By Boo Kok Chuon

With the Johor Bahru–Singapore Rapid Transit System Link approaching operation, confidence has returned to Johor Bahru’s high-rise property market. The logic is not difficult to understand. The RTS will connect Bukit Chagar to Woodlands North in approximately six minutes, replacing one of the least predictable portions of the present cross-border commute with a considerably faster and more reliable rail connection. For Malaysians working in Singapore, this naturally strengthens the proposition that one could earn a Singapore income while living across the border in Johor Bahru.

Developers appear equally persuaded by that proposition. New serviced-apartment projects are rising across Johor Bahru, with improved connectivity to Singapore forming an important part of the wider investment narrative. The more interesting question, however, is not whether the RTS will improve Johor Bahru’s attractiveness. It almost certainly will. The question is whether the amount of new housing demand created by the RTS is remotely comparable with the amount of new housing now being built.

In our earlier article, Will the RTS Link Empty Woodlands?, we examined the same proposition from the Singapore side. We argued that the hundreds of thousands of people already crossing the Causeway should not be mistaken for hundreds of thousands of Singapore residents waiting to move to Johor. This time, we turn the telescope around and ask the mirror question: if the RTS will not necessarily empty Woodlands, will it fill Johor Bahru?

According to CIMB Research, citing National Property Information Centre data, Johor had approximately 108,863 existing serviced apartments in the first quarter of 2026, with another 41,832 units incoming and 18,712 units planned.^[4]

Johor serviced apartments	Units
Existing stock	108,863
Incoming supply	41,832
Planned supply	18,712
Incoming + planned	60,544

The incoming and planned pipeline therefore amounts to approximately:

of the existing serviced-apartment stock. Put differently, for roughly every two serviced apartments that already exist, Johor presently has approximately one additional unit incoming or planned. That is a substantial supply response, and it raises the central question of this article: how much of that additional housing demand can reasonably be attributed to the RTS?

Passengers Are Not Houses

Cross-border movement between Johor and Singapore is already enormous. Singapore’s Immigration & Checkpoints Authority recorded an average 327,000 travellers per day through Woodlands Checkpoint alone in 2024, up from 269,000 in 2023.^[1] A commonly cited estimate also puts the number of Malaysians commuting into Singapore daily at more than 300,000, although the latter should be treated as a directional estimate rather than an audited count of unique commuters. These figures are not directly interchangeable because checkpoint movements, passenger journeys, commuters and unique persons are different statistical units, but the order of magnitude is sufficiently clear: hundreds of thousands of cross-border movements already take place every day.

That distinction is important because the RTS does not create a cross-border labour market where none previously existed. It improves the transport infrastructure serving a labour market that has existed for decades. If a Johor resident who presently rides a motorcycle into Singapore switches to the RTS, the railway has gained a passenger but Johor has not gained a household. Likewise, if an existing Johor commuter subsequently moves from Tebrau to an apartment closer to Bukit Chagar, housing demand may have shifted geographically without increasing the aggregate number of homes required in Johor.

The population relevant to the “move back to JB” property thesis is therefore narrower. It consists principally of people who presently live in Singapore because the existing cross-border commute is too burdensome, but who would choose to live in Johor once the RTS reduces that burden sufficiently. In simplified form, the causal chain is:

That mechanism is entirely plausible. What we do not presently have is a sufficiently reliable estimate of how many Singapore-resident Malaysian workers sit close enough to that residential decision margin to move once the RTS opens. One could attempt to estimate it using rental savings, salaries, household composition, travelling time, workplace location and individual commuting tolerance, but doing so would require a behavioural model built upon several uncertain assumptions.

There is a simpler way to test the proposition. Rather than trying to predict how many people will choose to move, we can ask how many conventional Singapore-bound commuters the RTS could physically carry during the period in which they actually need to travel, and then deliberately structure the assumptions in favour of the bullish property thesis.

Maxing Out the RTS

MRT Corp states that the RTS is designed to carry up to 10,000 passengers per hour per direction at ultimate peak capacity.^[2] For the present exercise, we therefore do not need to rely upon MRT Corp’s projected opening ridership of approximately 40,000 passengers per day.^[3] We can go further and assume that the railway has already reached its ultimate operating capacity.

The relevant constraint, however, is not total theoretical passenger capacity over an entire day. Conventional workers do not commute uniformly throughout the day. A worker whose Singapore job begins at 9am cannot simply arrive at Woodlands North at 9am, because Woodlands North is not his workplace. He must still travel through Singapore’s domestic transport network to the CBD, Jurong, one-north, Changi or wherever his workplace happens to be. The realistic commuting chain is therefore not merely Bukit Chagar to Woodlands North, but:

RTS services are expected to begin from 6am.^[2a] If a conventional 9am worker budgets approximately one hour after reaching Woodlands North to complete the Singapore leg of the commute, he would ideally need to reach Woodlands North by around 8am. There is, however, no reason to make our conclusion depend upon that single assumption. We can instead run a sensitivity analysis by progressively stretching the usable morning commuting window.

How Much Difference Does the Commuting Window Make?

Suppose the RTS operates continuously at its ultimate capacity of 10,000 passengers per hour in the Singapore-bound direction. We can then test three different morning commuting windows:

Assumed Singapore-bound commuting window	Hours	Maximum passengers	Share of 60,544-unit pipeline
6am–8am	2.0	20,000	33.0%
6am–8.30am	2.5	25,000	41.3%
6am–9am	3.0	30,000	49.6%

The first scenario assumes that our conventional 9am worker needs to reach Woodlands North by 8am. The second gives him another half an hour. The third stretches the assumption considerably further by allowing Singapore-bound workers to continue entering Singapore right up until 9am, notwithstanding that our hypothetical conventional working day begins at 9am.

Under that final scenario, the arithmetic is straightforward:

For the purpose of stress-testing the RTS property thesis, we shall use the 30,000-passenger case, precisely because it is the assumption most favourable to the proposition being tested.

What Would Have to Be True for 30,000 Passengers to Become 30,000 Homes?

The next step is where the distinction between passenger capacity and housing demand becomes important. To convert the full 30,000-passenger capacity into 30,000 units of incremental serviced-apartment demand, we would first have to assume that every passenger travelling from Johor to Singapore between 6am and 9am previously lived in Singapore and relocated to Johor solely because of the RTS. That effectively assigns no morning peak capacity whatsoever to the substantial population of existing Johor residents who may switch from motorcycles, buses, cars or the KTM Shuttle Tebrau to the new railway, and it likewise assumes away non-work passengers using the line during the same period.

Even that would not be sufficient. We would also have to assume that each relocating worker creates a separate household, rather than returning to a family home, living with parents, sharing accommodation or forming part of a household containing more than one Singapore-employed worker. Since the supply figure being tested comprises serviced apartments, we must further assume that none of those households chooses landed housing or another form of residential accommodation and that every relocating worker ultimately occupies one serviced apartment.

These are plainly unrealistic assumptions, but that is precisely why they are useful. They are not intended as a forecast. They are designed to establish a generous upper boundary by systematically pushing the model in favour of the proposition that the RTS can generate sufficient housing demand to absorb the new serviced-apartment pipeline.

Under that deliberately extreme scenario:

against:

which gives:

The result is quite revealing. Even if the RTS operates at its ultimate Singapore-bound capacity continuously from the first train at 6am until 9am, every available passenger slot is allocated to someone relocating from Singapore to Johor, every relocating worker forms a separate household and every household occupies one serviced apartment, the resulting hypothetical demand still accounts for less than half of Johor’s incoming and planned serviced-apartment pipeline.

This is also why the 30,000 figure should not be mistaken for our estimate of actual RTS-induced migration. It is the opposite. Once existing Johor commuters and other passengers occupy part of the same peak-period capacity, and once normal household formation is taken into account, the number of genuinely incremental serviced-apartment households attributable to RTS-induced relocation must necessarily fall below that upper bound. We do not attempt to estimate how far below because reliable data on those behavioural and household variables is not presently available.

There is another reason why the 30,000 upper bound should be regarded as generous. Cross-border commuters do not choose between the RTS and road transport in a vacuum; each morning they are effectively comparing competing generalised commuting costs, including fare, travelling time, waiting time, reliability, crowding and physical comfort. If the RTS becomes sufficiently crowded, the cheaper bus may become relatively more attractive; conversely, if enough passengers shift from road to rail, congestion on the Causeway may ease and the relative travelling-time disadvantage of the bus may narrow. Transport economics describes this tendency through Wardrop’s user-equilibrium principle, under which traffic distributes itself among available routes such that, at equilibrium, no traveller using the network can reduce his or her generalised travel cost simply by switching unilaterally to another available route.^[6]

The choice here is arguably even less deterministic because the RTS and bus alternatives are not presented to commuters within a single unified checkpoint. A commuter will generally have to commit to one route before fully observing congestion conditions on the other, so the morning decision contains an element of uncertainty based upon expected rather than perfectly observed journey conditions. This resembles the later concept of stochastic user equilibrium, which relaxes the perfect-information assumption in the classical Wardrop model.^[6] We do not attempt to estimate that equilibrium utilisation. Instead, our stress test deliberately assumes the entire problem away by filling the RTS continuously to its ultimate 10,000-passenger hourly capacity from 6am until 9am. Shift work, hybrid work, staggered hours and modal substitution may all alter the actual distribution of commuters, but none is required for our conclusion; if anything, they reinforce why the 30,000 figure should be read as an intentionally generous upper bound rather than a forecast of actual RTS-induced relocation.

The analysis also assumes conventional daytime employment. Shift workers, employees with staggered hours and others who can commute outside the conventional morning peak may utilise RTS capacity at different times. Without reliable data on the distribution of Singapore-employed Malaysian workers by working hours, it would be inappropriate to manufacture a numerical adjustment for that group. The point of the exercise is narrower: to test the familiar proposition that a large population of conventional Singapore workers can move to Johor and commute daily by RTS.

What About the 40,000-Passenger Opening Forecast?

MRT Corp separately estimates that the RTS will carry approximately 40,000 passengers per day upon opening.^[3] That figure remains useful as a reference point, although our analysis no longer depends upon how it should be interpreted. MRT Corp’s published material does not appear to expressly define whether the 40,000 figure represents passenger journeys, boardings or unique daily users.^[3a] An earlier version of this analysis therefore faced an unnecessary methodological problem: if 40,000 represented daily passenger journeys, one might derive approximately 20,000 conventional round-trip commuters; if it represented unique daily users, the calculation would be different.

The peak-capacity approach avoids that ambiguity altogether. We instead begin with MRT Corp’s independently published capacity of 10,000 passengers per hour per direction, identify the hours during which conventional Singapore-bound workers would actually need to travel, and ask how many such passengers could physically be carried within that window. The distinction matters because future growth in all-day ridership does not necessarily imply a proportionate increase in the number of conventional workers who can use the RTS for residential relocation. Spare capacity in the middle of the afternoon is of little assistance to someone whose employment requires him to be at a Singapore workplace by 9am.

For this particular property thesis, therefore, peak-direction capacity is more relevant than theoretical all-day throughput.

The RTS May Redistribute Demand More Than It Creates It

None of the foregoing suggests that the RTS will have little effect on Johor Bahru property values. Its impact may simply be more geographically concentrated than a broad “buy JB because of RTS” narrative implies. An existing Johor household may choose to move from Tebrau, Mount Austin or another part of Johor Bahru to a location closer to Bukit Chagar once the railway opens. That creates additional demand for housing within the effective RTS catchment, but it does not necessarily increase Johor’s aggregate housing requirement because the household already existed.

This distinction between aggregate demand and spatial redistribution is important for investors. The RTS can materially increase the relative attractiveness of properties with genuine accessibility to Bukit Chagar even if the increase in Greater Johor Bahru’s total number of households is much smaller. A proposition based upon owning scarce and genuinely well-connected property is therefore analytically different from a proposition that assumes the RTS will lift all Johor Bahru high-rise property more or less uniformly.

Sold Is Not the Same as Occupied

A related distinction arises between developer sales and actual residential absorption. If an investor purchases a serviced apartment because he expects the RTS to cause future capital appreciation, the developer records one unit sold. If a Malaysian family presently living in Singapore relocates to Johor and purchases an identical unit as its home, the developer also records one unit sold. During construction, the two transactions look the same in the sales statistics, but economically they represent different things.

The first transaction absorbs developer inventory through investment demand; the second absorbs housing through actual household formation. The distinction becomes more visible after vacant possession. If a sufficiently large population of occupiers eventually materialises, occupancy rises, rents remain supported and yields provide fundamental support to valuations. If household demand proves weaker than investor demand, the adjustment occurs through higher vacancy, rental competition, weaker yields and potentially greater resale supply.

This is why launch sales measure the absorption of investment inventory, whereas occupancy measures the absorption of housing. During a construction boom, both matter, but the latter ultimately determines whether the completed stock is supported by enough households to use it.

Where Does the Rest of the Demand Come From?

Our stress test does not establish that the remaining serviced apartments cannot be absorbed. Johor housing demand can arise from many sources other than Singapore workers relocating because of the RTS, including local population growth, ordinary household formation, employment expansion, the Johor-Singapore Special Economic Zone, Malaysians relocating for reasons unrelated to the RTS, Singapore and other foreign purchasers, retirees and wider economic development. CIMB itself expects the RTS and JS-SEZ to support parts of the Johor property market while warning that the high-rise residential segment could face oversupply risk if demand does not keep pace.^[4]

Those alternative sources of demand may ultimately prove substantial. The analytical point is simply that they are separate demand propositions and should be evaluated on their own numbers rather than quietly bundled under the three letters “RTS”. Once we max out the RTS at ultimate peak capacity and still derive an upper-bound housing figure below half the incoming and planned serviced-apartment pipeline, the burden of explanation necessarily shifts toward those other sources of household demand.

That is a much narrower conclusion than declaring Johor Bahru to be in a property bubble, but it is also a more defensible one.

A Final Reality Check: The RTS Is Not New Information

There is one final number worth considering. As of the first half of 2026, Malaysia’s Valuation and Property Services Department reported that Johor had 9,946 completed but unsold serviced apartments.^[5] Unlike our peak-capacity stress test, this is not a hypothetical calculation but an observed market outcome, and its timing is relevant because the RTS is not new information.

The project has been publicly known for years, its construction is physically visible, developers have launched projects into a market where RTS connectivity forms part of the investment narrative, and buyers have similarly had the opportunity to incorporate its anticipated benefits into their decisions. In economic terms, we have moved well beyond a theoretical T=0, when market participants did not know about the project, into a T+1 environment where the information is publicly available and market participants have had time to react.

The existence of 9,946 completed but unsold serviced apartments does not mean that the RTS has failed to generate demand. We do not observe the counterfactual and cannot know whether the overhang would have been larger without the RTS. Neither does the figure imply that those units will remain permanently unsold, since completed inventory changes continuously as existing units are sold and new projects reach completion.

It nevertheless provides a useful reality check when placed beside the amount of new supply under development.

Market position	Units
Completed unsold serviced apartments, H1 2026	9,946
Incoming serviced apartments	41,832
Planned serviced apartments	18,712
Incoming + planned	60,544

It would be incorrect simply to add the two figures and declare 70,490 units of oversupply. They represent different stages of the development cycle, some planned projects may be delayed or altered, existing completed units will continue to sell, and Johor will continue generating new households through ordinary economic and demographic processes. The more modest observation is that Johor already carries close to 10,000 completed but unsold serviced apartments while preparing for an incoming and planned pipeline equivalent to approximately 55.6% of its existing serviced-apartment stock.

Our transport analysis asks what the RTS could theoretically contribute to new housing demand through Singapore-to-Johor residential relocation. The completed-unsold figure tells us what the market has actually absorbed so far in an environment where the RTS story has already been known for years. The two numbers answer different questions, but they are worth reading together.

The RTS Can Move People. It Cannot Manufacture Them.

The RTS is a major piece of infrastructure and there is little reason to understate its significance. It should materially reduce travelling friction between Johor Bahru and Singapore, improve the predictability of cross-border commuting and persuade some Malaysians presently living in Singapore to move to Johor. It can also redistribute housing demand towards locations with better accessibility to Bukit Chagar, increase the relative value of genuinely scarce RTS-connected property and support deeper economic integration between Johor and Singapore.

What it cannot do is create households without limit. Johor presently has 60,544 incoming and planned serviced apartments.^[4] To test how much of that pipeline could conceivably be supported by RTS-induced relocation, we deliberately ignored the lower opening ridership forecast and assumed instead that the railway was already operating at its ultimate peak capacity of 10,000 passengers per hour per direction. We then allowed it to run at that maximum Singapore-bound capacity continuously from 6am until 9am, allocated every passenger to someone previously living in Singapore, assumed no existing Johor commuter occupied any of those places, converted every relocating worker into a separate household and placed every one of those households into a serviced apartment.

Even after doing all of that, the calculation produces:

The conclusion is therefore not that Johor Bahru necessarily has a property bubble, nor that the remaining apartments will sit empty. It is simply that the RTS story, by itself, is insufficient to explain the scale of serviced-apartment supply now coming into the market. The remaining demand must be supported by other sources, whether the JS-SEZ, local household formation, new employment, foreign and Singapore purchasers, retirees or wider economic development, and each of those propositions should be examined on its own merits rather than assumed into the calculation.

Ultimately, every property market faces a fairly mundane test. Developers may launch the units and investors may buy them; infrastructure can change accessibility and expectations can lift prices. But once construction is completed, somebody still has to occupy the homes. That leaves us with the same question with which we began: Who is going to live in all these homes?

Footnotes

^[1] Singapore Immigration & Checkpoints Authority. “Continuous heavy traffic expected at Woodlands Checkpoint.” Data shows 327,000 daily travellers in 2024, up from 269,000 in 2023. Accessed from ICA press releases and TAFEP reports.

^[2] MRT Corp. “RTS Link Project.” Official specifications: route length approximately 4 km; estimated station-to-station journey time 6 minutes; ultimate peak frequency 3.6 minutes; and capacity of up to 10,000 passengers per hour per direction. Source: MRT Corp official RTS Link project material, accessed October 2026.

^[2a] The sensitivity analysis assumes RTS passenger service is available from 6am. The operating-hours assumption should be checked against the final published timetable before publication if the operator subsequently changes the service hours.

^[3] Malaysia Rapid Transit System. Official ridership projection: approximately 40,000 passengers per day upon opening. Source: MRT Corp published figures and project documentation.

^[3a] MRT Corp publishes the 40,000-passengers-per-day figure but does not appear in the cited project specifications to expressly define whether it represents passenger journeys, boardings or unique daily users. Unlike an earlier version of this analysis, the present stress test does not divide this figure by two or rely upon any particular interpretation of it. The central calculation instead uses the independently published ultimate capacity of 10,000 passengers per hour per direction.

^[4] CIMB Research. “Johor high-rise segment faces oversupply risks.” July 2026. Data from National Property Information Centre (NPIC) Q1 2026: existing stock 108,863 units; incoming supply 41,832 units; planned supply 18,712 units. Published in The Star and The Edge Malaysia.

^[5] Malaysia Valuation and Property Services Department (JPPH). Property Market Report for the First Half of 2026. Released 10 September 2026. Johor recorded 9,946 completed but unsold serviced-apartment units as of H1 2026. Source: Malay Mail reporting on the JPPH H1 2026 report, 10 September 2026.

^[6] J.G. Wardrop, “Some Theoretical Aspects of Road Traffic Research”, Proceedings of the Institution of Civil Engineers, Vol. 1, Issue 3 (1952), pp. 325–362, doi:10.1680/ipeds.1952.11259. Wardrop’s first principle describes a user-equilibrium condition in which no traveller can improve his or her travel cost by unilaterally changing route. On stochastic user equilibrium under imperfectly perceived travel costs, see Carlos F. Daganzo and Yosef Sheffi, “On Stochastic Models of Traffic Assignment”, Transportation Science, Vol. 11, Issue 3 (1977), pp. 253–274, doi:10.1287/trsc.11.3.253.

This analysis isolates the RTS variable to test whether RTS-induced migration can explain the scale of Johor’s incoming and planned serviced-apartment supply. It does not forecast total market absorption or suggest that particular projects will fail. The peak-capacity sensitivity analysis assumes conventional daytime employment and does not attempt to estimate shift-worker demand. Its purpose is to establish a deliberately generous upper-bound baseline against which other sources of housing demand may be assessed.