**3. General assumptions**

**Regional effects:** The survey was carried out in three different agro-ecological zones: Kandi, the northern region has one growing season starting at the end of May to September, with a temperature ranging from 28 to 45°C and a low rainfall averaging 800 to 900 mm. However, Savalou and Abomey-Bohicon both have two growing seasons (April to July and September to November) with higher rainfall between 1,300 and 1,500 mm, and temperature from 25 to 35°C (Setamou et al., 1997). Because of its dry climate, Kandi is the most productive region and is also the least prone to AF production.

**Risk analysis:** Parameters such as price of output, inputs and quantity are manipulated to examine how changes in parameters affect peanut production and revenues. A total of 5,000 iterations of the model are executed to generate all probability distributions that are used to establish stochastic dominance. All parameters used to develop the risks models are present‐

Aflatoxin and Peanut Production Risk and Net Incomes

http://dx.doi.org/10.5772/51913

383

Here we assume that net returns from peanut sales are affected by the costs of production

Cost is the cost of production; cost includes seed quantity and price, equipment, cost of pre-

Stepwise least squares regression is conducted between the collected input distribution val‐ ues and the selected output values. The assumption is that there is a relationship between each input and output. The output of the stepwise regression is expressed in the form of a

Tornado chart is used to show the influence an input distribution has on the change in value of the output. Its main use is to enable the researcher to determine which variable contrib‐

Therefore, the coefficient for any of the variables is standardized and will vary from -1.0 to +1.0. Variables contributing zero to the cost will be eliminated. Variation in cost, each year

Socio-demographic information, knowledge of AF on peanut and farming practices were collected during the survey. Age of the respondents ranged from 35 to 55 years old, and over 55 years old (Table 1). Peanut production is done mostly by men in Kandi (63.3%), Sa‐ valou (100%), and in Abomey-Bohicon (54.4%). A large number of peanut producers in Be‐ nin have not received any formal education, and have never heard of AF contamination of

Cost = p1\*β1 drying cost + p2\*β2 storing cost + p3\*β3 sorting cost +…+ pn\*βn costs of n

and post-harvest handling. Hence, we use the formula:

\*Cost)]

harvest, harvest, sorting, storage, bagging, winnowing…

utes more to the output. It is also used for model diagnostic.

will be kept and their importance to cost will be explained.

**4.1. Demographics and socio-economic results**

\*(Pp\*Qp) – (pi

is the probability of the occurrence,

ed in Appendix 1.

\*N.R = [pi

N.R is net return,

tornado chart.

**4. Results**

peanut.

Pp is the price for peanut,

Qp is the quantity for peanut, and

pi

pi

where,

In the southern regions, AF production is due to the high rainfall and high temperature. Un‐ der these tropical conditions, development of fungi and AF proliferation are facilitated. A higher concentration of AF in Abomey-Bohicon than in Savalou and Kandi.

**Decision on drying, sorting and storing peanut:** Based on previous studies, drying, sorting and storing methods are reported as the most important factors that encourage AF produc‐ tion. Farmers are recommended to dry peanut immediately after harvest, importantly to bring the moisture level of less than 8% (ICRISAT, 2008).

Sorting is considered as one of the ultimate solutions for the AF problem. This method has been reported as a post-harvest intervention strategy successful in reducing AF levels in peanut. An essential question was how much farmers or market participants will lose if they decide to sort peanut. In case the decision was "no sorting", not only quantity is affected but also labor cost. Hence, based on the answers obtained during the survey, the probability to throw away some nuts was estimated at one to five percent of quantity produced if farmers decided to sort them. However, if not, the risk of fungal growth and from nuts (molded or contaminated with AF) will increase.

Long-term storage in warm environment results in *Aspergillus* growth and increased in AF contamination. Previous research has yet to suggest a safe period in which peanut can be stored. We assumed, therefore, that after two months, with a risk of having bad nuts (mold, insect damage, and AF contamination), the percentage of rejection will be one percent and will increase by one percent more after each of two months. This percentage is applied on the quantity harvested as the percentage representing the loss in quantity if the storage length exceeds two months. This period (two months) was chosen based on survey reports.

**Enterprise budgets:** Budget analyses are used to evaluate the profitability of peanut enterpris‐ es in the short run. Costs and returns were estimated for each region. Most information used to develop each enterprise budget was obtained during the survey. Data such as seed quanti‐ ty, seed price, quantity of peanut harvested, peanut selling price, material and equipment, la‐ bor hours and costs were obtained from the survey. They are the averages for the various size farms. All lands included in the budgets are treated as owned by farmers. Material and equip‐ ment are the same in each region and are depreciated according to the useful life, using the straight-line method. Costs for repairs and maintenance are assumed to be \$1.00 for a one-hec‐ tare farm. Labor costs include land preparation, planting, harvest, drying, sorting, bagging, and transport costs. Labor costs and hour of use vary depending on the farm size.

**Risk analysis:** Parameters such as price of output, inputs and quantity are manipulated to examine how changes in parameters affect peanut production and revenues. A total of 5,000 iterations of the model are executed to generate all probability distributions that are used to establish stochastic dominance. All parameters used to develop the risks models are present‐ ed in Appendix 1.

Here we assume that net returns from peanut sales are affected by the costs of production and post-harvest handling. Hence, we use the formula:

$$\mathbf{p\_i^\*}\mathbf{N}.\mathbf{R} = \left[\mathbf{p\_i^\*} (\mathbf{P\_p}^\* \mathbf{Q\_p}) - (\mathbf{p\_i^\*} \mathbf{Cost})\right].$$

where,

**3. General assumptions**

382 Aflatoxins - Recent Advances and Future Prospects

and is also the least prone to AF production.

contaminated with AF) will increase.

bring the moisture level of less than 8% (ICRISAT, 2008).

**Regional effects:** The survey was carried out in three different agro-ecological zones: Kandi, the northern region has one growing season starting at the end of May to September, with a temperature ranging from 28 to 45°C and a low rainfall averaging 800 to 900 mm. However, Savalou and Abomey-Bohicon both have two growing seasons (April to July and September to November) with higher rainfall between 1,300 and 1,500 mm, and temperature from 25 to 35°C (Setamou et al., 1997). Because of its dry climate, Kandi is the most productive region

In the southern regions, AF production is due to the high rainfall and high temperature. Un‐ der these tropical conditions, development of fungi and AF proliferation are facilitated. A

**Decision on drying, sorting and storing peanut:** Based on previous studies, drying, sorting and storing methods are reported as the most important factors that encourage AF produc‐ tion. Farmers are recommended to dry peanut immediately after harvest, importantly to

Sorting is considered as one of the ultimate solutions for the AF problem. This method has been reported as a post-harvest intervention strategy successful in reducing AF levels in peanut. An essential question was how much farmers or market participants will lose if they decide to sort peanut. In case the decision was "no sorting", not only quantity is affected but also labor cost. Hence, based on the answers obtained during the survey, the probability to throw away some nuts was estimated at one to five percent of quantity produced if farmers decided to sort them. However, if not, the risk of fungal growth and from nuts (molded or

Long-term storage in warm environment results in *Aspergillus* growth and increased in AF contamination. Previous research has yet to suggest a safe period in which peanut can be stored. We assumed, therefore, that after two months, with a risk of having bad nuts (mold, insect damage, and AF contamination), the percentage of rejection will be one percent and will increase by one percent more after each of two months. This percentage is applied on the quantity harvested as the percentage representing the loss in quantity if the storage length exceeds two months. This period (two months) was chosen based on survey reports.

**Enterprise budgets:** Budget analyses are used to evaluate the profitability of peanut enterpris‐ es in the short run. Costs and returns were estimated for each region. Most information used to develop each enterprise budget was obtained during the survey. Data such as seed quanti‐ ty, seed price, quantity of peanut harvested, peanut selling price, material and equipment, la‐ bor hours and costs were obtained from the survey. They are the averages for the various size farms. All lands included in the budgets are treated as owned by farmers. Material and equip‐ ment are the same in each region and are depreciated according to the useful life, using the straight-line method. Costs for repairs and maintenance are assumed to be \$1.00 for a one-hec‐ tare farm. Labor costs include land preparation, planting, harvest, drying, sorting, bagging,

and transport costs. Labor costs and hour of use vary depending on the farm size.

higher concentration of AF in Abomey-Bohicon than in Savalou and Kandi.

pi is the probability of the occurrence,

N.R is net return,

Pp is the price for peanut,

Qp is the quantity for peanut, and

Cost is the cost of production; cost includes seed quantity and price, equipment, cost of preharvest, harvest, sorting, storage, bagging, winnowing…

Cost = p1\*β1 drying cost + p2\*β2 storing cost + p3\*β3 sorting cost +…+ pn\*βn costs of n

Stepwise least squares regression is conducted between the collected input distribution val‐ ues and the selected output values. The assumption is that there is a relationship between each input and output. The output of the stepwise regression is expressed in the form of a tornado chart.

Tornado chart is used to show the influence an input distribution has on the change in value of the output. Its main use is to enable the researcher to determine which variable contrib‐ utes more to the output. It is also used for model diagnostic.

Therefore, the coefficient for any of the variables is standardized and will vary from -1.0 to +1.0. Variables contributing zero to the cost will be eliminated. Variation in cost, each year will be kept and their importance to cost will be explained.
