Abstract/Summary

The existence of convenience yield has become an axiom in the analysis of intertemporal price relations for commodities. However, the link between the macro level convenience yield concept and the micro-level sell/store decisions of stock holders lacks empirical evidence. The main objective of the study is to establish the drivers behind timing and volume of stock holders' actual commodity sales and to establish whether the decision-making process can be linked to convenience yield. Regression analysis of intertemporal price spreads, monthly dummies and stocks for Hungarian com are performed both to establish the relevance of the convenience yield concept, and to confirm its relationship to stocks. The GSADF (Generalized Sup-ADF) method is applied on consecutive expiries of corn futures, in order to validate convenience yield related assumptions of explosiveness. An alternative explanation of the convenience yield concept formulated in the spatial-temporal interpretation of storage is also tested on a large volume of actual Hungarian com logistics data. Finally, a choice model with hypothetical sales scenarios, filled by a large sample of Hungarian farmers is performed in order to validate whether a two-step Heckman model offers a good approximation of the actual decision-making process behind stock sales. Both convenience yield's existence and its inverse relation to stocks are demonstrated for Hungarian com. However, evidence does not support the hypothesis that nearby prices show greater tendencies for bubble formation than later expiries. Contrary to expectations, spatial¬temporal storage conclusions are validated on Hungarian corn, as delivery distance to central markets does increase through the season. Some variables proposed are found to be individually significant in driving stock sales, and the expectation that selling is best explained by a two-step Heckman model is also partly confirmed. Besides testing existing models on the data of a new geography, it is the additional step taken towards real decision-makers, where this study of commodity sales is expected to contribute both to convenience yield theory and to our current knowledge of how the physical grain markets work.