Recently, the effect of immigrants on international trade has been one of the important research topics in international economics. To examine this issue in connection with the agricultural export of Korea, relevant studies has used the panel tobit model to control the observations having zero value. However, theoretically, the tobit model is highly restrictive because it assumes the same consumption and participation effect of immigrant on trade volume. In this paper, with panel data sets covering three important fruit exports of Korea, we estimate these effects separately using the Cragg’s double hurdle model. The important empirical findings are: First, the Cragg’s model is statistically more proper than the tobit model to handle agricultural export data when there are significant zero observations in the sample. Second, the participation effect of immigrants on trade volume is significant while the consumption effect is not so important. Third, the effect of immigrants on Korea’s fruit exports are much bigger than the general goods suggested in the relevant literature.