The Expectation Gap: What Happens When Choosing a Partner Becomes a Strategy

Here is an observation that gets made constantly, usually badly.

As women have become more educated, better paid and more independent, the pool of men who meet the criteria they were raised to apply has shrunk. The gap between what a woman expects and what the market in front of her contains has widened. Partnership has stopped feeling like an instinct and started feeling like a decision with consequences — financial, social, reputational. And loneliness has increased inside that gap.

That description is substantially accurate, and it is also one short step away from the most tedious argument on the internet, which is that women’s standards are too high and they should lower them.

We are going to make the accurate version and refuse that one, and the refusal is not politeness. It is that the cheap version gets the arithmetic wrong. The gap was not produced by women deciding to want more. It was produced by one variable moving enormously and another variable not moving at all, and nobody chose either. What follows is an attempt to say which variables, what that does to a person’s body, and what gets buried underneath it.

Across the OECD, women now outnumber men among tertiary graduates. That is not a marginal shift or a projection; it is the current state of the educated population in most wealthy countries, and it happened inside two generations.

At the same time, and this is the half that gets discussed less, the economic position of men without higher education deteriorated in a great many of those countries. Manufacturing employment fell, stable long-tenure jobs became less common, and the wage premium attached to credentials rose. The research literature on this in the United States is extensive, and one of its oldest findings — going back to work on what was then called the pool of marriageable men — is that when male economic stability falls in a community, marriage rates fall with it.

So two things happened at once. Women’s educational and earning positions rose sharply. Men’s positions polarised, with the top rising and the middle and bottom eroding.

Purely as arithmetic, that narrows the set of men who are higher-earning and better-educated than any given woman. It narrows it a great deal. And it narrows it for reasons that have nothing to do with what anybody wants.

Now the other half, which is the part the cheap argument never reaches.

The expectation that a woman should partner with a man who is at least her equal in education and earnings — and preferably above it — is not something women invented recently. It is one of the most durable norms in modern societies, and it was taught to everybody.

It was taught to women as prudence: you will bear the greater share of any career interruption, so secure someone who can carry the household. It was taught to men as the definition of adequacy: a man who earns less than his wife has failed at something, and a great deal of popular culture continues to say so. It was taught to families as the standard by which a match is judged.

In much of Western Europe and North America that norm has genuinely weakened; couples where the woman out-earns the man are now common and increasingly unremarkable. In Japan it has weakened considerably less. Research on educational assortative mating finds the expectation of marrying an equal or higher-status man persisting more strongly here than in many comparable countries, and it persists on both sides — it is not only that women seek it, it is that men frequently do not want the alternative.

That is the structural sentence of this essay. One variable moved enormously. The other barely moved. The gap between them is not anybody’s personality. It is subtraction.

Japan makes the mechanism unusually visible because the official surveys ask the question directly.

The national fertility survey has consistently found that the overwhelming majority of unmarried Japanese people say they intend to marry eventually. This is not a population that has rejected partnership. And when unmarried people are asked why they are not married, the most-cited reason, year after year, is that they have not met a suitable partner.

Not “I do not want to.” Not “I prefer my freedom.” The stated obstacle is supply.

Set that beside the marriage-arranging industry that has grown up here, where income bands are stated openly as search criteria, and the picture is of a population that wants to partner, is actively looking, and is filtering against a criterion that a shrinking share of the available population meets.

And notice the second-order cruelty in it. A woman who has done exactly what she was told — studied, qualified, become financially independent — has, by succeeding, reduced the number of men who satisfy a norm she was also told to apply. Her achievement and her difficulty have the same cause. Nobody warned her, because nobody designed this.

It would be dishonest to describe an escalation of expectations as a female phenomenon, so let us be precise about the other direction.

Male expectations escalated too, along a different axis. The standard of appearance a woman is assumed to meet has risen steadily, and the tools for measuring it have multiplied — filters, curated images, an infinite comparison set available at any hour. The Korean essay in this chamber described a movement organised specifically around the hours and money that standard extracts. Japan has the same obligation in a softer, more deniable form.

There is also an age asymmetry that is rarely stated plainly: a woman’s perceived value in the partnering market is widely assumed to decline with age in a way a man’s is not, which compresses her decision window while leaving his open. That assumption does real damage whether or not it is defensible, because people act on it.

So what has actually happened is not that women became demanding. It is that both sides began evaluating each other against rising, increasingly measurable, increasingly public standards — economic on one side, aesthetic on the other — and both sides feel the same shortage from their own vantage point. The man who believes there are no women who will accept him and the woman who believes there are no men worth accepting are describing the same machine from opposite ends of it.

There is a reason the criteria hardened, and it is not that people became shallow.

Housing costs relative to income rose in most large cities. Stable employment became less reliable. Care for both children and ageing parents remained largely privatised and largely female. Pensions became less certain. In that environment, choosing a partner stopped being only a question about affection and became also a question about whether you will be able to afford a life.

That is a rational response to conditions, not a failure of romance. A woman evaluating a potential partner’s income is doing risk assessment in a society that has told her, accurately, that she will absorb the financial cost of any child, any illness, any parent who needs care, and any interruption to her own career.

But something specific happens when partnership carries that much weight. It has to be justified. A relationship becomes an outcome you can be judged on — by family, by colleagues, by yourself at three in the morning — and the question shifts from does this feel right to can I defend this choice.

Desire does not compete well with justification. It is not an argument. It cannot produce reasons. Asked to explain itself in a conversation about salary bands and life plans, it simply goes quiet, and the person experiencing it concludes that she does not feel very much, when the truth is closer to the opposite: she feels something that has no standing in the room where the decision is being made.

We have used the word market several times and it is worth stopping on, because the metaphor is doing something to the people inside it.

A market implies a set of goods with comparable attributes, a price, a rational buyer, and a better option one search away. Every dating platform is built on that model, and it is a superb model for selling things.

What it does to a person is convert her into a listing. Attributes, photographs, a summary, a stated range. It also converts everybody she meets into a listing, which means each encounter is not an experience but an assessment, and an assessment has a comparison set attached — the ones already dismissed, the ones not yet met.

Under those conditions, the failure of a meeting to produce a spark gets recorded as a data point about the other person. It is at least as likely to be a data point about the conditions. A person performing evaluation is not in a state where much can develop, and two people evaluating each other across a table are doing something that closely resembles a job interview conducted by both parties simultaneously.

People do not fall in love in interviews. This is not a mystery about modern romance. It is a description of what evaluating does.

Here we want to be careful, because this is where a plausible-sounding essay usually starts inventing neuroscience.

What is reasonably well established is this. Sexual response in humans is not a single drive that is either on or off; it involves both excitatory and inhibitory processes, and inhibition is a substantial part of the picture. The dual control model, developed in sexuality research, describes exactly this: alongside whatever is accelerating, there is a braking system that responds to threat, to distraction, to concern about performance, and to the possibility of being judged.

Related work on desire has made a second point that matters enormously here: for many people, especially many women, desire is more often responsive than spontaneous. It arrives in a context rather than preceding one, which means that the conditions surrounding an encounter are not incidental to arousal — they are a substantial part of what produces it.

Put those two together and apply them to a woman who has spent her day managing her appearance, her tone and her competence, and her evening assessing whether a stranger meets criteria that will determine her financial security for thirty years. The brakes are engaged. Not because of anything wrong with her, and not because she has lost touch with her body. Because a person conducting an evaluation, under observation, with consequences attached, is in almost exactly the condition the inhibitory system exists to respond to.

And then she concludes that she does not want anything much. That conclusion is the injury. The want is not gone; the conditions for it to appear have simply not been present for a very long time.

Now the part this house names directly, and it is the reason this essay exists.

Underneath the strategy, in almost everyone, there is something much simpler and much harder to say. Skinship — the word both Japanese and Korean use for physical closeness that is not sexual performance. Being held. Romance in the unfashionable sense: being taken somewhere, being paid attention to, being wanted specifically rather than approved of generally. Pleasure that is not evidence of anything.

These are not embarrassing wants. They are close to universal, and there is good physiological reason to take them seriously: affective touch has its own dedicated pathways in the human nervous system, distinct from the pathways that register pressure and texture, and it is implicated in how safe a person feels around another person. Touch is not a luxury layered on top of relating. It is part of the equipment.

So why are they so hard to say out loud?

Because every one of them sounds like weakness in the register a woman is required to speak in. She has been fluent for twenty years in competence, independence, self-sufficiency and not needing anything. That fluency was necessary; it is what got her the qualification, the salary and the standing. And it has no grammar for I would like to be held for an hour and not have it mean anything about my future.

So the want stays underneath, and the surface keeps performing. And because the surface is extremely convincing — she is competent, she is fine, she is busy — nobody, including her, hears the other thing.

This is a cross-border essay because the same arithmetic produces a different vocabulary in each society, which is the clearest evidence that it is arithmetic rather than culture.

In South Korea it produced an explicit politics: a named refusal, a backlash, and the lowest fertility rate in the world. In China it produced a state-adjacent shaming vocabulary, with educated unmarried women over a certain age labelled with a term meaning left over — a label aimed precisely at the women whose achievement had shrunk their own pool. In the United States it produced an enormous discourse about dating app fatigue and a loneliness advisory from the national health authority. In Japan it has produced a marriage-arranging industry, a stable official statistic about not having met a suitable partner, and very little public argument.

Four idioms. One machine. And note what the Chinese case in particular demonstrates: when a society finds itself with a large number of accomplished unpartnered women, the first instinct of the institutions is frequently to locate the defect in the women. That is worth remembering the next time the explanation on offer is that standards are too high.

The gap is real. We think the standard explanations of it are both wrong, in opposite directions.

The conservative explanation is that women overreached and should recalibrate. It fails because it asks one side to absorb a structural change by wanting less, and because the norm it quietly defends — that a woman should partner upward — is the very thing generating the shortage. If you genuinely want the gap to close, the norm to abandon is not women’s ambition. It is the idea that a man earning less than his partner is a diminished man. That norm is held as firmly by men as by women, and it is the load-bearing wall.

The progressive explanation is that the men simply need to improve. It fails for a similar reason: it treats a labour-market polarisation and a set of inherited scripts as a collective moral failing, which is both unfair and useless as a plan.

What we think is actually true is smaller and more usable. Two questions got fused that were never the same question, and separating them is most of the work.

The first question is: who would be a good person to build a life with. That is a strategic question. It legitimately involves money, values, stability and how somebody behaves when things are hard. Nobody should apologise for taking it seriously, and a woman who evaluates carefully is not being cynical; she is being accurate about what a society has arranged to cost her.

The second question is: what do my body and my heart actually need, and when did either of them last get any of it. That question has a different answer, on a different timescale, and — this is the part that goes missing — it does not have to wait for the first question to be resolved.

Almost everyone has been behaving as though the second question is a reward for solving the first. Hold out, choose correctly, and then be touched, wanted, unhurried, attended to. But the first question can take a decade, or may not resolve, and the body does not have a decade of patience. Years of waiting for permission is how a person arrives at forty having been extremely competent and barely touched.

You do not owe tradition obedience. You do not owe modernity performance. You are allowed to discover what actually fits your life.

Which brings us to something we have a direct interest in, so we will state the interest and then be as honest as we can.

Some women, having noticed the gap between their surface and what is underneath it, go looking for a bounded, explicit, paid setting in which the underneath part is the point. That describes what Moonlight is. We would rather explain the mechanism plainly than pretend the phenomenon does not exist.

The mechanism is not mysterious and it is not romantic. In an ordinary encounter, a woman is simultaneously being assessed and assessing, and the evaluation is exactly the condition under which the inhibitory side of her response is most active. In a bounded arrangement, several of those variables are removed by design. Nobody is auditioning. There is no future being decided. The terms were agreed in advance, in writing, by her. Nothing about the evening is evidence about her worth, her age, her income or her prospects, because none of those things are what the evening is about.

What that removes is not intimacy. It is the assessment, and the assessment was the thing engaging the brakes.

Now the limits, which we hold to firmly and will not blur.

This is not therapy and it is not treatment. It does not treat trauma, sexual dysfunction or depression, and anyone looking for treatment should see a clinician — that is a better outcome and we will say so directly rather than take the booking.

It does not fix loneliness. Loneliness is about being known over time by people who remain, and an evening cannot manufacture that. What an evening can do is interrupt a specific deprivation — of touch, of attention, of being wanted without being evaluated — and interrupting a deprivation is a real thing that is not the same as curing a condition.

It is not a relationship and it is not a substitute for one. Anybody who tells you a paid arrangement is the same as being loved is selling something, and we are in a position to sell something, which is exactly why we are saying it.

And it is not the answer for most people. Most people reading this essay will do better with a difficult conversation with somebody they already know. We would rather say that than not.

What it is: a bounded setting in which saying what you want is the activity rather than the risk. For a person who has spent twenty years being competent and has lost the habit of asking for anything, that is not a small thing, and it is also not a cure. Both of those are true at once.

This essay can be misread in two directions and we would rather name them than hope.

The first misreading is the manosphere one: that we have conceded that women are too picky and the whole problem is female hypergamy. We have not. The escalation is symmetrical, the norm being enforced is held by both sexes, the shortage is generated by a labour-market change nobody chose, and the single norm most responsible for the arithmetic is the belief that a man who earns less than his partner is diminished — which is a male-directed norm before it is a female-directed one.

The second misreading is the resigned one: that a woman should stop being strategic and simply follow her feelings. Also no. The strategic question is legitimate and the risks she is assessing are real, because the costs really will fall on her. Nothing here suggests she should evaluate less carefully. It suggests that the evaluation and the wanting are different activities that cannot be done in the same hour, and that she has been trying to do them simultaneously for years and concluding that she is broken.

And one further caution. This whole analysis is about aggregate arithmetic. It describes a population-level squeeze, and it explains nothing about any particular relationship, any particular man, or any particular reader. A structural account is a poor guide to an individual life and an excellent guide to why an individual life has felt unaccountably hard.

The claim that women now outnumber men among tertiary graduates across most OECD countries should be verified against the current OECD education release; Japan is a partial exception depending on whether junior colleges are counted alongside four-year universities, and the essay avoids a Japanese figure for that reason.

The male labour-market material rests on a large and contested literature, most of it American. The link from male economic stability to marriage rates is long-established in that literature and the size of the effect is argued about; it should not be treated as a mechanical law and does not transfer to Japan without qualification.

The claim that expectations of marrying an equal or higher-status partner persist more strongly in Japan than in comparable countries comes from the assortative-mating research literature. It is a comparative claim about norms and should be checked against current work; the further claim that men frequently do not want the alternative is supported in attitude surveys but is stated here as a pattern rather than a measured rate.

The Japanese survey material — that most unmarried people intend to marry and that not having met a suitable partner is the most-cited obstacle — comes from the national fertility survey and is a stable finding across waves. Percentages move; none is claimed here.

The dual control model and the responsive-desire account are established frameworks in sexuality research rather than settled neuroscience. The essay applies them to a social situation, which is an interpretation and is flagged as one. No claim is made that evaluation causes a measurable physiological effect in any individual.

The affective-touch material refers to research on dedicated pathways for gentle touch in the human nervous system. That research is real; the leap from it to what a person needs emotionally is an interpretation, and the essay should not be read as claiming that touch has a demonstrated clinical effect.

The Chinese “leftover women” term, the Korean material and the American loneliness advisory are described from widely reported public discourse. The Chinese term’s official history in particular is contested in its details.

And the whole essay is about a population-level pattern in wealthy societies with high female educational attainment. It does not describe every woman, every man, or any marriage, and a reader who finds that it does not describe her has not found an error in her life.

The thing we would most like a reader to take from this is a subtraction rather than an addition.

You are not difficult. You did what you were told, which was to become excellent, and the norm about who you should stand beside did not update to match. The resulting shortage is arithmetic performed on you by a generation of structural change nobody planned. It is not a verdict on your standards and it is not a verdict on the men either.

And underneath the arithmetic, the thing you actually want is not embarrassing and not unusual. To be held. To be wanted specifically. To be paid unhurried attention by somebody who is not assessing you. To feel pleasure that is not evidence of anything. Those wants are equipment, not weakness, and the reason they are hard to say is that you have been speaking a different language all day, for years, extremely well.

What is hard is not the wanting. It is the sentence. And the sentence is a skill, which means it can be practised somewhere the cost is low before it is needed somewhere the cost is high.

That is what an evening with Moonlight is: unhurried time in which saying what you want is the activity rather than the risk, where the terms are yours and set in advance, where a boundary is part of the conversation, and where nothing is inferred from your silence. Not a relationship. Not a treatment. Not an answer to the arithmetic. An hour in which nobody is evaluating you, including you.