USDA NASS Crop Surveys: The Federal Agricultural Data Behind Every Corn, Soybean, and Wheat Market

Since 1867 the federal government has conducted systematic surveys of American farms. Today the National Agricultural Statistics Service runs more than 400 surveys a year, reaches 3 million respondents, and publishes the numbers that determine whether corn futures open higher or lower on any given Monday morning. Every commodity trader, agricultural lender, food processor, and grain exporter in the world builds their models on the same underlying source: NASS.

What NASS is and why it exists

The National Agricultural Statistics Service is a principal federal statistical agency within the USDA, with a mandate established under the Census of Agriculture Act and the Agricultural Statistics Act of 1962. Its core function is to produce unbiased, statistically rigorous estimates of what American farmers are growing, raising, and selling. It operates 12 regional field offices covering all 50 states and Puerto Rico, each staffed with agricultural statisticians who design surveys, conduct outreach to farm operations, and validate responses against independent administrative sources.

The statistical program spans crops (planted and harvested acreage, yield per acre, total production, stocks in all positions), livestock (inventories, slaughter weights, milk and egg production), prices (prices received by farmers and prices paid for inputs), labor (hired farm workers, wages), and chemical use (pesticides, fertilizers, irrigation). No private data provider can replicate the breadth or the mandatory survey authority that NASS carries: farms above threshold size are legally required to respond to selected NASS surveys under 7 U.S.C. 2204g.

NASS data is free and public. Every estimate, along with historical series going back decades, is available through the QuickStats portal at quickstats.nass.usda.gov and its RESTful API. The release schedule is announced months in advance and adhered to with the same precision that the BLS uses for CPI releases—because commodity markets depend on it.

The Crop Production report: the flagship

The most market-moving document NASS produces is the monthly Crop Production report. During the growing season—May through November—NASS releases updated estimates of planted acreage, harvested acreage, yield per acre, and total production for every major field crop. The January final report closes out the marketing year. These are not forecasts in the econometric sense; they are survey-based statistical estimates derived from direct grower contacts, objective field measurements, and satellite-derived crop condition data.

Three other NASS crop reports anchor the annual calendar alongside Crop Production:

Each Crop Production report revises every prior estimate in the same growing season. The August report is typically the most consequential for the corn and soybean complex: it incorporates objective yield surveys—physical measurements of ears and pods in randomly selected plots—that establish the first high-confidence yield estimate of the season. Traders watch the August report with the same attention that bond markets give to Federal Reserve decisions.

Geographic hierarchy: crop reporting districts

NASS does not publish only national aggregates. The geographic reporting structure follows a three-level hierarchy:

At the finest level, each state is divided into nine crop reporting districts (CRDs), numbered 10 through 90 (Northwest, North, Northeast, West, Central, East, Southwest, South, Southeast). Districts are contiguous geographic groupings of counties with similar agricultural production patterns. A district in Iowa will be predominantly corn and soybeans; one in California may be dominated by tree fruits and vegetables. District-level estimates for acreage and production are published in the state statistical bulletins that accompany each Crop Production release.

State-level estimates aggregate the nine districts. National estimates aggregate all states. The QuickStats database stores all three levels—national, state, and district—and the API accepts the agg_level_desc parameter to select among them. County-level estimates are published for some crops and years in the Census of Agriculture (conducted every five years), though not in the annual survey program, because sample sizes at the county level do not support statistically reliable annual estimates for most commodities.

Key crops: the numbers behind the markets

NASS covers dozens of field crops and specialty crops. The following five account for the overwhelming majority of US cropland and commodity market activity:

Corn

Corn occupies roughly 35 percent of total US harvested cropland in a typical year— around 82–85 million acres. Iowa, Illinois, and Nebraska consistently rank as the top three producing states and together account for nearly half of national corn production. The NASS core corn series tracks planted acres, harvested acres (which differs from planted when fields are abandoned due to weather or pests), yield per acre (bushels per acre), and total production (bushels). The national average corn yield has risen from roughly 100 bushels per acre in the 1980s to over 180 bushels per acre by the early 2020s, reflecting hybrid seed improvement, precision agronomy, and irrigation expansion. The 2012 drought punctured that trend line dramatically: national average yield fell to 123.4 bushels per acre from 147.2 the prior year.

Soybeans

Soybeans are the second-largest row crop by planted area, typically 85–90 million acres, and the largest US agricultural export by value in most years. The global soybean market is now genuinely bipolar: Brazil surpassed the United States as the world's largest soybean producer around 2012 and has continued widening its lead. Brazil's Conab crop estimates (the Brazilian NASS equivalent) are now tracked as closely as NASS soybean reports by traders positioning in CBOT soybean futures, because a large Brazilian crop can offset a US shortfall—and vice versa. During the 2012 drought, soybean futures briefly exceeded $17 per bushel on the Chicago Board of Trade, a level not seen before or since, as both US and South American production were simultaneously stressed.

Winter and spring wheat

NASS tracks three wheat classes separately: hard red winter wheat (Kansas, Oklahoma, Texas—the primary bread wheat), soft red winter wheat (eastern Corn Belt and Mid-Atlantic), and spring wheat (North Dakota, Minnesota, Montana). Winter wheat is planted in fall, goes dormant through winter, and is harvested the following June and July; the Small Grains Summary provides final production estimates in September. Spring wheat is planted after snowmelt and harvested in August. The Hard Red Winter Wheat Seedings report in January establishes the first look at winter wheat area for the upcoming marketing year.

Cotton

Cotton acreage has declined substantially from its 1990s peaks but still occupies 10–12 million acres in the Belt states (Texas, Georgia, Arkansas, Mississippi, North Carolina). NASS tracks upland cotton separately from Pima (extra-long staple) cotton, reporting planted acres, harvested acres, yield per acre (in pounds), and total production in running bales (480 pounds). Cotton futures on ICE (Intercontinental Exchange) respond to NASS Crop Production releases on the same release calendar as CBOT corn and soybeans.

Rice

US rice production is relatively small by global standards but strategically significant: the United States is among the world's top exporters despite modest domestic production of roughly 20–25 million hundredweight (cwt). Arkansas dominates, accounting for roughly 45 percent of national rice production, followed by California. NASS reports rice in units of hundredweight per acre (yield) and total hundredweight (production), unlike grain crops that use bushels.

WASDE: the world supply-demand balance sheet

NASS produces the survey estimates, but the most closely watched monthly USDA release—the World Agricultural Supply and Demand Estimates (WASDE)— is produced by a separate USDA entity: the World Agricultural Outlook Board (WAOB). WASDE is released on the same day as the Crop Production report, typically in the late morning Eastern time, and incorporates NASS estimates into global supply-demand balance sheets for every major commodity.

A WASDE balance sheet for, say, US corn consists of: beginning stocks (carried over from the prior marketing year) + production (from NASS Crop Production) + imports = supply; feed and residual use + food/seed/industrial use + exports = total use; supply minus total use = ending stocks. The stocks-to-use ratio derived from this balance sheet is the single most important number in commodity price analysis. When ending stocks as a share of total use fall below roughly 10 percent for corn or 5 percent for soybeans, prices typically spike nonlinearly because any incremental demand shock or supply miss cannot be covered by inventory drawdown.

The WASDE is produced under strict information security procedures comparable to those the BLS uses for CPI releases. USDA staff with access to pre-release WASDE numbers are sequestered on the day of release; they cannot leave the building or communicate with the outside world until the report goes public via the USDA website at the scheduled time. The lockup procedure exists to prevent early leakage into commodity markets—a recurring concern given the billions of dollars in futures positions that move on WASDE release day.

How NASS reports move commodity markets

CBOT corn and soybean futures are among the most liquid commodity contracts in the world, with daily volume exceeding 500,000 contracts on major report days. The price discovery role of NASS is direct and measurable: academic research consistently finds statistically significant abnormal returns in corn and soybean futures in the minutes surrounding USDA report releases, with magnitudes proportional to the surprise in yield estimates relative to pre-release analyst consensus.

“Crop report trading” refers to the strategies commodity hedge funds and grain trading companies build around the NASS release calendar. Pre-report positioning is common; post-report volatility is predictable. CME Group's position limits and special margin requirements apply specifically during USDA report windows. Some years the August Crop Production report is the most significant individual piece of economic data for agricultural commodity prices of the entire calendar year—a title that competes only with the March Prospective Plantings release for the planting season outlook.

The 2012 drought provides the canonical example of NASS report impact on markets. The sequence: late June 2012 Acreage report showed fewer corn acres than expected. The July Crop Production report dropped the national yield forecast sharply. By the August report—which established the first objective yield survey numbers— the national corn yield estimate had fallen to 123.4 bushels per acre. Corn futures hit $8.49 per bushel, the highest price in the contract's history. Soybean futures simultaneously rose above $17. The NASS estimate sequence is the paper trail documenting that entire market event.

Weekly Crop Progress: the report that moves futures every Monday

While the monthly Crop Production report is the heavyweight, the weekly Crop Progress report is what commodity traders watch most closely during the growing season. Released every Monday afternoon (or Tuesday if Monday is a federal holiday) from April through November, Crop Progress is a snapshot of where the crop stands across the top producing states.

For each of 18 major crops, NASS reports the percentage of the crop at each developmental stage—planted, emerged, silking (corn only), setting pods (soybeans), headed (wheat), dough, dented, mature, harvested—compared to the five-year average for that date. A crop that is running significantly behind the average developmental pace entering a freeze-risk period or heading into a dry spell is a meaningful supply threat that shows up immediately in the Crop Progress data.

The condition rating component is the most market-sensitive element. For each crop and state, NASS reports the percentage rated Excellent, Good, Fair, Poor, or Very Poor based on field observations by NASS enumerators. The market convention is to collapse these into a “Good/Excellent” percentage—the sum of the top two categories—and track its week-over-week change. A Good/Excellent reading of 70 percent or above for corn in the first week of August (when corn is in its critical pollination and grain-fill period) implies a high probability of a trend-line or better yield. A reading of 50 percent or below is a supply signal that corn futures will typically price in immediately after the Monday afternoon release.

State-by-state breakdowns in Crop Progress allow traders to distinguish regional stress from national averages. A drought in the western Corn Belt (Nebraska, Kansas) may drag the national Good/Excellent number while Iowa and Illinois—which together account for a disproportionate share of production—remain in good condition. Crop Progress by state is the data that separates the regionally informed trader from one relying on national averages alone.

Livestock surveys: cattle, hogs, and poultry

NASS survey coverage extends well beyond field crops. Three livestock surveys are particularly significant for commodity and food industry analysis:

Cattle on Feed is a monthly survey of feedlots with capacity of 1,000 head or more. It reports three numbers: total cattle and calves on feed at the start of the month (the inventory figure), placements during the month (cattle moved into feedlots from ranches and backgrounding operations), and marketings during the month (cattle sold to packing plants). The on-feed inventory is the most forward-looking indicator of beef supply: feedlot cattle typically take 120–150 days to reach slaughter weight, so on-feed numbers today predict packing-plant kill volumes four to five months out. Cattle futures on CME respond to Cattle on Feed releases the Friday of its publication.

Hogs and Pigs is published quarterly (March, June, September, December) and covers breeding herd inventory (sows kept for breeding), the market hog inventory by weight class, and farrowing intentions (how many sows producers expect to farrow in the next two quarters). Because hog production cycles are shorter than cattle (pigs go to market in roughly six months from birth), the quarterly Hogs and Pigs report provides approximately two quarters of forward supply guidance. Lean hog futures on CME trade on this release calendar.

Poultry Production covers broiler (meat chicken) production, table egg production, and turkey production. NASS publishes monthly egg production (total and by production type—cage, cage-free, organic), hatching egg data, and monthly broiler and turkey slaughter numbers. The poultry data is particularly useful for tracking the impact of avian influenza outbreaks, which trigger mass depopulations of affected flocks and show up as sharp dips in the production series.

Agricultural prices: the parity ratio

NASS publishes two monthly price series that together define the economic position of US farmers:

Prices Received by Farmers covers every significant agricultural commodity—corn, soybeans, wheat, cotton, cattle, hogs, milk, eggs, and dozens more—at both the national and state level. The monthly report shows average prices received by producers for each commodity during the reference month, which are the prices at the farm gate (not the futures price). These are the input numbers for calculating farm revenue and for tracking the spread between futures prices and the actual prices farmers receive.

Prices Paid by Farmers tracks input costs: fertilizer (nitrogen, phosphate, potash), pesticides and herbicides, diesel fuel, seed, hired labor, and machinery. The ratio of the prices-received index to the prices-paid index is the parity ratio, a concept dating to the 1920s that tracks whether farmers' purchasing power has kept pace with their production costs. A parity ratio below 100 means farmers must sell more output than they would have in the reference period to afford the same inputs. The parity ratio is published monthly and remains politically significant in Farm Bill debates: periods of low parity are historically associated with farm sector financial stress, rural bank failures, and political pressure for commodity support programs.

The QuickStats API: accessing the full database

All NASS survey data is accessible through the QuickStats portal at quickstats.nass.usda.gov and its RESTful API at quickstats.nass.usda.gov/api. A free API key is required and available through the same portal. The API handles the full NASS database including historical series going back 50 or more years for major commodities.

The primary endpoint is:

GET https://quickstats.nass.usda.gov/api/api_GET/?key={YOUR_KEY}&commodity_desc=CORN&statisticcat_desc=YIELD&unit_desc=BU%20/%20ACRE&agg_level_desc=STATE&format=JSON

Key query parameters:

The API returns at most 50,000 records per call. For large pulls—all commodities across all states across all years—use the bulk download option on the QuickStats website, which exports the full filtered dataset as a CSV without the record limit. The QuickStats web interface also has a parameter-builder that generates the corresponding API URL, which is the fastest way to identify the correct parameter values for an unfamiliar series.

The Value field in the returned records is always a string. Non-disclosable values are represented as ' (D)' (withheld to avoid disclosing individual operations), ' (Z)' (less than half the unit shown), or ' (NA)'. Strip commas and filter for numeric strings before casting to float.

Python: state corn yield per acre since 2004—with the 2012 drought

The script below queries QuickStats for annual corn yield per acre at the state level for the top five producing states, plots a 20-year time series, and annotates the 2012 drought year—the most severe single-year yield collapse in the modern corn production era.

import requests
import pandas as pd
import matplotlib
matplotlib.use("Agg")
import matplotlib.pyplot as plt

# USDA NASS QuickStats API
# Register for a free key at: https://quickstats.nass.usda.gov/api
# Documentation: https://quickstats.nass.usda.gov/api#param_define

QUICKSTATS_API = "https://quickstats.nass.usda.gov/api/api_GET/"
API_KEY = "YOUR_NASS_API_KEY"  # free at quickstats.nass.usda.gov/api

def fetch_nass(params, api_key):
    """
    Query the NASS QuickStats API.
    Returns a list of records as dicts.
    Common parameters:
      commodity_desc   - e.g. 'CORN'
      statisticcat_desc - e.g. 'YIELD' or 'AREA HARVESTED'
      unit_desc        - e.g. 'BU / ACRE' or 'ACRES'
      agg_level_desc   - 'STATE' or 'NATIONAL' or 'COUNTY'
      year             - e.g. '2012' or leave blank for all years
      state_alpha      - two-letter state abbreviation, e.g. 'IA'
      short_desc       - pre-built series label (optional alternative to above)
    """
    params["key"] = api_key
    params["format"] = "JSON"
    resp = requests.get(QUICKSTATS_API, params=params, timeout=60)
    resp.raise_for_status()
    body = resp.json()
    if "data" not in body:
        raise ValueError("Unexpected NASS response: " + str(body))
    return body["data"]

# Pull corn yield per acre at the state level for all available years
records = fetch_nass(
    {
        "commodity_desc": "CORN",
        "statisticcat_desc": "YIELD",
        "unit_desc": "BU / ACRE",
        "agg_level_desc": "STATE",
        "sector_desc": "CROPS",
        "domain_desc": "TOTAL",
        "freq_desc": "ANNUAL",
    },
    API_KEY,
)

df = pd.DataFrame(records)

# QuickStats returns Value as a string; suppress non-numeric (e.g. ' (D)', ' (Z)')
df = df[df["Value"].str.strip().str.match(r"^[0-9,\.]+$")].copy()
df["Value"] = df["Value"].str.replace(",", "").astype(float)
df["year"] = df["year"].astype(int)

# Filter to past 20 years and the top corn-producing states
CURRENT_YEAR = 2024
START_YEAR = CURRENT_YEAR - 20
TOP_STATES = ["IOWA", "ILLINOIS", "NEBRASKA", "MINNESOTA", "INDIANA"]

df_filtered = df[
    (df["year"] >= START_YEAR)
    & (df["state_name"].isin(TOP_STATES))
].copy()

# Pivot to state columns for plotting
pivot = (
    df_filtered
    .pivot_table(index="year", columns="state_name", values="Value", aggfunc="mean")
    .sort_index()
)

# --- Plot ---
fig, ax = plt.subplots(figsize=(11, 6))

for state in TOP_STATES:
    if state in pivot.columns:
        ax.plot(pivot.index, pivot[state], marker="o", markersize=3, linewidth=1.5,
                label=state.title())

# Annotate the 2012 drought year
ax.axvline(x=2012, color="#c0392b", linewidth=1.2, linestyle="--", alpha=0.7)
ax.text(2012.15, ax.get_ylim()[0] + 5, "2012 drought", color="#c0392b",
        fontsize=8, va="bottom")

ax.set_title("Corn Yield per Acre by State (top 5 producing states)", fontsize=13)
ax.set_xlabel("Year")
ax.set_ylabel("Yield (bu / acre)")
ax.legend(fontsize=9, loc="upper left")
ax.grid(axis="y", alpha=0.3)
fig.tight_layout()
fig.savefig("nass_corn_yield_by_state.png", dpi=150)
print("Saved nass_corn_yield_by_state.png")

# Print summary table
print("\nCorn yield per acre (bu/acre) - top 5 states, past 20 years")
print(pivot.round(1).to_string())

# Identify the 2012 drought year in Iowa
if "IOWA" in pivot.columns and 2012 in pivot.index and 2011 in pivot.index:
    iowa_2011 = pivot.loc[2011, "IOWA"]
    iowa_2012 = pivot.loc[2012, "IOWA"]
    iowa_2013 = pivot.loc[2013, "IOWA"] if 2013 in pivot.index else float("nan")
    print("\n2012 drought impact on Iowa corn yield:")
    print("  2011: " + str(round(iowa_2011, 1)) + " bu/acre")
    print("  2012: " + str(round(iowa_2012, 1)) + " bu/acre  <-- drought year")
    print("  2013: " + str(round(iowa_2013, 1)) + " bu/acre  <-- recovery")

The 2012 signal is unmistakable. Iowa, the top corn-producing state, fell from 172 bushels per acre in 2011 to 134 bushels per acre in 2012—a 22 percent single-year collapse. Illinois dropped from 158 to 105 bushels per acre. Nebraska, which has substantially more irrigated acreage than the eastern Corn Belt, held up relatively better but still fell sharply. The 2013 recovery was nearly complete for Iowa and Illinois. The pattern is exactly what shows up in the QuickStats data and drove the historic CBOT corn price spike.

Report release procedures and information security

NASS operates under strict information security rules codified in USDA's statistical policy directives. Staff with access to pre-release Crop Production, Prospective Plantings, Acreage, and Grain Stocks estimates are subject to lockup procedures on release day: no external communication, physical confinement to secured areas of the NASS building in Washington, and release only through the USDA website at the scheduled time (typically noon Eastern, though the exact time has shifted across years). Phones and internet-connected devices are collected.

The same lockup applies to the WASDE release, which is produced under similar procedures within WAOB. Both the Crop Production and WASDE releases occur simultaneously on the same day during the growing season, and both are published at the same URL on the USDA's Economics, Statistics, and Market Information System (ESMIS) repository.

Historical violations of the lockup—including a 2013 incident in which an Iowa USDA employee leaked WASDE data ahead of release—have resulted in criminal prosecution under the federal insider trading statutes as applied to commodity markets. The economic value of pre-release access to NASS estimates is not theoretical; it is directly convertible to futures market profits, and federal regulators treat premature disclosure accordingly.

How NASS data connects to other federal datasets

The NASS statistical program does not exist in isolation. Several cross-references expand its analytical reach:

The USDA Economic Research Service (ERS) builds on NASS estimates to produce farm income forecasts, food price outlook analyses, and the farm sector balance sheet. ERS's annual “Farm Sector Income and Finances” publication combines NASS production data with prices received and input cost data to estimate net farm income by state and nationally—the NIPA-equivalent for the agricultural sector. When net farm income collapses (as it did in 2015–2016 after the post-2012 commodity price spike reversed), the ERS farm income series, grounded in NASS data, is the primary federal record of that stress.

The Census of Agriculture, conducted every five years by NASS (not the Census Bureau, despite the name), provides the deepest cross-sectional look at US farm structure: farm size distribution, operator demographics, land ownership and rental patterns, sales by commodity class, and operating expenses at the county level. The Census of Agriculture is to the annual NASS surveys what the decennial Census is to the ACS— lower frequency but far greater geographic and demographic depth.

USDA's Risk Management Agency (RMA) publishes crop insurance indemnity data that joins naturally to NASS yield estimates. When actual yields fall below the insured guarantee—determined by NASS county average yields under the Actual Production History program—indemnity payments flow to insured producers. Mapping NASS yield shortfalls to RMA indemnity records by county and year reconstructs the full economic impact of drought, flood, or other yield-limiting events at the sub-state level.

Related writing: USDA SNAP participation data covers the other major USDA statistical series—monthly food assistance enrollment and benefits by state—produced by the Food and Nutrition Service rather than NASS.

Related writing: FHWA Highway Data: Bridge Conditions, Pavement Quality, and Traffic Counts covers another federal infrastructure dataset where a release calendar and geographic hierarchy drive both policy decisions and market analysis.

Related writing: BEA GDP and National Accounts covers the federal macro accounting framework that the agricultural sector feeds into through farm income, personal income, and trade flow data.