One-time snapshot

USDA Crop Insurance Indemnities 1989–2023

4,173,884 USDA crop insurance summary records across 35 commodity years, with corrected financial fields, exact decimal amounts and preserved source values.

Snapshot coverage

Records: 4,173,884 source summary records

Coverage: Commodity years 1989–2023

Format: CSV + Parquet

Package uploaded: 2026-09-19

Listing reviewed: 2026-09-19

The upload and review dates do not extend the source coverage. This is a dated snapshot; future updates are not included.

Get this snapshot

$99one time

CSV and Parquet files for the coverage described here. Review the limitations and sample before purchasing.

Buy dated snapshot

Commercial license terms

View 1,000-row sample

New annual update subscriptions are unavailable. Existing customers can use their purchase download links or contact support.

Known limitations

Dated source snapshot retrieved September 19, 2026: commodity years 1989–2023 only. One indemnity table; weather data, weather joins, inferred cause categories and risk scores are excluded. Future updates are not included.
Rows summarize commodity year, county, crop, insurance plan, coverage, stage, cause and reported loss month/year. They are not individual claims, farms or unique policies. Do not sum policy counts as unique customers or mix quantity units across crops.
Breaking schema repair: use indemnity_amount at official source position 29. The former indemnity field read the EFA premium discount position. Monetary fields now use exact decimal values; all 30 original fields remain in raw_ columns. CSV readers may discard raw NUL characters; prefer Parquet or the documented PyArrow reader for source fidelity.
All source rows are retained, including 74,886 negative and 47,524 zero indemnities, two exact duplicate records and 11 excess rows sharing the same source dimension key. Use source_file plus source_row as this revision’s row key. Two duplicate source amounts are $79,560 and $936,721; no silent deduplication occurred.
Loss year is missing in 1,442,370 rows and invalid in 2,320. Loss month is missing in 22 and invalid in 48,287. Typed values are null with status fields; 2,068 calendar-valid loss years differ from commodity year by more than two years. Use commodity_year for period coverage.
161,137 cause descriptions are missing. Historical codes, labels and program coverage vary; five-digit county code shape does not establish a contemporary geography crosswalk. Amounts are nominal USD and these loss summaries do not provide full-program exposure or validated risk rates.
The deterministic 1,000-row sample spans all 35 years but is not statistically representative. All delivered fields and annual money totals reconcile to the pinned annual source files; agency accuracy and historical revision completeness are not independently certified.

Files in this snapshot

Row counts are per table and should not be added as independent events. Expand a file to see its delivered Parquet field names and types.

usda_crop_insurance_indemnities.parquet — 4,173,884 rows · 65 fields

Field	Type
commodity_year	int32
state_code	string
state_abbr	string
county_code	string
county_name	string
commodity_code	string
commodity_name	string
insurance_plan_code	string
insurance_plan_name	string
coverage_category	string
stage_code	string
cause_of_loss_code	string
cause_of_loss_desc	string
month_of_loss_status	string
month_of_loss	int32
month_of_loss_name	string
year_of_loss_status	string
year_of_loss	int32
policies_earning_premium	int64
policies_indemnified	int64
net_planted_quantity	decimal128(30, 10)
net_endorsed_acres	decimal128(30, 10)
liability	decimal128(30, 10)
total_premium	decimal128(30, 10)
producer_paid_premium	decimal128(30, 10)
subsidy	decimal128(30, 10)
state_private_subsidy	decimal128(30, 10)
additional_subsidy	decimal128(30, 10)
efa_premium_discount	decimal128(30, 10)
net_determined_quantity	decimal128(30, 10)
indemnity_amount	decimal128(30, 10)
loss_ratio	decimal128(30, 10)
county_fips	string
source_file	string
source_row	int64
raw_commodity_year	string
raw_state_code	string
raw_state_abbr	string
raw_county_code	string
raw_county_name	string
raw_commodity_code	string
raw_commodity_name	string
raw_insurance_plan_code	string
raw_insurance_plan_name	string
raw_coverage_category	string
raw_stage_code	string
raw_cause_of_loss_code	string
raw_cause_of_loss_desc	string
raw_month_of_loss	string
raw_month_of_loss_name	string
raw_year_of_loss	string
raw_policies_earning_premium	string
raw_policies_indemnified	string
raw_net_planted_quantity	string
raw_net_endorsed_acres	string
raw_liability	string
raw_total_premium	string
raw_producer_paid_premium	string
raw_subsidy	string
raw_state_private_subsidy	string
raw_additional_subsidy	string
raw_efa_premium_discount	string
raw_net_determined_quantity	string
raw_indemnity_amount	string
raw_loss_ratio	string

Inspect the public CSV

import pandas as pd

df = pd.read_csv(
    "https://huggingface.co/datasets/claritystorm/usda-crop-insurance-indemnities-weather/resolve/main/sample_1000.csv"
)
print(df.shape)
print(df.columns.tolist())
print(df.head())

This CSV is a sample, not the full package. It does not establish complete historical coverage or represent every field’s missingness. License details are on the Hugging Face card.

Source: government source portal · Dataset changelog