Manjeet Rege headshot.
Liam James Doyle/University of St. Thomas

In the News: Manjeet Rege on the Computer Programming Language Used by the Social Security Administration

Manjeet Rege, data science and software engineering professor at the University of St. Thomas School of Engineering, recently spoke with The Washington Post for a story about the claims of Social Security fraud by the Department of Government Efficiency (DOGE).

From the story:

The Social Security Administration maintains its databases using COBOL, a nearly 70-year-old computer programming language that has long served as the backbone for data maintenance in many government and finance-related industries, said Manjeet Rege, chair of the software engineering and data science department at the University of St. Thomas. The Social Security Administration maintained more than 60 million lines of COBOL in 2016, according to a report that year from the agency’s inspector general.

The challenge with COBOL is that it doesn’t have a standardized way to store and work with dates – unlike most modern programming languages. Instead, many dates in COBOL must be coded to a reference number, often using an international standard, Rege explained.

The most common reference date number is May 20, 1875, the day this standard, known as ISO 8601, was established. That means if someone applies for Social Security without a birth date, they may be recorded as 150 years old in the database depending on how a programmer troubleshoots the issue.

Programmers can overcome these limitations in COBOL’s functionality, but they may also opt to use 1875 as a placeholder for unknown dates if it aligns with the database’s purpose, Rege said. For example, when the federal government began issuing regular monthly Social Security payments in January 1940, recipients needed to be at least 65 to qualify, meaning they needed to be born in 1875 or before.

Those who crafted the Social Security database may have therefore set 1875 as the default birth year for anyone who lacked that information at the time.