Every year, bad weather causes crops to fail. The federal government offers crop protection insurance to farmers, ranchers and other agricultural entities through the Federal Crop Insurance Corporation (FCIC), which is managed by the Risk Management Agency (RMA) of the U.S. Department of Agriculture (USDA). The FCIC insures against the loss of crops due to natural disasters such as hail, drought and floods, as well as the loss of revenue due to the decline in prices of agricultural commodities. With 1.3 million crop insurance policies spread out across 3,200 counties, it is unlikely that agents could physically visit each location to confirm submitted insurance claims, opening the barn door for fraud, waste and abuse. However, GCN reports that the RMA maintains a partnership with the Tarleton State University’s Center for Agribusiness Excellence to combat fraud through a Crop Insurance Program Compliance and Integrity Data Warehouse, which uses data from a variety of sources to identify fraudulent insurance claims.
The Crop Insurance Program Compliance and Integrity Data Warehouse resides on Tarleton’s Texas campus and runs on Teradata Database 14, which draws data from more than 170 data sources. Aerial imagery, crop data, soil surveys, high resolution satellite data and farm loan and disaster assistance information are just a few of the data sources that are cross-checked and scrubbed to identify questionable claims. Once an unusual pattern is identified, a letter is sent to the claimant stating that a USDA representative may visit to inspect the farm’s operation in the near future. The article states that once claimants were notified, claim rates dropped and became more consistent with other claims across the region. (Isn’t that ironic?)
Satellite data quality is improving the project’s success rate for identifying fraudulent claims. Today, a single pixel of data can cover a circle approximately 50 feet in diameter. The near-red and infrared scans can help to determine the health of plants based on the degree of greenness, which represents the amount of chlorophyll in plant leaves. These images can also distinguish between systematic crop growth and chaotic growth from weeds, which also help to determine if crops were viable on land that a farmer claimed he was not able to plant. (A picture is worth a thousand words.)
Congratulations to the USDA RMA for proving the value and power of data analytics and saving tax payers millions, if not billions of dollars. According to the article, the project has saved $975 million in unjustified claims payments since 2001. It is also estimated that the program has saved $2.5 billion in cost avoidance, while only costing the government $50.68 million. With proper insurance payments for legitimate reasons, farmers can get back to the business of sowing seeds and producing bumper crops.