Tian was looking up some data on names for the next step of our project on using names to estimate individual network sizes, and she found the number of newborns with each name who were given Social Security numbers in each decade. In recent years, she saw the expected 51% boys, but in the early 1900s, many more newborn girls than boys got Social Security numbers:
Tian writes, “I am looking forward to some freaknomics-type answer from your blog visitors.”
Um, I'm pretty surprised that they handed out any SSNs at all to newborns in the early 1900's.
The Social Security program was created in 1935. I doubt Social Security numbers pre-dated the Social Security program, so I don't think any of the pre-1935 data include newborns. I'm guessing that all of the data from before 1935, and some of the data through perhaps 1950, are for people applying for numbers retroactively (i.e. post-birth). If that's right, then "World War I" is the obvious answer for the shortage of males who were born between 1880 and 1901 or so. And if males born in, say, 1920 were not issued numbers until considerably after 1935, some of them would have been lost in WWII. It wouldn't be until all babies were registered at birth that we would expect the male:female ratio to be higher than 0.5. Maybe that didn't happen until after WWII.
All of this is speculation, except the part about Social Security starting in 1935.
Good points. Tian said specifically that the data were just for newborns–but maybe that's for the recent data and not for the data before 1950 or whatever. And, yeah, I should have realized about no Social Security before 1935. Or maybe the data were for some predecessor program…
I think I got my Social Security number when I was 6, so I don't know if that means that I'm not in this database.
Oops. Not "newborns." The data were on the births reported on SSN applications. See below.
"All names are from Social Security card applications for births that occurred in the United States after 1879. Names are restricted to cases where the year of birth, sex, State of birth (50 States and District of Columbia) are on record, and where the given name is at least 2 characters long. Many people born before 1937 never applied for a Social Security card, so their names are not included in our data. For others who did apply, our records may not show the place of birth, and again their names are not included in our data."
I am still not fully convinced that WWI alone could have led to such a big gap in numbers.
1880s: 1,177,345 male births and 1,399,723 female births.
1890s: 1,231,611 male births and 2,353,483 female births.
1900s: 1,467,529 male births and 3,104,799 female births.
1910s: 6,948,837 male births and 8,500,041 female births.
1920s: 11,363,452 male births and 12,395,538 female births.
Yeah, I didn't think it was WWI. When SS was first enacted, enrollment was optional. Although SS benefits were small at the time, there were survivor benefits; plus, although regular benefits were roughly pegged to contributions, in general women had fewer market wage opportunities than men so SS was the only game in town. More women signed up sooner.
Jason Ruspini writes,
There is an error preventing me from
posting on the blog, but survivors benefits (life
insurance) would be my guess.
Cheers, Jason Ruspini
1. What SSA file exactly is this from?
2. Is the first point significantly higher than the next two; i.e., what is the sample size for 1880?
3. Could it be that the file only holds persons who have survived to, e.g., 2000?
4. Is there a chance that a lot of men back then used an initial for a first name, while women didn't? ("Given name at least 2 characters long").
SS was also not strictly a retirement (really elederly anti-poverty) program when it first started. It also included disability and other elements. It is easy to imagine (finding out if actually true is up to Tian) that there were benefits that women younger than retirement age were eligible for.
Is it possible that people applying for SS, who were born before 1935 or so, were old enough that the old-age gender gap played a role? That is, as women live longer than men, the greater incidence of women applying for social security is because there were more older women than older men.
Do your data include railroad IDs? For a long time anyone who worked for a railroad (almost all men, and it was a large employer at the time) were given railroad IDs. When SS was enacted, they didn't have to apply for SSNs and just used their RR IDs. unlikely to explain your entire disparity, but it could account for some of it.
RR ID's look different than SSNs. I think they start with letters rather than numbers, so you can check quickly to see if this might be the case.
Routine application for SSNs for newborns is actually quite recent, dating from the year that the IRS began to require SSNs of dependents on income tax returns. Prior to this, the only reason a very young person would have needed a SSN was if he or she had income (e.g., from a trust or other such vehicle). My youngest chile is almost 30, and (careful! Memory is hazy) I don't believe that his SSN was required on our income tax returns at that time; we did apply for an SSN for him when he was young because we established a trust to pay for his college education.
I'm not at home right now, or I would just look at some of those old returns and note when we started listing our childrens' SSNs on them.
Girl names and boy names.
I posit a Boy named Sue effect: Boys with girl's names.