Virginia Tech® home

Executive Summary

Executive Summary

This study reviewed over 80 previous studies that evaluated the accuracy and market impact of USDA reports relevant to the corn markets.  These reports included Prospective Plantings, Acreage, Crop Production, Crop Progress, Grain Stocks, Export Sales and World Agricultural Supply and Demand Estimates (WASDE).  Our analysis of previous literature revealed

  • Extensive evidence of the lack of bias and consistent accuracy in USDA yield forecasts with accuracy improving over time.
  • Revisions to USDA production forecasts were sometimes positively correlated and directionally consistent, suggesting they could be smoothed.  However, it is difficult to anticipate this smoothing pattern ahead of time and market participants seem to be aware of it, thus its impact is likely minimal.
  • Additional sources of information, such as weather models, crop conditions and private precision agriculture data, demonstrated limited potential for improving USDA yield forecasts.
  • WASDE price forecasts were traditionally published as a range without a confidence level.  Several studies suggested appropriate methods for evaluating their accuracy as well as developed alternative methods for generating more accurate prediction intervals.  However, USDA switched to publishing point price forecasts in May 2019. Midpoints of WASDE corn price forecasts were typically more accurate than futures-based alternatives.
  • Errors in WASDE balance sheet forecasts were associated with structural changes in commodity markets that took place in the mid-2000s, challenges with predicting periods of economic growth and changes in exchange rates.  Errors in supply and use forecasts contributed to ending stocks forecast errors. Ending stocks forecasts were inefficient with predictability in revisions consistent with smoothing.
  • There was a notable decline in the ability of market participants to anticipate USDA grain stock estimates for corn through 2013.  These large surprises for corn grain stocks were likely due to unresolved sampling errors for corn production estimates, suggesting a need for a revision of January corn production estimate as well as a survey of corn feed use.
  • While USDA baseline corn price forecasts have large errors, they tend to perform better than futures-based alternatives up to three years in advance USDA’s corn Prospective Plantings, Acreage, and October Crop Production forecasts were consistently more accurate than private forecasts. The only evidence of private forecasts dominating the USDA was found for August corn production during 1990s and early 2000s. However, USDA has regained its advantage in August corn production forecasts since the mid-2000s.
  • Extensive evidence of futures market reaction to USDA reports in the form of larger return variance.  However, this reaction signals that these reports provide new information to the markets. Reaction tends to be stronger when multiple reports are released at the same time.
  • Extensive evidence of futures market reaction to unanticipated information in USDA reports measured as the difference between USDA announcement and private industry expectations, or “market surprise.”  Strongest reaction to October and November crop production surprises and January and March grain stock surprises.
  • Extensive evidence of WASDE reports reducing market uncertainty as measured by a drop in option implied volatility following report release. Market uncertainty in corn and soybean markets decreases not only immediately following the report release but remains low for up to five days. This decrease in market uncertainty was more pronounced when there had been greater disagreement among industry expectations prior to the reports.
  • Information value tests show that August and October crop production reports get us closer to knowing the final estimate, thereby reducing uncertainty in the markets. Prospective Plantings and Acreage reports play a much bigger role in reducing supply uncertainty than Crop Production reports since 1983. The informational value of USDA forecasts has increased over time and was the strongest in the most recent 2002-2019 subperiod.
  • Several studies demonstrated positive trading returns would be possible from knowing USDA’s crop production reports in advance.
  • Intraday data allowed to measure that the strongest price reactions to the releases were found immediately after markets opened and persisted for about ten minutes. Some subtle reactions in the last trading session before the release, suggested that traders adjusted their market exposure in anticipation of the release.
  • Extending trading hours resulted in brief shocks in corn futures price volatility around the release of WASDE reports that may pose a challenge for producers whose risk management strategies are affected by intraday price swings. Typically, however, the heightened volatility has not lasted more than 30 to 60 minutes, thus it should not affect long-term risk management positions.

It is important to recognize the role of USDA information in market price discovery and reduction of information asymmetry. In order to maintain this role and ensure the quality of this information, it is important to keep communication lines open. Greater efforts can be made to increase transparency of USDA outlook and data products.