A tale of two tweets

 This is Joseph.

There is a new claim that the US is filled with vampires by Elon Musk, who is concerned about a database query of social security numbers with the death field as being false showing Americans who are hundreds of years old:

And then this new post as to what happens if use a second field (in the same database) to refine the query and just find the people in the previous list who actually get benefits:

There was immediate skepticism about the magnitude of the claims here. It seems like the sort of thing that was unlikely to have been missed for decades. Even Megan McArdle noted that these older persons wouldn't have made any contributions and so wouldn't get benefits when Musk used the same chart in a second post:

So what can possibly be going on here? Well, it seems like the explanation is people using false social security numbers to pay into the fund but that these payments don't create any eligibility for benefits. So it is true that we've located fraudulent payments that reduce the cost of funding social security. Fortunately, this information was buried in a formal report from 2015 in which the SSA decided it wasn't worth the cost to update old databases that no expert would bungle. I suspect that they feel foolish now -- focusing on efficiency rather than clarity -- but getting a correct death date for long dead Americans seems like an expensive journey into old government archives (many of these deaths likely predate electronic recording as Deva Hazarika noted). 

Ironically, such a project might well have been a target of DOGE had it been approved in 2023:

But the real lesson here is that database work is complicated (just ask Mark about his work in really exciting databases) and the key to doing anything is deep, subject matter knowledge. So why don't we have subject matter experts engaged? Until quite recently, the idea was that civil servants should be professionals and provide this expertise. 

I am worried that the answer Josh Marshall provides is true as to the reasons for this approach and, hopefully, he will expand on these thoughts with further investigative journalism -- hopefully disproving them.

So kids, play safe with databases!