Measuring Farm Technical Efficiency using Stochastic 1 Frontier Production Function Model Approach

10 Agricultural productivity in Africa is the lowest in the world with many households not able to feed themselves. 11 In Africa women make up 70-80% of the labour forces in the agricultural sector and play a core role in 12 agriculture but underperform in terms of productivity largely because they lack access to physical and human 13 resources. Well-being, a health resource is an important asset in production because people can work when 14 they are healthy. The study is aimed to analyze farm technical efficiency of women farmers in Niger Delta, 15 Nigeria. 216 female farmers were randomly selected from 18 communities of the three states in Niger Delta 16 Nigeria. Stochastic production frontier function model was the analytical tools used. The result showed that 17 farm size and labour positively influenced technical efficiency and was significant at 1% with a mean value of 18 68.8%. Farm efficiency level in Delta and Akwa Ibom States are not significantly different. However, technical 19 efficiency level in both Delta and Akwa Ibom States are significantly different from Rivers State. Inefficiency 20 variables of age and number of years spent schooling were significant at 5% and 10 % level respectively. The 21 study recommends that women should increase the use of farm plots and labour resource for higher 22 productivity.


25
Agriculture is a major source of livelihood for several people in the developing world providing employment for 26 5.5 billion people ( Mehta et al. 2010). Smallholder farmers provide about 80% of the food supply in Asian and 27 sub-Saharan Africa. In Africa, women make up 70-80% of the labour force in the agricultural sector and 28 produce about 80% of the staple crops mostly used for household consumption (Gordon and Gordon 2007).

29
Women are key players in food production in most developing countries, accounting for 43% of the agricultural 30 labour force with 50% in eastern and southeastern Asia and sub-Saharan Africa (FAO 2012).

31
It has been observed that despite contributions made by women in food production, their performance is still 32 low in terms of productivity. This is largely because of poor access to resources such as finance, skills training, 33 and information services (FAO 2012). Women tend to spend longer hours than men working in farms in 34 developing countries but lack access to production resources which are responsible for their low farm 35 productivity. Farm operations such as sowing, weeding, fertilizing and harvesting staple crops such as rice, wheat and maize are mostly carried out by women. Women's contribution to secondary crops, such as legumes 37 and vegetables, is even greater. FAO further reports showed that women produce most of the food consumed 38 locally in the rural area where majority of the world's hungry people live. Contribution made by women in 39 terms of production could be much greater if their access to essential resources and services, such as land, credit 40 and training is improved. Eliminating obstacles that hamper women in food production is the key to achieving 41 better farm yields. Improving women farmers' access to productive resources could also help increase their 42 farms yields by 20-30% (FAO 2011). This is because women are seen as the quiet drivers of change towards 43 more sustainable production systems and a more varied and healthier diet (FAO 2011).

44
The underperforming agriculture sector in many developing countries is so for a number of reasons. One of 45 which is due to lack of resources and opportunities needed by women to make the most productive use of their 46 time (FAO 2011). It has been reported severally that Africa's agricultural productivity is the lowest in the 47 world (Nilsson 2013). Many are not able to feed themselves, leaving the people vulnerable to shocks. In 48 addition, domestic food production growth in Africa has remained low, about 2.7%, which is barely above the 49 population growth rate (Nilsson 2013). Several studies on farm productivity in developing countries were 50 centered on estimating technical efficiencies using direct farm resources like farmland, labour, seeds and other 51 materials used in production. Worthy to note is, despite intervention programmes and actions by the 52 governments aimed at improving farm-level production in developing countries, it is sad to note that overall 53 technical efficiencies of farms are still far away from the best frontier level ie 100%.

54
The report of Ben-Belhassen and Womack (2002) study on farm technical efficiency in Missouri hog 55 production adopted stochastic production frontier function in determining technical efficiency. The study 56 showed a mean farm technical efficiency of 82%, implying that a large (18%) proportion of production was lost 57 due to farm-specific inefficiencies. Furthermore, technology and managerial skills were the major determinants 58 of technical efficiency (Aminu et al. 2013). It was also reported that a mean technical efficiency (TE) of dry 59 season vegetable farmers in Ojo Local Government Area in Lagos State, Nigeria was 71.1% which showed a 60 huge possibility for improvement for some farmers. This suggests that an average farmer would need to a 28.9% 61 cost saving in the study area to achieve the best technical efficiency level. It was also found that coefficients of 62 both illness episode and number of days absent from farm work showed positive and significant influence on 63 technical efficiency of farms. The finding implied that farmers, who suffered prolonged numbers of illness 64 episodes, had long days of absent from farm work, which increased farmers' inefficiency levels.
In same vein, Kussa (2012) study on health and farm productivity found that parametric estimation of inputs 66 such as land, labour, soil fertility and fertilizer significantly increased crop production but illness showed 67 negative correlation and elasticity of 0.53. This is an indication that illness is an important factor that affects 68 farm level production. Still more, households exposed to illness achieved an average technical efficiency of 69 33.5% while households not exposed to illness achieved 48.9% suggesting that health affects agricultural 70 production.

71
According to Simon

78
Much of the farm technical efficiency studies were focused on farm resource and ill health variables. It is also 79 surprising noting that not many studies have been carried out on farm technical efficiency of farms owned by 80 women in the study area. It is on this background that this study adopted stochastic production frontier function

157
Stochastic production frontier function used in measuring the technical efficiency of farm productivity.

159
Yi is the production of crop yield of i th producer, Xi a vector of inputs used by the i th producer, a vector of

269
Availability of data and materials: The author hereby declare that they can submit the data at any time based 270 on publisher's request. The datasets used and/or analyzed during the current study will be available from the 271 author on reasonable request.

272
Competing interests: There is no conflict of interest in the conduct and use of data generated from the project.

273
The supervisor, Cranfield University UK and University of Port Harcourt, Nigeria is void of any such conflict.

274
No previous submission of this article had been made to any scientific journal publisher and no "Conflict of

275
Interest" among the authors and sponsors of the research.