Most people carry one of two beliefs here, and both are out of date. Either men still hold more degrees, or women pulling ahead is a recent shift that is still settling out.
The OECD started measuring this in 1981, and in that first year young women in Sweden, Denmark, and France were already ahead of young men. Not closing the gap, but past it. Whatever year those three countries actually crossed over, it happened before the dataset even opens, which makes this change older than the data we use to look at it.
In the countries where men were still ahead in the early 1980s, you can watch the crossover happen, and it lands around 1990. Canada, the United States, Norway and Spain all cross within a year or two of each other. For most of the rich world, in other words, the flip was finished before today's average undergraduate was born.
And then it kept going.
The OECD-wide average only starts in 1998, and women are ahead there too. From that point the two lines pull steadily apart, until by 2024 it is 55% of young women with a degree against 42% of young men. The lead women hold today is wider than any lead men ever had.
Two ways to count, and they disagree
Here is the part that trips up almost everyone, including the people writing the headlines.
"Women are more educated" is really two different questions wearing the same coat. The first: what share of young women hold a degree, compared with what share of young men? That is every number here, 55% against 42%, each sex measured against its own population. The second: what share of all graduates are women? That is the "60% of graduates are women" line you see quoted everywhere.
Those two can point in different directions. The graduate-share number reflects two things at once, how many young men and women a country has, and who among them studies. Where a cohort has more men than women, women can hold degrees at a clearly higher rate and still sit near half of all graduates, simply because there are more men in the pool to begin with.
We used the first one, the attainment rate, because it answers "are young women more educated than young men" without the population balance muddying the result. The habit worth keeping is to ask what any such stat is dividing by. The denominator tends to be the whole story.
So where is the gap widest? Not where most people would guess.
Every OECD country with 2024 data has a higher share of young women than young men holding a degree, without exception. And the widest gaps are not in the countries with the loudest reputations for gender progress. They are in Slovenia, Latvia, Israel, Norway and the Baltics, where young women lead by roughly 20 points or more.
Slovenia sits at the top, and its last two decades show the shape of it clearly.
Both sexes became far more educated over these twenty years. Women simply did it faster, and the space between the two lines kept opening until it reached 24 points, the widest in the OECD.
This is about as far as the data can take you. It shows the gap and how long it has been growing, but not why women pulled ahead, or why the lead runs widest in Ljubljana and Riga rather than Stockholm. Those questions stay open.
What the numbers do settle is the timing. This is not a new trend catching on, it has been the ordinary state of things since before most of the people it describes were born.
There is one thing the overall figure cannot show you, though. "Women are more educated" is an average, and averages tend to keep their disagreements quiet. It says nothing about which subjects women run away with and which ones they do not, and that turns out to be the more surprising question. It is where this goes next.
Data: OECD, share of 25-34 year-olds with a tertiary degree, by sex, from DF_LSO_NEAC_DISTR_EA, via the SDMX API. This is the attainment rate: the share of each sex that holds a degree, measured within each sex (unit: percentage of population in the same sex and age). It is not women's share of all graduates, a different measure that also depends on the male-female population balance. The comparable series begins in 1981 for a few countries and 1998 for the OECD aggregate, so the earliest crossovers cannot be dated precisely, they predate the data. A methodology break in some national series around 1991-92 makes exact single-year crossovers approximate. The OECD aggregate is not published for 2011-2019 (shown dotted). US and Iceland lack 2024 figures. Pulled with oecd-mcp.