Ghosts in the Database: How Legacy Code Made Social Security Records Seem Immortal

150-Year-Olds on Social Security? The Real Story Behind the Data Anomalies

Elon Musk’s Department of Government Efficiency (DOGE) audit results of the Social Security Administration (SSA) have been all over the news recently. The findings appeared to show payments being made to individuals supposedly 150 years old or older, leading to widespread speculation about errors or fraud. However, a closer look at the data suggests that the real culprit isn’t fraud, but outdated government technology and a misunderstanding of how legacy systems handle data.*

The Problem With Old-School Code

The Social Security Administration (SSA) still relies on COBOL, a programming language from 1959. It’s old but reliable, and still runs a ton of government systems. The catch? It wasn’t built to handle dates the way modern systems do. Back in the day, programmers had to get creative when data was missing. One common trick was using a placeholder date— often May 20, 1875, which corresponds to the date of the Convention du Mètre, an international standards conference that helped establish universal measurement systems. Fast forward to 2025, and that placeholder makes it look like some people are 150 years old when they’re not. That’s what DOGE’s auditors saw, and they understandably interpreted it as a sign of potential issues.

Musk also shared a chart illustrating Social Security payments being made to individuals in exceptionally high age brackets. While May 20, 1875, is one known placeholder date, the presence of payments to individuals supposedly aged 240 or older suggests that additional placeholder dates or default values may have been used. These could have been entered when birthdates were unavailable or to fulfill system constraints in the SSA’s legacy databases. But those numbers weren’t proof of fraud—they were just artifacts of outdated data entry. The SSA actually has policies in place to stop payments to people listed as 115 or older unless they can prove they’re still alive.

Why It’s Important to Understand Legacy Tech

The SSA’s systems are complex, and you need experience with older programming languages like COBOL to understand them. The DOGE team, primarily familiar with modern systems, may not have been accustomed to the unique quirks of legacy data storage. Without prior experience with these systems, they likely saw anomalies that seemed concerning at first glance. This kind of misunderstanding happens a lot when old and new tech collide.

What This Means for Business and Tech Today

I’ve spent years working with legacy systems and finding ways to merge older data formats with modern tools. Even as technology has advanced, we’ve occasionally encountered relics of much older systems that required specialized knowledge to interpret.

Twenty years ago, while not a standard practice, we would still occasionally come across survey data coded in column-binary format—a method dating back decades that required an understanding of historical data structures to process correctly*. Even a decade ago, we were still dealing with guest surveys filled out on Scantron bubble sheets (don't forget your #2 pencil!), another holdover from older data collection methods. These outdated but functional systems worked, but only if you understood their quirks.

The origins of column-binary data formatting trace back to punch cards, which were widely used in computing from the early 20th century through the 1980s. These cards typically contained 80 columns, each representing a single character or numeric value.

Fun fact: Even though punch cards are long obsolete, their influence remains—most notably in the 80-character line limit still found in many programming environments, terminal displays, and coding standards.

At the End of the Day, It’s All Ones and Zeros

Computers have changed a lot over the years, but one thing hasn’t: everything still runs on binary (ones and zeros). Sure, we have better programming languages and slicker interfaces, but at the core, it’s the same principle. That’s why it’s so important to understand legacy systems—whether you’re auditing government data or modernizing business operations.

The Takeaway

The Social Security mix-up is a perfect example of what happens when we don’t bridge the gap between old and new tech. Encouraging collaboration between experienced professionals and those familiar with modern technologies can help ensure a clearer understanding of legacy systems and prevent misinterpretations. Understanding history—especially in tech—can help avoid costly mistakes and keep things running smoothly in both government and business.

‍

*Sources:

FactCheck.org – Clarifies that data shared by Elon Musk was misinterpreted, leading to exaggerated claims about improper payments.
https://www.factcheck.org/2025/02/trump-musk-exaggerate-scale-of-improper-social-security-payments-to-the-dead

Fox Business – Features an interview with the Acting Social Security Chief discussing DOGE's access to the agency and addressing reports spurred by Elon Musk's claims.
https://www.foxbusiness.com/politics/acting-social-security-chief-opens-up-doge-claims-dead-people-receiving-benefits

Associated Press – Reports that the Trump administration's claims about tens of millions of centenarians receiving benefits are false.
https://apnews.com/article/social-security-payments-deceased-false-claims-doge-ed2885f5769f368853ac3615b4852cf7

AFP Fact Check – States that Musk's assertions lack evidence, noting that his figures likely represent past-issued Social Security numbers rather than current beneficiaries.
https://factcheck.afp.com/doc.afp.com.36Y83WL

Time Magazine – Highlights that while some improper payments have occurred, the scale is much smaller than claimed by Musk and Trump.
https://time.com/7258453/trump-musk-social-security-dead-fraud-fact-check

‍