Policy reforms and efficiency analysis in domestic agricultural markets. Evidence from an econometric analysis in Rwanda

Recent changes and reforms in Rwandan agricultural policies, formulated since 2000, have emphasized on transforming agriculture sector from subsistence level into a modernized and more industrial and market oriented agricultural sector. The Vision 2020 launched in 2000, considers the agricultural sector as the principal source of economic growth of the country with an anticipated growth of between 5 - 6 percent each year to reach an overall economic growth of 7 – 8 percent projected in 2020, it has therefore placed greater emphasis on improving agriculture productivity (MINAGRI, 2009). As widely documented in the economic literature, policy changes and reforms adoption should improve the functioning of agricultural markets and enhance commodity market performance. Therefore, agricultural policy changes and reforms in the country have contributed to the decline of yield and crop production instability, improvement in food distribution system, investing in new products that can generate more revenues and has facilitated households to access foodstuff at fair prices. Nonetheless, food prices from 1998 to 2014 show sudden and sometimes tremendous food price variations. Rwanda agricultural markets represent a vital opportunity for commodity modelers to analyze and provide information on food price volatility and transmission. The findings from this paper will help to inform policymakers and decision takers on food market dynamics and their causes. In order to handle volatility in selected 5-6 crop commodities prices, the generalized autoregressive conditional heteroskedasticity (GARCH) developed by Bollerslev's (1986) as extension from Engle (1982) is used. Price transmission will be measured by vector error correction model (VEC) developed by Engle and Granger (1987). This econometric model allows to establish relationship (long-run equilibrium, short-run dynamics) between prices from different crop markets in different provinces of Rwanda. The time series properties of each of the price variables will be examined by using the Augmented Dickey-Fuller (ADF) test (Fuller, 1976). Taking into account the order of integration between markets, VECMs or vector auto-regressions (VARs) are specified and estimated. Several criteria like Akaike Information Criterion (AIC) for lag lengths are verified before the VEC and VAR models. At this stage, we do not have yet results but we have some assumptions/hypothesis. The first part on the literature review is finished. As now, we have the time series data on different 5-6 crop commodities in different provinces of Rwanda, we will quickly try to have some results and submit as soon as possible our full paper.