Agricultural Research Management Information System

Home
Research Summary
Search
About Us
- ARMIS
- Brochure
Contact Us
Report
User Request
Data Input
Help
- Operation Manual
  - PDF
  - Video
- Program Area & Commodity
We have reached 37600 number of research entries at this moment.

- Logout

Research Detail

Home
Research
Detail

Determinants of Crop Choices by Bangladeshi Farmers: a Bivariate Probit Analysis

Sanzidur Rahman
The University of Plymouth, United Kingdom

Abstract/ Executive Summary:

Using a bivariate probit model, the study jointly determines the factors underlying the probability of Bangladeshi farmers adopting a diversified cropping system and/or modern rice technology. Results reveal that the availability of irrigation is the single most important determinant of the decision to adopt modern rice technology, and adoption is higher among the tenant farmers. The exact opposite is true for the likelihood of adopting a diversified cropping system, which is significantly higher in areas with no irrigation as well as among the owner- operators. Furthermore, the diversified cropping system has a significantly higher rate of adoption in regions with developed infrastructure. Farmers’ education, farming experience, farm asset ownership, and non-agricultural income all positively influence crop diversification. Also, small farmers are more likely to adopt a diversified cropping system. Significant regional variation exists in the level of crop diversification as well. The decision to adopt a diversified cropping system and/or modern rice technology is significantly correlated, implying that a univariate analysis of such decision is biased. Crop diversification can be promoted by investing in farmers’ education as well as rural infrastructure development. Also, land reform policies focusing on delegating land ownership to landless and marginal farmers, and tenurial reforms are noteworthy.

Keywords: Crop choices, Bangladeshi farmers, Bivariate probit analysis

Location: All over Bangladesh

Start Date: 00-00-1996

End Date:

Program Area: Socio-economic and Policy

Non-commodity: Trend analysis

Objectives:

The present study aimed to determine the underlying socio-economic factors influencing the decision of farmers in Bangladesh to adopt a diversified cropping system and/or modern rice technology.

Methodology:

Data and the Study Area: The study is based on farm-level cross-section data for crop year 1996 collected from three agroecological regions of Bangladesh. The survey was conducted from February to April 1997. The selected regions are Jamalpur (representing wet agroecology), Jessore (representing dry agroecology), and Comilla (representing both wet agroecology and an agriculturally developed area). Multistage random sampling technique is employed to locate the districts, then the thana (subdistricts), the villages in each of the three subdistricts, and finally the sample households. A total of 406 households from 21 villages (broken down into 175 households from eight villages of Jamalpur Sadar thana, 105 households from six villages of Manirampur thana, and 126 households from seven villages of Matlab thana) form the sample for the study. Detailed crop input output data at the plot level for individual farm households are collected for ten crop groups 1. The data set also includes information on the level of infra structural development in the study villages. The Theoretical Framework: Bivariate Probit Model Several studies have analyzed the determinants of adopting modern/improved technologies (including HYVs of rice, wheat and/or maize) by farmers in Bangladesh and elsewhere. These are largely univariate probit or Tobit regressions of technology adoption on variables representing the socioeconomic circumstances of farmers (e.g., Hossain 1989; Ahmed and Hossain 1990; Shiyani et al. 2002; Floyd et al. 2003; and Ransom et al. 2003). The implicit theoretical underpinning of such modelling is the assumption of utility maximization by rational farmers. Empirical model A bivariate probit model is developed to empirically investigate the socioeconomic factors underlying the decision to adopt a diversified cropping system and/or modern rice technology. The dependent variable is whether the farmer adopts a diversified cropping system and/or modern rice technology. For a diversified cropping system, represented by dv, the variable takes the value 1 if the farmer has grown any of the range of non-rice crops in the three growing seasons covering one crop year, and 0 otherwise. Similarly for modern rice monoculture, represented by mv, the variable takes the value 1 if the farmer has grown modern rice in any or all of the three growing seasons, and 0 otherwise. In other words, in a bivariate probit model, four possibilities are incorporated. These are: (i) the non-adoption of both cropping systems (dv = 0, mv = 0); (ii) the adoption of modern rice technology only (dv = 0, mv = 1); (iii) the adoption of a diversified cropping system only (dv = 1, mv = 0); and (iv) the adoption of both cropping systems (dv = 1, mv = 1).

Source of Information: Asian Journal of Agriculture and Development, Vol. 5, No. 1

Article Link (Online Journal): http://ageconsearch.umn.edu/bitstream/198978/2/AJAD_2008_5_1_2Rahman.pdf

Funding Source:

Budget:

Total Budget (Tk.) :

Achievement/Findings:

The model diagnostic revealed that the choice of a bivariate approach was more appropriate than the univariate approach that was commonly used in the literature. Availability of irrigation was the single most important determinant of adopting modern rice technology, and adoption was higher among tenant farmers. On the other hand, the probability of adopting a diversified cropping system was significantly higher in developed regions as well as areas with no irrigation. Also, farmers’ education, farming experience, farm asset as well as the share of nonagricultural income were all significantly related to the adoption of a diversified cropping system. The small farmers and owner-operators were more likely to adopt a diversified cropping system. Significant regional variation existed in the level of crop diversification. Based on the analysis of joint marginal probabilities in a bivariate probit model, it saw that ‘farmers’ education’ exerted the strongest positive influence on raising the probability of adopting a diversified cropping system conditional on modern rice variety adoption. The actual influence of ‘irrigation’ in reducing the probability of adopting a diversified cropping system was small. The key policy implication that emerges from this study is that crop diversification can be promoted significantly by investing in education targeted for the farming population as well as in rural infrastructural development. However, the investment in irrigation should be targeted towards small farmers characterized by poor land and asset endowments, low level of education, and those located in underdeveloped regions. Promotion of crop diversification is likely to have a positive impact on agricultural sustainability as it is clear from the literature that the Green Revolution technology based on modern rice monoculture is unsustainable in the long run. The thrust at the planning level to promote crop diversification is a step in the right direction. Another significant factor influencing crop diversification decisions is the ‘share of non-agricultural income’ of the household, which in turn improves with the development of rural infrastructure. Ahmed and Hossain (1990) have concluded that infrastructure had profound impacts on the income of the poor in Bangladesh, raising their income by 33 percent (which included the doubling of wages and increase in income from business and industries by 17 percent), thereby reinforcing our argument to improve rural infrastructure. Furthermore, appropriate land reform policies that focus on delegating land ownership to landless and/or marginal farmers and improving the existing tenurial system, which is now biased towards favoring modern rice technology adoption, would boost the number of small farmers and owner-operators, who are the most likely adopters of a diversified cropping system.

Knowledge Management: Journal