Predictions for route amendments under BCEP are all guesstimates based on given information in LTA press releases. Take with salt, as we do not have authoritative information on final route plans to be released under BCEP later this year.

I know, I’m supposed to be away, preoccupied with other matters during this time period. Given recent developments however, it’s important to clear up the conversation about the health of our bus system before I slink back into my hole again until November. Do bear with me for a bit, as I explain how and why, the recent “bus improvement” initiatives announced do not sound as optimistic as LTA wants you to think it to be.

On 30 July 2024, LTA and the Transport Ministry collectively announced the Bus Connectivity Enhancement Programme, or BCEP for short, aimed at expanding the coverage of the bus network. With S$900 million set aside for funding these service increases over the course of the next eight years, it would supposedly arrest the decline of our bus network, and as Chee Hong Tat claims, increase bus ridership.

The BCEP comprises four key tenets, all of which were touted as supposed to the public bus system which had suffered greatly from repeated service cuts since the pandemic began.

In short, the four key thrusts of BCEP are, namely:

• Bridging short network gaps between MRT stations and housing estates
• Introduction of peak express services (branded City Direct locally) to relieve overcrowded MRT lines
• Commitment to linking new housing estates to MRT stations
• Introduction of “express feeders” to link major nodes to further-out new estates.

Before diving straight into judging this new policy, it’s helpful to remember that what’s not being said is often more important than what is, although they can be easily inferred with given statements. Given this, the BCEP doesn’t paint the optimistic vision of the bus network’s future that LTA wants you to see, and good reasons exist to be very skeptical of what’s being hailed currently as the next “big fix” for buses.

To give them credit, the branding for this new programme has been done in an extremely clever manner, by using a name greatly similar to the BSEP of last decade (2012-2017), an exercise that got the largely still-private and disjoint bus industry together with its act. Many critical of BCM-era rollbacks on bus services caused by cost-cutting on the ends of LTA and the bus operators would be more nostalgic for BSEP, when the rollout of more buses and new routes truly meant beefing up our bus network to be one of the strongest and most robust in the world. So much so that when BSEP concluded in 2017, our bus network had sufficient slack to barely tide us through the dark days of re-signalling the legacy lines, where train disruptions were prevalent, to the extent that outside experts called our bus network “duplicative” in horror. BSEP worked marvels for bolstering the bus network, and branding the current new initiative as a nearly-identical acronym to the roaring successes of the BSEP (in terms of hardware) is certainly a smart tactic by LTA to quickly curry favour from a population who remembers it as the golden era of bus service in Singapore. Of course, the devil lies in the details, and to very unfortunately pop their bubbles, BCEP is nothing like the old BSEP, aside from the similar acronyms.

话中之话

The BCEP promises are built upon a fanciful hodgepodge of vocabulary aimed at convincing the public that the sum of S$900 million will truly be invested towards a stronger and more “connective” network, supposedly by enabling connections, and hopefully representing a shift away from the unhelpful and inefficient hub-and-spoke model prescribed in LTMP 2040. On closer examination, it turns out that BCEP not only fails to undermine the unpopular hub-and-spoke network structure which LTA intends to impose on the public transport system over the next decade or so, it is also instead a facilitator of the hub-and-spoke model! Of the four key initiatives rolled out under BCEP, only one of them is not about feeders, and even that single non-feeder initiative falls prey to a fallacy that no transit agency around the world should be falling for in 2024. More on that later.

The first initiative highlights the enhancement of bus services (on a case-by-case) basis to “address changing travel patterns”. Little protip for reading official media: if the wording is deliberately vague, be wary, as this almost always means something is up. How are our travel patterns changing? Ask a thousand people this question and you’d get back a thousand responses. How does the LTA perceive shifts in our travel patterns, especially since they do have access to data that isn’t necessarily made available in the open-source Datamall sequence? Given that a conflict of interest exists between the LTA’s own manifesto (the LTMP, last updated 2019) and the thrust of adjusting bus services according to available ridership data, to what extent will the former inevitably affect the latter?

Where LTA’s intent is revealed is in the examples cited for all their initiatives. In the case of “enhancing connectivity to meet changing demand patterns”, the example amendment cited was “a service extension to connect Toa Payoh East and Caldecott MRT”. To distill this, it’s basically a connection between a new HDB estate and a MRT station. The TEL and CCL feel really far from Toa Payoh, but it really is not, with an existing Toa Payoh feeder (235) looping at Caldecott from Toa Payoh while still legally defined as a feeder service based on geographical boundaries. So what more is said “service extension” than just a mere feeder connection to the rail network? At best, the most optimistic interpretation of the first point is an intent to enhance long feeder services such as 141, enabling more efficiently-operated bus services but still without the benefits of a robust interconnective trunk network in place. Even so, long feeders, especially of the network filler kind intended to increase coverage, are by design ineffective at enabling faster journeys by bus. Not a convincing assurance that the BCEP will represent a meaningful improvement to better access, which should be the ultimate goal of connectivity improvements.

Ignoring the point on new City Directs for now, the third initiative advocates for bus connections to be made available to new housing estates sooner. This comes amidst a lot of controversy around inadequate bus coverage to new estates in upcoming towns (such as Tengah). On paper, these are best realised with connections offered by long feeder routes or trunk services making slight detours from arterial corridors that do not compromise overall travel time. Looking at the examples however, it appears a more feeder-based solution is favoured by LTA as part of the BCEP, with the example cited — a link between Yishun East (the area around Avenue 1, where future housing estates will be soon complete) and Khatib MRT. Those familiar with Yishun may recognise that this route profile most closely fits that of a short feeder, akin to 801 introduced in 2022.

Again, their intent is clear from the cited example — the primary tool used to achieve the goal of linking new housing estates will be feeders, rather than more useful long feeders or trunks that allow direct connectivity to far more destinations beyond just the town center. 

Last but not least, the most absurd initiative of the BCEP, for which LTA very helpfully gave zero further elaboration upon: express feeders, a supposed non-stop version of feeder routes that serve the role of expresses within individual towns. There’s two possibilities of what this could be, as “express” is a very vague and slippery word that is often mistakenly used to subsume the definition of “Rapid”-stopping services.

In the case of the former, where a “limited stopping” scheme is applied within the context of feeder services, I really wonder how the inspiration for such came from. Chances are, it’s a bastardisation of rapid-stop services that I proposed a while back, where stop spacing is wider but remains generally consistent. What are the chances that an “express feeder” is but just a two- or three-stop rapid service, acting as standalone branches off the rail network at ITHs? At such small scales that maybe cross only a single town boundary, rapid services are highly ineffective at bringing about significant fast connectivity. A possible example, for instance, would be 298 (Tampines North – Tampines West via Tampines Central) being converted to a three-stop rapid service calling once in each part of Tampines, as hinted at in the poster. Another possible application would be the conversion of 992 (Bukit Batok – Tengah) to be a non-stop relay service between the two towns, to link (some) Tengah residents to the NSL quicker.

The key advantages of rapid-stop bus services, as I said before, is their cross-network span enabling them to build a second “MRT-lite” network supporting the main radial rail network today. (Express services fall flat in robust network formation, and hence a less stellar choice than rapids for the formation of strong core bus networks). If LTA would so want to introduce mere fragments of rapid service, as could be a possibility described by “express feeders”, they would really be better off extending them to form network-spanning rapid services that offer disproportionately large returns relative to a minimal increase in operating cost.

Another interpretation of “express feeders” is the literal one — local-stopping feeder services incorporating a comparatively long non-stop sector in between to guarantee local access in further areas (as mentioned in the LTA poster) while making it “faster” for them. This is most prevalent in Hong Kong, where residential estates outside Kowloon and Wan Chai are predominantly formed out of isolated exclaves surrounded by vast undeveloped land, giving rise to an iron-strong hub-and-spoke system formed out of express feeders travelling vast distances between these communities and the nearest MTR station, which can be as far away as 5km or more. Express feeders are formed as a necessity due to the highly cul-de-sac nature of suburban Hong Kong which makes it highly ineffective and money-losing to through-run trunk services to these estates, unless as a starting point.

This urban form comparatively unique to Hong Kong (super dense exclaves surrounded by nothingness, linked to main transport networks by a thread) also gives rise to a completely different attitude in operating public transport, heavily centered upon the peak CBD-bound commuter, resulting in cycles of reinforcement built in both public transport service and urban planning. (Good luck with the Northern Territories project to better integrate with Shenzhen) It’s why Hong Kong’s bus operators are fundamentally incompatible with meeting the public transport needs of Singapore (or even Mainland China), and why it’s an absolute godsend that HK-based Bravo failed to clinch the tender for the Seletar Package’s 2nd term earlier last month. A public transport system built upon peak commuter expresses and express feeders in a hub-and-spoke system yields highly wasteful uses of resources for two brief, acute peaks and lackluster service outside of it at best. If we want to take decentralisation seriously, and capitalise upon our unique city-in-city urban planning model built up over the decades, the HK model is something we must steer clear of.

Singapore also has a few “express feeders” of this sort in operation, although the effects of the express sector are minimal, and they’re officially not branded as such. Some Jurong industrial services, namely 194, 248 and 254 incorporate non-stop segments for part of the route to avoid duplication with other JIS routes and carry workers to their factories faster. Of course, this arrangement, and that of the JIS system in general, is not very sustainable when resources are scarce, because its hub-and-spoke structure lends itself to the buses being empty more than half the time thanks to uni-directional demand by design. 

Either way, the direction set out in either proposition of “express feeders” is still unmistakably that of a hub-and-spoke system, as outlined in LTMP 2040, putting greater emphasis on feeder connections to main (typically rail) routes at ITHs, rather than a balanced approach that takes into consideration access, which is the defining prerequisite of the connectivity that BCEP sets out to achieve! Both methods of achieving express feeders only reinforce the inefficient and time-consuming radial network. If this is what much of the $900 million invested is intended for, the money should have instead been put towards a more robust and connective system built around trunks and long feeders (eg. 18, 60, 73, 99, 119, 962, 983) which offer direct connectivity to a wider range of destinations! 

Last but not least, the second initiative of BCEP which hasn’t been covered thus far: peak expresses aimed at alleviating congestion on overloaded rail lines. Confirmed, are new City Directs for Punggol and hopefully other northeastern towns aimed at easing the crunch on the NEL. Finally, LTA is somewhat tackling this issue head-on, although the BCEP approach still leaves some to be desired. For one, the congestion that is experienced on the NEL (and will continue to grow with 10,000 flats planned for Fernvale and beyond) is experienced throughout the day, to the extent that one struggles to tell the difference between peak and offpeak crowding at first glance. I’m also sure that more peak expresses bound for the CBD only wasn’t the answer that Sengkang GRC MPs were expecting (for their constituents with diverse travel needs) when Chee Hong Tat slapped the BCEP in their face in his reply. Meaningful solutions to the perennial issue of northeastern congestion, bar the introduction of completely new rail lines there, will have to be full-day options to enable a general increase in all-day carrying capacity. 

A fallacious proposition, circa 2024

Together with the emphasis on a feeder-based bus system, the CDS enhancements without corresponding improvements to the trunk network represent a planning fallacy that has no place in 2024 transit planning. Collectively, the two major directions of BCEP form a radial hub-and-spoke system with services that are geared primarily to the peak commuter, rather than the variety of other profiles that ride public transport. Whether it’s introducing feeders (or its express variants, whatever it is) to send passengers to the rail network, or enhancing the CDS system to send passengers directly from home to Shenton Way, both make the grave mistake of assuming the CBD is the point of primacy in the transport network, and by extension the focal point of our society. That ship has sailed for more than 30 years now, and any visit to Maxwell Road would show the myriad of initiatives to establish new centers of business, commerce and innovation in new towns far from the CBD. Why isn’t the BCEP talking about enhancing connectivity to these new areas? Furthermore, express services, be it CDS or express feeders, are inherently peak-oriented due to their exclusionary nature (or as Jarrett Walker would have put it, the hazard of specialisation). It’s a gross underrepresentation of various groups who use public transport outside the peak, such as students, freelancers, blue-collar workers (whom create the unrecognised third peak phenomenon) and people headed for recreational activities! 

Correct me if I’m wrong, but the BCEP fails to guarantee an enhancement to the main trunk bus network the way BSEP did last decade, which is the crucial bit needed to truly achieve the effect of a rising tide lifting all boats. Other than a very vaguely-phrased mention of “service adjustments to meet changing demands”, little in the BCEP’s initiatives suggest improvements planned for the trunk and long feeder network, which is what really guarantees connectivity in public transport, not feeders or peak expresses. And with no guarantee that trunk routes will be protected, comes the next of my worries regarding BCEP. 

Between the fine print

The naming of this campaign is rather fishy. Yes, it’s called the “Bus Connectivity Enhancement Programme”, and it’s supposed to sound like an enhancement to bus services. 

That’s where you have to be cautious, because this is where exact wordings matter. This point is best made by contrasting it to the last major bus enhancement project — the BSEP, or Bus Service Enhancement Programme. Here is where the questions start cropping up. BSEP, which targeted bus service, was all about improving the quality and quantity of bus services nationwide leading up to the introduction of BCM (a different can of worms, and another story altogether). 

How does one achieve better bus service? Buy new buses, expand the fleet size. Use the bigger fleet to run existing routes more frequently. Establish new routes too, for redundancy and connectivity purposes! The central focus of BSEP, come to think about it, was really the acquisition of additional buses and infrastructure to enable much more bus service to be provided. That’s how the robust network pre-pandemic came to be, notwithstanding decades of revisions to the network that brought it to its present state. And it’s not rocket science — the BSEP’s slogan was literally “More Buses, Better Rides”! The goal of BSEP was to improve bus services from its 2011 baseline, and by its conclusion in 2017, this goal had been met, even if we consider it mediocre by today’s standards. 

Connectivity, on the other hand, is a trickier word. Unlike service, the focus of BSEP, the key word of BCEP says nothing about the extent to which its beneficiaries can rely on the service as a viable part of a car-lite lifestyle. Instead, the emphasis is on coverage, which takes the perspective of the network structure rather than the passengers’ experience. Rather than measuring how quickly the passenger gets to their destination, it’s about how much of the Singapore map gets covered in lines representing bus routes, with nothing said of their frequency that actually matters to its riders. 

It’s also tougher to measure outcomes for connectivity rather than service. For the latter, we’ve devised over the years many (imperfect and in need of reform, but still) metrics to measure bus service. The BCM-era BSRF comes to mind, and other metrics such as effective waiting times, average journey times also exist to give us a picture of how bus service is provided to the public. But how do you measure connectivity? You can measure access, which is determined heavily by service, and tools like isochrone maps are your best friend for them. Connectivity, which is determined by your network structure (ie radial, grid etc) and how its elements interact (ie connections between routes, or lack thereof) is a lot harder to measure. This is a difficult task, and one that’s arbitrary and meaningless to pursue without greater goals of access in mind. 

That’s the attitude that I’m getting from many of the initiatives under the BCEP — they introduce connections for connections’ sake. Rather than considering possibilities of enabling quick access to likely key amenities, the modus operandi of many BCEP “enhancements” is simply just directly connecting residential estates to MRT stations in the vicinity, resulting in heavily feeder-based solutions. LTA’s policy team is mistaking coverage for access, an arguably novice-level error that most respectable transit systems in the world make clear not to commit. Different writers in the field of transit theory and advocacy refer to this by different names — Reece Martin calls it the “it’s served” fallacy, while Jarrett Walker explores the “new route problem” (which I’ve also written about before, regarding feeder 801). 

Of course, in the BCM era, expecting a total network overhaul to properly incorporate these new areas which did not exist in the last Network Integration Exercise (1989-93) is unfortunately unrealistic today. That ball is no longer in LTA’s court; instead in the collective hands of 93 MPs who make for very shortsighted bus network planning. It’s why there’s been a recent trend of increasingly useless and unhelpful bus routes pushed out — it will get worse during BCEP as each MP fights for what a certain rival blog derides as “bus-to-order”. Some of these feeders introduced under BCEP will definitely be much more useless than even 146, failing to garner sufficient ridership, but kept alive due to the direction of BCEP. 

Where doth the money come from? 

The 900 million dollars from the Finance Ministry, comprising a mix of our taxpayer money and bits from our reserves. Okay, that’s not the real question, but glad that you’ve read till this point. 

The real question: From where will the resources to operate BCEP routes come from? Note that BCEP entails an expansion of coverage, without an apparent corresponding increase in fleet size. Assuming no change or redistribution in resources, an increase in the bus network mileage inevitably means a reduction in service quality across the board. It’s possible to maintain service for some key routes, but that just means hitting others even harder, which still translates to an overall net reduction in service across the board. Worse yet, unlike bus rationalisation exercises that we’ve sat through over the years, there is no rail expansion to brusquely justify such an act with the promise of “faster rides by train”. 

This to be the most likely outcome of BCEP, as there has been zip mention of additional bus fleets brought in to sustain the new routes to be introduced. If there were intent to buy additional fleets, this would have been accorded equal status to the four aforementioned initiatives of BCEP, because this is particularly crucial in ensuring the delivery of the former without compromising the existing bus network. At least, that’s how a competent sales pitch for BCEP would have worked, provided the intent for further fleet expansion is there, and not just a mere footnote within the Straits Times article!! Needless to say, said footnote is also left deliberately vague, so it’s anybody’s guess whether actual extra fleets will be ordered for new BCEP routes, besides existing replacements for outgoing legacy buses. 

The failure to mention fleet procurements as part of BCEP means either one of two things. Either the team responsible for promoting BCEP is simply terrible at communicating what’s important to the public consciousness, or worse still, no plans for further fleet acquisition exist for BCEP! Which is why it’s all the more important to ask the question: From where do your buses come from? 

Please do not! 

Important questions, as a modern Chinese adage goes, must be asked thrice: From where do your buses come from? Uncomfortable answers result when this question is probed. 

With signs that LTA is unenthusiastic about further bus purchases after fatigue from mothballed buses and their policy fallout, the logical conclusion will be that the buses needed will mostly, if not entirely come from the existing 5,800 or so buses currently in use. Sure, a handful of rigid buses remain in storage to this day, but they are definitely insufficient to meet operational requirements of all BCEP routes, given the scale and scope that has been set out. The uncomfortable reality hidden behind the blissful marketing for the BCEP is that buses will have to be poached from somewhere to support the BCEP’s feeder-based initiatives.

In tandem with the LTA’s biased attitude against trunk services, the unspoken roadmap that BCEP outlines is clear — in store for us public transport passengers is an acceleration of the network’s collapse into a radial hub-and-spoke structure. When push comes to shove, and when buses are needed to run LTA’s fancy feeder toys, trunk services will be the go-to for their proverbial axes. Two birds in one stone, by destroying the trunk bus network to strengthen feeders. How convenient, a reallocation of public transport service from a more diverse profile to favour the peak CBD-oriented white-collar worker. What public transport is there for those working elsewhere, at other times then? 

A last word, on political fire

For those who are still unaware: Election storm is brewing, even if the political parties keep silent about it. With such a context surrounding the BCEP’s announcement, there’s always reasonable suspicion that the launch of the BCEP is politically motivated to score points in the coming general elections. Bolstering this theory is the fact that Chee Hong Tat, the current transport minister, is simply too unpopular due to his utterly incompetent handling of public controversies around the 167 service cut and the forced introduction of SimplyGo. (And, if Reddit stories are to be believed, he is personally a vile character — numerous accounts exist of his arrogance and lack of intellect while helming previous jobs in the civil service) Is there not a possibility that the BCEP was a hurried initiative launched to curry favour for the PAP to prevent negative sentiment over the decline of public transport in the country from affecting their performance, akin to 2011? 

That was the daredevil question a reporter posed to the Minister at the press conference of BCEP’s launch. Obviously, he denied it, but like with every official statement coming from MOT and LTA nowadays, timing is a conveyance of intent. 

It’s known that BCEP is really a thinly-veiled attempt to reorient the public bus system in Singapore to favour the peak commuter even more, at the cost of the multi-destinational trunk network that serves diverse travel needs. Knowing this truth about BCEP, it’s intellectually dishonest to attempt branding it as an improvement to bus services (Note the fallacy again: equating coverage with service. They’re linked, but one doesn’t automatically mean the other)

To attempt to brand what runs against the evolution of public transport as an “enhancement” is simply an attempt to sway less informed members of the public into believing that buses are improving when LTA’s own manifesto clearly states the intent and methodology to take the bus network in the other direction! Is this not the definition of political motivation, considering the time at which BCEP was released? Of course, the chickens will come home to roost, and BCEP will bite public transport users hard one day when it has sufficiently destroyed the gridded trunk bus network. But that’s still some years from now, while the false promise of improved bus services is now, and it is now that the gullible fall for the sweet talk from Chee Hong Tat and his ilk. You decide if the BCEP is a politically motivated act. 

The things that matter

In a shellnut, the BCEP is not a cause for celebration, and should be treated with the same suspicion that we view acts of bus irrationalisation over the years. Service cuts caused by opening new rail lines alone are insufficient in deleting the robust trunk network in favour of the radial hub-and-spoke system explicitly favoured in LTMP 2040. An approach that blindly favours feeders, rather than a rational plan comprising trunks and long feeders that offer real connectivity, is a plan set up to fail, even by the goals of BCEP itself. The promised connectivity of BCEP will NEVER be realised if only feeders and peak expresses have a place in the project, and should trunk services have to be inevitably cut back or deleted to run these new feeders and “express feeders”, LTA will stray even further from these goals. 

Let’s recognise the BCEP for what it is — an accelerator of bus irrationalisation, driven by hub-and-spoke ideology, a programme of ill-founded deference to the belief that buses are second-class transport in Singapore that has characterised LTA’s attitudes* towards buses since the pandemic began. If you value being able to go places regardless of public transport mode, the BCEP represents an affront against our “evolving travel needs”, to quote its own promotional material. 

And Mr Chee, nice try trying to fool the public with this programme of yours. Perhaps the less informed would fall for this programme to embark our bus network on its slow death, but for people like us who have been studying transport affairs long before you ascended the ministerial post, we know what’s fishy and what’s not, and programmes like BCEP simply fail the sniff test. Just because you claim it’s not politically motivated, doesn’t make it so. 

Such are the uncomfortable truths about the BCEP. It’s not the big fix that policymakers imagine it to be. It never was, and it never will. Not unless the BCEP’s guiding principles and initiatives radically change course, that is. And that would require an understanding of what “connectivity” and “enhancement” really mean, in the context of our transport network.

Now that I’m done here, off to my other things I go. See you all in November. :>

*Or at the very least, the attitudes of their top bosses whose subordinates are made to do by their diktat. 

Join the conversation here, and don’t forget to press the like button! Thanks for reading STC.