LAULYN CULLCULLUOTL The data was gathered by face-to-face interviews in April of 2019 from four provinces; Khan Hoa, Ninh Thuan, Soc Trang, and Bac Lieu, where most of the white-legged shrimp production take: place. Each interview lasted approximately 20 minutes. The first part of questionnaire elicited genera information on farming activities such as existing farming methods, size of the farm, production level water exchange. The second part of the questionnaire was dedicated to perceptions and attitude: toward shrimp farming. The choice sets were presented in the third part followed by a set of standar« socio-economic variables. A sample of 205 completed surveys was obtained. Results Figure 1. Perceived benefits and capacities for adoption of improved shrimp production practice (measured in %). Farmers’ preferences towards the adoption of improved shrimp aquaculture methods SAC MIYUIED J Ur erewu SE Peer ae Ms bral ella Uruw i vorer blll! al See eae ee ee eee x& To better understand what underlies farmers’ decision to invest and adopt improved aquaculture practices, information about their perceptions on the methods' advantages as well as their financial means reports farmers' and their knowledge and capacity to adopt the techniques were collected. Figure | perceptions of improved shrimp farming methods. A large share of farmers believed that improved production methods can help increase the price of shrimp (66.2%), control the spread of disease (81.5%), and allow them to use fewer antibiotics and pesticides (81%). Almost all of the farmers state tha they are concerned about the water quality in the shrimp ponds. Furthermore, many farmers believe that they have the necessary knowledge to invest in, and adopt, improved production methods (69.3% farming practice , and that they can successfully apply improved practices on their own farm (57.1%). However, farmers’ financial capacity to adopt improved aquaculture methods was remarkably low. Only 20.5% of the farmers indicated they have sufficient financial capacity to invest in improved Ss. Notes: ***, **, * ==> Significance at 1%, 5%, 10% level of confidence; S.E.=standard error; SD=standard deviation Table 6: Estimation results from the Latent Class Model Results of the LC model with two latent classes are reported in Table 6. The LC model fits the data better compared to the MIXL model. Choosing the correct number of classes to estimate implies trading off economic feasibility with statistical fit. In this case, moving beyond two classes led to numerical instability and implausible signs for the interest rate coefficient in one of the classes. Farmers in Class | are likely to be in different to alternative investments as evident by the very large, negative and insignificant ASC, but they are very sensitive to changes in the interest rate. Except for wastewater reduction, none of the other parameters in Class | are significant. Those in Class 2, on the other hand, are likely to invest in improved production methods. This is also the largest class comprising roughly 67% of respondents. The fact that most parameters are significant and of the expected signs, indicate that farmer s in this class have made trade-offs. While models cannot be compared directly, it can be noted that the parameter estimates are of the same sign, significance and rough magnitude as the MNL resu convergence of estimates between di Table 6: Estimation results from the ts and means in the MIXL model, which suggests a certain fferent model specifications. Latent Class Model Table 7. Marginal willingness-to-pay The mean MWTP from the LC model are overall lower compared to those from the MIXL model. The class specific MWTP is reported, however, the sample level (weighted sum) is straight forward to calculate. Farmers in Class 1 would request a compensation equal to 4.5% of the investment in improved aquaculture methods to reduce wastewater compared to insignificant MWTP in Class 2. Farmers were willing to pay for an additional crop an interest rate of 0.9% and for an increase in crop yield an interest rate of 1.9%. They may also adopt the improved methods, accepting the loss in production in exchange for a compensation of 0.6% of the investment.
