Essays on efficiency and risk in rice production and food consumption patterns in Sri Lanka

Manoj Indika Thibbotuwawa Rajakaruna Mudalige

    Research output: ThesisDoctoral Thesis

    757 Downloads (Pure)

    Abstract

    [Truncated] The aim of this study is to investigate the determinants of rice yields and yield variability among smallholder farmers and estimate the demand for food for a representative sample of households in Sri Lanka. These aims are addressed through four objectives: (1) to measure the efficiency and technology gaps between irrigated and rainfed rice farming systems; (2) to investigate fertilizer use efficiency in rice cultivation relative to other input-use efficiencies and the determinants of efficiency; (3) to investigate the risk characteristics of inputs and risk preferences of rice farmers under rain-fed and irrigated farming conditions; and (4) to estimate the price and expenditure elasticities of demand for major food categories in Sri Lanka and investigate whether being a recipient of the Samurdhi food subsidy programme affects food demand. These objectives are addressed in four papers.

    In the first paper, a non-parametric bootstrapped metafrontier method is applied to estimate the technology gap between the irrigated and rain-fed rice farming systems of 588 farms surveyed by the Department of Agriculture of Sri Lanka from 2007 to 2010. The results show that irrigated and rain-fed farming systems use different technologies which explain the yield gap. This suggests that policies targeted at marginally improving farm management alone may not reduce the yield gap. Instead an investment is required in the technology available to farms, including the development of irrigation infrastructure. The results also show that not accounting for technological differences in benchmarking farm efficiency may lead to biased estimates of production efficiency and therefore erroneous policy recommendations.

    The second paper, using the same data set as paper one, applied the slack-based DEA model to measure fertilizer use efficiency in rice cultivation relative to the efficiency of other inputs. Although fertilizer is heavily subsidized, the measured overuse of fertilizer is less than the overuse of other inputs - seed, land, chemicals, labour and capital. However, on average, there is scope to increase efficiency by reducing fertilizer use by 13 percent. We observed that the expected efficiency improvement that can be achieved by a reduction in subsidy would result in substantial financial savings to the government. Considering the social and political consequences of subsidy removal, efficient targeting of the farmers who are in need of the subsidy is recommended as a possible short run option. Moreover, enhanced fertilizer use efficiency with better irrigation management, quality seeds and better fertilizer management could allow the government more leeway to phase out the fertilizer subsidy. The distribution of fertilizer use efficiency does not match that of technical efficiency suggesting a role for specific input-use efficiencies for appropriate policy designs to improve agricultural productivity.

    Original languageEnglish
    QualificationDoctor of Philosophy
    Publication statusUnpublished - 2016

      Fingerprint

    Cite this