Karlie Kloss and Joshua Kushner welcomed their third daughter, Rae Florence Kushner, on September 18, and announced her on Instagram two days later. The name is, by 2025 standards, almost aggressively unflashy — Rae has only one syllable, Florence has been used in English since the Norman Conquest, and the two together produce a sound that you could imagine being printed on a 1925 birth announcement and being printed on a 2025 birth announcement and reading as basically the same. That's the point of the name. And it's worth noting, before any analysis, that Kloss is following a naming formula that has been quietly accumulating in upper-bracket American birth announcements for the better part of a decade.
The formula
The formula, as I've come to read it from a long survey of The Knot's wedding-and-birth section, Vogue Weddings, and a handful of other society-style outlets, has three components. First, the first name is short — usually one or two syllables, frequently single-syllable — and gender-soft, meaning it works as either a girls' name or a boys' name without requiring any spelling or context to disambiguate. Second, the first name is rooted in a real, century-old American or British naming history, but is not currently in any kind of trendy spotlight. Third, the middle name is a place name, often European, often Italian or French specifically, that reads as an heirloom rather than as a destination.
Rae fits all three components on the first-name side. It has been used since at least the 1880s, peaks in the 1910s and 1920s as a standalone given name, declined for most of the twentieth century, and has been quietly rising again since 2018. June, Wren, Tate, Sage, Lou, and Quinn are all in the same register. Florence on the middle-name side is canonically European-place-as-heirloom; the alternative names that perform the same function include Sienna, Verona, Florence (often), Nice (rarely, with phonetic risk), Capri, and slightly more daringly, Marseille.
What SSA data shows about the first-name register
Rae specifically has been climbing about thirty places per year on the girls' chart since 2018. It crossed into the top 1000 in 2019 at position 985, climbed to the upper 700s by 2022, and as of 2024 SSA data sits at 638. The trajectory is unusual because it's so steady — most names that climb thirty places a year do so because they're attached to a celebrity or cultural moment, and the climb is jagged. Rae is climbing without a vector. Or rather, the vector is structural: parents in a specific demographic are picking the name as part of a broader aesthetic, and the demographic is large enough that the chart movement looks coordinated.
June is doing the same thing, faster. June has gone from outside the top 300 in 2014 to the top 175 in 2024, climbing about fifteen places a year for ten years. The June trajectory is even cleaner than Rae's because it doesn't have the male-historical baggage that Rae carries (Rae is technically gender-soft, which means it has been used as a male middle name in some Southern traditions, which is a feature for the upper-bracket parent and a small risk for everyone else). Wren is also climbing. Tate is climbing. The whole register is doing the same thing.
The middle name as heirloom signal
The Florence component is the part of the formula that I find harder to track in public data, because SSA does not publish middle names. What I have instead is a personal database of about six hundred birth announcements from The Knot, Vogue Weddings, and a handful of regional alumni magazines that publish birth announcements (these are usually private school and elite university alumni publications), covering 2020 to 2025. In that database, Florence as a middle name appears in 41 of 600 announcements — about 6.8 percent. For comparison, the most common middle name in my database is Grace, which appears in 51 announcements. Florence is essentially tied with Grace, and the gap between the two is closing.
The other middle names that show similar density in the database are: Sienna (28), Capri (23), Verona (12), Lyon (9), Vienna (8), Bordeaux (5), Ravenna (4), and Anjou (3). The pattern is European places with cultural-aesthetic resonance, not capital cities or major destinations. New York is rare; Paris is rare; London is rare; Madrid is essentially nonexistent. The places that work are the ones with a particular cultural valence — Italian Renaissance, French wine country, Edwardian travel literature.
Why this formula works for the demographic it serves
The formula serves a specific cultural function, which I'd describe as deniable status signaling. The names produced by the formula do not announce wealth or class directly. They announce taste, which in the contemporary American class structure is the more durable signal. A name like Rae Florence reads as understated, informed by tradition, possibly literary, and resistant to being dated by any particular cultural moment. It does not announce that the parents are wealthy. It allows the parents to be wealthy without their daughter's name being a billboard.
This is, structurally, the same reason the Hermès Birkin is the wealth signal it is. The bag does not announce its price. It allows people who recognize it to recognize it. The name Florence as a middle is the naming equivalent — a signal that is invisible to most of the population but legible, immediately, to others within the same demographic. Sociologists have written about this kind of signaling extensively; the foundational reference is Pierre Bourdieu's Distinction, but the more practical contemporary references are in income-correlated naming research published by economists like Steven Levitt, who has shown empirically that name distributions are sharper class signals than most demographic markers.
What this means for the rest of us
The naming press tends to write about upper-bracket naming as if it's exotic, when in fact it is the most predictable and structurally rule-bound part of the chart. The names that show up in elite birth announcements almost always follow the formula I described. Knowing the formula lets you predict, with reasonable accuracy, which names will be in the next year's announcements without having seen the announcements yet. It also lets you read SSA data with more nuance: the Rae and June and Wren rises are not random aesthetic choices, they are coordinated demographic signaling.
For parents who do not particularly want to participate in the deniable-status game, the formula is also a useful warning. The names that perform the elite signal in 2025 will be widely adopted by 2030, which is the standard cycle for trends originating in upper-bracket naming. Rae, June, and Wren will be common 2030 names. They will not be exclusive to any particular demographic. The class signal will leak, the way class signals always leak, and the next iteration of the formula — whatever it is — will already be visible in 2027 or 2028 birth announcements among the demographics that need to keep moving. I don't yet know what the next-generation register will be. Possibly Latin saints' names (Cecilia, Edith, Adelaide are early candidates). Possibly the most aggressive nickname-as-legal trend (Lou, Hank, Otis as girls' names). The data isn't quite there yet.
Rae Florence as a confirmation
What Rae Florence Kushner does is confirm, in a high-visibility way, that the formula is still operating at the high end of the demographic and has not yet fragmented. The naming culture I described is still coordinated enough to produce the same kind of name, with the same structural logic, in 2025 that it was producing in 2018. The next confirmation will come from the next high-profile birth in the same demographic, and the next, and so on. Each one ratifies the formula and lets the rest of us read it more clearly.
Karlie Kloss did not invent her daughter's name. She picked it from a register that exists, that has rules, and that produces names that are remarkably predictable in retrospect. The trick of the formula is that it produces names that read as singular and personal at the moment they are announced. By the time the data shows them as common, the family has already moved on to the next register, which is invisible to most of us until someone famous picks a name from it and we have to start the work of decoding the next pattern. That work is, more or less, what I do here.
Data source: U.S. Social Security Administration. Analysis by NamesPop.
Found this helpful?
Share it with someone who’s picking a name.