345
Views
5
CrossRef citations to date
0
Altmetric
Original Articles

Use of wealth ranking to analyse factors influencing smallholder farmers' market participation in northern Mozambique

Pages 669-683 | Published online: 08 May 2007

Abstract

This study analysed factors influencing smallholders' market participation, using wealth-ranking factors. Two hypotheses were tested: that (1) wealth status and (2) wealth-ranking factors are positively related to market participation. Significant and positive relationships were found between wealth-ranking factors (labour, number of livestock, implements, bicycles, food availability, area of land cultivated and crops sold) and wealth status. Wealth status and wealth-ranking factors were positively and significantly correlated with the number of different kinds of cash crops sold. However, household characteristics not indicated as wealth-ranking factors, such as age, gender and literacy level, related insignificantly to market participation. Labour was the most important wealth-ranking factor explaining market participation. This analytical tool can be used to assess the wealth-ranking factors that influence market participation. It can help identify strategies for improving this participation and may also be used to assess the way a cash crop development project affects a household's wealth status.

1Respectively, Professor and Head of Discipline of Community Resources, School of Agricultural Sciences and Agribusiness, University of KwaZulu-Natal, Pietermaritzburg; Agricultural Manager, World Relief International; Lecturer, Rural Resource Management; and Professor, Applied Plant Sciences, School of Agricultural Sciences and Agribusiness, University of KwaZulu-Natal, Pietermaritzburg. The authors gratefully acknowledge the co-operation of the farmers, the João Ferreira dos Santos Company, German Agro Action, OXFAM and Acção Cristã Interdenominacianal da Saúde. The authors also express gratitude to World Relief International and the Swedish International Development Agency in Mozambique for making available the WR-SempreVerde facilities to collect the data used in this study.

1. INTRODUCTION

Mozambique is one of the poorest countries in the world. Agriculture already accounts for about 28 per cent of GDP and households are heavily dependent on it for food and income (Arndt & Tarp, Citation2001). The unemployment rate for formal work was about 98 per cent at the time of this study (Information from Cuamba District Labour Officer). The authors observed that remittance money was insignificant because of the lack of employment opportunities in towns around southern Niassa. They also observed sales of local construction material, fuel, processed food, and handcraft, and provision of labour to others, small-scale mining of precious stones and trading activities, but these absorbed very few people.

As Mozambique has a population density of only 22 people per square kilometre, and southern Niassa only 13, land is a non-limiting factor (Instituto Nacional de Estatística, Citation1997). The economy of Niassa Province, the poorest of Mozambique's provinces, is likely to benefit if appropriate support is provided to improve smallholder farmers' market participation. Therefore, donors and investors were willing to financially support agricultural market development activities to improve smallholder farmers' income in southern Niassa.

To design suitable strategies for agricultural market development projects, it is necessary to understand the factors that influence market participation. These factors are often identified by surveying or measuring all those that could be influencing the situation, then using statistical tests to identify the ones that are significantly related to the assessed indicator of agricultural market participation (Makhura, Citation2001; Matungul, Citation2002). The authors of this study stress that a simple, quick and cost-effective way of identifying these factors is to ask the farmers themselves to indicate which factors they feel influence their participation in the agricultural market. This method is effective because the relevance of factors influencing smallholder agricultural market participation changes with time and space. In their study of farmers' cash crop cultivation decisions, Lukanu et al. Citation(2004) asked smallholder farmers in southern Niassa, Mozambique to indicate the factors that influenced their decision to cultivate certain cash crops. These farmers identified, in order of importance, the profitability of crops, the availability and reliability of the buyers, the availability and accessibility of inputs and implements and labour, and access to extension services.

However, there is a general tendency for smallholders to point to exogenous household constraints on agricultural market participation, such as lack of buyers, lack of extension services, lack of or excessive cost of inputs, lack of credit and low market prices for their outputs, rather than identifying endogenous household constraints. Yet many studies have indicated that household characteristics such as age, education, gender of the household head, household wealth and household size can influence household participation in the agricultural market (Kebede et al., Citation1990; Makhura, Citation2001; Matungul, Citation2002).

This study proposes the use of wealth-ranking tools as a way of assessing those sensitive factors or endogenous household factors that relate to market participation. In this study market participation is reflected as the number of different kinds of promoted cash crops sold. Promoted cash crops are crops such as tobacco, cotton, sesame, sunflower and paprika that have been promoted by specific institutions for income generation purposes only. Food cash crops are mainly maize, cassava, sorghum, boer bean, cowpea and vegetables. Where surplus food crops were planted for sales, they were included in the study. Where this was purely incidental, such sales were not included.

Mearns et al. Citation(1992), Ghirotti Citation(1992) and Simanowitz Citation(1999) suggested that with wealth-ranking techniques one can extract sensitive socio-economic aspects of households that influence wealth status. Key informants can provide the criteria used in their village to characterise households according to wealth status. In addition, wealth-ranking techniques also help researchers, through simple observation, to pinpoint those households that can be classified as poor, middle class or wealthy, depending on the levels used in each community.

The general objective of this study is to find out whether wealth-ranking tools can be used to analyse or identify those factors that affect smallholders' participation in the agricultural market. The study has the following specific objectives:

  1. to analyse the relationship between wealth status and market participation; and

  2. to analyse the relationship between specific wealth-ranking factors and market participation.

If wealth ranking as a tool is shown to be reliable, it will be a useful adjunct for assessing the valid and locally specific factors that influence smallholder participation in the agricultural market. Wealth-ranking tools may be more accurate than more formal data collection and analytical tools that are based on assumptions made by the researchers (Abalu et al., Citation1987). Use of simple, rapid and cost-effective tools will allow the identification of up-to-date data that can be used to design appropriate project strategies for improving smallholder participation in the agricultural market.

2. ANALYTICAL METHODS

This section discusses the basis for the theory of using community-based wealth factors to determine whether there are any relationships between these and market participation. It also explains the statistical methods used to examine this relationship.

2.1 Theoretical analysis

Two hypotheses were tested: that (1) wealth status and (2) wealth-ranking factors are positively related to market participation. In his study on transaction cost barriers to market participation, Makhura Citation(2001) argued that the availability of financial assets or household endowments, which are proxies for wealth status, influences household market participation behaviour. Even farm size, which many authors have found to significantly correlate with agricultural market participation, is a proxy for wealth status (Kalinda et al., Citation2000; Matungul, Citation2002). Factors influencing market participation can be grouped according to household characteristics and economic, institutional and environmental factors (). These factors will vary because these variables are time and space dependent. Nevertheless, the cause–effect relationship between wealth status and market participation may work in both directions. The[0] Banlina & Tung Citation(1992) study in the Philippines found that households' involvement in agricultural production activity (and therefore participation in the agricultural market) determined their wealth status. However, the present study is limited to analysing wealth status and wealth-ranking factors as influences on smallholder farmers' market participation.

Figure 1: Factors affecting farmers' decisions to cultivate cash crops. Source Lukanu et al. Citation(2004).

Figure 1: Factors affecting farmers' decisions to cultivate cash crops. Source Lukanu et al. Citation(2004).

2.2 Statistical analysis

The relationship between wealth status, wealth-ranking factors and market participation was first analysed using qualitative perceptions provided by key informants ().Footnote2 Descriptive statistics, correlation analyses and z-score test or confidence limits were used to discover whether there was a significant relationship between wealth status, wealth-ranking factors and market participation; and to observe whether the proportion or mean of wealth status, wealth-ranking factors and the number of different kinds of cash crops sold was statistically different for poor, middle class and wealthy households. All significance was reported at the α = 0.05 (P95 per cent) level. The wealthy category was used as the standard for comparison because middle class and poor households desired a wealthy household's standard of living (which in Mozambique is still a comparatively low standard).

Table 1: Village leaders' criteria for differentiating household wealth status in southern Niassa

3. METHODOLOGY

3.1 Sampling

A combination of cluster and stratified sampling was used in the study. The random selection of one or two villages for each of the seven major routes in Cuamba district ensured that the study results could be generalised for these districts. Interviewers were told to identify a specific number of households representing a proportional distribution of certain characteristics such as wealth status and the gender of the household head. Visual indicators of poverty (VIP), based on housing criteria such as the size, number and condition of the house or houses, the cleanness of the compound and the state of the occupants' clothing, were used for wealth-ranking (Ghirotti, Citation1992; Simanowitz, Citation1999). Each interviewer was instructed to interview 32 household heads from 19 poor, nine middle class and four wealthy households. Interviewers were also asked to select proportional numbers of male- and female-headed households for each wealth status category. These proportions reflected the populations as obtained by Lukanu et al. Citation(2004) in southern Niassa in the previous year, with ranking performed by the village leaders. The study assumed that these proportions would not have changed significantly a year later.

3.2 Research tools

The study was preceded by focus group discussions involving key informants: traditional, church, mosque, political, government and opinion leaders. Key informants were asked what factors they would use to rank households according to wealth status. Many similar concepts were identified, and these were used to develop the wealth-ranking factors and the description for each wealth category (). These discussions also served as the basis for the design of questionnaires. Households were selected to respond to a set of questions. Where household heads were absent, spouses were allowed to respond. Respondents were also allowed to consult the members of the household for answers to questions about which they were uncertain.

3.3 Data collected

The following household data were collected: the number of external or casual labourers, known as ganho-ganho,Footnote3 the number of agriculturally active members, the number of members providing services to other households' fields, the number of livestock owned, the number of implements, the number of bicycles, the wealth status rank (poor, middle class or wealthy), the number of promoted and food cash crops sold in that current year, farming orientation, whether for cash or for consumption and, finally, age, gender, marital status and literacy level.

The focus of this study was to observe how a wealth-ranking tool as proposed by local village leaders during focus group discussions could be used to identify and analyse those factors that in reality influence agricultural market participation. This tool also used housing conditions to identify smallholders' wealth status. However, it may have failed to identify the households' wealth status accurately because (1) the household head might have been unwilling to invest income in improving housing conditions and (2) a household might be living in a good-looking house despite actually being poor. Nevertheless, the present study considered that visual indicators were the simplest way of identifying households' wealth status.

4. RESULTS AND DISCUSSION

4.1 Wealth-ranking factors and wealth status

shows 13 wealth-ranking factors and their variations across various wealth-ranking tools as identified by village leaders. The first seven factors, related to agricultural market participation, are analysed in this study. The interviewers used some of the remaining wealth-ranking factors, such as housing and clothing conditions, to select households according to wealth categories. An attempt was made to confirm the relationship between wealth-ranking factors, household characteristics not indicated as wealth-ranking factors, and wealth status (). The expectation is that these factors should correlate positively with smallholders' agricultural market participation if they correlate positively with wealth status.

Table 2: The relationship between some wealth-ranking factors and household characteristics, and wealth status

The correlation coefficient (0.29) between effective household labourFootnote4 and wealth status was positive and significant (), in accordance with key informants' suggestions (). Wealthy households' effective labour was significantly higher than that of the poor (z-score = 4.77) and middle class (z-score = 2.93) households. The number of labourers was also found to be a criterion for identifying households in different wealth categories in Gambia (Mearns et al., Citation1992).

Wealthy households employed significantly more casual labour (1.69) than poor (0.32) and middle class (0.96) households, and significantly fewer members of wealthy households (0.31) were employed on other households' farms than members of poor (0.79) and middle class (0.66) households. By working on other people's farms, poor households reduced their own available household labour and, as a result, became more dependent on casual labour to feed themselves from their own fields, and were increasingly obliged to work on other people's farms for quick pay. These poor farmers often ended up entering the cycle of poverty in which each action for survival could potentially lead them to further impoverishment. It can thus be hypothesised that the number of effective labourers will be related to market participation because it relates significantly to wealth status.

shows that livestock (mostly chicken farming) was positively and significantly (r = 0.40) related to wealth status and there was a significant difference (α = 0.05) between poor (z = 7.10) and middle class livestock ownership (z = 5.14) when compared to wealthy households, in confirmation of the factors identified during the focus group discussions, shown in . Based on observation, farmers used livestock more for income generation than for consumption. For example, villagers supplied the city of Cuamba with locally produced chickens. It can thus be hypothesised that livestock, a wealth-ranking factor, is likely to be related to market participation.

The correlation coefficient between the number of implements owned and wealth status was positive and significant (r = 0.33). Wealthy households had significantly more implements than poor and middle class households. This supports the claim made in the focus group discussions that the number of implements was related to wealth status. It is expected that the number of implements will also be positively related to smallholders' market participation.

The wealthy households had significantly more bicycles than the poor and middle class ones (z-scores = 9.97 and 4.48), and significantly more wealthy households (94 per cent) owned bicycles than the poor ones (61 per cent, z = –3.87). One would expect that the number of bicycles would be positively related to market participation because bicycles correlated significantly (r = 0.30) with wealth status, in accordance with the leaders' information.

Key informants reported that wealthy households had enough food for consumption whereas poor ones often did not. Ideally the study should have assessed the amount of food that the households possessed. However, food availability was not assessed because it relied on other agricultural factors that were examined. shows that the majority of the households (93 per cent) cultivated primarily to produce food for their own consumption while only six per cent cultivated primarily to generate income. Although there was no significant difference, wealthy farmers tended to give lower priority to food cultivation than poor households, possibly because they had a larger food reserve. Guijit Citation(1992) also related wealth status to how long food lasted within the household. Food availability will dictate the status of the household, the partitioning of resources for food or cash crop cultivation and subsequently market participation. As Sarch Citation(1992) explained, wealthy households have year-round food security, while the poorest regularly beg for food.

Because of a lack of (quantitative) data, the relationship between the size of the cultivated land and wealth status is not discussed here on the basis of statistical analysis. However, the study assumes and supports key suggestions that the size of the cultivated area is positively related to wealth status. The size of the land should be dependent on the number of effective labourers and will probably also influence the number of cultivated cash crops and resultant food availability.

The farmers' ages were not significantly related to wealth status. Based on the converse hypothesis, age and literacy level will be poorly related to smallholders' market participation because these characteristics are not indicated as wealth-ranking factors. However, the relationship between gender/marital status of the household head and wealth status was significant and positive (r = 0.23). Data revealed that there were no wealthy households among female-headed households, while all wealthy households were headed by a married male. Further, the analysis shows that these households had higher numbers of effective labourers (2.4) than female-headed households (1.4), and the number of effective labourers in female-headed households (1.4) was smaller than in poor households (1.8). That is, female-headed households could be classified as the poorest of the poor in southern Niassa, and therefore less likely to participate in agricultural markets. Carrilho et al. Citation(2003) also found that female-headed households were in the low-income group in Mozambique, and Mukherjee Citation(1992) arrived at similar conclusions in a study undertaken in India. It can be concluded that gender/marital status is related to wealth status through effective available labour.

4.2 Wealth status and market participation

The income from the sale of cash crops is one of the best indicators of the degree of agricultural market participation (Makhura et al., Citation1997). In particular, it was assumed that farmers who cultivated a variety of cash crops could be classified as good participants in the agricultural market. This is because crop diversification serves as a central strategy in managing agricultural production risk (Lukanu et al., Citation2004). The number of different kinds of promoted cash crops sold was therefore used here as a good indicator of the level of market participation.

There was a significant and positive correlation between wealth status and cash crops sold (r = 0.27). The average number of different kinds of promoted cash crops sold by wealthy farmers was significantly higher than the ones from poor households (z = 3.88) (). Based on this analysis and the perceptions provided by key informants () it can be concluded that wealth status is positively related to market participation. Makhura Citation(2001) also concluded that households with more resources would be in a much stronger position to participate in agricultural markets.

Table 3: Significance and correlation between wealth status and number of different kinds of cash crops sold

4.2.1 Cash crop sold

Overall, 74 per cent of the sample sold at least one cash crop (). This proportion is considerably larger than the 29 per cent estimated by Heltberg & Tarp Citation(2002) for Mozambique. The difference between the datasets can be attributed to time lag, and the different economic and climatic factors represented in the two research projects. In the present study, the percentage of wealthy farmers participating (83 per cent) was significantly higher (z-score = 2.21) than the percentage of poor ones (64 per cent). Thirty-eight percent of the respondents sold at least one food cash crop compared to 60 per cent who sold at least one promoted cash crop. Households showed a greater inclination to sell promoted cash crops than was observed in the previous year, when 63 per cent of the respondents sold food cash crops, compared to 37 per cent who sold promoted cash crops (Lukanu et al., Citation2004). The high profitability of tobacco, and support by the João Ferreira dos Santos Company (JFS), influenced households to shift from selling food crops in the 2000/2001 season to selling tobacco in 2001/2002 ().

Table 4: Different wealth categories cultivating food crops or promoted cash crops

Table 5: Farmers participating in the sale of only promoted or food crops, both promoted and food cash crops, and not participating in the market at all

shows that significantly more respondents (36 per cent) marketed promoted cash crops only than respondents who marketed food cash crops only (13 per cent), both promoted and food cash crops (25 per cent) and none (26 per cent). Poor farmers participated more (17 per cent) in the sale of food crops than middle class farmers (11 per cent) and wealthy farmers (8 per cent). Significantly more of the poor farmers (37 per cent, z-score 2.28) market no crop at all than wealthy farmers (17 per cent).

4.3 Wealth-ranking factors and market participation

Four wealth-ranking factors (numbers of labourers, livestock, implements and bicycles) were statistically related to the number of cash crops sold. The relationship between the size of the cultivated land and market participation were also analysed qualitatively based on information from key informants. The relationships between household characteristics such as age, gender, marital status and literacy level and market participation were also analysed. It is assumed that these household factors would have a weak relationship with market participation because they were not indicated as wealth-ranking factors.

4.3.1 Number of labourers and agricultural implements

There were significant correlations between the number of effective labourers (r = 0.18), number of implements (r = 0.53) and cash crops sold. It is postulated that a positive correlation would have been obtained between field size and the number of cash crops sold, given the relationship between field size and the number of effective labourers, as found by Makhura Citation(2001). Promoters need to bear in mind that labour is a major constraint on cash crop cultivation, particularly for poor households. Therefore, to involve poor households, promoters need to look for solutions that will improve the access to labour for the poor. By increasing labour capacity or promoting less labour-intensive cash crops, poor households will be able to become involved in the cultivation of cash crops and subsequently become fuller participants in the agricultural markets.

4.3.2 Livestock ownership

The money from livestock sales was significantly and positively (r = 0.25) related to the number of cash crops sold, in confirmation of the second hypothesis of this study which stated that there was a relationship between specific wealth-ranking factors identified by the village leaders (in ) and actual market participation. (The first hypothesis tested the relationship between actual wealth status and market participation.) Funds raised from the sale of livestock, the ownership of which was proposed as a wealth-ranking factor by village elders, were used for urgent household needs. On average, all livestock owned by a wealthy farmer would be valued at about $US68 compared to $US18 for those owned by a middle class farmer and $US4 for those owned by poor households. This money realised from wealthy household livestock sales could be used to pay casual labourers to cultivate an area of 0.51 ha of tobacco (casual labourers are paid $US133 to cultivate one hectare of tobacco) compared to 0.14 ha for middle class and 0.03 ha for poor households. The income from a wealthy household's livestock could be used to buy 326 kg of maize grain, compared to 86 kg and 19 kg for middle class and poor households, respectively.Footnote5 These data give an indication of how secure a wealthy household is in terms of paying external labourers or buying food for consumption using money from the sale of its livestock. This security gives wealthy households more flexibility to participate in agricultural markets than poor ones.

4.3.3 Bicycle ownership

Heltberg & Tarp Citation(2002) indicated that there was a relationship between cash crop sales and bicycle ownership in Mozambique, and that the causality ran in both directions. Bicycles made it possible to transport crop products from the fields and villages to the buyers, local merchandise (such as construction material, firewood and charcoal) from the villages to the city, and manufactured products from the city to the villages. A combination of farm and non-farm income-earning activities has long been an adaptive strategy that allows farmers to reduce risks (Gladwin et al., Citation2001). Households with reduced risk are more likely to participate in the agricultural market. However, in this study the number of bicycles was not significantly related (r = -0.03) to the number of cash crops sold.

Nevertheless, the relationship improved (Person R = 0.12), but was still not significant at α = 0.05, when cash crop data were related to bicycle ownership. Fifty-four percent of the respondents who did not have a bicycle did not sell promoted cash crops. However, there were also 35 per cent of smallholders who had at least one bicycle but who did not sell cash crops. It is possible that bicycle ownership was not a determining factor if buying points were close to the village, as the buying brigades from large agribusiness companies and informal traders bought at buying points close to (7 km maximum distance) or within the villages. Nevertheless, farmers with a bicycle who could transport their products to the city could sell them at a higher price. It is therefore expected that people owning at least one bicycle were likely to make more profit and potentially participate efficiently in the agricultural markets.

4.3.4 Age, gender/marital status and literacy level

The r-values of 0.05 and 0.11 respectively suggest a non-significant relationship between age, gender, marital status and number of cash crops sold. This is in line with the converse hypothesis that factors not used for wealth-ranking are less likely to be related to market participation where agriculture is the main economic activity. Makhura Citation(2001) also found that gender differences did not affect the level of market participation of some agricultural commodities in the Northern Province of South Africa.

However, r-values of –0.15 between literacy level and cash crop sold reflect that the relationship was significant (α = 0.05) and negative. Households with a more educated household head did not sell more cash crops than those headed by uneducated heads of households. Makhura Citation(2001) found that literacy level related negatively (but not significantly) to maize sales. shows that educated people were predominantly younger, unmarried and poor. It is known that the labour requirements for younger farmers are high because they need to clear virgin land to start their field activities, whereas older farmers have already cleared most of their fields. At the same time, younger, single and poor-headed households may have limited financial resources to pay for external labourers, which negatively affects their potential for market participation.

Table 6: The relationship between household characteristics, agricultural factors and literacy levels

In addition, the educated have low livestock ownership. Lack of livestock is negatively related to market participation. Further, educated people may perceive agriculture as an activity for the uneducated. For example, 88 per cent of the educated did not market any promoted cash crop, whereas 36 per cent of the less educated did. Twelve percent of highly educated households did not cultivate a single food crop, whereas all of the less educated cultivated some food crops. Perhaps highly educated smallholders wait for ‘good’ jobs outside agriculture (but in vain, because job opportunities are limited).

4.3.5 Most important factors explaining market participation

A stepwise regression was used to analyse the factors that best explained smallholders' market participation. The number of effective labourers explained 20 per cent of the variation in the number of different kinds of cash crops sold, while the number of livestock, number of implements and ownership of bicycles together explained 35 per cent of the variation. Overall, the number of effective labourers was the most important factor if one takes into consideration isolated factors. However, the combined effect of the number of implements, number of livestock and bicycle ownership strongly contributed to the variation in the number of cash crops sold.

5. MODEL AND STRATEGIES FOR CASH CROP DEVELOPMENT IN NIASSA PROVINCE

This study confirmed that wealth-ranking factors are likely to be related to market participation where agriculture is the main economic activity. Agricultural crop promoters can therefore use this wealth-ranking tool to quickly assess factors related to smallholders' market participation and use these to design strategies to help farmers participate in the agricultural market. For example, farmers may respond to the following strategies:

  1. Provision of agricultural inputs and services, on a credit basis, to allow smallholders to increase crop profitability with limited labour.

  2. Promotion of less labour-demanding crops so that poor households can use their limited effective labour to cultivate food and cash crops.

  3. Shifting the cultivation of cash crops from the rainy to the dry season, using lowland or small-scale irrigation schemes, so as to lower labour requirements during the rainy season.

  4. Introduction of livestock, to increase smallholders' food availability and income. Income from livestock may be used to buy inputs, pay for labour and buy food. Livestock can also provide manure for improved soil fertility, to increase crop yield, profitability and subsequently market participation.

Wealth-ranking tools may also be used to evaluate a cash crop project by observing whether:

  1. wealth-ranking factors have changed after the implementation of the project, for example, more external labourers are being used or the number of livestock has increased; and

  2. farmers have moved from one social status category to another since the implementation of the project.

6. CONCLUSIONS

In places where agriculture is the main production activity, wealth can generally be improved only by participating in the agricultural market. This participation can be stimulated by applying strategies based on appropriate identification of the factors that influence it. Use of wealth-ranking tools can provide wealth-ranking factors for predicting market participation. Although the reverse was not analysed in this study (because this was not the aim), it is likely that market participation also influences wealth status and wealth-ranking factors. That is, wealth status is likely to have a positive influence on smallholder farmers' agricultural market participation and vice versa. Wealth-ranking tools also provide a quick, simple and cost-effective means of identifying the factors that can be used to design strategies for improving smallholder farmers' participation. For example, bearing in mind that the number of effective labourers is a wealth-ranking factor, one can improve participation by helping smallholder farmers, through credit, to hire more labourers, to use the right technology so as to increase their productivity, or to use their available household labour efficiently throughout the year (that is, using irrigation during the dry season).

Obviously, wealth-ranking tools can also be used to evaluate cash crop projects by checking whether the wealth-ranking factors have changed for better or worse, or whether households have moved from one wealth status to another after the implementation of an agricultural market development project. This is because agricultural market development projects aim to increase smallholder farmers' income. This increased income can be used for routine household expenses, while the extra income can be accumulated as household wealth in the form of more cultivated land, more livestock and more hired casual labour in the household fields.

Notes

1Respectively, Professor and Head of Discipline of Community Resources, School of Agricultural Sciences and Agribusiness, University of KwaZulu-Natal, Pietermaritzburg; Agricultural Manager, World Relief International; Lecturer, Rural Resource Management; and Professor, Applied Plant Sciences, School of Agricultural Sciences and Agribusiness, University of KwaZulu-Natal, Pietermaritzburg. The authors gratefully acknowledge the co-operation of the farmers, the João Ferreira dos Santos Company, German Agro Action, OXFAM and Acção Cristã Interdenominacianal da Saúde. The authors also express gratitude to World Relief International and the Swedish International Development Agency in Mozambique for making available the WR-SempreVerde facilities to collect the data used in this study.

2 shows what the village leaders perceived the wealth factors to be and the remaining data were collected to see whether these perceptions were accurate in reality.

3 Ganho-ganho are people employed by the household to provide agricultural work in exchange for money, food, drink, cigarettes or clothes.

4The effective household labour was estimated as the sum of active members and ganho-ganho minus the number of members working in other people's fields.

5The price of 0.21 $US/kg of maize was used in this calculation.

REFERENCES

  • Abalu , D OI , Fisher , N M and Abdullahi , Y . 1987 . Rapid rural appraisal for generating appropriate technology for peasant farmers: Some experiences from northern Nigeria . Agricultural Systems , 23 : 311 – 24 .
  • Arndt , C and Tarp , F . 2001 . Who get the goods? A general equilibrium perspective on food aid in Mozambique . Food Policy , 26 : 107 – 19 .
  • Banlina , F T and Tung , L . 1992 . Farming's experience of wealth ranking in the Filippines: Different farmers have different needs. Special issues on application of wealth ranking . RRA Notes , 15 : 48 – 50 .
  • Carrilho , J , Benfica , R , Tschirley , D and Boughton , D . 2003 . Qual é o papel da agricultura familiar commercial no desenvolvimento rural e redução da pobreza em Mozambique? , Maputo : Ministério da Agricultura e Desenvolvimento Rural . Relatório No. 53P
  • Ghirotti , M . 1992 . A simple method for scoring housing conditions as income proxy in Ethiopia. Special issues on application of wealth ranking . RRA Notes , 15 : 43 – 7 .
  • Gladwin , C G , Thomson , A M , Peterson , J S and Anderson , A S . 2001 . Addressing food security in Africa via multiple livelihood strategies of women farmers . Food Policy , 26 : 177 – 207 .
  • Guijit , I . 1992 . The elusive poor: Wealth of ways to find them. Special issues on application of wealth ranking . RRA Notes , 15 : 7 – 13 .
  • Heltberg , R and Tarp , F . 2002 . Agricultural supply response and poverty in Mozambique . Food Policy , 27 : 103 – 24 .
  • Instituto National De Estatística . 1997 . II recenseamento geral da população e habitação, resultados definitivos, Provincia de Niassa , Maputo, Mozambique : Instituto Nacional de Estatística .
  • Kalinda , T , Filson , G and James , S . 2000 . Resources, household decision making and organisation of labour in food production among small-scale farmers in southern Zambia . Development Southern Africa , 17 : 165 – 74 .
  • Kebede , Y , Gunjal , K and Coffin , G . 1990 . Adoption of new technology in Ethiopian Agriculture: The case of Tegulet–Bulga District, Shoa Province . Agricultural Economics , 4 : 27 – 43 .
  • Lukanu , G , Green , J M , Worth , S and Greenfield , P . 2004 . Farmers' cash crop cultivation decisions in Southern Niassa Province, Mozambique . Development Southern Africa , 21 ( 3 ) : 531 – 54 .
  • Makhura , M T . 2001 . Overcoming transaction costs barriers to market participation of smallholder farmers in the Northern Province of South Africa , South Africa : Department of Agricultural Economic, Extension and Rural Development, University of Pretoria . PhD thesis
  • Makhura , M T , Goode , M F and Coetze , G K . 1997 . Indexing participation in the market economy through factor analysis: Implication for food security . Agrekon , 35 ( 4 ) : 473 – 83 .
  • Matungul , M P . 2002 . Marketing constraints faced by communal farmers in KwaZulu-Natal, South Africa: A case study of transaction costs , Pietermaritzburg : University of Natal . PhD Thesis
  • Mearns , R , Shambodon , P J , Narangered , G , Turul , U , Enghamgalan , A , Bayanjargal , A and Bethsuren , B . 1992 . Direct or indirect uses of wealth ranking in Mongalia. Special issues on application of wealth ranking . RRA Notes , 15 : 29 – 38 .
  • Mukherjee , N . 1992 . Villagers' perception of rural poverty through the mapping methods of PRA. Special issues on application of wealth ranking . RRA Notes , 15 : 21 – 6 .
  • Sarch , M . 1992 . Wealth ranking in the Gambia: Which households participated in the FITT programme? Special issues on application of wealth ranking . RRA Notes , 15 : 14 – 20 .
  • Simanowitz , A . 1999 . Pushing the limits of mapping and wealth ranking. Learning from analysis . PLA Notes , 34 : 4 – 8 .

Reprints and Corporate Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

To request a reprint or corporate permissions for this article, please click on the relevant link below:

Academic Permissions

Please note: Selecting permissions does not provide access to the full text of the article, please see our help page How do I view content?

Obtain permissions instantly via Rightslink by clicking on the button below:

If you are unable to obtain permissions via Rightslink, please complete and submit this Permissions form. For more information, please visit our Permissions help page.