The effects of USDA crop announcements on commodity prices

Abstract

ecent studies on spot (stock, bonds, commodities) as well as on futures R markets have stirred academic interest on the effect of the economic impact of public announcements on prices. Scheduled releases of economic information (earnings, dividends, inflation, money supply, CPI, etc.) as well as non-scheduled announcements (tender offers, investment proposals, new patents and discoveries, etc.), were analyzed to determine whether prices reflected this information before it was released. These semi-strong market efficiency tests conclude that prices reflect all publicly available information and only unanticipated changes in economic data can surprise the market.This study is based on the notion that commodity markets are information driven mechanisms which determine equilibrium prices. If markets are active, the information is quickly disseminated among market participants who, upon trading, determine a fair price. It is also hypothesized that prices can reflect information which is not publicly announced by a governmental agency but yet successfully forecasted by private agents. The investigation of these hypotheses should shed light on three issues: (1). the existence of the semi-strong market efficiency and the speed of information dissemination with respect to the USDA crop forecasts, (2). the effectiveness of a governmental agency in providing the market with useful and reliable forecasts, and (3). the importance of private forecasting companies in predating official estimates and help traders form a direction of price changes, if not an accurate prediction. This article examines the first two issues explicitly and finds implications about the third.We chose five agricultural commodities traded at the Chicago Board of Trade (corn, wheat, soybeans, soybean meal, and soybean oil) to investigate the cash price behavior upon the (USDA) U.S. Department of Agriculture, crop size forecasts over 19 years (1966)(1967)(1968)(1969)(1970)(1971)(1972)(1973)(1974)(1975)(1976)(1977)(1978)(1979)(1980)(1981)(1982)(1983)(1984).The USDA crop announcements are made available on a monthly basis during the summer and fall months. Due to weather conditions during the critical stages of the crops the final size is very uncertain. The unique way in which USDA forecasts *This article has benefited from helpful discussions with Harry Grammatikos and Joel Arata. Any remaining errors are mine. The author thanks Panos Chrysanthis for the graph preparation.

The performance of event study approaches using daily commodity futures returns

“…One is a constant mean return (CMR) model that measures abnormal returns as prediction errors from some benchmark model of normal returns. Examples cited earlier that have used the CMR model include Milonas (1987), Schroeder, Blair, and Mintert (1990), Mann andDowen (1997), andMcKenzie andThomsen (2001). A second approach involves regression methods with abnormal returns being estimated as coefficients of dummy variables that correspond to event days.…”

“…Milonas (1987), Schroeder, Blair, and Mintert (1990), Mann andDowen (1997), andMcKenzie andThomsen (2001) all use an 8-day normal period.…”

“…1 The objective of these studies has been to test hypotheses relating to market efficiency, the extent to which commodity prices react to market related events, and/or the informational content of an event. For example, Milonas (1987), Schroeder, Blair, and Mintert (1990), Patterson and Brorsen (1993), and Mann and Dowen (1997), investigated the impacts of government reports on daily commodity cash or futures price changes. Robenstein and Thurman (1996) studied the response of meat futures prices to adverse health information.…”

Futures Price Response to Crop Reports in Grain Markets

“…In addition, the authors found that the impact of August, September, and October USDA harvest forecast releases were stronger than the forecasts announced in other months. Milonas () conducted a similar analysis with grain cash prices and the results pointed to a significant price adjustment after USDA report announcements.…”

Hedging House Price Risk in China

“…Two methods have been developed in the existing literature to calculate the normal return of commodity futures that serves as the benchmark, and both are adopted in the following analysis. First, following the mean return model by Milonas () and McKenzie, Thomsen and Dixon (), we use the average return over a length of time (from 30 trading days to 11 trading days before the event, which is in total 20 trading days) to determine what would otherwise be the normal return if the event had not occurred. The abnormal rate of return was calculated as the real return minus the normal return.…”