Estimating Damages under the False Claims Act

The False Claims Act (FCA) imposes liability on any person or corporation that “knowingly presents or causes to be presented a false or fraudulent claim for payment” to the federal government. The FCA allows both fines and treble damages. Litigation under the FCA can have serious economic consequences. According to the Justice Department, FCA enforcement resulted in $4.9 billion in settlements and civil judgments in fiscal year 2012.

Cases involving alleged false claims in government procurement are an important category of FCA litigation; they accounted for $427 million in settlements and civil judgments in fiscal year 2012. Perhaps the largest FCA procurement settlement involved Oracle Corp., the software company. The General Services Administration (GSA) negotiated a procurement contract on behalf of several federal government agencies. In the negotiated contract with GSA, Oracle agreed to disclose discounts it made to commercial customers and provide these discounts to the government. The Department of Justice alleged that Oracle “overcharged the government by failing to disclose substantially lower prices offered to its commercial customers.” Without admitting liability, Oracle settled the case in 2011 for $199.5 million.

Estimating damages in cases that concern false claims in federal government procurement of products and services through GSA contracts typically involves calculating the amount of the overpayment. A number of difficulties may be encountered when performing that calculation. Specific modeling tools, however, can aid in that calculation, and in reaching settlements or assessing litigation risks.

GSA requires a most-favored customer (MFC) and a price reduction clause (PRC) in most of its major procurement contracts or schedules. Each GSA contract is required to have a section titled “Price Reductions.” The contracting officer and the supplier must agree on the customer (or category of customers) that will be identified as most-favored for purposes of the contract. The contract establishes a relationship between the prices and discounts charged the most-favored customers and the prices and discounts charged the government. If the price falls to the most-favored customer, it must also fall to the government.

FCA cases related to GSA contracts often involve claims that the supplier has increased its discounts to most-favored customers without telling the GSA, thus violating the PRC. That was the claim in the Oracle case. Estimating damages in such cases involves (1) calculating the total amount paid by the government for goods and services procured under the GSA contract, (2) estimating the total amount the government would have paid but for the false claim, and (3) taking the difference to calculate the overpayment. Performing these tasks might require overcoming a number of obstacles.

Calculating the amount paid by the government under a particular contract depends on invoice data. Generally, one would identify all invoices for sales made under the contract. The total amount paid on all these invoices less any credits would be the total dollar value of government purchases. It may be difficult to identify sales under a particular GSA contract or schedule if invoices do not reference GSA contracts or schedules. In those situations, customer name, billed entity, or delivered entity might be used to filter government sales. In using names to identify relevant sales, care must be taken to see whether names are represented in more than one way in the data, or whether any listed purchasers are in fact subcontractors purchasing for the government. For each purchase of a particular product, one would need list price, discount amount in percentage or dollars, final purchase price, quantity, and total payment. Final purchase price and total payment may be calculated from the other information, but it is important to check if the calculated prices and payments match the values present in the data.

Estimating the amount the government should have been billed requires several steps. First, detailed information on actual products purchased by the government has to be identified. A detailed description of the product and the identity of the purchasing agency are also needed. Often the same item may be sold with different discounts to different agencies purchasing under the same GSA contract. Even if the contract applies a uniform discount rate to all government sales, the detailed information would be useful if the price reduction adjustments vary by product or over time.

Second, detailed information on individual sales to most-favored customers is required to identify the relevant discounts and prices to apply to government sales. For the relevant time period of the GSA contract, the corresponding comparable sales to most-favored customers are reviewed to find the lowest discounted price or highest discount rate for each of the products purchased by the government.

Third, once pricing information and product information are collected, the exact terms of the contract are used to identify price reductions for estimating damages. For example, if a particular set of products were sold to most-favored customers below the price offered to the government, then every government purchase of those products would be identified and the amount that would have been paid at the more favorable price would be calculated. In some cases, products have a different PRC discount for different time periods, which complicates the analysis.

In cases where the GSA contract involves a large number of products and services with a complex discount structure based on the historical discounts to MFCs, damage estimation may be quite difficult. Problems with invoice data often complicate the estimation process. For example, free samples in exchange for large orders, bundled products, non-meaningful product differentiation, free shipping, and discounts not tied to a particular product may make it difficult to calculate the appropriate discount that should have applied to the government purchases. Problems such as multiple product numbers or multiple names for the same product may make it difficult to determine the products that a particular discount should apply to.

Ultimately, the data would be used to calculate a properly discounted price under the terms of the PRC. If that price is lower than the actual price paid by government, then the difference in price is multiplied by the quantity purchased at that price, and the result is included in the overpayment.

This type of damage calculation may be used to facilitate settlement. For example, negotiators may want to know the estimated damages under different assumptions about the PRC discounts. Damage models that estimate the discounts and corresponding damages under various assumptions can be created and the results can indicate the possible consequences of litigation.

Richard Shin is a Senior Vice President at Economists Inc., and has an extensive background in analyzing competitive effects of mergers in regulatory proceedings and estimating economic damages in complex litigation. He was retained as a testifying expert in the qui tam action against Oracle to provide damage estimates.