Studying the Dynamics of Market Integration and Price Transmission of Agricultural Commodities

Abstract

Price volatility as well as understanding the spillover effect of one market on the others has been the main center of attention for the researchers. It is therefore important to extend the consideration univariate Generalized autoregressive conditional heteroscedastic (GARCH) model to Multivariate GARCH (MGARCH) model. Various aspects of cointegration and vector error correction model have been discussed. In the MGARCH model, Baba-Engle-Kraft Kroner (BEKK) and Constant Conditional Correlation (CCC) models are considered for modeling volatility of onion prices in two major markets of onion in Karnataka, India. It is concluded that that the two markets are cointegrated and there exists spillover effect among them. ARIMA models are fitted using monthly Onion price data of two different markets, Bangalore and Hubli of Karnataka. The residuals were investigated for possible presence of ARCH effect followed by fitting of univariate GARCH models. It is seen that the magnitude of ARCH effects are more than the GARCH effects for both the series. The cointegration among the two series were tested by using both Trace statistic and Eigen value statistic and it is found that there was one cointegrated vector among the two series. Accordingly, VEC model was fitted and possible presence of MARCH effect was investigated on the residuals of VEC model. To this end MGARCH model was applied for modeling the conditional variance of the bivariate series. The performances of MGARCH models namely BEKK and CCC have been studied. High persistence of volatility has been observed in each market price. The interdependence and volatility spillover of onion price between Bangalore and Hubli markets has been established. The linkages among the markets, amount and direction of spill over will help the policy makers to take proper policy decision in order to stabilize the price of the commodity. In the present study, presence of cointegration was tested by using Johansen’s approach. It was revealed that wholesale and retail price of wheat in all the market are cointegrated both horizontally as well as vertically. Asymmetricity in price transmission is investigated by means of Threshold Autoregressive (TAR) and Momentum Threshold Autoregressive (M-TAR) models of Enders and Granger (1998). The application of MTAR model reveals that most of the markets under consideration are asymmetric in terms of price transmission from wholesale to retail markets. The acceptance of cointegration between two series implies that there exists a long run relationship between them and this means that an error-correction model (ECM) exists which combines the long-run relationship with the short-run dynamics of the model. The results indicate that most of the error correction term (ECT) are statistically significant implying that the system once in disequilibrium tries to come back to the equilibrium state. Moreover, findings pointed out that there are nonlinearities in the studied price adjustment process. The significance presence of threshold cointegration was ensured by application of test by Hansen and Seo (2011). To take care of asymmetricity as well as nonlinearity in cointegration and price transmission between wholesale and retail price of wheat, TVECM model was applied. Application of the Two- Regime Threshold Vector Error Correction Model (TVECM) demonstrated that the coefficient of Error Correction Term (ECT) is significant in retail for both the regimes in Delhi; wholesale for both the regime and retail in typical regime for Jammu, retail and wholesale in extreme regime for Amritsar; retail in both the regime but wholesale in extreme regime for Ludhiana; retail in extreme regime in Lucknow; retail in both the regime for Dehradun; wholesale in typical regime and retail in extreme regime for Ahmedabad; retail in both the regime for Bhopal; retail in extreme regime and wholesale in typical regime for Mumbai; wholesale in typical regime and retail in extreme regime for Jaipur; retail and wholesale both in extreme regime for Patna; retail and wholesale in both the regimes; retail in extreme regime for Bengaluru; retail in typical regime for Thiruvananthapuram; wholesale in second regime for Chennai; retail in both the regime and wholesale in typical regime for Hyderabad. This implies that retailers respond significantly to the deviations from the long-run equilibrium. Impulse response analysis has shown that changes in wholesale prices in a market will cause change in retail prices in that market with varying rate and time lags to price stabilization. The present study investigated the effect of volatility spillovers in monthly potato price of five different markets Agra, Ahmedabad, Bangalore, Delhi, and Mumbai from January, 2005 to April, 2021. The empirical results support the presence of ARCH and GARCH effects. Finally, the VIRF demonstrated that the impacts of impulse responses on expected conditional variances and expected conditional covariances took almost same time of ten months to recover. To this end one can conclude that changes in the volatility of one market will often trigger reactions in other markets