Operational Policy Making for the Development of Pistachio Drought Insurance in Sabzevar County

The dependence of agriculture on environmental conditions has caused the activity in this sector to face natural and unnatural risks. After several years of agricultural insurance activity in Razavi Khorasan province, most of the pistachio farmers are not insured. Drought insurance is one of the methods that has become important to cover the risks of drought and lack of water resources in order to compensate part of the gardeners' losses. The main issue from a managerial perspective is risk management. The use of agricultural insurance, which is one of the risk management tools, will ensure financial security and stability for farmers. Given that insurance is a tool for risk management, and given the uncertainty and risks of climate change in agriculture, insurance can be a very adaptable tool to water scarcity. Agricultural insurance is considered as a useful and appropriate solution to deal with natural hazards. . Drought insurance is an important factor in off-farm drought risk management that can mitigate the effects of this inevitable phenomenon. Insurance as one of the risk management tools can increase the risk-taking of farmers and, consequently, increase the sense of security in farmers, the necessary ground for proper and efficient use of factors of production and investment in the use of new technology and thus increase productivity in agriculture provide. The effects of water scarcity can be summarized as follows; Loss of production and income, abandonment of busy crops (with high water demand) and decline in agricultural employment, on the other hand, intensifies the over-exploitation of groundwater aquifers, which has tempted many farmers to do so meet your water needs. Access to water in the study area is one of the important variables affecting pistachio yield and quality as well as the survival of pistachio trees. This variable directly affects the profitability of producers and gardeners may suffer losses from this vital input. For this reason, gardeners' behavior in relation to regular pistachio insurance can affect access to water and make more farmers inclined to drought insurance.

This research seeks to answer the question that with 5% reduction in available water, pistachio growers in Sabzevar city, whether these people are willing to accept pistachio drought insurance or not, and if so, what is the extent of this desire. The Probit pattern is one of the most suitable econometric patterns for censored observations. This model was first proposed by Tobin (1958) to estimate the demand for durable goods. Subsequently, Arab Mazar and Schmidt (1979), Brown and Mufit (1982), Madela and Nelson (1982), and Hard (1975) worked on and developed the model, validating its high capability. This pattern was named by Goldberger (1964) as the Tobit or Probin Tobin model. Assume that y is the level of activity or action desired and xi are factors that generally affect the level of activity or action in question, namely:Also assume that one group of the observed observations performs the desired activity and the other group (the rest) does not perform the desired activity. As mentioned earlier, the values of xi and yi are visible for the first group. While for the second group only xi values are available and yi values are zero.In Hackmann's proposed two-step method for estimating the Tobit model, it is assumed that one set of variables may influence the decision to participate in the activity and another set of variables may affect the amount of activity performed after the decision is made. Therefore, two different sets of variables can be included in the Probit model, which are not necessarily barriers to aggregation. Therefore, two different sets of variables can be included in the Tobit model, which are not necessarily barriers to aggregation. Because it does not have a one-step model of this flexibility, it assumes that the variables influencing a person's decision to engage in an activity are the same as the variables that determine the amount of activity, if this is not necessarily the case. Hackman's two suggested steps are:Step 1: In the first step, the variables that affect the decision of gardeners in accepting pistachio drought insurance are identified and placed in a model with a binary dependent variable (zeros and ones); This means that the positive values of the dependent variable that indicate the tendency to accept pistachio drought insurance become the number one, and the dependent variable that does not tend to accept the drought insurance is set to zero. The number one means the decision to perform the activity and zero means the non-performance of the activity. At this stage, in order to identify the factors influencing the individual's decision, the Probit Model is used and estimated by the maximumlikelihood method. The first step is to create a new variable inverse of the Mills ratio to enter the second step. In other words, this variable is the first and second stage communication bridge.Step 2: In the second stage, the measures affecting the willingness to participate in drought insurance after the decision is made along with the inverse Mills ratio variable are placed in a classical regression model. The dependent variable in the second stage is the amount of garden area likely to be allocated to drought insurance.Reasons to use the Tobit model: Many econometric models face two types of errors, either due to the use of specific observation data or due to the structural features of the models: first, the error due to incorrect sample selection, which usually occurs in using classical regression models, and second, the same error Assuming effective variables in the decision stage and the amount of activity performed after the decision is made (decision and action or intention and action), which usually occurs in regression models with binary and multiple responses. The Tobit model has been developed to prevent the occurrence of these two types of errors in studies.The first error is the error of incorrect sample selection; in the sense that in many econometric models, information is obtained only from observations that have acted on the activity and omits observations that have refused to do that activity. Therefore, these models are not able to assess the reaction of observations that did not act on the independent variable changes. Tobit model (type one) solves this problem in terms of observations that have performed the desired activity as well as other observations. Under these conditions, the effect of changes in independent variables on both the total observations and on the observations of the activity can be calculated separately.The second error means that the factors that influence a person's decision to perform an activity are not necessarily the same as the factors that determine the amount and level of activity desired, and can be two different sets of variables. The Tobit model (type two, Hackett or Hackman two-stage) solves this problem by separating the factors influencing the decision and the amount of activity.

The data show that the response of pistachio growers to the reduction of available water in the next 2 and 5 years is that all gardeners will insure their pistachio orchards with a 5% reduction in available water, but in terms of area under cultivation, only 39% Gardeners will increase their arable land in the next 2 years and 33% of gardeners in the next 5 years. The reaction of gardeners who did not have a history of pistachio insurance to accept pistachio insurance and increase or decrease the area under pistachio orchard in exchange for a 5%  reduction in available water in the next 2 and 5 years shows that about 51% of gardeners face a 5%  reduction in water in 2 And in the next 5 years, they will insure their pistachio orchards, and about 60%  of gardeners will increase their cultivation in the next 2 or 5 years in the face of a 5%  reduction in available water. The results of the evaluation of gardeners' reaction to the continuation of the horticultural profession in the face of a 5%  reduction in available water in the next 2 years will cause 34%  of gardeners not to continue this profession and 51%  of gardeners will not continue this profession in the next 5 years. In the long run, water shortages can reduce the incentive for gardeners to grow pistachios. The reaction of gardeners to pistachio insurance against the reduction of available water quality shows that only 1.38 percent of the total population in the face of reduced quality of available water reduce the level of their insured garden and about 30% of them faced with declining available water quality, they will increase the level of their insured garden; And the rest of the gardeners (about 68.6%) do not change their insured level in the face of declining water quality.

According to the obtained information, the variables as gardener's age, ownership, relationship between gardener's field of study and agriculture, location, variety of cultivation, existence of insured pistachio garden in the neighborhood, frequency of risk, total water available to each gardener and garden life of each gardener in the first stage (Probit Model) have positive coefficients; which indicates the positive effect of these variables on the probability of willingness to accept pistachio drought insurance. In the second stage (linear regression), the variables of pistachio horticulture history, frequency of risk, garden life and total number of water hours available to gardeners have positive coefficients, which indicate the positive effect of these variables on the dependent variable of the second stage, is the tendency to accept pistachio drought insurance.