Food vending: adaptation under difficult circumstances VICTOR NGONIDZASHE MUZVIDZIWA ABSTRACT The respondents discussed in this article depended on food vending as their main source of income. To succeed in this activity requires shrewd marketing and hard work. For the majority food vending was basically a hanging on and coping strategy, offering very limited surplus for investment. Food iiending allowed them merely to stay in town while maintaining a foot in their home villages. The paper presents both a descriptive and an analytical account of food vending activities by female heads of households in Masvingo. The officially imposed constraints on food vending demonstrate the existence of competing and conflicting rationalities between male decisionmakers and poor women. The inter-connections between food vendors and the formal markets are noted. Introduction OEconomi SIRIM (1998 c Structura ) notesl Adjustmen that the adoptio t Programm n i e n (ESAP 1990 o ) b f y th th ee Zimbabwean Government at the behest of the World Bank and International Monetary Fund contributed to an unprecedented rise in the number of the very poor in Zimbabwe. These numbered between 60 and 70 per cent of the population in the Southern African region. In 1991, a year after the adoption of the Economic Structural Adjustment Programme (ESAP), 6.6 million people out of a total of Zimbabwe's 10.3 million were estimated to be in absolute poverty (Kamidza 1994; Osirim 1995). Many of the poor were absorbed in the informal sector of the economy. Food vending is such an activity. My data on female food vendors is part of a much larger research project focusing on how female heads of households in Masvingo dealt with impoverishment. These food vendors constituted 38 percent of JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 69 my sample. The 19 women depended on food vending as their main source of income but only one was able to use it as an investment strategy. Food vending enabled most of my respondents to do no more than survive in town and simply hang on and cope with urban existence. They had to develop networks and strategies to deal with the predominantly harsh economic climate of the mid-1990s and by doing so they made interconnections between informal food vending and the formal sector operators such as commercial and small-holder irrigation farmers. The first part of the essay presents a profile of food vendors, together with a detailed presentation of findings. The sections that follow then focus on vending constraints including legal aspects, the experiences and lessons learnt from Masvingo food vendors selected case studies. The last section focuses on food vendors coping strategies. Research methodology TIH don E RESULT e over a S reporte period o d f 1 in 4 thi month s articl s fro e m ar earl e base y Novembe d on fieldwor r 1994 t k o December 1995 in Masvingo, a provincial town of 52 000 people, in southern Zimbabwe (CSO 1993 and Muzvidziwa 1997). To ensure anonymity respondents were assigned a letter symbol(s) instead of using their real names. Table 1 below shows the sources of income for respondents in the main survey. TABLE 1: MASVINGO RESPONDENTS' MAIN SOURCE OF INCOME Main source of income food vending cross border trade formal job prostitution carrier bag hairdressers other* Total Percent 38 26 12 8 6 4 6 100 *One in each of the following categories: a traditional beer brewer, a rentier and a recipient of maintenance money. 70 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000 Frequency 19 13 6 4 3 2 3 50 Muzvidziwa / FOOD VENDING (ADAPTATION) While the focus of this paper is mainly qualitative, both qualitative and quantitative data regarding the situation of female heads of households in Masvingo was collected. To a large extent I tried to see things from the point of view of the women in my study. The method of data collection adopted for the study made it possible to highlight the centrality of the points of view of poor women when studying them. Multiple research techniques were employed during the data-gathering phase. The use of these techniques enabled me to gain greater insights of the women's survival and income earning strategies. Apart from informal interviews and observations in the community, I selected for intensive in-depth interviews 58 female heads of households (eight married breadwinners were later on removed from the analysis of the larger project as they were not heads of households, and I also interviewed 31 officials. The women who participated in the in-depth interviews from January to May, who included the 19 food vendors, were found using snowball sampling. Snowball sampling relied on the women's social networks, mostly drawing on their friendship networks. The snowball sample drew all its study population from Rujeko and Mucheke. These were Masvingo's only low income or high-density areas where the bulk of the town's population lived. Two variables, urban permanency and double-rootedness, were constructed on the basis of eight variables (length of stay in urban area, housing status, split residence, visits to rural village, remittances, attitude toward kin, number of dependents and poverty index). A total of 72 per cent and 28 per cent of respondents were in the doublerooted and urban permanency categories respectively. I also developed criteria that provided a classification of respondents into "burnt-out", "hanging on", "coping" and "climbing out" of poverty (See Table 2 below). The sample characteristics and basic data on respondents Dcatalogu URING TH e t E dat o determin a analysi e th s phas e socioeconomi e I develope c categorie d an analytica s of the l respondents. My respondents fitted into four categories based on an objectively constructed poverty index (see Muzvidziwa 1997). JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 71 Women whose income did not allow them to pay rent in a single room, who had one meal, a day hardly remitted anything home, could not save, were in dire financial difficulties and lived in crowded conditions were placed in the "burnt-out" category, followed by the "hanging on" category. The group after this one was "coping", for those who managed to sustain urban existence in a meaningful way. Finally those who were able to save part of their income for investment purposes, remitted to their dependents if their children lived elsewhere, lived an independent life and owned or rented at least a room were placed in the "climbing out of poverty" category. Table 2 indicates the composite variables aggregated to produce the respondents' socioeconomic status. It shows the value assigned to the response in each score category. TABLE 2: THE POVERTY INDEX VARIABLES Variable Value = 1 Value = 2 Value = 3 Education (self) Housing Number of rooms Persons per room Clothing Number of meals per day Meat intake (last week) Number of dependents Total income Remittances per annum Income minus expenditure "burnt-out" = total score < 14: Those who could hardly sustain their urban existence. "hanging-on" = total score of 15-20: Those who were in difficult circumstances but struggled to stay in town. "coping" = total score of 21-26: Those who could balance the household budget but had no savings. "climbing out" = total score of >27: had a healthy domestic budget, had savings and investments. JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000 72 Forms 1-4 primary post-secondary or less owner two+ 2 some new rental one 3-4 only 2nd hand 2 1-4 times 2-3 lodger shared 5+ only donated 1 0 4+ >Z$600 3 5+ times 1 or less Z$100O+ Z$600-999 1-3 break-even 0 negative 4+ positive Muzvidziwa / FOOD VENDING (ADAPTATION) SUMMARY PROFILE VARIABLE VALUE = 1 VALUE = 2 VALUE = 3 Meat intake (last week) 0 1^4 times 5+times Number of dependents 4+ 2-3 times 1 or none Remittances per annum 0 1-3 times 4 or more Visits to village 3 or more 1-2 times 0 times Table 3 presents a comprehensive summary of the profile of the 19 food vendors. This data is useful for an understanding of food vending activities and food vendors in Masvingo. The meaning of the value assigned to the variables on the previous page is given below: the number of dependents, visit to the village over a period of at least twelve months, meat intake over the week preceding initial interviews and, lastly, regularity of remittances to the village over a period of six months: The age range of food vendors was from 19 to 57 years old. Generally food vendors were lodgers; the few owners were mainly widows. Meat intake over the previous week preceding the initial interview was moderate, mostly ranging from once to no more than four times a week, Up to five had no meat intake during the preceding week, even though they were not vegetarians. This shows that a significant number of female-headed households had a nutritionally deficient diet. On average many had two meals, breakfast and dinner. Food vending was the main source of income for my respondents. Despite the low returns of this sector, it offered a potential lifetime career for most of the women. Other sources of income like crossborder trading and prostitution, due to their demanding nature, tended to be short range and did not offer a lifetime career. However, of the 19 food vendors, none was in the burnt-out category and only one was in the climbing-out of poverty category. Food vending was mainly a hanging on activity and, to a lesser extent, a coping strategy for a quarter of the respondents. The overwhelming majority of food vendors, 16 out. of 19, had very little or no education and none had done any post-secondary training. The food vending sector posed few barriers to these women on JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 73 AGE AC 41 B 40 N 46 AG 28 OS 31 MX 19 BB 31 V 29 BC 32 GD 45 BN 28 ZA 35 K 43 D 55 E 36 HF 47 WD 39 R 35 S 29 * The number indicates the value assigned to the variable as noted on the following page immediately after the abbreviations. Abbreviations listed ialphabetically CLMB = climbing out of poverty D = divorcee DPND = no. of dependents ECOSTATUS = economic status E = highest educational leve HANGING = hanging on LWM = living with a man Ms = marital status week) 0WN = owner PRI = primarjr = no. of rooms REMIT = remittances per annum SEC = secondary SHA = sharing a room W = widow 74 TABLE 3: PROFILE OF FOOD VENDORS H RMS MEAT DPND WSTT REMfT ECOSTATUS MLS MS EDUC L one D L one NM LWM L one 28 L one L one NM L sha NM D L sha D L one LWM L sha LWM L sha W own 2 W own 2 L sha D W pri pri pri pri sec Pri pri sec pri pri sec pri pri pri own 2 W L one D L sha pri Pri W L one NM L sha L sha D 2* 1 2 2 2 2 2 2 1 2 3 3 2 1 1 2 2 1 2 2 2 2 2 2 2 2 3 1 2 3 2 2 1 2 2 2 2 2 pri pri pri LJ= H = housing odger Mis = number of meals per day MEAT = meat intake (last NM = never married RMS VISIT = visits to village JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000 hanging hanging coping coping coping hanging coping clmb hanging hanging coping coping hanging hanging hanging hanging hanging hanging hanging 1* T 2* 1 2 2 2 2 3 1 2 3 1 1 3 1 3 1 1 2 3 2 2 3 2 1 2 1 2 3 1 3 2 1 1 1 1 2 3 1 3 1 1 1 2 1 3 2 1 1 2 1 1 3 3 2 2 Muzvidziwa / FOOD VENDING (ADAPTATION) account of their limited levels of education as no formal training or elaborate licensing procedures were required. The skills they had leamt in their roles as women were useful at the point of entry. For a significant proportion of women in my study sample, food vending was the first port of entry into the labour market. While family size in terms of number of children was moderate, the women had substantial responsibilities in terms of extended family demands. Quite often they took responsibility for the children of their (often deceased) sisters and brothers as well as their own younger siblings. The number of women who remitted something to the village was very high: an indication of the responsibilities that the women had to shoulder. The majority of these women were double-rooted. They sought to stay in town for as long as possible but practical considerations made them realize that at some point in future they would return to the village. Their decision to keep one foot in the village and one in town was based on such practical considerations, and though many remitted money to their rural dependents, visits "home" were becoming scarce. The research findings AL -esponsibl L NINETEE e fo N foo r decision d vendor s that affecte s operate d thei d r vendin as sol g enterprises e traders . Seven got assistance in their day-to-day operations. This assistance came from their own children or relations and also from hired child labour. In fact three food vendors employed a child salesperson as "helper". For instance, V employed two assistants whom she paid $60 each per month. In addition she provided them with food and shared her premises with them until August when she entered into a mapoto (an informal conjugal union). The women's food vending experiences ranged from less than a year to more than ten years. One food vendor had been operating since 1972. The main reason for embarking on vending operations was to earn an income and to escape from dire poverty, as it did not demand a large capital outlay. For those who started vending when they were still married, it allowed them to operate from their homes in a manner acceptable to their husbands. JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 75 The women's place of work Ta H t E Council-designate MAJORITY OF the wome d place n in m s i y sampl n the tow e wer n centr e suppose e area d to . Twelv operat e e operated from Masvingo town centre. The other seven operated from either Rujeko or Miacheke markets. In many ways they operated according to the rules and expectations of the municipality. However, through observation I also noted the use of the home as an incomegenerating space or resource by some respondents. Food vendors used residential space as storage for their wares, making use of the multiple functions of housing. Considerations of security, not only of persons and property but also of their wares, influenced the choice of housing for some of those in my .sample. Limited verandah trading by some food vendors was observed in Masvingo. Although it appeared that verandah trading occurred only on a small scale, this was another illustration of the use of the home as a source of income. My respondents did not report police raids and harassment of verandah traders. The difficulties involved in trying to police people's homes seemed to discourage the police. Products sold Tspecialize HE FEMAL d E i head n the sal s o e f o househol f tomatoes d foo . Mos d vendor t also sol s d generall fruits an y d vegetables and a limited amount of fresh farm produce. This was quite lucrative during the off peak season. One food vendor put her refrigerator to economic use by selling freezits to her customers. This lucrative freezits business was illegal. Another middle-aged food vendor also specialized in the sale of sweets. Sources of products OJ OME OF MY respondents sourced their wares from farms and k3irrigation schemes as far away as Esigodini, Birchnough Bridge and Nyanyadzi, 200 km. from Masvingo, from irrigation schemes in places like Chivi less than a 100 km. away and from small farms within the town's peri-urban zone. Supplies purchased outside the city were half the Masvingo price. For instance a carton of tomatoes in February 1995 was $15 from farms in Esigodini but $35 in a wholesalers in Masvingo. Usually the women purchased in bulk at source, starting 76 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000 Muzvidziwa / FOOD VENDING (ADAPTATION) with a minimum of ten cartons or crates. Some women travelled to collect orders on a weekly basis whilst others made monthly trips to the centres of supply. A common practice amongst the female vendors who depended on external markets was the tendency to minimize transport costs by pooling efforts. They teamed up into groups of three to five and went out for orders in rotation. Each time a woman brought back supplies she would divide the lot among two to four other women. In all cases the women contributed towards the purchase of the fresh produce and sometimes, but not always, they assisted with transport costs. The degree to which the women used networks based on friendship for business purposes was impressive and showed a high degree of organization. Women vendors also purchased fresh produce from wholesale markets in Mucheke and in town from places like Tanaiwa and Vitagreen fresh food markets. Nearly half of them replenished their supplies from delivery lorries that sourced fresh produce at production sources. The lorry deliveries specialized in tomatoes, green vegetables, especially cabbages, and fruit. Lorry delivered from as far away as Mutare some 300 km. to the east of Masvingo. The destination site of the lorries was Mucheke, close to the long-distance bus terminus. Formal and informal linkages Fscal OOD e VENDOR producers S WER . The E linke y knew d t wher o commercia e and when l a certai s wel n good l as small s were - available. Cheater (197-9) argues that cumbersome state bureaucracies are least capable of providing efficient fresh produce marketing opportunities. A more effective distributive system is one that seeks to circumvent official marketing channels. Women food vendors circumvented bureaucratic channels by purchasing from source. Direct purchases benefited both the producers and the vendors, as they were able to circumvent sales tax, thereby reducing the production costs of fresh produce. One can also hypothesize that it lowered the cost of fresh produce items to consumers and the cost of subsistence in the urban setting. It is possible that many a household JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 77 might not have been able to subsist at the level at which they did, had it not been due to the hard work of these women. Brand, Mupedziswa and Gumbo (1995), CutrufeUi (1983) and Kanji and Jadowska (1993) acknowledge the importance of food vendors in urban markets. Food vendors pursued their careers despite official opposition to their work. Food vending was one area that was controlled by women. Women food vendors were an important distributive link that enabled many urban households to subsist under difficult circumstances. Benefits that flow from vending TMasving HROUGH o VENDIN female head G operation s of household s a significan s were able t t proportio o obtain for thei n of r children food, clothes, at times shelter, school fees and hospital fees. Though these activities were not always regarded as legal by municipal authorities, most of the food vendors gained a new type of confidence that enabled them to defy Council regulations on the use of vending space. Many noted that they had gained autonomy and financial independence. Hansen's (1980) assertion with reference to food vendors in Lusaka, Zambia, that food vending operations provided women with cash only on a short term basis and that it did not offer long term.economic support or benefit was not applicable to the female heads in my study. For 38 per cent of my respondents, food vending was a full-time activity that sustained their urban existence. Masvingo food vendors were not only servicing the urban market but had also managed to entrench themselves in meeting the fresh produce needs of a predominantly rural market. At the end of the month the vendors stocked up on fresh produce in order to take advantage of the presence rural public servants and people in formal employment who came to Masvingo for banking and shopping. These rural people purchased tomatoes and other fresh produce items from vendors, hi fact the Masvingo food vendors were referred so as vakadzi vemadomasi (tomato women). There were many reasons why tomatoes were of strategic importance in the fresh produce market. Tomatoes are more than a vegetable in women's strategies for food provisioning. They are a multi-purpose food that can be used with nearly every type or relish. 78 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000 Sources of funding Sources of income Marketing wares Muzvkfziwa / FOOD VENDING (ADAPTATION) In hot climatic conditions tomatoes tend to last longer than most other types of vegetables. Lastly the centrality of tomatoes in the fresh produce market is due to the law of scarcity which puts them in high demand. Loo IDENTIFIE d vending D thre . Nin e e mai wome n source n self-finance s of initia d thei l busines r vending ventur s capita e fro l fo m r their savings. Another nine were given a loan by either parents or a brother, sister, relative or friend. Only one vendor got a monetary grant as opposed to a loan from her brother for her start-up capital. However, none but two of my respondents had at any time obtained credit from formal credit institutions. My respondents were not attracted to credit. There was a strong feeling that one might lose one's business to moneylenders through indebtedness. The women had witnessed numerous examples of people who lost their businesses due to their inability to repay a loan. To them getting a loan was like tying a rope around one's own neck. Given the unpredictable business environment under ESAP these responses were perfectly rational. They preferred to turn to marounds (rotating credit and savings groups) and friends for assistance. No IN f incom E OUT e o othe f th r e tha 19 n foo foo d d vendo vending r respondent . A further nin s ha e als d n o o knitte sourc d e and crocheted on a part-time basis, mainly for women cross border traders. One woman, in addition to food vending, also engaged in mupedzanhamo or the sale of second hand clothes (also known as mazitye'). She sourced her mazitye from Beitbridge once every three months. Beitbridge is Zimbabwe's southern gateway to South Africa. MASVING nteractio O FOO n wit D vendo h their r customers respondent . Durin s had clos g lunc e face-to-fac h hours and e towards the end of the day I saw customers who regularly purchased from particular vendors. These customers talked to the vendors using JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 79 the idiom of kinship. I did not establish whether these relationships were real or fictional. Nevertheless they obligated the customer to buy from a particular vendor. Out of the 19 vendors only three catered exclusively for the urban market; the rest served both the urban and rural clients. Demands on food vendors WTwelv OMEN FOO e operate D vendor d six day s wer s a wee e an k extremel from Monda y bus y to Saturday y group . . Sundays were devoted to attending to household chores such as laundry and tidying up the premises. Some attended church and others visited friends or kin, and there were also those who used Sundays to replenish supplies. These women hardly knew the meaning of leisure. Pood vendors worked long hours, all without exception worked for more than thirteen hours a day. They started their working day quite early at around 5 a.m. This enabled them to attend the produce market in the early hours so that by the time customers woke up they could serve them. There was competition for the best produce in the market and those served first tended to get the best fresh produce. This was useful in attracting customers. Pood vendors also went to bed late at around 10 p.m. Despite obstacles 14 of these respondents believed that their operations would sustain them for some time to come; only three felt that the future was bleak and two were noncommittal in terms of their attitude towards the future. There was a general feeling amongst my respondents that things were going to improve in the near future and that business viability would be enhanced. Three vendors did not share this view These thought that vending was not a lucrative venture and that success was too dependent on the ability of the operator to conduct business. If one fell ill that could spell disaster. Most vendors were interested in maintaining current operational levels, which they could manage effectively. Half of the vendors were able to make some kind of savings: most lived on a stringent hand to mouth budget. Only seven had a bank account. The harsh conditions and demanding nature of food vending had a debilitating effect on the health of the food vendors. Sourcing and selling foodstuffs demanded a lot of mobility both in Masvingo town 80 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000 Vending constraints Muzvidziwa / FOOD VENDING (ADAPTATION) and beyond. Many of the women had to replenish their food items on a regular basis. This often involved travelling and carrying heavy loads on their backs and heads. Even within Masvingo the women had to travel on a daily basis. SIXTEE \t the N beginnin OUT of th g o e f 199 19 vendor 5 licence s fee in th s wer e stud e $21 y pe wer r month e licensed . This . was increased to $27 from August onwards. The municipality had very mean-minded methods of enforcing payments by registered vendors. In the event of non-payment of the vending fee, the water supply at the premises where the vendor lived was disconnected. This sometimes happened long after the lodger had moved to some other premises. The practice of water disconnection was still in force at the end of 1995. Masvingo Council saw food vendors as sedentary, but in practice most food vendors were semi-mobile. Although the vendors thought of a particular spot as their own operational area, from time to time they moved and set themselves up at spots they considered strategic to their business operations. These were places that attracted large numbers of people. Vending in Masvingo was criminalized by administrative procedures and rules that designated fixed sites from which vendors were supposed to operate. Female heads of households who operated as food vendors however often choose to sell their wares from nondesignated strategic sites that maximized sales. They normally located themselves at spots where there was a high flow of people's movement, for instance, outside busy supermarkets, wholesalers, bus termini and the post office. They saw the ever-present threat of police arrest and harassment as the single biggest obstacle to their operations. During the course of 1995, especially the April raids, there was a definite gender bias in police harassment. Only female food vendors were arrested although the round-ups were intended to be a clean-up operation of aE illegal vendors. Most of the women except those who operated from Rujeko had experienced official harassment in the course of their operations in the twelve months preceding the study. They lived in constant fear JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 81 of being subjected to official harassment. Even though most female vendors were legal operators who paid a monthly fee to Council they were often arrested by the police and municipal police for failing to operate from designated points. There were cases of vendors like QP and WZ who were fined more than three times in a space of less than three weeks in April 1995. Illegal vending carried a fine of $30 each time together with the confiscation of one's goods. QP subsequently moved out of the premises where she was residing and 1 did not see her again for the rest of 1995. Other vendors noted that she had left for the communal areas. These fines destroyed the viability of a woman's vending operations. The vendors in the study agreed that harassment and arrests should stop. They felt that the municipal by-laws and regulations should be relaxed so as to allow mobile vending. The women were not opposed to license fees altogether, but they wanted a reduction in fees commensurate with the level of services Council offered them. For instance the food vendors in town operated in the open. They depended on trees for shade. During storms and strong winds they took cover in nearby shop verandahs, yet they were required to pay the same amount in fees as those who were provided with market stalls. Council justified their payment of similar fees by noting that the town women used its water and toilet facilities in the course of their operations. In order to minimize loses due to confiscation of wares during the frequent raids, pavement food vendors left the bulk of their wares at the legal vending sites and carried only small quantities at a time. The women who operated in town left their wares at the Aroma site (the council designated vending site), largely unguarded. They spent a great part of the day moving back arid forth from the legally designated vending site to the illegal sites. The above restraints had important implications for vending in Masvingo where, during the course of 1995, a number of police raids were carried out on women in an attempt to rid the city of illegal vending. One rationale for harassing women was the notion that food vending posed a health risk to city residents. It was nevertheless notable that spots where vendors operated were kept clean by the 82 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000 Muzvidziwa / FOOD VENDING (ADAPTATION) vendors themselves. These women made a point at the end of each day of removing any rubbish in the vicinity of their operational spots, including that left by the shoppers in the town, who had nothing to do with vendors. In many ways the vendors therefore contributed to the general cleanliness of the town and provided an unrecognized subsidy of the general maintenance costs of the municipality. The Masvingo Council licensed food vendors engaged in the sale of largely perishable fresh produce commodities. Non-perishables such as mealie-meal and sweet specialties by respondents D and S were considered illegal goods although D paid the monthly vending fee. Like all other food vendors she was allowed to trade only in fresh market produce. Her solution was to display a few items such as matchboxes, oranges and popcorn so that when Council officials came around she could say that she was selling only these items. D even approached Council seeking a waiver to allow her to sell sweets legally at the shade built by Council. The request was turned down although this was her main source of income. Due to poor health D could only sit and wait for customers in the shade. I had an opportunity to man D's stall once in July 1995 at her request. I sat in the shade for nearly an hour. This opportunity enabled me to chat with the children and the two other market women who were in the shade. Interesting thoughts flashed my mind as I sat manning the shade. What if the police come around and found me selling sweets at the stall? These are the thoughts some of the market women live with at every stage of their business life. Items which were legal to sell in the shade ceased to be so when sold at the terminus. The criminalization of vending activities is something that authorities in towns like Masvingo need to re-examine. The official response to food vendors in Masvingo had a negative effect on poor women. It marginalized the poverty alleviation strategies of informal sector operators such as the women food vendors. The concern of officials was with orderly development and a clean city. Masvingo is a leading tourist town by virtue of its proximity to the Great Zimbabwe Monument. In October and early November 1995 meetings to discuss the relocation of vendors were held in town. All the officials who spoke at these meetings were men. They JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 83 represented different organs of the state and the local authority such as the police, the municipal council health department, and the housing and community services department. The vendors' association representatives, ironically, were also all men. The power conflicts that women encountered in the course of their operations reflected the orchestrated exercise of power in society in the idiom of male-female behaviour patterns perceived appropriate by men. Public meetings are therefore nothing more than a re-enactment of gender roles demonstrating the way women must negotiate a space for acceptable public behaviour. By behaving in ways perceived to be nonthreatening to male control women vendors were able to subvert some of the regulations without being seen to be doing so. This point is taken up later in terms of how women devised strategies to cope with official demands by giving the impression of going along with what was expected of them when in fact they continued to ignore the regulations. The state operated in a masculine guise, trying to control and to undermine any gains women vendors had made in asserting an independent existence outside male control. State and municipal officials, like the women's representatives in the vendors' association, were mostly men. There was an emphasis on the observance of rules and regulations mostly detrimental to the well being of the women such as the insistence that they operate from a designated vending point. In November the area facing Aroma bakery was designated by Council as a new site for all vending operations in town. This was the space previously allocated to the women in the study. The other vending operations that were relocated to the site include those who sold non-food items such as watches, toys, doilies and art and craft products. The immediate result for the food vendors was increased market opportunities. Sales for the group of four women who operated on a permanent basis at the site shot up drastically. Another four women were also spending more time at the site selling their wares. It appeared as if the place had become more lucrative as a vending site due to increased volume of traffic involving people coming to the place because of the other types of vending operations. 84 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000 MuzvtdzKva / FOOD VENDING (ADAPTATION) Coping strategies Io N n GENERA or copin L vendin g with g urba coul n d life be viewe . In an d effor as a t strateg to ensur y fo e continue r hangin d g residence in town some respondents adopted strategies to enhance their income earning opportunities. Use of multiple income sources was quite common amongst food vendors, though the returns from these other sources were quite small. The women also reorganized their food procurement patterns and cooking practices so as to minimize waste. Cutting back on the number of meals was quite common as well as reduced meat intake. Co-residence as well as room sharing and adoption of the practice of split residence was common among Masvingo food vendors. This also had the tendency of reducing the cost of urban subsistence. In terms of vending operations the women tried to increase their returns by operating at multiple vending sites as weE as being semi-mobile. More than a third of food vendors increased returns by engaging child vendors who worked mostly as mobile vendors. Whilst the logic of urban living meant the need to minimize kin connections, for most of my respondents kin connections were being continually recreated and acted to underpin rural-urban connections. My respondents continued to define their rural village as their home (kurnusha). Despite the reduced visits to the rural areas due to economic difficulties many tried to remit to the village (see Table 3). The majority of my respondents (72 per cent) adopted what I call a double-rooted strategy in their day-to-day living experiences. They kept one foot in the village and another in town. In times of difficulty this strategy proved useful as they could always relocate. As argued in Muzvidziwa (forthcoming), under conditions of poverty such as was the experience of the food vendors, social networks, based mostly on friendship patterns and at times on neighbourliness, are used to help poor women to stay in town and to support their households. The use of institutions such as marounds to mobilize resources was common amongst the food vendors. They belonged to these maroamds as a means of raising cash for immediate basic needs as well as cash for business orders. JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 85 Although Masvingo food vendors had a limited amount of education, they had through experience learnt to be resourceful in order to survival. This meant an ability to survive under tight bureaucratic controls. In Bailey's (193: 9) terms, food vendors could be seen as svejks who, by pretending to be going along with official suggestions such as those emanating from meetings with municipal officials and police officers in October and November 1995, were able to carry on as before without attracting undue attention from the male official machinery. At these meetings women ululated and nodded at suggestions by officials as if in agreement, but nevertheless continued to sell on pavements. It is by appearing to go along with official regulations that many in Masvingo were able to survive and continue to sell on pavements. It is on the basis of this behaviour of my respondents that I conclude that coping strategies adopted by female food vendors are many and varied. Food vendor case study examples Ran ESPONDEN d was th T e B mothe , a 40-year-ol r of four, was d i woma n the hangin n wh g o o neve n socio-economi r married c group. B saw food vending as a career, having started at the age of 19. She was a one-room lodger. ESAP had forced her two younger children aged 11 and 13 to drop out of school. The older children were still in school. She was solely responsible for the welfare of her children. During my fieldwork B's 11 and 13 year-old children worked as full-time vendors alongside their mother. The children operated as mobile vendors in the town centre area (they did not sell from a fixed vending site, but moved from one point to another in search of customers) and were always on the look out for police. B was one of the few food vendors who spend most of'their time sitting at the designated vending site. She was part of a three-person group that took turns to source tomatoes from outside Masvingo. Fruits and vegetables were sourced locally in Masvingo. She belonged to a maround. By temporarily pulling the 11 and 13 year-old out of school B was able to fund the other two children's schooling. It appeared that the two child vendors were going to re-enroll in early 1996, as evinced by the school uniform B had bought which she showed me in 86 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000 Muzvidziwa / FOOD VENDING (ADAPTATION) early December 1995. B noted that food vending was a hard life and she was investing in the education of her children so that they would not be food vendors like her. Respondent R, a 35-year-old woman who never married, sold her wares mainly at the Mucheke bus terminus outside buses to the travelling public. Sometimes she sneaked inside a bus, a practice that was strongly discouraged by conductors. The work was very demanding as it was considered illegal and one had to keep a watchful eye on law enforcement agencies. She sourced her goods mainly from Masvingo. Prom July 1995 she started knitting and crocheting doilies on commission for a cross border-trading woman. She was introduced to this woman through an acquaintance, a fellow food vendor operating at the terminus. However due to increases in the rent, water and electricity charges in at the beginning of September, R was forced to relocate to the village sometime in October. R did not have an effective network to back her up during times of difficulty. Respondent D, a 55-year-old widow, had one child who was in Form 4. Her vending income was $150-$259 a month. In addition she received a small monthly stipend of $100 from the Social Welfare Department. She started vending operations in 1988 after the death of her husband. The kiosk which she used to operate before her husband's death was seized by her stepsons. It was quickly run down and closed. D sold sweets mainly to primary school going children in Mucheke. She operated from a shed at a designated area close to the Mucheke long distance bus terminus and purchased her sweets from a town wholesale. Although turnover was high the profit margin was very low. D noted, "Whatever I get from the sweets is almost immediately spent on food purchases. For me I just work for my stomach". With her indomitable spirit D was in the hanging-on group. August and September were particularly bad months for D. She had a near mental coEapse. It seemed that her economic status was going to improve in November when her stepdaughter moved out of the house and started lodging. D was thinking of taking in lodgers to assist her with the bills. Respondent S, a 27-year-old divorcee, had moved by the end of 1995 from the hanging-on category to the coping group. In early 1995 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 87 she had an income of at least Z$600 per month. Her five-year-old daughter lived with her. She had started vending in March 1995, having borrowed her start-up capital of $100 from a maternal aunt. She specialized in selling mealie meal and peanut butter and operated from an open space near Mucheke long distance bus terminus as well as from her own home. She was a registered vendor who was supposed to operate from a Council-designated selling point. She operated seven days a week. Despite price increases, with the price of maize almost doubling in nine months from $18 in February 1995 to $35 per bucket in October 1995, the demand for mealie meal proved constant which had helped her to offset the loss of a market for peanut butter. She packaged the mealie meal in small quantities of a kilogram which was convenient for her customers' needs as commercially packaged mealie meal was sold in a minimum of five kilogram packs. Her hammer-milled mealie-meal was sold at about half the price of the commercially produced type1. S crocheted and knitted for a cross-border trading woman on commission. She had been introduced to this woman through her church contacts. In addition she also collaborated with another woman knitting for a business person who had an export contract for jerseys and clothing materials. This contact came as a result of a casual discussion with a client of hers. S had a thriving garden at her home. She sold up to $10 worth of vegetables to neighbours on a monthly basis. Up to June 1995 she also sold secondhand clothes which she ordered from Beitbridge, but had stopped this line of business. Respondent V, a 29-year-old divorcee and a mother of eight-yearold twins, depended on food vending as her only source of income. V self-financed her start-up operations. She had five years of food vending experience. She was an assertive and determined woman, who sourced her tomatoes from outside Masvingo. V also ordered bananas twice a week from a Mutare delivery truck. She sold 1 A study conducted by the Ministry of Public Service, Labour and Social Welfare, in collaboration with the United Nations Children's Fund and reported in the Herald of February 17, noted that 5 kg. of roller meal was selling at $10 whilst 5 kg. of hammer-milled mealie meal was going at $5.50. 88 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2000 Muzvidziwa / FOOD VENDING (ADAPTATION) vegetables and an assortment of fruit bought locally and outside Masvingo. She banked at least $200 a month in her Post Office Savings Bank account. V had registered on the Council's housing waiting list and was optimistic that, sooner rather than later, she would get a stand on which to build a house. V was the only food vendor in the climbing-out-of-poverty group with an estimated monthly income of $1000. She employed non-kin, two but mostly one at a time, to assist her with sales. They worked as town mobile vendors. She shared her accommodation with her helpers (as she calls them) up to September, when she started cohabiting with a man. She then stopped providing them with accommodation and food, but raised their wages from $60 to $100 per month but continued to provide her assistants with lunch in the form of a bottle of coke and a loaf of bread. Concluding remarks Wwome HILE TH n engage E POOLIN d chil G o d vendor f incom s i e wa n thei s no r operation t common s in orde , som r to e boost returns. Food vending was hard work. Food vendors worked for fifteen hours a day on average. The women were not interested in politics other than the politics of the stomach. Quite often those struggling to make ends meet combined vending with other secondary income-earning activities. A house, as in the case of D, was the most effective form of investment as it could be turned into a source of income at some stage. The single greatest threat to food vendors was official policy of harassment and arrest. Yet food vending had become an integral part of the city. Vendors' perceptions and those of officials differed greatly, hence the need to come up with policies and regulations that are less restrictive as far as income earning opportunities for the poor are concerned. While food vending offered a means for survival for many town women, increased competition was undercutting profitability and when combined with police harassment it forced some of the women to relocate to the village. This brings us to one of the most important observations. Food vendors adopted a strategy of double-rootedness in case they were forced to relocate to the village due to difficult economic circumstances. Rather than undermining social networks JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 89 ESAP had in fact strengthened some as it was increasingly becoming clear that these networks were a major resource that enabled many to hang on and to cope with urban life. Official constraints imposed on the women posed a greater risk to the success of their operations than any other factor. Although vending was not easy for many female-headed households it was all they had to depend on for a living. In Masvingo vendors were an integral part of the city and they interacted with formal sector operators in many ways. They were a vital cog in the fresh food distributive network. The way food vendors were treated by officials demonstrated the existence of competing and conflicting rationalities between decisionmakers and poor vendors. Yet Masvingo vendors' concern was to engage in activities that contributed to the livelihood of their households. Where they broke the law and regulations it was mostly because of the restrictive nature of these officially imposed constraints, rather than inherent criminality among my respondents. REFERENCES Bailey, F.G. 1993. The Kingdom of Individuals. An Essay on Self- Respect and Social Obligation, Ithaca, Cornell University Press Brand, V., Mupedziswa, R. and Gumbo, P. 1995. Structural adjustment. women and informal trade in Harare. In P, Gibbon (ed.) Structural Adjustment and the Working Poor in Zimbabwe. Uppsala, Nordiska Afrikainstitutet 132-214 Census 1992. Provincial Profile Masvingo 1993. Harare Central Statistics Office Cheater, A.P. 1979. The production and marketing of fresh produce among blacks in Zimbabwe, Supplement to Zambezia Harare, University of Zimbabwe Cutrufelli, M.S. 1983. Women of Africa; Roots of Oppression. London: ZED Hansen, K.T. 1980. The urban informal sector as development issue: poor women and work in Lusaka, Zambia Urban Anthropology 9, 2, 199-225 90 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 No. 2 JULY 2 0 0 0 Muzvidziwa / FOOD VENDING (ADAPTATION) Kamidza, R. 1994. Structural adjustment without a human face. Southern Africa Political and Economic Monthly Review 7, 6 Kanji, N. and Jazdowska, N. 1993. Structural adjustment and women in Zimbabwe. Review of African Political Economy 56 11-26 Muzvidziwa, V.N. 1997. Rural-urban linkages: Masvingo's doublerooted female heads of households. Zambezia 24, 2 97-123 Networks as survival strategy: female heads of households in Masvingo, Zimbabwe. South-South Journal of Culture and Development (forthcoming) Osirim, M.J. 1992. The status of women in the third world: the informal sector and development in Africa and the Caribbean. Social Development Issues 14, 2 & 3 74-87 —1995 Trading in the midst of uncertainty: market women, adjustment, and the prospects for development in Zimbabwe. African Rural and Urban Studies 2, 1 —1998 Negotiating identities during adjustment programs: women microentrepreneurs in urban Zimbabwe. In A. Spring and B. McCobe, (eds.) African Entrepreneurship: Theory and Reality. Florida: University of Florida. JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 15 NO. 2 JULY 2000 91