«Crop Forecasts Office of the Chief Economist, World Agricultural Outlook Board Miscellaneous Publication No. 1554 Understanding Crop Statistics. ...»
Office of the Chief
Understanding Crop Statistics. Frederic A. Vogel, National Agricultural Statistics Service, and
Gerald A. Bange, World Agricultural Outlook Board, Office of the Chief Economist, U.S. Department of
Agriculture. Miscellaneous Publication No. 1554.
Abstract Each month, the U.S. Department of Agriculture publishes statistics and related information about crop production in the United States and the world. Several USDA agencies are responsible for preparing these statistics. The methodology used by the National Agricultural Statistics Service to prepare the crop forecasts is described to give the user of these data a better understanding of the strengths and weaknesses. This is followed by a similar description of the world agriculture supply and demand estimates prepared by the World Agricultural Outlook Board.
Preface The crop supply and demand estimates prepared by the USDA are crucial to both policy makers in government and people involved in making decisions about marketing and investing.
In today’s information age, the statistical methods described within this paper provide the benchmark against which all other data sources are compared. The agencies involved have a solid record of objectivity and ability to meet established report dates. This objectivity is preserved because it is USDA policy to make it so. The Secretary of Agriculture’s office is briefed about the results only after the final results have been completed and prepared for distribution to the public.
The security of the data before release is fiercely defended with extraordinary efforts to ensure there is no premature disclosure of any of the information.
Hopefully this report will answer the many questions often raised about how the estimates of production, supply and demand are generated.
EEO statement The United States Department of Agriculture (USDA) prohibits discrimination in all its programs on the basis of race, color, national origin, gender, religion, age, disability, political beliefs, sexual orientation, and marital or family status. (Not all prohibited bases apply to all programs.) Persons with disabilities who require alternative means for communication of program information (braille, large print, audiotape, etc.) should contact the USDA’s TARGET Center at 202-720-2600 (voice and TDD).
To file a complaint of discrimination, write USDA, Director, Office of Civil Rights, Room 326-W, Whitten Building, 14th and Independence Avenue, SW, Washington, D.C., 20250-9410, or call 202-720-5964 (voice or TDD). USDA is an equal opportunity provider and employer.
Each month, the U.S. Department of Agriculture (USDA) publishes crop supply and demand estimates for the Nation and the world. These estimates are used as benchmarks in the marketplace because of their comprehensive nature, objectivity, and timeliness. The statistics that USDA releases affect decisions made by farmers, businesses, and governments, by defining the fundamental conditions in commodity markets. When using USDA statistics, it is helpful to understand the estimating procedures used and the nature and limitations of crop estimates.
Several agencies within USDA are responsible for preparing crop statistics. The National Agricultural Statistics Service (NASS) forecasts U.S. crop production based on data collected from farm operations and field observations. Forecasts for each crop season begin with a winter wheat and rye seedings report in early January followed by a March report that gives a first look at what farmers intend to plant. This is followed in late June by a report of the acreage actually planted. Monthly yield and production forecasts begin in May for winter wheat, in July for spring wheat and other small grains, and in August for other spring-planted crops, concluding with estimates of actual production at the end of the harvesting season.
NASS also conducts quarterly surveys of grain and soybeans stored on and off farms.
The World Agricultural Outlook Board (WAOB) coordinates an interagency process that prepares monthly forecasts of supply and demand for major crops, both for the United States and the world, and follows a balance-sheet approach to account for supplies and utilization. The major components of the supply and demand balance sheet are beginning stocks, production, domestic use, trade, and end-ofseason carryout stocks. Whereas forecasts of U.S. crop production and estimates of U.S. stocks on hand are independently prepared by NASS, U.S. and foreign supply and demand forecasts are developed jointly by several USDA agencies.
The NASS Crop Production report and the World Agricultural Supply and Demand Estimates (WASDE) report are prepared simultaneously in a secured area and released at 8:30 a.m. Eastern Time between the 9th and 12th day of each month. Joint preparation enables USDA analysts to incorporate the new NASS production forecasts of U.S. crops into supply and demand estimates. These estimates provide an overview for more detailed analyses published by other agencies of USDA, especially the Foreign Agricultural Service (FAS) and the Economic Research Service (ERS). Country-level data published by all USDA agencies must be consistent with the supply and demand numbers released by NASS and WAOB.
USDA strives to provide the agricultural community with estimates that are accurate, objective, reliable, and timely. Security measures are used to prevent leaks of market-sensitive information. The agencies conduct research to improve estimating techniques. The following discussion provides a closer look at how USDA arrives at its estimates, the data sources used, and the policies governing initial reports and revisions.
Preparing NASS Production Forecasts
Crop production forecasts have two components--acres to be harvested and expected yield per acre. For example, preliminary corn and soybean acreage estimates are made using data obtained from a survey of farmers conducted during the first 2 weeks in June. Expected corn and soybean yields are obtained monthly, August through November, from two different types of yield surveys. Data from the yield surveys reflect conditions as of the first of the month, as data are collected during the last week of the previous month and the first 2 or 3 days of the current month.
Crop production forecasts are based on conditions as of the survey reference date and projected assuming normal conditions for the remainder of the season. For example, the assumption of "normal conditions" is that temperatures and precipitation will be at historic averages for the remainder of the season. It is assumed that the first killing frost will occur on the historic average date. The crop maturity and conditions at the reference date are evaluated against the time remaining until the expected frost--if one third of the crop will not reach maturity until the frost date has passed, it is assumed that some frost damage will result. Long-range weather projections are not used as an indicator for final yield.
The reference point for crop forecast surveys is the first of the month, which is also usually close to the mid-point of data collection. Both grower-reported average yields and objective-measurement modeled yields contain a measurable forecast error based on the historic difference between these survey estimates and the final end-of-season yield. The review process followed to develop the monthly yield forecasts involves evaluating the relative ranges of the forecast errors of the grower yields and the objective measurement yields and the degree to which they overlap.
If there is a significant change in conditions between the survey period and the report date such as a killing freeze, serious heat wave, beneficial rains, etc., the primary goal is to provide the most accurate production forecast possible given the available survey data. The official estimate may represent a departure from the survey averages, but still will reflect the current crop conditions within the ranges provided by the data. When NASS states as policy that it is forecasting based on conditions as of the first of the month, it is saying that it will establish yields within the range of the survey estimates.
When forecasting crop yields, NASS does not attempt to predict future weather conditions. Long-range weather forecasts are not used in any forecast models. To the extent that conditions depart from normal, the forecasts also will fluctuate. Procedures used to prepare acreage estimates and yield forecasts are discussed in the following sections.
Base for Acreage Planted and To Be Harvested The largest single survey NASS conducts each year is the June Agricultural Survey. During the first 2 weeks in June, about 2,400 interviewers contact over 125,000 farmers, either by telephone or in person, to obtain information on crop acreages, grain stocks, and livestock inventories. These producers are asked to report the acreage, by crop, that has either been planted or that they intend to plant, and the acreage they expect to harvest as grain. Data from this survey are used to estimate, among other things, total acres planted to corn, soybeans, and other crops regardless of the intended uses. Preliminary projections of acres to be harvested for grain or soybeans, including seed, are also made using these data.
This survey also provides estimates of quantities of grain stored on farms and livestock inventories.
The sample design for this survey utilizes two different sampling frames. The area frame, which is essentially the entire land mass of the United States, ensures complete coverage of the U.S. farm population.
The list frame, a list of known farmers and ranchers, does not provide complete coverage of all farms, but allows the use of more efficient data collection methods.
Sampling from the area frame is a multi-step process. First, all land in each State is classified into land use categories by intensity of cultivation using a variety of map products, satellite imagery, and computer software packages. These land use classifications range from intensively cultivated areas to marginally cultivated grazing areas to urban areas. The land in each use category is then divided into segments ranging from about 1 square mile in cultivated areas to 0.1 square mile in urban areas. This allows intensively cultivated land segments to be selected with a greater frequency than those in less intensively cultivated areas. Segments representing cultivated areas are selected at a rate of about 1 out of 125. Sample segments in land use classifications with decreasing amounts of cultivated land are selected at rates ranging from 1 out of 250 to 1 out of 500.
About 10,000 area segments are selected nationwide for the survey conducted each June. Using maps and aerial photos showing the exact location and boundaries of each sample segment, interviewers locate and interview every operator with land inside the segment boundaries to identify crops planted in each field, and to obtain livestock inventory information, and quantities of grain in storage. A similar survey is conducted in early December which provides a measure of winter wheat acres planted.
Before sampling from the list, each farm is classified by various characteristics such as number of acres by crop. Large farms are sampled at high rates. For example, Illinois farms on the list with over 5,500 acres of cropland, or grain storage capacity exceeding 500,000 bushels, are selected with certainty, as are Iowa farms with over 5,000 acres of cropland. Smaller farms are selected at rates of 1 out of 25 to 50.
About 75,000 farms across the United States are selected from the list to be surveyed during the same time period in June. Farmers on the list sample are asked to provide total acres planted for each crop on all the land they operate, and quantities of grain stored on their operation. Most of the data from this sample are collected by telephone interviewers.
Data from the area and list samples are combined using multiple-frame statistical methodology developed jointly by NASS and Iowa State University, which ensures that all land areas in the United States can be accounted for once and only once. The June Agricultural Survey is subsampled for surveys in July, September, December, and March for the basic livestock inventory, crop production and quarterly stocks estimates.
Generally, estimates of planted acres from the June Agricultural Survey are not changed during the crop season. However, occasionally the planting season runs late and many fields are not yet planted with the intended crops at the time the survey is conducted. When this happens, adjustments to planted acres estimates may be made at the time of the first yield forecast in August. The preliminary projections for harvested acres may also be adjusted using data from the August yield surveys. When significant portions of the crops are not yet planted, during the survey farmers are asked what they intend to plant.
Then, in years such as 1993, 1995, and 1996 these farms are re-visited during late July to determine what was actually planted. If necessary, harvested acreage estimates will be revised and published in the August Crop Production report.
Yield Forecasts A subsample of farmers who respond to the list portion of the June Agricultural Survey is selected to provide monthly crop yield projections. This provides a way to screen farmers so that only those currently growing the commodities of interest are contacted during the monthly surveys. This subsample may be supplemented with other known growers randomly selected from the list frame when monthly district level production forecasts are required for some States.
The sampled farmers are asked what they expect their crops to yield before harvest, and actual yields obtained at harvest. All yield data for an individual report are weighted by the farm's crop acres for harvest.
Objective Yield Surveys are conducted monthly in States that contribute most heavily to total U.S.
production of corn, soybeans, cotton, and wheat. These surveys provide information for making forecasts and estimates of crop yields based on counts, measurements, and weights obtained from small plots in a random sample of fields.