%) 5.00 (3.9%) 7.20 (10.8%) 25.60 (41.0%) 32.80 (25.4%) 4.60 (6.9%) 4.60 (7.4%) 9.20 (7.1%) 7.50 (11.3%) 7.50 (12.0%) 15.00 (11.6%) 66.60 (100.0%) 62.40 (100.0%) 129.00 (100.0%) Source: Own draft based on surveys 2008–2010

Transportation Marketing Marketing

Consumers’ expenditure (in BDT/kg) 7.00 (5.4%)

Note: 1. Arising costs: Hatchery and pond owners’ costs comprise expenses for land lease, fertilizer, animal protection, lime, post-larvae/fry, water, permanent staff, and seasonal labor, and costs that accrue to fish loss. Beparis’ transportation costs comprise expenses for up- and downloading of trucks, fees for toll collection points (Jamuna Bridge, Faricha Ferry Ghat, etc.), labor and fuel, and costs that accrue to fish loss. The marketing costs of dalal and arotdar comprise business lease and labor, while the marketing costs of retailers comprise stall fees and transportation. 2. The net margin is calculated as the difference between purchase and sales price of respective actors less arising costs. The consumers’ expenditure is the sum of arising costs and net margins.

Fig. 19 shows how the customers’ expenditure is composed. It can be seen that 51.9 percent of the consumers’ expenditure accrues until the fish farm’s gate, of

118

Food Wholesale Markets

which 33.2 percent are costs for cultivation and 18.8 percent are the net margin of hatchery and pond owners. By leaving involved costs aside, the greatest share of the consumers’ price arises in the hands of the bepari with 23.9 percent, followed by the pond owner with 17.1 percent and the retailer with 5.8 percent. In some distance follow the arotdars with 3.6 percent, the dalals with 1.9 percent and the hatchery owners with 1.6 percent. Thus, similar to the case of rice, it is the beparis and the gher owners who are the dominant players of the supply chain in relative terms. Likewise to the case of peasants it must be added for the case of gher owners that they usually bear a higher risk of making loss (s. Box 3) as compared to beparis who are embedded in multiple networks to various supply regions by which they can assure a consant business despite local or regional supply bottlenecks. Figure 19: The value chain for carps to Dhaka

Source: Own draft based on survey 2009

Before closing this part, it must be added that fish is not only cultivated but also captured in rivers, estuaries, canals, beels, haors, baors, lakes and in the Bay of Bengal. Since supply chains of capture fish are organized slightly different, the findings of this section should be said to apply to aquaculture only.

119

Food Wholesale Markets

5.3 RICE AND FISH WHOLESALE TRADERS IN DHAKA 5.3.1 The Traders In order to get information about the key characteristics of the people who are active in Dhaka’s food wholesale business, about the involved organizations as social units, and about basic organizational patterns of wholesaling in Dhaka, one representative from each business organization was interviewed in his function as owner, manager, salesman or accountant. In so doing, it was always tried to conduct this interview with the representative of highest rank available, preferably with owners or managers of the respective stores. In this part, an overview of these representatives is given according to gender, age, education, religion and place of origin. In the subsequent part, the individual business units are presented according to basic economic parameters such as workforce, shop size, daily turnover (in terms of weight), starting capital, and annual business volume (in financial terms). Figure 20: Distribution of rice and fish wholesale traders by age 30%

25%

Rice Wholesalers n = 206

20%

Fish Wholesalers n = 241

15%

10%

5%

+ 70

-6 9 65

-6 4 60

-5 9 55

-5 4 50

-4 9 45

-4 4 40

-3 9 35

-3 4 30

-2 9 25

-2 4 20

15

-1 9

0%

Age Groups

Source: Own draft based on survey 2009–2010

From the 206 rice wholesale traders interviewed 116 (56.3 percent) were in the position of managers, 86 (41.7 percent) were the owners of the respective stores, and 4 (1.9 percent) were in a lower executive position, e.g. accountants, salesmen or cashiers. Similarly, from 242 fish wholesale traders that were interviewed 123 (50.8 percent) were managers, 118 (48.8 percent) were owners, and 1 (0.4 per-

120

Food Wholesale Markets

cent) was in a lower position. All interviewees were men, as the entire rice and fish wholesale business in Bangladesh is in the hand of males.50 The average age of the respondents was found to be 37 years (36.8 in case of rice traders and 36.9 in case of fish traders). The youngest interviewee was 19, while the oldest was 72. Fig. 20 shows the distribution of respective age groups. Accordingly, 56.8 percent of Dhaka’s rice merchants are in an age of 30 to 50 years, while 28.2 percent are younger, and 15.0 percent are older. The situation is similar among fish merchants. 67.6 percent are 30 to 50 years old, 21.2 percent are younger than 30, and 11.2 percent are older than 50. 1.0 percent of rice wholesalers and 1.2 percent of fish wholesalers are older than 65. The largest individual share of rice wholesalers is in an age of 25 to 29. The largest share of fish wholesalers is 30 to 34 years old. Figure 21: Distribution of rice and fish wholesale traders by educational level No Formal Education Fish Wholesalers n = 242 Class I-V Rice Wholesalers n = 206 Class VI-IX

S.S.C. & H.S.C.

% 70

% 60

% 50

% 40

% 30

% 20

% 10

0%

University

Source: Own draft based on survey 2009–2010

In regard to the formal education of these representatives it was found that, in both cases, the largest portion, i.e. 64.1 percent of rice merchants and 49.5 percent of fish merchants, went to school up to the level of the Secondary School Certificate (S.S.C.) and Higher Secondary School Certificate (H.S.C.) (s. Fig. 21). Both are public examinations that are known under this name not only in Bangladesh, but also in Pakistan and in parts of India. S.S.C. is usually passed after the successfull completion of ten years of schooling, while H.S.C. is usually passed two 50 Usually, poor women and children search Dhaka’s rice markets for so-called “sweeping rice”, i.e. wastage rice which accrues due to the use of hooks in handling the bags. Women collect this rice, clean it and sell it for 12.00 to 16.00 BDT per kg. Apart from them, only one woman was found on Dhaka’s rice markets who worked as night laborer at Badamtuli. On Dhaka’s fish markets, not a single woman was found.

Food Wholesale Markets

121

years afterwards. 11.2 percent of rice and 12.9 percent of fish merchants received higher education at a university. From them, 5.3 percent of rice wholesalers and 5.0 percent of fish wholesalers hold an university degree. From the rice merchants, 15.5 percent went to school up to class nine and 7.8 percent up to class five, while 1.5 percent did not receive any form of education at a governmental school. From the fish merchants, 14.9 percent went to school up to class nine and 16.8 percent up to class five, while 5.9 percent did not receive education at a governmental school. Thus, in the case of fish, there are slightly more wholesalers without formal education than those with a unversity degree. Nevertheless, when compared to the national distribution (cf. BBS 2007b), it must be stated that – from an educational point of view – the majority of Dhaka’s rice and fish merchants cleary belong to the better trained minority of the total population of Bangladesh. With a tradition that dates back to the Delhi Sultanate and the Mughal rule (cf. 4.1.1), today, the overwhelming majority of 89.7 percent of the people in Bangladesh are Muslims. 9.2 percent are Hindus, and a minority of Christians and Buddhists account together for 1.1 percent (BBS 2007b). As such it is not surprising that a clear Muslim dominance coins Dhaka’s food markets as well. Nonetheless, in terms of religion, the rice and fish wholesale business differ from each other. With 96.6 percent of all wholesalers, Muslims are clearly over-represented in the rice trade. Only 2.9 percent of Dhaka’s rice wholesalers are Hindus and no more than 0.5 percent follow other religions. A look at the individual markets reveals that there are pure Muslim markets in all segments: From the seven small markets of Uttor Badda, Kawran Bazar, Mirpur 10, West Nakhal Para, Boro Bazar, Mohakali and Modda Badda, only one comprises non-Muslims. The proportion of Muslims at medium and large size markets are all beyond the national average. At Badamtuli, Dhaka’s largest rice market, not a single member of another religion was found. The share of Muslims among Dhaka’s fish wholesalers, in contrast, is under-represented. Only 78.5 percent are Muslims, while 20.7 percent are Hindus and 0.8 percent are members of another religion. From the researched larger fish markets of Kawran Bazar, Jatrabari, Merul Badda, Sultan Shah, Mugdapara and New Market, only one is entirely in the hands of Muslims. Two out of three smaller markets reveal a mix of religions. Accordingly Dhaka’s fish business leaves space for religious minorities to participate economically, without being an exclusive niche for Hindus or other minorities. In regard to the respondents’ place of origin, it was found that only 10.7 percent of all rice wholesalers and 7.9 percent of all fish wholsalers are born in the district of Dhaka. The majority are migrants. Out of them, 88.6 percent of rice merchants and 92.8 percent of fish merchants are first-generation migrants who live in Dhaka not longer than 30 years. More than two-thirds of the wholesale traders (67.4 percent of rice merchants and 77.6 percent of fish merchants) live in Dhaka for no longer than 20 years, while still more than one third (35.3 percent rice merchants and 40.8 percent fish merchants) moved to Dhaka during the last ten years. These figures, depicted in detail in Fig. 22, reflect the extraordinary dynamic of Dhaka’s population growth with its affects on the city’s labor market.

122

Food Wholesale Markets

Figure 22: Wholesale traders’ residency in Dhaka 30% Rice Wholesalers n = 184

25%

Fish Wholesalers n = 223 20%

15%

10%

5%

0% 1 to 5

6 to 10

11 to 15

16 to 20

21 to 25

26 to 30

31 to 35

36 to 40

> 40

Residency in Dhaka (in years)

Source: Own draft based on survey 2009–2010

Dhaka’s rice wholesalers stem from 35 different districts (out of 64) and from 15 regions (out of 21) that are depicted in Map 3. When taking the regions as reference areas (s. Fig. 23), it turns out that 50.2 percent of the respondents are from the Dhaka region that comprises the districts of Dhaka, Gazipur, Manikganj, Munshiganj, Narayanganj and Narshingdi. The largest number of rice wholesalers comes from Narayanganj district with a share of 31.6 percent. This fact reflects Dhaka’s history. From colonial times onwards, Narayanganj serves as Dhaka’s port. At the beginning of the 19th century, the small town gained momentum when foreign companies started to export jute to western countries. With the growing importance of jute exports, Narayanganj evolved as the most important inland river port in East Bengal (later East Pakistan and Bangladesh). Until the end of colonial rule, it was Narayanganj that connected Dhaka to Calcutta. In the 1960s, imports of wheat flour from Calcutta started as part of food aid and a number of flour mills were established (Solaiman 2006). As such, it is basically the children and grandchildren of the flour millers of Narayanganj who opened up the first rice wholesale stores in Old Dhaka and elsewhere. Today, the largest concentration of people from Narayanganj can be found at Badamtuli where they make up 68.3 percent of all wholesalers, but also at Jatrabari and Krishi Market where they account for 50.0 and 21.4 percent, respectively. The geographical origin of Dhaka’s fish wholesalers, in contrast, is more diverse. Fish merchants originate from 41 different districts and from 18 regions (cf. Map 4). Nevertheless, almost two-thirds are from four regions only, namely Comilla (20.3 percent), Faridpur (20.3 percent), Dhaka (17.0 percent) and Barisal (14.1 percent) (s. Fig. 24). In more detail, the largest number of traders are from the districts of Barisal (10.7 percent), Shariatpur (Faridpur region; 8.3 percent), Dhaka (7.9 percent), Comilla (7.4 percent), and Gopalganj (Faridpur region; 6.3 percent). All

123

Food Wholesale Markets

other regions and districts play only marginal roles. The analysis of individual markets did not reveal any strong segregation patterns of wholesalers according to their place (region or district) of origin. Figure 23: Distribution of rice wholesale traders by region of origin 1.5% 1.5% 2% 2.5%

0.5% 0.5%

0.5% 0.5% 0.5%

4%

50%

5%

Dhaka Comilla Noakhali Barisal Faridpur Patuakhali Dinajpur Rajshahi

9%

Jamalpur Mymensingh Tangail Khulna Kushtia

9%

n = 206

13%

Rangpur Pabna

Source: Own draft based on survey 2009–2010

Figure 24: Distribution of fish wholesale traders by region of origin 1% 1% 2%1% 2%

1% 0.5%

0.5% 0.5% 0.5%

Comilla Faridpur Dhaka

3%

Barisal Mymensingh

20%

4%

Khulna Noakhali

6%

Sylhet Patuakhali

6%

Jessore Jamalpur 20%

Chittagong Rangpur Pabna Tangail

14% n = 242 17%

Bogra Dinajpur Rajshahi

Source: Own draft based on survey 2009–2010

In regard to the personal backgrounds of the owners of the wholesale stores, it can be said that family-owned enterprises and networks of kin do generally play an important role in Dhaka. There are families that can look back on a long history in wholesale trade and that are strongly committed to it. An example of such a family is provided by Shahin, the manager of one of the rice wholesale stores at Kawran Bazar. He explains:

124

Food Wholesale Markets

“I live in Narayanganj together with my grandfather, my grandmother and my brothers. I have four brothers, two of them are the owners of rice shops at Badamtuli. The other two – including myself – are managers of rice shops at other markets. My father Md. Shahidullah started his rice business already in the 1960s. He is already dead, but we continue his business.” Source: Own interview (13.02.2008)

However, not all rice and fish traders are born into the business. In fact, the recent process of differentiation of wholesale markets in Dhaka went hand in hand with new actors who decided to invest in this business on the basis of their personal knowledge and their financial endowments. The wholesale store of Mir Bikhtiar is a case in point. His father was a soldier of the British Army. After his retirement he received a pension of the Queen Elisabeth Fund, but since he soon suffered from paralysis, he was not able to invest the money personally. Thus, Mir Bikhtiar as his first born son assumed the management of the pension and opened up a fish wholesale store in 1997 at Merul Badda Fish Market at the eastern fringe of Dhaka City (cf. Map 2 and 5). As he bears responsibility not only for himself, but also for his greater family, he included his brothers in his business. Until he died in 2009, two of his younger brothers worked as managers and cashiers in his wholesale store, while a third brother opened up his own wholesale store at the same market. His three sisters were never personally involved in the fish business, as they are married and work at home. But three sons of the oldest sister live in Dhaka and are old enough to work in the business organizations of their uncles. After Mir Bikhtiar died, his two younger brothers assumed the fish store. Apart from these individual life stories, the dominant role of family ties within the rice and fish wholesale business is also reflected in numbers. Today, in every third rice store (29.9 percent) and in every second fish store (47.1 percent), relatives of the respective store owners are found among the employees in executive functions (manager, auctioneers, accountants). Besides, there are employees who are not related as Sahabuddin knows to refer: “My father was a banker. I went to school until he got retired in 1995. I completed S.S.C and H.S.C. Afterwards I went to graduate class, but I did not complete it. After the retirement of my father, I started to work in the rice business. First, I worked as a salesman in another shop. Later on, I became the manager of this shop. I knew the owner of this shop because we are neighbours. He employed me because I was familiar with doing accountancy.” Source: Own interview (06.02.2008)

5.3.2 The Stores The business organizations of Dhaka’s rice and fish wholesalers must be added to the country’s small enterprises. In the Industrial Policy, the Government of Bangladesh (FBCCI 2005) defines small enterprises for the trading and service sector to be those business organizations “in which fewer than 25 workers work” (ibid.: Chapter 4). Medium enterprises are those with 25 to 100 workers, while large businesses have more than 100 workers. With average number of employees from

Food Wholesale Markets

125

one (1.1) to six (5.9) and from three (3.2) to thirteen (12.7), respectively (s. Tab. 21), the rice and fish wholesale enterprises cleary belong to the first category. The business organizations of rice wholesalers comprise on average three (3.0) employees. Out of them, two (1.8) have executive functions as managers (sorkar) and accountants (kalami), while one (1.3) is usually employed as a permanent laborer (saomik). Only one rice wholesaler declared to have closed a written contract with his employees. The remaining 99.5 percent of the workers in the rice business are employed on the basis of oral contracts. The business organizations of fish wholesalers are generally larger with an average of seven (6.8) employees, i.e. four (3.8) in the executive functions of managers, auctioneers (koial), accountants and weigher (mabdar), and three (3.0) as laborer. 100.0 percent of the workers in the fish business are employed on the basis of oral contracts. By taking the mean figures and adding the owners (malik) of the individual enterprises, one can estimate that, in sum, there are currently 3,100 families in Dhaka City who make a living from rice and some 7,000 families who make a living from fish wholesaling. By taking into account the average family size of 4.4 persons (BBS 2011), there are roughly 44,440 persons who are directly dependent on the rice (13,640) and fish (30,800) business. In addition to that, there are the numerous daylaborer, the rickshaw pullers, truck drivers and their helpers, the tea vendors, ice deliverers, and peons who all would have to be included. Table 21: Workforce of rice and fish wholesale stores Quintile 1st 2nd 3rd 4th 5th Total

Rice wholesale stores Fish wholesale stores Workforce n Workforce n 1.1 41 3.2 48 2.0 41 4.8 49 2.6 42 5.8 48 3.4 41 7.7 49 5.9 41 12.7 48 Mean: 3.0 Sum: 206 Mean: 6.8 Sum: 242 Source: Own draft based on survey 2009–2010

The floor space of the stores of rice and fish wholesalers is comparatively small. Contrary to what one might think of, the majority of rice merchants just command their bare show rooms, where rice sacks are stored, samples of the offered rice varieties are displayed, and the sale is arranged. While on private markets, rice stores usually occupy a certain space in the ground floor of residential buildings (s. Photo 27), on governmental markets, there are usually larger market halls that comprise numerous compartments that are leased out to the merchants (s. Photo 28). Only in some cases, the merchants command additional godowns, be it a garage nearby or an extra compartment of a governmental market. The case of Badamtuli is an exception in this regard. Some powerful market player command considerable storage space under the 2nd Buriganga Bridge, which was found to be the largest single rice storage in Dhaka City. For most of Dhaka’s rice wholesalers, though, access to larger storerooms is denied. These disparities are reflected by the floor space of the rice stores (including the show room and the godown)

126

Food Wholesale Markets

ranging from five square meters to one hundred square meters (sqm). While the smallest stores (first quintile) are on average not bigger than 13.0 sqm, the largest stores (fifth quintile) have an average size of 70.8 sqm. The average is 37.4 sqm (s. Tab. 22). As such, the storerooms of Dhaka’s rice wholesalers seldomly suffice to store bags for longer than a week, given different amounts that are sold on a day to day basis. Due to these limitations, rice bags are often additionally stored on the sidewalk in front of the shops (s. Photo 27).

Photo 27: Rice wholesale store Dhaka City, Badamtuli, 13.11.2007 Source: Photo by author

Photo 28: Rice wholesale store Dhaka City, Kochukhet, 06.02.2008 Source: Photo by author

Table 22: Shop size of rice and fish wholesale stores Quintile 1st 2nd 3rd 4th 5th Total

Rice wholesale stores Fish wholesale stores Shop size n Shop size n (in sqm) (in sqm) 13.0 41 9.3 48 25.6 41 14.2 48 35.1 42 17.5 49 42.4 41 20.8 48 70.8 41 35.0 48 Mean: 37.4 Sum: 206 Mean: 19.3 Sum: 241 Source: Own draft based on survey 2009–2010

In contrast to the brick buildings (pucca) that shelter the rice stores, the vending sites of fish wholesalers are much less permanent with constructions that are mostly made of wood and bamboo (kutcha) with corrugated iron sheets as roofs (s. Photo 29). Due to all the water and ice that is used and spilled, the ground floor of fish markets is partly paved with concrete or bricks (s. Photo 30). Since fish trading usually last for no longer than four to five hours in the early morning and the premises are not used during the rest of the day, no more durable constructions are used. The few valuable objects, such as the pair of scales, knifes and water bowls are put in lockable superstructures above the vending sites or they are stored in a number of offices that the merchants command. For instance, at Kawran Bazar, some fish merchants possess compartments in one of the market buildings that have been erected by the government in the 1980s. These offices

Food Wholesale Markets

127

are used by the workers for the accountancy, to place new orders, and to rest after the rush hours of the market. Additionally, they serve as a showroom for suppliers and customers.

Photo 29: Fish vending sites Dhaka City, Kawran Bazar, 25.01.2008 Source: Photo by author

Photo 30: Fish vending sites Dhaka City, Merul Badda, 08.03.2009 Source: Photo by author

There is hardly any cold storage in the market. The maximum storage time of fish on the markets was stated to be seven days, but only 1.0 percent of the interviewees answered to have the ability to do that. In fact, 75.0 percent of Dhaka’s fish wholesalers have no storage capacities at all and need to sell their entire stocks within a few hours until the market day ends. In the rare case that fish cannot be sold, it is cooked and eaten; larger quantities are processed to dry fish. 15.1 percent of the traders are able to store fish at least over night, mostly done in metal boxes that are filled with ice and kept beside the vending sites. Not more than 9.9 percent are able to store fish for longer than a day. Given these circumstances, it is not surprising that fish wholesale stores occupy only 5.0 to 70.0 sqm. The average size is 19.9 sqm. With 9.3 sqm as the average store size of the first quintile and 37.9 sqm as the average store size of the fifth quintile, the disparities between fish wholesalers are slightly lower than in the case of rice (s. Tab. 22). Table 23: Daily turnover of rice and fish wholesale stores Quintile 1st 2nd 3rd 4th 5th Total

Rice wholesale stores Fish wholesale stores Daily turnover n Daily turnover n (in kg) (in kg) 1,040.3 41 203.8 48 1,990.2 41 337.8 49 2,810.4 41 516.8 48 3,856.1 41 742.3 49 9,182.5 41 2,104.1 48 Mean: 3,775.9 Sum: 205 Mean: 778.9 Sum: 242 Source: Own draft based on survey 2009–2010

128

Food Wholesale Markets

Despite the small number of employees and a limited physical space, the daily amounts of rice and fish that are traded at Dhaka’s wholesale markets are huge. Rice wholesale traders sell a daily average quantity of 3,775.9 kg. This is almost five times as much as the fish wholesale traders do, who sell an average of 778.9 kg per day. In both businesses, there are distinct disparities between the traders. The smallest rice wholesalers (first quintile) sell on average 1,040.3 kg per day, while the largest ones (fifth quintile) sell 9,182.5 kg per day. Likewise, small fish merchants (first quintile) sell an average daily amount of 203.8 kg, whereas large tradesmen (fifth quintile) sell an average amount of 2,104.1 kg (s. Tab. 23). Thus, in case of rice, the largest business organizations (fifth quintile) sell 8.8 times more quantities than their smallest competitors (first quintile). Being converted to the city’s total supply, it turns out that those 20.0 percent of enterprises that are at the bottom of the spectrum manage together only 5.5 percent of the city’s total supplies. In contrast, those 20.0 percent at the top of the spectrum manage together 48.6 percent of the total supplies. In terms of fish, disparities are even more distinct with the largest business organizations (fifth quintile) selling 10.3 times as much fish in quantity as their smallest competitors (first quintile). While the smallest 20.0 percent of fish wholesalers organize together 5.2 percent of Dhaka’s total supplies, the largest 20.0 percent organize 53.6 percent. This means that not more than 155 rice and 180 fish wholesale stores arrange half of the city’s supplies. These figures are impressive enough. And still, real disparities are even more pronounced, since 25.7 percent of the owners of rice stores51 and 17.8 percent of the owners of fish stores52 run more than just one single wholesale store.

51 From them, 65.4 percent run additional rice wholesale stores, while 34.6 percent run other types of business organizations, such as grocery shops, or stores for the sale of potatos, shoes, books, confectionary, furniture and oil. One interviewee stated to run a factory for bottled water. From 53 rice wholesalers with more than one store, there are 30 (56.6 percent) with a total number of two stores, 12 (22.6 percent) with three stores, six (11.3 percent) with four or five stores and five (9.5 percent) with six stores or more. 52 From them, 68.3 percent run additional fish wholesale stores, while 31.7 percent run other types of business organizations. Among those businesses mentioned were shops for confectionary, television and multimedia, ornaments, iron and steel, and meat, beside grocery shops, pharmacies and phone and fax shops. From 42 fish wholesalers with more than one store, there are 26 (61.9 percent) with a total number of two stores, five (11.9 percent) with three stores, six (14.3 percent) with four or five stores, and five (11.9 percent) with six stores or more.

Food Wholesale Markets

129

Table 24: Capital invested by rice and fish wholesale stores Quintile

1st 2nd 3rd 4th 5th Total

Rice wholesale stores Capital inCapital vested invested (in BDT) (in EUR) 113,677.00 1,125.00 366,667.00 3,627.00 537,097.00 5,314.00 973,333.00 9,629.00 3,109,677.00 30,764.00 Mean: Mean: 1,024,677.00 10,137.00

Fish wholesale stores Capital invest- Capital n ed (in BDT) invested (in EUR) 31 55,725.00 551.00 40 30 165,000.00 1,632.00 40 31 325,875.00 3,224.00 40 30 485,000.00 4,798.00 40 31 1,454,250.00 14,387.00 40 Sum: Mean: Mean: Sum: 153 497,170.00 4,919.00 200 Source: Own draft based on survey 2009–2010 n

Note: In the course of the survey, the curreny rate for 100.00 BDT changed from 0.973 EUR (15.12.2009) to 1.011 EUR (01.01.2010) to 0.984 EUR (15.01.2010) (cf. http://www.bankenverband.de/service/ waehrungsrechner). By taking the mean of these three, the average rate in the time of the survey was 0.989 EUR.

Compared to other actors of Dhaka’s food system, rice and fish wholesalers are of considerable size if measured by their economic value. This can be expressed by means of the capital invested and the annual sales. The capital invested by rice wholesale enterprises ranges from 113,677.00 BDT or 1,125.00 EUR (first quintile) to 3,109,677.00 BDT or 30,764.00 EUR (fifth quintile) with an average of 1,024,677.00 BDT or 10,137.00 EUR. Compared to that, fish wholesalers need roughly half the financial resources to start a business, with amounts that range from 55,725.00 BDT or 551.00 EUR (first quintile) to 1,454,250.00 BDT or 14,387.00 EUR (fifth quintile) and with an average of 497,170.00 BDT or 4,919.00 EUR (s. Tab. 24). As such, the invested capital of both business types exceeds that of most other food system actors by far. Table 25: Annual sales of rice and fish wholesale stores Quintile

1st 2nd 3rd 4th 5th Total

Rice wholesale stores Annual business Annual volume business (in BDT) volume (in EUR) 9,528,974.00 94,270.00 17,695,723.00 175,064.00 26,021,293.00 257,429.00 38,270,719.00 378,612.00 93,449,626.00 924,497.00 Mean: Mean: 36,993,267.00 365,974.00

Fish wholesale stores Annual business Annual n volume business (in BDT) volume (in EUR) 41 8,171,894.00 80,845.00 48 41 15,664,435.00 154,968.00 49 41 25,122,234.00 248,534.00 48 41 38,751,113.00 383,365.00 49 41 158,313,058.00 1,566,191.00 48 Sum: Mean: Mean: Sum: 205 49,022,755.00 484,982.00 242 Source: Own draft based on survey 2009–2010 n

Notes: 1. For the calculation of the annual sales, the average quantities (in kg) sold daily were multiplied with the number of days the wholesale business organization in question was open during the course of the last year, and with the average sales price of the rice or fish type that is primarily sold.

130

Food Wholesale Markets

2. In the course of the survey, the curreny rate for 100.00 BDT changed from 0.973 EUR (15.12.2009) to 1.011 EUR (01.01.2010) to 0.984 EUR (15.01.2010) (cf. http://www. bankenverband.de/service/waehrungsrechner). By taking the mean of these three, the average rate in the time of the survey was 0.989 EUR.

The annual sales of rice wholesalers is on average 36,993,267.00 BDT or 365,974.00 EUR per year, with a range from 9,528,974.00 BDT or 94,270.00 EUR (first quintile) to 93,449,626.00 BDT or 924,497.00 EUR (fifth quintile). The annual sales of fish wholesalers is even larger with an average of 49,022,755.00 BDT or 484,982.00 EUR. Also the disparities in the fish business are more pronounced with a sales volume that ranges from an average of 8,171,894.00 BDT or 80,845.00 EUR per year for the first quintile, to an average of 158,313,058.00 BDT or 1,566,191.00 EUR for the fifth quintile of merchants (s. Tab. 25). As such, in case of rice, the sales volume of those 20.0 percent of wholesalers at the top of the spectrum exceeds that of the 20.0 percent at the bottom of the spectrum by the factor 9.8. In case of fish this factor is even 19.4. These figures now bring to the fore the real dimensions of the disparities between the various wholesale enterprises in Dhaka and the economic power, not of all, but of some wholesale traders to dictate prices even along the entire value chain. In the fish business, only 7.9 percent of all wholesalers have a sales volume that exceeds 100,000,000.00 BDT per year. In the rice business, it is even only 4.9 percent. By taking these figures, one can estimate that the “big fishes” of Dhaka’s food markets are not more than 71 wholesalers in the fish business and 38 wholesalers in the rice business. 5.4 RÉSUMÉ Three major findings can be drawn from this section. First: For understanding Dhaka’s food system, one needs to first understand the city’s freight transportation system, which saw three broad development stages in the last centuries. The first stage was predominant more or less from the early times of the city until the end of the 19th century, when Dhaka’s food provision relied almost entirely on inland water transport. The prevalence of this transportation system explains not only the location of ancient Dhaka, but also the places of the city’s first wholesale markets. The second development stage started in 1885, when the Eastern Bengal Railway opened the Narayanganj-Dhaka-Mymensingh railway line. Since the early 20th centry the railway network has covered substantial shares of the northwestern part of Bangladesh which has allowed Dhaka to be supplied with rice from regions such as Dinajpur, Rangpur and Bogra. Yet, despite these extensions, the railway did never replace the country’s inland water transport system. Consequently, it affected the capital’s food system only marginaly. The third development stage started in the second half of the 20th century when the government took initiative to develop the road network with support of international donor agencies. The Jamuna Multi-Purpose Bridge has allowed to transport boro rice from north-western regions to Dhaka and serves since as the infrastructural backbone of

Food Wholesale Markets

131

the city’s rice supply. Today, all three kinds of transport are run side by side, which is good and necessary. However, when considering the often antiquated trucks that are in use for the bulk of food transportation and the high costs that accure for maintaining waterways in a region heavily affected by monsoon, it seems a pressing task for Bangladesh to resume a national initiative for the development of the railroad network. The much larger load capacity of trains allows food to be transported at reduced rates. In the long run, this would help reduce CO2 emissions in freight traffic and it would contribute to keep prices low for the ultimate consumer. Second: Despite Dhaka’s outstanding expansion since independence, the city’s food markets kept step with this rapid development. With the rise of the truck as Bangladesh’s main means for inland freight transportation, it became possible to establish wholesale markets throughout the entire city. In consequence, the last 40 years saw an intense differentiation process of markets in Dhaka. Today, a total number of 87 wholesale markets for food are located on the territory of DCC. This development can be understood as an expression of private businessmen’s agency in constantly responding to the successive physical expansion of the city. Notably, the majority of 67.8 percent of all food markets have been established on the basis of private initiatives. At the same time, however, it can also be understood as a manifestation of the municipality’s lacking capacities to govern its market system and to guide its future direction. In fact, the DCC became visible in the planning and establishing of markets not before the second half of the 20th century and yet, its commitment can be described best as being reserved. Therefore many governmental markets are today in a bad shape which makes it difficult for traders to keep up with the urban population’s growning concerns about food quality and hygienic standards. Recently, the national government, the DCC and the megacity’s central planning authority RAJUK (Rajdhani Unnayan Kartripakkha) resumed an initiative for establishing three new food wholesale markets at the urban fringe, i.e. at Aminbazar in the western part of Dhaka, Jatrabari in the Southeast, and Mohakali in the central North (cf. Map 2 and 5). If things are made right, this new effort to concentrate food wholesale activities at the urban fringe can be a step in the right direction, as scarce financial means can be bundled and used for placing investments in important infrastructures – most of all in cold storage facilities for traders of fresh produce. Third: Based upon calculations on the share of the consumer price, the value chain analysis showed that millers and beparis in rural Bangladesh are in the position to earn the highest rent as compared to other involved actors of the food system. For the case of rice, this result is also backed by a study that has recently been published by the Centre for Policy Dialogue (Rahman et al. 2008). These findings might be read as an indicator for millers and beparis to also dominate the buying and selling price of rice and fish to large extents through their links to producers via farias and to retailers via arotdars. However, such a statement is premature as long as absolute amounts are neglected that are earned by each individual actor. For this reason, some social and economic parameters of the traders were presented and discussed in some detail. Dhaka’s wholesale traders show specific

132

Food Wholesale Markets

characteristics. As compared to the national average, it was found that the majority of Dhaka’s rice and fish merchants belong to the better educated minority of Bangladesh with the largest portion having completed S.S.C. and H.S.C.. In regard to the respondents’ place of origin, the survey showed that only 10.7 percent of all rice wholesalers and 7.9 percent of all fish wholsalers are born in the district of Dhaka. The majority are migrants. For the case of rice it turned out that it is basically the children and grandchildren of the flour millers of Narayanganj who opened up rice wholesale stores in Old Dhaka and who make up large proportions of the traders at the markets of Badamtuli and Jatrabari even today, where they account for 68.3 and 50.0 percent respectively. In regard to the personal backgrounds of the traders, it can be said that family-owned enterprises and networks of kin do generally play an important role. Of all data presented, it was especially the findings on the trading stores’ daily and annual business volumes, which was enlightening. In case of rice, it was shown that those 20.0 percent of enterprises that are at the bottom of the spectrum manage together 5.5 percent of the city’s daily supplies, while those 20.0 percent at the top of the spectrum manage together 48.6 percent. The annual sales volumes of the top 20.0 percent of wholesalers exceeds that of the 20.0 percent at the bottom by the factor 9.8. In terms of fish, disparities turned out to be even more distinct with the largest business organizations having an annual sales volume that is 19.4 times higher than that of their smallest competitors. While the smallest 20.0 percent of fish wholesalers organize together 5.2 percent of Dhaka’s daily supplies, the largest 20.0 percent organize 53.6 percent. Based on these findings it can be suggested that not more than 155 rice wholesale stores (out of 775) and 180 fish wholesale stores (out of 898) arrange half of the city’s supplies and not more than 38 wholesalers in the rice business and 71 wholesalers in the fish business can be addressed as the “big fishes” of Dhaka’s food system with individual sales volumes that exceed 100,000,000.00 BDT per year. These results make clear that the question of power in value chains is not so much about a certain category of actors based on their position in the chain. It is rather a question of individual enterprises, their specific upstream and downstream linkages, their individual strategies and, last but not least, their business volume. Against this background it is correct to speak of Bangladesh’s value chains of rice and fish to be shaped like hourglasses, being characterized by the predominance of large numbers of peasants and gher owners, the channeling of trade through relatively few intermediaries like millers, beparis and traders in wholesale spot markets, and the ultimate distribution through large numbers of retailers like street vendors, kitchen markets and grocery stores. Against this background it is also correct to speak of powerful millers and beparis that, in part, even dominate Dhaka’s food system. However, it is similarly correct to speak of powerful wholesalers (arotdars) who possess power to determine prices to large extents. Thus, any classification of actors based on the actors’ position in the value chain alone provides only a coarse picture and might even lead to an oversimplified idea of power in a food system if the economic performance of individual enterprises is overlooked. Any form of generalizing about the wholesaler in

Food Wholesale Markets

133

Dhaka must consequently refrain from mixing up single large scale brokers with the masses of mundane commission agents that coin the general picture of Dhaka’s wholesale markets. Based on the description of Dhaka’s supply chains, the urban food system can be said to be characterized by a lack of formal coordination between the different links in the chain, since producers, wholesalers and retailers relate to each other individually and are not tied through written contracts or common ownership of their enterprises, whatever their size. This, however, does not mean that there is no coordination in Dhaka’s supply chains all together. Otherwise the notable efficiency of the food system, as discussed in 4.2.2 and 4.3.2, could not be explained. As a matter of fact, informal modes of coordination lie at the heart of the megacity’s food system. How does these informal coordination modes look like and how do they work? These and other questions will be addressed in the next chapter.

6. THE RESILIENCE OF FOOD WHOLESALE TRADERS IN DHAKA “Narrowing the focus on the heart of resilience, the key question is what enhances capacities of individuals, groups and [business] organizations to deal with threats more competently” Brigit Obrist, Constanze Pfeiffer and Robert Henley (2010: 291)

After having discussed the general role of food wholesalers in Dhaka’s food system in the section above, the ground is prepared now to arrive at the center piece of this empirical study. In this section, I am going to analyze the economic resilience of Dhaka’s food wholesale traders. In the theory chapter (s. 2.2.3), resilience has been defined from an actor perspective to comprise three dimensions: 1) Coping capacities: the ability of actors to cope with and overcome all kinds of adversities by the means of those resources that are directly available; 2) adaptive capacities: their ability to learn from past experiences, anticipate future risks, and adjust themselves to future challenges in their everyday lives; and 3) transformative capacities: their ability to access assistance from the wider socio-political arena, to participate in decision-making processes, and to craft institutions that both improve their individual welfare and foster societal robustness toward future crises. When studying real markets, all three types of capacities are important and need to be considered. In order to make the terms operable, they are related in the following section to three specific scales of analysis, i.e. practices, networks, and arenas. Furthermore they are connected to market actors’ coordination problems as they have been identified by Beckert (2009) and extended by myself (s. 2.4.2). Accordingly, the subsequent study will comprise the following sections: 1) The study of coping capacities: In order to analyze wholesale traders’ coping capacities, I am going to discuss the daily practices of the businessmen which involves three types of strategies, namely the delivery of stocks, the sales of goods, and the organization of restocking. In this discussion emphasis will be placed on the traders efforts to solve the problem of value and the problem of competition. The first of these problems refers to the value of commodities given their multiplicity and complex quality properties. Only if sellers can reliably demonstrate the value of their goods and potential buyers are in a position to distinguish between the values of goods, will uncertainty be reduced and a disposition to buy arises. The second problem is related to one of the insights of neoclassical theory and to the relations of market actors to their competitors. Only if the traders are able to establish structures that shield them from competitors are they in a position to reduce uncertainty with regard to their profit expectations. 2) The study of adaptive capacities: For the analysis of adaptive capacities I am going to put the focus on the traders’ business networks and investigate their

The Resilience Of Food Wholesale Traders

135

strategies to work towards solving what Beckert (ibid.) has called the cooperation problem. The problem of cooperation arises from business risks that market actors face because of their incomplete knowledge of the intentions of their exchange partners in combination with incalculable external factors of influence that might hinder the successful order or delivery of the product. Market relations are risky for instance when one exchange partner makes an advance payment without being sure whether the other party will actually fulfill the contractual obligations. Only if buyers are confident of not being exploited by their contract partners will they engage in market exchange. Creating this confidence is thus a fundamental precondition of stable markets (ibid.). 3) The study of transformative capacities: For the study of transformative capacities, finally, I am going to shift attention to the wholesale traders’ position in the wider urban arena and examine their potentialities to bring about positive outcomes in terms of both the traders’ individual welfare and the wider public. In doing so, my special interest is directed toward the businessmen’s problem of place. Markets need some kind of physical backbone for their functioning. Specific locations must be designated for markting, traders must have access to these places, and illegitimate access and usage must be hindered. As I will show, markets in Dhaka are highly politicized arenas, especially those markets that are located on public land. Only if Dhaka’s wholesale traders find ways to deal with this kind of politicized environment, markets can come into existence as institutional facts (Searle 1995). Before starting the examination of economic resilience, in the first part of this section, I am going to provide the reader with an overview of the business relations that Dhaka’s wholesale traders are embedded in. Based on that, the discussion of the three capacities which together constitute resilience will follow in part two, three and four. The guiding questions of the entire chapter are: What are the major threats food traders have to deal with? How do the traders cope with and adapt to these threats? What can be said about the traders’ potentialities to transform present conditions? And, finally, who benefits from the current system and who bears its costs? 6.1 THE EMBEDDEDNESS OF FOOD WHOLESALE TRADERS As it has been mentioned above (s. 3.3.3), the participatory method called Venndiagram (Kumar 2002) has turned out to be of special value for studying the personal business networks (Schnegg & Lang 2002) of Dhaka’s food wholesale traders. In the course of the fieldwork, this method was performed as comprising three phases: In a first step, the interviewee was asked to name all those actors in his social environ that are of some relevance for his business performance. These actors could be suppliers and customers, but also any other person who would play a role for the trader. In a second step, the interviewee was asked to place these actors on a scale which should reflect the wholesaler’s personal social proximity or distance to these actors. For this purpose, a marker was placed on the ground or

136

The Resilience Of Food Wholesale Traders

on a table for representing the interviewee. Around this marker, a number of concentric circles were drawn that displayed the social space in which he is embedded. Actors who are well known and with whom strong relations exist were placed close to the marker. Actors with whom only weak relations are maintained were placed more far away. A number of five concentric circles were used to express this social proximity in categories from “very close/familiar” to “very distant/unfamiliar”.53 Finally, in a third step, the interviewee was asked to evaluate the relevance of each and every actor or actor group that was placed on the diagram. For this purpose, the trader was equipped with symbols (in this case, cardboard circles) in five different sizes that should represent the degree of importance in categories from “very high” to “very low”.54 I am going to present two of these Venn-diagrams in detail in order to provide the reader “unfiltered” information in the sense that various topics are addressed at the same time, which are sorted out and discussed individually in the subsequent sections of this chapter. The first Venn-diagram (Box 5) will show the business relations of a rice wholesaler, while the second one (Box 6) will address those of a fish wholesaler. Box 5: A rice wholesale trader in Dhaka and his business relations Bashir is one of more than 200 rice wholesalers at Badamtuli, the megacity’s largest rice market (s. Map 5). He is a Dhakaiya, as he was born not far from his rice store in the old part of the city. In 1954, his father moved from Narayanganj to Dhaka and opened up a rice wholesale store at Badamtuli, at a time, when there were only few stores there. Some years later, he bought the entire building. Today, there are eight shops in this building, i.e. three groceries, one shop for mobile phones, one small bakery, and three rice wholesale stores. One of these wholesale stores is run by Bashir’s older brother, one by himself, and one is rented out. Bashir lives together with his wife and his two little sons not far from the rice market. He is in his mid 30s and his sons are in the age of five and one. He resides in one floor of his family-owned house. His mother and his younger brother live in the second story, while his older brother lives together with his wife, his son and his daugther in the third story. Bashir holds a Bachelor degree in Management, but he does not unveil his educational background by wearing pleat-front trousers, how many of his former fellow students do it who work in banks or in governmental offices. During daytime, he rather wears a shirt and the traditional wrap-around skirt (lungi), which is the typical clothing of blue-collar workers in Bangladesh. With this dress he appears to be a primus inter pares among his workers who wear the same type of cloths, even though his body shape clearly distinguishes him from the others. Only after he leaves his rice store, he wears the clothing of white-collar workers. In the evening hours, he likes to play his guitar and to surf the web or chat online with friends. To me as an outsider, it seems that 53 The remaining categories were: “rather close/familiar”, “neither close nor distant” and “rather distant/unfamiliar”. 54 The remaining categories were: “rather high”, “neither high nor low” and “rather low”.

137

The Resilience Of Food Wholesale Traders

he lives in two different worlds – the rustic rice market during daytime, the sophisticated social networks during night. For him, however, there is no disaccord. One afternoon, we sit together in his living room in front of the vitrine where he keeps tableware and figurines from foreign countries. He presents me pictures of himself that show him as a model for menswear, a part-time job that he held down when he was at college. Afterwards, we start conjointly drawing a Venndiagram and, in doing so, he explains to me how his business works (s. Fig. 25). Figure 25: The personal network of Bashir Market Association

7

Suppliers 9

5

Employees 3

4 2

6

Customers

8 10

1

11

Business Partners

13

Government, DCC

12

Bank, Moneylender 14

Bashir

Scale of social proximity: close – – – – – – – – distant

Relevance for business very low – – – – – – – very high Source: Own draft based on survey 2009

Bashir begins with his employees. From his perspective, there is a clear hierarchy of his laborer, as the manager (2) is most important for his business, while the accountant (1) comes second and the laborer (3) third. However, in regard to familiarity, he places the accountant (1) closer to the marker that represents himself

138

The Resilience Of Food Wholesale Traders

than the manager (2) and explains: “I employed my manager two years ago. I met him in Narayanganj, which is the home town of my father. He is a very trustworthy person and we are close to each other. But my accountant is my uncle. We are akin.” His two permanently employed laborer (3) are placed last on the scale of social proximity. Bashir explains this like that: “They are very important for carrying the heavy rice sacks. But they are replacable.” After his employees, Bashir turns to his suppliers. In this regard he builds two categories. The first group (4) comprises his five most important suppliers. All of them are placed very close to him. In fact, two of them are his uncles, who both own rice mills and perform large-scale rice trade. One lives in Bogra and one in Dinajpur (s. Map 3). But also the other beparis of this category are well known to him, since they share with him a long-lasting business history. In fact, Bashir’s father did already count them among his suppliers, which is the reason why no distinction is made between these two kinds of suppliers: “All of them are trustworthy. All of them are reliable.” The second group (5) comprises his other fifteen suppliers. Bashir places them in far distance to himself and attributes them with very low importance. The main reason for this is the fact that he deals with them only irregularly. With his close suppliers, on the contrary, he maintains permanent relations and performs regular transactions. In regard to his customers Bashir estimates that approximately 50 persons purchase rice in his store every week. From them, 15 to 20 are regular customers (8), while the rest are casual ones (9). Accordingly, he divides this group of actors into two. His argument for doing this goes like this: “The first group comprises my regular customers. They get rice on credit. This is important for them because they have large shops on their own. They need to purchase large quantities of rice and often they cannot pay for all bags at once.” However, according to Bashir, the most decisive aspect of this group is the following: “They have good will and they have name and fame.” All other customers do need to pay cash on delivery, and no mentionable social relation exists to them. Beside his employees, suppliers and customers, Bashir mentions four more actor groups that are of relevance for his business. First, he names four people whom he calls his business partners. The first one (10), most important to him, is his brother who runs the former store of their father, next door to Bashir’s store. The other three persons (11) are his cousins, who all run their own wholesale stores at Badamtuli: “In times of need, we help each other. We lend money or we borrow some.” Compared to his cousins, his brother is evaluated as being much more relevant for Bashir’s business. The second actor group that Bashir mentions is the bank (12): “My bank is important to me. I need the bank account for the money transfer to my suppliers”. As third actor group Bashir mentions the market association. In this regard, he distinguishes between the general secretary (6) and the rest of the market association (7). The market association (7) is assessed to be rather unimportant, but the general secretary (6) is evaluated as highly relevant: “He is a good person. Whenever we have a problem here at the market, he has the will and the capacity to help us. He personally discussed the issue of price hike with the government and saved us from interventions.”

The Resilience Of Food Wholesale Traders

139

On my query Bashir also discusses the role of DCC and the central government (14) for his business. He considers the DCC to be of no noteworthy relevance apart of its role in issueing trade licenses. For renewing his trade license, Bashir usually sends a peon (13) in order not to loose his personal time for waiting. For this service, his helper is paid 500.00 BDT, which is the same amount that the renewal of the trade license costs. In this regard, Bashir underlines: “The peon is worth every single paisha, because I don’t have to go there [to DCC] personally. This saves time and nerves.” The central government plays a role for Bashir only in exceptional cases, such as in times of rising prices. In 2007, for instance, the interim government ordered all wholesale traders in Dhaka to fix up price boards in their stores for more transparency: “At that time, they [the central government] put pressure on us. The price boards were no pressure. That was just a formality. But they called the office [of the market association] and tried to put pressure on them.” Apart from such activities, however, the central government is far away from Bashir’s daily live. This is the reason why this actor group is the only one that Bashir places outside the predefined circles that represent its social proximity to him. Source: Own interview (07.04.2009) Fig. 25 represents the outcome of the Venn-diagram that I drew together with Bashir. This diagram shows that Dhaka’s food wholesalers maintain social relations to various actors. The traders are connected not only to their suppliers and customers, but also to other wholesalers, to banks, market associations, and – if only indirectly – to representatives of the municipality and the central government. For presenting another merchant’s network and for comparing the trade of perishables and non-perishables, Box 6 illustrates the example of a fish wholesaler. Box 6: A fish wholesale trader in Dhaka and his business relations Gilgad is one of more than 270 wholesalers of Five-Star Fish Market at Kawran Bazar in the heart of Dhaka City. He is in his late 30s and comes from a small village (gram) not far from the city of Narail, located in the Southeast of the region of Jessore (s. Map 4). Gilgad moved to Dhaka in 1993. It is a long time since. Still, he does not regard Dhaka as his home (bari). That is the small village where he was born and where his parents live. He lives together with his wife and children in a small appartment in Mohammadpur (s. Map 2 and 5), some three kilometers away from his place of work. He has a son of eight years and a daughter of ten months. Gilgad went to school up to class 12, but from his perspective, schooling is only of secondary relevance for the success of his business. More important is what he learned from his father about trade in general. His father was a salesman for clothing and taught him everything he needed to know about how to manage a store. Gilgad started his business in 1993 with a vending stall at Sonali Market (one of the five compounds of Five-Star Fish Market, s. Map 11). He rented a second stall in 1997, and a third one in 2001. In 2007, he had the financial resources to

140

The Resilience Of Food Wholesale Traders

rent three further stalls. Today, he runs six fish vending sites with a total number of 20 employees. Each one is run by a manager, an auctioneer and a weigher. Two helpers rotate between the single stalls to make sure that there is always enough fish to be sold. For additional tasks, like for getting tea or cigarettes, two peons are employed on an irregular basis. The vending sites are located close to each other so that the workers can easily move from one stall to another according to requirements. Figure 26: The personal network of Gilgad Market Association Suppliers 12

8

Customers

Employees 5

7 4

3

2

11

10 6

9 13

1 18

Business Partner

15 14 Bank, Moneylender

Lessor, Land Owner

16

19

Government, DCC 17

Bashir

Scale of social proximity: close – – – – – – – – distant

Relevance for business very low – – – – – – – very high Source: Own draft based on survey 2009

After the fish market is closed and the vending sites are cleaned, Gilgad leaves the market and goes to his office, a room of approximately 25.0 sqm in the second storey of one of the market halls of Kawran Bazar. He uses the office for checking the accounts and for welcoming his business partners. For his employees, the of-

The Resilience Of Food Wholesale Traders

141

fice serves as permanent residence. In fact, many of them are married and while their families live in their village homes, they are migrant workers who send parts of their earnings home. By staying in the office, they can save a good portion of money. Also catering is cared for, as they together pay a maid for the daily purchase and preparation of food. In February 2009, Gilgad takes his time to explain me how his business works (s. Fig. 26). Similar to Bashir, Gilgad sees a clear hierarchy of his workers. From all his employees, he places one of his managers (1) first, giving him the highest score of importance. During the market hours, this manager is responsible for the same tasks as the other managers. But afterwards, he collects the records and earnings of all vending sites for the central book-keeping. As such, he has decisive control over the total income of Gilgad’s business. The six auctioneers (2) are placed second. From all other workers, they are most important due to their responsibility to realize good sales prices. As he explains, the auctioneers must be very experienced. They need to know the customers and their bargaining behavior, and they need to strategically manage the auctioning by forcing or reducing the pace. In third position are the managers of his remaining five vending sites and his six weighers (3). He underlines that all of them fulfill important tasks. Nonetheless, he evaluates them to be of lower importance compared to the hitherto mentioned employees. Interestingly, the night laborer (4), who are not directly employed by himself but by the market association, are assessed to be more important and socially closer to him than his own laborer (5). He explains this by the fact that the night laborer bear the responsibility for the unloading of the trucks and for the correct allocation of his supplies, a task that they perform independently at night when his workers are not yet at the market. The responsibilities of his own laborer are much less. After the explanation of the role of his employees Gilgad turns to his suppliers. He divides his 16 suppliers into three groups. Each group is regarded with different degrees of importance and placed in different distances to himself: “There are times, when I deal only with ten suppliers. But there are also times when I deal with up to 16 suppliers at the same time. Ten of my suppliers are very important for my business. They all deliver fish around the whole year. The others are far less important. I know them but we do business only irregularly from time to time”. In fact, the first (6) and the second (7) group of suppliers are those whom Gilgad provides with advance payments (dadon) that range from 50,000.00 to 150,000.00 BDT per year. As return service they are bound to deliver a certain amount of fish of a specific type in regular intervals. Every New Year, at hal katta (s. 6.1.3), the contracts are anewed. With four of them (6), very close bonds exisit, as they stem from Gilgad’s home district Narail. These four suppliers are those partners with whom he started his business. To share the same home district is seen as the fundamental pillar of these relationships. The relations to the six following suppliers (7) rest on a long joint business history. In fact, Gilgad deals with all of them for ten years and more. These long-lasting business relations and the shared experience are enough proof for the good will of each of them. The remaining six suppliers (8) are placed far behind. These agents constitute the

142

The Resilience Of Food Wholesale Traders

stock of additional beparis that can be harked back in times of need or that can be dealt with in case of good offers. As Gilgad underlines, these suppliers must be trustworthy as well for fullfilling mutual agreements. Nonetheless, these six latter suppliers are not as familiar to him as his other suppliers are. Gilgad mentions further that there are approximately 250 customers who come to his vending sites every day. “Until last year, I had a contract with the Central Jail. We closed a contract for the period of one year at a time when fish prices were low. After a couple of months, the prices rose so that the whole contract became a loss business. Today, I have some large-scale customers. But I don’t engage in long-term contracts anymore.” Gilgad divides his stock of customers into two categories. The first category (11) consists mainly of retail shop owners that can be located easily. All actors of this group are regular customers. They are well known to Gilgad and his employees, and they usually purchase large quantities of fish. There is usually not much bargaining with these customers, and most of them get fish on credit. The second category (12) consists of all casual customers who are not personally known to Gilgad and with whom no social relation exists. For purchasing fish, they need to participate the auctions. None of them gets fish on credit. Beside his employees, suppliers and customers, Gilgad mentions five more actor groups that are of relevance for his business performance. The first one is his brother (13), whom he calls his closest business partner. “In case I need a oneday credit, I can get the money from him without any interest rates”. The second actor is his bank (14) that he needs to transfer money his suppliers. He never took any credit from his bank yet. The third actor group is the DCC (16) and the central government (17). The DCC (16) is of relevance insofar as Gilgad needs to anew his trade license every year. The central government (17) plays a role for him, since he needs to pay tax. One of the responsible officers is his uncle (15), which is the reason why he locates him very close to himself. As fourth actor group Gilgad mentions the market association. In fact, there are six market associations at Five-Star Fish Market, one for each market compound, and one umbrella association for the market as a whole. For Gilgad, the main purpose of the association of Sonali Market (9) is to collect the wholesalers’ stall fees on behalf of the owners of the market ground, and to run an informal credit system. Each wholesaler usually pays a daily amount of 20.00 BDT into a common pool that is available as credit in case of need. “If one of us has a problem and he needs money, then all wholesalers hold a meeting and discuss his case. If his query is accepted, he gets a certain amount of money for a certain time – without interest. Sometimes we also sacrifice money. We give money without any chance to get it back”. The umbrella association (10) is of relevance for Gilgad mainly in its role to represent his interests politically. The last actor group that Gilgad mentions is the land owners of the market (19) and his lessor (landlord) (18). Actually, the market buildings of Kawran Bazar were built by the municipality to serve the merchants as offices. Nonetheless, there is another player (18) in between himself and the DCC (19) that is responsible for the building. In case of his vending stalls, the situation is likewise. The owners of the ground of Sonali Market (19) are in part the Railway

The Resilience Of Food Wholesale Traders

143

Department and in part the Roads and Highway Department. However, the market association is not paying this money directly to the governmental agencies. There is again an intermediary (18), who commands over this piece of public land and skims off the profits. Source: Own interview (23.02.2009) The Figures 25 and 26 show that the personal business relations of rice and fish wholesale traders are fairly comparable. In sum, eight actor groups can be distinguished that are relevant for the traders’ businesses, i.e. suppliers, customers, employees, business partners, the market association, banks or moneylenders, DCC and the central government, as well as land owners and lessors. For analytic purposes, I am going to distinguish the merchants’ business networks in a narrow sense and their extended business networks which connects them to the wider urban arena. Business networks comprise all those actors who are of interest for understanding wholesalers’ coping and adaptive capacities, namely employees, suppliers, customers, and business partners. The extended business networks comprise all other relevant actors and need to be considered for understanding the merchants transformative capacities. 6.2 MARKETS AS PRACTICE: FOOD WHOLESALE TRADERS’ COPING CAPACITIES 6.2.1 Delivery of Goods Markets are more than the abstract forces of supply and demand. A real market (Mackintosh 1990) is a small universe in itself, comprising hundreds of actors from the owners of the stores and their managers who keep the accounts, over the auctioneers who try to sell the goods for the best price, to the laborer who do the physical work. From the early morning hours onwards, markets in Dhaka are frequented by hundreds of customers who want to buy food either for direct consumption or for business purpose. During night time, the markets awake for a second time, when one truck after another arrives at the gates to replenish the stocks. In this part, the question is asked of how Dhaka’s wholesalers organize their business. By focusing on their daily practices, the general aim of the traders can be said to be the sale of as much rice and fish as possible at most favorable prices. This involves three types of coping strategies that are discussed on the following pages, namely the delivery of stocks, the sales of goods, and the organization of restocking. As it will be shown, there are considerable differences in how rice and fish markets work. Rice markets can be called “slow markets”. Since rice can be stored for month, there is no temporal bottleneck that would force wholesalers to sell their stocks within the shortest possible time. This is the reason why rice markets fairly follow the rhythm of the day. Usually, they are opened at 7.00 a.m. or 8.00 a.m. in the morning and last until 4.00 p.m. or 5.00 p.m. in the late afternoon.

144

The Resilience Of Food Wholesale Traders

Photo 31: Laborer rest for night turn Dhaka City, Badamtuli, 25.01.2008 Source: Photo by author

Photo 32: Laborer replenish rice stocks Dhaka City, Badamtuli, 25.01.2008 Source: Photo by author

Fish markets, on the contrary, need to be addressed as “quick markets”. In the absence of sufficient financial means for guaranteeing uninterrupted cold chains, the perishability of fish is most decisive for how they work. Delivery of fish is a just-in-time-business and the merchants are forced to sell their supplies rapidly, as they loose considerable value within hours. Thus, besides the problems of value and competition (Beckert 2009), the natural attributes of rice and fish and involved technologies of handling must be seen as most decisive factors for explaining the general organizational patterns of rice and fish markets in Dhaka. In regard to the delivery of goods to Dhaka’s wholesalers, the rice and fish business is generally organized alike. In both cases, the majority of goods are delivered by truck, the restocking takes place at night, and the suppliers (or the truck companies) bear the responsibility for the loads until they reach the market. However, there are differences between both businesses that are related to the handling of goods after they arrived. At Badamtuli – which is taken as example for the rice business – the first trucks arrive around 10.00 p.m.. Due to the limited space and the narrow lanes of the market (s. Map 9), it takes a while until all trucks are in the right position. The unloading does usually not start before 12.00 p.m.. The laborer stay over night in the wholesale stores to guard the rice sacks and the safe (s. Photo 31). When the lorries arrive, they organize themselves in small teams and unload the rice sacks, a mechanism that saves a lot of time. As Photo 32 shows, the trucks are unloaded by hand. In the absence of any lifting facilities, the workers carry the rice sacks on their heads. The sacks they deliver are counted by using thin twigs which the carriers drop to the ground, one at a time as they carry the sack to the storeroom. By means of this simple method, the truck driver and his helpers are able to keep track so that in the end, each particular store is delivered with the correct number of sacks. For one of these rice bags, the laborer earn 4.00 BDT. Usually, the restocking lasts until 3.00 a.m. to 6.00 a.m. in the morning. After the laborer took some rest, they are paid by the store owners who arrive at the market on the next morning. The wholesalers also bear responsibility to pay off the truck drivers and their helpers, even though the cost for transport are final-

The Resilience Of Food Wholesale Traders

145

ly borne by the supplier. In this way, the merchants assure to receive the right number of bags of the ordered rice type and quality.

Photo 33: Laborer unload frozen fish Dhaka City, Kawran Bazar, 29.01.2009 Source: Photo by author

Photo 34: Sleeping accommodation Dhaka City, Kawran Bazar, 29.01.2009 Source: Photo by author

At Kawran Bazar – which is taken as example for the fish business – the delivery of fish starts around 12.00 p.m. at night, when one truck after another comes to the market and stops on the southern side of the market on Panthopath Road (s. Map 11). The unloading of metal boxes (s. Photo 33), drums and baskets with frozen and fresh fish is done by night laborer who usually sleep and live on the markets – often under miserable conditions (s. Photo 34). The night laborer are not employed by individual wholesalers but work for the market as a whole (s. Box 6). As its name indicates, the Five-Star Fish Market at Kawran Bazar consists of five individual market compounds (s. Map 11) and each of these compounds has its own team of night laborer.55 Each team has a leader (sordar) who gives the instructions, who collects the service charge from the wholesalers (after the market is closed), and distributes the earnings among the workers. This system is not based on any written contracts but on oral agreements between the sordars and the associations of the respective markets. The laborer are paid per weight and can be said to earn on average about 100.00 BDT per night. A business card is fixed at each load that indicates the name and address of the respective businessman. In the absence of motorized vehicles or conveyor belts, the laborer use rickshaw vans to transport the fish from the trucks to the merchants’ vending sites. The night laborer are responsible for guarding the loads of fish until the owners arrive, a fact that explains why wholesalers do acknowledge them with high confidence (s. Box 6). Ice is delivered in blocks from nearby factories and the laborer manu-

55 At the time of the fieldwork, there were in total 78 night laborer at the market, 25 worked at Akota Market, 14 at Sonali Market, 14 at Jashim Market, 18 at Mukti Jogha Market and 7 at the Mosque Market.

146

The Resilience Of Food Wholesale Traders

ally crush it and put it on the piled fish for keeping it fresh.56 Immediatly after the night laborer’s work is done, the sale of fish is started. It needs to be mentioned that the laborer bear the hardest burden of all interviewed market agents. This burden stems from several factors: First of all, as Photo 32 and 33 showed, their work is highly demanding. Second, with 4.00 BDT per rice bag (that equals about 0.04 EUR) or 100.00 BDT per day (that equals about 1.00 EUR), their earnings are extremely low. Third, their irregular time slots for taking rest adds to their overstrained physical condition. All these factors in combination are reflected in the generally poor shape of the workers’ bodies which makes them distinguishable from managers and owners of the wholesale stores. As such, the bitter irony of Dhaka’s food system is that the people who carry the food for the entire city on their shoulders are those who are most vulnerable to food insecurity due to their low payment, demanding work loads and related health risks. 6.2.2 Sale and Price Building Each sale comprises basically three steps on Dhaka’s wholesale markets, i.e. the bargaining over 1) the quality of goods, 2) the price, and 3) the commission. In this regard, rice and fish markets work similarly. Nonetheless, both types of markets differ in the particular way of how sales are organized. In case of rice markets, most of the customers come to the wholesale stores in the morning hours until 12.00 a.m. to purchase goods for business purposes. After lunch time, the number of customers decreases. If during morning hours 20 to 30 customers buy rice at a particular store, it is usually not more than one or two in the afternoon – mostly people who buy rice for their private consumption. The small platform (godi) at the entrance of each store plays a crucial role for all steps of bargaining in the rice business. This platfrom serves to display the rice types on offer in form of samples, i.e. small heaps of grain that can be inspected by potential buyers (s. Photo 35); the owners, managers and accountants have their desks on this platform (s. Photo 36); and it is the place where the purchasing price and the commission is negotiated. Encompassed by lockers for documents and a safe, these platforms are the center of each individual rice store. Initially, the quality of rice is bargained over. This is usually done on the basis of samples that are displayed in front of the stores. Each sample consists of a specific rice type that stems from one particular supplier.

56 Alam et al. (2010) mention that unchlorinated river water is often used for making the ice. Given the fact that the water of Buriganga and other rivers around Dhaka is highly polluted, this practice means a severe danger for the food safety of urban consumers.

The Resilience Of Food Wholesale Traders

Photo 35: Displayed rice samples Dhaka City, Badamtuli, 13.11.2007 Source: Photo by author

147

Photo 36: Godi Dhaka City, Badamtuli, 07.04.2009 Source: Photo by author

For inspection, the customers take a handful of rice and hold it to the light in order to prove its color and shininess. They check the grain size and whether it stems from the variety or whether it is the result of the milling process. As it is stated by many traders, it is a popular practice of millers to mechanically break the grain in order to increase the sales price, because “good” rice in Bangladesh is said to be short grain rice.

Photo 37: Bonga Dhaka City, Kawran Bazar,04.02.2008 Source: Photo by author

Photo 38: Hook Dhaka City, Badamtuli, 13.11.2007 Source: Photo by author

Furthermore, a small portion is tasted in order to test the flavor, which is after all the most decisive factor that decides upon the willingness of the customers to buy

148

The Resilience Of Food Wholesale Traders

a certain rice type or not. Eventually, the potential buyers try to find out the proportion of discolored rice (mora chal). First, this is done by inspecting the displayed sample. But in the course of bargaining, the customers usually ask for getting a sample of rice from a specific bag they are about to purchase. For this purpose a sample taker (bonga), that looks like a long drawn-out gouge (s. Photo 37), is inserted into the bags in whose channel the grain is collected and pulled out. This process of tapping a bag is accompanied by the emergence of little holes through which grain is lost during handling. Even though the loss is insignificant, this procedure is often taken by the seller and the potential buyer as a clear sign that both parties are seriously interested in closing the deal.

Photo 39: Laborer lifting bag of 85 kg Dhaka City, Badamtuli, 25.01.2008 Source: Photo by author

Photo 40: Delivery of rice bags Dhaka City, Badamtuli, 26.02.2008 Source: Photo by author

The inspection of the grain quality goes hand in hand with the bargaining over the price that takes place at the same spot where the goods are displayed. The allotments are always whole rice bags (bosta) whose weight ranges from 50 to 85 kg. The price is negotiated either in the form of the price for the entire bag or in terms of the rice price per kg. Usually, the wholesaler makes the first offer that is often said to be the fixed price of the supplier (bepari or miller). Actually, the commission agent is not bound to his suppliers’ price. Since beparis and millers are physically absent, they heavily depend on the knowledge and experience of their commission agents in the city. While prices of beparis and millers mainly go by the general supply and by import prices, the wholesalers in Dhaka know the actual demand in the city. Due to recent developments in telecommunication, all information can be gathered and shared in a view minutes by mobile phone. As such, the starting price can be said to be based on a mutual agreement between the supplier and the wholesaler which is based on the general assurance of the bepari or miller not to stop the re-supplies and of the wholesaler to try to sell the supplier’s goods at the best possible price. Eventually, of course, the customers have some

The Resilience Of Food Wholesale Traders

149

scope of influencing the price, as it is their decision to buy the goods for the stated price or to go to one of the competitors. The purchasing price then evolves as the result of the negotiation of all three actors. This constellation leaves the wholesalers with some room for maneuver as they are able to make better offers to their major customers who purchase in large quantities, they are in a position to distinguish between regular and floating customers, and between those who get the bags on credit and those who pay in cash. After the purchasing price is negotiated, the wholesalers add their commission. Contrary to the public opinion and often also neglected by academic studies (cf. Rahman et al. 2008), Dhaka’s wholesalers charge for their services a commission not only from suppliers, but also from customers. This “double commission” is usually said to be a fixed amount of 15 paisha (0.15 BDT) that the wholesaler receives per kg that is sold, once from the bepari or miller who is responsible for the supply and once from the retailer who purchases the good. There are many people in Dhaka, also some of the wholesalers themselves, who call this practice an instrument of cheating. This stems from the fact that buyers can never be fully sure about what portion makes the purchasing price and what the commission. Since the option to line own pockets cannot be neglected, the commission from the buyer’s side must be seen as a second instance of bargaining and thus a second option of both parties to influence the price according to their wishes. However, it should not be overseen that the wholesalers’ commission from the buyer’s side seldomly exceeds a proportion of 0.5 percent of the purchasing price. When an agreement is reached, the wholesaler writes a receipt (memo) of which the buyer gets a copy. Afterwards, his laborer carry out the transport of the purchased bags to the vehicle of the customer. Given the weight of the rice bags and the absence of any lifting facilities – the workers usually use hooks (s. Photo 38) for this purpose – this is a demanding task (s. Photo 39). For this service they receive a service charge of 4.00 BDT per bag. From now on, the buyer is responsible for the transport. In case his vending site is close by, the wholesalers’ laborer can take over the complete transportation (s. Photo 40) not without being paid an extra service charge of around 8.00 to 10.00 BDT per bosta. In other cases, a CNG or a rickshaw is rented that usually takes 30.00 to 60.00 BDT for a ride of one to five kilometers. In times of heavy traffic congetions, this price can go up to 80.00 BDT or 100.00 BDT for one way. Extra waiting charges can come along. The sales at Dhaka’s fish markets work differently. Usually, the wholsalers’ employees arrive at the market already in between 3.00 a.m. and 4.00 a.m. to arrange the supplies to be displayed to the customers. Prior to their arrival, the market’s night laborer organize the unloading of goods and the distribution of loads to the stores. The fish is sorted by type, size and quality. As Photo 45 shows, this task of handling fish can be quite dirty. Since usually no protective clothing or single-use rubber gloves are used, it is also often unhygienic and thus endangers the food safety of the final consumers. After these initial preparations, frozen fish is usually displayed on iron plates (s. Photo 41), while live fish is put in small water-filled basins (s. Photo 42). For sale the fish is apportioned in certain allotments that basically depend on the size of fish. Small varieties (e.g. Punti, Mola,

150

The Resilience Of Food Wholesale Traders

Taki, Magur, Chingri) are usually offered per heap, mostly per maund (37.3 kg). Medium size fish (all the major and exotic carps, such as Rui, Catla, Mrigal, Kalbasu, Silver Carp, Grass Carp, and Mirror Carp, but also most cat fishes and snakeheads) is sold per hali (a batch of four pieces). Single large fishes are sold piecewise. Around 5.30 a.m., the market opens.

Photo 41: Displayed frozen fish Dhaka City, Kawran Bazar, 31.03.2008 Source: Photo by author

Photo 42: Displayed live fish Dhaka City, Kawran Bazar, 31.03.2008 Source: Photo by author

With the aim to reduce time, the general sales mechanism on fish markets is the open ascending price auctioning (English auction) that takes place directly at the vending stalls of the individual wholesalers (s. Photo 46). There is no extra accommodation for centrally organized auctions. By open outcry, the auctioneers try to arouse the interest of as many potenial buyers as possible. The auctioning starts when enough people are gathered around the vending stall. In some cases, the auctioneer may announce a price for high quality fish, but usually the first bid stems from the customers. As such, it is the customer (influenced by the commission agents’ auctioneers) who sets the price and not the supplier as it is the case on rice markets. The fish business is organized conversely. Participants bid openly; each subsequent bid must be higher than the previous one. The auctioneer repeats each bid and calls it out loudly in the presence of the potential buyers. Bids are usually made in the form of price per kg or for the full heap or hali. There are no limitations for how much higher a bid must be to be valid apart from the rule that it must be a full amount of BDT (not paisha). The auction ends when no participant is willing to bid any longer, at which the highest bidder becomes the buyer. Not all auctions proceed exactly alike since auctioneers have possibilities to intervene and manipulate. In case a heap of fish has not been sold, they can increase the offer by adding further fish. They can slow the pace for letting hesitant buyers to make their offer. They can accelerate the pace in order to avoid distress sales at the end of the market day. And they can even interrupt an auction and assign a

The Resilience Of Food Wholesale Traders

151

portion of fish to a well-known regular customer, even though others would be willing to pay a higher price.57 As such, the auctioneers bear a huge responsibility, a fact that makes them often those employees in which the owners of the vending sites puts most confidence (s. Box 6).

Photo 43: Weigher at work Dhaka City, Kawran Bazar, 31.03.2008 Source: Photo by author

Photo 44: Accountant at work Savar, Bank Town, 18.01.2009 Source: Photo by author

After the auction is over, the fish is put into a basket or drum to be weighted (s. Photo 43). In the case of halis or single fishes (or undefined heaps of small fish), it is only after the weighing process that the buyer really knows what amount he bought. As such, the auctioning is a tricky business, as there is only a short time for the potential buyers to inspect the fish and to decide upon a price. The large price variability of fish (cf. 4.3.3) is thus also rooted in the institution of auctioning (besides the differences of fish in quality and size). For the weighing service, the buyer is usually charged a fixed amount of 5.00 BDT, whatsoever the quantity of fish is. The bargaining of the commission is in the hand of another actor, the accountant.58 His task is to carefully supervise the whole process of auctioning and weighing. For this reason, he usually remains behind the auctioneers (s. Photo 44), sometimes heightened at the back of the vending stall to be able to overlook the entire scene (cf. Photo 29). He is responsible for noting down the price and quanity of fish that is sold as well as the name and address of the buyer. As Photo 29 57 However, this is only an exception. Most sales to regular customers are closed after direct mutual negotiations without any auctioning to take place. 58 All mentioned tasks, from auctioning over weighing to the book keeping of the sales, are usually done by individual workers who are exclusively responsible for this one task. The main reason for this is to save time, since the next auction can already be started when the fish of the previous auction is still to be weighted.

152

The Resilience Of Food Wholesale Traders

and 44 indicate, the task of keeping track of the numerous buyers and often several sales that take place at one and the same time should not be underestimated. It is the accountant’s responsibility to collect the sales price and the commission. For their services and costs, the wholesalers charge the buyers 3.0 to 4.0 percent of the sales price. Just like in the case of rice, charging commission can be seen as another step of bargaining, since it provides the accountant options to distinguish customers who are unknown from regular customers who usually buy in large quantities and get fish on credit. With 3.0 to 4.0 percent, the commission of fish wholesalers is much higher than that of rice wholesalers, which makes it more important for the final price. Once the sale is done, the buyer bears the responsibility to organize the transport just like in the case of rice. Due to the perishability of fish, the payment of extra waiting charges to CNG drivers and rickshaw pullers who queue up in front of markets is a common practice. The fish market closes in between 9.00 a.m. and 10.00 a.m., just when most other city dwellers start to work.

Photo 45: Helper amidst live fish Dhaka City, Kawran Bazar, 31.03.2008 Source: Photo by author

Photo 46: Auctioning of small fish Dhaka City, Kawran Bazar, 31.03.2008 Source: Photo by author

6.2.3 Accountancy and Restocking The wholesalers’ organization of resupply involves basically four different activities, i.e. the collection of money from customers, the accountancy, the money transfer, and the placing of new orders. All these four activities are usually pursued in the afternoon hours after the fish markets are closed and the number of customers on rice markets have decreased. Around 4.00 p.m., rice and fish wholesalers start to collect money from those customers who received goods on credit. Credits are paid back in cash either in the evening of the day of the transaction or

The Resilience Of Food Wholesale Traders

153

at the end of the week. Usually, an employee of the wholesaler is sent to collect the money. However, it happens that buyers are not liquid for a longer time so that their debt accumulates. For this reason, there is the rule to clear all accumulated debt each New Year (in Bangladesh, this is 14 April) when the old accounting books are closed and new books (hal katta) are opened. For the customers of wholesalers it is important not to miss this date as they would lose face if they did not appear. For this reason, they usually come personally to the wholesalers, often with small gifts such as the locally famous milk-based sweets (mishti), to pay their debt and to show their good will. Thus, hal katta plays an important role for the business relations of wholesalers and their customers. Once the money is collected, the arotdars do the accountancy for that particular day.

Photo 47: Manager at work Dhaka City, Jatrabari, 06.02.2008 Source: Photo by author

Photo 48: Account books Dhaka City, Krishi Market, 10.03.2008 Source: Photo by author

For this purpose, they use four (sometimes five) different books (s. Photo 48). First, all memos are collected in the memo katta, the memo book (s. Photo 47). From their, sales are transcribed to the cash book (chota katta). Basic information are the quantity of rice and fish, the name of the buyer and the actual sales price. The most important information, though, is the absolute value of each single transaction and the method of payment. The third book is the inventory or stock book (tali katta) that is only used by rice wholesalers since fish wholesalers do usually not command cold stores that would make such a measure necessary. The stock book of rice merchants provides information about the input and output of rice sacks according to the ownership of bags and rice types. Each supplier usually gets one row or a full page in which stocks and sales are entered. In this way, the wholesaler keeps up-to-date in regard to his current stocks. The fourth book, called supplier book (bonash katta), is for the calculation of the commission from the suppliers’ side. Like in the inventory, the total amount of daily sales is entered according to ownership which provides information about the total amount of

154

The Resilience Of Food Wholesale Traders

commission that each supplier needs to pay. Both, rice and fish wholesalers keep this type of books. Finally, there is a book called dag katta which literally means a book of something like a mark or a kerf. Besides the bonash katta, this book is most important as it is used to calculate the average price per rice and fish type and the average profits over a longer period. Based on this information the wholesalers are in a position to see which rice and fish type is most lucrative for them at what time, which enables them to adjust their business accordingly. Based on the data of their account books, the wholesalers calculate their profits and place new orders. There are sometimes representatives of large-scale suppliers available at the market for face-to-face talks, but in most cases the day-today communication between the arotdar and his suppliers is done by phone. Personal visits of both parties usually happen not more often than once or twice a year. On these occasions, the supplier is usually hosted by the arotdar and vice versa, as it is the best chance to renew their oral agreements, to prove the mutual good will and to harden the business relations. It is said that suppliers do adhere a clear hierarchy by visiting their commission agents sequentially according to their relative importance for their own business. The earnings of the day are brought to a bank in cash where they are deposited on the merchant’s bank account. From there they are wired to the bank account of the supplier via telegraphic transfer (TT). The telegraphic transfer is a means to send money from one bank account to another that works today like a fax or the internet. Its name stems from a time, when a telegraph was used for this process. Even though there are no such telegraphs involved anymore, the use of the term remained until today. Telegraphic transfer usually involves a small fee that is charged from the sender and from the recipient. Financial transactions across the national border are organized in a similar way. However, for transboundary exchange, an additional import permit and a “letter of credit” (LC) is necessary. Usually, a customs clearing and forwarding agent is assigned to obtain the import permission from the government and to calculate the dues that the importer needs to cede.59 The letter of credit is an official document that the applicant – in this case, the wholesaler in Dhaka who wants to import goods from abroad – gets from his bank by depositing a certain amount of money. Such a transnational deal involves basically five actors, i.e. the buyer, the bank of the buyer, the seller, the bank of the seller, and the carrier: The wholesaler (buyer) in Dhaka is in contact with a supplier (seller) abroad. Before the closing of a deal, both parties meet personally, they send a representative to each other, or they exchange little samples of goods on offer. After an agreement is made, the 59 The import of rice (“rice in the husk”, “husked (brown) rice”, “semi-milled or wholly milled rice” and “broken rice”) is exempted from customs duties. Only the “para-tariff” (World Bank 2006b: 21) of 5.0 percent advanced income tax must be paid (s. http://www.nbr-bd.org/ operative_tariff/SECTION2/CHAP10.pdf). For importing fish, 25.0 percent customs duty and 28.0 percent para-tariffs must be paid, i.e. 15.0 percent value added tax, 5.0 percent advanced income tax, 5.0 percent regulatory duty, and 3.0 percent advance trade value added tay (s. http://www.nbr-bd.org/operative_tariff/SECTION4/CHAP16.pdf).

The Resilience Of Food Wholesale Traders

155

wholesaler in Dhaka authorizes his bank (bank of buyer) to issue a letter of credit in the specific amount of the deal to the supplier. Once the supplier has received the wholesaler’s letter of credit, he consigns the requested goods to a carrier or a clearance and forwarding agent in exchange for a bill of lading (BL). Afterwards, he forwards this bill of lading to his own bank (bank of seller) that credits the buyers payments to his account. The bill of lading is then handed on from the bank of the seller to the bank of the buyer and finally to the buyer himself who needs to pay it. After the buyer has paid, he provides the bill of lading to the clearance and forwarding agent who calculates landing charges, customs duties, taxes and the insurance, and finally hands over the goods.60 While the above mentioned tasks of money collection, accountancy and money transfer are executed more or less similarly by rice and fish wholesalers, the ways how the merchants arrange their re-supplies differ significantly. In the rice business, today, the main share of agreements are made for each truckload (called tip) anew. This does not mean that no long term relations would exist. It only means that the concrete conditions are negotiated tip by tip and all involved financial transfers are organized alike. For this reason, the arotdar in Dhaka and the supplier in the rural area compare prices in Dhaka with purchasing prices in rural areas, the milling and transport cost, and the import prices at the BangladeshIndian border. Based on this information, the sales price is orally agreed upon for each and every rice type that is delivered. These negotiations in between the commission agent and his supplier are more time consuming than those in between the arotdar and his customers, because it is the supplier who dominates the price building while the arotdar needs to keep in mind his customers’ expectancies. This is affirmed by Shahin, one of the rice wholesalers at Kawran Bazar: “There is only little bargaining with my customers. They just come, ask for prices, and decide upon it. But I negotiate a lot with the agency [the supplier].” Source: Own interview (04.02.2008)

After the agreement is made, it is in the hand of the commission agent to fulfill the expectations of his supplier. This involves a considerable amount of trust between the two parties. The supplier sends the goods at once, while the selling goes piece by piece. Until all goods are sold, his physical and financial resources are thus “locked” in the storerooms of his commission agents. Likewise, the arotdar needs to trust his supplier to get the right amount of the correct type and quality of rice that both agreed upon. If wrong quantities or a wrong rice type is delivered, the wholesaler would inform his supplier and the sales price would be renegotiated. The organization of re-supplies admits the rice wholesaler some room for maneuver. This is not related to cheating with the accounts (even though this cannot be fully forclosed by the supplier). It is related to the question of when the whole60 It should be added that – according to the own survey – not more than 1.0 percent of Dhaka’s rice wholesalers and 3.7 percent of fish wholesalers possess a letter of credit, because it involves a good standing with the bank and financial resources that only few, powerful players can afford.

156

The Resilience Of Food Wholesale Traders

saler pays his supplier. The dag katta allows him to distinguish between more or less profitable rice types. In case of less profitable rice or when he is not sure whether he will be able to sell the full amount of bags at the price that the supplier aims at, he transfers the money piece by piece, that is every evening or once a week. But in case of more lucrative rice or when he is sure to sell the entire load at a price that might even exceed what the supplier had in mind, he pays for the entire truckload at once (given that he commands sufficient financial resources) and sells the bags on his own account. By this practice, the narrow definition of commission agent obviously becomes blurred. To sum up, in the rice business, it is usually the supplier who becomes active beforehand by delivering goods in advance, while the commission agent pays for the supplies by installments. Rashid, rice trader at Babubazar, describes it like this: “It is like a chain. I get rice on credit from my suppliers. And I give rice on credit to my customers.” Source: Own interview (04.04.2009)

In the fish business, in contrast, it is the wholesaler who usually provides the supplier with payments in advance, that are called dadon. Dadon can be said to be a derivate instrument that takes the form of a contract between the supplier and the commission agent. This contract specifies the amounts of goods that are delivered at a particular date for a particular period. The amounts of these advance payments range widely from 50,000.00 to 2000,000.00 BDT that are paid for an entire year or more. This equals a proportion of 10.0 to 50.0 percent of the value of fish the contract is agreed upon. As return service, the supplier is bound to deliver fish to his commission agent in regular intervals. As such, fish wholesalers are connected to their suppliers by two parallel financial cycles. The first one comprises the financial transactions for each single truckload – just like in the case of the rice business. The second one is dadon that is paid at the beginning of a longterm agreement, and that is repaid after this contract expired. Such kind of future mainly serves the purpose to hedge the business of both involved parties. It does not so much serve as an instrument to reduce price fluctuations. In case of the wholesaler, dadon serves to assure preferably frictionless resupplies which enable the trader predictability. In case of the supplier, it serves to increase his liquidity which is necessary for him to organize fish from producers.61 The discussion above has shown that Dhaka’s wholesalers are embedded in two resource cycles, one that connects them to their suppliers and one that connects them to their customers. These resource cycles comprise the flow of goods and the flow of finances. In case of fish, the financial flows are aligned in both directions, as the commission agents provide credit to their suppliers (dadon) and to their customers. In case of rice, credit is provided only to customers. In this regard, two things must be added. First: The payment of dadon was also a common practice in the rice wholesale business in Dhaka some decades ago, as the 61 Suppliers usually need to provide gher owners with advance payments – similarly pursued in form of dadon – who need them to cover their expenses.

The Resilience Of Food Wholesale Traders

157

arotdars know to report. Nonetheless, with few exceptions, it is not practiced anymore in Dhaka. Second: It must be mentioned that fish wholesalers do not pay dadon to every one of their suppliers, but only to a certain number. To the rest of their suppliers, they maintain relations that function similarly as those of rice wholsalers. As it has been said before, reality is diverse. 6.3 MARKETS AS NETWORKS: FOOD WHOLESALE TRADERS’ ADAPTIVE CAPACITIES62 6.3.1 Exposure and Sensitivity to Business Risks From a long-term perspective, the aim of wholesalers is to run their business preferably frictionless on as many days per year as possible and thus to safeguard the survival of their enterprises (Fligstein 1996). This long-term goal, however, is jeopardized from time to time as certain threats put the wholesalers’ businesses at risk. On the next pages, two questions are raised. First: What types of risk are Dhaka’s rice and fish wholesalers exposed and susceptible to? And how do Dhaka’s rice and fish wholesalers adapt to these risks? For studying these questions empirically, I proceeded twofold. First, I organized four group discussions on the particular topic of business risks, two with rice and two with fish wholesalers, in order to grasp the emic perspective of the merchants. Afterwards, I inquired the identified risks by means of the standardized survey in order to get information on their general frequency, their perceived severity, and the tradesmen’s strategies to overcome them. In this section, the focus will be on the traders’ networks. In the course of fieldwork, three types of business risks were identified: 1) Risks that are related to customers; 2) risks that are related to suppliers; and 3) risks that are related to the delivered goods. The first type of risk is related to behavior of the wholesale traders’ customers. Interestingly, the most frequent risk and the one perceived to most severly threaten the performance of the interviewed tradesmen was the default of credit paybacks by customers. 88.8 percent of rice traders and 95.0 percent of fish traders faced this kind of problem at least once in the course of the last three years (s. Fig. 27 and 28). A rice trader at Malibag explains what it is about: “The main problem is the payback behavior of our customers. Many customers pay for the bags little by little. Sometimes, they cannot pay the money back in time. So, in the end, I run out of money by myself and cannot pay my suppliers.” Source: Own interview (04.02.2008)

In this regard, a participant of a group discussion that was organized together with the market association of Badamtuli Rice Market adds:

62 Parts of this section have meanwhile been published in Keck et al. (2012).

158

The Resilience Of Food Wholesale Traders “A credit [to customer] can easily sum up to half a lakh [100,000.00] Taka. And then, sometimes, the customer vanishes and the money is gone. If this happens more often than once, it can break our neck.” Source: Own interview (08.04.2009)

The second type of risk is related to the supply of goods. In this regard, one has to keep in mind that all risks that are related to the acquisition of rice and fish from rural producers affect the commission agents only indirectly, since millers and beparis are prepended. The same holds true for any problems that are related to the transport of goods from their place of origin to Dhaka’s markets. One participant of a group discussion that was held at Merul Badda Fish Market explains this: “Sometimes there are transportation problems. The reason for that might be dense fog, water logging on the roads, long waiting periods at Jamuna Bridge, a road accident, or a damaged bridge. All these incidents are first of all the problem of the bepari, because he is responsible for the fish until it reaches the market gate. However, in the end, I loose as well, because I have less amounts for sale.” Source: Own interview (01.04.2009)

In fact, for fish merchants and their rapid markets, already the delay of supplies has severe negative impacts on their business performance, as prices decrease remarkably in the course of only a few hours (cf. 4.3.3). In the absence of alternative suppliers, the default of a single truckload can even force traders to leave their vending stores closed for this day. Such a default can basically have two reasons. As it is stated by the interviewer above, the first reason is rooted in a number of different uncertainties that might happen during transport. The second reason is that the supplier is not able to deliver goods because he himself is not in the position to purchase necessary amounts from local markets. As fish merchants mention, a single supplier is usually able to deliver fish not more often than on 200 days per year. Even though dadon is paid in advance, the wholesaler in Dhaka goes away empty-handed if he is not able to get supplies from alternative sources in these days. The survey revealed that 84.0 percent of rice arotdars and 93.4 percent of fish arotdars experienced problems with late supply trucks. The default of a single truckload due to transport related incidents (e.g. traffic jam, waiting periods at ferry, etc.) happened to 33.0 percent of rice traders and to 54.0 percent of fish traders. 81.6 percent of the fish merchants faced the problem that one of their suppliers was not able to supply goods for some days despite the fact that dadon had been paid in advance. Since the payment of dadon is widely abandoned in the rice business, only 4.4 percent of rice merchants mentioned this kind of problem. 45.6 percent of fish arotdars had to leave their business closed for some days because of such supply cuts. Because of their warehouses, only 1.0 percent of the rice traders faced this problem (s. Fig. 27 and 28). The third risk type is the deterioration of goods. Interestingly, this problem was mentioned predominantly by rice traders. In fact, there are cases that rice bags become wet on the journey to Dhaka. If they are not properly dried, the infiltrated moisture can lead to fungal decay which bears the risk to spread throughout the entire batch of stored bags in the wholesalers’ storerooms. 33.5 percent of rice

159

The Resilience Of Food Wholesale Traders

wholesalers had to deal with the delivery of deteriorating or already deteriorated goods in the last three years. In case of fish, the loads can deteriorate on their way if proper cooling or sufficient water and oxygen is not guaranteed. Surprisingly, only 7.5 percent of fish wholesalers had to deal with this type of adversity. 63 Figure 27: Rice wholesale traders’ exposure to business risks Default of Credit Payback by Customer

Delay of Supply Truck

Delivery of Deteriorated Goods

Default of Truckload

Default of Supply after Payment of Dadon

Business Closure for a Day due to Supply Cuts 0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Source: Own draft based on survey 2009–2010 (n=203)

By comparing Fig. 27 and 28, one can clearly speak of a higher exposure of fish merchants to the mentioned business risks. But how severe are these incidents to the traders? In order to understand the susceptibility of Dhaka’s rice and fish wholesalers, the interviewees were asked to appraise the enumerated business risks in regard to their frequency, the perceived severity of their impact, and the time and effort that is necessary to cope with them. Four questions were raised in the survey. The first question was: “How often did this incident happen in the course of the last three years”? Five response categories were given, i.e. “never” (that equals a value of 1), “once” (2), “2–3 times” (3), “4–5 times” (4) and “more than 5 times” (5). The second question was: “How do you assess the severity of the impact of this/these event/s on your business performance”? Possible answers ranged from “not severe” (1) to “highly severe” (4). As third question, it was 63 After the fieldwork, an additional issue became acute that is related to the handling of fish. Residua of formalin were found, especially in import fish. Formalin is generally used for the preservation of tissue. In aquaculture, it is used to treat fish parasites, which bears the risk that residua can end up in the food chain. This implies a high risk to the final consumer and to food safety in general, since concentrations of formalin are toxic to the human body. In Dhaka, “formalin-laced” fish was detected on the city’s markets, which caused a chorus of public outrage in which traders were accused to medicate the perishable in order to make it look more fresh. The english speaking newspaper The Daily Star reported on this issue.

160

The Resilience Of Food Wholesale Traders

asked: “After this/these event/s, how long did you need to recover from its negative impact”? Answer categories were: “at most a week” (1), “a month” (2), “several months” (3), “a year” (4) and “more than a year” (5). Eventually, it was asked: “After this/these events, how much extra management effort did you need to overcome the problem”? Answers ranged from “no extra effort” (1) to “great extra effort” (4). Fig. 29 shows the results of this appraisal. Depicted are the mean figures of rice and fish wholesalers individually. While the severity and frequency of the mentioned risks are plotted on the x- and y-axis, the recovery time and the management effort are expressed in a combined way as “cost of coping” that is indicated in the diagram as symbol size. This cost of coping are calculated by summing the scores of recovery time (1 to 5) and the scores of management effort (1 to 4), which results in an index that ranges from 1 to 9. Figure 28: Fish wholesale traders’ exposure to business risks Default of Credit Payback by Customer

Delay of Supply Truck

Default of Supply after Payment of Dadon

Default of Truckload

Business Closure for a Day due to Supply Cuts

Delivery of Deteriorated Goods

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Source: Own draft based on survey 2009–2010 (n=234)

By comparing Fig. 27 and 28, one can clearly speak of a higher exposure of fish merchants to the mentioned business risks. But how severe are these incidents to the traders? In order to understand the susceptibility of Dhaka’s rice and fish wholesalers, the interviewees were asked to appraise the enumerated business risks in regard to their frequency, the perceived severity of their impact, and the time and effort that is necessary to cope with them. Four questions were raised in the survey. The first question was: “How often did this incident happen in the course of the last three years”? Five response categories were given, i.e. “never” (that equals a value of 1), “once” (2), “2–3 times” (3), “4–5 times” (4) and “more than 5 times” (5). The second question was: “How do you assess the severity of the impact of this/these event/s on your business performance”? Possible answers ranged from “not severe” (1) to “highly severe” (4). As third question, it was asked: “After this/these event/s, how long did you need to recover from its nega-

161

The Resilience Of Food Wholesale Traders

tive impact”? Answer categories were: “at most a week” (1), “a month” (2), “several months” (3), “a year” (4) and “more than a year” (5). Eventually, it was asked: “After this/these events, how much extra management effort did you need to overcome the problem”? Answers ranged from “no extra effort” (1) to “great extra effort” (4). Fig. 29 shows the results of this appraisal. Depicted are the mean figures of rice and fish wholesalers individually. While the severity and frequency of the mentioned risks are plotted on the x- and y-axis, the recovery time and the management effort are expressed in a combined way as “cost of coping” that is indicated in the diagram as symbol size. This cost of coping are calculated by summing the scores of recovery time (1 to 5) and the scores of management effort (1 to 4), which results in an index that ranges from 1 to 9. As one can see, the general logic of the wholesalers’ risk appraisal is that more frequent risks are evaluated as less severe in their impact, while less frequent risks are assessed to be more severe. At the same time, less frequent risks do usually involve higher cost of coping. As such, a delay in the delivery of goods happens more often than, for instance, the supply of deteriorated goods. However, the delivery of deteriorated goods is assessed to be more severe in its impact on the wholesalers’ business performance and it urges the trader to spend more time and effort to overcome it. Figure 29: Rice and fish wholesale traders’ susceptibility to business risks Fish wholesalers (n=234) Rice wholesalers (n=203) A: Default of credit payback by customer B: Default of delivery by supplier after payment of dadon in advance C: Late delivery of goods D: Default of truckload E: Shop closure for a day due to supply cuts F: Delivery of deteriorated goods Costs of coping: ______7.05 _______2.82

Source: Own draft based on survey 2009–2010

The approximation curves show that fish wholesalers do generally face more frequent and more severe risks in comparison to rice wholesalers. Three examples shall be explained in detail. First: On average, every single fish wholesaler was

162

The Resilience Of Food Wholesale Traders

subject to delayed truckloads more often than five times in the last three years (4.58). Rice wholesalers were subject to this kind of risk only four to five times (3.96). Due to the limited time on Dhaka’s fish markets, a delayed truck was evaluated by fish merchants to be a “severe” problem (2.52), whereas it was only “slightly severe” (1.90) for rice merchants. Finally, a delayed truck was said to cause fish traders considerable cost of coping (4.12), whereas the time and effort that rice traders need to deal with this type of risk were stated to be comparatively low (2.82). Second: Fish arotdars were forced to leave their stores closed for some days at least two to three times in the last three years (2.51). Because of their warehouses, rice arotdars hardly ever faced this problem (1.03). Nevertheless, to all traders who faced this problem, it meant “severe” losings (2.77 in case of fish and 2.51 in case of rice) and considerable costs of coping (4.05 in case of fish and 4.50 in case of rice). Third: Despite all difficulties in maintaining cold chains, deteriorated goods reached Dhaka’s fish markets only in very rare cases (1.19). In case of rice merchants, this happended at least once in the period of the last three years (1.57). However, rice traders evaluated this problem to be only “slightly severe” (2.43), whereas it meant a “severe” risk for fish traders (3.29). In both cases, the costs of coping were addressed to be considerable, but these costs were higher for fish arotars (5.18) than for rice arotdars (4.30), since their reputation is much more dependent on the freshness of their goods. Two business risks turned out to be exceptional in this context. The risk of customers who do not pay back their credit was evaluated to be not only highly frequent, but also highly severe in its consequence. This risk befalled rice wholesalers on average four to five times (4.16) and fish wholesalers even more than five times in the last three years (4.66). To both business types, it meant a “severe” risk (3.22 in case of rice and 3.11 in case of fish). In case of rice, this risk was associated with the highest general cost of coping (7.05). But efforts and time of fish merchants to cope with this problem were considerable (6.10) as well. The risk that fish wholesalers evaluated to be most severe (3.42) was the longer-term default of supplies despite preceding dadon payments. This risk was stated to demand the highest cost of coping (7.05) from the fish traders. The extraordinary positions of these outstanding risks are seen to be an indication for their importance for the everyday business of the merchants. Given the wholesalers’ widespread lack of insurances64, how do they adapt their enterprises to these kinds of risk?

64 According to the own survey, not more than 9.7 percent of rice wholesalers and 7.0 percent of fish wholesalers have contracted an insurance for business purposes.

The Resilience Of Food Wholesale Traders

163

6.3.2 Trust in Business Relations As revealed by the Venn-diagrams (s. Box 5 and 6), Dhaka’s rice and fish merchants maintain close bonds to a stock of regular suppliers and customers who guarantee delivery and sale of goods throughout the whole year. Such strong ties (Portes 1998) to regular business partners involve other price building mechanisms and other methods of payment and need thus to be distinguished from pure business relations that the merchants maintain to the rest of their suppliers and to floating customers. In order to understand the difference between regular and occasional business partners, it is necessary to reveal the underlying rationale of the involved actors and the degree of their mutual knowledge, that make up the “glue” of these relations. The aim of this part is to uncover this glue by studying the meaning or “content” (Beckert 2007) of the wholesalers’ business relations. According to Dhaka’s rice and fish merchants, the entire wholesale business rests on trust-based relations. This was expressed by numerous interviewees during fieldwork, such as by Yunus, a fish wholesaler at Kawran Bazar: “Our entire business works on the basis of trust.” Source: Own interview (30.01.2009)

However, a closer look reveals that there are distinct differences in the underlying governance modes, depending on whether regular or occasional business partners are concerned. This fact finds its expression in the everyday language of the merchants as they characterize the relations to their regular suppliers and customers with various words such as “trust” or “faith” (biswas), “commitment” (protigha, protisruti), “courage” (utshaho) or “honesty” (dan). By contrast, for characterizing relations to their occasional suppliers or customers, interviewees state that only “pure business (bebsha) relations” are maintained or that no relation exists at all. The different governance modes of the wholesalers’ business relations can be comprehended by resorting to Lewicki & Bunker’s (1996) typology of trust (s. 2.4.3). As it will be shown, all three types of trust, i.e. calculus-based, knowledgebased and identification-based trust, are prevalent features of Dhaka’s food markets: Relations to wholesalers’ occasional suppliers and customers are well described by the “calculus-based” type of trust (ibid.). These relationships are characterized by a large social distance of the business partners, which is the reason why all transactions are entirely agreed upon on the basis of price negotiations. Transactions assume a minimum of mutual trust insofar as the wholesaler relies on the customer’s immediate and correct payment after delivery, while the customer relies on the wholesaler’s promise not to manipulate the weight of purchased goods or to add lower quality (or even already rotten) items. Between the wholesaler and the supplier, the need for a minimum of trust is even more obvious, since both tradesmen interact across a considerable spatial distance, which necessitates them to rely on the good will of one another. The main precondition of a transaction between otherwise not well-known businessmen to take place is to bring immediate rewards to both involved actors. This makes it necessary for the

164

The Resilience Of Food Wholesale Traders

merchants to carefully compare the involved risks with expected profit margins. In the absence of other factors, this calculation of risk and profit is the decisive pillar of these relations. Those relations that have matured and have become consolidated over the course of many years, are in line with Lewicki & Bunker’s (ibid.) notion of “knowledge-based trust” that is grounded in the parties’ sufficient knowing of one another. In these relations, the joint business history offers both exchange partners plenty of first-hand experience that makes their actions predictable to one another. A statement of Kazi, owner of a fish wholesale store at Kawran Bazar, reveals the difference between calculus-based and knowledge-based relations, which holds relevance for the wholesalers’ dealings with both their suppliers and their customers: “I usually call my permanent customers to sell them their preferred fish. Floating customers come to me independently. They buy fish only in case they can make profit. I have a pure business relation to them. With my permanent customers this is different: I tell them the purchase price and my expenses. I add my personal profit and tell them the price. And they have to buy [the fish] for that price. Even though other wholesalers here at the market have better prices, they would never buy fish from someone else. Because they have one hundred percent trust in me!” Source: Own interview (03.04.2009)

This statement should not be misinterpreted in the sense that an attitude of profit maximization would only be inherent in calculus-based relations. In fact, calculations on profit margins and related risks are also pursued in knowledge-based relations. The difference of these two types of relations is rather rooted in different time horizons that become bearing. Calculus-based relations follow a short-term logic which puts the focus on nothing but immediate rewards. The rationale of knowledge-based relations, in contrast, is to prolong the specific relation and to assure a preferably frictionless business in the long run. In this latter type of relations, the traders share much more fine-grained information with their exchange partners, e.g. on the acquisition of goods, on their personal expenses, or on hidden quality features (cf. Uzzi 1996). This sharing of knowledge is this relations’ fundamental reproduction mechanism. Some of the wholesalers’ business relations are consolidated in such a way that “identification-based trust” (Lewicki & Bunker 1996) becomes the fundamental rationale of both involved business partners. Identification-based trust means that “the parties effectively understand and appreciate the other’s wants; this mutual understanding is developed to the point that each can effectively act for the other” (ibid.: 122). By following Li (2007), such kind of relations can be said to be associated with a “sentimental” and an “instrumental” value enhancement. The sentimental value of these relations stems from their inherent versatility. This became obvious when drawing the Venn-diagram together with Bashir (cf. Box 5). On the question of how he perceives his five most important suppliers he answered briefly:

The Resilience Of Food Wholesale Traders

165

“To some extent, they are businessmen. But they are also my brothers [bhai, which literally means ‘brother’, but is also used to express social proximity and appreciation of males older than the speaker]. They are both, brothers and businessmen.” Source: Own interview (07.04.2009)

As such, Bashir does not perceive the two roles of “businessman” and “confidant” as two opponents. He does not draw a clear cutting line that would separate them. It is rather both roles that are accepted as being constitutive for the relationship. Calculus-based relations are, in essence, uni-dimensional with rather clear and fixed role perceptions, such as that of the “seller” and the “buyer”. Identificationbased relationships, on the contrary, are much more diversified and, in essence, multiplex (Boissevain 1974). By means of ongoing, iterative interaction, the former “supplier” can thus become a person with various social roles such as “business partner”, “bread earner”, “trusted friend”, and “supplier”, each one bringing up another logic and code of conduct. It is this comparatively rich mutual knowledge of one another that bears the potential to bring about the instrumental value of those relations. This instrumental value is a specific commitment to prolong the relationship (Li 2007), expressed in a “joint problem solving attitude” (Uzzi 1997). This attitude was addressed by many interviewees. In case of rice, for instance, wholesalers stated to allow some of their customers to postpone the payback of installments in times of financial bottlenecks. If necessary, they would even increase the granted credit that is to be paid back at the end of the year. Likewise, in case of fish, it is a popular practice of suppliers in times of supply shortfalls, to distribute the limited quantity of available goods among their most valued commission agents only, whereas orders of others are ignored. Such strong ties play a pivotal role for wholesalers to secure their supplies. Ruhul provides an explanation for that: “In case my supplier fails to deliver fish, I am in trouble. Because I am not able to do business at that day. Sometimes all of my three suppliers are not able to deliver fish at the same day.” Source: Own interview (03.04.2009)

Supportive efforts on part of a bepari to safe a wholesaler in Dhaka from being forced to leave his store closed for a day, or of a wholesaler to bail one of his customers out of his temporary financial bottleneck might be asymmetric in the single case. In sum, however, such efforts help to assure the business performance of both, the wholsalers and their business partners, since both act on the assumption that their partner will fulfill his obligation and return the favor next time. Rashid aptly sums up this point: “I am reliable to them [the supplier], and they are reliable to me.” Source: Own interview (04.04.2009)

In this regard, it must be added that a strong feeling of belonging that might be based on kinship, religion or a shared place of origin is a potential but not a necessary precondition for the emergence of identification-based trust. In fact, kinship plays an important role in fish wholesalers’ business networks. Relatives were found among the arotdars’ regular suppliers in 18.2 percent of the cases and regu-

166

The Resilience Of Food Wholesale Traders

lar suppliers who share the same home district with arotdars were found even in 42.0 percent of the cases. In case of rice, relatives were found among the wholesalers’ regular suppliers only in 3.4 percent of the cases; regular suppliers who share the same home district with the traders were found only in 6.3 percent of the cases. A similar pattern was found in regard to the wholesalers’ customers.65 Table 26: Assessment of rice wholesale traders’ business relations “Please consider the relations to your customers and suppliers: What role plays ... for these relations?” In Relations to Regular In Relations to OccaDifference Business Partners sional Business Part(Mean) (Mean) ners (Mean) The quality of goods Very important Very important 0.00 (1.00) (1.00) To push prices for perImportant Important 0.14 sonal gain (2.18) (2.32) To follow same religion Not important Not important 0.15 (3.81) (3.96) To be relatives Not important Not important 0.28 (3.60) (3.88) To share same home Not important Not important 0.32 district (3.67) (3.99) The reputation of your Very important Important 0.71 partner (1.15) (1.86) To get credit* Important Not important 1.12 (2.48) (3.60) To understand other’s Important Rather not important 1.47 business* (2.01) (3.48) To do each other a faVery important Rather not important 1.56 vor* (1.47) (3.03) To help in difficult Very important Rather not important 1.59 times* (1.50) (3.09) The balance of favors* Important Rather not important 1.84 (1.59) (3.43) To give credit* Very important Rather not important 1.99 (1.42) (3.41) Source: Own draft based on survey 2009–2010 Note: The single-factor (one-way) analysis of variance (ANOVA) revealed a significant difference between relations to regular and occasional business partners in regard to the inherent “joint problem solving attitude” with p