FDIC Study of Bank Overdraft Programs
In 2006, the Federal Deposit Insurance Corporation (FDIC) initiated a two-part study to gather empirical data on the types, characteristics, and use of overdraft programs operated by FDIC-supervised banks. The study was undertaken in response to the recent rapid growth in the use of automated overdraft programs, defined as programs in which the bank honors a customer's overdraft obligations using standardized procedures to determine whether the nonsufficient fund (NSF) transaction qualifies for overdraft coverage. Little empirical data have been available on these programs, their features, their managing practices, the fees imposed, and consumer usage patterns.
Data and information for the FDIC's study were gathered through a survey of a sample of banks that represented 1,171 FDIC-supervised institutions, and a separate data request of customer account and transaction level data from a smaller set of 39 institutions.
The two-part study was designed to obtain the following types of information related to overdraft programs:
- characteristics, features, and fees of overdraft programs;
- transaction-processing policies;
- marketing and disclosure practices;
- internal controls and monitoring practices;
- the role of vendors and third parties in overdraft program implementation;
- NSF-related fee income and growth.
The customer account and transaction-level data collection was designed to gather information on the provision of overdraft services on customer accounts, the occurrence of NSF activity covered under automated overdraft programs, and the characteristics of customer accounts that tend to incur the highest volume of overdraft fees.
The FDIC believes that objective information on these programs will help policymakers make better informed policy decisions and will help the public better understand the features and costs related to automated overdraft programs. The study results also will help the banking industry develop more effective overdraft programs to better serve consumers.