Financial profitability and technical efficiency of horticultural crops in the Nakanbé River Watershed in Burkina Faso

In the watershed of the Nakanbé River in Burkina Faso, horticultural crops are financially profitable with high gross incomes, cost/revenue ratios and attractive hourly returns to labor. Based on a 6-h work day, profitability indicators of horticultural crops show that farmers are better-off compared to the ‘less than one dollar’ situation of people in Burkina Faso; horticultural crop production could thereby be an important part of a poverty reduction strategy in the studied sites and other rural areas. The proportion of efficient producers ranged from 35% for tomato in Ouonon to 100% for onion in Zekeze. Considering efficiency for individual crops, onion production in Zekeze was the most efficient, with a maximum score of (100%), followed by the production of cucumber in Ouonon and green beans in Kongoussi with respective scores of 98%. Inputs (seeds, fertilizers and pesticides) are used in excessive quantities which tend to reduce technical efficiency and could have negative environmental impacts. Improving horticultural crop producers' capacities to use appropriate rates of chemicals and organic inputs or biological treatments is recommended since the survival as well as the development of this kind of farming increases rural households’ incomes and supplies fresh food to an increasing urban population at a reasonable price.


INTRODUCTION
Burkina Faso is one of the poorest countries in the world; the incidence of poverty is 43.9% at the national level (INSD, 2010). Poverty is greater in rural than urban areas. In order to ensure their economic and nutritional survival, rural, urban as well as peri-urban populations produce horticultural crops under irrigation, particularly during the dry season. In Burkina Faso, only 12% of the total irrigable land is under irrigation (Government of Burkina Faso, 2005), and 30% of the irrigated area produces horticultural crops. Burkina Faso has been an exporter of fruit and vegetables such as fresh green beans since 1970, but in recent years the competitiveness of this sector has sharply declined. This decline in competitiveness of horticultural products and the sub-sequent reduction of exports has been 'the result of an intrinsic decline in the quality of the product" (Government of Burkina Faso, 2005). Any loss of competitiveness leads to a loss of market and therefore, to a reduction of exports earnings. Since the daily income per capita is less than one U.S. dollar in Burkina Faso, *E-mail: dpkabore@gmail.com. Tel: 00 226 70 26 88 73, 00 226 78 25 78 28. Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License any decline in export earnings reduces per capita income and then worsens poverty.
Reduced quality of horticultural products not only has a negative influence on the competitiveness of the sector, but also leads producers to use excessive amounts of inputs (fertilizers and pesticides) during production, which poses environmental problems.
According to the definition of Koopmans (1951), 'a producer is efficient if he maximizes output given the input he uses; that is, transforming inputs to outputs he needs a certain kind of technology. Since this type of efficiency deals solely with technology this type of efficiency is called technical efficiency', while allocative efficiency measures the ability of the farmer to use inputs in optimal proportions, given input prices.
Technical efficient is rather the ability of the farmer to produce maximum output from a given level of inputs, whatever these inputs are.
It is therefore important to identify bottlenecks and weaknesses in horticultural production to understand the production decline in recent years in order to improve production in the future. This objective of this study is to analyze the profitability and technical efficiency 1 of horticultural crop production in the Nakanbé River watershed in Burkina Faso.

Data collection
The first step in data collection was the choice of study sites in the watershed with farmers' cooperatives and associations. The selection criteria were (i) the volume of horticultural crop production during the dry season of candidate sites, (ii) ease of access to the production area, and (iii) the level of cooperation of the producers in the sites.
Using the above criteria, four (4) sites were chosen ( Figure 1): Kongoussi, in the Bam province; Titao, in the Loroum province; Ouonon, in the Passore province and Zekeze in the Boulgou province.
The survey was conducted during the 2004 dry season to a total of 131 randomly selected farmers: 31 in Kongoussi, 24 in Titao, 36 in Zekeze and 40 in Ouonon. The questionnaire used for data collection included the (i) farmers' socio-demographic characteristics (ii) land parcel identification, crops and cropped areas, and land tenure, (iii) cropping calendar describing the succession of farming operations (iv) quantity and cost of inputs used, (v) family labor indicated by the number of persons and amount time spent working by gender, age, crop produced, and cropping operation, (vi) amount of hired labor in hours and payments in cash and in-kind (vii) inventory of equipment used (purchased or borrowed), (viii) cost of equipment repair and maintenance, (ix) quantities harvested and sales by crop, and (x) marketing problems.

Method of analysis
The method of analysis included profitability, technical efficiency and measures of excess inputs use.

Indicators of financial profitability
Production budgets were used to assess profitability, especially cost/revenue ratio and hourly payment, during the dry season. For a given crop, the household average production was computed and valued at average sale prices to get the gross income (GI).
The corresponding cost/revenue ratio (C/RR) is calculated as: C/RR= GI/VC where VC is total variable costs to produce the crop. The related hourly payment (HP) is obtained, dividing GI by the number of family hours (FL), that is HP=GI/FL.
The higher the gross income and hourly payment, the more profitable is the crop 2 .

Technical efficiency
The Data Envelopment Analysis (DEA) method was used to analyze the technical efficiency of horticultural crops producers which is a non-parametric method, a deterministic measurement of technical efficiency. This method is based on models of optimization and linear programming techniques. The analysis was performed using the DEAP (Data Envelopment Analysis Program) (Coelli, 1996); in addition to the level of efficiency it provides, this software also gives information on the variables which are the sources of the inefficiency.
The method is based on the measurement of the ratio output/input in the form u'yi/v'xi where u is a Mx1 vector of relative weights of outputs and v a Kx1 vector of relative weights of inputs. Optimal weights are obtained by solving a problem of mathematical programming, specified as follows: The objective is to find values for u and v such that the measure of the efficiency of the unit of decision i (a farmer in our case) is maximized under the constraint that all measures of efficiency are less than or equal to 1. This formulation of ratio has the disadvantage of having an infinite number of solutions since if (u*, v *) is a solution of the program, then so is (u*, v*).
To overcome such a difficulty, an additional constraint is imposed and the problem is reformulated after transformation; the DEAP model developed by Coelli (1996) uses the dual program as reformulated (Coelli, 1996).
Some authors such as Boles (1966) and Afriat (1972) proposed methods of linear programming to estimate the isoquant frontier; such methods did not receive attention until Charnes et al. (1978) introduce the term "Data Envelopment Analysis". The model developed by Charnes et al. (1978) focused on the inputs 3 and based on the assumption of the constancy of returns to scale. Later on, alternative hypotheses were introduced; Banker et al. (1984) proposed a model rather based on the assumption of variable returns to scale.

Inputs excess
Referring to Figure 2 illustrating this idea of input excess, it can be seen that on the same linear segment of the isoquant convex frontier SS', you can have an optimal point with the possibility of reducing the quantity of inputs (x 1 and x 2 ) used without reducing the level of production (y). We may then wonder if such a point is optimal; it is an excess of inputs used, that could have been saved, since this does not increase the level of production. The technical efficiency for decision units A and B are respectively of OA'/OA and OB'/OB; the question that can be raised is whether point A' is efficient since the quantity of input x 2 can be reduced by CA' without reducing the level of production; CA' is therefore the input excess of x 2 . The Data Envelopment Analysis software provides options for the calculation of the excess inputs.

Main crops production
The main crops in Titao are onion and potato; green beans predominate in Kongoussi while at Zekeze, onion, tomato and banana are the most important vegetable crops. In Ouonon, the main crops are tomato, eggplant and cucumber (Table 1).

Profitability of horticultural crops 4
Potato and onion in Titao have the highest gross incomes per hectare, followed respectively by onion in Zekeze, green beans in Kongoussi, banana in Zekeze, onion and tomato in Ouonon, tomato in Zekeze, eggplant and cucumber in Ouonon (Table 2). However, the cost/revenue ratio of 8.77 reveals the greatest profitability for onion in Ouonon. Potato and green beans, despite good gross incomes, recorded relatively poor cost/revenue Measure of efficiency and excess inputs. x 1 /y : quantity of input x 1 to produce a unit of production y; x 2 /y: quantity of input x 2 to produce a unit of production y. Source: Coelli (1996). ratios due to high input costs, especially for seeds and water. Green beans producers had to pay a water fee of 400,000 FCFA/year and the seed cost was estimated at 225,650 FCFA per hectare, while potato required 2,126,600 FCFA of seed per hectare for its production. The cost/revenue ratio was significantly higher for tomato in Ouonon than in Zekeze. Eggplant and cucumber had the lowest gross revenues, but still financially rewarding with revenue/cost ratios of 1.72 and 1.77, respectively.
Potato had the highest gross margin per hour (1733 FCFA). The hourly remuneration of green beans was intermediate in Kongoussi (525 FCFA), and lowest for cucumber in Ouonon (206 FCFA) and banana in Zekeze (260 FCFA) which was the most labor-demanding.
Onion had a higher hourly payment in Ouonon than in Titao and Zekeze. Tomato recorded a higher hourly remuneration in Ouonon than in Zekeze.
Based on a 6-h work day, profitability indicators show that farmers are better-off compared to the 'less than one dollar' situation of people in Burkina Faso. Pro-poor rural economic policy could therefore be promoted through the production of horticultural crops.

Scores of efficiency
Scarce studies are available on horticulture efficiency in Africa and in Burkina Faso. However, Jema (2008) conducted an interesting one in Ethiopia and found that mean technical efficiency for vegetable production was 96% in two districts of eastern Ethiopia.
In our study, it is found that technical efficiency scores (Table 3) showed that the proportion of efficient producers ranged from 35% for tomato in Ouonon to 100% for onion in Zekeze.
Cucumber production in Ouonon, green beans in Kongoussi, tomato in Zekeze, potato in Titao and eggplant in Ouonon also recorded a technical efficiency score greater than 70%.
Considering efficiency for individual crops, onion production in Zekeze was the most efficient, with a maximum score of 100%, followed by the production of cucumber in Ouonon and green beans in Kongoussi with respective scores of 98%. Onion had a lower efficiency score in the northern site of Titao, and in Ouonon (88% and 84% respectively); the lowest technical efficiency score was recorded in Ouonon for tomato with a score of 68.4%. Green beans and potato producers were less efficient than onion producers in the southern site of Zekeze and cucumber producers in Ouonon. The reasons for this inefficiency should be investigated in order to improve the profitability and competitiveness of horticultural crops.

Excess inputs
The use of excessive levels of inputs is one source of technical inefficiency. In Kongoussi, the inputs that most farmers used in excessive quantities, included seeds (probably because of poor quality) and fertilizers (NPK and urea) ( Table 4).
The use of fertilizers in excess quantities reflects the decline in soil fertility that producers tried to compensate yield decrease by increasing application rates which might result in environmental pollution. Less important excess uses were observed on family labor, energy and  depreciation for 43% inefficient farmers. In Titao, excess use of inputs was observed in the production of potato and onion: excess use of seeds and NPK fertilizer for 513,059 FCFA/ha and 150,111 FCFA/ha respectively for potato. For onion production, excess levels were of 293,697 FCFA/ha for seed, 203,695 FCFA for NPK fertilizers, 61,919 FCFA for urea, 201,739 FCFA for the organic manure and 106,795FCFA for pesticides. On the production site of Ouonon, not only inputs were used in excess quantities; hired labor also recorded a peak of excess level of 39,778 FCFA/ha and 50,117 FCFA/ha for onion and tomato, respectively.
In Zekeze, banana inefficient producers recorded on average an excess inputs use of 112,076 FCFA/ha for the NPK fertilizer, 138,237 FCFA for urea and 125,064 FCFA for energy. Unlike banana, the recorded major excess input use was in tomato production: 169,889 FCFA/ha (100% of farmers) for energy and 116,780 FCFA