Planting to feed the city? Agricultural production, food security and multi-spatial livelihoods among urban households in Ghana Ayerakwa, Hayford Mensah Published: 2017-05-11 Document Version Publisher's PDF, also known as Version of record Link to publication Citation for published version (APA): Ayerakwa, H. M. (2017). Planting to feed the city? Agricultural production, food security and multi-spatial livelihoods among urban households in Ghana Lund, Sweden LUND UNIVERSITY General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the PO Box 117 legal requirements associated with these rights. 221 00 Lund +46 46-222 00 00 • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal Take down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim. Download date: 26. Sep. 2018 PLANTING TO FEED THE CITY? 11 2 2 PLANTING TO FEED THE CITY? Agricultural Production, Food Security and Multi- Spatial Livelihoods among Urban Households in Ghana Hayford Mensah Ayerakwa DOCTORAL DISSERTATION by due permission of the Faculty of Social Sciences, Lund University, Sweden. To be defended at Varlden (111) Geocentrum I, Solvegatan 10, Lund on 7th June, 2017 at 13:15. Faculty opponent Pay Drechsel International Water Management Institute (IWMI) Sri Lanka 33 Organization Organization Document namDeo cument name LUND UNIVERLUSNITDY UNIVERSITY Doctoral DisseDrtoacttionra l Dissertation Department ofD Heupmaratmn eGnet oogf rHapuhmya n GeograpDhayt e of issue Date of issue 15th May 20171 5th May 2017 Author(s) Author(s) Sponsoring orSgapnoinzsaotiroi ng organization Hayford MensHaha yAfoyredr aMkwenas ah Ayerakwa Title and subtiTtleitl e and subtitle Planting to FePedla tnhtein gC ittoy ?F eAegdri cthuelt uCraitly P? rAogdruicutioltunr, aFl oPordo dSuecctiuornit,y F aonodd M Suelctiu-Sriptya atianld L Mivuelltiih-Soopdasti aalm Loivnegli hoods among Urban HousehUorlbdas nin H Gohuasenhao lds in Ghana Abstract Abstract The 2010 popTuhlea ti2o0n1 a0n pdo hpouulastinogn caennds uhso uisni nGg hcaennas uresv eina lGedh atnha t rmevoerael etdh atnh aht amlf oorfe ththea nG hanlfa oiafn t he Ghanaian population livepdo pinu luartbioann licveendt einrs u. rCbraitnic cael ntote trhse. Cphrieticnaolm toe nthoen pohf eunrboamneiznaotnio no fi su rtbhaen qizuaetsiotino nis otfh eh oqwu etos tion of how to sustainably feesuds tthaein uarbblay nfe peodp tuhlaet iuornb aens ppeocpiaullayt itohne eusrpbeacni aplolyo trh aes u rrabpaind uporboar naizsa rtaiopnid h uarsb athneiz taetniodne nhcays the tendency to urbanize potvoe urtryb.a Tnhizise hpaosv elertdy .t oT hreisn ehawse dle pdo tloic rye dnebwaetde paobloicuyt dtheeb aimtep laicbaotuiot ntsh eo fim faprlmiciantgio inns coitfi efasr tmoi ng in cities to the food securtihtye ofof oudrb saenc ruersitiyd eonf tusr.b an residents. This thesis aimTsh iast tchoenstirsi bauimtinsg a tto c tohnet rdibeubtainteg btoy tdhel idnebaatinteg bthy ed enloi ne-matainrgk etht seo nuorcne-ms aorfk feoto sdo aunrcde asn oafl yfozoed and analyze their implicatiothnes irf oimr uprlbicanti ohnosu sfoerh uorldbsa nin h toeurmsesh ofld fso oind tseermcusr itoyf. fIono d osinegc utrhitiys., tInhe d othinegs isth aisn,a tlhyez etsh etshies analyzes the interplay that einxtisetr pbleatyw teheant ethxeis vt abreiotwuese ang trhiceu vltauriaolu esn aggargiceumlteunratsl ebnyg uargbeamn ehnotuss beyh oulrdbsa inn hboouths euhrboaldns ai n db oth urban and rural areas asr uwreall arse faoso da st rwanesllf ears rfeocoedi pttrsa ntos fuerb raenc ehipotuss teoh oulrdbsa na nhdo uhsoewh othldesy acnodn thriobwut eth teoy h coounsterihboultde to household food security ifno osmd asell caunrdit ym ine dsimumal ls aiznedd m cietideisu min sGizheadn ac.it i es in Ghana. The thesis emTphloey sth ae smisi xeemdp mloeytsh oad ms iaxpepdr omaecth-oqdusa anptitpartoivaec ha-nqdu aqnutaitlaittaivtiev ea nmde qthuoadlista- ttiove i nmvesthtiogdaste- toh ein vestigate the concept of urbcaonn cfoeopdt osfe ucrubriatyn. fTohoed asencaulyrtiitcya. lT fhraem aenwaolyrtkic eaml fpralomyewd oisrk g eromupnldoeyed do nis tghreo uancdcesds o pni ltlahre toa ccess pillar to household foohdo suesceuhroitlyd. food security. The findings oTf hthee f itnhdeisnigss a oref tphree stheensteisd a irne t hpreese anrtteicdle isn pthrerece daertdic bleys a p ‘rkeacpepdae’d. Ib ayr gau ‘eka tphpaat ’t.h Ie a drgeubea tteh at the debate on the contribuotnio tnh eo fc uornbtraibnu atigornic uoflt ureb aton uargbraicnu lhtuorues teoh oulrdb afono hdo suesceuhroitlyd sfoeoedm se tcou orivtye rs ceoenmcse ntotr aotvee ro nco ncentrate on urban agricultureb aanlo anger iwcuitlhtuoruet alcocnoeu wntiitnhgo ufot ra tchceo uonthtienrg f ofoord t hpero ovtihsieorn fionogd o prpoovritsuionintiensg aovpapiolarbtulen itoie sth aev ailable to the household inchluoduinsge hfolodd i npcroludduincgtio fno oind perorid-urcbtiaon ai n dp erurir-aulr bsapna caensd. Eruxrpaal nsdpiancge tsh. eE sxcpoapned ianlgs oth hee slpcso ptoe also helps to account for othaecrc nooun-t mfoar roktehte for ondo ns-omuarcrkeest sfouochd asos uforcoeds t rsauncshf ears r feocoedip ttrsa nthsafet ra rec feoiupntsd tthoa pt larye i mfopuonrdt aton tp lay important roles in the foordo leses cinu rtihtye o fof uordb saenc huoriutys eohf oulrdbsa.n T hoeu tshehsoisld ess.t Tabhleis thess itsh aets,t haobulisheehso ltdhsa tw, ihtho umsuelhtiopldes f owoitdh multiple food provisioning oprpoovritsuionintiensg, oepspoerctuianlliyti ethso, sees pwehcoia lelyn gthaogsee i nw hbo the nugrabgaen iann bdo rtuhr aulr baagnri caunltdu reu rhaal vaeg rbiceuttlteurr e have better food security ofouotldo oske cthuarinty t hooustleo owkh toh adno tnhoots. e who do not. The implicatioTnsh efr oimmp ltihcea tifoinndsi nfgrosm is t hthea fti npdoilnicgies si sa timhaetd p aotli caideds raeismsiendg autr baadnd rfeososdin sge ucurbriatyn tfhoroodu gshe couwrinty through own food provisionfionogd s phroouvldis inoonti nbge sthreoautledd n iont ibsoe latrteioante. dR aint hiseor,l astuiocnh. pRoalitchiers, shuochu ldp oalciccieosu nsth four ldth aec accotuivnet for the active rural-urban intreurraacl-tuiornbsa nc hinatrearcatcetriiosntisc cohf amr acntye rcisotuicn otrfie ms ain ys ucbo-uSnatrhieasra in Asufrbic-aS aahnadr ahno wA frthiceay acnodu lhdo bwe they could be harnessed to choamrnpelsesmeedn tto e caocmh poltehmere fnotr e baecthte or tfhoeord f osre cbuertitteyr afonodd l ivselcihuoriotyd aonudtc loivmeleihso. od outcomes. Key words Key words Urban agricultUureb,a rnu raagl raicgurlitcuurletu, reu,r aolw ang rfiocoudlt uprreo,d ouwctnio fno,o fdo opdro sdeuccutiroitny, food streacnusrfietyr, rfeocoedip ttrsa,n Gsfhear nreac eipts, Ghana Classification Csylasstesmifi caantido/no rs iynsdtexm t earnmds/o (ri fi nadneyx) terms (if any) SupplementarSy ubpibpllieomgreanpthaircya bl inbflioorgmraptihoinc al information Language Language English English ISSN and key ItSitSleN and key title ISBN 978-91-7IS7B53N- 299758--89 1(P-7r7in5t)3 -295-8 (Print) ISBN 978-91-7IS7B53N- 299768--59 1(E-7le7c5t3ro-2n9ic6)- 5 (Electronic) Recipient’s noRteesc ipient’s notes Number of pagNeusm 1b7e9r of pagPerisc e Price Security classiSfieccautiroitny classification I, the undersigIn, ethde, buenidnegr sthigen ceodp, ybreiginhgt othwen ceor poyf rtihgeh ta obwstnrearc to of ft hthee a abbsotrvaec-tm oef nthtieo naebdo vdeis-mseerntatitoionne,d h deirsesbeyr tgartaionnt, hereby grant to all referenceto s aolul rceefesr penercme isosuiornc etos puebrmlisishs aionnd tdoi spsuebmlisinha aten dth deis asbesmtriancatt eo ft hthee a abbsotrvaec-tm oef nthtieo naebdo vdeis-mseerntatitoionne.d dissertation. Signature Signature Da te: 28th AprDila, t2e0: 1278 t h April, 2017 4 4 4 4 PLANTING TO FEED THE CITY? Agricultural Production, Food Security and Multi- Spatial Livelihoods among Urban Households in Ghana Hayford Mensah Ayerakwa 55 Cover photo by Copyright (Hayford Mensah Ayerakwa) Faculty of Social Sciences Department of Human Geography ISBN 978-91-7753-295-8 (Print) ISBN 978-91-7753-296-5 (Electronic) Printed in Sweden by Media-Tryck, Lund University Lund 2017 6 6 To Steve and Clara Opare-Obisaw 77 With God all things are possible Luke 1:37 8 8 Content Acknowledgement ............................................................................... 11 List of Articles ..................................................................................... 14 List of Figures ..................................................................................... 15 List of Tables ....................................................................................... 16 List of Acronyms ................................................................................. 17 1. INTRODUCTION ............................................................................... 18 1.1 Introduction ........................................................................... 18 1.2 Research background and questions ...................................... 20 1.3 Structure of the thesis ............................................................ 22 2. AGRICULTURAL PRODUCTION AMONG URBAN RESIDENTS .................................................................................................. 25 2.1 Introduction ............................................................................... 25 2.2 Agricultural production in urban and peri-urban areas .............. 26 2.3 Urban residents food production in rural areas and multi-spatial livelihoods ........................................................................................... 28 2.4. Gender dynamics of urban households’ agricultural production............................................................................................ 30 2.5 Motivation for engagement in urban and peri-urban agriculture............................................................................................ 32 2.6 Motivation for engagement in rural agriculture ..................... 33 2.7 Conclusions ........................................................................... 35 3. RESEARCH SETTING: GHANA ...................................................... 37 3.1 Introduction ........................................................................... 37 3.1 Overview of the Ghanaian economy ..................................... 38 3.2 Urban and peri-urban agriculture in Ghana ........................... 41 3.2.1 Regulatory framework guiding urban agriculture in Ghana 42 3.3 Urbanization trends in Ghana ................................................ 44 3.4 Institutional arrangements governing urban lands in Ghana ..... 46 3.5 Institutional arrangements governing urban land use in Techiman and Tamale .......................................................................................... 47 4. CONCEPTUALIZING URBAN FOOD SECURITY ........................ 49 99 4.1 Introduction ........................................................................... 49 4.2 Household Food Security .......................................................... 49 4.2.1 The three pillars of food security ................................... 50 4.2.2 Household food security and productive assets ............. 52 4.3 Normative approaches to food security ..................................... 53 4.4 Explanations for lacking food security ...................................... 55 4.4.1 The food availability decline (FAD) theory .................. 55 4.4.2 The entitlement theory ................................................... 56 4.5 Household food security, assets and urban livelihoods ......... 58 4.5.1 Elements of Sustainable Livelihood ............................. 59 4.5.2 The Urban Context and the Sustainable Livelihood Framework .................................................................................. 60 4.6 Conclusions – linking concepts to the research questions ..... 63 5. METHODOLOGY .............................................................................. 65 5.1 Introduction ........................................................................... 65 5.2 Study site selection and description ....................................... 66 5.2.1 Techiman Municipality and Tamale Metropolitan Assemblies .................................................................................. 66 5.2.2 Characteristics of Techiman and Tamale ...................... 70 5.3 Sampling Technique .............................................................. 73 5.4 The survey instrument and survey administration ................. 76 5.4.1 Quantitative data collection ........................................... 76 5.4.2 Interviews and Consents ................................................ 77 5.4.3 Qualitative data collection ............................................. 78 5.5 Method of Analysis ....................................................................... 81 5.6 Ethical considerations ................................................................ 82 5.7 Data quality and limitations of the study ............................... 82 6. SYNTHESIS OF THE FINDINGS .................................................... 85 6.1 Contextualizing Research Findings ....................................... 85 6.2 Contextual Characteristics of the Sample .............................. 87 6.3 Summary of Articles .............................................................. 89 7. CONCLUSIONS ................................................................................. 93 References ..................................................................................................... 97 10 10 Acknowledgement ‘Whether you think you can or cannot, you’re right’. This quote from Henry Ford, has be a source of inspiration to me for many years. Indeed the PhD journey has come quicker than I anticipated it to be. I recall quite well the Skype interview I had in June 2013 for the only social science slot on the Swedish-Africa Urban Agriculture Project in Ghana-It feels like that happened only a few weeks ago but it’s almost four years since that conversation took place, and about three and half years when the actual PhD journey began. The PhD journey has been an interesting yet humbling experience. The journey began with lots of enthusiasm and energy. There were really wonderful moments but also very low points along the way. There were deviations and difficult moments on the journey. Through the high and low moments, I learnt to be determined and ambitious yet modest and realistic. The journey has taught me the relevance of keeping the right people who can stir you on to the next level when it matters most. Several people have made this journey possible. I would therefore like to express my gratitude to them for the diverse roles they played in making this journey a success. My sincere appreciation goes to Agnes Andersson Djurfeldt who was the principal supervisor for this thesis. Agnes has been more than a supervisor keeping me on track for all these years and taking personal interest in my life and career as a researcher. Working mostly from Ghana was never a problem because Agnes always made time to catch up with my progress every fortnight. I’m grateful for going out of your way to secure housing and all other entry requirements anytime I had to travel to Sweden from Ghana. I recall your visit to Ghana to participate in the qualitative fieldwork in August 2015 and how supportive you were during that phase of the project. Working with you on this project has been more fulfilling than I thought it was going to be. Thank you for nurturing me to this stage! Special thanks to Magnus Jirstrom who was my co-supervisor on the project and team leader on the Urban Agriculture Project. Your words of encouragement, taking time to read my manuscripts and all those wonderful comments after my seminars were simply wonderful and well appreciated. Thank you for the many wonderful tips you provided me that has become a part of me 1111 I’m grateful to Fred Dzanku, a co-supervisor, for providing great support during my estimations as well as reading my manuscripts. I recall time and again how I walked into your office to seek clarification on some estimation related issue-thanks for your patience and investments in me! While I close a chapter on this, I certainly look forward to greater collaborations and estimating new equations. I am also thankful to Daniel Bruce Sarpong for keeping your doors opened whenever I needed your help. Your comments on the manuscripts as well as the platforms you offered me to share my work with your PhD class and senior members at the department of Agricultural Economics and Agribusiness at the University of Ghana inspired new ideas. The financial support from the Swedish Research Council for Environment, Agricultural Sciences and Spatial Planning under the Swedish African Urban Agriculture Project is duly appreciated. I’m also thankful to AFSUN for permission to use the AFSUN Household Food Security Baseline Survey instrument for the research. Special thanks to the respondents in Techiman and Tamale who made it possible to have the needed data used for this thesis. The insights received from the discussants at various stages of my research have contributed to shaping the work. -Ellen Hillbom at the research proposal seminar; Tomas Germundsson at the mid-term seminar, and Karl Johan Lagerkvist at the final seminar. I’d like to also thank my colleagues at the Human Geography Department for their support! Thank you Karl-Johan Lundquist, head of department, for being supportive and providing great and inspirational leadership. Arvin Khoshnood provided excellent administrative support. Linda Stihl ensured that I had an office and a desk space at every visit. To all the administrative staff, thank for being so excellent at your work. Sarah Alobo L. and Chai Sui have been simply phenomenal-the former was for a long time my office colleague-I recall the number of times we have had to share ideas in the office and also at lunch-In you I found a sister and family. Chai was simply amazing in the organization of the DiP seminars but also read my manuscript and provided comments at her own free will-Thank you! Ibrahim Wahab has been more than a brother. Samuel Omondi and Heather Mackay are colleagues working on the same project in Kenya and Uganda respectively-thank you for sharing ideas! The following individuals and groups made my stay in Lund really refreshing –Henry Ampah-Korsa, Benedict and Charity Asamoah, The Ghana Community in Lund, Torbern and Pamela Klason and the International 12 12 Fellowship. Thank you all for the demonstration of Christian love during my stay in Lund! I gained a lot of moral strength from my parents Steve and Clara Opare-Obisaw throughout my study. Indeed I have come this far because of their belief and support-Dad, I remember our discussions and they’ll continue to guide my steps. Special thanks to Rev. Dr. Ivy Drafor Amenyah, Rev. Dr. Gyang-Duah, Rev. Daniel Amoako-Nyarko, Rev. Bawua Bonsafo, Rev. Otuo Acheampong and Rev. Phillip Tutu for standing with me in the gap on this journey. Aunty Hannah Okyere Dankwa, your support and regular checks on my family whenever I was away is duly appreciated. Samuel Adu acted as the Clerk of PCG, Covenant Congregation, Dzorwulu and discharged the duties with excellence for all the times that I was away-God bless you brother! I owe Grace Lincoln Quarshie a lot of appreciation for being supportive and of great help- thank you! My friend Stuart McDonalds from Cocoa Life was a great blessing- thank you for the many boxes of chocolates you sent me every now and then- It’s time to team up to support our cocoa farmers to improve their yields-so we can produce more chocolates! Thank you Dr. Gepi-Attee for your support and love to my family when it was needed most. To my brother Yaw Sakyi-Bekoe and his family, this is it-thank you! Finally, I cannot find enough space to express my appreciation to my family for the support and encouragement I received from you. You had to endure my absence from home and long hours of work at night. I’m indebted to you Maame Ekua Ayerakwa and I hope I’ll find the space to give back all the time you spent taking care of our son. Nenyeh Yaw Opare Ayerakwa, now you have all the time and hopefully, we can resume our football and cycling lessons. This thesis is about understanding the food security outlook of the increasing urban population in Ghana from the perspective of own food production and food transfer receipts. I wish to thank and acknowledge the hard work of the urban farmers who produce and supply vegetables to the increasing urban population. To those who spend ours commuting to rural areas to produce food for their families, this work is about you and thus acknowledge your efforts. We can collectively work to make life and this world a better place for all. All these would not have been possible without God-To Him alone be all the glory! 1133 List of Articles Article 1. Ayerakwa, H.M. (2017). Urban Households’ Engagement in Agriculture: Implications for Food Security in Ghana’s Medium Sized Cities. Journal of Geographical Research Article 2. Ayerakwa, H.M., Dzanku, F.M. and Sarpong, D.B. (submitted to a peer reviewed journal) Effects of food production on household food security in a small and medium-sized city in Ghana Article 3. Ayerakwa, H.M. (Under peer review) Food Transfer Receipts as Multi-Spatial Livelihoods among Urban Residents in Ghana 14 14 List of Figures Figure 3.1 Map of Ghana showing regional population distribution Figure 3.2 Ghana’s GDP growth rate (2008-2015) Figure 3.3 Sectorial contribution to the Ghanaian economy (share of GDP) Figure 5.1 Map of Techiman township highlighting major roads Figure 5.2 Map of Tamale township highlighting major roads and neighbourhoods Figure 5.3 Map showing the distribution of sampled households in urban Techiman Figure 5.4 Map showing the distribution of sampled households in urban Tamale 1155 List of Tables Table 3.1 Ministries with policy regulatory mandate in urban and peri- urban agriculture Table 3.2 Trends in population growth and urbanization in Ghana Table 3.3 Proportion of urban population by region (1960-2010) Table 5.1 Agricultural activities in the Techiman and Tamale Assemblies Table 5.2 Demographic characteristics of urban households in Techiman and Tamale Table 5.3 List of participants during qualitative interviews 16 16 List of Acronyms CAADP Comprehensive Africa Agriculture Development Programme FAO Food and Agriculture Organization GDP Gross Domestic Product GPS Global Positioning System GSS Ghana Statistical Services HFAP Household Food Insecurity Access Prevalence ISSER Institute of Statistical, Social and Economic Research METASIP Medium Term Agriculture Sector Investment Plan MoFA Ministry of Food and Agriculture MoTI Ministry of Trade and Industry NGO Non-Governmental Organization SAP Structural Adjustment Programme UA Urban Agriculture UPA Urban and Peri-Urban Agriculture UN United Nations UNDP United Nations Development Programme WFP World Food Programme 1177 1. INTRODUCTION 1.1 Introduction It was mid-morning on August 5, 2015 when I interviewed the Chief of Sisala Line, Mr. Issaka Ibrahim in Techiman on his understanding of the urbanization trends towards Techiman and the nature of non-market sources of food the city’s residents were engaged in. He has been living in what has become his new home for nearly 40 years since he moved to Techiman as a youth in his early twenties in search of better living conditions. At the time of his arrival, his present home was then a farm village but now sits at the center of the town. He narrates how things have changed quickly such that the once rural town that served as farmland to his family now buzzes with various economic activities and population growth even though his profession as farmer remains unchanged. The cost of remaining a farmer continues to increase. He narrates how he and other people in his community need to travel nearly 20 kilometers to rural areas in order to secure agricultural land for farming. The cost of shuttling between the farm and Techiman on a daily basis is beyond the means of his household. As a result, he has decided to spend four days (Monday to Thursday) on the farm and return on Friday to spend the weekend with his family. The account of Mr. Ibrahim was not different from many other people I interviewed on the subject of urban agriculture, urbanization and urban food security. As population increases, demand for food also increases. Interestingly, much work seems to have been done on how producing food in the urban space can contribute to address the concerns of feeding the urban population. This notwithstanding, existing studies on urban agriculture have been biased in favor of metropolitan areas and do not situate urban agriculture in a wider context of other self-produced sources of food available to urban residents. What we do not know is whether the contextual factors (including cultural, environmental and institutional) are the same in small and medium- sized cities. 18 18 The rate of urbanization and whether food production can keep pace with population growth have been an issue of concern and analysis dating back to the past couple of centuries since Malthus’ days (Salih, 1995). The 2010 population and housing census in Ghana suggests that more than half of the Ghanaian population (50.9%) reside in urban centers (GSS, 2013). Whereas this development presents lots of potential benefits, most of the towns that are increasingly being classified as urban are yet to experience the economic transformation that will impact on the livelihoods of the citizenry. As a result, most livelihoods in these newly classified cities continue to hinge around agriculture in various ways (GSS, 2014, 2013). Agriculture also continues to play a significant role in centers with a longer urban history. Agricultural production in Ghana is subsistence and rainfed in nature. It is also a function of land and other productive inputs available to the particular farmer or household in question. It is possible to find people resident in urban areas who derive their livelihood from cultivating plots in the cities or outside the city either for own consumption or commercialization purposes. While people in rural areas can in most instances produce their own food, those in urban areas are constrained to do so and mostly depend on food purchases while others engage in urban agriculture or rely on food transfer receipts or both. One way of addressing the concerns of urban food security is thus through own production to supplement urban households purchases. This however is encouraged when urban residents have access to agricultural productive lands and are supported by appropriate legislations that will promote investments in this type of agriculture. Such policies also need to account for geographical diversity and context specific strengths and weaknesses (Riley and Legwegoh, 2014). The question of how to feed the growing urban population through paths of production that connect rural and urban areas and whether these sources can contribute to urban households’ food security is the entry point of this thesis. What can we learn about the different food provisioning arrangements among urban households and their implications for the food security of these households? How does the interplay of urban, peri-urban and rural food production systems as well as food transfer receipts contribute to provide better food security outcomes to urban agricultural households in the changing dynamics of the city? These questions are explored in detail in the pages of this thesis. It is important however to note that, the focus of the thesis is on small and medium sized cities in Ghana. 1199 1.2 Research background and questions Globally, there is a renewed interest in urban food security research in sub- Saharan Africa partly due to high rates of urbanization and the urbanization of poverty (Riley & Legwegoh, 2014) which places a strain on urban food systems. Recent hikes in global food prices (2007-2008) together with consumer protests in some African cities reignited the debate about where food should be produced for the urban population and how it can be produced in a sustainable manner (Lerner & Eakin, 2011). This phenomenon has resulted in policy recommendations in favor of urban and peri-urban agriculture. The practice of urban and peri-urban agriculture in African cities has been an age old activity. With farming as the mainstay of most African economies, present day cities were once farmlands that supplied food to households. However with population growth comes alternative demands on urban agricultural land for development into infrastructure including housing, schools, roads, and for government’s own uses. The research and policy debate on whether to permit or prohibit urban agriculture has received varied reactions from both sides of the divide. The discussions have been divided along the school of thought that believes urban agriculture should be encouraged and supported for its contributions to food security and livelihoods, and as a poverty alleviation mechanism in urban centers (Mwangi & Foeken, 1996; Obosu-Mensah, 1999; Armar-Klemesu, 2000; Mougeot, 2000; Nugent, 2000; Maxwell, 2001; Danso, et al., 2002; Drechsel, et al., 2006, FAO, 2001). The other side of this divide by contrast believes urban agriculture is not an efficient pro poor initiative that should be encouraged in cities. This is because urban agriculture is not necessarily practiced by the poor who generally lack access to the most important resource-land to engage in own food production (Zezza & Tasciotti, 2010; Crush, et al., 2011; Lee-Smith, 2013; Stewart, et al., 2013; Frayne, et al., 2014). Food production by urban residents however goes beyond the urban boundaries to include production in peri-urban and rural areas. Another dimension to the discussion of urban food security relates to food transfer receipts from other peri-urban and rural households that are found to constitute an important part of the urban households’ food basket. These dynamics are overlooked in many studies of urban food security. In this thesis therefore, the food security of residents of small and medium sized towns in Ghana is contextualized in relation to their own food production 20 20 in a variety of localities and transfers of food to the households. How these various arrangements contribute to urban household food security is traced in this thesis. The overarching aim of the thesis therefore is to delineate the non- market sources of food and analyze the implications of these sources for urban households in terms of their food security. In particular, the thesis is guided by the following research questions: RQ#1: What are the sources and types of own food production arrangements available to urban households and how do they contribute to urban household food basket in the context of small and medium sized cities in Ghana? RQ#2: How does own food production affect household food security in the context of small and medium sized cities in Ghana? RQ#3: How does the interplay of food production in rural areas and food transfers contribute to urban households’ food basket in the context of small and medium sized cities in Ghana? RQ#4: What are the determinants of household food (in) security and food transfer receipts among urban households in the context of small and medium sized cities in Ghana? By answering these questions, the thesis aims to contribute to the understanding of urban food security by looking at the entirety of urban residents own food production opportunities either on urban or rural land and its effect on household food security. In addition, the thesis contributes by exploring other sources of food provisioning arrangements such as rural-urban and intra-urban food transfers available to households and their resultant implications for the urban household food basket. The focus of the thesis on small and intermediate sized cities is novel in that almost all works in the field of urban agriculture and urban food security to date has focused on large urban areas. As a result, the thesis contributes significantly to addressing the knowledge gap in the field of urban agriculture and urban food security in Ghana from the context of small and intermediate urban areas which have been generally understudied in Ghana. 2211 1.3 Structure of the thesis The thesis is a compilation of three journal articles which forms the core of the document preceded by the kappa (translated from Swedish as the ‘coat’). The contents of the kappa have been grouped into chapters and discusses the empirical, methodological and theoretical approaches employed in the thesis. In addition, the main arguments from the individual articles are synthesized and discussed in relation to established empirics. The under listed are the chapters in the kappa. Chapter 1 presents the general introduction to the thesis which outlines the questions to be answered in the thesis. In particular, the research questions guiding the whole thesis as well as the potential contributions of the thesis are also discussed in the first chapter. Chapter 2 reviews and discusses relevant literature in the fields of urban and peri-urban agriculture, rural food production by urban residents as well as other multi-spatial livelihood opportunities available to urban households and their implications for urban household food security. The goal of the chapter is to provide the reader with appropriate empirical background that will help to put the discussions in the thesis in context. Chapter 3 takes a deeper look into the nature of urban and peri-urban farming in Ghana as well as other forms of own food provisioning arrangements among urban households. The chapter also introduces the reader to Ghana by highlighting important socio-economic characteristics as well as some institutional arrangements governing urban land use that makes it an important case for the study. Chapter 4 provides a discussion of the conceptualization of urban food security. In particular, the discussions focus on the normative approaches to measuring food security as well as other explanations for lacking food security. The latter deals with the food availability theory and the entitlement theory. The various components in the food security equation - that is availability, access and utilization are also discussed in the chapter. Other theories related to urban livelihoods that can be used to analyze differences in urban food security are also discussed in the chapter. In Chapter 5, the methodological approaches adopted in the thesis are discussed. The chapter begins with a description of the study sites as well as 22 22 the sampling design and sample selection in the study. The types of data and how they were collected and analyzed are discussed in this chapter. Chapter 6 contextualizes the research findings by synthetizing the findings in relation to the various research questions posed in Chapter 1. In doing this, the main arguments from the articles regarding the effects of urban households’ food production on household food security are summarized in relation to the theoretical underpinnings presented in the thesis. Finally, Chapter 7 presents the main conclusions from the thesis highlighting the need to contextualize the findings of the study in the discussions on urban food security.-thus, policies relating to urban food security through food production should not be generalized but be localized to account for specific socio-cultural contexts. The second part of the thesis focuses on the individual articles which directly answers the different research questions in Chapter 1. Each article aims at addressing at least one of the research questions in the main body of the thesis. Additionally, the papers are written to account for the different geospatial dynamics associated with the findings. In this regard, the data is analyzed along city specific contexts, Techiman and Tamale. The focus of the different articles are presented below: Paper 1 descriptively analyzes the different sources of food production opportunities available to urban households and their contribution to the household food basket. The types of food, and where foods are produced as well as the uses of these foods by urban households are discussed in the paper. The income earning opportunities tied to agricultural production are also discussed. In paper 2, the effects of own food production in urban and rural space on food security are estimated. The particular contributions of the different forms of agricultural production to household food security are discussed in the paper. That is the importance of agricultural production – whether carried out in rural or urban areas - to urban household food security is estimated. The general determinants of urban household food security in small and medium sized cities in Ghana are also estimated. Paper 3 considers the interplay that exist among food transfers, urban agriculture and rural agriculture and how they relate to urban households food provisioning and security. In particular, the types of food transfers received and their uses among households as well as the factors determining food transfer receipt among urban households are discussed in the paper. 2233 24 24 2. AGRICULTURAL PRODUCTION AMONG URBAN RESIDENTS 2.1 Introduction The sources of food available to urban residents and their resultant implications on urban household food security vary across specific households. In this thesis, three main sources of urban foods are observed and discussed-that is engagement in own food production arrangements in urban and peri-urban areas as well as in the rural areas; the second is food transfers from family and friends in rural and urban areas; and finally market channels. The focus in the thesis is on the non-market food sources, that is own production and in kind receipts of food. The decision to choose a particular channel as a main source of food to the household or as a complementary source depends on assets and social capital1 available to the household. Details of these are discussed in the review in this section. This section therefore reviews relevant literature associated with urban households’ food production in both urban and rural spaces and how these practices relate to their food security situation. In particular, the review focuses on food production in urban and rural areas by urban residents, urban households’ food production in rural areas and other multi-local livelihoods. The literature on gender dynamics of food production and the various motivations for engaging in own food provisioning-whether for subsistence or commercialization purposes – is also reviewed. Finally, the effects of own food production on the livelihoods of producing households as well as the more general determinants of urban food security are discussed. 1 This is discussed in Chapter 4 2255 2.2 Agricultural production in urban and peri- urban areas Urban agriculture (UA), defined as the growing of crops and/or raising livestock in an urban space for own consumption and/or sale to other city dwellers (Ayerakwa, 2017) has received considerable interest in the literature measured by its growing importance as a research and policy issue (Badami & Ramankutty, 2015). The number of people engaged in the practice globally according to the United Nations Development Programme (UNDP) stood at 800 million in 1996 and represented almost a third (30%) of the global urban population at the time. This figure though widely cited in the literature (see for example Smit, et al., 1996; Koc, 1999; Lee-Smith, 2010; Mougeot, 2010, 2011; Redwood, 2012) has in recent times been criticized as being exaggerated. Based on available data from 15 developing countries, a more plausible estimate of urban households engaged in agriculture ranges from 11% in Indonesia to about 70% in Vietnam and Nicaragua (Zezza & Tasciotti, 2010). Nonetheless, the first status report on Urban and Peri-Urban Horticulture by the Food and Agriculture Organization (FAO) suggests that UA provides locally grown fresh fruits and vegetables to over 22 million people in Africa’s cities. Thus, the report argues urban agriculture plays an important role in food supply and income generation opportunities in Africa. As a result, the practice of urban and peri-urban horticultural production could potentially see expansion and contribute to achieving zero hunger in African cities provided the sector receives the necessary support (FAO, 2012). Recent studies in selected African cities reveal varied degrees of engagement in urban and peri-urban agriculture (UPA) suggesting that the potential of UPA in addressing poverty as well as food insecurity is highly contextual. Frayne et al. (2014) report in their study of 11 Southern African countries that engagement in urban agriculture varies significantly across cities with participation ranging from as low as 6% in Windhoek, Namibia to over 60% in Blantyre, Malawi among the urban sample. Zezza and Tasciotti (2010) in their work on urban agriculture, poverty and food security based on evidence from a sample of developing countries conclude that, although no clear regularity is found in the participation rates of countries, the four African countries2 in the sample demonstrated the highest shares of cash income 2 The African countries are Ghana, Nigeria, Madagascar and Malawi 26 26 derived from urban agriculture in the sample ranging from 26% in Malawi to 71% in Nigeria. In Ghana for example, urban agriculture is practiced in backyards of households, on own plots outside the home, private plots awaiting development, government reserved plots as well as areas clearly demarcated as non-habitable (for example wetlands, areas along rail lines etc.). These locations are largely private lands, public lands or semi-public lands. The private lands include homestead or land away from the farmer’s residence; these lands may be owned by the farmer or leased. The public lands include national parks and conservation areas whilst the semi-public lands are for example the school yards, hospitals and prison farms. At the turn of the millennium, urban and peri-urban agriculture has received considerable recommendations by activists for its potential to contribute to initiatives of poverty alleviation, especially in developing countries (Smit, et al., 1996; UNDP, 1996; Armar-Klemesu, 2000; Cofie & Drechsel, 2007; Lee- Smith, 2010; Mkwambisi, et al., 2011; Mougeot, 2011; FAO, 2012). These important roles of urban and peri-urban agriculture to the livelihoods of practicing urban households and by extension the cities in which they are practiced have well been documented in the literature to include increased access to nutritious food; provision of fresh fruits and vegetables to urban dwellers; reducing food purchases and supplementing farming households income from surplus sales and serving as a source of employment to sellers of farm products from urban and peri-urban plots etc. (Smit & Nasr, 1992; Smit, et al., 1996; Maxwell, et al., 1998; Armar-Klemesu, 2000; Cofie & Drechsel, 2007; Dubbeling, et al., 2010; Lee-Smith, 2010; Satterthwaite, et al., 2010; Mkwambisi, et al., 2011; FAO, 2012; Badami & Ramankutty, 2015). Aside from such positive effects on urban livelihood, UA is also associated with a number of eco-system services as well as social and environmental aspects such as waste recycling, reduction in soil erosion, beatification of cities etc. (Badami & Ramankutty, 2015). Even though there is not outright condemnation of the potential benefits associated with urban agriculture, it continues to receive criticism in the literature. Central to the criticism is the premise that, the potential benefits of urban agriculture to urban households have been over exaggerated by activists. Additionally, information on the scale and scope of the practice in Africa remains piecemeal with inconsistent data and methodologies (Zezza & Tasciotti, 2010; Crush, et al., 2011; Lee-Smith, 2013; Stewart, et al., 2013; Frayne, et al., 2014). The lack of empirical evidence on the scale and scope as 2277 well as the impact of urban agriculture on food security further validates the skepticism (Stewart, 2013). In this thesis however, a new perspective to the debate is introduced. The thesis argues that, assessing urban food security from the perspective of urban and peri-urban food production alone leaves a gap about other food provisioning arrangements available to the household. In effect, analysis of the effect of own production on urban household food security should encompass all agricultural production opportunities available to the household. This subject is explored in detail in papers 1 and 2 where the various sources of food to the urban household and its effects on household food security are examined. 2.3 Urban residents food production in rural areas and multi-spatial livelihoods Although urban-rural linkages (especially regarding food production) are receiving increasing attention in the urban food security literature, very little is known about the scale of urban households’ agricultural production in rural areas in Ghana (Ayerakwa, 2017). Data from a handful of other African countries confirms that rural farming by urban residents is perceived as an important source of food to producing households. The approach adopted to cultivate rural farmlands include owning land in rural areas that are cultivated by other people or the owners themselves used in rearing livestock and crop cultivation (Krüger, 1998; Andersson, 2002; Foeken & Owuor, 2008). Apart from the fact that access to rural land is fairly easy and cheaper (monetarily), the motivation for rural farming is grounded on access to cheap rural labour (Makoka, 2005; Bryceson, 2006). As noted by Foeken and Owuor (2008), food produced in rural areas was found to be the most important source of livelihood among urban farmers in Nakuru, Kenya. The practice of rural farming by urban residents is part of a strategy adopted by urban residents to minimize risks associated with households’ livelihoods in urban areas (Sarpong & Asuming-Brempong, 2004). In most instances, this engagement does not lead to a spilt in the household structure, but may require the main farmer (mainly the man) to be absent from home a few days in a week if the farmland is considered too far to warrant daily commuting (Frayne, 2004a; Foeken & Owuor, 2008; Frayne, 2010). In other cases, women may travel seasonally to engage in rural agriculture (Andersson Djurfeldt, 2012). 28 28 The safety net role of rural agriculture is manifested in the extent to which urban farm families are prepared to travel to cultivate rural land in order to produce food crops. In the city of Tamale in Ghana for example, some urban residents prefer to migrate during the main farming season to rural areas in order to have access to arable land to grow staple crops for household consumption (Chagomoka, et al., 2015; Chagomoka, et al., 2016). Similar findings are observed for urban residents in Nakuru, Kenya where urban households own land in rural areas which are used to raise livestock or grow crops (Foeken & Owuor, 2008)As an emerging field of literature, not much is known about the proportion of urban residents involved in rural food production, however. Whereas a number of urban household resort to rural farming as a source of food for their households, available evidence suggests a number of households rely on food transfers from rural areas as a source of food in meeting the household food requirements in urban areas (Crush and Caesar, 2017). This phenomenon has led to what is referred to in the literature as multi-spatial or multi-local livelihoods where households maintain linkages between urban and rural areas as sources of food or income (Foeken & Owuor, 2001; Start, 2001; De Haan & Zoomers, 2005; Foeken & Owuor, 2008; Tacoli, et al., 2008; Dick & Schmidt-Kallert, 2011; Crush, 2013). The motivation for engagement in multi-spatial/local livelihood is also connected to risks associated with increased food prices due to globalization and volatility in food prices and incomes among urban residents (Andersson Djurfeldt, 2015; De Haan & Zoomers, 2005). Economic pressure may lead people to migrate to cities in search of better conditions of life, mainly through employment opportunities in the informal sector. As a livelihood strategy, the whole family does not migrate at the same time, especially for married migrants. For fear of uncertainty, households prefer at least one of the members (mostly men) to first relocate to the urban area in order to secure appropriate housing. Families of married migrants tend to follow later. Between the times a household member migrates to the urban area and when they finally settle with appropriate employment opportunities, they maintain strong linkages between the newly formed household in urban area and their rural relatives (Krüger, 1998; Potts, 2010; Andersson Djurfeldt, 2015a, 2015b). In the urban areas where migrant farmers do not have access to food production on a subsistence basis but have to depend on almost entirely on the market, the need to maintain strong links with rural households become more important. 2299 These relationships translate into food remittances that cushion the urban household from total dependence on the market economy for food. This occurrence supports the assertion that rural to urban food transfers are important components of the discussion of urban food security (Frayne, 2004b) as urban households increasingly depend on such food sources for their livelihoods. This thesis though not nationally representative, makes the first attempt at providing the scale of this practice from the perspective of the study areas. Another important gap that the thesis seeks to fill relates to multi-spatial livelihoods. Presently, not much data exists on multi-spatial livelihood in Africa. The thesis therefore contributes to enriching understanding of the subject from the perspective of small and medium sized cities in Ghana. In addition, the estimated underlying factors explaining food transfer receipts also provides knowledge relevant for policy. 2.4. Gender dynamics of urban households’ agricultural production The practices of urban agriculture in Africa and the people involved generally vary from country to country. While researchers in some countries like Ghana and to a large extent West Africa conclude that the practice is male-centered, females are found to dominate the practice in the Southern African countries (Andersson, 2002; Foeken & Owuor, 2008; Obuobie, et al., 2014). In Ghana, less than 10% of all urban open-space farmers are women who cultivate primarily indigenous vegetables (Obuobie & Hope, 2014). However, marketing of produce from urban space is done by the wives of the male farmers or other market intermediaries who are mostly women retailing in leafy vegetables (Gbireh, 1999; Obosu-Mensah, 1999; Armar-Klemesu, 2000; Obuobie, et al., 2014). Andersson (2002) report in her study on food security in Rusape, Zimbabwe that urban agriculture is mostly practiced by vulnerable groups of persons including elderly females with many dependents (Andersson, 2002). Similar findings are reported from Nakuru, Kenya where women were found to engage in multiple livelihood activities including urban agriculture in their quest to cope with the declining purchasing power of their household’s income (Foeken & Owuor, 2008). A study of urban agriculture in 11 Southern African countries 30 30 reveal that, urban agriculture is practiced most in countries that have economic fragility like Malawi, Zimbabwe and Lesotho where there is increasing hardship and urban poverty. (Kutiwa, et al., 2010; Tawodzera, 2010; Crush, et al., 2011; Frayne, et al., 2014) The participation of women in urban agriculture does not preclude the contribution of men to farming activities in the cities. Generally, men provide the labor force for commercial agriculture activities in return for wages compared to women whose primary goal is own consumption and as a source of revenue to supplement household incomes (Hovorka & Lee-Smith, 2006). In cities where women dominate urban agriculture, the majority of men have casual employment as artisans in various sectors of the urban economy not related to agriculture. In certain southern African countries, urban agriculture is generally perceived to be associated with women, discouraging male participation in such activity (Mudimu, 1997). Generally, women are predominant among urban farmers in countries in Eastern and Southern Africa, like Uganda, Kenya, Namibia, Zimbabwe, Malawi, Zambia, Mozambique etc., whereas men predominate in Ghana and Nigeria (Mascarenhas, 1999; Anosike & Fasona, 2004; Dima & Ogunmokun, 2004; Nabulo, et al., 2004; Obuobie, et al., 2004; Frayne, et al., 2014). Other studies from Port Harcourt, Nigeria, and Senegal reveal that women predominate as agricultural laborers with men as owners of horticultural enterprises (Ba Diao, 2004; Oruwari & Jev, 2004). The dominance of women urban farmers in many parts of Africa according to Hovorka (2005) is related to reasons that the main responsibility for household sustenance and well-being is still borne by women and that women also tend to have lower educational status than men and therefore more difficult in finding formal wage employment. Food production in rural areas by urban residents however falls in the domain of male household members. In Ghana and most African countries, men generally assume the headship roles of household and are therefore expected to provide food to feed the household members while women concern themselves with preparing the food for family consumption. In addition, the labour intensive nature of farming in rural areas does not encourage women resident in urban areas to consider rural farming. Rather, they tend to serve as market intermediaries who buy food and supply same to urban markets (Foeken & Owuor, 2008; Chagomoka, et al., 2015; Chagomoka, et al., 2016). 3311 In this thesis however, female participation in agriculture and its effect on household food security are explored using female headed households as a proxy. Papers 2 and 3 for instance capture the gender dimensions to urban food security. 2.5 Motivation for engagement in urban and peri-urban agriculture The reasons for engaging in UPA may vary across different regional, geographical and cultural contexts. Generally however, the reasons for participating in UPA food production relates to household consumption and commercialization purposes. In the literature, a principal motivation for engaging in urban agriculture generally in Africa and in particular to Southern African countries has to do with economic hardships. Historically, these hardships have been associated with Structural Adjustment Programmes (SAP) that most African countries subscribed to in the 1980s due to deficits in the fiscal balances of the economy. This led to retrenchment of civil servants in many countries in Africa (Smart, 2015). In most instances, SAPs led to loss of jobs with widespread urban poverty, leading to the search for alternative livelihood strategies which includes urban agriculture. The decision for people to resort to urban agriculture was found in Zimbabwe to correlate with the difficult economic conditions of 1991/1992 partly due to the IMF/World Bank Structural Adjustment Programmes (SAPs) (Mbiba, 1995). The implementation of the SAP was perceived to have contributed to unemployment leading to the decline in income levels of residents (Andersson, 2002). (See also Rakodi, 2002; Tacoli, 2002). These conditions compelled urban residents to find alternative means of detaching their households’ food security from the market economy that became very inflationary, by engaging in UA to cushion the effects of rising food prices. There are many factors which have contributed to the expansion of UA, but of particular importance is the desire to ensure food and economic security (Smart, et al., 2015) through food availability, access, and utilization, and also through income and employment generation. Peoples’ motivation to participate in urban and peri-urban agriculture is part of a coping mechanism (Burger et al., 2009) and a response to inadequate food access. It is observed in Uganda for example that participation in urban agriculture is as a result of insufficient access to food and that 95 percent of urban farmers began farming 32 32 in order to increase their access to food (Bukusuba, et al., 2007). The practice thus strengthens resilience against social, political, economic and other external shocks which tends to affect one’s ability to have sufficient food (Maxwell, 1995; Maxwell, et al., 1998; Warren, et al., 2015). According to Hovorka (2003) women are actively engaged in urban farming and the sale of agricultural produce with the motivation of contributing to household income and curbing the impact of poverty and food insecurity. Smart et al., (2015) argue that in the unfortunate occurrences such as economic meltdown, natural disaster and civil wars, the most at risk are the urban poor. As a result, survival strategies such as urban and peri-urban agriculture become an essential element to ensure food and economic security to such groups. When food is grown it is either consumed by the producer and his or her family or sold, therefore creating a larger pool of people getting access to food in addition to the direct nutritional and employment benefits to producers. Depending on the preferred needs of households, the ranking of factors that influence the decision of households or individuals to participate in urban agriculture may differ from income to food security and employment reasons. For instance in a study conducted in Freetown, Sierra Leone, 20 percent of the participants indicated that they were not engaged in urban agriculture on full time basis as they had other sources of employment (Lynch, et al., 2013). The motivation to participate in agricultural production is analyzed in paper 1 by assessing the different uses of own food provisioning to the household- whether it is motivated by consumption or commercialization purposes. 2.6 Motivation for engagement in rural agriculture Similar to the motivations to cultivate urban and peri-urban plots, the decision to cultivate rural plots by urban households may vary across space and household. This notwithstanding, cultivating rural plots over time has been found to serve as accumulation across space as a way of developing multi-local livelihoods (Andersson Djurfeldt, 2012). These practices not only cushion the household against urban price increases but also provides opportunity for such farmers to reduce their reliance on the market as well as providing them the opportunity to remit food to other urban food insecure households (Crush & Caesar, 2017). 3333 Even though consumption may be the underlining reason for most urban households who engage in rural agriculture, some urban residents do same for profit motives. These individuals with investment capacity prefer larger land areas to be able to cultivate for commercialization purposes. More so, labour is generally cheaper in the rural areas than in the urban centers. Other motivations to engage in rural agriculture stems from the declining real income and the increasing economic insecurity (Devereux, 1999) in urban areas that tend to put a strain on household income and food security. In addition, farming in rural areas provides an incentive for urban households to maintain family and kinship ties and contribute their part to the social contract among rural families. It is important to note that, construction of households in Ghana and most countries in sub Saharan Africa is characterized with spatial and functional features (Yaro, 2006; Andersson Djurfeldt, 2012). Even though most households in urban centers are nuclear in nature, they maintain a wider spatially intertwined relations with other members of the extended family in both urban and rural areas (Guyer, 1981; Budlender, 2003; Adepoju, 2005; Andersson Djurfeldt, 2012). Maintaining these relations have both social and economic implications forming part of the social contract among families. Beyond urban-rural cash remittances, rural farming provides opportunity for this social and economic obligation to be fulfilled. Similarly, Sarpong and Asumeng-Brempong argue that, farming in rural areas is a form of social insurance among some urban workers who wish to return to their kinsmen after retiring from active formal wage employment mostly located in urban areas (Krüger, 1998; Sarpong & Asuming-Brempong, 2004). Maintaining a rural plot or farm thus facilitates reintegration into the rural community in such situations. Beyond these reasons, households who engage in rural farming improves their access and possibly utilization of food. The above notwithstanding, engaging in rural agriculture is not an option opened to everyone, however. Rather, the practice thrives on social and kinship capital that can be drawn by households that have a connection to such established channels. This backdrop makes it imperative for urban households to continue to belong to rural extended households-making the concept of multi-spatial livelihoods more relevant today than ever. 34 34 2.7 Conclusions Overall, the review concludes that urban and peri-urban farms are mostly characterized by the production of fresh produce (mostly vegetables) as part of subsistence by the producing households or motivated by commercial purposes or both. (FAO, 2012; Obuobie and Hope, 2014). The review identifies a number of benefits associated with UPA including its ability to clean up the city through the use of recycled waste (Mougeot, 2005), provision of direct cash incomes and access to food (Ayerakwa, 2017) and contribution to urban employment and reduction of inequalities (Dubbeling et al., 2010; Armah- Klemesu, 2000). Generally because the urban poor form a significant proportion of urban populations in low-income countries and are more likely to be highly vulnerable to food insecurity (Badami & Ramankutty, 2015), participation in urban agriculture serves as a supplement or as a means to diversify livelihoods (Arku, et al., 2012). The general belief that it is the urban poor that are involved in urban agriculture or depend on food transfers from rural areas as a livelihood strategy has been disputed by many authors with evidence to show that many well to do households in society are equally involved (Van Veenhuizen & Danso, 2007; Zezza & Tasciotti, 2010; Mkwambisi, et al., 2011; Padgham, et al., 2015; Ayerakwa, 2017). It is further argued that, the political economy of urban land holding makes it difficult for the poor to have access to urban farm lands even if they have the desire to do so. Urban farmers are therefore not “the poorest of the poor” but represent a large spectrum of income classes with some having obtained significant resources and networks to engage in urban and peri-urban agriculture. A number of knowledge gaps are identified which the thesis seeks to address. They include the over emphasis on urban and peri-urban food production in the discussion of urban food provision and security. This thesis seeks to bring all agricultural production opportunities to bear in the discussion of urban food security moving away from the over concentration on urban and peri-urban agriculture. As observed in the literature review, cash remittances from urban to rural households has received much attention in economic literature. However, much less is known about the reverse where rural households remit food to urban residents. This thesis therefore contributes to addressing the gap by 3355 examining this emerging trends from the perspective of Ghana. The determinants of urban households food transfer receipt are also estimated. Of particular interest is to empirically establish the driving factors for engagement in agriculture across space-whether driven by consumption or income related motives. 36 36 3. RESEARCH SETTING: GHANA 3.1 Introduction This section presents the background to the country of study-Ghana! Ghana presents an interesting case for the study of food production among urban residents for several reasons. In particular, Ghana is one of few middle income countries in sub-Saharan Africa. The nation is also going through the urbanization transition with a number of small and medium sized cities increasingly expanding mainly due to high fertility rates and rural-urban migration. The growth in population brings expansion in access to urban markets but also leads to loss of agricultural lands in towns and cities. The important question of the role of non-market sources of foods in small and medium sized cities to the food security of urban households remain unanswered. The role of the market in providing food for urban residents remains unknown as long as households receive food transfers or continue to provision through agriculture in rural and urban areas. With much empirical works and policies that turn to bias large urban cities, the contextual differences tied to multi-locality among small and intermediate sized cities bring new perspective to policy that can improve the food security situation of urban households in Ghana. The chapter therefore discusses the overview of the Ghanaian economy with emphasis on the agricultural sector. The section takes a deeper look into the nature of urban and peri-urban agriculture in Ghana as well as some urbanization trends in the country. 3377 3.1 Overview of the Ghanaian economy Ghana is a West African country and occupies a total land area of about 239,460 square kilometers (km2). The country is divided into 10 administrative regions. It is bordered to the north by Burkina Faso, Togo to the east, Côte d'Ivoire to the west, and the Gulf of Guinea to the south (Figure 3.1). Fig 3.1 Map of Ghana showing regional population distribution 38 38 On average, annual rainfall patterns vary across regions but ranges between 800 mm in the southeast along the coast in Accra to about 2,200 mm in the extreme southwest with annual temperatures averaging about 30oC. The country’s population is estimated to be about 28 million as at 20153 based on 2010 census data (GSS, 2013). Over the period 2005-2013, Ghana recorded an impressive economic growth with per capita GDP reaching US$1858 in 2013. This growth trend has however slowed down in recent years. For instance, from a high GDP growth rate of 9.1% in 2008, GDP growth peaked at 14% in 2011, mainly due to oil revenues (Figure 3.2). Since then, the growth rate has consistently declined to as low as 4% in 2014 and a further decline to 3.9% in 2015 (ISSER, 2016). Although the agricultural sector remains the smallest contributor to GDP (20.3%), it continues to employ nearly half of the total country’s labour force (44.7%) (GSS, 2014; ISSER, 2016). GDP GROWTH RATE 16 14 14 12 10 9.1 9.3 8 7.9 7.3 6 4.8 4 4 3.9 2 0 2008 2009 2010 2011 2012 2013 2014 2015 YEARS Figure 3.2 Ghana’s GDP growth rate (2008-2015) 3 This is based on calculations of the annual population growth rates 3399 GDP GROWTH RATE (%) As presented in Figure 3.3, the contribution of the agricultural sector to the Ghanaian economy declined from over 30% in 2008 to as low as about 20% of GDP by 2015. At the same time, the share of the services sector to the economy increased from about 49% in 2008 to over 54% in 2015. Industry’s share of GDP has been mixed averaging about 23.3% over the period 2008-2015. As indicated earlier, however, in terms of employment agriculture continues to serve as the backbone of the Ghanaian economy, while the sector also provides access to food for the majority of urban and rural dwellers. 60 50 40 30 20 10 0 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Year Agric Industry Service Figure 3.3 Sectorial contribution to the Ghanaian economy (share of GDP) Agricultural production however is generally done outside of cities but with pockets of farms scattered in open areas in cities. Foods for urban consumers are generally sourced from rural areas with the help of market intermediaries. Urban and peri-urban agriculture in Ghana has become an important component of Ghana’s agricultural development programme in the recent past and contributes significantly to the food needs of urban households, although actual productivity levels remain difficult to measure as a result of the uncoordinated nature of the practice in urban areas. Policies in favour of food production over the years has been biased in favour of rural production. With 40 40 Percent the emphasis on rural agricultural production and the uncoordinated nature of urban and peri-urban agriculture, there is the tendency to underestimate the contribution of urban agriculture to national gross domestic production. Ghana, like many other African countries is a signatory to the Comprehensive Africa Agriculture Development Programme (CAADP) which is a policy framework of the African Union aimed at transforming agriculture in the Africa region. Under the CAADP agreement, member countries commit to invest 10% of discretionary budgeted expenditure to agriculture sector development. This is expected to translate to an annual agriculture sector growth of 6%. Ghana’s performance in relation to these targets averaged 10.4% from 2008-20114. The average sector growth rate between 2008 and 2015 is 4.6%, falling short of the CAADP targets. This notwithstanding, Ghana is the first country in sub-Saharan Africa to have achieved Millennium Development Goal 1-halving extreme poverty by 2015. Available data suggests that the number of poor population has also reduced from about 7 million in the early 1990s to about 1 million in 2015. Nevertheless, over a quarter of the population are still below the poverty line of US$ 1.25/day (MoFA, 2016). 3.2 Urban and peri-urban agriculture in Ghana Urban farming in Ghana traces its root to colonial rule in the 16th century (Anyane, 1963). The practice however gained prominence in the post- independence era as a means to address the food crises of the 1970s. The launch of ‘Operation Feed Yourself’ contributed to the production of food everywhere in the 1970s and a general acceptance of the practice of urban farming (Anyane, 1963; Asomani-Boateng, 2002; Danso, et al., 2002; Drechsel, et al., 2014). Urban farming in Ghana focuses on crops with some livestock rearing. Drechsel et al., (2014) identify three common systems associated with urban farming- intensive market production on larger open spaces; rainfed farming on plots designated for construction which is usually present in low density settings but also at the urban periphery; and home gardens (or back yards) 4 Data from 2012 to date are not available although some evidence suggests a consistent decline in the agriculture sector growth rates. 4411 cultivated primarily for home consumption. The greatest challenge to raising livestock in cities concerns the indiscriminate roaming of livestock along the streets and in the city centers which disturbs city dwellers and disrupts the urban landscape. Although the practice of urban agriculture is accepted nationwide, the enterprise has thrived in the big cities like Accra, Kumasi and in recent times Tamale and Takoradi. Farmers in these cities cultivate mainly exotic leafy vegetables such as lettuce, cabbage, spring onions, carrot as well as some indigenous crops (common in Tamale). Generally, most urban farmers have had previous farming experiences in rural areas before migrating to urban centers, mainly to pursue economic improvement. The farmers are generally illiterate with others having some level of primary education. The illiteracy rates are higher in the northern region (62.82%) relative to the Greater Accra and Ashanti regions (10.7% and 17.4% respectively) (GSS, 2014; Obuobie, et al., 2014). Since most urban residents have previous knowledge of agriculture, rural farming appeals to urban residents in the event of loss of agricultural production on urban land or difficulties with access to food. Food production in rural areas by urban residents may not be a common phenomenon in big cities such as Accra and Kumasi. However, in medium and intermediate sized cities where distance and access to rural land is possible, households more easily draw on their kinship ties to cultivate food. These foods are usually grains and tubers that are stored over long periods for consumption in urban areas. These arrangements form part of the household’s mechanisms to minimize or neutralize any form of risk that may destabilize their consumption patterns. 3.2.1 Regulatory framework guiding urban agriculture in Ghana Agriculture in Ghana is largely rainfed with pockets of irrigation across the country. This makes farmers vulnerable and prone to food insecurity in the event of changes in weather and climatic conditions. The overall policy framework guiding agricultural production in Ghana is anchored on the Ghana Shared Growth and Development Agenda (GSGDA) which covers the period 2014-2017 and has agricultural modernization in a sustainable environment as its core objective (Commission, 2014; ISSER, 2016). 42 42 Sustainable agricultural production will require appropriate policy formulation and implementation. Even though urban agricultural production is widely acknowledged in different policy documents across ministries and agencies, the policies seem fragmented and uncoordinated (MoFA, 2016). Presently, there are several institutions that have policy mandate to work together in harnessing the potential of urban and peri-urban agriculture in Ghana (Obuobie & Hope, 2014). These are presented in Table 3.1 Ministry Type of policy mandate Ministry of Food and Development and growth of agriculture in the country including Agriculture setting of standards and extension support (including urban and peri-urban agriculture). Ministry of Trade Formulation, implementation and monitoring of internal and and Industries external trade. Bodies under the MoTI set standards for local and imported fresh and processed foods and drugs. They also ensure chemical safety by monitoring quality and usage. Ministry of Health The ministry works through the Ghana Standards Authority to establish standards on food products that conform to international standards. Sections of the standards on food safety touch on hygiene, microbiological, packaging and labelling requirements. On urban UPA, they ensure that hazardous chemicals are not used for agricultural production. Ministry of Local The ministry has the responsibility to promote the establishment Government and and development of a vibrant and well-resourced decentralized Rural Development system of local government including the facilitation and the promotion of a clean and healthy environment and of horticultural development. Ministry of The ministry through the Department for Town and Country Environment, Planning has an advisory role to the assemblies in planning and Science, Technology zoning for different land uses including spaces for urban and peri- and Innovation urban agriculture. Ministry of Tourism, This ministry has the mandate to supervise and regulate practices Culture and Creative and standards of catering enterprises and ensuring that food sold Arts are safe for consumption. Ministry of Water The main responsibility of the ministry is ensuring access to safe Resources, Works water and sanitation including water for irrigation. They therefore and Housing work with MoFA to ensure safety in the use of waste water for irrigation purposes. Table 3.1 Ministries with policy regulatory mandate in urban and peri-urban agriculture 4433 For example, the Ministry of Food and Agriculture (MoFA) which has the primary responsibility for food production in Ghana acknowledges in their Medium Term Agriculture Sector Investment Plan (METASIP) the important contribution of urban and peri-urban agriculture in Ghana in terms of employments, livelihoods and poverty alleviation (MoFA, 2010). The METASIP acknowledges practical challenges including access to land and quality of water for irrigation as some of the constraints affecting the practice of urban agriculture. In particular, there are three by-laws that specifically make provisions for urban and peri-urban agriculture in Ghana-two at the national level and one at the district level. The national level laws are found in the Medium Term Agriculture Sector Investment Plan (METASIP) developed in 2010 and the National Irrigation Policy of 2011. The only district level by-law governing the practice of urban agriculture is the Accra Metropolitan Assembly’s by-law on urban agriculture and food safety. The by-law recommends that all urban agriculture activities must take place on backyards of the individuals involved. It also requires all persons interested in participating in urban agriculture in Accra to obtain approval from the metropolitan Public Health unit before doing so. These by-laws have largely remained on the books of the respective assemblies and ministries. The uncoordinated nature of urban and peri-urban agriculture is not about the lack of appropriate legislations to regulate the sector. Rather, it is the lack of enforcement of the regulations which is a function of resources. In other words, the lack of budgetary allocations for the effective enforcement of regulations is the reason for policy failure in UPA. 3.3 Urbanization trends in Ghana Ghana’s population has witnessed significant increases in the last 50 years or so. With less than 8 million population in 1960, Ghana’s population growth has increased to about 28 million as at 2015. Closely tied to the increasing population growth are corresponding urbanization trends with the urban population crossing the 50% mark (GSS, 2013). The share of urban population over the same period has increased from 23% to over 53% (Table 3.2). Annual urban population growth rates continue to vary but averages about 4%. Overall, about two-thirds (65%) of the population resides in localities of their birth with 44 44 another 15% living in localities other than their birth but from the same region. Nearly a fifth (18.7%) of the population live in regions outside of their birth- an indication that urbanization is driven both by natural growth and migration. Migration from one locality to another in the same region or outside the region is a common phenomenon in both urban and rural Ghana. Migration to urban areas has generally been influenced by employment motives (formal and informal). Year Population '000 % Urban Annual urban growth rate 1960 6,727 23.1 - 1970 8,559 28.9 4.7 1984 12,296 32 3.3 2000 18,912 43.8 4.6 2010 24,659 50.9 4.2 2015* 27,899 53.0 3.5 Source: Ghana Statistical Services (2013). * Based on own calculations Table 3.2 Trends in population growth and urbanization in Ghana Generally, classification of communities in Ghana into urban or rural is based on the population sizes. After every census, every community that has about 5,000 inhabitants or more are classified as urban. As shown in Table 3.3, the greater Accra region remains the most urbanized in the country with the Upper East region being the least urbanized. The Central, Ashanti and Brong Ahafo regions are found to be the most rapidly urbanizing in the last 15 years. Over the period 1960-2010, all regions experienced significant growth in urban population in the country. The population growth trends in Techiman and Tamale like many other cities in Ghana, is attributed to the post-independence relaxation of the restrictive rural- urban migration laws which resulted in increased inter regional movements (Drechsel et al., 2014). 4455 Region 1960 1970 1984 2000 2010 Western 24.7 26.9 22.6 36.3 42.4 Central 28.0 29.1 28.8 37.5 47.1 Greater Accra 72.6 85.3 83.0 87.7 90.5 Volta 13.1 16.0 20.5 27.0 33.7 Eastern 21.1 24.6 27.7 34.6 43.4 Ashanti 25.0 29.7 32.5 51.3 60.6 Brong Ahafo 15.6 22.1 26.6 37.4 44.5 Northern 13.0 20.4 25.2 26.6 30.3 Upper West 5.0 7.3 10.9 17.5 16.3 Upper East 3.9 6.7 13.9 15.7 21.0 Total 23.1 28.9 32.0 43.8 50.9 Source: Ghana Statistical Services (GSS, 2013) Table 3.3 Proportion of urban population by region (1960-2010) 3.4 Institutional arrangements governing urban lands in Ghana Land in Ghana is mostly owned by families and chiefs with about a tenth owned by the state (Kasanga, 1988; Abdulai & Ndekugri, 2007). As a result, access to land for various uses is generally through traditional institutions. In most urban centers (and in some cases rural areas), government continue to acquire lands from the requisite traditional institutions for purposes of establishing state institutions and agencies as well as decentralizing governance. In most instances, the government acquires a lot more land than it requires in anticipation of future developmental needs. This means most state acquired lands are not immediately used for their intended purposes and thus makes it possible for other uses such as urban farming. The traditional and state systems of landownership complement one another (Abdulai & Antwi, 2005; Abdulai, 2006). Whereas families and chiefs can lease their interest to individuals and businesses, certificate of ownership can only be issued by the state following the submission of requisite documents from the party leasing their interest. 46 46 Even though processes to obtaining land for urban and peri-urban farming across cities in Ghana are not clearly defined, individuals and families from whom such lands were acquired may continue to have the opportunity to cultivate the land. In other instances, employees of the various state institutions in whose merit the land was acquired may decide to cultivate the land or permit same from other persons known to them. On the other hand, individuals and businesses are also able to acquire land from the traditional institutions and defer development of such property. As a way of securing their investment, they permit family and friends to cultivate the land to register presence on the property. A common phenomenon in such scenarios is the erection of brick walls around the plot sending a signal to everyone that the interest in that land has been transferred to someone. 3.5 Institutional arrangements governing urban land use in Techiman and Tamale As noted earlier, nearly 90% of all lands in Ghana belong to individuals, families, and stools/skins (Abdulai & Ndekugri, 2007). The remaining 10% are held by the state, which is generally an acquisition from the traditional authorities who are the allodial owners. Ownership of land in both Techiman and Tamale are vested in the hands of chiefs who hold it in trust of the community. Natives of particular communities are generally permitted to cultivate such lands without paying ground rent while non-natives are permitted based on shared cropping arrangements or payment of ground rent. These arrangements affect the types of agriculture households can engage in. For instance, in Techiman, as a result of the land ownership structure and the fairly small size of the city, farmers are able to move to peri-urban and rural areas to cultivate other family or community lands for food at no extra cost. In the event that farmers are made to pay ground rent, the fees are usually moderate and serves as an incentive for urban farmers to travel to rural localities to cultivate land. In contrast, the rapid nature of urbanization and infrastructural development in Tamale (mostly by private developers) continue to shift the urban boundary to include the peri-urban spaces. Urban farmers who may have the intentions of engaging in peri-urban or rural farming still have to travel long distances to access land for production. These peri-urban lands can only be used for 4477 agricultural production purposes as long as they remain outside of the urban development boundary. This means that, rural lands continue to be counted as peri-urban lands and then urban lands depending on the speed with which development spreads to these areas. As Tamale continues to expand, access to peri-urban or rural land becomes a challenge and expensive to acquire for agricultural production. This may partly explain why male households will migrate (temporally) several kilometers to rural areas to cultivate land during the main production season and return with the food items after the season. 48 48 4. CONCEPTUALIZING URBAN FOOD SECURITY 4.1 Introduction In this chapter, the key concepts relating to urban household food security and livelihoods are discussed. While the discussion on food security has many parts (i.e. availability, access and utilization), the thesis focuses on the access component. In addition, the conceptualization focuses on food production by urban residents within the urban boundary and also outside the urban boundaries in the rural areas. Using the urban livelihoods framework, the various contextual conditions that can affect asset ownership of household members and their productive capabilities are discussed. The unit of analysis for the thesis as a whole is the household. To this end, the concepts will be connected to the household level. In addition, the different normative approaches to studying food security are also discussed in this section. 4.2 Household Food Security As mentioned in the introductory chapter, food security has been on the global radar since 1948 with the signing of the Universal Declaration of Human Rights. Food security assumed global attention after the world food crises in the 1970s-1972 to 1974 (UN, 1948). The concept and interest in food security however received a surge in the 1980s partly due to the famine in African countries; structural adjustment programs which led to deterioration in basic needs; and encouraged general intellectual enquiry in the field (Maxwell & Smith, 1992). The concept of food security has since undergone several definitional changes. However, the mostly widely cited relates to the definition agreed to at the 1996 4499 World Food Summit, which states that food security represents ‘a situation that exists 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, 1996). 4.2.1 The three pillars of food security Building on this food security definition, food security has generally been conceptualized in terms of what has been commonly referred to as the three pillars of food security: availability, access, and utilization (Barrett, 2010). Fundamentally, the concepts relate to each other. For instance, availability is necessary but not sufficient to ensure access to ‘sufficient, safe and nutritious food’. On the other hand, access is necessary, but not sufficient for effective utilization (Webb et al., 2006; Barrett, 2010). Similarly, availability, access and utilization could be disrupted in an unstable environment such as civil unrest, harsh climatic conditions etc. The World Food Programme defines food availability as ‘the amount of food that is present in a country or area through all forms of domestic production, imports, food stocks and food aid’ (WFP, 2009: 170). The term availability as observed by Riely et al., (1995) has the potential to focus on national and regional related analysis instead of the household level. However, availability at the national level does not guarantee access at the micro or household level. For example, although the national food balance sheets in the 1980s indicated food availability in most African countries, there were distributional challenges that prevented households from accessing food and hence remained food insecure. Access to food on the other hand is defined as a situation where the household has the ‘ability to acquire adequate amounts of food regularly through a combination of purchases, barter, borrowings, food assistance or gifts’. (WFP, 2009: 170). Access thus relates to choice sets available to the household, community or region subject to economic and markets conditions such as prices and income earning opportunities; policies and production related opportunities available to the household. These conditions depend on the types of policies that support production opportunities as well as the sociocultural belief systems including societal norms and practices through which food can be appropriated by the household. The close relationship between food security and poverty as well as to the socio-economic and political disenfranchisement becomes clearer using the access lens (Barrett, 2010). 50 50 Another pillar relevant to food security is utilization of food. This concerns the types of foods that households can afford and use. Critical to this concept is whether foods consumed are safe and can deliver the appropriate nutrition to the individual or household. In essence, utilization is concerned with dietary quality, especially accessing micronutrients associated with adequate intake of essential minerals and vitamins (Barret, 2010). A household is therefore considered food insecure when its members are unable to obtain sufficient quantities of food necessary for a normal and healthy life. This condition may be the result of lack of food or income to purchase food, or other factors such as poor distribution channels (FAO, 2012). Generally, urbanization poses complex challenges to poor households to satisfy the three conditions that could make them food secure. This notwithstanding, urban residents with access to productive resources to engage in own food production are able to overcome the challenge of food availability which has generally been a macro level analysis and appropriate food at the household level. Such households also reduce their dependence on the cash economy for food and thus are able to access food for domestic consumption. At the same time, the greatest challenge of the urban poor who may feed on cheap imported foods that lack the requisite micronutrients are addressed by the consumption of own produced foods including vegetables and fruits. Again, there is the interplay that exist between availability, access and utilization of food through the mechanism of food transfers. Increasingly, food transfers are becoming an important component to urban residents and contributes substantially to meeting their food security needs (Krüger, 1998; Andersson, 2002; Frayne, 2004a; Foeken & Owuor, 2008; Andersson Djurfeldt, 2012). Food transfers received by urban households satisfies the condition of accessibility of food to the household. However, the motivation for these transfers remains largely speculative. One thing that is obvious however is the fact that, these foods are sent to close relatives and friends (Krüger, 1998; Frayne, 2004a). Food security could also be viewed as a continuum such that, at the lower range, the focus is on access to food- making sure that households have adequate amounts of food at all times while at the upper range, the focus could be on food utilization-ensuring that, households obtain all requisite micronutrients needed from the consumption of particular foods. This continuum may influence households’ decision and choice of what to produce and where to produce. The decision of a household to produce food in either 5511 rural or urban space therefore may be a reflection of this continuum and how households perceive their choices to complement their food security needs. In general therefore, the concept of food security/insecurity can be related to the socio-economic, political, cultural and environmental factors available to the household, community, nation or region. Unfortunately, sub-Saharan Africa remains the only continent in the world to still have severe food production shortfalls and this has largely been blamed on economic development failure, even though evidence abounds to the effect that economic growth of a nation does not necessarily translate into food security for all (Sen, 1981; Salih, 1995). 4.2.2 Household food security and productive assets Conceptually, food insecurity could be classified as an issue of general decline in food access, entitlement failure, and as a human rights issue. Improvements in the asset wealth of households (including the provision of infrastructure and production equipment) are means which enable people to pursue livelihoods to improve their access to food. Access to productive assets for instance enables households to invest in the production of food for own consumption and also store values which serves as an insurance in the event of shocks. Assets available to households or communities constitute a ‘stock of capital’ that can be stored, accumulated, exchanged or depleted and put to work to create an income flow (Rakodi, 1998). The natural capital is the wealth of natural resource stocks including land, water and other resources (Carney, 1998; Rakodi, 1998; Booth et al., 1998). Households can also exchange the labour assets for income either through direct channels such as monetary exchange through formal wage employment or through indirect channels such as production of goods and services for sale (World Bank, 1991). The household is therefore able to appropriate their entitlements based on the type of assets they command. Ownership of land in urban space provide households the opportunity to engage in agricultural production, rent the space for income, or provide other investments that bring returns to the households. In such cases where the household is able to command resources, there is a higher likelihood of food security relative to other households with no command over assets or other resources. 52 52 The factors influencing exchange entitlement given particular ownership are identified as whether the household member in question can find employment or not which is tied to income earning opportunities among members of the household (Sen, 1981; Dreze & Sen, 1989). For household members to fully engage in trading their labour assets for income, it will depend on how long they are prepared to work and at what wage rate and how much it costs them to buy whatever they may wish to buy (Sen & Dreze, 1999). Also, an assessment of what could be produced with the household’s own labour assets and resources as well as the costs of purchasing the resources and the value of the products that can be produced all contribute to making the household decide whether to exchange their labour assets or invest it in production of goods. 4.3 Normative approaches to food security The human rights approach to studying food security derives from the Universal Declaration of Human Rights (1948) which recognizes the inherent dignity and equal rights of all people of the human race to have the right to live and access to basic human livelihood in a dignified way. Article 25(1) of the declaration states that ‘everyone has the right to a standard of living adequate for the health and wellbeing of himself and of his family, including food, clothing, housing and medical care and necessary social services, and the right to security in the event of unemployment, sickness, disability, widowhood, old age or other lack of livelihood in circumstances beyond his control’. In addition, the right of people to freely participate in the cultural life of their community, enjoy the arts and to share in scientific advancement and its benefits are also guaranteed in article 27(1). These articles embody the totality of the vulnerable in society and how their human dignity ought to be guarded to create an environment that enables them to fully participate in society. A good point to start in the food security discussion will be an advocacy, based on human rights that seek to provide some level of assurance to the vulnerable of a certain minimum food entitlements that will guarantee their human dignity. These guarantees work in an environment where there is rule of law and good governance through democracy which allow all people in society to participate in making decisions that affect them (Eyben & Ferguson, 2000; Mikkelsen, 2005; Li, 2007). However, civil and political rights granted through democratic elections and freedoms alone do not achieve development without 5533 the granting of economic, social and cultural rights although these rights were more easily achieved under democratic rule than dictatorship (Yaro, 2004). This kind of contract (what Sen refers to as a political contract) makes it a demand on political leadership to act and make decisions that will be in the interest of the people (Dreze & Sen, 1989; Yaro, 2004) especially decisions about hunger, production, distribution and utilization of natural resources. The right based approach essentially argues that, the war on fighting hunger cannot be left to a certain marginalized group in society to battle their way to survival. Rather, it should be a deliberate and collective effort to ensuring that, everyone is entitled to a fair share of food and other resources enough for a meaningful life. This will require comprehensive global, continental, regional, national, societal, household and individual level contributions (Effah-Abedi, 2014) to create an enabling environment that will birth individual and collective potentials in a well-defined social, economic, environmental and political framework. The approach advocates a non-discriminatory methodology in the fight against hunger. Although the right based approach sounds superior in the food security debate, the main bottleneck has to do with the implementation of these international ratifications which although countries are signatories to, sometimes find it difficult to implement. The problem with this approach is the extent to which social, economic, and cultural rights are enforced. Resource constraints, inappropriate policies and the dilemma in assigning rights in societies with conflicting interests groups impedes the agenda of this approach. Often times, groups with the ‘loudest’ voice or representation turn to have a major stake in deciding what the landscape can be used for at the expense of the poor and vulnerable. Following the above, ‘access to food could be determined by the food entitlements derived from the human and physical, assets and stores, access to communal property and a variety of social contracts at the household, community and state level’ (Maxwell & Smith, 1992). Entitlements failure will therefore increase the likelihood of vulnerability and food insecurity. 54 54 4.4 Explanations for lacking food security This sub-section discusses the various theories that seek to provide insights on why households remain food insecure. Specifically, the food availability decline and the entitlement theories to food security are discussed. 4.4.1 The food availability decline (FAD) theory This theory which is largely national and regional in scope focuses on food supplies as a major cause of food insecurity. This has generally been analyzed as a supply and demand nexus with great emphasis on quantities of available food versus needs and net imports needs versus capacity to import (Von Braun, 1992; Salih, 1995). This foundation led to the crusade for countries to invest in agriculture with the firm belief that, once supply is enhanced, food will be in abundance and thus address the insecurity problem. The approach called for investment and policies that built on the ‘complementarity of food and crop production’. This according to Yaro (2004) led to huge investments in green revolution technologies designed to increase food supplies for both national self-sufficiency and for export. The returns on these investments have been encouraging; especially in Asia and Latin America where productivity per acre increased many fold. The food availability decline approach argues that, an acute decline in the supply of food is a necessary condition for famine to emerge (Fine, 1997). Hence under normal circumstances a growing supply of food relative to population is sufficient to guarantee the absence of famine (Yaro, 2004). This approach largely depended on the use of supply and demand methods to measure deficits in relation to aggregate requirements resulting in supply related factors such as irrigation, land access, innovation and technology receiving greater attention with one central goal of improving food production to match the needs of the growing population. The approach however fell short in addressing the problem of food insecurity in several countries despite the increasing world food production. At the macro level, higher production was recorded but most poor households and individuals could not access available foods due to poor distribution channels. This led to the harsh realization that food availability alone does not ensure food access by all. The revelation led to the decline of this approach and the emergence of the entitlement approach (Nyborg & Haug, 1995). 5555 4.4.2 The entitlement theory A major conceptual shift was experienced with the introduction of the entitlement approach to food security/insecurity analysis which moved the focus from a short-term phenomenon to a long-term perspective. The entitlement approach finds its root in the work of Amartya Sen (1981) which has since been regarded as a fundamental theoretical underpinning of the concept of food security. In his essay on poverty and entitlements Sen (1983) summarized the entitlement approach which concentrates on the ability of people to command adequate food given the established societal procedures including production possibilities, opportunities through trade, state allocations etc. The ability to command enough food through these channels enable a person to avoid starvation. If a person is found starving it is a reflection of the individual’s inability to command enough food, or the lack of ability to appropriate the established channels to their advantage. The entitlement framework has its analytical structure grounded in neoclassical general equilibrium theory (Fine, 1997) and seeks to examine what conditions are necessary to ensure an individual’s set of entitlements (Yaro, 2004). Entitlements according to Sen are the set of alternative commodity bundles that a person can command in society using the totality of rights and opportunities that he or she faces. Entitlements, he argues depend on the legal, political, economic and social characteristics of the society in question and the person’s position in that society (Sen, 1983). These commodity bundles in themselves must be enough to meet the minimum daily calorie and nutritional requirements in a sustainable way without which a person may face starvation. In a later work Sen and Drèze, argue that the understanding of entitlements should be at the individual level where the ability of the individual to acquire food and other commodities within the prevailing economic, social and legal arrangements is the focus (Sen & Dreze, 1999). The approach also analyzed the concept of poverty and starvation and indicated that, starvation has to do with the kind of relationship that exists between a person and their ability to access essential commodities in a sustainable manner. Entitlement theory has however received diverse criticism including the fact that the approach is vague and has several interpretations and associations attached to the term entitlement. As a result, the approach does not specify one particular cause of famine but a general conclusion that famine is a reflection of failure in entitlements (Sen, 1995). This has led to the use of the concept in 56 56 different applications and interpretations way beyond its relevance as a framework for analyzing famines within the boundaries of an economy. Secondly, the approach seeks to imply that individuals could translate their endowments into food to address the problem of famine but there is no linear progression in the conversion of assets into food (Faridi & Wadood, 2010; Daie, 2014; Davies, 2016). People may rather preserve their assets for future vulnerabilities (Waal, 1989; Al-Hassan, et al., 1997). In some cases, the poor would rather choose to be food insecure and put their assets in productive investments than to lose the possibility of future livelihood (Waal, 1989). The entitlement theory also has a weakness in confining its analysis into one final outcome - food Yaro (2004). The approach is only a part of a whole and an analysis of entitlements as an entity does not give a complete picture. In the words of Davis (1996), food security is only a part of livelihood security strategy and the decision of a people within any geographical boundaries are borne out of a complex web of economic, social and political interrelationships that develops over time. The entitlement approach primarily considers the household as the unit of analysis but this becomes problematic once the argument is stretched to the larger population. The approach therefore fails to outline a framework that would link the micro level issues to the macro variables. Whereas Sen argues that famine does not occur in a democracy, the approach in a way does not fully explain the complex cause of famine in certain regions especially, Africa where victims of famine are only the result of selfish political interests that render them powerless in the distribution and access to resources. 5577 4.5 Household food security, assets and urban livelihoods Conceptually, the livelihood framework has been used to assess and understand poverty and deprivation, mostly in rural settings but it has been generally accepted to have applicability in urban and peri-urban contexts as well (Rakodi, 1997; Carney, 1998; Moser, 1998; Rakodi, 1998; Tacoli, 1998a; Rakodi, 2002). A livelihood according to Carney (1998) consists of capabilities, assets (including both material and social resources), and activities necessary for a means of living. When the livelihood in question is able to withstand complex occurrences such as shocks and stresses without compromising the natural resources base, we classify such a livelihood as sustainable (Carney 1998:2) Broadly speaking the point of departure for the livelihood framework focuses on identifying what the poor have (the wealth of the poor) instead of what they lack (UNDP, 1998; Moser, 1998). Even though rural and urban are generally treated as two distinct settings, there are commonalities in the principles underpinning the livelihood approach in these two settings. This notwithstanding, contextual differences relating to economic, social, governance and environmental factors exist between rural and urban settings and also among different urban settings (Meikle, et al., 2001). It is therefore imperative to understand and account for these contextual differences as they have the potential to influence the livelihoods of the urban poor. To have an appreciation of the urban sustainable livelihood framework requires an understanding of the fundamental predisposition that poverty is not a static or a permanent condition. Rather, people may move in and out of relative poverty subject to the opportunities, threats, stresses and shocks that confront them-these could be social, economic, governance or even environmental aspects (Chambers, 1995; Moser, 1998; Meikle, et al., 2001). Survival among the urban poor therefore is based on different multifaceted coping strategies that draw on their assets (Chambers, 1995; Rakodi, 1997; Burnell, 1998). Such assets can be classified into tangible and intangible (Sen, 1992; Moser, 1998). As indicated earlier, livelihood options are generally different across rural and urban and inter-urban settings depending on contextual factors (Meikle, et al., 2001). Depending on the opportunities, threats or shocks that may confront households, they respond by allocating their asset portfolios including social 58 58 capital, and capabilities either as a short-term or long-term strategy (Meikle, et al., 2001; Development, et al., 1994). Swift analyzed vulnerability and insecurity as a function of assets which is classified as investments (human investments including education and health; individual productive assets such as houses and land; and collective assets such as irrigation systems and access to common property resources); stores (food, jewelry and money); and claims on other networks including kinship relationship, friendships, community, government or international community (Swift, 2006:44). 4.5.1 Elements of Sustainable Livelihood For any livelihood approach to be described as sustainable, it must have the following characteristics as espoused by Meikle et al., (2001). First, the livelihood must take into account how to translate available assets (both material and social), capabilities and all the activities that the household engages in that can help to give them a livelihood (Sen, 1992; Chambers, 1995). The opportunities may be diverse depending on the spatial location of the household and the type of resources at their disposal (Wratten, 1995; Moser, 1998; Tacoli, 1998b). Additionally, the livelihood must account for social networks and any institutions that enhances access to resources (De Haan, 1999). Secondly, the livelihood must have the potential to respond to change. That is, it should be flexible and adaptable and must have the ability to withstand and recover from shocks in a sustainable manner, at least in the long term (Chambers & Conway, 1992; Singh & Titi, 1994). Thirdly, the livelihood must be based on the choices, capabilities and priorities of the affected people keeping in mind that the people with capabilities are at the core of the framework. As a result, they should be treated as such but not as helpless vulnerable. Fourthly, the livelihood must also be household and community centered. Thus it must be sensitive to the community and the context specificities. This is important because different members of the household will contribute differently depending on their capabilities and specific roles assigned them which also draws on the social capital and kinship relations available to members of the household (Putnam, et al., 1994; Moser, 1996). 5599 4.5.2 The Urban Context and the Sustainable Livelihood Framework In order to have a complete appreciation of the sustainable livelihood framework and its applicability to the urban context, I revisit the key issues relating to the urban context outlined in earlier sections. In particular, the social, economic, governance and environmental contexts are discussed in this section. 4.5.2.1 The Urban Social Context As noted by Meikle et al., (2001), the cosmopolitan nature of cities makes them diverse and fluid in structure, departing from the otherwise more organized and stable rural structure. Social capital presents itself as an important asset available to both urban and rural households. Social capital is used to represent the organizational trust, norms and networks with the potential to improve societal efficiency. Social capital also relates to the wider social relations and networks available to both urban and rural people. This is considered an important resource for both urban and rural families and has received considerable attention in the literature as a major asset in the event of crises (Moser, 1996; Douglass & Friedmann, 1998). One important component of social capital is the remittance of cash from urban households to rural networks. Even though this type of remittance has gained considerable attention in the economic literature, recent trends suggest that there is also a flow of resources in the form of food from rural areas to urban households (Andersson Djurfeldt, 2012, 2015a). This is done as part of the social contract that may exist between individuals, groups or communities. In effect, the underlining characteristics of social capital is grounded in relationships and context specific scenarios. Social capital thus is expressed through strong rural-urban linkages that are often activated to support each other in crises. Social capital therefore transcends beyond the boundaries of the city to include the wider rural-urban interactions (Tacoli, 1998). Thus, households with stronger social capital would have a lower likelihood to experience food insecurity relative to those with weak social capital. 60 60 4.5.2.2 The Urban Economic Context Urban areas and cities present a variety of opportunities for residents to generate wealth but at the same time also a myriad of challenges to the urban poor as they tend to depend on the cash economy for their livelihoods. The commercial nature of urban areas where basic necessities of life such as food and shelter are obtained through the market places extra burden on poor households to get cash income in order to command food and other needs (Wratten, 1995; Satterthwaite, 1997; Meikle, et al., 2001). This leads the urban poor to consider a range of activities including urban farming and engagement in informal related businesses as a means of livelihood. Notwithstanding the challenges confronting the urban poor, most rural people, especially the youth continue to migrate to urban centers for the opportunities that it presents them. As noted by Meikle et al., (2001), the urban informal economy is not the reserve of the poor as other non-poor households are also found to actively engage in that sector. In fact, evidence abound that, urban agriculture is not necessarily practiced by the poor as they lack access to land which is considered an important resource of production (Ayerakwa, 2017; Mkwambisi et al., 2011). It is noteworthy however that, not all households are able to find productive land for own food production or actively engage in the informal industry. As a result, the food security situation of the urban poor worsens when poverty is urbanized. 4.5.2.3 The Urban Governance Context In the spirit of decentralization, management and the provision of infrastructure are the responsibilities of city authorities. The urban poor like most city dwellers depend on these infrastructure and services to map out their livelihood strategies. However, in the provisions of infrastructure, the poor are often left out or discriminated against for reasons that, they live in unauthorized structures or engage in either unauthorized business or operate businesses in unauthorized locations (Katepa-Kalala, 1997; Walker & Meikle, 1999). For example, in Ghana, urban food production mostly takes place on state reserved lands which does not provide any form of tenure security for the farmers to make any meaningful investments. These practices are considered a nuisance to city development by the local government and thus refuse to provide any form of services that will promote the practice to such farmers (Ayerakwa, 2017). 6611 4.5.2.4 The Urban Environmental Context In developing countries experiencing increasing urbanization, access to decent housing and related services become expensive to the poor as towns and cities grow. The quality of housing and working conditions and its related health implications of the poor have received attention in the literature (Meikle et al., 2001). In most developing countries like Ghana, the urban poor are often concentrated in densely populated neighborhoods or uncompleted structures with minimal or no social amenities. The lack of basic social amenities such as water and toilet facilities exposes the poor to the inhuman sanitary conditions leading to infectious diseases (Werna, 1998). The decision of poor men and women to live in ‘low scale’ neighborhoods is often a trade-off to sacrifice their health and ‘luxury’ in order to preserve their assets to engage in productive activities that will guarantee some form of livelihood. On the contrary, if a household member falls ill due to insanitary conditions, it could also deny the same household the opportunity to engage in any form of livelihood activity (Miekle et al., 2001; Douglas 1998). 4.5.2.5 The Urban-Rural Multi-Spatial Context Increasingly there is a consensus that the urban and rural context must not be treated in isolation as happenings in either urban or rural areas have implications for each other. For instance, in Ghana and by extension most developing countries in Africa, there is a strong interaction between urban and rural areas in the form of people and goods. Some urban residents live in areas classified as rural but work in urban areas while a number of urban residents live in areas classified as urban but source their livelihoods from rural areas including farming in rural areas and seasonal migration. Other households depend on food transfers from rural areas to supplement their food needs in urban areas. As a result, rural-urban interactions play an important role in understanding how urban residents, especially the urban poor construct their livelihoods (Tacoli, 1998) Multi-spatial livelihoods may result from the command over natural resources either in urban or rural areas in combination with social networks. Urban households who wish to participate in rural farming for example will need to draw on this important relationship to be able to command land for production. Similarly, households draw on the social network capital to receive cash remittances from urban networks as part of their livelihoods in the rural areas. In essence, the lack of social networks hinders the ability of households to command social capital for food production or food transfers which has 62 62 implications on their food security (Andersson, 2002; Andersson Djurfeldt, 2012, 2015a). 4.6 Conclusions – linking concepts to the research questions Urban households are faced with the option of appropriating their assets into production, consumption and exchange. Assets can be transformed into production directly or through exchange. Similarly, production and exchange can be used to create assets. This includes urban food production available to households for consumption and exchange of products for income through the market. As indicated earlier, households without productive resources such as land can still exchange their labour to acquire other forms of assets (investments and stores) or exchange for consumption. Understanding urban food security needs therefore requires that, we analyze the contextual factors that independently or collectively explain the food security situation of the specific households in question and whether the underlining factors are the same across specific cities. For example, among urban residents who admit to engaging in own food production, the context differ from engagement in only urban agriculture to growing food solely in rural areas. Obviously, the factors underlining participation in own food production will differ by the type of production and also by city specific contexts. Similarly, the food provisioning arrangements tied to social or kinship relations may differ across different sociocultural settings. In effect, the different agricultural production arrangements and food transfer receipts may have different implications for the producing household’s food security situation which is the focus of this thesis. 6633 64 64 5. METHODOLOGY 5.1 Introduction The quantitative data for this thesis forms part of a broader collaborative research project between Lund University, the University of Ghana and the University of Nairobi being carried out in Ghana and Kenya with the aim of understanding the social, economic and environmental challenges and prospects of urban agriculture under changing global and demographic realities. For the Ghana case study, Techiman municipality and Tamale metropolitan assemblies were selected. Interviews were however conducted in the urban areas of Techiman and Tamale. The two towns were jointly selected by the research team in Lund University and their Ghanaian collaborators based on the rapid population growth in the two cities tied to changing marketing structures and diets away from traditional consumption to modernized diets including exotic vegetables. These standards tend to favour urban agriculture production as proximity matters in the production of such vegetables. The choice of the two cities (Techiman and Tamale) therefore allows for comparison and assessment of the scale and contribution of urban agriculture to urban food security. The study employed a mixed method methodology to analyzing the phenomenon of food security and urban agriculture. Specifically, the thesis is written mostly based on the quantitative data. The qualitative data however provides the setting and background to which the results of the quantitative data are interpreted. The use of these two approaches give the advantage of obtaining information that supplement each other. In addition, the qualitative information aids in contextualizing findings of the study as well as other useful information such as socio-cultural information that may be impossible to capture quantitatively. This study therefore is biased towards quantitative methodology. 6655 5.2 Study site selection and description As indicated earlier, two medium sized cities, Techiman municipality and Tamale metropolitan assemblies in the Brong Ahafo and Northern regions respectively were purposively selected for the study in Ghana, breaking away from the metropolitan bias that has characterized earlier studies on agriculture in urban areas (Obosu-Mensah, 1999; Obuobie, et al., 2004; Probst, et al., 2012; Ayerakwa, et al., 2014; Danso, et al., 2014; Drechsel, et al., 2014). These earlier studies generally lacked context and scale. This gap however is filled in the thesis. 5.2.1 Techiman Municipality and Tamale Metropolitan Assemblies Techiman municipality is located in a central area that connects to the Northern, Upper West and Ashanti regions as well as other big cities in the Brong Ahafo region of Ghana (Figure 5.1). The unique location of this municipality makes it very attractive to several traders of primary food stuff to converge to do business. The main market in Techiman is projected as the largest primary food market in Ghana, which also attracts several traders from neighboring West African countries (mainly from Togo and Cote d’Ivoire) on a weekly basis. 66 66 Figure 5.1 Map of Techiman township highlighting major roads 6677 The municipality has a total land area of 669.7 square kilometers and has an annual rainfall ranging between 1250mm and 1650mm with an average temperature of 28ºC (MoFA, 2014). The main agricultural activities include crop farming and livestock rearing (MoFA, 2014; GSS, 2013). Despite being defined as urban area in the census data, the municipal’s economy is mainly rural and dominated by the agriculture sector with nearly half of the municipal population engaged in agriculture (46.2%). The Ghana Statistical Services classifies households as agricultural if at least one member of the household participates in agricultural production (GLSS, 2012). Nearly all urban and rural agriculture households (93% and 98% respectively) in Techiman are crop farmers. Even though rural farming is the dominant agricultural household type, comparing urban and rural households shows that the proportion of urban residents engaged in urban farming is substantial (about 49%) while a little over half (51%) of agricultural households remain rural Tamale Metropolitan Assembly on the other hand is the administrative and regional capital of the Northern region of Ghana. The metropolis has a population of about 223,252 as at 2010 with more females (50.2%) than males. The share of the population in the metropolis who are urban is estimated at 80.8% with an average household size of 6.2 members but higher for rural Tamale (7 members). The main economic activity in the metropolis include formal wage employment, trading in food crops (common among women) and farming on urban spaces. Some urban households also have farms in rural areas although they reside in the city. For the farmers who engage in urban agriculture, vegetable production is their main focus, with products usually sold to urban consumers. More so, with Tamale as the administrative capital of the Northern region, it provides home to hundreds of government employees and other visitors who seek to do business. Tamale can also boast of the presence of financial institutions, several dozens of Non-Governmental Organizations (NGOs) as well as private for profit organizations. Farming in the city mainly focuses on vegetable production whereas rural farming focusses on staples such as maize, millet, yam and groundnuts. 68 68 Figure 5.2 Map of Tamale township highlighting major roads and neighbourhoods 69 69 5.2.2 Characteristics of Techiman and Tamale The choice of Techiman and Tamale for this study provides insight into the contextual nature of food security since the prospects for engagement in urban and rural agriculture in the two cities are different. Available estimates based on census data suggest that more than three-quarters (88.4% and 83.2%) of households in Techiman and Tamale respectively are employed in the private informal sector. Higher proportions of households are employed in the public formal sector (11.3%) in Tamale than Techiman (5.7%). Whereas the household structure in Techiman is mostly nuclear (head, spouse and children) (27.9%), the household structure in Tamale is mostly extended (household head, spouse, children and other family relatives). Table 5.1 suggests that nearly half (46.2%) of households in Techiman Municipality participate in agriculture compared to about a quarter (26.1%) in Tamale Metropolitan Assembly. Of this however, about 57% are engaged in urban farming in Tamale compared to about 49% as found in Techiman- an indication that, urban farming is more pronounced in Tamale relative to Techiman. Most of these farmers practice mixed cropping and supplement their farming activities with animal rearing which serves as investment or insurance among family members for payment of hospital bills during illness, payment of school fees, and even providing a befitting burial in the event of death of a member of the household. The data also suggests that, one in every four (24.5%) households in Techiman are involved in rearing livestock compared to about 12.9% in Tamale. The share of urban households engaged in animal rearing is however higher in Tamale than in Techiman (49.8% and 42.4% respectively). Livestock are usually kept in urban areas and allowed to graze on open grounds or kept within an enclosure and fed. Unlike Techiman, the proportion of agricultural households engaged in urban agriculture relative to rural agriculture is 57% and 43% respectively in Tamale. Crop production and tree planting and rearing of livestock remain the three most practiced types of agricultural activities in the metropolis. 70 70 Techiman Total Urban Rural Type of activity N % N % N % Total households 34,137 100.0 23,566 69.0 10,571 31.0 Households engages in agriculture 15,781 46.2 7,769 49.2 8,012 50.8 Crop farming 15,073 95.5 7,222 47.9 7,851 52.1 Tree planting 146 0.9 94 64.4 52 35.6 Livestock rearing 3,870 24.5 1,639 42.4 2,231 57.6 Fish farming 7 0.0 6 85.7 1 14.3 Tamale Agricultural activities Total Urban Rural N % N % N % Total Households 35,408 100.0 29,322 82.8 6,086 17.2 Households engages in Agriculture 9,251 26.1 5,244 56.7 4,007 43.3 Crop Farming 7,842 22.1 4,150 52.9 3,692 47.1 Tree Planting 48 0.1 27 56.3 21 43.8 Livestock Rearing 4,575 12.9 2,278 49.8 2,297 50.2 Fish Farming 7 0.0 5 0.0 2 0.0 Source: Calculations based on GSS 2010 population and housing census data Table 5.1 Agricultural activities in the Techiman and Tamale Assemblies While households in small cities like Techiman could have access to food from own production from surrounding villages, households in intermediate sized cities like Tamale on the other hand may not have the same opportunity. This is because proximity and access to rural land for production may be easier in small towns than intermediate sized towns. Even if urban households do engage in rural agriculture in the intermediate sized cities, it may come at a higher cost, since land is farther away. Another important characteristic across the sample cities relates to the context of engagement in agriculture. For example, less than a third of households in Tamale engage in agriculture compared with nearly half of households in the case of Techiman. Similarly, the share of rural households engaged in 7711 agriculture is higher in Techiman than Tamale even though they all participate in agriculture. Moreover, Techiman and Tamale fall within two different agricultural ecological zones, the transition belt (Techiman) and the northern savanna belt (Tamale) with different production potentials. Techiman Tamale N % N % Household head population by sex 34,137 23.5 35,408 16.1 Household structure Nuclear (head, spouse and children) 40,590 27.9 42,817 19.5 Single parent nuclear 16,202 11.2 10,654 4.8 Single parent extended 21,330 14.7 28,150 12.8 Extended (head spouse(s) children head's 34,212 23.5 101,495 46.1 relatives) Marrital status 101,020 155,046 Married 43,742 43.3 75,352 48.6 Widowed/divorced or seperated 8,385 8.3 10,233 6.6 Religious affiliation Non religious 7,843 5.3 374 0.2 Islam 40,545 27.4 202,003 90.5 Christianity 97,362 65.9 19,749 8.8 Source: Calculations based on GSS 2010 population and housing census data Table 5.2 Demographic characteristics of urban households in Techiman and Tamale Other demographic characteristics based on census data across the two cities are presented in Table 5.2. Nearly a quarter (23.5%) and a fifth (16.1%) of the households are headed by women in Techiman and Tamale respectively. The households are structured differently across cities, however. In Techiman, nearly a third (27.9%) of households are structured as nuclear (head, spouse and children) followed by those structured along the extended family system (head with spouse, children and head’s family). In Tamale on the other hand, 72 72 households are structured primarily along the extended family structure (46.1%) followed by the nuclear family structure (19.5%). On the marital statuses of household heads, nearly half of household heads in Tamale are married (48.6%) compared to about 43.3% in Techiman. In contrast however, about 9.0% and 7.0% of household heads in Techiman and Tamale respectively are either widowed/divorce or separated. About two-thirds (65.9%) of household members indicate their religious affiliation as Christianity in Techiman. Households in Tamale on the other hand subscribe to the Islamic faith with over 90% indicating their allegiance to Islam. 5.3 Sampling Technique As part of the broader research design, a total sample of 1,000 households from each sampled city was estimated to be interviewed. To start with, the focus of the study was to identify and interview households in urban Techiman and Tamale. As a result, the assistance of the local Planning Office was sought to give clear demarcations of the boundaries of the urban center. The Research Team and Officers from the assembly drove first to identify the city center. It was possible to locate the center of each city because the cities in question are not mega but medium sized in nature. Based on that information, each city was put into four quadrants with the center serving as the epicenter. The boundaries of each quadrant was also identified. The goal was to have proportional representation of households across the city. In the second stage, sub communities within each quadrant were identified and proportionately stratified for equal representation. In the third and final phase, enumerators were divided into teams of four with each team responsible for a sub community. The spread of surveyed households in Techiman and Tamale are presented in Figures 5.3 and 5.4 respectively. 7733 Figure 5.3 Map showing the distribution of sampled households in urban Techiman 74 74 Figure 5.4 Map showing the distribution of sampled households in urban Tamale 7755 At the sub community level, the sampling design was repeated by first identifying the center of the community with the help of the Assembly Member or other Opinion leaders. Similarly, each sub community was also divided into four quadrants with each team member surveying a quadrant, starting at the center of the community and moving in a serpentine path. Enumerators surveyed and interviewed one household in every third dwelling. Two households were however selected from each dwelling in densely populated sub communities. The questions were administered mostly to the heads of household. In the absence of the household head the most knowledgeable adult was interviewed. Interviews were conducted in the language the respondent was most comfortable with. In Techiman, the local dialect Twi and English were mostly used while Dagbani and English were used in Tamale. A total of 2,020 households were surveyed in both cities distributed as: Techiman (1,019) and Tamale (1,001). 5.4 The survey instrument and survey administration 5.4.1 Quantitative data collection To begin with, a standard questionnaire containing several household indicators on food (in) security was jointly designed for uniformity across project countries. Some of the captured indicators include questions on household demographic and housing characteristics, household cash incomes from all sources etc. Others include questions relating to household food access, dietary diversity, months of adequate food provisioning, food price changes, sources of food, urban agriculture, food transfers and food aid. As part of the project design and working closely with researchers from the University of Ghana, an advertisement was placed at the campuses of the University of Ghana in Accra and the University for Development Studies in Tamale to recruit enumerators for the study. Individuals were shortlisted based on agreed criteria including qualification (at least Bachelor’s degree); experience with data collection and ability to speak any of the local languages spoken in the study area. A total of about 60 enumerators were finally hired to assist with data collection. Training of enumerators was put into two. 76 76 First, a two day training workshop was organized for enumerators that were hired in Accra at the University of Ghana. Discussions at the training focused on three things: understanding the meaning of each question, accurately translating the questions into the local language; and learning to efficiently use the survey tablets, including picking GPS coordinates of all households that were interviewed. This process was repeated for the second cohort of enumerators that were hired in Tamale who were used to survey the city of Tamale. My role at the training was in two parts-assist with explaining questions to enumerators and to help translate questions into the local language and vice versa during ‘role play’ sessions. In the field however, I led a team of enumerators and assisted with community entry protocols. Thereafter, I supervised enumerators-moving from one household to another to address any difficulties that arose during interviews. Each day after the interviews, I assisted in coordinating the debriefing sessions. At these meetings, enumerators raised difficulties encountered in the field for redress. The use of tablets instead of paper had an added advantage of potentially minimizing the error margin on the part of enumerators as we were able to build in logics in a sequence that questions were expected to be asked and answered. Supplementary paper questionnaires were however printed and handed to each enumerator in the unlikely event of machine breakdown. The above notwithstanding and for the fact that the questionnaire was designed to be replicated in all project countries, it had an inherent limitation to address all the research questions in this thesis. To ameliorate this, a new design was developed to allow the collection of additional qualitative data that could help shed more light on the specific research questions of interest. The qualitative data collection was therefore designed to focus on the opinion leaders in the respective communities from which quantitative data was collected. 5.4.2 Interviews and Consents There are sub-communities in each of the urban areas we visited with local leaders for each community. These leaders (chiefs/sub chiefs) are responsible for their people. With this in mind, the research team first visited the palace with the aid of the Assembly Member for the electoral area in question and introduced the project to the sub community leadership. After receiving the consent from these individuals, the teams then applied the study design to 7777 recruit households into the study. In order to pick the GPS coordinates, all quantitative interviews were conducted at the residence of the selected household. Each interviewer began by introducing him/herself and the project to the household, highlighting that it was meant for research purposes and had no affiliations with any political groupings. Once consent was sought, interviews commenced. The respondent who in most cases was the head of the household, was encouraged to consult other household members for issues that other members could best provide answers to. 5.4.3 Qualitative data collection After analyzing the quantitative data, there was a need to provide more context to explain the observed patterns as well as address the overall research objective of the thesis. To do this, a qualitative follow up study was designed to account for the limitations in the quantitative datasets, in the attempt to put findings of the thesis in context. With the qualitative interviews in mind, names and contacts of key informants including Chiefs, Assembly Members as well as other Opinion Leaders were purposively collected during the quantitative data collection to aid any future engagement. During the follow up, selection of respondents to participate in the interview was conditioned on their locality and willingness to be interviewed. First, the sub communities were grouped according to quadrants and respondents selected from each quadrant. This was done to have a fair representation of opinions across the city. In each city, efforts were made to also solicit opinions of principal officers working in the field of agriculture as well as those with the mandate to plan and manage the city. In this regard, the qualitative interviews focused on institutions and individuals that control the use of land in urban settlements- that is local authorities who have the mandate of planning the city, the chiefs who are the owners of land and households who find their livelihoods in urban agricultural landscapes. Representatives of the Ministry of Food and Agriculture (MoFA) and key informants from non-governmental organizations and development partners working in the field of urban agriculture were also interviewed in each city. As indicated earlier, key informants were selected from a list of opinion leaders compiled in each city during the quantitative data collection phase. A total of 21 interviews were conducted in both cities in August 2015. Personal consents were sought first for the interview and also for audio recording the interview 78 78 sessions to aid transcription. For persons who were not comfortable with recording, notes were taken instead. All qualitative interviews were personally conducted, mostly in the Akan (Twi) dialect in Techiman and a blend of English and Twi in Tamale. Interviews began by fixing an appointment with a respondent, usually a day or two before the interview. Respondents were given the opportunity to decide the meeting place as may be convenient to them. Before the interview, they were reminded of the quantitative study in 2013. Using interview guides designed in line with the research questions of the thesis, questions (and probing questions) were asked and recorded in addition to taking of notes as a backup. Data was collected on several indicators including land ownership structure, how such lands could be accessed in the cities for different uses, perception of urban agriculture among the different opinion leaders as well as the threats and opportunities, and other known policy frameworks (if any) guiding the practice of urban agriculture in the selected cities. These questions, together with several other questions were very important in placing the contribution of urban food production to household food security in context. The list of participants during the qualitative interviews are presented in Table 5.3. 7799 Name Designation Date of interview Participant 1 Municipal Planning Office, Techiman Aug. 4, 2015 Municipal Agriculture Office, Ministry of Food and Participant 2 Aug. 4, 2015 Agriculture, Techiman, Participant 3 Chief, Techiman Aug. 4, 2015 Participant 4 Assembly Member, Techiman Aug. 5, 2015 Participant 5 Unit Committee Chairperson, Techiman Aug. 5, 2015 Participant 6 Imam, Techiman Aug. 5, 2015 Participant 7 Chief, Techiman Aug. 5, 2015 Participant 8 Assembly Member, Techiman Aug. 6, 2015 Participant 9 Chief, Techiman Aug. 6, 2015 Participant 10 Chief, Techiman Aug. 6, 2015 Participant 11 Chief, Techiman, Aug. 7, 2015 Participant 12 Lecturer, University for Development Studies, Tamale Aug. 10, 2015 The Executive Director, Urban Food Security Network Participant 13 Aug. 10, 2015 (URBANET) Tamale Participant 14 Tamale Metropolitan Assembly, Tamale Aug. 10, 2015 Participant 15 Works and Housing Department. Tamale Aug. 10, 2015 Participant 16 Metropolitan Department of Agriculture Aug.t 11, 2015 Participant 17 Assembly Member, Tamale Aug. 11, 2015 Participant 18 Assembly Member, Tamale Aug. 11, 2015 Participant 19 Assembly Member, Tamale Aug. 12, 2015 Participant 20 Assembly Member, Tamale Aug. 12, 2015 Participant 21 Assembly Member, Tamale Aug. 12, 2015 Table 5.3 List of participants during qualitative interviews 80 80 5.5 Method of Analysis The analytical methods employed in this thesis are mostly quantitative in nature but with some qualitative underpinning that provides contextual insights into the econometric estimations. After successfully developing the appropriate models that fits the various research questions, the STATA statistical software (v 14.1) was used to analyze the data. The outputs of the quantitative analysis are descriptive in nature with econometric computations where necessary. The qualitative data on the other hand was first transcribed and analyzed along themes, looking for similarities and differences in relation to the research questions. The output of the qualitative data is largely used to provide city specific contexts and explanations for why particular patterns are observed in the quantitative datasets. In article 1, the paper employs both quantitative and qualitative methods to analyze participation in agriculture among the different urban household types- that is urban agriculture based households, rural agriculture based households, urban and rural agriculture based households and non-agricultural households. Similar methods were used to analyze the cropping patterns, uses of own food produced, as well as food security and income shares accruing to households from different activities. In article 2, a combination of descriptive statistics and econometric analysis were employed to ascertain the effect of participation in agriculture on the household’s food security. In particular, the probit model was first employed to group households as food secure or otherwise. In the second phase, households were grouped along a continuum of food severity indicators (i.e. food secure, mildly food insecure, moderately food insecure and severely food insecure). This enabled us to fit an ordered probit model for the sample. The qualitative data was used to describe the research context of the paper, which allows the reader to interpret the results of the analysis in context. In the third paper which looks at food transfers receipts and its associated determinants, households were classified into food transfer recipients or otherwise. Assuming a normal standard distribution of the error term, a probit model was fixed to estimate the determinants of rural-urban and intra-urban (urban to urban) food transfer receipts. The results are presented mainly in the form of tables, graphs and charts to show observed patterns and differences in the indicators of interest. 8811 In the kappa however, the qualitative data is used to introduce the reader to the research background. Thus, setting the tone to enable the reader to have an appreciation of the settings in which the research was conducted. 5.6 Ethical considerations The ethics of research requires that confidentiality of respondents are respected at all times without identifying participants by name, gender or providing any lead that will help third parties to identify particular respondents. In the quantitative data, all responses were aggregated and analyzed at the city level. Similar considerations were made for the qualitative data. However, in the event of direct quotation from a particular interview, reference is made to the individual. These consents were sought during interviews to publish portions of the interviews as part of the research output. 5.7 Data quality and limitations of the study The survey instrument used for the study adapts several scales in measuring the food security status of sampled households. The benefit to this approach is that, one is able to capture a broad array of indicators. As in the case of this thesis, the approach enabled us to capture large number of households which would have been practically impossible with other forms of capturing food security data. Even though the measurement of food security is not limited to food access, the data is limited in its ability to measure nutritional security of urban households. In addition, the data used for the thesis is cross sectional. However, a longitudinal study would have provided rich insights in the food security situations of selected households over time. Nonetheless, this type of studies can be constrained by funds and time. In focusing on food transfers as part of multi-spatial livelihoods analysis, an important variable of interest would have been to see the quantities and frequencies of food transfers received by households. Lack of this important variable places a limitation on the kind of analysis that can be done. 82 82 Food security could be treated at the household level or at the intra-household level to test for levels of security within the same household. The data used for this thesis does not account for intra-household variables, making it impossible to address food security at that level. 8833 84 84 6. SYNTHESIS OF THE FINDINGS 6.1 Contextualizing Research Findings In this section, I revisit the research questions posed in Chapter 1. To start with, I choose to recap the research context to guide the discussions of the research findings. Ghana as a country has historical antecedents of rural-urban and north-south migration dating back to colonial days. This to some extent relates to the unequal distribution of infrastructure between the north and south as well as urban and rural areas. For example, most educational institutions were located in urban areas. As a result, all school going children who passed major promotion exams and had to continue their studies were compelled to relocate several kilometers away from home. Additionally, the lack of market access and market integration makes it convenient for other market intermediaries to travel to large cities to sell mostly agricultural produce and also purchase other commodities for use or resale in rural areas. As indicated earlier, food production in the study cities-Techiman and Tamale- have been a major source of food supply to households in those cities for decades. The farmlands were mostly located behind their homes and so were fairly cheaper (in terms of time and transportation cost) to engage in agriculture. Production was inspired purely by subsistence purposes and not commercial motives. However, population expansion has led to loss of agricultural lands for production. In the case of Ghana, and Techiman and Tamale in particular, population expansion has not been accompanied with corresponding expansion in economic growth and infrastructure development- economic growth has been slower relative to population growth. With farming as the mainstay of most displaced urban residents in the study area, some farmers prefer to move their farming practices outside the city boundary in order to have arable and adequate land for their production. 8855 In addition, the definition of urban and by extension city in Ghana relates more to population growth and is not necessarily based on infrastructure or economic development. Historically, Techiman has been a farming community but with its strategic location, it serves as a good converging point for both farmers and market intermediaries to do business as it links to several other cities. Over time, the population of Techiman has increased and become a municipal area. This notwithstanding, proximity to rural areas and access to land for farming purposes remain fairly easy compared to residents of Tamale who have to travel several kilometers to access land for agricultural purposes. Tamale on the other hand has a historical and cultural antecedent that encourages farming in and around houses. With increasing urbanization and migration of several thousands of people to the northern regional capital, the city continues to expand very fast and pushes agricultural lands farther away. This sometimes leads to temporary migration during farming seasons to distant rural areas for cultivation of rural lands, but return after the farming season. Historically and increasingly, rural-urban interactions have existed and continue to take new forms in the wake of urbanization and the urbanization of poverty. The research questions posed in the thesis provide an indication of how the different types of food production in different contexts can impart on household food security Following the contextual issues surrounding acquisition and use of land in Ghana and the study areas in particular, the sources of food constituting urban food baskets would be affected based on where land could be acquired for production. For instance, in Tamale where urban food production is considered as a cultural phenomenon, the practice of urban agriculture receives support from the metropolitan department of agriculture and other stakeholders including NGOs, academic community and the traditional authorities. As a result, access to essential foods such as green leafy vegetables with relatively shorter lifespan can be and continues to be produced and consumed in the city. This has implications on the producing households’ food security in terms of access and utilization. In contrast however where not much urban agriculture is practiced, and access to peri- urban and rural land is fairly easy, there is little incentive for farmers to invest their resources in urban farming. As a result, production of certain types of food including vegetables are limited in Techiman. In such contexts, the factors that explain the food security situations will differ across cities. 86 86 6.2 Contextual Characteristics of the Sample To place the articles in context, I present some contextual differences between Techiman and Tamale as found in the dataset. Overall, there are more households in Techiman who practice solely urban agriculture than in Tamale (26.3% and 15.0% respectively). The higher rate of participation in urban agriculture in Techiman is a result of the nature of the city. Techiman is still expanding and so even though most lands have been sold for residential purposes, both in the city and the peri-urban areas, they are yet to be developed. This provides opportunity for adjoining homes to grow crops (mostly maize) on such plots during the rainy season. Urban agriculture in Tamale on the other hand is intensive in nature and focuses on the production of vegetables for consumption and also the local market. The practice is more pronounced in the dry season where most farming activities are not possible due to climatic conditions. The proportion of farmers in Tamale who practice both urban and rural agriculture (8.0%) and, only rural agriculture (19.9%) are higher than found in Techiman (2.4% and 15.3% respectively). As the city expands, open spaces for agriculture purposes shrinks and pushes urban farmers outside the city to peri- urban and rural areas. The data also suggests that nearly a third (30%) of household heads in Techiman are females compared to only about 14% in Tamale with mean ages of 44.4 and 47.1 years respectively. This is slightly higher than the mean age of male headed households for the two cities (42.3 years and 46.2 years respectively). Giving the cultural and religious orientation in Ghana, the male spouse automatically assumes the headship role of the household. In the event of death or the absence of the male spouse however, the woman assumes that role. It is also common to find females as head of their households when they are not married or married but live in separate cities from their spouse. I argue that, the high proportion of female headed households in Techiman could be the result of the busy trading enterprise in food stuff which generally falls in the domain of females. Generally in Ghana, access to agricultural lands for food production is culturally biased in favour of males who often assume the headship roles of households. Assumption of these roles have implications for decision making on access and use of resources such as land which is an important factor of food production. All things being equal, cities with higher proportions of females as household heads potentially should have better access to productive resources such as land. 8877 Comparing the education levels across the cities show that nearly half (47.3%) of all household heads in Tamale have no formal education compared to about a quarter (22%) in Techiman. This notwithstanding, more than a third (35%) of households in Tamale have either secondary level education or higher compared to Techiman that has about 31%. Educational attainment generally is tied to increased formal employment and income earning opportunities. It is thus perceived that, persons with little or no education would most likely be involved in agricultural production. 88 88 6.3 Summary of Articles Article 1 addresses the first research question which seeks to understand the various sources and types of own food production arrangements available to urban households and how these sources contribute to the urban households’ food basket in small and medium sized cities in Ghana. The paper had an overarching objective to quantitatively measure and analytically assess ways in which urban agriculture could be placed in the wider context of own food production. Three main contributions are made to this objective. First, the paper quantifies the various sources of own produced food; discusses the food security contributions to the households and the income earning opportunities tied to agricultural production in small and medium sized cities. To do this, the sample is disaggregated into four categories of households (urban agriculture only, rural agriculture only, urban and rural agriculture only, and households with no agriculture) allowing for in-depth analysis of the contribution of the different types of agriculture engaged in by urban households to self- provisioning and food security. The results show that nearly half (43.4%) of surveyed respondents engaged in the production of their own food in the urban or rural areas or both, primarily for household consumption. Households who do not engage in agriculture raised their incomes through engagement in informal businesses. Engagement in agriculture by urban residents either on urban or rural plots are found to be a livelihood mechanism adopted to stabilize or improve the household’s food security. For households who engaged in both urban and rural production, cash income from agriculture was found to be their single most important source of income. The paper concludes that the contribution of agriculture to urban food security should not be narrowed to considering urban agriculture only but be placed in the wider context of the different food production arrangements including rural agriculture and the linkages that exist with other production spaces outside the city. Research question two is answered in article 2 which seeks to answer the question on the effects of own food production on household food security in small and medium sized cities in Ghana. Parts of research question 4 (determinants of food security along the urban household food (in) security continuum) are also analyzed and answered in this article. The aim of the article is to quantitatively examine the potential effect of urban households’ food production in urban and rural spaces on urban food security. The article applies econometric estimations to determine the effects of own food 8899 production on urban household food security. It begins by highlighting the research debate on the contribution of urban agriculture on food security and the corresponding oppositions in the literature. The research context and conceptual considerations relating to urban food production (i.e. urban agriculture, food production in rural areas by urban residents) are also discussed. To position the findings in relation to appropriate literature, a review of the determinants of food security is done which guides the selection of variables into the model. In the analytical framework and following on household categorization in article 1, households are classified by the type of activity whether they engage in urban agriculture, rural agriculture or both, or no production at all. The effects of participation in own food production on food security are estimated. Two main types of analysis are conducted in this paper. First, households are classified as either food secure or otherwise using the Household Food Insecurity Access Prevalence (HFAP). At the second level, the household food security status which is calculated as a graduated continuum (food secure, mildly food insecure, moderately food insecure and severely food insecure) serves as the dependent variable which allows for calculation of the effect of agricultural production on food security. Analyzing food security by the type of household along the different food (in) security indicators brings a new perspective to the debate in the literature by helping to identify which particular type of agricultural production activity brings the most food security related benefits to urban households in the context of intermediate sized cities. Furthermore, the paper shows overwhelming evidence that own food production impacts on household food security. However, the findings, ought to be interpreted in context. Some important predictors of household food security include household and individual level characteristics such as number of years of schooling of mother(s), income, education and the marital status of household head. Female headed households were generally found to be less likely to be food secure. The main conclusion of the paper is that, while there is some merit in the research debate, urban agriculture or rural agriculture alone is not enough to guarantee food security among own food producers in urban areas. Rather, the evidence suggests that households that participate in food production in both urban and rural areas have better food security outlook relative to the other household categories, all things being equal. 90 90 Additionally, the paper demonstrates that city context is important in the debate on impact of urban agriculture on household food security. For example, whereas UA has no effects on household food security in Techiman, the practice shows strong effect on household food security in Tamale. It concludes by recommending that policies and discussion aimed at addressing urban poverty through agriculture should be placed in city context. Also, policies related to urban and rural agriculture should be complementary to another as the paper demonstrates a clear link between engagement in urban and rural agriculture by urban residents. Following the estimations in article 2, article 3 responds to the third and fourth research questions by contributing to the discussions on an emerging body of knowledge in the literature -multi-spatial livelihoods. The aim of the article was to analyze the determinants of food transfer receipts (both rural and urban) among receiving households and how transfers are connected to own production of food whether in rural or urban areas. The article links multi- spatiality in relation to rural agriculture by urban residents with rural-urban food transfer and intra-urban food transfer phenomena which have been treated in isolation in previous literature. Applying econometric techniques, the article hypothesizes that, participation in agriculture reduces the likelihood of receiving food transfers. The results suggest that participation in agriculture does not influence the likelihood of a household receiving food transfer, at least for the total sample. However, presence of working age males decreases the chances of food transfer receipt. Female headed households, ageing household heads and household cash remittance expenditure were all found to increase the likelihood of household food transfer receipt. City specific effects suggest that participation in urban agriculture, age of household head, being a female headed household (sex), attainment of higher education by household head and spouse as well as household remittance expenditure jointly determine food transfer receipts in Techiman. In Tamale on the other hand, food transfer receipt is determined by age of household head and gender (females). The determinants along the types of food transfer receipts are also estimated. The paper concludes by applying qualitative insights in providing explanations about the causes of the identified determinants but questions the non-existence of social insurance policy for the aged and the limitations in property rights of female heads in Ghana. These aspects, the paper argues, makes the aged and female headed households vulnerable and raises food security concerns, impinging on the basic human right of all persons to food. 9911 92 92 7. CONCLUSIONS Insofar as unequal development continues to exist in developing countries such as Ghana, movement of people in search of better living conditions will persist. The increasing rate of urbanization brings several benefits to urban dwellers but at the same time may pose as a threat to food and livelihood security especially for the urban poor. Farm households whose livelihoods are lost to infrastructure development have a task to adopt new livelihood strategies including farming open spaces. However, the lack of enforcement of laws governing the practice of urban agriculture places the tenure security in question and dampens interest in investing in such lands. Increasingly, households with rural connections prefer to move away from farming in and around the city to neighboring rural areas as part of the urban livelihood strategy. Maintenance of kinship ties with rural relatives serves as insurance mechanisms that could be relied on as a buffer in the events of failed entitlements in urban areas. The concept where cities are defined based on population figures places expectations on both traditional and city authorities as to what an ideal city should look like. In the process, critical sectors of the economy such as agriculture that relate to the livelihood needs of the citizenry are abandoned in exchange for merchandize where cheap imported commodities are sold. This raises several food and nutrition security concerns and livelihood threats for the urban poor. The entry point of this thesis is whether producing food in urban area or in the rural area can have any role in improving the food security needs of the households in question. Even though the thesis demonstrates the important contributions of own food production on food and livelihood security among urban households, I submit that, the potential effects could be higher and impact greater depending on the lenses with which stakeholders, especially city authorities and owners of land see urban and rural agriculture and how these relate to food and nutrition security. At present, there is very little interaction and lack of extension support services to urban farmers. In Techiman for example, where city authorities are still planning their ‘ideal’ city, agriculture 9933 in the city is considered alien to modern day city development and this notion of city development prevents appropriate policies and investments in such sectors which ultimately affect both urban farmers and urban residents of the potential economic and nutritional benefits associated with farming in the city. In returning to the debate in the literature, and especially the lack of adequate empirical evidence on the effects of urban agriculture on household food security, my conclusion after analyzing the data is that, such debates cannot be made without accounting for contextual factors that might be driving the results. I submit therefore that, as much as it will be a good policy directive to have a national food security policy, local authorities must be empowered to model their own food security policies that take into account the contextual factors such as socio-cultural and geographical variables that could best be explained from a local context. On the research debate on whether urban agriculture should be promoted as a poverty alleviation strategy for its nutritional and economic benefit or not, I submit based on evidence from this thesis that, over concentration on urban agriculture only presents part of the argument but not the whole. Expanding the scope to encompass rural food production opportunities of the households paints a brighter picture and also informs what kinds of policies needed in support of the practice. As alluded to earlier, the disproportional development and the lack of decent and basic infrastructure mostly in the countryside would always make living in urban areas more attractive than rural areas. This is primarily because, the urban centers are perceived to have better opportunities than the rural area. However, the uncertainties regarding access to food, which is a basic human right creates an incentive among some urban households to maintain strong reciprocal ties with kinship relations in rural areas. This serves as an insurance for urban households either by cultivating food on rural plots or relying on food transfer receipt from the rural area to supplement food purchases in the urban area. In connecting the conclusions to the conceptual underpinnings, there is the need to understand what type of resources are available to households in addressing their food security needs. As indicated in Chapter 4, household asset ownership helps to command resources, including production resources. What is seen in this study however is that, access to land is critical in making households food secure! In Tamale for instance where urban agriculture is a celebrated enterprise, household access to agricultural plot means an opportunity to cultivate high value crops (mainly vegetables) for the market 94 94 and own consumption. Similarly, access to agricultural land in peri-urban and rural areas, both in Techiman and Tamale, provides opportunity to households to produce grains and tubers which could last several months and make the household more food secure. Nonetheless, access to land for rural agriculture is not available to all, neither is urban agriculture, nor social relations that permit food transfers. For a household to be able to command land largely depends on their ability to make claims on their social networks. This is because, most open space lands in the cities belong to the state and it is the ability of the household to leverage and make claim on government that gives them access. Similarly, households’ rural networks, based on family ties or friendship gives them the opportunity to make claim to rural and community lands which are used for production, and thereby making them food secure. Contextually, apart from state lands, ownership of land is vested in the hands of chiefs in Tamale whereas ownership of land is vested in family heads and chiefs in Techiman. The question of who is able to appropriate the land for production relates to the household’s ability to make claims based on kinship ties. From a socio-cultural perspective, women in Tamale do not have access to agricultural lands. They however are able to cultivate portions of farmland belonging to their husband’s family. In the event that the husband dies, access to the agricultural land becomes problematic and highly dependent on the relationship that exist between the woman and her in-laws. In contrast, women in Techiman can make claims to agricultural lands, even on behalf of their spouses. With rapid urbanization comes the demand for land by private real estate companies, making land sales a lucrative venture in recent times. In the event that agricultural lands have to make way for city expansion, the family head usually demarcates portions of the land to all members of the family (mainly the original members) either to keep for housing or sell to raise capital to start a new line of business. In some respect, there is equality in accessing urban and rural lands in Techiman than Tamale. Can urban agriculture be expected to feed the cities? This question has often come up in my discussion on urban agriculture and its potential benefits to food security in the household and the cities in general. Based on empirical evidence presented so far in this thesis, I submit that, the response to this question will vary across cities. Increasingly, as cities become cosmopolitan and income levels improve, dietary patterns also change in favor of healthier and nutritious meals. This increases the demand for vegetables which are mostly supplied to urban consumers through urban agriculture. This trend however is weakened 9955 when the rural areas are easily accessible for food production. My response to the question of whether urban agriculture can feed the cities therefore relates to the kind of systems including legislations, infrastructure and support systems local and city authorities put in place to produce food in the city. With this in place, urban agriculture, especially in Tamale could potentially feed the city with the requisite vegetables needed for a healthy life. The same can however not be said for Techiman because of the general lack of appreciation of farming in the city by the stakeholders including the traditional institutions headed by the chiefs who also double as owners of land. Planting to feed the city therefore is connected to the interplay that exist in urban food provisioning systems and how to they various production and non- production arrangements can be harnessed for urban food security. This will require the involvement of all the relevant stakeholders in the formulation of appropriate city level legislations and enforcement. In addition, urban based policies relating to food security must account for the increasing multi-spatial activities of urban residents, including farming in rural areas and rural-urban food transfers and how they are connected to food security. In conclusion, this thesis makes a unique contribution to the discussion on urban food security by expanding the scope from the concentration on urban agriculture and its contribution to household food security to include agricultural productions opportunities available to urban households and how these are appropriated for food. In addition, the focus on small and intermediate sized cities brings a new perspective that departs from the urban bias that has characterized urban agriculture studies in Ghana. The research also fills in an important gap that relates to the lack of empirical data in the policy debate. Connecting urban households to rural livelihoods also brings a new perspective on the nature of urban-rural interactions and how these multi-locality engagements contribute to the urban food security discussions. Another important contribution of the thesis is the ability to disaggregate urban households by the type of agricultural engagements and how these specific types of activities contribute to household food security. 96 96 References Abdulai, R. T. (2006). Is land title registration the answer to insecure and uncertain property rights in sub-Saharan Africa? In): RICS (The Royal Institution of Chartered Surveyors). Abdulai, R. T., & Antwi, A. (2005). Traditional landholding institutions and individual ownership of land rights in sub-Saharan Africa. World Review of Science, Technology and Sustainable Development, 2(3-4), 302-319. Abdulai, R. T., & Ndekugri, I. E. (2007). Customary landholding institutions and housing development in urban centers of Ghana: Case Studies of Kumasi and Wa. Habitat international, 31(2), 257-267. Adepoju, A. (2005). African families in the twenty-first century: Prospects and challenges: iUniverse. Al-Hassan, R., Famiyeh, J., & De Jager, A. (1997). Farm household strategies for food security in Northern Ghana: a comparative analysis of high and low population farming systems. In Sustainable food security in West Africa, (pp. 129-152): Springer. Andersson, A. (2002). The Bright Light Grow Fainter: Livelihoods, Migration and a Small Town in Zimbabwe. Stockholm University, Sweden. Andersson Djurfeldt, A. (2012). Virtuous and Vicious Cycles in Rural-Urban Linkages: Cases from Zimbabwe. Africa Review, 4(2), 25-51. Andersson Djurfeldt, A. (2015a). Multi‐Local Livelihoods and Food Security in Rural Africa. Journal of International Development, 27(4), 528-545. Andersson Djurfeldt, A. (2015b). Urbanization and linkages to smallholder farming in sub-Saharan Africa: Implications for food security. Global Food Security, 4, 1- 7. Anosike, V., & Fasona, M. (2004). Gender dimensions of urban commercial farming in Lagos, Nigeria. Urban Agriculture Magazine, 12, 27-28. Anyane, S. L. (1963). Vegetable gardening in Accra. The Ghana Farmer, 1(6), 228- 230. Arku, G., Mkandawire, P., Aguda, N., & Kuuire, V. (2012). Africa's quest for food security: what is the role of urban agriculture? Armar-Klemesu, M. (2000). Urban Agriculture and Food Security, Nutrition and Health. In D. M. Bakker N, Gundel S, Sabel-Koschella U, de Zeeuw H (Ed.), 9977 Growing Cities, Growing Food, Urban Agriculture on the Policy Agenda). Feldafing, Germany: DSE Feldafing. Asomani-Boateng, R. (2002). Urban cultivation in Accra: an examination of the nature, practices, problems, potentials and urban planning implications. Habitat international, 26(4), 591-607. Ayerakwa, H. M. (2017). Urban households' engagement in agriculture: implications for household food security in Ghana's medium sized cities. Geographical Research. Ayerakwa, H. M., Egyir, I. S., & Asuming-Brempong, S. (2014). Consumer Health Concerns and Willingness-to-Pay Premium Prices for Organic Vegetables in Accra. In F. A. Asante, A. Y. Owusu & C. Ahiadeke (Eds.), Placing health at the centre of development.). Accra: Institute of Statistical, Social and Economic Research (ISSER), University of Ghana. Ba Diao, M. (2004). Women and periurban agriculture in the Niayes zone of Senegal. Revista de Agricultura Urbana, 12, 23-24. Badami, M. G., & Ramankutty, N. (2015). Urban agriculture and food security: A critique based on an assessment of urban land constraints. Global Food Security, 4, 8-15. Barrett, C. B. (2010). Measuring food insecurity. Science, 327(5967), 825-828. Bryceson, D. F. (2006). Ganyu casual labour, famine and HIV/AIDS in rural Malawi: causality and casualty. The Journal of Modern African Studies, 44(02), 173-202. Budlender, D. (2003). The Debate about Household Headship. Social Dynamics, 29(2), 48-72. Bukusuba, J., Kikafunda, J. K., & Whitehead, R. G. (2007). Food security status in households of people living with HIV/AIDS (PLWHA) in a Ugandan urban setting. British Journal of Nutrition, 98(01), 211-217. Burnell, P. (1998). Britain's new government, new White Paper, new aid? Eliminating world poverty: a challenge for the 21st century. Third World Quarterly, 19(4), 787-802. Carney, D. (1998). Sustainable rural livelihoods: What contribution can we make? : Department for International Development London. Chagomoka, T., Drescher, A., Glaser, R., Marschner, B., Schlesinger, J., & Nyandoro, G. (2015). Vegetable production, consumption and its contribution to diets along the urban–rural continuum in Northern Ghana. African Journal of Food, Agriculture, Nutrition and Development, 15(4), 10352-10367. Chagomoka, T., Unger, S., Drescher, A., Glaser, R., Marschner, B., & Schlesinger, J. (2016). Food coping strategies in northern Ghana. A socio-spatial analysis along the urban–rural continuum. Agriculture & Food Security, 5(4), 1-18. Chambers, R. (1995). Poverty and livelihoods: whose reality counts? Environment and Urbanization, 7(1), 173-204. 98 98 Cofie, O., & Drechsel, P. (2007). Water for food in the cities: the growing paradigm of irrigated (peri)-urban agriculture and its struggle in sub-Saharan Africa. African Water Journal, 1(1), 26-50. Commission, N. D. P. (2014). Medium-term national development policy framework: Ghana shared growth and development agenda (GSGDA) II, 2014-2017. Crush, J. (2013). Linking food security, migration and development. International Migration, 51(5), 61-75. Crush, J., & Caesar, M. (2017). Food remittances: Migration and food security in Africa: Southern African Migration Programme. Crush, J., Hovorka, A., & Tevera, D. (2011). Food security in southern African cities: the place of urban agriculture. Progress in Development Studies, 11(4), 285-305. Daie, G. F. (2014). Ethno-culture Disparity in Food Insecurity Status: The Case of Bullen District, Benishangul-gumuz Regional State, Ethiopia. African Journal of Food Science, 8(2), 54-63. Danso, G., Drechsel, P., Obuobie, E., Forkuor, G., & Kranjac-Berisavljevic, G. (2014). Urban vegetable farming sites, crops and cropping practices. In P. a. K. Drechsel, B. (Ed.), Irrigated urban vegetable production in Ghana: characteristics, benefits and risk mitigation, (pp. 7-27). Colombo, Sri Lanka: International Water Management Institute. Danso, G., Drechsel, P., Wiafe-Antwi, T., & Gyiele, L. (2002). Income of farming systems around Kumasi, Ghana. In Urban Agriculture Magazine, vol. 7 (pp. 5- 6). Davies, S. (2016). Adaptable livelihoods: Coping with food insecurity in the Malian Sahel: Springer. De Haan, A. (1999). Livelihoods and poverty: The role of migration‐a critical review of the migration literature. The journal of development studies, 36(2), 1-47. De Haan, L., & Zoomers, A. (2005). Exploring the frontier of livelihoods research. Development and change, 36(1), 27-47. Development, I. I. f. S., Titi, V., & Singh, N. C. (1994). Adaptive strategies of the poor in arid and semi-arid lands: In search of sustainable livelihoods: International Institute for Sustainable Development. Devereux, S. (1999). 'Making less last longer': informal safety nets in Malawi. Dick, E., & Schmidt-Kallert, E. (2011). Understanding the (Mega-) Urban from the Rural: Non-Permanent Migration and Multi-Locational Households. disP-The Planning Review, 47(187), 24-36. Dima, S., & Ogunmokun, A. (2004). Agricultura urbana y periurbana en Namibia. Revista agricultura urbana, 44-46. Douglass, M., & Friedmann, J. (1998). Cities for citizens: Planning and the rise of civil society in a global age: John Wiley & Sons New York. 9999 Drechsel, P., Adam-Bradford, A., & Raschid-Sally, L. (2014). Irrigated urban vegetable production in Ghana: A farming system between challenges and resilience. In P. Drechsel, Keraita, B. (Ed.), Irrigated urban vegetable production in Ghana: characteristics, benefits and risk mitigation, vol. 2 (pp. 247). Colombo, Sri Lanka: International Water Management Institute (IWMI). Drechsel, P., Graefe, S., Sonou, M., & Cofie, O. O. (2006). Informal irrigation in urban West Africa: An overview (Vol. 102): IWMI. Dreze, J., & Sen, A. (1989). Hunger and public action: Oxford University Press on Demand. Dubbeling, M., Zeeuw, H. d., & Veenhuizen, R. v. (2010). Cities, poverty and food: multi-stakeholder policy and planning in urban agriculture: Practical Action Publishing. Effah-Abedi, T. (2014). Smallholder farmers and the question of food security in the Asante Akim North Municipality of Ghana. Eyben, R., & Ferguson, C. (2000). Realizing human rights for poor people: Strategies for achieving the international development targets. Target Strategy Paper. London: Department for International Aid. FAO. (1996). World Food Summit. Food and Agriculture Organization, Rome. FAO. (2012). Growing greener cities in Africa: first status report on urban and peri- urban horticulture in Africa. Rome: FAO. Faridi, R., & Wadood, S. N. (2010). An econometric assessment of household food security in Bangladesh. The Bangladesh Development Studies, 97-111. Fine, B. (1997). Entitlement failure? Development and change, 28(4), 617-647. Foeken, D., & Owuor, S. O. (2001). Multi-spatial livelihoods in sub-Saharan Africa: Rural farming by urban households–The case of Nakuru, Town, Kenya. Mobile Africa: Changing patterns of movement in Africa and beyond, 125-140. Foeken, D. W., & Owuor, S. O. (2008). Farming as a livelihood source for the urban poor of Nakuru, Kenya. Geoforum, 39(6), 1978-1990. Frayne, B. (2004a). Migration and urban survival strategies in Windhoek, Namibia. Geoforum, 35, 489-505. Frayne, B. (2004b). Migration and urban survival strategies in Windhoek, Namibia. Geoforum, 35(4), 489-505. Frayne, B. (2010). Pathways of food: mobility and food transfers in Southern African cities. International Development Planning Review, 32(3-4), 291-310. Frayne, B., McCordic, C., & Shilomboleni, H. (2014). Growing Out of Poverty: Does Urban Agriculture Contribute to Household Food Security in Southern African Cities? Urban Forum, 25(2), 177-189. Gbireh, Z. (1999). Development of urban and peri-urban irrigation for food security within and around Accra Metropolitan Area. FAO Technical Report.(unpublished). 100 100 GSS. (2014). 2010 Population and Housing Census: District Analytical Report, Techiman. Ghana Statistical Services, Accra, Ghana. Guyer, J. I. (1981). Household and community in African studies. African studies review, 24(2/3), 87-137. Hovorka, A., & Lee-Smith, D. (2006). Gendering the urban agriculture agenda. 2006), Cities Farming for the Future–Urban Agriculture for Green and Productive Cities, RUAF Foundation, IDRC, Ottawa, and IIRR, Leusden, The Netherlands. ISSER. (2016). The State of the Ghanaian Economy Report in 2015. Institute of Statistical, Social and Economic Research, Accra, Ghana. Kasanga, R. (1988). Land tenure and the development dialogue. The myth concerning. Katepa-Kalala, P. (1997). Sustainable Livelihood approaches in operation: A gender perspective. UNDP International Working Group on Sustainable Livelihoods. Retrieved January, 20, 2011. Koc, M. (1999). For hunger-proof cities: Sustainable urban food systems: IDRC. Krüger, F. (1998). Taking advantage of rural assets as a coping strategy for the urban poor: The case of ruralurban interrelations in Botswana. Environment and Urbanization, 10(1), 119-134. Kutiwa, S., Boon, E., & Devuyst, D. (2010). Urban agriculture in low income households of Harare: an adaptive response to economic crisis. Journal of Human Ecology, 32(2), 85-96. Lee-Smith, D. (2010). Cities feeding people: an update on urban agriculture in equatorial Africa. Environment and Urbanization, 22(2), 483-499. Lee-Smith, D. (2013). Which way for UPA in Africa? City, 17(1), 69-84. Lerner, A. M., & Eakin, H. (2011). An obsolete dichotomy? Rethinking the rural– urban interface in terms of food security and production in the global south. The Geographical Journal, 177(4), 311-320. Li, T. M. (2007). The will to improve: Governmentality, development, and the practice of politics: Duke University Press. Lynch, K., Maconachie, R., Binns, T., Tengbe, P., & Bangura, K. (2013). Meeting the urban challenge? Urban agriculture and food security in post-conflict Freetown, Sierra Leone. Applied Geography, 36, 31-39. Makoka, D. (2005). The Emergence of supermarkets in Malawi: implications for agrifood markets and the small farmer. Agricultural and Rural Development in Malawi: Macro and Micro Perspectives, AR Series (11), 121-139. Mascarenhas, O. (1999). Kigogo and Hananasif in Dar-es-Salaam, Tanzania: Gender Aspects of Urbanisation and Natural Resource Management. Maxwell, D. (2001). The importance of urban agriculture to food and nutrition. Annotated biography, ETC-RUAF. In): CTA publishers, Leusden, Netherlands. Maxwell, D., Levin, C., & Csete, J. (1998). Does urban agriculture help prevent malnutrition? Evidence from Kampala. Food policy, 23(5), 411-424. 110011 Maxwell, D. G. (1995). Alternative food security strategy: A household analysis of urban agriculture in Kampala. World Development, 23(10), 1669-1681. Maxwell, S., & Smith, M. (1992). Household food security: a conceptual review. Household Food Security: concepts, indicators, measurements. Edited by S. Maxwell and T. Frankenberger. Rome and New York: IFAD and UNICEF. Mbiba, B. (1995). Urban agriculture in Zimbabwe: implications for urban management and poverty: Avebury. Meikle, S., Ramasut, T., & Walker, J. (2001). Sustainable urban livelihoods: Concepts and implications for policy. Mikkelsen, B. (2005). Methods for development work and research: a new guide for practitioners: Sage. Mkwambisi, D. D., Fraser, E. D., & Dougill, A. J. (2011). Urban agriculture and poverty reduction: evaluating how food production in cities contributes to food security, employment and income in Malawi. Journal of International Development, 23(2), 181-203. MoFA. (2010). Medium Term Agriculture Development Plan (METASIP) 2011-2015. Ministry of Food and Agriculture Accra. MoFA. (2016). 2015 Agricultural Sector Annual Progress Report. Ministry of Food and Agriculture, Accra. Moser, C. O. (1996). Confronting Crisis. A Comparative Study of Household Responses to Poverty and Vulnerability in Four Poor Urban Communities. Environmentally Sustainable Development Studies and Monographs Series No. 8: ERIC. Moser, C. O. (1998). The asset vulnerability framework: reassessing urban poverty reduction strategies. World Development, 26(1), 1-19. Mougeot, L. (2000). Achieving urban food and nutrition security in developing countries: The hidden significance of urban agriculture. IFPRI, Brief paper (6). Mougeot, L. J. (2010). Agropolis:" The Social, Political and Environmental Dimensions of Urban Agriculture": Routledge. Mougeot, L. J. (2011). International Support to Research and Policy on Urban Agriculture (1996–2010): achievements and challenges. International Development Research Centre, Ottawa. Mudimu, G. D. (1997). Urban agricultural activities and women's strategies in sustaining family livelihoods in Harare, Zimbabwe. Singapore Journal of Tropical Geography, 17(2), 179-194. Mwangi, A. M., & Foeken, D. (1996). Urban agriculture, food security and nutrition in low income areas in Nairobi. African urban quarterly, 11(2/3), 170-179. Nabulo, G., Nasinyama, G., Lee-Smith, D., & Cole, D. (2004). Gender analysis of urban agriculture in Kampala, Uganda. Urban Agriculture Magazine, 12, 32-33. 102 102 Nugent, R. (2000). The impact of urban agriculture on the household and local economies. Bakker N., Dubbeling M., Gündel S., Sabel-Koshella U., de Zeeuw H. Growing cities, growing food. Urban agriculture on the policy agenda. Feldafing, Germany: Zentralstelle für Ernährung und Landwirtschaft (ZEL), 67- 95. Nyborg, I., & Haug, R. (1995). Measuring household food security: a participatory process approach. In Forum for Development Studies, vol. 22 (pp. 29-59): Taylor & Francis. Obosu-Mensah, K. (1999). Food production in urban areas: a study of urban agriculture in Accra, Ghana: Ashgate Publishing Ltd. Obuobie, E., Drechsel, P., Danso, G., & Raschid-Sally, L. (2004). Gender in open- space irrigated urban vegetable farming in Ghana. Urban Agriculture Magazine, 12(1), 13-15. Obuobie, E., & Hope, L. (2014). Characteristics of Urban Vegetable Farmers and Gender Issues. In P. Drechsel & B. Keraita (Eds.), Irrigated Urban Vegetable Production in Ghana: Characteristics, Benefits and Risk Mitigation, vol. 2). Colombo, Sri Lanka: International Water Management Institute. Obuobie, E., Keraita, B., Hope, L., & Agodzo, S. K. (2014). Health Risk Perceptions of Stakeholders in Irrigated Urban Vegetable Farming. In P. Drechsel & B. Keraita (Eds.), Irrigated Urban Vegetable Production in Ghana, vol. 2 (pp. 116). Sri Lanka: International Water Management Institute (IWMI). Oruwari, Y., & Jev, M. (2004). Integration of gender in municipal policies in Port Harcourt, Nigeria. Urban Agriculture Magazine, 12, 29. Padgham, J., Jabbour, J., & Dietrich, K. (2015). Managing change and building resilience: A multi-stressor analysis of urban and peri-urban agriculture in Africa and Asia. Urban Climate, 12, 183-204. Potts, D. H. (2010). Circular migration in Zimbabwe & contemporary sub-Saharan Africa: Boydell & Brewer. Probst, L., Houedjofonon, E., Ayerakwa, H. M., & Haas, R. (2012). Will they buy it? The potential for marketing organic vegetables in the food vending sector to strengthen vegetable safety: A choice experiment study in three West African cities. Food policy, 37(3), 296-308. Putnam, R. D., Leonardi, R., & Nanetti, R. Y. (1994). Making democracy work: Civic traditions in modern Italy: Princeton University press. Rakodi, C. (1997). The urban challenge in Africa: Growth and management of its large cities: United Nations Univ. Rakodi, C. (1998). Review of the poverty relevance of the peri-urban interface production system research. Report for the DFID Natural Resources Systems Research Programme, 2. Rakodi, C. (2002). A livelihoods approach–conceptual issues and definitions. Urban livelihoods: A people-centred approach to reducing poverty, 3-22. 110033 Redwood, M. (2012). Agriculture in urban planning: generating livelihoods and food security: Routledge. Riley, L., & Legwegoh, A. (2014). Comparative urban food geographies in Blantyre and Gaborone. African Geographical Review, 33(1), 52-66. Salih, S. A. (1995). Food security in Africa: Concepts, measurement, policy and reality: UNU/WIDER. Sarpong, D. B., & Asuming-Brempong, S. (2004). Responding to economic shocks in Ghana: the agricultural sector as a social safety net. Electronic Journal of Agricultural and Development Economics, 1(1), 117-137. Satterthwaite, D. (1997). Sustainable cities or cities that contribute to sustainable development? Urban Studies, 34(10), 1667-1691. Satterthwaite, D., McGranahan, G., & Tacoli, C. (2010). Urbanization and its implications for food and farming. Philosophical Transactions of the Royal Society B: Biological Sciences, 365(1554), 2809-2820. Sen, A. (1981). Ingredients of famine analysis: availability and entitlements. The quarterly journal of economics, 433-464. Sen, A. (1992). Inequality reexamined: Clarendon Press. Sen, A., & Dreze, J. (1999). The Amartya Sen and Jean Dreze Omnibus:(comprising) poverty and famines; Hunger and public action; and India: Economic development and social opportunity. OUP Catalogue. Smart, J., Nel, E., & Binns, T. (2015). Economic crisis and food security in Africa: Exploring the significance of urban agriculture in Zambia’s Copperbelt province. Geoforum, 65, 37-45. Smart, N. (2015). The science of religion and the sociology of knowledge: Some methodological questions: Princeton University Press. Smit, J., & Nasr, J. (1992). Urban agriculture for sustainable cities: using wastes and idle land and water bodies as resources. Environment and Urbanization, 4(2), 141-152. Smit, J., Nasr, J., & Ratta, A. (1996). Urban agriculture: food, jobs and sustainable cities. New York, USA, 2, 35-37. Start, D. (2001). The Rise and Fall of the Rural Non‐farm Economy: Poverty Impacts and Policy Options. Development policy review, 19(4), 491-505. Stewart, R., Korth, M., Langer, L., Rafferty, S., Da Silva, N., & van Rooyen, C. (2013). What are the impacts of urban agriculture programs on food security in low and middle–income countries? Environmental Evidence, 2(7), 1-13. Swift, P. (2006). Livelihoods in the Srepok River Basin in Cambodia: A baseline survey. In The NGO Forum on Cambodia, Phnom Penh, vol. 3). Tacoli, C. (1998a). Rural-urban interactions: a guide to the literature. Environment and Urbanization, 10(1), 147-166. 104 104 Tacoli, C. (1998b). Rural-urban interactions; a guide to the literature. Environment and Urbanization, 10, 147-166. Tacoli, C. (2002). Changing rural-urban interactions in the Sub-Saharan Africa and their impact on livelihoods: a summary (Vol. 4): IIED. Tacoli, C., McGranahan, G., & Satterthwaite, D. (2008). Urbanization, poverty and inequity: Is rural-urban migration a poverty problem or part of the solution. The new global frontier: Urbanization, poverty and environment in the 21st century, 43. Tawodzera, G. (2010). Vulnerability and resilience in crisis: Urban household food insecurity in harare, Zimbabwe. University of Cape Town. UNDP. (1996). Urban Agriculture. Food, Jobs, and Sustainable Cities. Series for Habitat, United Nations Development Programme, New York. Van Veenhuizen, R., & Danso, G. (2007). Profitability and sustainability of urban and periurban agriculture (Vol. 19): Food & Agriculture Org. Von Braun, J. (1992). Improving food security of the poor: Concept, policy, and programs: Intl Food Policy Res Inst. Waal, A. d. (1989). Famine that kills: Darfur, Sudan, 1984-85: Clarendon press. Walker, J., & Meikle, S. (1999). The changing face of the Chinese city. in 2G: International Architecture Review, 10(II). Warren, E., Hawkesworth, S., & Knai, C. (2015). Investigating the association between urban agriculture and food security, dietary diversity, and nutritional status: A systematic literature review. Food policy, 53, 54-66. Werna, E. (1998). Urban management, the provision of public services and intra-urban differentials in Nairobi. Habitat international, 22(1), 15-26. Wratten, E. (1995). Conceptualizing urban poverty. Environment and Urbanization, 7(1), 11-38. Yaro, J. A. (2004). Theorizing food insecurity: building a livelihood vulnerability framework for researching food insecurity. Norsk Geografisk Tidsskrift- Norwegian Journal of Geography, 58(1), 23-37. Yaro, J. A. (2006). Is deagrarianisation real? A study of livelihood activities in rural northern Ghana. The Journal of Modern African Studies, 44(01), 125-156. Zezza, A., & Tasciotti, L. (2010). Urban agriculture, poverty, and food security: Empirical evidence from a sample of developing countries. Food policy, 35(4), 265-273. 110055 Paper I Urban households’ engagement in agriculture: implications for household food security in Ghana’s medium sized cities Hayford Mensah Ayerakwa1,2 1Department of Human Geography, Lund University, Lund, Sweden 2Department of Agricultural Economics and Agribusiness, University of Ghana, Accra, Ghana Email: hayford_mensah.ayerakwa@keg.lu.se Received 1 February 2016 • Revised 15 June 2016 • Accepted 7 July 2016 Abstract Urban and peri-urban agriculture plays an important role in meeting the food and nutrition needs of those living in the increasingly urbanised cities in Africa. The extent and scope of the practise of urban and peri-urban agriculture differs from one city to another depending on the economic, environmental, socio-political, and contextual conditions at play. Using household data drawn from urban Techiman and Tamale in the Brong Ahafo and Northern regions of Ghana respec- tively, this paper descriptively analyses the contribution of households’ engagement in agriculture on urban households’ food security. The results show that, nearly half (43%) of urban residents are involved in the production of food either in the urban or rural areas or both, primarily for household consumption and sale of surplus pro- duce. Households who do not engage in agriculture raise their cash incomes through engagement in informal businesses. The picture is however different for households that engage in urban and rural agriculture. For such households, income raised from agriculture (rural and urban combined) is the highest in both cities contributing nearly half (43%) and about a third (33%) of total cash incomes in Techiman and Ta- male respectively. The results underscore the need to place the discussion on the contribution of urban agriculture to urban food security in the broader context of the different food production arrangements available to urban households, both in urban and rural areas. Keywords food security; urban agriculture; rural agriculture; Ghana; medium sized city; Techiman; Tamale Introduction The nutritional health and economic benefits ac- cruing to households practising urban agriculture Proponents of urban agriculture argue that it pre- has received considerable attention in the literature sents an innovative way of optimising access, (Armar-Klemesu, 2000; Badami & Ramankutty, quantity, and quality of food for the urban poor 2015; Chagomoka et al., 2015; 2016; Cofie & (FAO, 2012; Koscica, 2014; Mkwambisi et al., Drechsel, 2007; Maxwell et al., 1998; 2000; 2011). Increasing urbanisation in sub-Saharan Mougeot, 2011) although the actual number of Africa brings additional challenges for how to feed people engaged in the practice globally remains the growing population in a sustainable way and as debatable (Crush et al., 2011; Frayne et al., 2014; such makes the discussion on urban agriculture Lee-Smith, 2013; Stewart et al., 2013; Zezza & evenmore timely (Gupta&Gangopadhyay, 2013). Tasciotti, 2010). This issue notwithstanding there Geographical Research • 2017 1 doi:10.1111/1745-5871.12205 2 Geographical Research • 2017 remains a knowledge gap as to the importance of multiple sources: urban agriculture, rural agricul- urban agriculture across different types of urban ture (where households may farm areas of land settings. Its role differs from one city to another outside the city), food transfers from relatives and depending on the economic and contextual condi- friends, and market purchases. This paper provides tions at play. Thus, initiatives or policies aimed at quantitative data to unravel the contribution of addressing urban poverty through urban agricul- urban agriculture in medium sized cities, while ture need to be city specific. placing these in the broader spatial and social con- In this paper urban agriculture is contextualised text of other self-produced food and food transfers. in relation to households’ own food provisioning While some quantitative case studies exist these in two medium-sized cities in Ghana: Techiman are piecemeal and do not cover small and medium and Tamale in the Brong Ahafo and Northern re- sized cities (for example, Ellis & Sumberg, 1998 gions, respectively. The aim is to quantitatively and van Veenhuizen, 2006). measure and analytically assess the ways in which Fulfilling the paper’s objectives first requires urban agriculture is placed in the wider context of clarification of the concepts of urban agriculture market-procured and other self-produced, non- and food and nutrition security, and consideration market sources of food. This perspective has not of how they link together. Urban agriculture is been systematically generated from previous re- growing crops and/or raising livestock in an urban search (but see Andersson, 2002). Thus, this paper space for one’s own consumption and/or sale to addresses the knowledge gap about the actual and other city dwellers. Urban agriculture and backyard potential contributions of urban agriculture to food gardening may be a response to inadequate or security by considering emerging cities in Ghana, costly food (crops and livestock) in contexts of in- and the role that own food provisioning and the creased urban food demand because of population market economy play in helping people to meet expansion (Cofie et al., 2003). Urban agriculture the food and nutrition needs of households. serves locally grown, fresh fruits and vegetables to over 22 million people in Africa’s cities, playing Research objectives, rationale, and conceptual an important role in food supply and income op- background portunities according to the Food and Agriculture Organization’s (FAO) first status report on urban Urbanisation has grown rapidly in Ghana in recent and peri-urban horticulture (FAO, 2012). years, with Techiman and Tamale identified as be- Urban agriculture can potentially contribute to ing among the fastest growing cities in the country the food requirements of urban households while (GSS, 2013). As urbanisation continues to pro- enhancing access to cash incomes from the sale ceed, access to safe and nutritious food to feed of surplus food produced (Warren et al., 2015), the growing urban population continue to receive thereby increasing dietary diversity and food secu- attention from all stakeholders. The capacity of ur- rity This process allows producers to reduce their ban agriculture to meet these challenges is a key is- reliance on purchased food (Badami & sue for the government and population. Ramankutty, 2015) and, when surplus is sold, po- This paper makes three contributions: to quan- tentially it serves as a source of income that can tify and analyse various sources of self-produced help meet other pressing household needs and purchased food in small and medium sized cit- (Armar-Klemesu, 2000; Zezza & Tasciotti, ies; to discuss the food security contribution of all 2010). It can also allay food access problems dur- of these sources in the cities in question; and to dis- ing periods of price spikes or seasonal shocks cuss the income earning opportunities tied to agri- (Warren et al., 2015). The practise of growing cultural production in the cities and surrounding food crops and livestock has persisted over time rural areas and ask how such opportunities relate as part of the urban landscape (McClintock, to food and nutrition security. Most empirical re- 2010; Steel, 2008), enhancing more ecologically search performed in Ghana in the past has focused sound and resilient urban food systems (Viljoen, on the largest cities of Accra and Kumasi, with lit- 2005). Estimates of the extent of impact and the tle or no work in small and medium cities. Yet as contribution of urban agriculture to household Ghana urbanises, what is happening in these cities food security have received criticism because there becomes of greater interest, not least because they is a lack of uniform continent-wide empirical data represent the places where rural-urban transforma- on the extent and scope of urban agriculture. An tions would appear to be more clearly identifiable. exception is the work of Zezza and Tasciotti Householders in small and medium sized cities in (2010) who, after analysing nationally representa- sub-Saharan Africa potentially derive food from tive data from 15 countries, conclude that, with © 2017 Institute of Australian Geographers H.M. Ayerakwa, Urban household food security in Ghana 3 the exception of Ghana, Madagascar, Malawi and agriculture crops in Tamale and Techiman. Amoah Nigeria, participation in urban agriculture has been et al. (2007), for example, conclude that about 95 generally overestimated and that urban agriculture per cent of all lettuce consumed in Accra and is not necessarily practised by the very poor. The Kumasi is produced from urban agriculture activi- study further concludes that urban residents’ par- ties. This kind of production also contributes to ticipation in crop and livestock agriculture in employment, livelihoods and poverty alleviation Ghana in 1998 was 41 per cent. The shares of in- (MoFA, 2010). Indeed, the Ministry of Food and come accruing to urban residents from agriculture Agriculture projects a 20 per cent increase in the was 18 per cent, but when only households en- output obtained from urban and peri-urban areas gaged in urban agriculture are taken into consider- across the country from 2011 to 2015. ation, average shares of income from agriculture These considerations are important in light of are more substantial (44%). the fact that urban and rural populations are Food and nutrition security, sometimes known increasingly connected to one another. Cash remit- by its briefer variation food security, is the situa- tances from urban to rural households have often tion where ‘all people, at all times, have physical been described as vital to address the problem of and economic access to sufficient, safe and nutri- poverty in rural areas. Little or no attention has tious food that meets their dietary needs and food been given to social linkages; to reverse rural- preferences for an active and healthy life’ (FAO, urban flow of resources (Moorsom, 1995; Potts 2012). This definition emphasises that food & Mutambirwa, 1998); or to establish why—in security is not just about the physical availability spite of increasing unemployment—migration to of food, but includes social and economic access, urban areas persists. In Ghana, very little is known food safety, and the nutritional utilisation of food about how rural–urban linkages affect much- for a healthy life. needed food and nutrition requirements through Urban agriculture can have a positive impact on food transfers and, in some cases, rural farming household food security as it generates direct in- to urban residents. What is important, argues come for the household concerned, and provides Frayne (2005), is the socio-economic relationship direct access to the food produced. Especially in between the rural and urban components of the situations where markets are inefficient, urban same household. It is noteworthy that food trans- agriculture households may have access to a wide fers from rural areas are not restricted to migrants variety of relatively inexpensive nutritious foods, but form an important part of the urban food secu- such as vegetables and products of animal origin rity situation (de Haan & Zoomers, 2003; Kaag (milk, eggs, meat, and so on). Yet, gaining these et al., 2004). This point underscores the fact that benefits can be challenging. Farmers engaged in rural and urban areas should not be viewed as mu- urban agriculture in Ghana, like those in most de- tually exclusive entities (Foeken & Owuor, 2008). veloping countries, are faced with several chal- As will be shown in this paper, there is a dynamic lenges including but not limited to input costs, interaction between the two populations that im- theft, and lack of extension support services pacts on the food security needs of each other. (Hillbruner & Egan, 2008). Other challenges con- cern the use of land and the legal issues informing Methodology the practise of urban agriculture. The government of Ghana, unlike its counterparts in many other This paper forms part of a broader collaborative African countries, has formulated a policy on research project between Lund University in urban agriculture. This policy is contained in the Sweden, the University of Ghana, and University Medium Term Agriculture Development Plan of Nairobi, which is being carried out in Ghana 2011–2015 of the Ministry of Food and Agricul- and Kenya to understand the social, economic, ture (MoFA). In the Plan, urban and peri-urban ag- and environmental challenges and prospects of ur- riculture has been acknowledged as a major ban agriculture under changing global and demo- contributor to national food security. The Plan graphic realities. In Ghana, according to the further acknowledges the practise of urban and Ghana Living Standards Survey Six, rural and ur- peri-urban agriculture to be a year-round activity ban poverty has fallen rapidly to an urban poverty that takes place in open spaces, on government re- level of around 11 per cent as at 2013 (GSS, 2014; served lands, and on non-habitable areas in and World Bank, 2012). Techiman municipality and around high tension poles, and contributes to the Tamale metropolitan assemblies in the Brong supply of vegetables to city dwellers which also Ahafo andNorthern regions respectively were pur- constitutes one of the many important urban posively selected for the study in Ghana, breaking © 2017 Institute of Australian Geographers 4 Geographical Research • 2017 away from the metropolitan bias that has densely populated communities, where two house- characterised earlier studies on urban agriculture. holds out of three were selected. The choice of Techiman and Tamale for this An important implication of this sampling study provides particular insight into the method is that the data identify the location of the contextual nature of food and nutrition security household and not where any farming activity discussion in medium sized cities insofar as the takes place. Urban agriculture in this paper there- prospects of urban agriculture in both cities are dif- fore refers to the agricultural activities that take ferent. The population size between the two cities place in each of the urban centres. Following from varies, with Tamale having a population of this definition, the sample can be divided by the 223,252 and Techiman 67,241 (GSS, 2014). type of agriculture that the household is engaged Households in small cities could have access to in the following: households engaging solely in ur- food from their own production from surrounding ban agriculture; households combining urban and villages; on the other hand, households in medium rural agriculture; households involved only in rural sized cities may not have the same opportunity— agriculture but resident in urban centres; and and even if they do, it may come at a higher cost. households that have no participation in agriculture In addition, Techiman and Tamale have been iden- at all. A total of 2,020 households were surveyed tified as major supply centres for the national food as follows: Techiman (1,019) and Tamale (1,001). basket. The two cities also fall within two different agricultural ecological zones, the transition belt Classification of variables (Techiman) and the northern savannah belt (Tamale). Household respondents who reported some en- gagement in food production for their own con- Data sources and sampling techniques sumption in the 12months preceding the survey either on urban or rural plots were asked to give Data for this paper come in two forms. First, a their opinion about the importance of the food they quantitative questionnaire on several food security produced to their household from the different indicators (the main source of data for the paper) plots and to describe the use to which such foods was administered to the head of household (or an were put by the households. available and willing adult) in October 2013. The The Household Food Insecurity Access Scale second source of data comprised follow-up (HFIAS) developed by the Food and Nutrition qualitative interviews with representatives of the Technical Assistance under the auspices of the institutions that control the use of land in urban Academy for Educational Development was used settlements (the local authorities which have the to classify households as food secure or not. The mandate to plan the city, the chiefs who are the HFIAS is a set of nine questions that allows for owners of land, representatives of the MoFA and classification of a household as food secure or non-governmental organisations). A total of 21 not. Each question is asked with a recall period interviews was undertaken in August 2015. of four weeks (30days) with the respondent first For the first stage of the questionnaire, each city answering to the occurrence—that is, whether the was divided into four quadrants with the centre condition happened rarely (once or twice), some- serving as the epicentre to allow for proportional times (three to ten times) or often (more than ten spread of households in all four quadrants (250 times) in the past four weeks (Coates et al., households in each). The second stage involved 2007). This scale is used to classify the prevalence identifying the number of communities in each status of a household. It is noteworthy that the quadrant and proportionately stratifying the num- HFIASmeasures the access component of food se- ber of communities for equal representation. In curity but does not provide an indicator of food the third stage, the sample size was determined utilisation which also concerns nutrition. on the basis of the estimated population sizes of communities. In the final phase, enumerators were Results and discussions divided into teams of four with each team respon- sible for a community. Each community was fur- Demographic characteristics of respondents ther divided into four quadrants with each team member surveying a quadrant, starting at the cen- Demographic characteristics of the sample are tre of the community and moving in a serpentine described in Table 1. On the whole, men dominate path. Enumerators surveyed and interviewed one the headship of surveyed households (77.6%) household in every third dwelling, except for across groups and cities with higher proportions © 2017 Institute of Australian Geographers H.M. Ayerakwa, Urban household food security in Ghana 5 © 2017 Institute of Australian Geographers Table 1 Demographic characteristics of household members across groups and cities Techiman Tamale Total sample UA RA UARA No agric UA RA UARA No agric Sex of HH heads n = 268 n = 154 n = 24 n = 512 n = 141 n = 199 n = 79 n = 487 n = 1,991 Female 25.8 23.4 25.0 35.0 9.2 4.0 1.3 22.2 22.4 Male 74.3 76.6 75.0 65.0 90.8 96.0 98.7 77.8 77.6 Mean HH age 45.1 44.0 53.0 36.1 44.5 45.0 45.0 41.8 41.6 Average HH sizes Average HH size 5.1 5.3 5.1 4.1 5.4 6.0 6.4 4.1 4.8 Average HH size (18+) 3.0 2.9 3.1 2.2 3.1 3.7 3.8 2.5 2.8 Min HH size 1 1 1 1 1 1 1 1 1 Max HH size 17 11 11 15 13 24 24 17 24 HH housing structure Female centred 20.5 19.9 20.8 31.3 8.1 2.0 1.3 16.4 18.4 Male centred 8.2 7.7 8.3 10.5 2.7 6.5 12.5 10.0 8.8 Nuclear 49.3 53.2 62.5 46.6 70.5 67.3 53.8 60.3 56.0 Extended 22.0 17.3 8.3 11.1 18.8 24.1 32.5 13.0 16.3 Under 18 male centred HH – – – 0.2 – – – – 0.1 Under 18 female centred HH – – – – – – – – 0.1 Other – 1.9 – 0.4 – – – 0.2 0.4 Education of HH head Number 189 101 18 399 106 166 59 457 1,495 No formal schooling 25.9 32.7 27.8 18.6 55.7 59.6 62.7 37.4 35.3 Some primary school 20.6 15.8 16.7 18.6 8.5 6.0 8.5 4.2 11.7 Primary completed 28.0 24.8 27.8 30.8 12.3 10.2 11.9 14.7 20.7 Some high school 5.3 5.0 5.6 3.5 0.9 3.6 5.1 3.1 3.6 High school completed 10.6 12.9 11.1 15.5 7.6 12.1 3.4 14.7 13.0 Post-secondary (not univ) 6.4 4.0 5.6 8.0 9.4 4.8 6.8 18.6 10.4 Some university 0.5 1.0 – 1.0 – 0.6 – 1.5 0.9 University completed 2.7 4.0 5.6 3.8 4.7 1.8 1.7 4.6 3.7 Postgraduate – – – 0.3 0.9 1.2 – 1.3 0.7 Agric, agriculture; HH, household; RA, rural agriculture; UA, urban agriculture; UARA, urban agriculture and rural agriculture. 6 Geographical Research • 2017 reported across groups in Tamale. In Techiman phenomenon where traditionally, the people of however, nearly a third (27.3%) of household Dagbon in the Northern region generally farm in heads are women. An average household size of and around their houses in both rural and urban 4.8 is observed for the overall sample. Larger areas. This practise has remained with them and household sizes are observed in Tamale. In been passed on from one generation to another. Techiman, the minimum household size is found Additionally, as a regional capital Tamale con- among households whose members do not engage tinues to witness rapid expansion in infrastructure in agriculture (4.1) while the maximum is found making it difficult for farm households to have ac- among households engaged only in rural agricul- cess to land to grow crops. ture. Similar patterns are observed for households in Tamale. Those who do not engage in agriculture have the smallest household size of 4.1, while Cropping patterns households engaged in urban and rural agriculture have the largest households, consisting of 6.4 The food groups grown by urban agriculture members on average. This finding is consistent households in Techiman and Tamale are described with those generated in the Ghana Statistical Ser- in Table 3. Among households in Techiman, pro- vice’s Population and Housing Census of 2010 duction of roots and tubers, and grains are the (GSS, 2014). The households are mostly nuclear mostly cultivated crops (88.1% and 67.4%, respec- in composition across cities and groups (56.0%). tively). In Tamale however, production of grains and vegetables are the crops mostly cultivated. Household food production Among urban agriculture households in both cit- ies, vegetable production constitutes a very impor- Table 2 presents household participation in the tant part of their farming activities, with more than production of food either in urban or rural areas. half of households in Techiman (59%) and nearly Overall, 20.6 per cent of households were involved half (42%) of households in Tamale cultivating in the production of food in the cities in which they it, and by extension improving the dietary diversity live. Another 17.6 per cent grew food in rural areas of households. Similar patterns are observed for but resided in urban centres, while about six per rural food production in Techiman and Tamale cent of the sample was involved in both urban with nearly all (97.5%) of rural farming house- and rural agriculture. Significant differences were holds in Tamale producing grains/cereals and less found in the proportions of households engaged of vegetables and fruits. While information on in urban agriculture in Techiman (26.3%) and the different types of meat production would have Tamale (14.9%). More than half of the sample been able to shed more light on the access to ani- (56%) did not produce any food either in urban mal source of foods found among households, un- or rural areas and relied on other sources of food fortunately the dataset poses limitations to this to meet their food requirements. A noteworthy dif- type of analysis because it does not contain this ference between the two cities is that a sizable data. Follow-up qualitative interviews reveal that number of people in Techiman are actively in- cereal and grains traditionally constitute much of volved in agriculture in and around their immedi- the household diet because they can be stored over ate communities because of agricultural lands at longer periods to cover the six months’ lean/no their disposal which, while now sold for infrastruc- production season experienced by farm house- ture, await development. The practise of urban holds in northern regions of Ghana. These crops agriculture in Tamale is seen as a cultural are also relatively easy to cultivate, compared, for Table 2 Household’s participation in different type of agriculture in Techiman and Tamale Type of production Techiman Tamale Total n % n % n % Urban_Only 268 26.3 149 14.9 417 20.6 Rural_Only 156 15.3 199 19.9 355 17.6 Urban_Rural 24 2.4 80 8.0 104 5.2 No_Prod 571 56.0 573 57.2 1,144 56.6 Total 1,019 100 1,001 100 2,020 100 © 2017 Institute of Australian Geographers H.M. Ayerakwa, Urban household food security in Ghana 7 Table 3 Types of food grown among urban households in Techiman and Tamale Type of food produced UA RA UARA Total n % n % n % n % Techiman Cereal and grain products 176 67.4 111 71.2 13 54.2 300 68.0 Starchy, roots, tubers, and legumes 230 88.1 144 92.3 24 100.0 398 90.3 Vegetables 153 58.6 66 42.3 19 79.2 238 54.0 Fruits 45 17.2 27 17.3 8 33.3 80 18.1 Others (including meat and eggs) 47 18.0 23 14.7 6 25.0 76 17.2 Tamale Cereal and grain products 105 76.1 192 97.5 79 100.0 376 90.8 Starchy, roots, tubers, and legumes 38 27.5 91 46.2 34 43.0 163 39.4 Vegetables 58 42.0 30 15.2 44 55.7 132 31.9 Fruits 8 5.8 3 1.5 5 6.3 16 3.9 Others (including meat and eggs) 27 19.6 18 9.1 8 10.1 53 12.8 RA, rural agriculture; UA, urban agriculture; UARA, urban agriculture and rural agriculture. Table 4 Importance of own food production to urban households in Techiman and Tamale Techiman Tamale Total n % n % n % Importance of food from UA Not important at all 2 0.8 – – 2 0.5 Somewhat important 32 11.9 30 20.1 62 14.9 Important 51 19.0 39 26.2 90 21.6 Very important 101 37.7 66 44.3 167 40.1 Critical to our survival 75 28.0 14 9.4 89 21.3 Don’t know 7 2.6 – – 7 1.7 Importance of food from RA Not important at all 1 0.6 – – 1 0.3 Somewhat important 12 7.7 7 3.5 19 5.4 Important 39 25.0 25 12.6 64 18.0 Very important 77 49.4 135 67.8 212 59.7 Critical to our survival 27 17.3 32 16.1 59 16.6 RA, rural agriculture; UA, urban agriculture. example, to livestock, and their production is compared with only nine per cent in Tamale. favoured. A possible reason is the basic characteristics of the local economy in Techiman, which Consumption of foods cultivated by households hinges on the main weekly food market. Ta- male on the other hand has a vibrant local Table 4 shows that one fifth of households in economy and relatively developed market sys- Tamale and Techiman consider that food from tems, which enable residents to engage in some urban agriculture is critical to their survival. form of trading activities at least six days a An additional 40 per cent and 60 per cent of week. The consideration by households to par- households in Techiman and Tamale respec- ticipate in urban agriculture is thus influenced tively consider own food production as very greatly by the potential contribution the practise important to meeting the food requirements of can make to the consumption needs of their their households. Nearly a third (28%) of the households. Rural agriculture was considered sample in Techiman regards food from urban comparatively more important in Tamale; how- agriculture activities as critical to survival ever, with 16 per cent of households engaged © 2017 Institute of Australian Geographers 8 Geographical Research • 2017 Table 5 Use of own food production from different sources Techiman Tamale Total n % n % n % Food produced from UA HHs Eat it 261 100.0 147 98.7 408 99.5 Sell it 150 57.5 82 55.0 232 56.6 Give it away to friends and relatives 183 70.1 65 43.6 248 60.5 Feed it to livestock 36 13.8 20 13.4 56 13.7 Other 0 0.0 1 0.7 1 0.2 Food produced from RA HHs Eat it 155 99.4 197 99.0 352 99.2 Sell it 103 66.0 141 70.9 244 68.7 Give it away to friends and relatives 88 56.4 101 50.8 189 53.2 Feed it to livestock 25 16.0 43 21.6 68 19.2 Other 0 0.0 2 1.0 2 0.6 UA food produced from UARA HHs Eat it 24 100.0 78 97.5 102 98.1 Sell it 13 54.2 46 57.5 59 56.7 Give it away to friends and relatives 8 33.3 24 30.0 32 30.8 Feed it to livestock 2 8.3 11 13.8 13 12.5 Other 0 0.0 1 1.3 1 1.0 RA food produced from UARA HHs Eat it 22 91.7 79 98.8 101 97.1 Sell it 21 87.5 60 75.0 81 77.9 Give it away to friends and relatives 17 70.8 42 52.5 59 56.7 Feed it to livestock 4 16.7 20 25.0 24 23.1 Other 0 0.0 5 6.3 5 4.8 HH, household; RA, rural agriculture; UA, urban agriculture; UARA, urban agriculture and rural agriculture. in rural agriculture reported that it was critical produce to other urban households as gifts to fam- to their survival and a further 68 per cent stated ilies and friends. that it was very important. As illustrated in Table 5, almost all (99.5%) Food security and domestic production crops produced from urban agriculture in the sam- ple are consumed by members of the producing According to the FAO (FAO, 2012), ‘a household households and another 60.5 per cent and 56.6 is considered food insecure when people lack se- per cent, respectively are given out as gifts to fam- cure access to sufficient amounts of safe and nutri- ilies and friends or sold. While consumption re- tious food for a normal growth and development mains the primary use of foodstuff produced and an active and healthy life’. The food security through urban agriculture in both Tamale and situation of urban households correlates closely Techiman, the secondary uses varies between the with own food-provisioning arrangements, the two towns. For households engaged in rural agri- rural-urban linkages that exists among households, culture or both urban and rural agriculture, con- and households’ abilities to command food from sumption remains the primary motivation the market to meet their nutritional and dietary followed by commercialisation and food remit- needs (Table 6). Overall, nearly half of the total tance motives. The primary role of food produced sample is food secure in both Techiman (48%) in both urban and rural areas, hence, is related to and Tamale (47%). The proportions of households consumption rather than commercialisation. This considered severely food insecure in Techiman are finding establishes the relative importance of agri- found among households practising only urban ag- culture to urban households as measured by its riculture (21%) and those with no agriculture contribution to household food provision in the (20%), while severe food insecurity is found two cities. A point noteworthy is the fact that more among households practising only rural agriculture than half of all households give part of their (13%). The qualitative data suggest that the © 2017 Institute of Australian Geographers H.M. Ayerakwa, Urban household food security in Ghana 9 explanation for this variance may be found in the nature of urban agriculture production in Techiman, because residents there concentrate on maize cultivation (during the major cropping sea- son) with very little cropping activities during the minor season. (personal communication, Nana Abankwa, Chief, Techiman, 6 August 2015). Although urban agriculture contributes sizable proportions to food provisioning among participat- ing households, the volumes are generally inade- quate to cater for all food requirements. In Tamale, by contrast, households practising only rural agriculture are found to be the most severely food insecure, with nearly a quarter (23%) of these households being severely food insecure, followed by those without access to any form of agricultural production capacity (21%). Because of the rapid expansion of Tamale, most farm lands continue to be converted into residential properties, a trend which contributes to the distance over which urban residents have to travel to access arable land for cultivation in the surrounding rural areas, with sig- nificant travel costs incurred as a result (personal communication, Razak Gombila, AssemblyMem- ber, Tamale, 11 August 2015). In Tamale, those experiencing the lowest levels of severe food insecurity severely food insecure households number among households practising both urban and rural agriculture (11.3%). The food security situation in Tamale thus differs signifi- cantly for households depending on their agricul- tural involvement and the type of agriculture that they are engaged in. The food security situation in Techiman, by contrast, is roughly similar for all the different household types in the city: regard- less of the degree and type of engagement in agri- culture, around 50 per cent of the households reported being food secure. Again, this finding points to the importance of situating urban agricul- ture in relation to rural agriculture as well as con- textual factors related to the city environment itself. Sources of cash income As suggested by the data on uses of own production, farm households generally sell around a third of their produce. The shares of cash incomes accruing to urban households from the different sources of income and in particular from the various types of agriculture in the two cities are presented for Techiman and Tamale in Tables 7 and 8. For the sample as a whole income from informal businesses (including petty trading and selling of household consumables) and wage © 2017 Institute of Australian Geographers Table 6 Prevalence of food insecurity among urban households Prevalence of food insecurity UA RA UARA No agric Total n % n % n % n % n % Techiman Food secure 130 48.5 73 46.8 11 47.8 249 48.6 482 47.8 Mildly food insecure 32 11.9 16 10.3 6 26.1 46 9.0 104 10.3 Moderately food insecure 50 18.7 47 30.1 3 13.0 117 22.9 227 22.5 Severely food insecure 56 20.9 20 12.8 3 13.0 100 19.5 195 19.4 Prevalence of food insecurity Tamale Food secure 76 51.0 73 36.9 51 63.8 237 47.5 468 47.0 Mildly food insecure 21 14.1 34 17.2 11 13.8 58 11.6 132 13.3 Moderately food insecure 26 17.5 46 23.2 9 11.3 101 20.2 198 19.9 Severely food insecure 26 17.5 45 22.7 9 11.3 103 20.6 198 19.9 Agric, agriculture; RA, rural agriculture; UA, urban agriculture; UARA, urban agriculture and rural agriculture. 10 Geographical Research • 2017 © 2017 Institute of Australian Geographers Table 7 Urban households cash income sources from all activities in Techiman Sources of HH income UA RA UARA No agric Total n % n % n % n % n % Wage work 88 17.2 50 18.0 7 14.3 204 23.8 349 20.6 Casual work 39 7.6 15 5.4 1 2.0 44 5.1 99 5.8 Remittances –money 44 8.6 24 8.6 3 6.1 97 11.3 168 9.9 Remittances – goods 7 1.4 2 0.7 0 0.0 16 1.9 25 1.5 Remittances – food 20 3.9 13 4.7 1 2.0 65 7.6 99 5.8 Income from rural farm 0 0.0 63 22.7 14 28.6 3 0.4 80 4.7 Income from urban farm 115 22.4 0 0.0 7 14.3 2 0.2 124 7.3 Income from formal business 10 2.0 11 4.0 1 2.0 44 5.1 66 3.9 Income from informal business 164 32.0 89 32.0 12 24.5 344 40.1 609 35.9 Income from renting dwelling 6 1.2 3 1.1 1 2.0 4 0.5 14 0.8 Income from aid – food 0 0.0 0 0.0 0 0.0 1 0.1 1 0.1 Income from aid – cash 0 0.0 0 0.0 0 0.0 1 0.1 1 0.1 Income from aid – vouchers 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 Pension/disability/other social grants 6 1.2 3 1.1 1 2.0 7 0.8 17 1.0 Gifts 2 0.4 1 0.4 0 0.0 15 1.8 18 1.1 Other 12 2.3 4 1.4 1 2.0 11 1.3 28 1.7 Agric, agriculture; HH, household; RA, rural agriculture; UA, urban agriculture; UARA, urban agriculture and rural agriculture. H.M. Ayerakwa, Urban household food security in Ghana 11 © 2017 Institute of Australian Geographers Table 8 Urban households cash income sources from all activities in Tamale Sources of HH income UA RA UARA No agric Total n % n % n % n % n % Wage work 61 19.9 43 12.0 14 8.6 214 23.8 332 19.2 Casual work 44 14.3 39 10.9 39 24.1 116 12.9 238 13.8 Remittances –money 25 8.1 16 4.5 11 6.8 100 11.1 152 8.8 Remittances – goods 4 1.3 4 1.1 1 0.6 12 1.3 21 1.2 Remittances – food 14 4.6 14 3.9 2 1.2 50 5.6 80 4.6 Income from rural farm 0 0.0 116 32.4 28 17.3 3 0.3 147 8.5 Income from urban farm 53 17.3 0 0.0 25 15.4 0 0.0 78 4.5 Income from formal business 3 1.0 6 1.7 0 0.0 32 3.6 41 2.4 Income from informal business 79 25.7 103 28.8 36 22.2 278 30.9 496 28.7 Income from renting dwelling 5 1.6 4 1.1 0 0.0 5 0.6 14 0.8 Income from aid – food 0 0.0 0 0.0 0 0.0 1 0.1 1 0.1 Income from aid – cash 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 Income from aid – vouchers 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 Pension/disability/other social grants 3 1.0 0 0.0 0 0.0 23 2.6 26 1.5 Gifts 9 2.9 3 0.8 3 1.9 27 3.0 42 2.4 Other 7 2.3 10 2.8 3 1.9 40 4.4 60 3.5 Agric, agriculture; HH, household; RA, rural agriculture; UA, urban agriculture; UARA, urban agriculture and rural agriculture. 12 Geographical Research • 2017 Table 9 Average household cash income sizes HH type Obs Mean Obs Mean Techiman Tamale UA 238 642.5 147 916.1** RA 121 736.6 186 729.5 UARA 23 792.4 76 905.7 No production 464 540.6** 522 696.8** City average 846 604.1*** 931 755.0*** Overall sample 683.2 HH, household; RA, rural agriculture; UA, urban agriculture; UARA, urban agriculture and rural agriculture. **(P = 0.05). ***(P = 0.01). related work constitute the two main sources of for agricultural households in Techiman and Ta- income available to households in Techiman male, underscoring the important role agriculture contributing more than half (52%) of total cash plays both as a source of food and income. income. Households whose members do not engage in agriculture raise their incomes primarily Size of cash incomes through engagement in informal businesses, with income from this source constituting about 40 per The gross mean monthly income for the sample cent in Techiman and 31 per cent in Tamale. shows that in Techiman higher incomes are ob- For households that engage in agricultural pro- served for households that engage in both urban duction, however, the picture is different. For and rural agriculture (GHS 792), followed by households engaged in urban and rural agriculture, households that practise only rural agriculture the combined income raised is the highest in both (GHS 734), and urban agriculture households cities, contributing about 43 per cent and 33 per (GHS 643). Non-agricultural households in cent of total cash incomes in Techiman and Techiman are found to have the lowest gross Tamale, respectively (see Table 9). monthly income (Table 9). In Tamale, households In Techiman, households whose members en- that practise only urban agriculture are found to gage in urban agriculture generate nearly a quarter have the highest average monthly incomes (22%) of their cash income from urban agriculture, (GHS 916), followed by those engaged in both even if income from informal business is the larg- urban and rural agriculture (GHS 906), and only est individual source of income (32%). The impor- rural agriculture (GHS 730). While household tance of urban agriculture in income terms is participation in rural agriculture records the slightly lower in Tamale, where incomes from highest monthly cash income in Techiman, such productive activities are the third most impor- households who participate in urban agriculture tant source of income to households, contributing tend to have the highest average monthly incomes nearly a fifth to household income (17.3%) after in Tamale. A point noteworthy is the fact that incomes from informal businesses (25.7%) and participation in both urban and rural agriculture wage income (19.9%). For households engaged in Techiman and Tamale contributes significantly in rural agriculture in Tamale, income from rural to household incomes. A test of means shows that agriculture is the highest, contributing 32 per cent gross monthly income from households involved of the total household cash incomes. The impor- only in urban agriculture is higher and signifi- tance of rural agriculture in income terms is cantly different from all other household catego- slightly lower in Techiman. Income generated ries in Tamale. Similarly, the average monthly from rural agriculture is the second most important income observed for non-agricultural households source of income contributing nearly a quarter is found to be the lowest and significantly differ- (23%) of the total household cash incomes in ent from other household categories in Techiman. Techiman. The result corroborates the relative importance of Cash income from agriculture—whether rural both urban and rural agriculture to livelihoods of or urban—serves as an important complement to urban households involved in such cultivation agricultural production used for own consumption and further underscores the importance of rural- © 2017 Institute of Australian Geographers H.M. Ayerakwa, Urban household food security in Ghana 13 urban linkages in the discussion of food and nu- Cities Growing Food: Urban Agriculture on the Policy trition security. Agenda. Feldafing: DSE Feldafing. Badami, M.G. and Ramankutty, N., 2015. Urban agriculture and food security: a critique based on assessment of urban Coda land constraints. Global Food Security, 4, pp.8–15. Chagomoka, T., Drescher, A., Glaser, R., Marschner, B., This paper makes an important contribution to Schlesinger, J. and Nyandoro, G., 2015. Vegetable existing research on urban agriculture, by con- production, consumption and its contribution to diets along textualising this agricultural practice in terms of the urban–rural continuum inNorthernGhana.African Journalof Food, Agriculture, Nutrition and Development, 15(4), its effects on different classes of households in pp.10352–10367. medium sized urban landscapes in Ghana. The Chagomoka, T., Unger, S., Drescher, A., Glaser, R., Marschner, paper emphasises the point that urban agriculture B. and Schlesinger, J., 2016. Food coping strategies in northern should not be treated in isolation from other Ghana. A socio-spatial analysis along the urban–rural mechanisms that households employ to meet continuum. Agriculture & Food Security, 5, 4.Coates, J., Swindale, A. and Bilinsky, P., 2007. Household their food and nutrition needs. The contribution Food Insecurity Access Scale (HFIAS) for Measurement of of urban agriculture to household food security Food Access: Indicator Guide. Washington, DC: Food and should be viewed in the context of understanding Nutrition Technical Assistance Project, Academy for Educa- the values that households involved in urban ag- tional Development. Cofie, O. and Drechsel, P., 2007. Water for food in the cities: riculture place on produce and considering how the growing paradigm of irrigated (peri)-urban agriculture produce is utilised. In effect, food and nutrition and its struggle in sub-Saharan Africa.AfricanWater Journal, security in small and medium sized cities should 1(1), pp.26–50. be studied in the context of broader arrange- Cofie, O.O., vanVeenhuizen, R. and Drechsel, P., 2003.Contri- ments, entitlements and linkages available to the bution of urban and peri-urban agriculture to food security inSub-Saharan Africa. Colombo: International Water Manage- households in question and mindful of the extent ment Institute. to which they supplement household food and Crush, J., Hovorka, A. and Tevera, D., 2011. Food security in nutrition needs. southern African cities: the place of urban agriculture. Progress in Development Studies, 11(4), pp.285–305. de Haan, L. and Zoomers, A., 2003. Development geogra- Acknowledgements phy at the crossroads of livelihood and globalization. Tijdschrift voor Economische en Sociale Geografie, 4(3), The author is grateful for the financial support from the Swedish pp.350–362. Research Council for Environment, Agricultural Sciences and Ellis, F. and Sumberg, J., 1998. Food production, urban areas Spatial Planning under the Swedish African Urban Agriculture and policy response. Food Policy, 26(2), pp.213–225. Project. I am also grateful to the African Food Security Urban FAO, 2012.Growing greener cities in Africa: first status report Network (AFSUN) for the cooperation in the use of several on urban and peri‐urban horticulture in Africa. Rome: sections of their questionnaire to collect the urban agriculture United Nations Food and Agriculture Organization. baseline data. The useful comments from Professors Agnes Foeken, WJD. and Owuor, S., 2008. Farming as a livelihood Andersson Djurfeldt, Magnus Jistrom and the Special Issue source for the urban poor of Nakuru, Kenya. Geoforum, 39 Editor Prof. Bill Pritchard are appreciated. (6), pp.1978–1990. Frayne, B., 2005. Rural productivity and urban survival in Note Namibia: eating away from home. Journal of Contemporary African Studies, 23(1), pp.51–76. US$1 =GHS 4.2075 as at 28 November 2016. Frayne, B., McCordic, C. and Shilomboleni, H., 2014. Growing out of poverty: does urban agriculture contribute to house- Conflict of interest hold food security in southern African cities? Urban Forum, 25(2), pp.177–189. None declared. Ghana Statistical Services, 2013. 2010 Population and HousingCensus: National Analytical Report. Accra: Ghana Statistical Services. References Ghana Statistical Services, 2014. Ghana Living Standard Survey Round 6 (GLSS 6). Accra: Ghana Statistical Services. Amoah, P., Drechsel, P., Henseler,M. and Abaidoo, R.C., 2007. Gupta, R. and Gangopadhyay, SG., 2013. Urban food security Irrigated urban vegetable production in Ghana: through urban agriculture and waste recycling: some lessons microbiological contamination in farms and markets and for India. Vikalpa, 38(3), pp.13–22. associated consumer risk groups. Journal of Water and Hillbruner, C. and Egan, R., 2008. Seasonality, household food Health, 5(3), pp.455–466. security, and nutritional status in Dinajpur, Bangladesh.Food Andersson, A., 2002. The Bright Light Grow Fainter: and Nutrition Bulletin, 29(3), pp.221–231. Livelihoods, Migration and a Small Town in Zimbabwe. Kaag, M., van Berkel, R., Brons, J., de Bruijn, M., van Dijk, H., Stockholm: Department of Human Geography, Stockholm de Haan, L., Nooteboom, G. and Zoomers, A., 2004. Ways University. forward in livelihood research. In D. Kalb, W. Pansters and Armar-Klemesu, M., 2000. Urban agriculture and food security, H. Siebers, eds. Globalization and Development: Themes nutrition and health. In N. Bakker, M. Dubbeling, S. and Concepts in Current Research. Dordrecht: Kluwer, Guendel, U. Sabel-Koschella and H. de Zeeuw, eds.Growing pp.49–74. © 2017 Institute of Australian Geographers 14 Geographical Research • 2017 Koscica, M., 2014. Agropolis: the role of urban agriculture in Mougeot, LJ., 2011. International Support to Research and addressing food insecurity in developing cities. Journal of Policy on Urban Agriculture (1996–2010): Achievements International Affairs, 67(2), pp.177. andChallenges. Ottawa: International Development Research Lee‐Smith, D., 2013. Which way for UPA in Africa? City, 17 Centre. (1), pp.69–84. Potts, D. and Mutambirwa, C., 1998. Basics are now a luxury: Maxwell, D., Levin, C. and Csete, J., 1998. Does urban agricul- perceptions of the impact of structural adjustment on rural ture help prevent malnutrition? Evidence from Kampala. and urban areas in Zimbabwe. Environment and Urbaniza- Food Policy, 23(5), pp.411–424. tion, 10(1), pp.55–66. Maxwell, D.G., Levin, C.E., Armar‐Klemesu, M., Ruel, M.T., Steel, C., 2008. Hungry City: How Food Shapes Our Lives. Morris, S.S., Ahiadeke, C., 2000. Urban livelihoods and London: Chatto & Windus. food and nutrition security in Greater Accra, Ghana. Stewart, R., Korth,M., Langer, L., Rafferty, S., Da Silva, N. and International Food Policy Research Institute. Available at: van Rooyen, C., 2013. What are the impacts of urban agricul- http://www.ifpri.org/publication/urban‐livelihoods‐and‐ ture programs on food security in low and middle‐income food‐and‐nutrition‐security‐greater‐accra‐ghana. [Accessed countries? Environmental Evidence, 2, 7. 28/11/2016]. van Veenhuizen, R., 2006. Cities Farming for the Future: McClintock, N., 2010. Why farm the city? Theorizing urban Urban Agriculture for Green and Productive Cities. Ottawa: agriculture through a lens of metabolic rift. Cambridge RUAF Foundation, IDRC. Journal of Regions, Economy and Society, 3(2), pp.191–207. Viljoen, A., 2005. Continues Productive Urban Landscapes: Mkwambisi, D.D., Fraser, E.D.G. and Dougill, AJ., 2011. Designing Urban Agriculture for Sustainable Cities. Oxford: Urban agriculture and poverty reduction: evaluating how Elsevier. food production in cities contributes to food security, Warren, E., Hawkesworth, S. and Knai, C., 2015. Investigating employment and income in Malawi. Journal of International the association between urban agriculture and food security, Development, 23(2), pp.181–203. dietary diversity, and nutritional status: a systematic literature MoFA, 2010. Medium Term Agriculture Development Plan review. Food Policy, 53, pp.54–66. (METASIP) 2011–2015. Accra: Ministry of Food and World Bank, 2012. World Development Indicators 2012. Agriculture. Washington, DC: World Development Indicators. Moorsom, R., 1995.Underdevelopment and Labour Migration: Zezza, A. and Tasciotti, L., 2010. Urban agriculture, poverty, The Contract Labour System in Namibia. Windhoek: Depart- and food security: empirical evidence from a sample of ment of History, University of Namibia. developing countries. Food Policy, 35(4), pp.265–273. © 2017 Institute of Australian Geographers Paper II Effects of food production on household food security in a small and medium-sized city in Ghana Hayford Mensah Ayerakwa, Department of Human Geography, Lund University Fred M. Dzanku Institute of Statistical, Social and Economic Research, University of Ghana Daniel B. Sarpong Department of Agricultural Economics and Agribusiness, University of Ghana 11 Abstract The debate on the contribution of urban agriculture to food security seems to over concentrate on the effects of urban agriculture on household food security alone without considering the fact that, there are farmers living in urban areas who have other farming engagements in peri-urban and rural areas. Using data collected on 2,000 urban households in two intermediate-sized cities in Ghana, we examine the effects of urban households’ food production in urban and rural spaces on urban food security. The results demonstrate that, overall, urban residents who produce food in both urban and rural space achieve better food security outcomes. Key words: urban agriculture, rural agriculture, own food production, intermediate sized cities, household food security, Ghana 2 Introduction Urban and peri-urban agriculture (UPA) has in recent times received considerable attention for its importance in the provision of food both to producing households and also to city dwellers. UPA also serves as a source of income to a chain of beneficiaries including farmers, market sellers, input suppliers and invariably contributing to the livelihoods of people (Armar-Klemesu 2000; Danso et al. 2002; Drechsel et al. 2006; Henn 2000; Obosu-Mensah 1999; Sabates et al. 2001; Sanyal 1985; Smit et al. 1996). Probably due to the relative ease of access to water for irrigation in cities UPA is usually an all year round activity that takes place in open spaces, government reserved lands, and on non-habitable areas in and around high tension poles, contributing to the supply of food to city dwellers, and to poverty alleviation among producers (MoFA 2010). The practice of agriculture by urban residents in Ghana dates back to the colonial days with the arrival of Europeans. The practice then focused on cultivation of crops, especially vegetables in and around the castles and forts in the 16th century (Anyane 1963; Danso et al. 2014). A government supported initiative of own food production called Operation Feed Yourself (OFY) in the 1970s brought a general acceptance of agriculture in urban areas where citizens were encouraged to grow food in their backyards (Girdner et al. 1980). This was possible at the time because the cities had lower population pressures to deal with, but also because the practice received executive approval. However, Ghana’s urban population has been growing much faster than the rural population – averaging 4.2 % and 1.7 % annual growth between 1960 and 2014, respectively (Fig. 1). At such a pace, more than half of Ghana’s population were living in urban areas before 2010, with the urban population share having risen from just 23.1 % in 1960 to about 53.4 % in 20141 (Fig. 2). Increasing urbanization coupled with rapid expansion of infrastructure places extra demand on agricultural lands to be converted for other uses including housing and the provision of essential social amenities. Urban agriculture has been promoted by activists including Civil Society Organizations (CSOs), researchers, government agencies, NGOs and development agencies as a positive pro-poor initiative to be encouraged in the poverty alleviation agenda in developing countries (Cofie and Drechsel 2007; FAO 2012; Lee-Smith 2010; Lee-Smith 2013; Mougeot 2011; Smit et al. 1996; Gockowski, et al., 2003). Available evidence suggests that urban agriculture provides an important source of food in most developing countries and serves as a critical food security strategy among the urban poor (Armar- Klemesu 2000; Maxwell 2001; Mougeot 2000; Nugent 2000) while improving nutritional status of practicing households (Maxwell 2001; Mwangi and Foeken 1996). Recent evidence suggests that the importance of urban agriculture far exceeds what has been previously understood. For example, Thebo et al. (2014) shows that approximately 11% and 5% of all irrigated and rainfed cropland, respectively are situated in urban areas. If the urban boarder radius is extended by 20km the share rises to 60% and 35% respectively. While there is no outright condemnation of urban agriculture, its relative importance in meeting urban household food requirements has been questioned by several researchers (Crush et al. 2011; Frayne et al. 2014; Lee-Smith 2013; Stewart et al. 2013; Zezza and Tasciotti 2010). The lack of comparable data on the scale as well as the impact of urban agriculture on food security seems to further validate the scepticism (Stewart et al. 2013). Zezza and Tasciotti (2010), for example, believe the potential impact of urban agriculture has been exaggerated by activists. They also question what has been widely cited in the literature as the estimated number of people involved in urban agriculture globally (about 800 million). What is missing from their analysis is failure to provide alternative estimates of what numbers are engaged in urban agriculture worldwide. 1 The inter-census population growth and the 2015 urban-rural population share figures are estimates from World Bank 2016. 33 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 Year Urban Rural Fig. 1 Ghana’s urban-rural population growth, 1960-2014 2014 53% 2010 51% 2000 44% 1984 32% 1970 29% 1960 23% 0% 20% 40% 60% 80% 100% Urban Rural Fig. 2 Ghana’s urban-rural population share Although the literature on urban agriculture and its contribution to household food security has received considerable attention, little is known about urban households’ participation in agricultural production (both urban and rural) and how these engagements contribute to their food security. More so, while studies on urban agriculture exist in Ghana, with the exception of a few (Chagomoka et al. 2015; Chagomoka et al. 2016) most of the literature focus on Accra and Kumasi (the two largest cities). In this study however, we analyse the contribution of own food production in both urban and rural areas to urban household food security from the perspective of a small and an intermediate sized city. The rest of the paper is structured as follows: The study context is presented in the next section. This is followed by a discussion on the conceptual linkages that exist between urban and rural space in the 4 Census year Growth rate (%) 1960 1963 1966 1969 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002 2005 2008 2011 2014 production of food. Section four presents the methodology. The results from our empirical model is presented and discussed in section five. The conclusions are presented in section six. Study context As could be expected, both polity and broader institutional contexts matter for agricultural production in general and urban agriculture in particular. The government of Ghana, unlike their counterparts in many other African countries, has a policy on urban agriculture. This is contained in Ghana’s Medium Term Agriculture Development Plan (METASIP) 2011-2015 (MoFA 2010). In the METASIP policy document, urban and peri-urban agriculture have been acknowledged as major contributors to national food security. METASIP does not consider the interplay between agricultural production by urban dwellers in urban space on the one hand and production by urban dwellers in rural space on the other, and how this relates to urban household food security. In this sense, the policy does not contextualise urban agriculture in small and intermediate sized cities (Ayerakwa 2017). Studying this gap is an important contribution of this paper. We use empirical data to show that context matters in discussing the contribution of own food production to household food security. In addition, we show that own food production contributes to urban households food security and that, urban residents’ participation in agriculture could be used as an effective poverty reduction strategy in specific contexts. To put the research in context, we discuss some differences between the two cities where this study is sited (Techiman and Tamale). The discussion is based on qualitative interviews with key informants (including Chiefs, Assembly Members, Officers from the Ministry of Food and Agriculture (MoFA), and other agencies working in the field of urban agriculture). The two cities demonstrate some contextual differences that shape the results of the study. First, an assessment of the land ownership structure reveals that whereas Techiman lands are mostly vested in the paramount chief (stool lands) a significant share of lands in Tamale is under state control. The situation in Tamale is mainly as a result of the city being the administrative capital of the Northern region. As has become the convention in Techiman, agricultural lands revert to the chief once urban development reaches a particular sub-community. Whoever cultivated the parcel prior to the reversal receives a portion of the land as compensation. The beneficiary has the opportunity to continue cultivating the land or use it for other purposes including the option of outright sale. In Tamale, state lands are sometimes leased out to private developers whereas those remaining undeveloped are cultivated by the original families until such a time that the government decides to develop it. As a result of the different land ownership structures, the nature of urban agriculture in the two cities are also different. The qualitative interviews show that farming in urban spaces in Techiman is an important strategy for ensuring land tenure security. The rule is that, if purchased land is left undeveloped over a two year period, the chiefs reserve the right to resell the land to other interested persons who are ready to develop the land. This is done to avoid weeds growing on plots in the city center. In Tamale on the other hand, urban farming is part of the livelihoods of several families who devote substantial amount of resources to the production of high value crops for consumption and sale to other city dwellers through market intermediaries. This takes place mostly in open spaces belonging to the state. Due to the relatively high population of public servants and expatriates in Tamale, there is higher demand for freshly produced exotic vegetables in Tamale than Techiman. Farmers take advantage of the high demand for exotic vegetables and the availability of open spaces within the city to grow such vegetables, especially during the dry season when rainfed agriculture is impossible in rural and peri- urban areas. Apart from the rule guiding the purchase and development of land in Techiman, it is easy to access peri- urban and rural lands for own food production. As a result, farming in the city remains generally 55 unattractive. Farmers in Tamale on the other hand tend to spend several hours commuting to peri-urban and rural areas to access land for farming. As a result, where possible, every available space is utilised for farming. Additionally, farmers in Tamale tend to benefit from extension support systems which makes them more efficient in their production activities. It is thus expected that, the impact of urban agriculture on household food security in Tamale would be stronger than in Techiman. Since it could be argued that discussions of urban food security from the perspective of household own food production should be done in the context of access to the most valuable resource (i.e. land), the context provided above helps shape our understanding of possible differences in the effect of urban agriculture on food security. Determinants of food (in) security The Food and Agriculture Organization (FAO 1996) defines food Security as the situation in which all people, at all times, have physical and economic access to sufficient, safe and nutritious food that meets their dietary needs and food preferences for an active and healthy life. The multidimensional nature of food (in) security makes measurement complex with a lack of agreement across disciplines. In the 1980s, food security was perceived as a problem of food shortage that required increased production. This view resulted in investments in food aid to developing countries like Africa and in Green Revolution technologies in Asia. In that era, food availability was tracked using food balance sheets which provided estimates of available food (Hendriks 2015; Pinstrup-Andersen 2009; Webb et al. 2006). However, increased food production only addressed the challenge of food availability but did not guarantee access by individuals and households who needed food the most. For example, the Food and Agriculture Organization (FAO, 2002) suggests nearly a third (33%) of people living in Sub-Saharan Africa (SSA) are undernourished. With Ghana’s current urban population share already in excess of the continent’s projected growth by 2030, the concerns on urban food security and poverty alleviation is further heightened (Haddad et al., 1998; Armar-Klemesu, 2000; Frayne, 2005; United Nations, 2005; UN Habitat, 2007). The recent food hikes in some African cities is an indication of the extent of the problem of urban food security and why urgent solutions are needed to address them, especially in the area of distribution. As noted by Sen (1981) people experience food deprivation due to their difficulty in accessing food and not necessarily because of the non-availability of it. Measurement of food security has since evolved with different interests and foci across disciplines (Barrett 2010). A household’s food security is affected by their purchasing power and other resources such as production possibilities, transfers and gifts, market conditions (including market integration), and policies (Sen 1981; Hendriks, 2015). Household and household head characteristics such as age, education and household size are also important because they influence the income earning opportunities of households (Ibekwe 2010; Talukder 2014). Dzanku and Sarpong (2011) used data from rural Ghana to show that resource allocation to staple crop production, household characteristics such as sex, age, household composition and education; physical asset wealth, remittances, other non-farm income and distance to main market outside the village are important predictors of rural household food security. Some studies have found that female headed households are more likely to be food insecure than male headed households (e.g., Bashir et al. 2012; Dzanku and Sarpong 2010) Evidence from Mozambique show that household food access depends on income, access to land and resources for own food production (Garrett and Ruel 1999). The presence of small children or the elderly, education of household members, and location were all found to influence the food security situation of households. Additionally, household income from participation in non-farm activity was found to lower the likelihood of household food stock decline during critical moments of food shortages (Aidoo et al. 2013; Owusu et al. 2011) as well as reduce the prevalence of child stunting, underweight 6 and wasting (Babatunde and Qaim 2010). Feleke et al. (2005) group the determinants of household food security in Ethiopia into demand and supply side factors and conclude that supply side variables including technology adoption, farming system, farm size and land quality better explain household food security than demand side variables such as household size, per capita aggregate production and market access (Feleke et al. 2005). Other variables found to influence household food security include credit access and marital status (Aidoo et al., 2013); fertilizer application, cattle ownership and access to irrigation (Sikwela, 2008). The determinants of food security therefore vary across geographical, cultural, sociopolitical contexts. Methodology Sample This paper is part of a broader multidisciplinary research project carried out in Ghana, Kenya and Uganda with the aim of understanding the social, economic and environmental challenges and prospects of urban agriculture under changing global and demographic realities. For the Ghana case study, Techiman municipality and Tamale metropolitan area in the Brong Ahafo and Northern regions, respectively, were purposively selected for the study. The rational was to break away from the metro bias that has characterised earlier research on urban agriculture in Ghana. The two cities have a population of 67,241 and 223,252 inhabitants respectively (GSS 2014b, 2014a), and have been identified as major supply blocks to the national food basket. Techiman is located within the transition belt whereas Tamale is in the northern savanna zone. The choice of Techiman and Tamale also provides insight into the contextual nature of food security since the prospects for participation in urban and rural agriculture in the two cities are different. For example while households in small cities like Techiman could have access to own-produced food from surrounding villages, households in intermediate sized cities such as Tamale may not have the same opportunity. Surveys were conducted in the two cities in October 2013. This was complemented by qualitative interviews used to elicit city specific contexts. The quantitative surveys on the other hand aimed at a representative sample of households in each city. Overall, about 1,000 households in each city were interviewed as part of the quantitative survey in addition to 21 key informant interviews. The food security measure The outcome variable of interest to this article is a food (in) security indicator. Following Coates et al., (2007), we use the ‘Household Food Insecurity Access Scale’ (HFIAS) to classify households as food secure or not based on a set of nine questions. The questions take into account a thought reflection on several indicators including hunger, anxiety (about household food access, satisfaction of food preferences) worry and wellbeing (Headey and Ecker, 2013). The information generated by the HFIAS can be used to assess the prevalence of household food insecurity and to detect changes in the household food insecurity situation of a population over time (Coates et al. 2007, p. 2). Respondents are asked each of the questions with a recall period of four weeks and allowed to assess first an occurrence of food insecurity and then a frequency of different types of food insecurity occurrence question is asked to determine whether the condition occurred rarely (once or twice), sometimes (three to ten times) or often (more than ten times) in the past four weeks (Headey and Ecker, 2013; Coates et al. 2007). It is noteworthy that, the HFIAS measures the access component of food security but does not provide an indicator of food utilization which relates to nutrition. Following the classification of Coates et al., (2007), we classify as food secure any household that does not experience anxiety about food access, satisfaction of food preferences, or shortages of food in daily 77 lives, or just experiences worry, but rarely. Households that are found to worry about inadequate food for consumption sometimes or on a more frequent basis, or without the ability to eat preferred foods, and/or eats a more monotonous diet than desired are classified as mildly food insecure. The households in this category do not reduce the quantities of food consumed nor run out of food, going to bed on an empty stomach or going a whole day without food. For the moderately food insecure households, they turn to sacrifice quality on a regular basis and also eat monotonous or undesirable foods sometimes or often. In some instances, they have started to reduce the size or number of meals. A severely food insecure household runs out of food, have members going to bed hungry, or going a whole day and night without eating on a regular basis. They turn to cut back on meal size or number of meals often. (Coates et al. 2007). Using the HFIAS we are able to reclassify households into the different household food insecurity access prevalence (HFIAP). Based on the above, let FSSi denote the food security severity status of household i. A househld belongs to one of four food security groups: Severely Food Insecure (FSSi = 1), Moderately Food Insecure (FSSi = 2), Mildly Food Insecure (FSSi = 3), and Food Secure (FSSi = 4). Aside this food security scale, we also define two broad categories of food security status, where a household is either food secure (FSSi = 4) or food insecure (FSSi = 1, 2, 3). Validity and limitations of the food security measure As indicated earlier, the HFIAS measures only the access component of food (in) security. The validity of the scale has been tested to have several advantages including the ability of the scale to capture psychological dimensions of food insecurity which also has the ability to elicit expectations that potentially can impart on the food security status of households. The scale has also been seen as fairly easy to measure relative to other food insecurity measures. This also makes it possible to account for seasons based on the reference category identified. Several other literature on the veracity of the scale suggests it is a simple and valid tool for measuring the access component of food security (Becquey et al. 2010; Gebreyesus et al. 2015; Salarkia et al. 2014). In all the studies cited above, the HFIAS showed a good internal consistency. The above notwithstanding, Headey and Ecker (2013) found that the ordering of the questions in the scale affects the reported food insecurity situation in China. He argues further that, there is a lack of comparability across wealth and education groups such that there is a general lack of a common reference point. The next weakness associated with the HFIAS relates to the lack of cross cultural comparability across the set of questions. For example, literature on the test of cultural comparability of the HFIAS in six countries reveal only three out of the nine questions demonstrate cross country comparability (Deitchler et al. 2010). Headey and Ecker (2013) argue that HFIAS has the tendency to underestimate food insecurity situation due to feelings of shame associated with admission of hunger. Similarly, expectations of public transfers and other material incentives has the potential to overestimate food insecurity situation. Furthermore, the period of interview could possibly cloud the true food security situation of the household. Periods prior to harvesting seasons are usually characterised by shortages compared to periods during and immediately after harvest. As a result, conducting an interview at a particular time of the year may skew responses in favour of a particular indicator. For this last case, we attempt judging how this could influence our results and conclusions by doing the analysis using alternative indicators that sought to assess household food (in) security over a 12 month period. The drawback with this is the length of the recall period. Using multiple indicators, however, helps us check the robustness of our estimates. 8 Empirical Model specification We first specify a binary food security equation as K FS*i  kownproduction ik   X i it , i 1,2,..., N (1) k1 FSi 1 if FS * i  0; FSi  0 otherwise, where FS* is the latent unobserved food security level of urban household i; FSi is the food security i level which is only observed as a binary response variable which takes on the value 1 for a household that does not experience any of the food insecurity episodes (i.e. FSSi = 4), 0 otherwise (i.e. for FSSi = 1, 2, 3); ownproduction, the main explanatory variable of interest, is the categorical variable which defines the four types of food producing urban households (ownproduction = 1 if household does not produce any food, ownproduction = 2 if the household produces food in urban space only, ownproduction = 3 if household produces food in rural space only, and ownproduction = 4 if household produces food in both urban and rural space); the vector X contains all other explanatory variables (mainly individual and household characteristics); α, the βs andk (k 1,2,3,4) are the unknown parameters to be estimated; εi is the idiosyncratic error term. Assuming a standard normal distribution for εi leads us to fit a probit model to equation (1). The food security severity indicator, FSSi, is amenable to an ordered choice regression. We thus specify K FSS*i kownproductionik   X i i , FSSi  J if  j1  FSS*i   j , i 1,2,..., N (2) k1 where  is the threshold parameter, and all other variables and parameters are as defined in equation (1). An ordered probit model is estimated for equation (2) assuming a standard normal CDF of the unobserved. Our hypothesis in both equation (1) and (2) is that, on average, urban households who produce food in both urban and rural space have better food security status than urban households who either do not engage in agriculture at all or engage in food production in urban space only or rural space only. Thus, relative to 4 , we expect 1,2 and 3 to all be negative and statistically different from zero at the 5% level. Equations (1) and (2) are also estimated replacing ownproduction with an urban household food production dummy (ownprod), that is, a variable that takes on the value 1 if an urban household produces their own food irrespective of the spatial location where the food is produced, and 0 otherwise. In this instance, the models are: FS*i  ownprodi   X i it , i 1,2,..., N FSi 1 if FS * i  0; FSi  0 otherwise, (3) and FSS * i  ownprodi  Xi i , FSS * i  J if  j1  FSSi  j , i 1,2,..., N (4) 99 A priori, we expect urban households who produce their own food to be more food secure than those who rely solely on the market for food, after controlling for other determinants of households food (in) security. Thus, we expect γ to be positive and statistically different from zero. As mentioned earlier we attempt gauging the possible sensitivity of the estimates to survey dates. First, equation (1) is re-estimated using as dependent variable the response to the survey question ‘Over the past year, how often, if ever, have you or your household gone without enough food to eat?’ Here, a household is defined as food secure (FSi = 1) if the answer to this question is never; the household is food insecure otherwise. Similarly, equation (2) is re-estimated with the dependent variable being the full set of responses to the same question above, where the choice possibilities are: always (FSSi = 1), many times (FSSi = 2), several times (FSSi = 3), just once or twice (FSSi = 4), and never (FSSi = 5). However, in the actual estimation FSSi took on only the last three values because 98 % of observations were in this range. Results and Discussion Descriptive statistics Figure 3 provides a distribution of the food insecurity prevalence indicator (i.e. the HFIAP) among the sample in the two cities. In both cities, more than half of all households (52% for Techiman and 53% for Tamale) fell in one of three food insecurity states, the rest had secure access to food. Table 1 presents the descriptive statistics for the models in equations 1 and 2. The four categories of urban households based on participation in food production are ownproduction_none, ownproduction_urban, ownproduction_rural, and ownproduction_urban_rural. By our main hypothesis, one would expect the proportion of ownproduction_urban_rural households to be increasing significantly as one moves from the severely food insecure state (Severe) to the food secure state (Secure). Conversely, it is expected that the proportion of ownproduction_none households would increase across the scale. Indeed, the results show that the proportion of ownproduction_none households is decreasing across the three food insecurity states (Severe, Moderate, and Mild), approximately 60 %, 57 % and 49 % respectively. A similar pattern is not observed for ownproduction_urban_rural households, but we see that a greater proportion of them are food secure (about 6 %) than are severely food insecure (3 %). Overall, there is a statistically significant difference in food (in) security status by type of participation in food production (χ2 = 27.69, p = 0.001). Nearly half (49%) of the other variables show statistically significant differences across the food security scale at the 5 % level. For example, household head’s years of schooling (years_schooling_hh_head) is significantly increasing as one moves from Severe to Secure on the scale. Similarly, the proportion of educated mothers is increasing across the scale. For instance, only about 11 % of mothers are educated among the severely food insecure whereas among the food secure about 20 % are educated.2 As could be expected, there are significant wealth differences by the HFIAS scale. For example, monthly per capita income is increasing significantly as one moves from the Severe status (USD 51.23) to Secure status (USD 107.39).3 2 We define an educated mother here as one who has attained secondary education or higher. 3 At the time of the survey the US dollar was exchanging for about 2.03 GH¢ (the GH¢ is the local currency unit of Ghana). 10 Techiman Tamale 60% 50% 47% 45% 50% 48% 40% 35% 40% 30% 30% 25% 23% 20% 20% 19% 20% 20% 15% 13% 10% 10% 10% 5% 0% 0% Fig. 3 Food insecurity severity scale across the two cities Food security scale Severe Moderate Mild Secure (n = 393) (n = 425) (n = 236) (n = 950) variable Mean SD Mean SD Mean SD Mean SD p-val. Participation in agriculture: ownproduction_none 59.5% 57.4% 49.2% 56.4% 0.08 ownproduction_urban 20.9% 17.9% 22.5% 21.7% 0.38 ownproduction_rural 16.5% 21.9% 21.2% 15.4% 0.01 ownproduction_urban_rural 3.1% 2.8% 7.2% 6.5% 0.00 female_headed_hh 29.8% 24.2% 17.8% 19.2% 0.00 age_of_hh_head 45.2 15.5 44.6 14.1 45.8 14.5 44.1 13.9 0.31 married_hh_head 68.4% 76.2% 83.1% 76.4% 0.00 years_schooling_hh_head 4.4 4.7 5.3 4.9 5.7 5.4 6.4 5.3 0.00 years_schooling_mother 3.4 3.2 4.5 3.7 3.5 3.6 5.3 4.2 0.00 other_educated_adult_present 46.8% 45.9% 46.6% 45.5% 0.97 age_of_mothers 39.7 12.8 38.5 11.2 37.3 11.7 37.9 12.1 0.04 educated_mother 11.2% 13.2% 14.4% 19.7% 0.00 hh_size 4.7 2.6 4.9 2.3 4.8 2.3 4.7 2.6 0.59 members_under15 1.7 1.6 1.8 1.5 1.6 1.4 1.7 1.5 0.37 members_above64 0.2 0.5 0.2 0.4 0.2 0.4 0.2 0.4 0.47 working_age_members 2.7 1.8 2.8 1.5 2.9 1.6 2.8 1.6 0.44 dependant_proportion 37.1% 25.6% 37.6% 23.8% 34.8% 22.9% 35.1% 24.5% 0.22 under5_child_present 43.5% 49.6% 43.2% 44.9% 0.24 nuclear_hh 56.0% 57.6% 60.2% 54.6% 0.42 cash_transfer_dummy 19.6% 21.2% 11.8% 47.4% 0.06 Food transfers: no_food_transfer 71.0% 72.0% 67.8% 67.3% 0.26 food_transfer_from_urban 12.5% 11.3% 14.4% 12.9% 0.69 food_transfer_from_rural 14.8% 14.8% 15.7% 16.8% 0.71 food_transfer_from_both 1.8% 1.9% 2.1% 2.9% 0.49 nonfarm_income_earners 1.4 0.9 1.6 1.1 1.6 1.0 1.6 1.0 0.00 livestock_index 0.5 0.9 0.4 0.8 0.6 1.0 0.5 0.9 0.05 household_income_pc 104 116 105 122 147 202 218 282 0.00 own_house 28.5% 27.3% 30.9% 34.1% 0.04 Shocks: chronically_sick_head 22.9% 14.4% 20.8% 12.3% 0.00 chronically_sick_partner 12.0% 10.1% 7.6% 8.6% 0.19 chronically_sick_hh_member 14.0% 14.6% 17.4% 10.7% 0.02 Indeptedness 8.9% 9.4% 12.3% 11.6% 0.33 unemployed_head 10.2% 7.1% 2.5% 3.6% 0.00 unemployed_spouse 10.2% 7.5% 6.8% 6.7% 0.17 unemployed_adult 23.2% 16.2% 16.9% 13.3% 0.00 City: Techiman 19.4 22.5 10.3 47.8 0.14 Tamale 19.9 19.9 13.3 47.0 1111 Note: The p-values are based on chi-squared tests for discrete variables, and ANOVA for the roughly continuous variables. Table 1 Sample descriptive statistics, by food (in) security scale Discussion of regression results Fitting one model for both cities assumes that all coefficients do not vary between the two cities. This is a strong assumption given the differing characteristics of the two cities in terms of size and agroecology, for instance. We employed a likelihood-ratio test to formally test the null hypothesis that the coefficients of our models do not differ significantly across the two cities. The null hypothesis is rejected in all cases at the 1 % level, so for each case we fit a model to the full sample as well as city- specific models. Combined Techiman Tamale VARIABLES Coef. AME Coef. AME Coef. AME Participation in agriculture (ref. is ownproduction_urban_rural): ownproduction_none –0.424*** –0.153 0.010 0.004 –0.546*** –0.190 ownproduction_urban –0.355** –0.128 –0.024 –0.008 –0.354* –0.123 ownproduction_rural –0.503*** –0.181 –0.070 –0.025 –0.688*** –0.239 female_headed_hh –0.399*** –0.144 –0.740*** –0.260 –0.192 –0.066 age_of_hh_head 0.003 0.001 0.014** 0.005 –0.005 –0.002 married_hh_head 0.050 0.018 0.034 0.012 0.084 0.029 age_of_mother –0.002 –0.001 –0.010 –0.003 0.002 0.001 years_schooling_hh_head –0.006 –0.002 –0.014 –0.005 –0.002 –0.001 years_schooling_mother 0.064*** 0.023 0.134*** 0.047 0.023* 0.008 other_educated_adult_present 0.045 0.016 –0.038 –0.013 0.168 0.058 hh_size –0.059 –0.021 0.091 0.032 –0.155 –0.054 members_under15 0.091 0.033 –0.078 –0.028 0.204 0.071 members_above64 0.165 0.059 0.064 0.022 0.265 0.092 working_age_members 0.078 0.028 –0.015 –0.005 0.135 0.047 under5_child_present 0.003 0.001 0.042 0.015 –0.015 –0.005 nuclear_hh –0.216*** –0.078 –0.121 –0.043 –0.319*** –0.111 cash_transfer_dummy 0.024 0.009 0.122 0.043 –0.107 –0.037 Food transfers (ref. is food_transfer_from_both): no_food_transfer –0.430** –0.155 –0.307 –0.108 –0.492* –0.170 food_transfer_from_urban –0.403* –0.145 –0.246 –0.086 –0.640** –0.222 food_transfer_from_rural –0.321 –0.116 –0.194 –0.068 –0.425 –0.147 nonfarm_income_earners 0.023 0.008 –0.055 –0.019 0.096* 0.033 livestock_index –0.046 –0.016 –0.045 –0.016 –0.067 –0.023 own_house 0.130** 0.047 0.030 0.011 0.183* 0.063 household_income_pc_100 0.153*** 0.055 0.126*** 0.044 0.202*** 0.070 chronically_sick_head –0.216** –0.078 –0.176 –0.062 –0.237* –0.082 chronically_sick_partner –0.086 –0.031 –0.178 –0.063 0.046 0.016 chronically_sick_hh_member –0.185** –0.066 –0.191 –0.067 –0.165 –0.057 Indeptedness –0.065 –0.024 –0.279** –0.098 0.268 0.093 unemployed_head –0.354** –0.127 –0.534*** –0.188 –0.241 –0.084 unemployed_spouse –0.174 –0.063 –0.397** –0.140 0.015 0.005 unemployed_adult –0.204** –0.074 –0.061 –0.022 –0.329** –0.114 Techiman (ref. is Tamale) 0.010 0.004 Intercept 0.281 –0.581 0.654 Observations 2,004 1,008 996 Pseudo R-squared 0.093 0.114 0.125 Log-likelihood value –1,257 –618.1 –602.4 Note: AME denotes Average Marginal Effect *** p < 0.01, ** p < 0.05, * p < 0.1 Table 2 Probit estimates of urban household participation in different types of agriculture and the probability of being food secure Table 2 reports parameter estimates from the probit model (equation 1). With ownproducti on_urban_rural being the reference category for the main explanatory variable of interest, we expect ownproduction_none, ownproduction_urban, and ownproduction_rural to be negative and statistically different from zero, based on our hypothesis. Ceteris paribus, our null hypothesis is at the 1% level of 12 significance for ownproduction_none and ownproduction_rural and at the 5% level for ownproduction_urban, implying that, households who participate in both types of agriculture (urban and rural) have better food security outlook relative to all the other three types of households. Other important determinants of food security include female headed households who are found to be less food secure relative to male headed households and the number of years of schooling of mothers which is found to positively correlate with household food security level. That is, an additional year of schooling of mothers increased the likelihood of households being food secure. Generally, women provide care to members of the household as well as manage the home. As such, a sizeable proportion of decisions are made by them regarding dietary and food requirements of household members. In addition, additional year of schooling is generally tied to increased income earning opportunities. This tied to the roles of women in the home explains why education of mothers can improve the food security situation of household members. A prior, it was expected that, income, which is an important determinant of food security, at least in the cities would have a positive effect on our food security indictor. Our parameter estimates suggests that, average monthly income per capita has a positive effect on household food security. In other words, increases in income has the potential of improving the food security status of the household. Combined Techiman Tamale VARIABLES Coef. AME Coef. AME Coef. AME Participation in agriculture (ref. is ownproduction_urban_rural): ownproduction_none –0.405** –0.100 –0.358 –0.087 –0.306 –0.074 ownproduction_urban –0.408** –0.100 –0.385 –0.095 –0.360 –0.088 ownproduction_rural –0.334* –0.080 –0.321 –0.077 –0.320 –0.077 female_headed_hh –0.254** –0.071 –0.348** –0.097 –0.229 –0.061 age_of_hh_head 0.009* 0.002 0.019*** 0.005 –0.001 –0.000 married_hh_head 0.166 0.047 0.155 0.043 0.240 0.064 age_of_mother –0.009 –0.002 –0.016** –0.005 –0.003 –0.001 years_schooling_hh_head –0.012 –0.003 –0.016 –0.005 –0.009 –0.002 years_schooling_mother 0.064*** 0.018 0.107*** 0.030 0.033** 0.009 other_educated_adult_present –0.032 –0.009 –0.026 –0.007 –0.037 –0.010 hh_size –0.116 –0.032 0.163 0.045 –0.436*** –0.116 members_under15 0.168* 0.047 –0.131 –0.036 0.514*** 0.137 members_above64 0.243** 0.068 –0.019 –0.005 0.616*** 0.164 working_age_members 0.142* 0.040 –0.055 –0.015 0.421*** 0.112 under5_child_present –0.101 –0.028 –0.119 –0.033 –0.033 –0.009 nuclear_hh –0.175** –0.049 –0.057 –0.016 –0.275** –0.073 cash_transfer_dummy –0.015 –0.004 –0.057 –0.016 0.055 0.015 Food transfers (ref. is food_transfer_from_both): no_food_transfer –0.565** –0.158 –0.319 –0.088 –0.686** –0.182 food_transfer_from_urban –0.480* –0.134 –0.169 –0.047 –0.754** –0.201 food_transfer_from_rural –0.533** –0.149 –0.374 –0.103 –0.566 –0.151 nonfarm_income_earners 0.015 0.004 –0.079 –0.022 0.089 0.024 livestock_index 0.060 0.017 –0.041 –0.011 0.148** 0.039 own_house 0.102 0.028 –0.078 –0.022 0.278** 0.074 household_income_pc_100 0.121*** 0.034 0.077** 0.021 0.200*** 0.053 chronically_sick_head –0.087 –0.024 –0.233* –0.064 0.073 0.019 chronically_sick_partner –0.113 –0.032 –0.191 –0.053 –0.064 –0.017 chronically_sick_hh_member –0.209** –0.058 –0.181 –0.050 –0.227 –0.060 Indeptedness 0.016 0.005 0.141 0.039 –0.166 –0.044 unemployed_head –0.359*** –0.100 –0.617*** –0.171 –0.191 –0.051 unemployed_spouse –0.199* –0.056 –0.431** –0.119 0.047 0.013 unemployed_adult –0.198** –0.055 –0.157 –0.043 –0.213 –0.057 Techiman (ref. is Tamale) 0.008 0.002 Intercept 1.209*** 0.713 1.251** Observations 2,004 2,004 1,008 1,008 996 996 Pseudo R-squared 0.073 0.086 0.113 Log-likelihood value –995.4 –496.0 –470.7 Note: AME denotes Average Marginal Effect *** p < 0.01, ** p < 0.05, * p < 0.1 Table 3 Probit estimates of urban household participation in agriculture and the probability of being food secure 1133 Fitting the model with the same covariates at the city level shows similar pattern in terms of effects of agricultural production on food security for Tamale city but not Techiman. City-specific effects matter in the sense that, as alluded to early on, Tamale metropolitan serves as the regional capital of the northern region of Ghana, and home to several type of businesses and individuals who have the taste and preference for exotic vegetables and thus provide market access for the farmers. Additionally, a cultural phenomenon encourages the cultivation of crops (especially vegetables that are central in the preparation of traditional diets) around homes. This practice potentially explains why urban agriculture is more pronounced in Tamale apart from other contextual variables such as access to market, the general lack of space for other forms of agricultural production around the city, and the harsh climatic conditions that allows for only one season of production. In Techiman on the other hand, proximity to rural areas where land for farming abounds makes farming in urban space unattractive in addition to the fact that, urban agriculture is a relatively new concept in the city. Qualitative interviews on the institutional frameworks guiding the practice of urban agriculture in both cities suggest that, urban agriculture in Tamale is considered an integral source of vegetables to the city and thus receives extension support from the Metropolitan Assembly’s Department of Agriculture. On the contrast, urban agriculture, though exist in Techiman, is considered alien to modern city development by both municipal authorities and local opinion leaders. To check for the robustness of our results, we estimate equation 1 again using as the dependent variable ownprod (a variable that takes the value of 1 if the household produces their own food irrespective of locality (either in urban or rural space) (Table 3). Whereas the combined effect of participation in both urban and rural agriculture remains unchanged for the sample, the city level effects are lost. That is, reclassifying households as agricultural or otherwise has no effect on household food security at least at the city level. This further corroborate the need to consider effect of own production in specific contexts. Our ordered choice regression estimates which uses the food security severity indicator as dependent variable (equation 2) shows that, households engaged in both urban and rural agriculture have better food security status relative to those who have no production at all and those who engage in only UA or only rural agriculture (Table 4). This we find to be consistent with our aprior expectations and significant at the 1% level. The city specific models show similar results for Tamale with households engaged in both urban and rural agriculture having better food security status relative to the other household types. Per contra, we find no statistically significant difference between households that practice both urban and rural agriculture and the other household types in Techiman. 14 15 Combined Techiman Tamale Average Marginal Effect Average Marginal Effect Average Marginal Effect VARIABLES Coef. FSS = 1 FSS = 2 FSS = 3 FSS = 4 Coef. FSS = 1 FSS = 2 FSS = 3 FSS = 4 Coef. FSS = 1 FSS = 2 FSS = 3 FSS = 4 Participation in agriculture (ref. is ownproduction_urban_rural): ownproduction_none –0.438*** 0.096 0.052 0.010 –0.158 –0.113 0.026 0.012 0.001 –0.039 –0.550*** 0.116 0.061 0.015 –0.192 ownproduction_urban –0.370*** 0.078 0.045 0.010 –0.133 –0.182 0.043 0.019 0.001 –0.063 –0.328* 0.062 0.039 0.013 –0.114 ownproduction_rural –0.436*** 0.095 0.052 0.010 –0.157 –0.133 0.031 0.014 0.001 –0.046 –0.562*** 0.120 0.062 0.015 –0.196 female_headed_hh –0.340*** 0.087 0.034 0.002 –0.122 –0.670*** 0.164 0.067 0.002 –0.233 –0.126 0.031 0.011 0.001 –0.044 age_of_hh_head 0.007* –0.002 –0.001 –0.000 0.003 0.019*** –0.005 –0.002 –0.000 0.007 0.000 –0.000 –0.000 –0.000 0.000 married_hh_head 0.146* –0.037 –0.015 –0.001 0.053 0.133 –0.033 –0.013 –0.000 0.046 0.152 –0.038 –0.014 –0.001 0.053 age_of_mother –0.005 0.001 0.001 0.000 –0.002 –0.013* 0.003 0.001 0.000 –0.005 –0.004 0.001 0.000 0.000 –0.001 years_schooling_hh_head 0.001 –0.000 –0.000 –0.000 0.000 –0.010 0.002 0.001 0.000 –0.004 0.006 –0.002 –0.001 –0.000 0.002 years_schooling_mother 0.060*** –0.015 –0.006 –0.000 0.022 0.141*** –0.035 –0.014 –0.000 0.049 0.015 –0.004 –0.001 –0.000 0.005 other_educated_adult_present 0.035 –0.009 –0.003 –0.000 0.013 –0.003 0.001 0.000 0.000 –0.001 0.108 –0.027 –0.010 –0.001 0.038 hh_size –0.027 0.007 0.003 0.000 –0.010 0.159* –0.039 –0.016 –0.001 0.055 –0.201 0.050 0.018 0.001 –0.070 members_under15 0.040 –0.010 –0.004 –0.000 0.014 –0.136 0.033 0.014 0.000 –0.047 0.200 –0.050 –0.018 –0.001 0.070 members_above64 0.083 –0.021 –0.008 –0.000 0.030 –0.072 0.018 0.007 0.000 –0.025 0.293 –0.073 –0.027 –0.002 0.102 working_age_members 0.056 –0.014 –0.006 –0.000 0.020 –0.093 0.023 0.009 0.000 –0.032 0.205 –0.051 –0.019 –0.001 0.071 under5_child_present 0.016 –0.004 –0.002 –0.000 0.006 0.017 –0.004 –0.002 –0.000 0.006 0.053 –0.013 –0.005 –0.000 0.018 nuclear_hh –0.231*** 0.059 0.023 0.001 –0.083 –0.087 0.021 0.009 0.000 –0.030 –0.372*** 0.093 0.034 0.002 –0.129 cash_transfer_dummy –0.010 0.002 0.001 0.000 –0.003 –0.018 0.004 0.002 0.000 –0.006 –0.053 0.013 0.005 0.000 –0.018 Food transfers (ref. is food_transfer_from_both): no_food_transfer –0.400** 0.102 0.040 0.002 –0.144 –0.188 0.046 0.019 0.001 –0.066 –0.480* 0.120 0.044 0.003 –0.167 food_transfer_from_urban –0.344* 0.088 0.034 0.002 –0.124 –0.137 0.034 0.014 0.000 –0.048 –0.540** 0.135 0.049 0.003 –0.188 food_transfer_from_rural –0.305 0.078 0.030 0.002 –0.110 –0.092 0.023 0.009 0.000 –0.032 –0.407 0.102 0.037 0.003 –0.141 nonfarm_income_earners 0.016 –0.004 –0.002 –0.000 0.006 –0.038 0.009 0.004 0.000 –0.013 0.077 –0.019 –0.007 –0.000 0.027 livestock_index –0.038 0.010 0.004 0.000 –0.014 –0.018 0.004 0.002 0.000 –0.006 –0.071 0.018 0.006 0.000 –0.025 own_house 0.094 –0.024 –0.009 –0.001 0.034 0.032 –0.008 –0.003 –0.000 0.011 0.099 –0.025 –0.009 –0.001 0.034 household_income_pc_100 0.147*** –0.037 –0.015 –0.001 0.053 0.123*** –0.030 –0.012 –0.000 0.043 0.195*** –0.049 –0.018 –0.001 0.068 chronically_sick_head –0.192** 0.049 0.019 0.001 –0.069 –0.155 0.038 0.015 0.001 –0.054 –0.209* 0.052 0.019 0.001 –0.073 chronically_sick_partner –0.170* 0.043 0.017 0.001 –0.061 –0.188 0.046 0.019 0.001 –0.066 –0.117 0.029 0.011 0.001 –0.041 chronically_sick_hh_member –0.095 0.024 0.009 0.001 –0.034 –0.097 0.024 0.010 0.000 –0.034 –0.093 0.023 0.009 0.001 –0.032 Indeptedness –0.038 0.010 0.004 0.000 –0.014 –0.281** 0.069 0.028 0.001 –0.098 0.343** –0.086 –0.031 –0.002 0.119 unemployed_head –0.398*** 0.101 0.039 0.002 –0.143 –0.546*** 0.134 0.054 0.002 –0.190 –0.297* 0.074 0.027 0.002 –0.103 unemployed_spouse –0.230** 0.059 0.023 0.001 –0.083 –0.469*** 0.115 0.047 0.002 –0.163 –0.007 0.002 0.001 0.000 –0.002 unemployed_adult –0.264*** 0.067 0.026 0.002 –0.095 –0.154 0.038 0.015 0.001 –0.054 –0.372*** 0.093 0.034 0.002 –0.129 Techiman 0.005 –0.001 –0.001 –0.000 0.002 Constant cut1 –1.152*** –0.221 –1.655*** Constant cut2 –0.468 0.533 –0.992** Constant cut3 –0.140 0.829* –0.610 Observations 2,004 1,008 996 Pseudo R-squared 0.060 0.081 0.079 Log-likelihood value –2362 –1149 –1160 *** p < 0.01, ** p < 0.05, * p < 0. Table 4 Ordered probit estimates of urban household participation in agriculture and food insecurity in Techiman and Tamale 15 Estimating the model using the ordered variable that we believe is less sensitive to survey dates decreases the precision of our estimates. We find household that participate in both urban and rural agriculture showing statistically significantly better food security outlook than households who do not participate and those who produce in urban space at the 10% level of significance (Table A2). No statistically significant difference is found between participation in rural agriculture and participation in both urban and rural agriculture. Thus, it appears timing of the surveys matter for the results observed. Furthermore, we reclassify our households into agricultural and non-agricultural households and estimate the effects of participation in agriculture on household food security. The results are summarised in Appendix Table A4 (full estimates are available on request). In general, we find no statistically significant difference in the food security status among households that participate in own food production and those who do not. This finding is inconsistent with our a prior expectation but provides an indication that, own food production and food security need not be generalised but be contextualised to the specific type of agricultural production a particular household is engaged in. As discussed earlier, some important predictors of urban household food security include household characteristics such as the number of years of schooling of mother(s), income, and the marital status of household head. Female headed households were generally found to be less likely to be food secure, a result that is consistent with other studies using Ghana data (e.g. Dzanku and Sarpong, 2011). Income as a predictor of household food security suggests ability to access food on the market. Mother’s education (years) on the other hand correlates with higher socioeconomic status and incomes. Although these findings are consistent with the literature, (see for example Sen, 1981; Dzanku and Sarpong, 2011; Ibekwe et al., 2010) we argue that, food is a basic necessity that should be accessible to every person. As a result, income and education should not be important predictors of household food security, more so in urban settings. However from the policy perspective, this finding is relevant to guiding policy formulation in the food security discussions. Compared with households who receive food remittances from both urban and rural areas, both those that do not receive food remittances and those that received food remittances from urban areas only are less likely to be food secure. The health status of the household (i.e. whether there is a chronically sick member) and the employment status of working adults, particularly, the household head and spouse also significantly predict household food (in) security. Conclusion This paper contributes to the debate on urban agriculture’s potential contribution(s) to household food security by examining the association between urban households’ agriculture production in urban and/or rural space on the one hand and food security on the other. The paper brings a new dimension to the debate by departing from the focus on urban agriculture only (i.e. agricultural production in urban space only) to include all agricultural production options available to households. Additionally, the study is contextualised within intermediate sized cities, breaking away from the metropolitan biases that has characterised urban agriculture studies in the past. We conclude from our results that, urban households’ participation in both urban and rural agriculture affects their food security situation. Neither agriculture production in urban nor rural space alone has a significant effect on urban household food security. Our robustness check further validates the result that the combined effect of participation in both urban and rural agriculture significantly affects household food security although we do not find a significant city-specific effects. Reclassifying households into agricultural and non-agricultural households, we find participation in agriculture to have no effect on household food security across survey cities. As a result, we argue that urban household food security should not be limited to food production in urban space only but be broadened to include all own production opportunities available to the household including production of food in rural areas. 16 The findings should be interpreted in context – city location is important in the debate on the effects of agriculture on household food security. Our results confirm, for example, that in Tamale, urban agriculture contributes substantially to household food security. This result is attributable to the cosmopolitan nature of the city of Tamale which creates ready market for freshly produced vegetables creating an incentive for urban farming. Additionally, the harsh climatic conditions makes farming impossible without irrigation and thus push farmers to pursue urban farming in a more modern manner than in Techiman. On the contrary, urban households’ proximity to rural areas in Techiman makes farming in neighbouring rural communities more convenient than urban farming, hence the lack of a significant urban agriculture effect on household food security in Techiman. It is thus recommended that policies and discussion aimed at addressing urban poverty through agriculture should be city- specific. In addition, urban and rural agriculture policies should be complementary and not be treated as mutually exclusive because the synergies could be harnessed in specific contexts as an effective poverty alleviating strategy. Conflict of interest None References Aidoo, Robert, Mensah, James Osei, and Tuffour, Thomas (2013), 'Determinants of household food security in the Sekyere-Afram plains district of Ghana', European Scientific Journal. Anyane, S. L. (1963), 'Vegetable gardening in Accra', The Ghana Farmer, 1 (6), 228-30. Armar-Klemesu, M. (2000), 'Urban Agriculture and Food Security, Nutrition and Health', in Dubbeling M Bakker N, Gundel S, Sabel-Koschella U, de Zeeuw H (ed.), Growing Cities, Growing Food, Urban Agriculture on the Policy Agenda (Feldafing, Germany: DSE Feldafing). Ayerakwa, Hayford Mensah (2017), 'Urban households' engagement in agriculture: implications for household food security in Ghana's medium sized cities', Geographical Research. Babatunde, Raphael O and Qaim, Matin (2010), 'Impact of off-farm income on food security and nutrition in Nigeria', Food policy, 35 (4), 303-11. Barrett, Christopher B (2010), 'Measuring food insecurity', Science, 327 (5967), 825-28. Bashir, Muhammad Khalid, Schilizzi, Steven, and Pandit, Ram, 2012. The determinants of rural household food security in the Punjab, Pakistan: an econometric analysis. Available at: (accessed Becquey, Elodie, Martin-Prevel, Yves, Traissac, Pierre, Dembélé, Bernard, Bambara, Alain, and Delpeuch, Francis (2010), 'The household food insecurity access scale and an index-member dietary diversity score contribute valid and complementary information on household food insecurity in an urban West-African setting', The Journal of nutrition, 140 (12), 2233-40. Boayke-Yiadom, L and McKay, Andrew (2007), 'Rural-urban linkages and welfare: the case of remittance flows in Ghana', Livelihoods and well-being in developing countries: Multidisciplinary approaches Conference, St Catherine’s College, University of Oxford, 18-20. Chagomoka, T., Drescher, A., Glaser, R., Marschner, B., Schlesinger, J., and Nyandoro, G. (2015), 'Vegetable production, consumption and its contribution to diets along the urban–rural continuum in Northern Ghana', African Journal of Food, Agriculture, Nutrition and Development, 15 (4), 10352-67. 1177 Chagomoka, T., Unger, S., Drescher, A., Glaser, R., Marschner, B., and Schlesinger, J. (2016), 'Food coping strategies in northern Ghana. A socio-spatial analysis along the urban–rural continuum', Agriculture & Food Security, 5 (4), 1-18. Coates, Jennifer, Swindale, Anne, and Bilinsky, Paula (2007), 'Household Food Insecurity Access Scale (HFIAS) for measurement of food access: indicator guide', Washington, DC: Food and Nutrition Technical Assistance Project, Academy for Educational Development. Cofie, O. and Drechsel, P. (2007), 'Water for food in the cities: the growing paradigm of irrigated (peri)-urban agriculture and its struggle in sub-Saharan Africa', African Water Journal, 1 (1), 26-50. Crush, J., Hovorka, A., and Tevera, D. (2011), 'Food security in southern African cities: the place of urban agriculture', Progress in Development Studies, 11 (4), 285-305. Danso, G., Drechsel, P., Wiafe-Antwi, T., and Gyiele, L. (2002), 'Income of farming systems around Kumasi, Ghana', Urban Agriculture Magazine, 7, 5-6. Danso, G., Drechsel, P., Obuobie, E., Forkuor, G., and Kranjac-Berisavljevic, G. (2014), 'Urban vegetable farming sites, crops and cropping practices', in P. and Keraita Drechsel, B. (ed.), Irrigated urban vegetable production in Ghana: characteristics, benefits and risk mitigation (Colombo, Sri Lanka: International Water Management Institute), 7-27. de Haan, L. and Zoomers, A. (2003), 'Development geography at the crossroads of livelihood and globalization', Tijdschrift voor Economische en Sociale Geografie, 4 (3), 350-62. Drechsel, Pay, Graefe, Sophie, Sonou, Moise, and Cofie, Olufunke O (2006), Informal irrigation in urban West Africa: An overview (102: IWMI). Dzanku, F and Sarpong, DANIEL (2010), 'Agricultural diversification, food self-sufficiency and food security in Ghana–the role of infrastructure and institutions', African Smallholders: Food Crops, Markets and Policy. CABI, Wallingford, UK, 189-213. FAO, 1996. World Food Summit. Food and Agriculture Organization, Rome. Available at: (accessed FAO (2012), Growing greener cities in Africa: first status report on urban and peri-urban horticulture in Africa (Rome: FAO). Feleke, Shiferaw T, Kilmer, Richard L, and Gladwin, Christina H (2005), 'Determinants of food security in Southern Ethiopia at the household level', Agricultural Economics, 33 (3), 351-63. Foeken, Dick WJ and Owuor, Samuel O (2008), 'Farming as a livelihood source for the urban poor of Nakuru, Kenya', Geoforum, 39 (6), 1978-90. Frayne, B., McCordic, C. , and Shilomboleni, H. (2014), 'Growing Out of Poverty: Does Urban Agriculture Contribute to Household Food Security in Southern African Cities? ', Urban Forum, 25, 177-89. Garrett, James L and Ruel, Marie T (1999), 'Are determinants of rural and urban food security and nutritional status different? Some insights from Mozambique', World Development, 27 (11), 1955-75. Gebreyesus, Seifu Hagos, Lunde, Torleif, Mariam, Damen H, Woldehanna, Tasew, and Lindtjørn, Bernt (2015), 'Is the adapted Household Food Insecurity Access Scale (HFIAS) developed internationally to measure food insecurity valid in urban and rural households of Ethiopia?', BMC Nutrition, 1 (1), 1. Girdner, Janet, Olorunsola, Victor, Froning, Myrna, and Hansen, Emmanuel (1980), 'Ghana's agricultural food policy: Operation feed yourself', Food Policy, 5 (1), 14-25. Gockowski, J., Mbazo’o, J., Mbah, G., & Fouda Moulende, T. (2003). African traditional leafy vegetables and the urban and peri-urban poor. Food policy, 28(3), 221-235. GSS, 2014a. 2010 Population and Housing Census: District Analytical Report, Techiman. Ghana Statistical Services, Accra, Ghana. GSS, 2014b. 2010 Population and Housing Census: District Analytical Report, Tamale. . Ghana Statistical Services, Accra, Ghana. Hendriks, Sheryl L (2015), 'The food security continuum: a novel tool for understanding food insecurity as a range of experiences', Food Security, 7 (3), 609-19. Henn, Patrick (2000), 'User benefits of urban agriculture in Havana, Cuba: An application of the contingent valuation method', (McGill University). Ibekwe, UC (2010), 'Determinants of income among farm households in Orlu Agricultural Zone of Imo State, Nigeria', Report and Opinion, 2 (8), 32-35. Krüger, Fred (1998), 'Taking advantage of rural assets as a coping strategy for the urban poor: The case of ruralurban interrelations in Botswana', Environment and Urbanization, 10 (1), 119-34. Lee-Smith, D. (2013), 'Which way for UPA in Africa? ', City, 17 (1), 69-84. 18 Lee-Smith, Diana (2010), 'Cities feeding people: an update on urban agriculture in equatorial Africa', Environment and Urbanization, 22 (2), 483-99. Maxwell, DG (2001), 'The importance of urban agriculture to food and nutrition. Annotated biography, ETC- RUAF', (CTA publishers, Leusden, Netherlands). Mkwambisi, David D, Fraser, Evan DG, and Dougill, Andy J (2011), 'Urban agriculture and poverty reduction: evaluating how food production in cities contributes to food security, employment and income in Malawi', Journal of International Development, 23 (2), 181-203. Mkwambisi, David Dalison (2007), 'Urban agiriculture in Malawi: poverty reduction, waste management and institutional barriers', (University of Leeds). MoFA, 2010. Medium Term Agriculture Development Plan (METASIP) 2011-2015. Ministry of Food and Agriculture Accra. Available at: (accessed Mougeot, LJA (2000), 'Achieving urban food and nutrition security in developing countries: The hidden significance of urban agriculture', IFPRI, Brief paper, (6). Mougeot, Luc JA (2011), 'International Support to Research and Policy on Urban Agriculture (1996–2010): achievements and challenges', International Development Research Centre, Ottawa. Mwangi, ALICE M and Foeken, Dick (1996), 'Urban agriculture, food security and nutrition in low income areas in Nairobi', African Urban Quarterly, 11 (2/3), 170-79. Nugent, Rachel (2000), 'The impact of urban agriculture on the household and local economies', Bakker N., Dubbeling M., Gündel S., Sabel-Koshella U., de Zeeuw H. Growing cities, growing food. Urban agriculture on the policy agenda. Feldafing, Germany: Zentralstelle für Ernährung und Landwirtschaft (ZEL), 67-95. Obosu-Mensah, Kwaku (1999), Food production in urban areas: a study of urban agriculture in Accra, Ghana (Ashgate Publishing Ltd.). Owusu, Victor, Abdulai, Awudu, and Abdul-Rahman, Seini (2011), 'Non-farm work and food security among farm households in Northern Ghana', Food Policy, 36 (2), 108-18. Pinstrup-Andersen, Per (2009), 'Food security: Definition and measurement', Food Security, 1 (1), 5-7. Potts, Deborah and Mutambirwa, Chris (1997), 'The government must not dictate’: rural‐urban migrants’ perceptions of Zimbabwe's land resettlement programme', Review of African Political Economy, 24 (74), 549-66. Sabates, Ricardo, Gould, Brian W, and Villarreal, Hector J (2001), 'Household composition and food expenditures: a cross-country comparison', Food Policy, 26 (6), 571-86. Salarkia, Nahid, Abdollahi, Morteza, Amini, Maryam, and Neyestani, Tirang Reza (2014), 'An adapted Household Food Insecurity Access Scale is a valid tool as a proxy measure of food access for use in urban Iran', Food Security, 6 (2), 275-82. Sanyal, Bishwapriya (1985), 'Urban agriculture: who cultivates and why? A case study of Lusaka, Zambia', Food and Nutrition Bulletin, 7 (3), 15-24. Sen, Amartya (1981), 'Ingredients of famine analysis: availability and entitlements', The quarterly journal of economics, 433-64. Smit, Jac, Nasr, Joe, and Ratta, Annu (1996), 'Urban agriculture: food, jobs and sustainable cities', New York, USA, 2, 35-37. Stewart, R., Korth, M., Langer, L., Rafferty, S., Da Silva, N., and van Rooyen, C. (2013), 'What are the impacts of urban agriculture programs on food security in low and middle–income countries?', Environmental Evidence 2(7), 1-13. Talukder, Dayal (2014), 'Assessing Determinants of Income of Rural Households in Bangladesh: A Regression Analysis', Journal of Applied Economics and Business Research, 4 (2), 80-106. Thebo, AL, Drechsel, Pay, and Lambin, EF (2014), 'Global assessment of urban and peri-urban agriculture: irrigated and rainfed croplands', Environmental Research Letters, 9 (11), 114002. Webb, Patrick, Coates, Jennifer, Frongillo, Edward A, Rogers, Beatrice Lorge, Swindale, Anne, and Bilinsky, Paula (2006), 'Measuring household food insecurity: why it's so important and yet so difficult to do', The Journal of nutrition, 136 (5), 1404S-08S. World Bank (2016), World Development Indicators, . Zezza, Alberto and Tasciotti, Luca (2010), 'Urban agriculture, poverty, and food security: Empirical evidence from a sample of developing countries', Food policy, 35 (4), 265-73. 1199 Appendix Notation Description Participation in agriculture: ownproduction_none Household does not produce any food ownproduction_urban Household produces food in urban areas only ownproduction_rural Household produces food in rural areas only ownproduction_urban_rural Household produces food in both urban and rural areas female_headed_hh Female headed household age_of_hh_head Age of household head married_hh_head Marital status of household head years_schooling_hh_head Years of schooling of household head years_schooling_mother Years of schooling of mother other_educated_adult_present Number of educated adults in the household age_of_mothers Age of mother educated_mother Mother is educated hh_size Household size members_under15 Number of household members less than 15 years members_above64 Number of household members above 64 years working_age_members Number of household members 15-64 years dependant_proportion Dependent proportion of household members under5_child_present Number of children less than 5 years in the household nuclear_hh Household is nuclear in structure cash_transfer_dummy Cash transfer dummy Food transfers: no_food_transfer Household does not receive food transfer food_transfer_from_urban Household receives food transfers from urban areas food_transfer_from_rural Household receives food transfers from rural areas food_transfer_from_both Household receives food transfers from both urban and rural areas nonfarm_income_earners Household receives income from non-farm activities livestock_index Household livestock index own_house Household own the house in which they live in household_income_pc Household income PC Shocks: chronically_sick_head Household has a chronically sick head chronically_sick_partner Household has a chronically sick partner chronically_sick_hh_member Household has a chronically sick member Indeptedness Household is indebted unemployed_head Household head is unemployed unemployed_spouse Spouse is unemployed unemployed_adult Number of unemployed adults in the household Table A1 Variable notation and description 20 21 Combined Techiman Tamale Average Marginal Effect Average Marginal Effect Average Marginal Effect VARIABLES Coef. OFS = 1 OFS = 2 OFS = 3 Coef. OFS = 1 OFS = 2 OFS = 3 Coef. OFS = 1 OFS = 2 OFS = 3 Participation in agriculture (ref. is ownproduction_urban_rural): ownproduction_none –0.341* 0.047 0.039 –0.086 –0.379 0.055 0.037 –0.092 –0.209 0.026 0.027 –0.053 ownproduction_urban –0.315* 0.043 0.036 –0.079 –0.362 0.052 0.035 –0.087 –0.244 0.031 0.032 –0.062 ownproduction_rural –0.246 0.032 0.028 –0.060 –0.329 0.047 0.032 –0.078 –0.169 0.020 0.022 –0.042 female_headed_hh –0.250** 0.041 0.029 –0.070 –0.363** 0.064 0.036 –0.100 –0.217 0.030 0.028 –0.058 age_of_hh_head 0.008 –0.001 –0.001 0.002 0.017** –0.003 –0.002 0.005 –0.001 0.000 0.000 –0.000 married_hh_head 0.196** –0.032 –0.023 0.054 0.129 –0.023 –0.013 0.036 0.338** –0.046 –0.044 0.090 age_of_mother –0.009* 0.001 0.001 –0.003 –0.018** 0.003 0.002 –0.005 –0.002 0.000 0.000 –0.000 years_schooling_hh_head –0.011 0.002 0.001 –0.003 –0.015 0.003 0.001 –0.004 –0.008 0.001 0.001 –0.002 years_schooling_mother 0.064*** –0.010 –0.007 0.018 0.113*** –0.020 –0.011 0.031 0.029* –0.004 –0.004 0.008 other_educated_adult_present –0.012 0.002 0.001 –0.003 –0.009 0.002 0.001 –0.003 0.001 –0.000 –0.000 0.000 hh_size –0.134* 0.022 0.016 –0.037 0.127 –0.022 –0.012 0.035 –0.392*** 0.054 0.051 –0.105 members_under15 0.182** –0.030 –0.021 0.051 –0.086 0.015 0.008 –0.024 0.451*** –0.062 –0.059 0.120 members_above64 0.291** –0.047 –0.034 0.081 0.100 –0.018 –0.010 0.027 0.566*** –0.078 –0.073 0.151 working_age_members 0.167** –0.027 –0.019 0.046 –0.001 0.000 0.000 –0.000 0.372*** –0.051 –0.048 0.099 under5_child_present –0.072 0.012 0.008 –0.020 –0.088 0.016 0.009 –0.024 –0.005 0.001 0.001 –0.001 nuclear_hh –0.169** 0.027 0.020 –0.047 –0.025 0.005 0.002 –0.007 –0.324*** 0.044 0.042 –0.086 cash_transfer_dummy –0.035 0.006 0.004 –0.010 –0.079 0.014 0.008 –0.022 0.024 –0.003 –0.003 0.006 Food transfers (ref. is food_transfer_from_both): no_food_transfer –0.548** 0.089 0.064 –0.153 –0.308 0.055 0.030 –0.085 –0.657* 0.090 0.085 –0.175 food_transfer_from_urban –0.493** 0.080 0.057 –0.137 –0.199 0.035 0.020 –0.055 –0.758** 0.104 0.098 –0.202 food_transfer_from_rural –0.504** 0.082 0.059 –0.140 –0.314 0.056 0.031 –0.086 –0.567 0.078 0.073 –0.151 nonfarm_income_earners 0.009 –0.001 –0.001 0.002 –0.078 0.014 0.008 –0.021 0.078 –0.011 –0.010 0.021 livestock_index 0.049 –0.008 –0.006 0.014 –0.057 0.010 0.006 –0.016 0.135** –0.019 –0.018 0.036 own_house 0.086 –0.014 –0.010 0.024 –0.045 0.008 0.004 –0.012 0.199* –0.027 –0.026 0.053 household_income_pc_100 0.128*** –0.021 –0.015 0.036 0.088*** –0.016 –0.009 0.024 0.203*** –0.028 –0.026 0.054 chronically_sick_head –0.031 0.005 0.004 –0.009 –0.098 0.017 0.010 –0.027 0.053 –0.007 –0.007 0.014 chronically_sick_partner –0.145 0.024 0.017 –0.040 –0.194 0.034 0.019 –0.053 –0.109 0.015 0.014 –0.029 chronically_sick_hh_member –0.228** 0.037 0.027 –0.064 –0.240* 0.043 0.024 –0.066 –0.211 0.029 0.027 –0.056 indeptedness 0.035 –0.006 –0.004 0.010 0.094 –0.017 –0.009 0.026 –0.058 0.008 0.008 –0.015 unemployed_head –0.430*** 0.070 0.050 –0.120 –0.663*** 0.118 0.065 –0.183 –0.267 0.037 0.035 –0.071 unemployed_spouse –0.215* 0.035 0.025 –0.060 –0.421** 0.075 0.041 –0.116 0.017 –0.002 –0.002 0.005 unemployed_adult –0.199** 0.032 0.023 –0.055 –0.164 0.029 0.016 –0.045 –0.224* 0.031 0.029 –0.060 Techiman –0.043 0.007 0.005 –0.012 Constant cut1 –1.719*** –1.216** –1.808*** Constant cut2 –1.139*** –0.716 –1.106** Observations 2,002 1,006 996 Pseudo R-squared 0.063 0.074 0.091 Log-likelihood value –1294 –646.5 –617.0 *** p < 0.01, ** p < 0.05, * p < 0.1 Table A2 Ordered probit estimates of urban household participation in agriculture and food insecurity in Techiman and Tamale 21 Equation 3a Equation 4a Combined Techiman Tamale Combined Techiman Tamale Estimates 0.029 –0.050 0.061 0.062 –0.043 0.128 Std. Err. (0.079) (0.108) (0.123) (0.070) (0.093) (0.110) AME [0.011] [–0.018] [0.021] [–0.016] [0.011] [–0.032] [–0.006] [0.004] [–0.012] [0.000] [0.000] [–0.001] [0.023] [–0.015] [0.045] Equation 3b Equation 4b Estimates 0.051 0.010 –0.002 0.072 0.042 0.024 Std. Err. (0.086) (0.116) (0.139) (0.084) (0.114) (0.133) AME [0.014] [0.003] [–0.001] [–0.012] [–0.007] [–0.003] [–0.008] [–0.004] [–0.003] [0.020] [0.012] [0.007] Table A3 Effect of urban household participation in food production on food security 22 22 Paper III Food transfer receipts as multi-spatial livelihoods among urban households in Ghana Hayford Mensah Ayerakwa Department of Human Geography Lund University 1 1 Abstract This study uses data from two intermediate sized cities in Ghana to analyze the importance and determinants of food transfer receipts among urban households in Ghana. The results show that grains and tubers are the most received food groups, usually from rural and urban relatives. All (100%) rural- urban and intra-urban food transfer receipts are found to be consumed by the receiving households while about a fifth of rural-urban transfers are giving out as gifts to other urban households. More than two-thirds of food receipts are considered as important/very important to the receiving urban household. The results further show that, female headed households and ageing household heads had an increasing likelihood of receiving food transfers. At the city level, household cash remittance expenditure strongly influenced the likelihood of food receipt from rural relatives in Techiman but not Tamale. This the paper attributes to the agglomeration of small villages around Techiman. The paper concludes that multi-spatiality and food transfer receipts are an important part of the urban household food basket that must be accounted for in the urban food security discussions, taking into account city specific contexts. Key words: Food transfer receipt, multi-spatial livelihood, rural-urban, intra-urban, own food production, Ghana 2 Introduction Global population growth and urbanization have implications for food production and household food security, especially among urban residents in Africa. Urbanization in sub Saharan Africa, though not a new phenomenon, has received considerable interest in recent literature (Andersson Djurfeldt, 2015). While increased population growth and urbanization bring expansion in economic activities and growth, the phenomenon also raises food security concerns for the urban population as rapid urbanization may be connected to widespread urban poverty. In response to these concerns, government policies have promoted urban and peri-urban agriculture production as an efficient pro-poor activity that allows urban farm households to participate in the market economy by reducing cash expenditures on food from the market and also selling freshly produced food (mostly vegetables) to urban dwellers (Ayerakwa, 2017; Armar-Klemesu, 2000; Cofie & Drechsel, 2007; Zezza & Tasciotti, 2010; Mougeot, 2011; Chagomoka, et al., 2015; Chagomoka, et al., 2016). This assertion has however received a number of criticisms across the literature, in the sense that, the potential impacts of urban agriculture on household food security is over estimated (Zezza & Tasciotti, 2010; Crush, et al., 2011; Frayne, et al., 2014). More so, the lack of comparable data across countries to measure the potential impacts seems to support this claim. Recent evidence however suggests that, urban agriculture alone is not enough to explain food security levels of urban households. Rather, the broader contextualization of all production opportunities available to households is necessary (Ayerakwa, 2017). Other empirical literature suggests urban households in sub Saharan Africa rely on food transfers from rural areas (Potts & Mutambirwa, 1998; Tacoli, 1998 Kamete, 1998; Rakodi, 1998; Smit, 1998; Potts, 2000 ) and own food production in rural areas (Andersson, 2002; Foeken & Owuor, 2008; Foeken & S., 2008) as a livelihood source and contributes significantly to urban household food security (Frayne, 2004; Owuor, 2006; Frayne, 2010). Even though available literature suggests an active interaction between rural and urban spaces in the form of production of food in rural areas and the movement of people, goods and services as parts of livelihood strategies that cover both spheres (Tacoli, 1998; Frayne, 2005; Baker, 2005; Foeken & Owuor, 2008; Fraser et al., 2008) this is not reflected in policy. Indeed, government and policy makers in developing countries and in particular sub Saharan Africa over the years have classified urban and rural areas as mutually exclusive in the development agenda with little consideration on the linkages that exist between the two. Whereas many urban enterprises rely on raw materials from rural areas to feed their industries, households in rural areas also rely on cash incomes from the sale of their goods and services in urban centres to meet their livelihood needs (Tacoli, 1998; Mkwambisi, 2011; 2008). In most developing countries, rural and urban are described based on the population sizes. Agriculture is perceived to be the primary occupation of rural dwellers while urban dwellers engage in manufacturing and service related jobs. In reality however, the increasing interrelations that exist between urban and rural space makes the discussion more complex. An urban centre according to the Ghana Statistical Services is any settlement with a population of 5,000 or more persons. This definition in essence classifies a number of villages as towns not because of parallel development in the provision of infrastructure or economic growth but simply on the basis of population growth. Increasing urbanization brings about shifts in land ownership in favour of infrastructure and residential development. This leads to loss of agricultural productive landscapes in urban centres and increased participation in informal activities, especially in the petty trading sub-sector (Potts & Mutambirwa, 1997; Potts, 2010 Hansen and Vaa, 2004). In other words, urban residents construct an increasingly diverse portfolio to maintain or improve their livelihoods (Ellis, 2000). For most urban residents, maintaining strong links with rural ties provides access to rural land for own food production and food transfers in addition to other livelihood sources available to the household. Increasingly, households in both urban and rural areas continue to find their livelihoods in multi-spatial settings as a livelihood diversification strategy. In instances where these activities are spatially separated, there remains a varied exchange of resources (Tacoli, 1998). ‘Urban centres may provide markets as well as social and producer services for the rural population whereas, for many urban 3 3 individuals, access to rural land or produce through family or reciprocal relationships can be crucial’ (Tacoli, 1998:149). While some literature on food transfers and own food provisioning exists, the concept of food transfers, urban agriculture and rural agriculture have been treated in isolation. This study however considers the interplay that exists among these concepts and how they relate to urban households food provisioning and security. In this article, multi-spatial livelihoods are analysed through the lenses of food transfers across space and own production of food. The paper in this sense considers three types of multi-locality based on food transfers: own production of food in rural areas, rural-urban1 food transfers and intra- urban2 food transfers (food transfers among urban households within the same city). Using data from two intermediate sized urban areas in Ghana, the article aims to analyse the determinants of food transfers (both rural and urban) among receiving households and how transfers are connected to own production of food whether in rural or urban areas. A sub question of interest is to investigate whether poor and female headed households are more likely to receive food transfers than their male counterparts. The rest of the article is structured as follows: the next section discusses the conceptual considerations underpinning multi-spatial livelihood. Section three discusses the survey methodology which highlights the source and type of data used as well the empirical model employed in the article. The results and discussions are presented in section four while the conclusions and policy recommendations are presented in section five. Conceptual considerations The analysis is grounded in concepts related to multi-spatial livelihood, own food production in rural and urban areas, and food transfers to urban areas which is viewed as a response to urban households food security arrangements. Insights from the sustainable livelihoods concepts based on the notion that poor households apply a portfolio of capabilities, assets and activities for a means of living are applied (Carney, 1998). The concept, though mostly applied to rural setting (Scoones, 1998), has applicability in transition urban areas (Rakodi, 2002; 1998, Ellis, 1998). The assets available to households from which livelihood decisions are made are classified into tangible (land, labour, human capital, housing, store of food) and intangible (household relations and social capital) (Rakodi, 2002, 1995; Moser, 1998). In the sections that follow, the detail conceptual descriptions are presented. Multi-spatial livelihoods An aspect of urban-rural multi-spatial linkages is exhibited through food remittances. Yet, as observed by Andersson (2002), the literature on urban food security exhibits a surprising neglect of foodstuffs remitted from rural to urban areas. This is in spite of the increasing awareness that urban and rural areas are generally closely linked through remittances, which serves as a risk minimizing strategy for both urban and rural households (Sarpong & Asuming-Brempong, 2004; Frayne, 2010; Andersson Djurfeldt, 2012). Multi-spatial livelihoods refer to a situation in which households have a livelihood footing in both urban and rural areas, either as a source of food or income (Foeken & Owuor, 2001). The reason for this may vary but one central theme in multi-spatial livelihoods has to do with minimizing risks associated with the household’s livelihood sources. Multi-spatial livelihoods do not necessarily mean a split in the household but a demonstration of commitment to allocating resources to urban-based or rural-based households which are grounded in reciprocal support systems across space (Tacoli, 1998, Foeken & Owuor, 2008). 1 Food transfer receipt from rural households to urban households 2 Food transfer receipt from other urban households within the same city 4 Available literature regarding the flow of cash remittances from urban to rural households and the crucial role these transfers play in the livelihood security of rural households have been well documented (Frayne, 2010; 2004). Nonetheless, recent evidence suggests, increasingly, there is flow of resources, mostly in the form of food from rural to urban networks. This partly explains why most people maintain strong ties with their kinship networks both in urban and rural areas, in order to lay claim on resources that will guarantee access to food. As a result of this, many urban households maintain a close relationship with their families in rural areas with some keeping part of their assets in rural areas (Foeken & Owuor, 2008). Multi-spatial livelihood partly explains the cultural phenomenon where Africans in general maintain close ties with friends and family in rural areas through remittances, investments in housing and social services such as funerals. These activities continue to bond, strengthen and lubricate the social relationships that exist between households in rural and urban areas. Incoming cash remittances to rural areas can therefore be reciprocated by the transfer of food from rural dwellers to friends and families in urban areas. The literature on rural-urban linkages shows a general decline in the frequencies and importance of remittances from urban areas to rural areas while the reverse is true for food transfers from rural to urban areas in Namibia (Frayne, 2004). Frayne (2004) also established from Namibia that, urban households with limited social connections to rural areas were the most vulnerable to hunger. Increasingly, a number of households in urban areas depend on food transfers from friends and relatives in rural areas to mitigate their food insecurity needs (Frayne, 2010; 2004). Multi-spatial livelihood is therefore tied to growing insecurity in both urban and rural areas (Andersson Djurfeldt, 2015) making the once existing division between urban and rural areas increasingly irrelevant (De Haan & Zoomers, 2005; Lerner & Eakin, 2011). As important as this may be, much attention has not been given to this social linkage in the literature (Moorsom, 1995; Potts & Mutambirwa, 1998). This notwithstanding however, it is not every household that can make claim on rural lands to produce food. Rather, it is the privilege of those who can lay claim to social networks through friendship and family ties (Tacoli, 1998; Rakodi, 1998). Households who have this social asset stand the chance of improving their food security situation compared to those with no such opportunities. Urban residents’ food production in rural areas Rising costs of living as a result of increased urbanization and rising food prices pushes urban households to consider own food production to supplement their purchases and also raise income from surplus sales. Part of the strategy to doing this is tied to finding rural plots that are used for own food production by urban households. This is so because rural assets are associated with monetary returns and social value that can be relied upon as forms of social insurance by urban households in difficult times such as commodity price hikes. Rural farming which is part of a multi-spatial livelihood activity among urban residents is correlated to increasing economic pressure associated with livelihoods in both urban and rural areas (Andersson Djurfeldt, 2015; Lerner and Eakin, 2011; de Haan, 2005). One strategy among urban residents is to obtain rural plots for rearing animals or cultivating crops that are transported to the city for consumption or sale. These strategies can have implications for urban food systems as well as rural food production (Mkwambisi, 2011). The motivation for these arrangements sometimes is grounded on cheap rural labour (Makoka, 2005; Bryceson, 2006). In Botswana, Kruger (1998) reports that about a third of low income households in Gaborone city own cattle while half retain land in their home villages that serves as a buffer during difficult periods. In the Kenyan city of Nakuru, 95% of urban residents have access to rural land and 80% and 50% used those lands for crop cultivation and animal rearing respectively (Foeken & Uwuor, 2008). Foeken and Owuor further conclude that, rural food production by urban residents was found to be more important as a livelihood source than urban farming in Nakuru, Kenya. Recent literature from Ghana suggests that during the main farming season, some urban residents in the city of Tamale migrate to rural areas (but leave their families behind in the cities) to cultivate crops and return after the main farming season (Chagomoka et al., 2016; 2015). 5 5 The strong interaction between urban and rural space has affected population mobility-such that permanent migration has been replaced with seasonal migration (Potts, 2010) or counter migration ( Potts, 2005, 2009). Although urbanization may bring better prospects for urban residents, compared with their rural counterparts, these prospects are not realizable for all urban residents. With most urban households having a rural or farming background, increases in food expenditure compels urban households to consider production of food as part of a livelihood diversification strategy. Own food production in urban areas Food production in urban areas, also referred to as urban agriculture (UA) has received considerable attention in the literature in recent decades. Households engage in own food production in urban areas for several reasons including consumption and income from surplus sales. The nutritional, health and economic benefits attributed to urban agriculture (see for example Ayerakwa, 2017; Chagomoka et al., 2016; 2015; Badami and Ramankutty, 2015; Cofie & Drechsel, 2007; Maxwell et. al., 2000; 1998; Mougeot, 2011; 2005; Armah-Klemesu, 2000) has led to the promotion of the practice as a beneficial enterprise that can be harnessed to help the poor, especially the urban poor (Smit, et al., 1996; Cofie & Drechsel, 2007; Lee-Smith, 2010; Mougeot, 2011; FAO, 2012; Lee-Smith, 2013). Recent literature suggests the scale of urban agriculture exceeds what has been understood previously. It is estimated that the share of all irrigated and rainfed croplands can be found in urban areas (11% and 4.7% respectively). Higher proportions are observed (60% and 35% respectively) when the urban boarder is extended by a 20km radius (Thebo, et al., 2014). Notwithstanding these benefits, opponents of urban agriculture have questioned the potential of urban agriculture to contribute to the food requirements of the urban poor who do not own land which has been identified as a critical factor in own food production in urban areas. (Zezza & Tasciotti, 2010; Crush, et al., 2011; Lee-Smith, 2013; Stewart, et al., 2013; Frayne, et al., 2014). Tacoli (1998) concludes that the possibility of engaging in urban agriculture was central to the decision of migrants to move to small towns. Increasingly, there is awareness that rural- urban spaces are intertwined and that the involvement in linkages between them serves as a risk minimizing strategy among households-urban or rural (Andersson Djurfeldt, 2012; Frayne, 2010). Food transfers to urban areas As indicated earlier, rural links and engagement in agriculture in rural areas by urban residents are vital safety nets employed by urban households in response to increasing food prices and population pressures. Although much is known about urban residents’ remitting incomes to rural households, little is known about the reverse. However, there is some evidence to the effect that, food transfers from rural to urban residents is gaining prominence while cash remittance from urban residents to rural kinsmen are on the decline (Owuor 2006; Foeken and Owuor, 2001). Assessment of the importance of transfers and gifts as ‘risk sharing strategy’ has also received considerable interest, especially in economic literature (Andersson Djurfeldt, 2015; Rosenzweig & Stark, 1989; Fafchamps & Lund, 2003). Central to the discussion of food transfers is the question of which types of households are the recipients of food from rural and urban areas? Moreover, the source of food transfers, either from rural or urban areas influences the frequency of transfers. Frayne (2010) conclude from a survey of 11 Southern African countries that, there is greater frequency when the food transfer comes from the urban area to other urban households than when it came from the rural area to urban households. However, the share of food transfers from rural to urban households was greater than those received from other urban households. Diversification in the form of multi-spatial livelihoods may be connected to improved household food security, especially in times of shocks and also provide a hedge against escalating food prices in the urban food economy. Increasingly, rural-to-urban food transfers have become important in understanding urban food security (Frayne, 2004). Urban household own food provisioning arrangements in urban areas through urban and peri-urban agriculture or rural agriculture or reliance on food transfer from rural to urban or urban to urban will 6 continue to occupy the attention of researchers as an important urban livelihood phenomenon requiring further investigations (Andersson Djurfeldt, 2015, Andersson, 2002; Foeken and Owuor, 2008; Frayne, 2004; 2005; 2010). The implications of multi-local food transfers provide the basis of exploring how food transfers relate to own production across space. While several studies look at food transfers, urban agriculture and rural agriculture in isolation, this study is unique in that it tries to grasp the interaction between the different components of food transfers and the interplay that exists with own food provisioning among urban households. Methodology Sample The data used in this article is part of a larger survey conducted in October 2013 in Ghana as part of a collaborative and multidisciplinary research project being carried out in Ghana, Kenya and Uganda. In selecting cities for the Ghana case study, Techiman municipality and the Tamale metropolitan assemblies were purposively selected. The two cities fall within different agro ecological zones in Ghana. Techiman is located in the transition belt in the Brong Ahafo region of Ghana while Tamale falls within the northern savanna belt in the northern region of Ghana. Recent Population and Housing Census estimates the total population of urban Techiman to be about 67,241 while the population of urban Tamale is estimated at about 223,252 (GSS, 2014b, 2014a). Although Techiman is classified as a city based on census data, about a third of households in the municipality are rural and constitute about 51% of all households that participate in agriculture in the municipality. Tamale on the other hand has less than a fifth of total households belonging to rural households (17.2%). The proportion of urban agricultural households however is about 57% (GSS, 2014). In addition, Tamale is the regional capital of the northern region of Ghana, which makes it home for several government agencies and private businesses and therefore provide a huge market base for freshly produced vegetables cultivated in urban areas. A common phenomenon is the active trading sub sector in both cities although this is more pronounced and of larger scale in Tamale. By contrast, Techiman has closer proximity to rural areas which makes it convenient for urban agriculturalists to farm in rural areas or transport food from rural to urban space for consumption or sales or both. The sampling criteria for the study involved the identification of the urban boundary, ensuring that the focus of the study (urban households only) were strictly adhered to. With the help of the District Planning Officers in each city, the urban boundary was classified into four quadrants, ensuring that there was proportional spread of households in each quadrant and the city as a whole. An average of 250 households were drawn from each quadrant adding up to 1,000 urban households in each city and 2,000 households for both Techiman and Tamale. However after data collection, the final sample is distributed as follows: Techiman-1019, Tamale-1001. The total number of households who reported receiving food transfers were about 722. To address the research questions of determinants of multi-spatial livelihood requires household level characteristics such as household type (whether the household participates in agriculture), income and household population distribution (proportions of working adult household members, dependency proportion, etc.) and individual level characteristics such as gender, age, education etc. of household heads. Although the data allows for analysis of foods transferred, the quantities of food transferred were not captured in the dataset. This is considered to be the main limitation of the study. 7 7 Empirical Model To determine the causal determinants of food transfers (intra-urban and rural-urban), households are classified into two groups-households that received food transfers and those that did not, and estimate a binary probit equation as follows, assuming a standard normal distribution of the error term. ⁡ Pr⁡( FTi ⁡ = ⁡1⁡|⁡𝑋) ⁡= ⁡𝛷(𝑋 𝑇𝛽) Where 𝑃𝑟 denotes probability, and 𝛷⁡is the Cumulative Distribution Function (CDF) of the standard normal distribution. FTi is the observable binary food transfer response variable which takes on the value 1 for a household that receives food transfer (i.e. FTi = 1), 0 otherwise; the vector X contains all other explanatory variables (mainly individual and household characteristics). In the first level of the estimation, households are classified as agricultural households or otherwise (i.e. ownproduction, the main explanatory variable of interest takes on the value of 1 if the household is engaged in agricultural production, and 0, otherwise). In the second level of analysis, ownproduction becomes a categorical variable which defines the four types of food producing urban households (ownproduction = 1 if household does not produce any food, ownproduction = 2 if the household produces food in both urban and rural space, ownproduction = 3 if household produces food in rural area only, and ownproduction = 4 if household produces food in urban space only). This is estimated for the combined sample and also for the city specific sample for both rural-urban food recipients and intra-urban food recipients. The hypothesis is that, on average, urban households who produce food have a less likelihood of receiving food transfers (intra-urban or rural-urban food transfer). A priori, we expect urban households who participate in agriculture to relatively be more food secure and less likely to receive intra-urban and rural urban food transfers than those who rely solely on the market, ceteris paribus. In addition, it is expected that female headed households and households with lower monthly cash income per capita will have an increasing likelihood of receiving food transfers, all things being equal. Results and Discussions Types and sources of food transfer received Before returning to the research questions of determinants of food transfer, let us consider some contextual differences relating to food transfer receipts. To begin with, the following questions set the stage for the discussions- what are the types of food that that are sent to urban households and from where do they come from? To what use are these foods put to and how relevant are they in the food security needs of the household? To answer the first question, Table 1 presents the different types of food received by urban households. Broadly speaking, even though different foods are received, tubers remain the most food group received by urban households accounting for more than three quarters (77.6%) of all food transfers received. This is followed by grains accounting for about two-thirds (65.7%) of all food transfer receipts. Nonetheless, there exist some contextual differences across cities and the type of transfer. Although the number one food group received in both cities remain tubers (88.0% and 66.0% respectively for Techiman and Tamale), lower proportions are observed among households receiving grains in Techiman (55.8%) compared to those in Tamale (76.5%). In contrast, the proportion of households receiving fruits in Techiman is higher (8.3%) than what is found in Tamale (1.7%). Even though fruits are generally grown in both cities, the harsh climatic conditions in Tamale and its surroundings makes the production of fruit unconducive relative to Techiman where different types of fruits could be 8 cultivated. As a result, the share of fruits in the food transfers receipt in Techiman is expected to be higher than what is reported in Tamale. Techima Tamal Rural- Intra- Total n e urban urban sample Type of variable n=35 n=371 n=383 n=339 n=722 1 grains 55.8 76.5 68.1 63.3 65.7 tubers 88.0 66.0 76.4 78.7 77.6 vegetables 19.9 11.2 19.6 12.2 15.8 fruits 8.3 1.7 7.3 3.1 5.2 other foodstuffs (including meat and egg) 17.8 8.2 15.3 11.3 13.2 rural relatives 42.0 45.6 85.0 0.0 43.7 urban relatives 40.5 31.0 0.0 74.1 36.0 rural friends 6.1 9.5 15.1 0.0 7.7 urban friends 11.4 14.0 0.0 25.9 12.6 Table 1 Types of food received and their sources Moreover, comparing differences by the types of food transfer received (either rural-urban or intra- urban) follow observed patterns in the dataset. That is, more than two-thirds (76.4% and 78.7%) of rural-urban and intra-urban food transfer receipts respectively come in the form of tubers (mainly yam), followed by grains. An important point noteworthy however is the proportion of households receiving vegetables which is highest among rural-urban food recipients. The general pattern in the types of food transfers received suggests that commodities with longer shelf lives constitute what are commonly received as food receipts, either from rural or urban areas across cities. The sources of transfers to urban households broadly are categorised into four-rural relatives, urban relatives, rural friends, and urban friends. The data suggests that transfers are generally sourced from relatives (whether rural or urban) which jointly account for more than three-quarters (79.7%) of all transfer receipts in the sample. Similar patterns are observed across cities with sources from relatives accounting for about 82.5% and 76.6% for Techiman and Tamale respectively. More so, rural-urban transfers mainly come from rural relatives (85%) with only about 15% coming from rural friends. Although the pattern is similar for intra-urban transfers, a little over a quarter (25.9%) of urban transfer receipts come from friends. Food transfers, consumption and food security As noted in the theoretical literature, food transfers provide important safety nets that complement food needs of receiving urban households. All things being equal, households with kinship ties in rural or urban areas who receive food transfers would relatively be food secure than their counterparts without such opportunities. As observed in Table 2, food transfers received either from rural or urban areas are consumed by the household (99.7%). In addition, over a fifth (21.9%) of food receipts are given out as gifts to other urban households but this is somewhat influenced by rural-urban food transfer. With just about 4% of households selling food received from rural areas, the notion that rural-urban food transfer receipts is motivated by commercial purposes is refuted. 9 9 With all households consuming foods received from urban and rural sources, the paper attempts to understand the relative importance receiving households attach to such food receipts. With less than 2% of households suggesting food receipts are not important to them, the share of households indicating the relevance of food transfers receipts is most striking. As suggested in Table 2, about 98% of recipients consider food transfers as critical to the household’s survival (5.2%); very important (38.7%); important (28.7%); or somewhat important (25.8%). Scaling down to the different cities suggest that all households in Tamale consider food transfer receipts as an important source of food to the household compared to about 97% in Techiman. The proportion of households who consider food transfer as critical to their survival in Tamale is found to be relatively higher (5.8%) compared to Techiman (4.6%). Similar importance is associated with rural-urban and intra-urban food transfers. Techiman Tamale Rural-urban Intra-urban Total sample Type of variable n=371 n=351 n=383 n=339 n=722 eat 99.8 99.7 100.0 100.0 99.7 sell 2.7 2.2 3.4 1.3 4.1 given away 19.7 18.7 20.8 17.3 21.9 feed to livestock 4.2 1.8 6.8 4.7 3.8 not important 3.1 0 0.9 2.3 1.6 somewhat important 26.7 24.8 29.8 21.6 25.8 important 26.4 31.3 30.4 26.9 28.7 very important 39.3 38.1 35.7 41.9 38.7 critical to HH survival 4.6 5.8 3.1 7.3 5.2 Table 2 Uses of food transfers and their importance to urban households Agricultural production, gender and food transfer receipts In this section and the sections that follow, the parameter estimates from the probit model are presented in Table 3. The variables of interest include household participation in agriculture, cash income per capita, and gender of household head as well as city dummy. The null hypothesis that participation in agriculture, either in urban or rural areas increases the food security situation of the households in question and thus have a decreasing probability of receiving intra-urban or rural-urban food transfer is tested, using households who do not engage in food production as the reference category. Ceteris paribus, we fail to reject the null hypothesis at the 5% significance level for intra-urban food transfer. Generally, we find no statistical difference between households who do not produce food and those who produce food in relation to total food transfers receipts in the sample. Although participation in agriculture on rural-urban food transfer carries a positive sign, no statistical difference is found relative to those who do not produce their own food. In effect, urban households engaged in agriculture demonstrate a decreasing likelihood of receiving food transfers from other urban residents Another hypothesis of interest is the probability that, female headed households are more likely to receive intra-urban and rural-urban food transfer. Relative to male headed households, we fail to reject the null hypothesis that female headed households have higher likelihood of food transfer receipt for both total food transfers and intra-urban food transfer at the 1% level of significance. Several reasons may account for this result. Generally in Ghana, the cultural phenomenon is to have males as the head of households. It is however common to find females whose spouses are deceased acting as the heads of their respective households. In such instances, the deceased husband’s family have the obligation (unwritten social contract) of supporting the widow (now the female head) to have a decent livelihood, 10 and this can come in the form of food transfer. This is the case because, the children from the marriage are considered the assets of the husband’s family. Intra-urban food transfer Rural-urban food transfer Total food transfer VARIABLES Coef AME Coef AME Coef AME hh_produce_food -0.183** -0.041 0.034 0.009 -0.100 -0.036 age_of_hh_head -0.031** -0.007 -0.039** 0.010 -0.042*** -0.015 agesq_of_hh_head 0.038** 0.008 0.038** 0.010 0.045*** 0.016 female_headed_hh 0.368*** 0.091 0.090 0.024 0.272*** 0.100 higher_educ_hh head 0.001 0.000 0.074 0.020 0.066 0.023 higher_educ_spouse 0.121 0.028 0.071 0.019 0.066 0.024 working_age_female 0.002 0.000 0.067** 0.018 0.036 0.013 working_age_male -0.106** -0.024 -0.000 -0.000 -0.071** -0.025 members_under15 0.006 0.001 0.003 0.001 -0.007 -0.003 members_above64 -0.035 -0.008 -0.163 -0.043 -0.061 -0.022 food_secure_hh 0.064 0.014 0.120* 0.032 0.103* 0.037 log_income_pc 0.035 0.008 -0.056 -0.015 -0.007 -0.003 hh_remmitance_expd 0.091 0.020 0.201*** 0.052 0.203*** 0.072 Techiman(ref=Tamale) 0.078 0.018 -0.140** -0.037 -0.013 -0.005 Intercepts -0.678* 0.011 0.351 Observations 2,004 2,004 2,004 2,004 2,004 2,004 Pseudo R-squared 0.0414 0.0140 0.0244 chi2 67.85 27.75 61.35 Log-likelihood value -814.4 -939.3 -1248 Note: AME denotes Average Marginal Effect *** p < 0.01, ** p < 0.05, * p < 0.1 Table 3 Probit estimates of the determinants of food transfer receipt All things being equal, the presence of working age female adults increases the chances of a household receiving rural-urban food transfers while the presence of male working age adults decreases the likelihood of the household receiving intra-urban food transfers. This result suggest that, the presence of working age males in an urban household reduces the vulnerability level of the household while the reverse is true for the presence of female working adult members of the household. Generally, farming is a labour intensive activity in Ghana and mostly done by men while women provide care for children and the family. As alluded to from the empirical literature, as part of the multi-locality, some families prefer to keep their wives in the city together with the children while the men spent time in the villages cultivating rural land. Such households turn to depend largely on food transfer receipts as a livelihood source from their partners. On the contrary, male presence in the urban household setting provides some guarantees as men are perceived to be hardworking with the ability to providing the needs of their households. The results of the parameter estimates on whether households remittance expenditure is a determinant of food transfer receipt suggests that households who remit cash incomes to rural areas (either to friends or relatives) have an increasing likelihood of receiving food transfer from rural areas. Generally, reciprocal exchange is part of the Ghanaian culture where rural dwellers seek to give food to anyone who voluntarily gives money to them. Since food crops are what they generally have, they find great satisfaction in giving part of their farm produce to urban dwellers whenever they got the opportunity. In fact, in some instances, a rejection of a gift from a rural kinsman is considered a rejection of the person and not the gift in question. 11 11 Cash income and food transfer receipt In this section, we also test the hypothesis that, households with lower monthly cash income per capita have higher likelihood of receiving intra-urban and rural-urban food transfers. Based on the parameter estimates which shows no statistical effect, the null hypothesis is rejected and conclude that, contrary to the assumption that poor urban households are the recipients of intra-urban and rural-urban food transfers, income is found to not be a determinant of food transfer receipt. In other words, the poor are not necessarily the recipients of food transfers. Other determinants of food transfer receipt in the model include age and age square of household heads which are found to be statistically significant at least at the 5% level of significance. The implication of this result is that, a unit increase in age decreases the probability of food transfers receipt. However, as household heads grow older, they become less economically active and more susceptible to receiving both intra-urban and rural-urban food transfer. This to a large extent reveals the weak social pensions system in Ghana where the elderly largely depend on the goodwill of family, friends and well-wishers for a livelihood. Determinants of multi-locality and food transfer receipts in Techiman and Tamale To analyse multi-spatial livelihoods specifically across cities, we turn our focus to those households who sourced their livelihoods in both rural and urban areas and, estimate the determinants of intra-urban and rural-urban food transfer using the same covariates in the model for both Techiman and Tamale. To start with, we disaggregate households into four groups namely no agricultural households, both urban and rural agricultural households, rural agricultural households and urban agricultural households (Table 4). At the city level, we test for the hypothesis that has been employed in the main model and conclude as follows: The hypothesis that participation in both urban and rural agriculture decreases the likelihood of receiving intra-urban food transfer is rejected for Techiman but we fail to reject same for Tamale. On the contrary, participation in rural agriculture is found to decrease the likelihood of intra-urban food transfers receipt in Techiman but not Tamale. Similar to earlier findings, female headed households in both cities (Techiman and Tamale) have an increasing probability of receiving intra-urban food transfers, significant at the 1% level. The finding suggests vulnerability on the part of female headed households, especially in Tamale where females do not own land even though a sizeable proportion of women help their husbands on their farms. In the event of death of a male spouse, the farmland reverts to the clan of the deceased husband, making the woman vulnerable and dependent on family and friends for food as a means of livelihood. The determinants of rural-urban food transfer receipt in Techiman are found to include the age of household head and cash remittance expenditure of the household. As people advance in age, their chances of depending on food transfer receipt increases. It is instructive to note however that, cash remittance expenditure is a strong determinant of rural-urban food transfer receipt but not for intra- urban food transfers suggesting that, some degree of reciprocity. In Tamale on the other hand, the determinants of rural-urban food transfer receipt are found to be participation in rural agriculture by urban residents and the sex of household head (female headed households), all other things being equal. As indicated earlier, there seems to be some correlation between rural-urban food transfer and cash remittance to rural households from relatives and friends in urban areas. In essence, rural-urban food transfer may be some form of reciprocal relationship that benefits both urban and rural households involved in this relationship. In addition, rural-urban food transfer seems to thrive on participation in rural agriculture. Households that participate in rural agriculture have the chance of receiving gifts from family and friends, especially during the harvest season when there is surplus produce available. 12 13 Intra-urban food transfer Rural-urban food transfer Total food transfer Techiman Tamale Techiman Tamale Techiman Tamale VARIABLES Coef AME Coef AME Coef AME Coef AME Coef AME Coef AME hh_produce_food_urban&rural -0.212 -0.050 -0.655** -0.102 0.319 0.092 0.268 0.075 -0.187 -0.066 -0.031 -0.011 hh_produce_food_rural only -0.347** -0.077 -0.099 -0.021 0.115 0.031 0.242* 0.067 -0.189 -0.067 0.078 0.028 hh_produce_food_urban only -0.177 -0.042 -0.186 -0.038 -0.197 -0.046 0.049 0.013 -0.211** -0.074 -0.110 -0.037 age_of_hh_head -0.019 -0.005 -0.053** -0.011 -0.046** -0.011 -0.028 -0.007 -0.038** -0.014 -0.048** -0.017 agesq_of_hh_head 0.029 0.007 0.056** 0.011 0.039* 0.010 0.034 0.009 0.041** 0.015 0.053** 0.018 female_headed_hh 0.368*** 0.093 0.405*** 0.095 0.012 0.003 0.294** 0.085 0.278*** 0.100 0.350*** 0.129 higher_educ_hh head 0.140 0.034 -0.178 -0.035 0.168 0.043 0.038 0.010 0.233** 0.084 -0.079 -0.028 higher_educ_spouse 0.321** 0.085 -0.088 -0.017 0.072 0.018 0.134 0.037 0.235* 0.085 -0.027 -0.009 working_age_female 0.005 0.001 -0.009 -0.002 0.073 0.018 0.056 0.015 0.023 0.008 0.037 0.013 working_age_male -0.097 -0.023 -0.108* -0.022 -0.045 -0.011 0.016 0.004 -0.096* -0.034 -0.064 -0.023 members_under15 -0.049 -0.012 0.064* 0.013 -0.004 -0.001 0.005 0.001 -0.036 -0.013 0.021 0.007 members_above64 0.043 0.010 -0.127 -0.026 -0.139 -0.034 -0.208 -0.056 0.005 0.002 -0.141 -0.049 food_secure_hh 0.139 0.033 0.011 0.002 0.163* 0.040 0.127 0.034 0.188** 0.066 0.044 0.016 log_income_pc 0.067 0.016 -0.004 -0.001 -0.061 -0.015 -0.053 -0.014 -0.007 -0.002 -0.006 -0.002 hh_remmitance_expd 0.115 0.027 0.043 0.009 0.328*** 0.081 0.083 0.022 0.283*** 0.100 0.120 0.042 Intercepts -1.082** 0.153 0.167 -0.471 0.254 0.472 Observations 1,008 1,008 996 996 1,008 1,008 996 996 1,008 1,008 996 996 Pseudo R-squared 0.058 0.0552 0.0392 0.0180 0.0467 0.0223 chi2 51.52 38.05 34.57 18.42 57.78 27.05 Log-likelihood value -430.8 -368.1 -448.5 -476.7 -622.3 -611.4 Note: AME denotes Average Marginal Effect *** p < 0.01, ** p < 0.05, * p < 0.1 Table 4 Probit estimates of multi-spatial livelihood and the probability to receive food transfers across cities 13 Conclusions Urban food security will continue to attract attention from both researchers and policy makers, especially in sub-Saharan Africa where urbanization is on the rise above global averages. Critical to meeting the urban household food requirements include the reliance on food transfers from family and friends domiciled in either urban or rural areas or both. To this end, urban households keep very close ties with family and friends in order to keep the social support system operational. Other families resort to participation in agriculture in rural areas as a means of supplementing the household food requirements. Although food transfers and other foods received from participating in rural agriculture play an important role in household diets, the aggregate effect are not noticeable as these activities take place outside the market channels. This article therefore had as its primary objective to identify the determinants of food transfer receipt from urban and rural areas in Ghana. Contrary to the widely held view that relatively poor households are the recipients of food transfer, the results suggest income is not a determining factor in either intra-urban or rural-urban food transfers. Additionally, the paper concludes from the results that participation in agriculture decreases the likelihood of receiving intra- urban food transfers but not rural-urban food transfers. The article further establishes that, female headed households have a positive and higher likelihood of receiving food transfers relative to their male counterparts. The weak land tenure and ownership structure in Ghana places so much power in the hands of male spouses which is a demonstration of inequality in the allocation of wealth at the local level. Access to essential resources such as land for agricultural production by women would greatly improve on the welfare and bridge the inequality levels associated with female access to productive resources. This in way will reduce the female vulnerability levels while improving on household food security outlooks. At the city level, participation in urban agriculture, age of household head, female headed household (sex), attainment of higher education by household head and spouse as well as household remittance expenditure were all found as determinants of food transfers in Techiman. In Tamale on the other hand, food transfer is determined by age of household head and gender. Similar to the vulnerability argument concerning female headed households, there is over concentration of pension schemes to the formal sector of the economy. However, with nearly 70-80% of the Ghanaian economy being informal, most people do not have any form of pension to depend on as they age. As a result, ageing correlates with increased dependence, including relying on food transfers from friends and relatives to augment food needs of household members. It is however noteworthy that, reciprocal multi-spatial livelihoods are an important part of everyday life in urban areas in Ghana particularly in smaller and intermediate sized urban centres where it may be easier to uphold relationships with rural areas as well as engage in rural agriculture. With all households consuming food received through intra-urban and rural-urban food transfers, the article concludes that, food transfers plays an important role in the food security needs of urban households and should therefore be accounted for in the food security discussions of urban residents. The concept of multi-locality/spatiality should be given the requisite space and attention in the quest to understanding the food security needs of urban households. With Techiman and Tamale increasingly expanding, the potential benefits associated with multi-locality as a livelihood source could be harnessed to improve on the food requirements of urban residents. Conflict of interest None 14 References Andersson Djurfeldt, A. (2015). Urbanization and linkages to smallholder farming in sub-Saharan Africa: Implications for food security. Journal of Global Food Security, 4, 1-7. Andersson, A. (2002). The Bright Light Grow Fainter: Livelihoods, Migration and a Small Town in Zimbabwe. Stockholm University, Sweden. Andersson Djurfeldt, A. (2012). Virtuous and Vicious Cycles in Rural–Urban Linkages: Cases from Zimbabwe. Africa Review, 4(2), 136-156. Andersson Djurfeldt, A. (2015). Multi‐Local Livelihoods and Food Security in Rural Africa. Journal of International Development, 27(4), 528-545. Armar-Klemesu, M. (2000). Urban Agriculture and Food Security, Nutrition and Health. In D. M. Bakker N, Gundel S, Sabel-Koschella U, de Zeeuw H (Ed.), Growing Cities, Growing Food, Urban Agriculture on the Policy Agenda). Feldafing, Germany: DSE Feldafing. Ayerakwa, H. M. (2017). Urban households' engagement in agriculture: implications for household food security in Ghana's medium sized cities. Geographical Research. Chagomoka, T., Drescher, A., Glaser, R., Marschner, B., Schlesinger, J., & Nyandoro, G. (2015). Vegetable production, consumption and its contribution to diets along the urban–rural continuum in Northern Ghana. African Journal of Food, Agriculture, Nutrition and Development, 15(4), 10352-10367. Chagomoka, T., Unger, S., Drescher, A., Glaser, R., Marschner, B., & Schlesinger, J. (2016). Food coping strategies in northern Ghana. A socio-spatial analysis along the urban–rural continuum. Agriculture & Food Security, 5(4), 1-18. Cofie, O., & Drechsel, P. (2007). Water for food in the cities: the growing paradigm of irrigated (peri)-urban agriculture and its struggle in sub-Saharan Africa. African Water Journal, 1(1), 26-50. Crush, J., Hovorka, A., & Tevera, D. (2011). Food security in southern African cities: the place of urban agriculture. Progress in Development Studies, 11(4), 285-305. De Haan, L., & Zoomers, A. (2005). Exploring the frontier of livelihoods research. Development and change, 36(1), 27-47. Ellis, F. (2000). Rural livelihoods and diversity in developing countries: Oxford university press. Fafchamps, M., & Lund, S. (2003). Risk-sharing networks in rural Philippines. Journal of development Economics, 71(2), 261-287. FAO. (2012). Growing greener cities in Africa: first status report on urban and peri-urban horticulture in Africa. Rome: FAO. Foeken, D., & Owuor, S. O. (2001). Multi-spatial livelihoods in sub-Saharan Africa: Rural farming by urban households–The case of Nakuru, Town, Kenya. Mobile Africa: Changing patterns of movement in Africa and beyond, 125-140. Foeken, D. W., & Owuor, S. O. (2008). Farming as a livelihood source for the urban poor of Nakuru, Kenya. Geoforum, 39(6), 1978-1990. Foeken , W. J. D., & S., O. (2008). Farming as a livelihood source for the urban poor of Nakuru, Kenya. Geoforum, 39, 1978-1990. Frayne, B. (2004). Migration and urban survival strategies in Windhoek, Namibia. Geoforum, 35, 489-505. Frayne, B. (2005). Rural productivity and urban survival in Namibia: eating away from home. Journal of Contemporary African Studies, 23(1), 51-76. Frayne, B. (2010). Pathways of food: mobility and food transfers in Southern African cities. International Development Planning Review, 32(3-4), 291-310. Frayne, B., McCordic, C., & Shilomboleni, H. (2014). Growing Out of Poverty: Does Urban Agriculture Contribute to Household Food Security in Southern African Cities? . Urban Forum, 25, 177-189. GSS. (2014a). 2010 Population and Housing Census: District Analytical Report, Tamale. . Ghana Statistical Services, Accra, Ghana. GSS. (2014b). 2010 Population and Housing Census: District Analytical Report, Techiman. Ghana Statistical Services, Accra, Ghana. Kamete, A. Y. (1998). Interlocking livelihoods: farm and small town in Zimbabwe. Environment and Urbanization, 10(1), 23-34. Lee-Smith, D. (2010). Cities feeding people: an update on urban agriculture in equatorial Africa. Environment and Urbanization, 22(2), 483-499. Lee-Smith, D. (2013). Which way for UPA in Africa? . City, 17(1), 69-84. Lerner, A. M., & Eakin, H. (2011). An obsolete dichotomy? Rethinking the rural–urban interface in terms of food security and production in the global south. The Geographical Journal, 177(4), 311-320. Mougeot, L. J. (2011). International Support to Research and Policy on Urban Agriculture (1996–2010): achievements and challenges. International Development Research Centre, Ottawa. 15 15 Owuor, S. O. (2006). Bridging the urban-rural divide: Multi-spatial livelihoods in Nakuru town, Kenya. Potts, D. (2000). Urban unemployment and migrants in Africa: evidence from Harare 1985–1994. Development and change, 31(4), 879-910. Potts, D. (2005). Counter-urbanisation on the Zambian copperbelt? Interpretations and implications. Urban Studies, 42(4), 583-609. Potts, D. (2009). The slowing of sub-Saharan Africa's urbanization: evidence and implications for urban livelihoods. Environment and Urbanization, 21(1), 253-259. Potts, D., & Mutambirwa, C. (1997). The government must not dictate’: rural‐urban migrants’ perceptions of Zimbabwe's land resettlement programme. Review of African Political Economy, 24(74), 549-566. Potts, D., & Mutambirwa, C. (1998). “Basics are now a luxury”: perceptions of structural adjustment's impact on rural and urban areas in Zimbabwe. Environment and Urbanization, 10(1), 55-76. Potts, D. H. (2010). Circular migration in Zimbabwe & contemporary sub-Saharan Africa: Boydell & Brewer. Rakodi, C. (1998). Review of the poverty relevance of the peri-urban interface production system research. Report for the DFID Natural Resources Systems Research Programme, 2. Rosenzweig, M. R., & Stark, O. (1989). Consumption smoothing, migration, and marriage: Evidence from rural India. The Journal of Political Economy, 905-926. Smit, J., Nasr, J., & Ratta, A. (1996). Urban agriculture: food, jobs and sustainable cities. New York, USA, 2, 35- 37. Smit, W. (1998). The rural linkages of urban households in Durban, South Africa. Environment and Urbanization, 10(1), 77-88. Stewart, R., Korth, M., Langer, L., Rafferty, S., Da Silva, N., & van Rooyen, C. (2013). What are the impacts of urban agriculture programs on food security in low and middle–income countries? Environmental Evidence, 2(7), 1-13. Tacoli, C. (1998). Rural-urban interactions; a guide to the literature. Environment and Urbanization, 10, 147-166. Thebo, A., Drechsel, P., & Lambin, E. (2014). Global assessment of urban and peri-urban agriculture: irrigated and rainfed croplands. Environmental Research Letters, 9(11), 114002. Zezza, A., & Tasciotti, L. (2010). Urban agriculture, poverty, and food security: Empirical evidence from a sample of developing countries. Food policy, 35(4), 265-273. 16 Appendix Notation Description Participation in agriculture: ownproduction_none Household does not produce any food ownproduction_urban Household produces food in urban areas only ownproduction_rural Household produces food in rural areas only Household produces food in both urban and rural ownproduction_urban_rural areas female_headed_hh Female headed household age_of_hh_head Age of household head agesq_of_hh_head Square age of household head higher educ_hh_head Head has higher education highr_educ_spouse Spouse has higher education members_under15 Number of household members less than 15 years members_above64 Number of household members above 64 years working_age_female Number of working age females in the household working_age_male Number of working age males in the household food_secure_hh Food secure household log_income_pc Log of household income PC hh_remmitance_expd Household remittance expenditure Table A1 Variable notation and description 17 17 MEDDELANDEN FRÅN LUNDS UNIVERSITETS GEOGRAFISKA INSTITUTION. AVHANDLINGAR I. Herman Richter: Skånes karta från mitten av 1500-talet till omkring 1700 : bidrag till en historisk-kartografisk undersökning. (1929) II. Josef Westin: Kulturgeografiska studier inom Nätra-, Näske- och Utbyåarnas flodområden samt angränsande kusttrakter. (1930) III. Herman Richter och Wilhelm Norlind: Orbis Arctoi Nova et Accurata Delineatio Auctore Andrea Bureo Sueco 1626. (1936) IV. Sven Björnsson: Sommen-Åsundenområdet : en geomorfologisk studie. (1937) V. Arne Sandell: Tektonik och morfologi inom dalformationen med omgivande urbergsterräng. (1941) VI. Sven Dahl: Torna och Bara : studier i Skånes bebyggelse- och näringsgeografi före 1860. (1942) VII. Karl Erik Bergsten: Isälvsfält kring norra Vättern : fysisk-geografiska studier. (1943) VIII. Carl Erik Nordenskjöld: Morfologiska studier inom övergångsområdet mellan Kalmarslätten och Tjust. (1944) IX. Sven Björnsson: Blekinge : en studie av det blekingska kulturlandskapet. (1946) X. Karl Erik Bergsten: Östergötlands bergslag : en geografisk studie. (1946) XI. Tor Holmquist: Den halländska vinterfiskehamnsfrågan. (1947) XII. Olof Ängeby: Landformerna i nordvästra Jämtland och angränsande delar av Nord-Tröndelag. (1947) XIII. Axel Wennberg: Lantbebyggelsen i nordöstra Östergötland 1600-1875. (1947) XIV. Lars Bjerning: Skånes jord- och stenindustri : dess utveckling, lokalisering och betydelse ur näringsgeografisk synvinkel. (1947) XV. Allan Weinhagen: Norbergs bergslag samt Gunnilbo och Ramnäs till omkring 1820 : studier i områdets närings- och bebyggelsegeografi. (1947) XVI. Helge Stålberg: Smålands skogs- och träförädlingsindustrier : en närings- geografisk studie. (1947) XVII. Folke Lägnert: Veteodlingen i södra och mellersta Sverige. (1949) XVIII. Yngve Nilsson: Bygd och näringsliv i norra Värmland : en kulturgeografisk studie. (1950) XIX. Olof Ängeby: Evorsionen i recenta vattenfall. (1951) XX. Karl Erik Bergsten: Sydsvenska födelseortsfält. (1951) XXI. Folke Lägnert: Valmanskåren på skånes landsbygd 1911-1948. (1952) XXII. Olof Nordström: Relationer mellan bruk och omland i östra Småland 1750-1900. (1952) XXIII. Arvid Bergdahl: Israndsbildningar i östra Syd- och Mellansverige med särskild hänsyn till åsarna. (1953) XXIV. Sven E Behrens: Morfometriska, morfogenetiska och tektoniska studier av de nordvästskånska urbergsåsarna, särskilt Kullaberg. (1953) XXV. Torsten Hägerstrand: Innovationsförloppet ur korologisk synpunkt. (1953) XXVI. Gunhild Weimarck: Studier över landskapets förändring inom Lönsboda, Örkeneds socken, nordöstra Skåne. (1953) XXVII. Ingemar Larsson: Structure and landscape in western Blekinge, southeast Sweden. (1954) XXVIII. Sven Godlund: Busstrafikens framväxt och funktion i de urbana influensfälten. (1954) XXIX. Folke Lägnert: Syd- och mellansvenska växtföljder. Del I : de äldre bruknings- systemens upplösning under 1800-talet. (1955) XXX. Olof Ängeby: Toppkonstans, erosionsytor och passdalar i Jämtland och Tröndelag. (1955) XXXI. Gunnar Johnsson: Glacialmorfologiska studier i södra Sverige. (1956) XXXII. Folke Lägnert: Syd- och mellansvenska växtföljder. Del II : 1900-talet. (1956) XXXIII. Olof Nordström: Befolkningsutveckling och arbetskraftsproblem i östra Småland 1800-1955. (1957) XXXIV. Sven Godlund: Befolkning - regionsjukhus - resmöjligheter - regioner. (1958) XXXV. Elis Pålsson: Gymnasiers rekrytering och lokalisering. (1958) XXXVI. Harald Svensson: Glaciation och morfologi : en glacialgeografisk studie i ett tvärsnitt genom Skanderna mellan södra Helgelandskusten och Kultsjödalen. (1959) XXXVII. Erik Ljungner: Nahuel Huapi : ein geographischer Querschnitt durch die Anden in Patagonien. (1959) XXXVIII. Nils Lewan: Om pendling mellan bostad och arbetsplats : en undersökning med material från sydvästra Skåne. (1960) XXXIX. Åke Mattsson: Morphologische Studien in Südschweden und auf Bornholm über die nichtglaziale Formenwelt der Felsenskulptur. (1962) XL. Stig Nordbeck: Framställning av kartor med hjälp av siffermaskiner. (1964) XLI. Olof Nordström: Svensk glasindustri 1550-1960. (1962) XLII. Jan Davidsson: Littoral processes and morphology on scandian flatcoasts. (1963) XLIII. Martin Markgren: Detaljmorfologiska studier i fast berg och blockmaterial : geomorfologisk studie inom Fennoskandia med Skåne. (1962-1963) XLIV. Martin Markgren: Geomorphological studies in Fennoscandia. II : chute slopes in northern Fennoscandia. A : regional studies. (1964) XLV. Martin Markgren: Geomorphological studies in Fennoscandia. II : chute slopes in northern Fennoscandia. B : systematic studies. (1964) XLVI. Lennart Améen: Stadsbebyggelse och domänstruktur. (1964) XLVII. Arvid Bergdahl: Det glaciala landskapet. (1961) XLVIII. Olof Nordström - Solveig Mårtensson: Turism på Öland (1966) XLIX. Jan O Mattsson: The temperature climate of potato crops. (1966) L. Nils Lewan: Landsbebyggelse i förvandling. (1967) LI. Gösta Nordholm: Skånes äldre ekonomiska geografi. (1967) LII. Sven Godlund - Torsten Hägerstrand - Bengt Svanström: Samhälls-utvecklingen och samhällsplaneringen. (1967) LIII. Tor Fr Rasmussen: Storbyutvikling og arbeidsreiser. (1966) LIV. Erik Fagerlund - Harald Svensson - Sven Lindqvist m fl: Infrarödtermografi : principer och naturgeografiska tillämpningar. (1967) LV. Lars Eldblom: Structure foncière, organisation et structure sociale. (1968) LVI. Knut Norborg: Jordbruksbefolkningen i Sverige. (1968) LVII. Gunhild Weimarck: Ulfshult (1968) LVIII. Rune Järvenstedt - Sven Lindqvist - Jan O Mattsson m fl: Televisionssystem i naturgeografisk forskning. (1968) LIX. Arne Jakobsson: Omflyttningen i Sverige 1950-1960. (1969) LX. Åke Hillefors: Västsveriges glaciala historia och morfologi. (1969) LXI. Sven Lindqvist: Bebyggelseklimatiska studier. (1970) LXII. Torsten Hägerstrand, Gunnar Törnqvist m fl: Urbaniseringen i Sverige. (SOU 1970:14) LXIII. Bengt Sahlberg: Interregionala kontaktmönster : personkontakter inom svenskt näringsliv – en flygpassagerarstudie. (1970) LXIV. Björn Hedberg: Kontaktsystem inom svenskt näringsliv : en studie av organisationers externa personkontakter. (1970) LXV. Mats G Engström: Regional arbetsfördelning : nya drag i förvärvsarbetets geografiska organisation i Sverige. (1970) LXVI. Torsten Persson: Geomorphological studies in the south-Swedish highlands. (1972) LXVII. Dewitt Davis Jr: A factoral ecology of Malmö 1965 : the social geography of a city. (1972) LXVIII. Zoltan Petery: Studier i svensk regionplanering : regionplanering enligt byggnadslagen i mindre regioner. (1972) LXIX. Tommy Book: Stadsplan och järnväg i Norden. (1974) LXX. Hans Abrahamson: Studier i jordbrukets omstrukturering. (1974) LXXI. Christer Persson: Kontaktarbete och framtida lokaliseringsförändringar. (1974) LXXII. Ulf Helldén: Karst : en studie av Artfjällets karstområde samt jämförande korrosionsanalyser från Västspetsbergen och Tjeckoslovakien. (1974) LXXIII. Jànos Szegö: Befolkningstäthet, markanvändning, planering (vol 1 o 2). (1974) LXXIV. Raul Nino Guerrero: Rural to urban drift of the unemployed in Colombia. (1975) LXXV. Ulf Erlandsson: Företagsutveckling och utrymmesbehov. (1975) LXXVI. Sture Öberg: Methods of describing physical access to supply points. (1976) LXXVII. Bo Lenntorp: Paths in space-time environments : a time-geographic study of movement possibilities of individuals. (1976) LXXVIII. Richard Åhman: Palsar i Nordnorge : en studie av palsars morfologi, utbredning och klimatiska förutsättningar i Finnmarks och Troms fylke. (1977) LXXIX. Björn Gyllström: The organization of production as a space-modelling mechanism in underdeveloped countries. (1977) LXXX. Anders Järnegren - Fosco Ventura: Tre samhällens förändringshistoria : exploateringen av den fysiska miljön i historisk belysning. (1977) LXXXI. Tommy Book: Stadsplan och järnväg i Storbritannien och Tyskland. (1978) LXXXII. Jan O Mattsson - Leif Börjesson: Lokalklimatiska temperaturstudier inom ett skånskt fruktodlingsdistrikt med särskilt beaktande av frostläntheten. (1978) LXXXIII. Bjørn Terje Asheim: Regionale ulikheter i levekår. (1979) LXXXIV. Solveig Mårtensson: On the formation of biographies in space-time environments. (1979) LXXXV. Erik Wallin: Vardagslivets generativa grammatik - vid gränsen mellan natur och kultur. (1980) LXXXVI. Reinhold Castensson: Välja för framtid - om markanvändningsval och förtroendemannainflytande i kommunal planering. (1980) LXXXVII. Kerstin Montal: Industri och vatten : den vattenförorenande industrins lokaliseringsproblem i Malmöhus län. (1980) LXXXVIII. Tommy Carlstein: Time resources, society and ecology : on the capacity for human interaction in space and time in preindustrial societies. (1980) LXXXIX. Jonas Åkerman: Studies on periglacial geomorphology in west Spitsbergen. (1980) XC. Leif Engh: Karstområdet vid Lummelunds bruk, Gotland, med speciell hänsyn till Lummelundagrottan. (1980) XCI. Karna Lidmar-Bergström: Pre-quaternary geomorphological evolution in southern Fennoscandia. (1982) XCII. Lars-Olof Olander: Staten, kommunerna och servicen : tiden kring kommun- reformen i ett ekonomiskt – geografiskt perspektiv. (1984) XCIII. Bo Malmström och Owe Palmér: Glacial och periglacial geomorfologi på Varangerhalvön, Nordnorge : geomorfologisk kartering med analys av glaciala former och blockhav. (1984) XCIV. Franz-Michael Rundquist: Hybrid maize diffusion in Kenya : policies, diffusion patterns and consequences. (1984) XCV. Girma Yadeta: Dynamic processes of development in marginal areas : a case study from the pokot of north west Kenya. (1985) XCVI. Anders Sporrek: Food marketing and urban growth in Dar Es Salaam. (1985) XCVII. Rolf Nyberg: Debris flows and slush avalanches in northern Swedish Lappland : distribution and geomorphological significance. (1985) XCVIII. Lennart Olsson: An integrated study of desertification - applications of remote sensing, GIS and spatial models in semi-arid Sudan. (1985) XCIX. Mikael Stern: Census from heaven? : population estimates with remote sensing techniques. (1985) C. Katarina Olsson: Remote sensing for fuelwood resources and land degradation studies in Kordofan, the Sudan. (1985) CI. Göran Loman: The climate of a sugar beet stand - dynamics, impact on the crop and possibilities of improvement. (1986) CI. Eric Clark: The rent gap and urban change : case studies in Malmö 1860-1985. (1987) CII. Karin Hall-Könyves: Remote sensing of cultivated lands in the south of Sweden. (1988) CIII. Eva Ahlcrona: The impact of climate and man on land transformation in central Sudan : applications of remote sensing. (1988) CIV. Kerstin Cederlund: Offentlig verksamhet : sysselsättning territoriellt och funktionellt. (1988) CV. Per Olof Hallin: Tid för omställning : om hushålls anpassningsstrategier vid en förändrad energisituation. (1989) CVI. Jens Möller: Godsen och den agrara revolutionen : arbetsorganisation, domän- struktur och kulturlandskap på skånska gods under 1800-talet. (1989) CVII. Juha Uitto: The Kenyan conundrum : a regional analysis of population growth and primary education in Kenya. (1989) CVIII. Ola Jonsson: Informationsteknologi och arbete : fallstudier inom svensk sjukvård. (1989) CIX. Tora Friberg: Kvinnors vardag. Kvinnors arbete och liv : anpassnings-strategier i tid och rum. (1990) CX. Tomas Nihlén: Eolian processes in southern Scandinavia and the Mediterra-nean area. (1990) CXI. Anders Löfgren: Att flytta hemifrån : boendets roll i ungdomars vuxenblivande ur ett situationsanalytiskt perspektiv. (1990) CXII. Irma Guillén: Cuidad Guayana – en stad, två världar : en studie av ett regionalt utvecklingsprojekts lokala effekter. (1991) CXIII. Hans Holmén: Building organizations for rural development : state and cooperatives in Egypt. (1991) CXIV. Petter Pilesjö: GIS and remote sensing for soil erosion studies in semi-arid environments : estimation of soil erosion parameters at different scales. (1992) CXV. Ann-Cathrine Åquist: Tidsgeografi i samspel med samhällsteori. (1992) CXVI. José da Cruz: Disaster and society : the 1985 Mexican earthquakes. (1993) CXVII. Tomas Germundsson: Landsbygdens egnahem : egnahemsrörelsen, småbruket och landskapet i sydsvenskt perspektiv. (1993) CXVIII. Ann-Katrin Bäcklund: JUST-IN-TIME : hur industriella rationaliseringsstrate-gier formar arbetsdelning och kompetens. (1994) CXIX. Jon Knudsen: Kulturspredning i et strukturelt perspektiv : eksemplifisert ved politisk og religiøs endring under moderniseringen av det norske samfunn. (1994) CXX. Tone Haraldsen: Teknologi, økonomi og rom : en teoretisk analyse av relasjoner mellom industrielle og territorielle endringsprosesser. (1994) CXXI. Peter Jönsson: Wind climate during the instumental period and recent wind erosion in southern Scandinavia. (1994) CXXII. Peter Schlyter. Palaeo-wind abrasion in southern Scandinavia : field and laboratory studies. (1995) CXXIII. Jun Yamashita: Spatial interaction and spatial structure : a study of public facility location. (1995) CXXIV. Mats Riddersporre: Bymarker i backspegel : odlingslandskapet före kartornas tid. (1995) CXXV. Anders Schærström: Pathogenic paths? : a time geographical approach in medical geography. (1996) CXXVI. Lars Eklundh: AVHRR NDVI for monitoring and mapping of vegetation and drought in east African environments. (1996) CXXVII. Magnus Jirström: In the wake of the green revolution : environmental and socio- economic consequences of intensive rice agriculture – the problem of weeds in Muda, Malaysia. (1996) CXXVIII. Stefan Anderberg: Flödesanalys i den hållbara utvecklingens tjänst : reflektioner kring en ”metabolism” – studie av Rhenområdets utveckling. (1996) CXXIX. Karl-Johan Lundquist: Företag, regioner och internationell konkurrens : om regionala resursers betydelse. (1996) CXXX. Badr-Eldin Taha Osman: GIS-hydrological modelling in aridlands : a geographical synthesis of surface waters for the African Red Sea region in the Sudan. (1996) CXXXI. Marie Stenseke: Bonden och landskapet : ägares och brukares relationer till markerna och förutsättningarna för en uthållig markanvänding. (1997) CXXXII. Kristina Blennow: Spatial variation in near-ground radiation and low temperature – interactions with forest vegetation. (1997) CXXXIII. Lennart Runesson: Tomträtt : ett markpolitiskt instrument i upplösning. (1997) CXXXIV. Johan Hultman: The eco-ghost in the machine : reflexions on space, place and time in environmental geography. (1998) CXXXV. Jonas Ardö: Remote sensing of forest decline in the Czech Republic. (1998) CXXXVI. Per Hillbur: The knowledge arena : approaching agroforestry and competing knowledge systems – a challenge for agricultural extensionl (1998) CXXXVII. Tom Mels: Wild landscapes : the cultural nature of Swedish national parks. (1999) CXXXVIII. Carolyn Hannan-Andersson: Promoting equality between women and men in bilateral development cooperation : concepts, goals, rationales and institu-tional arrangements. (2000) CXXXIX. Nikolaus Solakius: The parnassus zone, central Greece. (2000) CXL. Jonathan Seaquist: Mapping primary production for the west African Sahel using satellite data. (2001) CXLI. Karin Book och Lena Eskilsson: Stadens struktur : varför och hur? (2001) CXLII. Maria Wikhall: Universiteten och kompetenslandskapet : effekter av den högre utbildningens tillväxt och regionala spridning i Sverige. (2001) CXLIII. Rannveig Olafsdottir: Land degradation and climate in Iceland : a spatial and temporal assessment. (2002) CXLIV. Marie Ekström: Relationships between atmospheric circulation and wind erosion in southern Sweden and Australia. (2002) CXLV. Maj-Lena Finnander Linderson: The spatial distribution of precipitation in Scania, southern Sweden : observations, model simulations and statistical downscaling. (2002) CXLVI. Richard Ek: Öresundsregion - bli till! : de geografiska visionernas diskursiva rytm. (2003) CXLVII. Olivia Louw: Exploring the culture of non-payment in the post-apartheid South Africa. (2003) CXLVIII. Cecilia Kjellman: Ta plats eller få plats? : studier av marginaliserade människors förändrade vardagsliv (2003) CXLIX. Christina Scholten: Kvinnors försörjningsrum : hegemonins förvaltare och murbräckor (2003) CL. Micael Runnström: Land degradation and mitigation in northern China : evaluated from the biological production (2003) CLI. Sara Brogaard: Recent changes in land use and productivity in agro-pastoral Inner Mongolia, China (2003) CLII. Jan_Henrik Nilsson: Östersjöområdet : studier av interaktion och barriärer (2003) CLIII. Thomas Hickler: Towards an integrated ecology through mechanistic modelling of ecosystem structure and functioning (2004) CLIV. Andreas Persson: Hydrological modelling, topographical influence and yield mapping in precision agriculture (2004) CLV. Maria Olsrud: Mechanisms of below-ground carbon cycling in subarctic ecosystems (2004) CLVI. Sandra C. Fernández: Farewell to the peasantry? : (Post)modernising rural Mexico : the case of the ejido peasants in the Isthmus of Tehuantepec (2004) CLVII. Andrés P. Leskó: To own the phone : spatial diffusion, ownership and regulation of telephone service in Argentina, 1878-1990 (2004) CLVIII. Henrik Svensson: Öppna och slutna rum : enskiftet och de utsattas geografi : husmän, bönder och gods på den skånska landsbygden under 1800-talets första hälft (2005) CLIX. Pablo Morales: Modeling carbon and water fluxes in European terrestrial ecosystems (2006) CLX. Emmanuel S. Gritti: Global changes and European terrestrial ecosystems (2006) CLXI. Ola Thufvesson: Kreativitetens yttre villkor : miljöer, rörlighet och nobelpristagare (2006) CLXII. Deniz Koca: Impacts of regional climate change on Swedish forests : an evaluation using process-based regional ecosystem modelling approach (2006) CLXIII. Bodil Elmqvist: Livelihood diversification and land use change in the Sahel : an interdisciplinary analysis of gum arabic in Sudan (2006) CLXIV. Jan Vang-Lauridsen: Industrial dynamics and the spatial organization of industries (2006) CLXV. Heidi Wiig Aslesen: Innovation in an urban context (2006) CLXVI. Torbjörn Johansson: Temporal and spatial variability of carbon cycling in a subarctic landscape (2006) CLXVII. Anders Lund Hansen: Space wars and the new urban imperialism (2006) CLXVIII. Lars Coenen: Faraway, so close! : the changing geographies of regional innovation (2006) CLXIX. Pontus Olofsson: Remote sensing of carbon balance across Scandinavian forests (2007) CLXX. Margareta Rämgård: The power of place : existential crises and place security in the context of pregnancy (2006) CLXXI. Helena Eriksson: Leaf area index of Scandinavian forests : methods using in situ and remotely sensed data (2007) CLXXII. Ransom Lekunze: Corporate social responsibility and development : the case of the Chad Cameroon Oil Pipeline Project (2007) CLXXIII. Alla Yurova: Hydrological aspects of the carbon balance in a boreal catchment : a model study (2007) CLXXIV. Jerker Moodysson: Sites and modes of knowledge creation : on the spatial organization of biotechnology innovation (2007) CLXXV. Yahia Mohamed-Mahmood: Chinese development assistance and West African agriculture : a shifting approach to foreign aid? (2007) CLXXVI. Høgni Kalsø Hansen: The urban turn and the location of economic activities (2008) CLXXVII. Helene Bogren: Flytta eller stanna? : betydelsen av plats och platsförankring för den kvalificerade arbetskraftens internationella migration (2008) CLXXVIII. Lotten Jönsson Johansson: Semi-natural grasslands : landscape, history and plant species diversity : a spatio-temporal study of an agricultural landscape on Öland, Sweden (2008) CLXXIX. Carin Nilsson: Windstorms in Sweden : variations and impacts (2008) CLXXX. Margareta Johansson: Changing lowland permafrost in northern Sweden : multiple drivers of past and future trends (2008) CLXXXI. Martin Svensson Henning: Industrial dynamics and regional structural change : geographical perspectives on economic evolution (2009) CLXXXII. Brynhildur Bjarnadóttir: Carbon stocks and fluxes in a young Siberian larch (Larix sibirica) plantation in Iceland (2009) CLXXXIII. Magnus Lund: Peatlands at a threshold : greenhouse gas dynamics in a changing climate (2009) CLXXXIV. Marcin Jackowicz-Korczyński: Land-atmosphere interactions of a subarctic palsa mire (2009) CLXXXV. Nicklas Guldåker: Krishantering, hushåll och stormen Gudrun : att analysera hushålls krishanteringsförmåga och sårbarheter (2009) CLXXXVI. Nicodemus Mandere: Alternative agriculture and rural development: a case study of sugar beet cultivation in Kenya (2009) CLXXXVII. Anna Wramneby: The role of vegetation-climate feedbacks in regional earth system dynamics (2010) CLXXXVIII. Mikhail Mastepanov: Towards a changed view on greenhouse gas exchange in the Arctic : new findings and improved techniques (2010) CLXXXIX. Evelin Urbel-Piirsalu: The Estonian forest sector in transition to sustainability? : capturing sustainability with the help of integrated assessment (2010) CXC. Marie Vandewalle: Effects of past and current land use on biodiversity : from a semi-natural grassland on Öland (Sweden) to a European perspective (2010) CXCI. Maria Andrea Nardi: Rural development and territorial dynamics in the provinceof Misiones, Argentina (2011) CXCII. Torbern Tagesson: Land-atmosphere exchange of carbon in a high-Arctic wet tundra ecosystem (2011) CXCIII. Per Schubert: Model development for estimating carbon dioxide exchange in Nordic forests and peatlands with MODIS time series data (2011) CXCIV. Mabel Ndakaripa Munyuki-Hungwe: In search of ‘community’ in Zimbabwe's fast track resettlement area of Mazowe district (2011) CXCV. Oliver Purschke: Plant community assembly and biodiversity : a spatio-temporal perspective (2011) MEDDELANDEN FRÅN INSTITUTIONEN FÖR KULTURGEOGRAFI OCH EKONOMISK GEOGRAFI. AVHANDLINGAR I. Johanna Bergman Lodin: Engendered promises, gendered challenges. Changing patterns of labor, control and benefits among smallholder households growing NERICA in Uganda (2012) II. Dobloug Tore Anstein: Høyvekstbedrifter og regionale finansieringssystemer (2012) III. Patrik Olsson: Ömse sidor om vägen; Allén och landskapet i Skåne 1700-1900 (2012) IV. Ingrid Helene Garmann Johnsen: Social Capital and Regional Innovation Systems – Bridging approaches and broadening understanding of knowledge sharing and innovation (2012) V. Roman Martin: Knowledge Bases and the Geography of Innovation (2012) VI. Monica Plechero: The changing geography of innovation – Chinese and Indian regions and the global flows of innovation (2012) VII. Erik Jönsson: Fields of Green and Gold: Territorial hunger, rural planning, and the political ecologies of high-end golf (2013) VIII. Elena Zukauskaite: Institutions and the Geography of Innovation: A Regional Perspective (2013) IX. Carl-Johan Sanglert: Att skapa plats och göra rum – Landskapsperspektiv på det historiska värdets betydelse och funktion i svensk planering och miljövård (2013) X. Tiina Lovisa Solbär: Anthropogenic Open Land in Boreal Landscapes – Investigations into the Creation and Maintenance of Arable Fields on Swedish Farms (2014) XI. Ståle Holgersen: The rise (and fall?) of post-industrial Malmö – Investigations of city-crisis dialectics (2014) XII. Wim Carton: Fictitious Carbon, Fictitious Change? Environmental Implications of the Commodification of Carbon (2016) XIII. Hanna Martin: Innovation for tackling grand challenges: Cleantech industry dynamics and regional context (2016) XIV. Srilata Sircar: Between the highway and the red dirt track: Subaltern Urbanization and Census Towns in India (2016) XV. Mikhail Martynovich: General purpose technology diffusion and labour market dynamics: A spatio-temporal perspective (2016) XVI Hayford Mensah Ayerakwa: Planting to Feed the City? Agricultural Production, Food Security and Multi-Spatial Livelihoods among Urban Households in Ghana (2017)