Poverty and technical efficiency in presence of heterogeneity in household behaviours

Purpose - The purpose of this paper is to investigate the relationship between poverty and technical efficiency (TE) of paddy farmers in presence of their heterogeneous selling behaviours. This paper explains how such behavioural heterogeneity affects this relationship in south-western Bangladesh. Design/methodology/approach - Translog production frontier model was used to estimate TE since it fitted the data set better. On the other hand, poverty indices were constructed by using P-αmethod. Then, multinomial logit models examined the existence of heterogeneous selling behaviours. It revealed adequate evidences in favour of behavioural heterogeneity. Finally, the authors employed a series of two stage instrumental variable regression models to relate poverty and TE with and without considering the behavioural heterogeneity. Findings - The study finds that around 18, 39 and 44 per cent of households exhibit autarkic, non-wholesaling and wholesaling behaviour, respectively. Market failure due to transaction cost and credit constraints leads to emergence of such heterogeneity. Across these heterogeneous behaviours, impact of improving TE on poverty status significantly differs. Without controlling behavioural heterogeneity, TE significantly improves the poverty status of the rural farm households. However, scenario is changed after controlling this heterogeneity. After behavioural segregations, TE improves poverty status only for wholesalers. In contrary to prior expectation, it worsens the poverty situation for both autarkic and non-wholesaling households. Simultaneous failure in both credit and product market for these households might be the plausible reason behind this heterodox finding. Credit market failure compels these households to borrow from local money lenders with costlier terms. This effort might improve their TE. But, product market failure makes their additional production due to improved TE unsold. Thus, repayment of credit directly reduces their consumption expenditure. Therefore, an effort to improve TE might increase prevalence and depth of poverty when market failure exists. Henceforth, the improvement of TE appears as an effective policy instrument only when households exhibit wholesaling behaviour. Originality/value - The earlier studies show the relationship between TE and poverty status but did not account behavioural heterogeneity. The authors attempt to overcome this shortcoming and show how market failure induced behavioural heterogeneity affects the effectiveness of TE on improving poverty status of farm households.