Dynamic Relationships among U.S. Wheat-Related Markets: Applying Directed Acyclic Graphs to a Time Series Model

Article excerpt

Using advanced methods of directed acyclic graphs with Bernanke structural vector autoregression models, this article extends recent econometric research on quarterly U.S. markets for wheat and wheat-based value-added products downstream. Analyses of impulse response simulations and forecast error variance decompositions provide updated estimates of market elasticity parameters that drive these markets, and updated policy-relevant information on how these quarterly markets run and dynamically interact. Results suggest that movements in wheat and downstream wheat-based markets strongly influence each other, although most of these effects occur at the longer-run horizons beyond a single crop cycle.

Key Words: Bernanke structural VARs, directed acyclic graphs, quarterly wheat-related markets

JEL Classifications: C22, QIl

The value-added side of the food industry, unlike farm commodity markets, has been neglected as an empirically researchable area, in part because of a lack of published data on these industries. Unlike commodities such as corn or soybeans, the United States Department of Agriculture (USDA) and other Federal agencies often do not publish highly periodic (monthly or quarterly) data on quantities (demanded or supplied) or stocks of value-added products (Babula and Rich, p. 1). Moreover, food industries typically classify information on own prices, production, and distribution as confidential and thereby preclude it from the public purview (Babula and Rich, p. 1). Consequently, there are few studies that estimate the market-driving elasticity parameters for wheat-based value-added products, and that illuminate the dynamic nature of quarterly interactions among U.S. wheat and wheat-based value-added markets. Two recent empirical contributions to this generally lacking area of value-added, commodity-based market research are Rich, Babula, and Romain's (hereafter, RBR's) time series econometric study on the U.S. markets for wheat and wheat-using value-added products downstream, and Babula and Rich's analysis of the U.S. markets for durum wheat, semolina, and pasta. However, RBR is a conference proceedings paper published in a largely unavailable volume, and without any benefit of peer review.

This paper contributes to the literature in two ways. First, a time series analysis is performed on the U.S. wheat market and for wheat-using markets downstream (downstream markets). Given the lack of empirical econometric research just cited, this article helps to fill this gap of research by providing an analysis on these important U.S. markets in the peer-reviewed literature. We ultimately provide updated estimates of important wheatrelated market parameters and updated results that illuminate the dynamic nature with which upstream and downstream wheat-related markets interact, particularly for the RBR analysis.1 second, this paper extends the studies of RBR and Babula and Rich with the application of new methods of directed acyclic graphs (hereafter DAGs). These new graphical procedures generate evidentially based lines of causality among variables in contemporaneous time for the U.S. markets of wheat and wheatusing value-added products. These downstream markets include wheat flour, mixes and doughs (hereafter, mixes/doughs), bread, breakfast cereals, and cookies and crackers (hereafter, cookies/crackers). More specifically, our analysis has the advantage of having adapted the recently developed econometric procedures, as applied to U.S. beef and pork prices by Bessler and Akleman, to a quarterly system of U.S. wheat-related markets. This set of procedures combines DAGs with Bernanke's methods of structural vector autoregression (VAR) modeling and is hereafter denoted as the "DAG/Bernanke VAR" procedures.

The remainder of this paper is comprised of several sections. First is a summary of why updating and extending the research of RBR and Babula and Rich on U.S. wheat-related markets is important and newsworthy, and hence of interest to this journal's readership. …