Background
20–40% of city space sits idleSpace, enterprise, and talent are all waiting on an answer
Cafes and coworking spaces have large blocks of idle time, enterprises pay high fixed costs to run an office, and remote and hybrid workers can't find a stable third space to use. These are three faces of the same supply-demand gap.
NexCircle Research continuously compiles official statistics and case studies on housing and workspace in Taiwan. This is the first phase of the research summary, to be updated as new data becomes available. Full sources are listed at the bottom of the page for verification.
Beyond home and the office, somewhere you can focus deeply on work and still connect with people naturally. Cities often hold plenty of spaces like this, just scattered, sitting idle, never properly discovered.
Pain point one · Space owners: idle time, no stable foot traffic
Taipei already has a substantial density of cafes and coworking spaces, but they operate independently, each with 20–40% of their time slots sitting idle, with no way to pick up each other's overflow. What space owners actually need is steady, predictable foot traffic.
EnterprisePain point two · Enterprises: office costs are high, but employees don't always want to come in
Read more
Per JLL's Q1 2026 Taipei office market report, average monthly rent for Grade-A office space in the core business districts reached NT$3,271 per ping, up 0.39% quarter over quarter; full-year 2025 growth was around 1.7%, and the market has entered a slower-growth "plateau." Submarkets vary widely: Dunbei runs around NT$2,891/ping, Songjiang-Nanjing and the western districts around NT$2,694/ping, and Nangang around NT$2,367/ping. This is a fixed cost under a signed lease, due every month regardless of actual seat utilization, the flip side of the same coin as the "20–40% idle time" problem on the space-owner side: the capacity sitting idle for space owners is, for enterprises, fixed spending that never gets fully used.
Read more
From Cushman & Wakefield's Global Flexible Office Trends survey, showing that enterprises are making real purchasing decisions around flexible office space, not just the individual subscription market. This is a global figure; no equivalent Taiwan-specific survey exists yet for comparison.
Pain point three · Individuals: need an outside third space, without being tied to one
Remote workers, hybrid workers, digital nomads, freelancers, and founders look different on the surface, but share the same underlying need: an external third space to work from, without being tied down to one specific store or office.
The five profiles below are real, anonymized cases drawn from the research team's interviews and on-site observation, used to connect the statistics above to concrete situations.
Read more
The population estimate is an industry estimate, not an official census figure; MBO Partners, DemandSage and similar organizations cite one another, so the actual number varies considerably. The composition is also shifting: it no longer resembles the freelancer-dominated pattern from before 2020, meaning employers of this group are also potential 2B targets, not just the individual subscription market.
Read more
According to 104 Job Bank, Taiwan's largest job platform, over 5,300 job postings mentioned WFH by January 2022, still only about 1.5% of the overall job market but up 4.6x from three years earlier. This is Taiwan's only remote-job statistic, a one-time survey with no public follow-up tracking since 2022 (see "Data still missing" further down the page). The more recent signal comes from FlexJobs, a US job platform: remote job postings grew 20% in Q1 2026 versus Q4 2025, a sign of the market warming back up after cooling earlier. That is a US figure, a separate matter from Taiwan's own statistics. Both figures point the same direction: remote work opportunities keep growing, though Taiwan's own trajectory between 2022 and 2026 remains a data gap.
Space, enterprise, and talent are each stuck for a different reason, but the root is the same supply-demand gap. What NexCircle wants to do is connect all three.
Our Approach
In one line: NexCircle starts from idle urban space, researching and validating a third-space collaboration model for space owners, companies, and talent.
NexCircle Research set out to answer one specific question: can a city's idle third-space resources be organized into a usable service that solves the problem for all three parties above? The research focuses on Taipei as the first city to validate this, given its combination of the highest housing pressure, solid cafe density, and a steadily growing remote workforce.
Right now, NexCircle is building up a list of users who need third space through community events and the residency program. The next step is field-testing the subscription mechanism with a small group, to validate whether it can be win-win, users paying in and space owners earning a fair share. Only once that assumption holds up would integrating enterprise procurement demand be worth considering.
Turning idle time into a subscription service
The NexCircle Coworking Pass is a subscription service we've designed, not yet formally launched. Its core logic follows the sharing economy: seats that already exist in the city but sit idle get redistributed to the people who need them through a single pass, cutting down on the waste of space that gets built but never fills up. It's also a circular-economy way of thinking, reusing a resource that already exists rather than building more. Here's how the mechanism is designed to work.
- Cafes and coworking spaces, with an idle-time share estimated at 20–40% (first-hand observation)
- Space partners are joining on a rolling basis, expanding the network
- Remote, hybrid workers, and digital nomads
- Common thread: need to work from outside space, without being tied to one place
The design comes down to a spread: the monthly fee a subscriber pays, converted into an effective cost per visit, needs to come in below what they'd pay hunting for space one venue at a time; the wholesale rate a space receives needs to exceed the opportunity cost of leaving that time slot idle. Whether that gap holds up is what determines whether the mechanism can run sustainably, and actual rates will move with each space's wholesale terms and usage frequency.
Activating space, making transactions repeat
Croissant and Deskpass show that the "subscription, cross-space access" model already has a stable operating track record in mature markets: Croissant currently covers 100+ cities and 700+ locations, and Deskpass covers 22 countries and 290 cities, with both still expanding. Nobody has actually tested this model in Taiwan yet. What NexCircle wants to do is bring a mechanism already proven in international markets into Taiwan as the first real test, taking a first-mover position in this market. The closest local comparison in the Taiwan market is Sio-Tshiu, but its model is built around operating its own venues, which differs from NexCircle's approach of integrating existing idle time from independent space owners.
NexCircle wants to layer a community-connection mechanism on top of that: spaces release idle time, the Pass matches occupancy, and participants who build community ties through that connection return and refer others, giving the space steady foot traffic and drawing in more space owners to join the network. This mechanism is drawn from actual observation during the residency program, and has only been validated in one space so far; whether it can be replicated into newly onboarded spaces is the core hypothesis to prioritize validating next.
- Space releases idle time: Cafes and coworking spaces open up unused time slots
- Pass matches occupancy: Users are matched by demand; check-in at the venue starts the timer
- Community-connection mechanism: Drawing on residency-program-style community building, regular gatherings and mutual support
- Belonging drives return visits: Once users build community ties, they return and refer others
- Space gains steady foot traffic: More brand exposure, drawing in more space owners to join the network, back to step 1
Method
NexCircle Research has, since July 2026, continuously compiled official statistics and case studies related to housing and workspace, building up the Field Notes series. This summary extends the same methodology: only official statistics and verifiable primary sources, with data gaps flagged directly rather than papered over with secondhand blog numbers. Figures will be revised as official statistics update; the page footer notes the verification date for each.
Field NotesField observation: the community-connection variable
The community-connection variable in this research scope comes from on-the-ground fieldwork in Taipei, an extension of the same methodology.
Community fieldwork through Ouer Guanghe
Running the "Ouer Guanghe" community, testing community and space activation across Taipei's cafes on an ongoing basis. The working assumption: space and community need to be designed together for either to stay active long-term.
Residency program at a converted coworking space
Running a residency program at a coworking space converted from an old clock-shop building, observing firsthand how community-connection mechanisms affect a space's return-visit rate.
Coworking Day interviews
Regular conversations with remote workers, hybrid workers, and digital nomads newly arrived in Taipei, building up recurring patterns in what they need, which carries more weight than conclusions from a single survey.
Impact Assessment
This section covers the early results from fieldwork, along with the two key variables still being validated going forward.
These two figures come from ongoing interviews and observation at multiple cafes and coworking spaces in Taipei since 2022 (see the Field Notes timeline above). The sample size is limited and this isn't a citywide census; it's included to show the basis for validating Hypothesis 2, not as an official statistic, so please note that when citing it.
Read more
A survey study of Taiwanese remote workers (N=195) found that horizontal peer relationships significantly reduce workplace loneliness (workplace friendship β=-0.49, p<.01), with a significantly stronger effect than vertical manager-subordinate relationships. A separate study on digital nomad lifestyles similarly recorded loneliness as the most commonly cited challenge, typically eased by joining communities that form naturally within coworking or coliving spaces; one recurring format has organizers bring together nomads with shared interests for a 15+ day itinerary, working independently during the day and sharing meals in the evening, a rhythm similar to the residency program's weekly co-working day.
The residency program has built up a small, steadily growing group of long-term participants to date, spanning fields from dream interpretation to art therapy to sound-healing music, holding regular in-person gatherings each month, replacing solo experimentation with collectively accumulated experience, and continuing to validate early users' feedback on the model.
Research ScopeLong-term research scope: two key variables
The data above answers "how much demand, how much pressure exists." Whether this gap can be closed by a business mechanism is the research scope NexCircle Research is tracking long-term; the two variables below are still being validated as more data comes in.
Can the revenue-share mechanism scale while working for both sides
How the residency program actually runs today: organizers don't have to cover venue rent, bringing in limited-run programming; the space owner gets otherwise-idle time filled without having to plan the content themselves, a win-win mechanism. Whether this logic can extend into a subscription-based revenue-share structure that works for both sides at Taipei price points is what comes next to validate.
Can community connection lift return-visit rates
Pure space rental tends to become a one-off transaction; spaces where users build community ties see higher return and referral rates. Whether this can be decoupled from a specific operator or space and replicated elsewhere is the direction being validated with more samples. The academic literature supports the premise that connection itself reduces loneliness (see the Field Notes citation above), but whether it can be replicated in a new space is still something NexCircle has to validate itself; the literature doesn't have a ready answer.
Full Data
The figures cited in the Background and Our Approach sections above, with full sources and supporting detail, are collected in this section for verification and citation.
Demand · Detail| Indicator | Figure | Source |
|---|---|---|
| Workers with remote work experience | 42.7% (2024; as of the August 2026 check, no more recent equivalent survey was found) | Ministry of Labor / DGBAS, compiled by 104 Job Bank |
| Regular hybrid work (1+ day of remote work per month) | 22.8%, with 5–10 days/month at 6.0%, more than 10 days/month at 4.0%, fully remote at 1.6% | Same as above |
| Cumulative Employment Gold Card approvals | 15,378 (approx. 8,184 currently valid) | Taiwan Employment Gold Card official site, Dec 31, 2025 |
| Taiwan Digital Nomad Visa | Eased starting January 2026, allowing stays of up to 2 years | Verified, see /visa/ |
Enterprises already offering flexible work benefits
This 104 Job Bank survey dates to early 2022 (after the pandemic peak) and reflects the share who had ever worked from home, not who still do; the same survey also found the share continuing WFH after the pandemic eased dropped back to about 5%, included here for context.
Enterprise BudgetAn existing budget mechanism for enterprises: the Employee Welfare Fund Act
Taiwan enterprises don't need a new budget line to fund flexible work or third-space benefits, since a statutory mechanism already exists. Taiwan's Employee Welfare Fund Act (職工福利金條例) requires companies to contribute to a welfare fund at set rates, managed and disbursed by an in-house Employee Welfare Committee, typically under its recreation/leisure or welfare-assistance spending categories.
See full contribution rates
| Contribution source | Rate | Basis |
|---|---|---|
| Capital at company founding | 1%–5% (one-time) | Employee Welfare Fund Act, Art. 2 |
| Monthly revenue | 0.05%–0.15% | Same as above |
| Monthly employee salary (payroll deduction) | 0.5% | Same as above |
| Proceeds from scrap/waste sales | 20%–40% | Same as above |
Welfare committees also face category caps on how this fund can be spent: recreation and leisure, up to 50%; education and grants, up to 40%; welfare assistance, up to 30%. The actual usable share still depends on each committee's own disbursement rules.
Supply: the state of housing space compression
| Indicator | Figure | Source |
|---|---|---|
| National price-to-income ratio | 9.47x (rose in Q1 2026 after several quarters of decline through 2025, still far above the World Bank's recommended 3–5x range) | Ministry of the Interior Real Estate Information Platform, Q1 2026 |
| Taipei City price-to-income ratio | 14.72x (same quarter, trending opposite the national figure, down from the previous quarter) | Ministry of the Interior Real Estate Information Platform, Q1 2026 |
| Households without owned housing, Taipei City | 347,339 households, 48.2% of which rent (approx. 167,000 households), for an estimated renter population of approximately 467,000 (as of the August 2026 check, this is a low-frequency census-type figure and no version newer than 2018–2022 was found) | Construction and Planning Agency, MOI / Taipei City Government DBAS, 2018–2022 |
| Per-capita living space, Taipei City | Approximately 10.5 ping (approx. 34.7 sqm) | Same as above |
Supply: coworking spaces and cafes
Taipei already has a substantial density of cafes and coworking spaces. The problem is that these spaces operate independently: users have to search one by one, and the spaces themselves aren't connected to each other.
| Indicator | Figure | Source |
|---|---|---|
| Number of cafes nationwide | 4,824, up 42.5% from 2018 (as of the August 2026 check, still the most recently published figure) | Ministry of Finance Statistics Bulletin No. 2, Jan 23, 2025 |
| Cafe density, Taipei City | Approximately 4.2 per 10,000 people | Same as above |
| NexCircle space directory entries | 121 digital-nomad-friendly spaces across 11 cities (Taipei 28 / Tokyo 23 / Seoul 22 lead) | NexCircle's own database, see /spaces/, not an official census figure Self-reported |
Data still missing
Taiwan has no official count of coworking spaces nationwide, no idle-seat utilization figures for cafes and shared offices, and remote job posting counts exist as only a one-time snapshot with no ongoing tracking.
Read more
The Ministry of Finance only publishes cafe counts and revenue, not customer dwell time or the share of use for work purposes. We can currently size "how much demand exists" but not "how much idle capacity is available to absorb it," a gap to fill through primary research (surveys, operator interviews) going forward.
For reference, there's a global industry figure: Deskmag's Global Coworking Survey 2025 found average occupancy at global coworking spaces of about 68% in early 2025, higher than 70% in major cities. This isn't a Taiwan figure, and the calculation isn't exactly equivalent to an "idle-seat rate" either; it's included only as a rough point of comparison.
Taiwan's remote job posting count exists only as a single January 2022 snapshot from 104 Job Bank (see "Signal from job postings" above), with no public follow-up since, so it isn't possible to confirm whether postings kept growing or pulled back between 2022 and 2026. The only current point of comparison is from FlexJobs, a US platform.
Join Us
This is the first-phase output of NexCircle Research's ongoing study of housing and workspace in Taiwan, with more data updates and extended research to follow. The current data answers whether the problem exists; whether people are willing to pay for this approach still needs more participants to validate, which is also what this page is inviting you into. To get the full dataset, discuss a collaboration, take part in the research, or share data of your own, leave your contact details below.
Figures on this page are compiled from official public statistics for reference only, and do not constitute a formal publication. Please check the original source links on the page before citing. Last updated: 2026-08-19.
To learn more about the team behind this research, see About NexCircle.