Government, donors and NGOs in southern Niassa have been, after the 1992 peace agreement, extensively involved in agricultural development programmes to improve smallholders' food security. A study of the area and literature review revealed that many factors limited the benefits of agricultural market development programmes. Yet, opportunities in southern Niassa suggested that appropriately designed programmes could improve the standard of living of smallholders if these programmes were designed on a solid understanding of factors and strategies influencing agricultural market participation by smallholders. The main research hypothesis of this study was that: smallholders would participate in agricultural markets when their wealth status was high, when they had enough available household labour and when cash crops were profitable." Four main hypotheses were investigated: (i) factors and strategies identified through smallholder perceptions would provide local and time specific information on the constraints and solutions to smallholder market participation; (ii) wealth status and wealth-ranking factors were positively related to market participation where agriculture was the main economic activity as in southern Niassa; (iii) labour aspects such as crop labour requirements (CLR) could be negatively related, while available household labour (AHL) and the ratio AHL/CLR could be positively related to smallholders cultivation of cash crops and subsequent participation in agricultural markets; and (iv) aspects of profitability and indicators could be used to predict smallholder cash crop preferences. Data for this study were collected in Cuamba district of Mozambique from nine focus group discussions (FGDs) with community leaders, 287 household-head questionnaires and staff interviews during September 2002. Nine villages were randomly selected. The leaders' FGDs provided the criteria utilised to rank households according to wealth status and much of the qualitative information of this study. The wealth-ranking tool was used to identify and analyse the socio-economic factors that influenced smallholder market participation. A follow-up interview of managers of promoting institutions also provided greater insight on some aspects raised by smallholders. The study employed (i) descriptive statistics such as means and frequencies; (ii) correlation analysis and standard scores (iii) qualitative analysis was also used for some wealth-ranking, perceived labour demand and aspects of profitability influencing cash crop cultivation, preference and market participation based on information from FGD, farmers and staff; and (iv) simple mathematical expressions for analysis and interpretation of the research findings. This study relied on perceptions, knowledge and experience of smallholders, leaders and leaders of promoting institutions. Smallholder-suggested factors and strategies were in line with the limitations of socio-economic characteristics such as low effective household labour, particularly for females. These strategies included an improvement in outputs and inputs markets, agricultural services and credit at a subsidised prices or low interest rates. Other strategies for improving smallholders' participation in agricultural markets included promotion of profitable cash crops, household food security, provision of extension support services and information about cultivation and agricultural markets. However, smallholders did not identify some factors that have been acknowledged to influence agricultural market participation: ecological and natural resources, policies, institutional infrastructures and physical infrastructures. Smallholders also did not mention socio-economic factors (except household labour) as influencing their decisions to participate in agricultural markets in spite of the fact that researchers assume these factors in almost every study on smallholder market participation. The findings of this research confirmed that a wealth-ranking tool could be used to identify the socio-economic factors affecting smallholders' participation in agricultural markets. The identified wealth-ranking factors such as labour, livestock number, implements and bicycles significantly correlated with wealth status and subsequently to smallholder agricultural market participation. Conversely, household socio-economic characteristics not indicated as wealth-ranking factors such as age and gender related poorly to market participation. The wealth-ranking tool could also be used to identify strategies for improving smallholder participation in agricultural markets, and to evaluate an agricultural market development programme. The study found that, other factors being held constant, CLRs were negatively related to market participation. Weeding was the most labour intensive operation followed by harvesting, soil preparation, transportation, land clearing and seedling preparation. It also found that AHL and the ratio AHL/CLR were positive and significantly related to market participation. The ratio AHL/CLR together with household consumption requirements and yield were used to estimate the total area a household could cultivate, both for food crops for consumption and for cash crops; the proportion of farmers likely to participate in the market; and those unable to cultivate enough for consumption. The research also confirmed that profitability-related aspects correlated to cash crop preferences. Yield was the most important factor that influenced smallholders' preference for cash crops. It was also found that indicators incorporating more aspects of profitability correlated strongly with cash crop preferences. The correlation increased as more aspects were incorporated. A crop, such as tobacco, with a profit of more than twice the profit for food cash crops was preferred more than food cash crops. The indicators and underlying aspects of profitability were used to interpret the current and projected cash crop preference. / Thesis (Ph.D.)-University of KwaZulu-Natal, Pietermaritzburg, 2005.