Determinants of adoption and spatial diversity of wheat varieties on household farms in Turkey

Abstract

The Heckman two-stage estimation procedure was used to investigate factors influencing the adoption of modern and/or landrace wheat varieties and spatial diversity of wheat varieties in Turkey. In the first stage, the multinomial logit choice model (MNLM) was used to determine factors influencing farmers’ adoption of modern varieties (MVs) and/or landrace varieties (LVs) of wheat. Conditional on the choice of a given wheat variety or combination of MVs and LVs, a Tobit regression model was used to assess the determinants of on-farm spatial diversity of wheat varieties in the second stage. Our empirical approach allows for the analysis of partial adoption decision of wheat varieties and controls for self-selection problem in analyzing the determinants of spatial diversity of wheat varieties. The empirical model was conceptualized based on random utility model (RUM). The analysis was based on cross-sectional survey data collected on 486 sample households in six provinces of Turkey. Results showed that household size, the number of owned cattle, the number of buildings on farm, farm size, farm land fragmentation, the percentage of irrigable farm plots and regional variations are the important factors in determining the farmers’ first-stage choice of wheat variety types. The self selection problem was significant only in one of the three cases for the landrace wheat varieties. In the second-stage, the farm size and land fragmentation were found to be the key variables influencing the level of on-farm spatial diversity of wheat varieties. The results showed that considerable spatial wheat genetic diversity was maintained on-farm at the household level, mainly through the simultaneous adoption of modern and traditional wheat varieties. Growing a combination of modern and landrace wheat varieties was observed to yield significantly higher level of spatial diversity of wheat genetic resources as compared to growing modern varieties alone or landrace varieties alone. This result suggests that the modern and landrace wheat varieties can coexist and could still support more on-farm spatial diversity of wheat genetic resources. This finding has significant implications for future extension, research and policy efforts for on-farm conservation and utilization of wheat genetic resources in Turkey. There is a need for the government and private sector research and extension efforts to support farmers’ use of both modern and landrace varieties, for example, in terms of seed supply, provision of extension and credit services and marketing support instead of just giving undue priority to popularization and adoption of modern varieties alone.