Two Industries: A Primer on Japan’s Male-Facing and Women-Facing Services

This is a primer rather than an argument. It is written for readers who want the actual shape of the thing — and for anyone doing research who has discovered how little of this is laid out in one place in either language.

A warning about that, first, because it governs everything below. Almost nothing in this field is properly measured. There is no reliable public figure for the size of either sector, for how many people use them, for repeat rates, or for growth. What exists is operator marketing, journalism working from a handful of interviews, and survey work with severe self-selection problems, because the honest answer to a survey about this carries a social cost that varies enormously by who is being asked.

So this primer states structure, which is knowable, and is explicit about magnitude, which is not. Where a number would ordinarily appear, there is a description of what would have to be measured instead. A document that filled those gaps with plausible-looking figures would be more satisfying and considerably less useful.

Both sectors sit on exactly the same law, and this is the first thing that surprises people, because the two look so different that most readers assume different rules.

Japan’s 1956 statute prohibits prostitution, attaches no criminal penalty to the act for either party, and defines the offence around one specific act. Everything that is not that act falls outside the statute entirely — not as a tolerated evasion but because the law does not reach it.

A second statute, decisively amended in the mid-1980s, names the sex-business category, requires operators to notify the prefectural authorities rather than obtain a licence, zones them by local ordinance at distances from schools, hospitals and childcare facilities, and froze new premises-based openings. Existing shops continued under acquired rights.

Employment of anyone under eighteen is prohibited outright and is the sharpest enforcement line in the field, with the buyer rather than the seller as the primary target where a minor is involved. This is the one place where the law is unambiguous, and it applies identically to both sectors.

Hold that identity in mind. Whatever explains the differences between these two industries, it is not that they are permitted to do different things.

It is useful to separate three legally distinct things that casual discussion runs together, because conflating them produces most of the confusion in English-language writing on this subject.

The first group is the premises-based sex businesses: the private-room bathhouse, the various shop-based services organised around a room and a fixed session, and the older theatre and cinema formats. These are the businesses the 1980s freeze applies to. In most of the country a new one cannot legally open, which means this tier is a fixed and ageing stock.

The second group is the non-premises tier — services delivered to a hotel or residence, arranged by phone or online. This is where the male-facing industry has grown, for the same regulatory reason the women-facing sector exists only here.

The third group is legally not a sex business at all, and it is enormous. Hostess clubs, snack bars and girls’ bars are regulated as hospitality and entertainment businesses under a different provision entirely. Nothing sexual occurs. What is sold is conversation, attention and the experience of being interesting to someone for two hours. It is entirely legal, entirely mainstream, advertised openly, and a great many men spend a great deal of money on it.

That third group is the single most instructive fact in this primer, and we will come back to it, because an industry that sells no sexual contact whatsoever and thrives is telling you something the sexual categories cannot.

The women-facing side has a shorter list, and the shape of the list is itself the finding.

The sexual-service category — commonly abbreviated in Japanese to a two-character shorthand — operates almost entirely as non-premises delivery, to a hotel or the client’s home, arranged online. It emerged in recognisable, named, organised form only in roughly the last fifteen to twenty years.

Alongside it sit companionship and rental services, which supply an outing, an occasion or a companion for an afternoon, and which generally involve no sexual contact and are often not registered as sex businesses at all, because there is nothing in them for that framework to catch.

And then there is the host club, which is considerably older than either and is the direct structural counterpart to the hostess club: hospitality, conversation and attention, legally not a sex business, culturally enormous, and with a well-documented problem around escalating spending that the Japanese press has covered extensively.

What is absent from this list is what matters. There is no women’s premises tier. No district, no shopfront, no equivalent of the bathhouse. Not a single one.

The explanation for that absence is the most commonly mistaken thing in this whole subject, so it is worth stating as plainly as possible.

It is usually attributed to culture: Japanese women are discreet, female desire is private, a women’s red-light district would be socially impossible. Those claims may or may not be true, and they are not the reason.

The reason is that the women-facing sector arrived after the door closed. Premises-based openings were frozen in the mid-1980s. A category that only organised itself in the 2000s and 2010s could not have obtained premises even if every party involved had wanted them, because the legal route to a new shop was shut a generation before the demand became visible.

This single fact — regulatory timing, not cultural temperament — cascades into nearly every other difference in this primer. The male-facing industry inherited geography. The women-facing industry was structurally confined to delivery from its first day, and everything about how it operates follows from having no building to operate in.

It is worth noticing how much explanatory work a cultural story was doing here, and how little it was needed. This happens often when people reason about Japan.

Set the legal frame aside and look at the products as products. The differences are systematic rather than incidental, and they cluster.

Session length. The male-facing premises and delivery services are organised around comparatively short fixed blocks. The women-facing services run substantially longer as a norm, with multi-hour bookings ordinary rather than exceptional. When a market’s standard unit is three hours instead of one, it is not selling the same thing more slowly. It is selling something else.

What the brand attaches to. In the male-facing sector the shop is the brand: it advertises, it holds the reputation, and the individual is presented from a roster. In the women-facing sector the individual practitioner is the brand, frequently with a personal social-media following that travels with them between operators. The client is choosing a person and the operator is closer to infrastructure.

How the menu is written. Male-facing menus tend toward enumeration — a time block and a specified set of what it contains. Women-facing menus tend toward the person and the shape of the encounter, with far more weight on prior communication, stated preferences and what the client wants the evening to feel like. A pre-session consultation is a normal feature on one side and unusual on the other.

Accountability infrastructure. The male-facing sector has a large, public, long-established review ecosystem — dedicated information sites carrying extensive user reports. The women-facing sector has very little of this, and relies instead on practitioner self-presentation and private word of mouth. Note carefully which way this cuts, because the intuitive reading is backwards.

And vocabulary. The women-facing sector very commonly calls its practitioners therapists.

This deserves its own section because it is the most consequential piece of language in the sector and it is routinely misread, including by people who should know better.

It is a marketing term. It is not a clinical credential, it does not correspond to any licensure, and it should not be read as indicating training in psychology, counselling, trauma or any regulated therapeutic practice. Nothing in the regulatory framework described above requires or verifies any such training, and nothing in it could.

The word was adopted because the sector needed a vocabulary that was not the male-facing sector’s vocabulary. That was a real problem with no good solution available: the existing words carried associations that did not describe what was being offered, and inventing a word from nothing rarely works. Reaching for a borrowed register was an understandable move.

It was also a costly one, and the cost falls on clients. A woman who books someone called a therapist because she is struggling may reasonably believe she is accessing something adjacent to care. She is not, and the person arriving is generally not equipped to recognise when she needs something he cannot provide. Anyone in this field who takes the word at face value has made an error with consequences, and anyone using it has an obligation to be clear about what it does not mean.

For a researcher, the practical implication is narrower and still important: the term cannot be used as a proxy for anything. It marks a market positioning, not a qualification.

Four things do most of the explanatory work, and they compound.

Regulatory timing, already covered, which produced the delivery-only structure and everything downstream of it.

Accumulated capital and infrastructure. The male-facing sector has had centuries to build premises, supply chains, advertising channels, recruitment pipelines, review platforms and a body of operational convention. The women-facing sector has had roughly two decades and began with none of it. Differences that look like deliberate choices are frequently just the difference between an old industry and a young one.

Stigma asymmetry, which is real and which runs in a specific direction. Male use of these services is socially jokeable in Japan — referenceable among colleagues, present in mainstream comedy. Female use is not, or is only beginning to be. That asymmetry pushes the women-facing sector toward discretion in a way that reinforces the delivery model, suppresses public review culture, and makes the market harder to observe from outside. It is worth being precise that this is a difference in social cost, not a difference in what people want.

And the shape of the demand itself, which does appear to differ. The evidence for this is indirect but it is consistent: the women-facing sector’s premium attaches to duration, to the specific practitioner, and to the quality of communication before anything begins. Those are the features you would price highly if what you were buying were the encounter rather than the act. The male-facing sector prices differently, and its own conversation-only tier suggests the same appetite exists there too, separated into a different category rather than absent.

Both sectors are moving, in opposite directions, for reasons that are largely structural.

The male-facing premises tier can only contract. The freeze means no new shops, so the stock is fixed and ageing, and it faces a shrinking young-male population, competition from digital material that costs nothing, and payment and advertising infrastructure that has become progressively less willing to serve the category. Consolidation and closure are the expected shape.

The women-facing sector is expanding, and the mechanism is visible even though the magnitude is not. Operators have multiplied, recruitment advertising aimed at prospective practitioners has become conspicuous, franchise-like structures have appeared, and the practitioner-as-brand model has produced individuals with substantial independent followings. The most important input was not commercial: mainstream manga and television treatments in the early 2020s supplied a vocabulary and a depiction, and a category grows fastest at the moment women acquire a word they can say to themselves.

Both sectors face the same two external pressures. Payment processing and advertising platforms continue to tighten against adult categories, which advantages operators with the scale to absorb it. And synthetic companionship is arriving as a genuine competitor for the conversational and attentive portion of the demand — cheaper, always available, and improving quickly at exactly the part the conversation-only businesses have been selling for decades.

The unresolved question worth watching is whether the resilient categories turn out to be the ones organised around a physical act or the ones organised around attention. The conversation-only tier has survived every previous technological substitution, which is some evidence. This may be the first substitute that competes with it directly.

This section exists because the honest boundary of the field is itself information, and it is rarely printed.

There is no reliable market size for either sector. Figures in circulation derive from industry estimates with undisclosed methodology, or from extrapolations built on operator counts that cannot capture non-premises activity. Treat any confident number as unsourced until it shows its working.

There is no dependable customer demographic profile. What circulates comes from operator self-reports, which are marketing, and from journalism built on small interview samples selected precisely for willingness to talk — which is the most non-representative selection criterion available in a stigmatised field.

Repeat and retention behaviour is entirely unmeasured in public. So is spending distribution, which matters enormously, because a market with a small number of heavy users and one with many occasional users have completely different social meanings and the aggregate cannot distinguish them.

Labour conditions research is thin on the male-facing side and close to absent on the women-facing side. Practitioner recruitment, pay structure, retention, exit and harm are the least documented part of the entire field, and they are the part where the human stakes are highest.

And the comparative evidence on regulatory outcomes — the question of which of the world’s models produces less harm — is genuinely unsettled everywhere, for the structural reason that stigmatised and partly criminal activity is badly measured in every jurisdiction. Anyone claiming a settled answer is arguing from conviction.

What would actually be needed: anonymised operator transaction data under a research agreement, a properly sampled population survey with the design work to handle the sensitivity, and longitudinal practitioner interviews with real protections. None of this exists. A researcher entering this field should know that the gap is the finding.

A reader of this Library will have assembled a flattering conclusion somewhere around the middle of this piece, and it needs dismantling before the end.

The conclusion is that the women-facing sector is the enlightened one — longer sessions, communication first, the person rather than the act, no sordid district — and that the male-facing sector is the crude inheritance. It is a tempting reading and it is wrong in a way that matters.

The women-facing sector is not more ethical. It is younger, and youth is not a virtue. Consider what it lacks. It has no meaningful accountability infrastructure: the male-facing sector’s review ecosystem is coarse and often unpleasant, and it does make a bad operator visible, which a private word-of-mouth market does not. It places the practitioner under a personal brand, which means the individual absorbs the risk, the exposure and the reputational consequence personally, with an operator standing further back than in the older model. It borrows a clinical vocabulary that promises a competence nothing verifies. And it has no accumulated operational convention, which sounds like freedom and also means every safeguard has to be invented by whoever happens to care.

The older sector’s ugliness is at least legible. A newer sector’s problems are not yet visible, and being invisible is not the same as being absent.

The accurate statement is narrower and duller: these are two industries at different ages standing on identical law, each with a characteristic failure mode, and neither has solved the thing that actually determines whether a person is treated well inside it.

The statutory outline here is a summary of structure and not legal advice. Provisions are amended, and the prefectural ordinances that do the zoning differ across Japan, so national-level statements are generalisations. Anyone with a practical reason to rely on any of it should read current text.

The category descriptions are drawn from how these businesses present and organise themselves publicly. They are accurate as a map of how the sectors are structured and they are not a survey: individual operators vary enormously, and any statement beginning “these services are” describes a centre of gravity rather than a rule.

The dating of the women-facing sector’s emergence is approximate and reflects when it acquired a public name, visible operators and a recognisable shape — not when any first service existed. Earlier and scattered precursors did exist.

The structural comparisons — session length, where the brand attaches, how menus are written, the density of review infrastructure — are read from public self-presentation and market organisation. They are consistent and checkable by observation, and they are not quantified, because the data to quantify them is not public.

The claim that regulatory timing rather than culture explains the absence of a women’s premises tier is an inference from the sequence of the freeze and the sector’s emergence. It is the essay’s own reading. It does not establish that cultural factors are absent, only that they are not needed to explain this particular fact.

The account of market direction is qualitative throughout. No growth rate, market size or customer count appears anywhere in this piece, because none is publicly measured, and the section above says so at length rather than leaving it implied.

If there is one thing to carry out of a primer like this, it is probably the third category — the one that sells no sexual contact at all.

Hostess clubs, host clubs, snack bars: entirely legal, entirely public, advertised without euphemism, and sustained for decades on conversation and attention alone. They are the part of this landscape nobody has to explain away, and they are also the part that reveals what the rest is organised around. A society in which enormous sums are spent, openly, on being listened to for two hours is a society telling you something quite specific about how difficult it is to be listened to for free.

That is the fact a researcher should probably start from, and it is also the one a reader can check against her own week without consulting any literature at all.

What this Library can offer is the smaller half of it: language for the things that have become difficult to say, in a register that does not require anyone to be a case.

And the other half is more specific than anything in this primer. An evening in which somebody is paying attention, the terms set in advance and by you, where saying what you want is the activity rather than the risk, and where nothing is inferred from your silence. Not a category on the map above, not a solution, and not a therapy. A few hours of being accompanied, described accurately.