Journey to the West

If you live in the West and care about public transport because you ride it daily (or selfishly want emptier roads to drive your cars in), the big day that you should be taking note of is here.

Just released, on Friday, 18 Aug 2023, were the results of not one, but two bus packages’ tender results, for an effective term from 2024 to 2028. The Jurong West Bus Package, currently based in Soon Lee Bus Park (SLBP) and the Bukit Merah Package hailing from the current Ulu Pandan Depot (UPDEP), currently both operated by incumbent SBS Transit, will encounter some major operational shifts within the following year. In the case of the Jurong West Bus Package, operations will be gradually shifted to SMRT Buses in September 2024.

The good news here which much of the bus community is celebrating, is SMRT’s first win in the entire BCM’s history since the first competitive tender in 2015. We look forward to SMRT’s higher-quality service being provided to workers in the Jurong industrial area for, at least, the next 5 years. As one of those who previously mentioned that SMRT was being sidelined in BCM, now’s the time for me to eat my words as I watch everyone celebrate this momentous first win for a once-thought-to-be-done-for bus operator. At least, we are assured of SMRT’s presence in the industry for up until 2028, up from the previous 2025 (tender-out date for the Choa Chua Kang-Bukit Panjang package).

Alas, the good news from Friday pretty much ends there, because while SMRT winning its first package in the BCM is worth celebrating, the operational implications of such a win for bus service elsewhere is certainly not, and in fact should be a cause for concern, be it in the short or long term. I’ve previously written a (pretty cringe, looking back) similar reaction piece when TTS won the Sembawang-Yishun Bus Package in 2021 to highlight the flaws of BCM, and while that piece is getting overhauled, much of the issues that surface in this newly-announced tender for the two bus packages are in fact problems intrinsic to BCM and similar competitive tendering schemes. Examining the implications of these bids on future operations hence makes for a good case study for troubleshooting our bus industry, if we even intend to do so seriously in the forseeable future.

Before we start, here’s a quick overview of shorthands used when discussing the Jurong West and Bukit Merah packages, to prevent confusion:

• BCM – Bus Contracting Model
• GCM – Gross Cost Model, of which BCM is a subset of
• UPDEP1 – Ulu Pandan Bus Depot (current site opposite Tradehub 21)
• UPDEP2 – Ulu Pandan Bus Depot (future site co-located with Ulu Pandan MRT depot expansion)
• PPDEP – Pasir Panjang Bus Depot (recently announced)
• SLBP – Soon Lee Bus Park
• BBDEP – Bukit Batok Depot
• HGDEP – Hougang Depot
• JWP – Jurong West Bus Package, based out of SLBP
• BMP1 – Bukit Merah Bus Package (1st term, effective November 2018 – September 2023), based out of UPDEP1
• BMP2 – Bukit Merah Bus Package (2nd term, effective September 2023 – 2028), based out of UPDEP1
• CLP – Clementi Bus Package, based in BBDEP, then subsequently UPDEP2 from 2029
• SYP – Sembawang-Yishun Bus Package, based in Mandai Depot since 2021

Show me the money

Before we dive headfirst into the rabbit hole known as BCM package tendering, it is important to note that the essence of BCM is just the financing framework of the public bus industry that LTA aspires to transition all of Singapore towards by the end of the decade. The gist of it is just to convert our bus system from a national franchised model, where operating rights are directly conferred to state-owned operators by locality, towards a “free-market” competitive tendering model. Ideally, it seeks to reduce the x-inefficiency (basically: bureaucratic crap) of the industry by forcing them to operate more leanly to stay competitive with foreign players who dominate far greater markets overseas coming in for their slice of the pie locally. It’s a bit like an auction, where aspiring players bid for the item being sold on stage with higher prices willing to be paid, except with a twist: Competitive tendering favours the lowest bidder, especially in the context of BCM. Why is this so? It’s all about the money, like they say.

In any case, it is not hard to see the first fatal flaw of a financing framework that incentivises the lowest bidder more in a more or less open competition: it encourages a race to the bottom, where bidders (i.e. your aspiring bus operators) put out ever lower bids in an effort to one-up each other for better chances of winning the next tendered package. In more economics terms, it simply means that competitive tendering fosters a culture of price warfare, a price war being one where one firm cutting prices drastically leads to others following suit in order to not lose market share.

It is thus important to understand, even as a passenger, what this “price” being bidded for entails. Under BCM, a form of GCM (gross cost model) or gross cost contract model, operators are paid a fixed sum (hence, “gross cost” in the name) for their operations across an agreed duration as stipulated in the package contract. This fixed sum will cover all expected costs that the operator is expected to incur across the contract period, including manpower, maintenance, asset management (although this job is now mostly borne by LTA), internal accounting, and of course, as a private firm, expected profit margins. This runs under the assumption of course, that the operator has no other source of income which is capable of cross-subsidising their bus operations in the particular package in the event of inability to break even.

Well, when we look at the tendencies for price warfare to happen under such a financing framework as BCM, this inevitably leads us to the first long-term problem: can these funds tide them through the full period?

Have a look at the bidding information for BMP2 and JWP, as shown below.

Unlike past packages, where the lowest bidder need not necessarily be the one awarded the bus package, the tenders for JWP and BMP2 were clearly awarded to the lowest bidder, and for the matter, the margins for which the winning bids won out over their competitors were quite significant, suggesting a certain extent of lowballing prevailing in the industry in the mid-stage of BCM. In the JWP, SMRT underbidded the next lowest option by $10 million, while SBST underbid the others by a full $20 million for BMP2. A culture of underbidding just for the sake of winning the package is afoot, and I don’t see it to go away anytime soon.

In particular, this is the first time (since the Seletar package in 2018) where the winning bids for bus packages were also the lowest bids. Despite the margins being far larger for previous packages, the lowest bidder did not necessarily win the package, unless in extreme cases. (SBST for instance won the Seletar package in 2018 with a staggering $35 million win margin over the next lowest bid) Even TTS’ controversial win of Sembawang-Yishun in 2021 arose from a bid that was third-lowest among all of them. It seems to suggest that the tides of bus package awarding criteria have truly changed, to give more weightatge towards the amount bidded, rather than other operations performance standard criteria, as LTA claims to be using for BCM evaluation.

Of course, it’s a happy moment when your company wins a new package, or gets to retain an existing package for a further period of time. However, with the limited resources that have gone into the bid, free lunch hardly comes easy. Remember, the BCM operates on a gross cost model, hence lowballing the tender means there is less to go around, which can become a particular issue a few years down the road. Consider two existing cases where aggressive underbidding have come to bite the operator hard in the neck, the Seletar package and BMP1 (how ironic).

Though SBST managed to secure BMP1 not entirely through underbidding (a joint consortium between Travel GSH and Jiaoyun Express bidded a full $60 million lower, in what I think of as a pretty suicidal move), its aggressive price strategy in bidding for BMP1 has cost it dearly, especially as revenue streams collapsed during the pandemic. As SBST tried to get things going again after the pandemic, it quickly ran into a whole host of funding issues that forced it to adopt unpopular cost-cutting measures as financial damage control. This included tightening schedules unacceptably, where BCs were only given eight-minute breaks at the non-anchor end of some long trunk routes, cutting at mealbreak durations, and tightening runtime to reduce manpower and capital operating cost. Of course, bus service levels took a significant hit too, which passengers felt. *cough cough service 5* Nonetheless, that still did not stop the death spiral, which allegedly got so bad that internal management was desperately seeking an off-ramp to minimise losses, which makes it quite painfully ironic that they won the tender for BMP2. Good luck and have fun managing those finances in the deep reds :skull:

Over in Seletar Depot, the effects of underbidding (which was actually indeed the case, with a $35 million margin to the next bidder) were far more pronounced.

With a looming manpower crisis as the public transport industry exited the shadow of COVID-19, SEDEP management adopted some serious austerity measures (which BCM encourages especially when things turn rough) which created even more resent among both staff and passengers. To cope with fewer BCs out and available to drive buses, duties were dropped, resulting in longer wait times for passengers, and at the same time, runtimes (especially for trunk routes, though some feeders also bore the brunt at some point) were tightened, further exacerbating the runtime-safety contradiction and narrowing that delicate tightrope that BCs were walking. All in all, its safety record this year and the last hasn’t been particularly great, precisely due to an overworked workforce causing more errors in fatigue, and stress being placed by various levels of management. While layover times have not been as terribly affected as it has been for UPDEP1, mealbreak time took a greater hit, which some BCs spoken to viewed as a direct insult to their dignity. Of course, we do not exclude the middle management’s attempts to scrimp on further operating expenses by finding various (and often unreasonable) excuses to book bus captains for minor offences, thus carrying financial penalties on their salaries. (Translation: less money needs to be doled out for employee salaries) It’s no surprise thus, that turnover rates are up — resignations in SEDEP are constantly aplenty, which forces management to seek even more damage control measures that only make existing staff even more unhappy. It doesn’t help, that the Seletar Package, which was already financially constrained by significant margins due to this exact practice of aggressive underbidding from 2018, received a two-year extension to continue operations to 2025. (The Package’s 2nd term would otherwise have been called for tender this year too, alongside JWP and BMP2) While the BMP2 will likely get some financial respite (at least in the short term), its Seletar counterpart will have to continue bleeding money (and face an uphill managerial battle) for another two years until the start of its next term.

The post on long trunk runtimes, in fact, was directly inspired by the effects of SEDEP tightening runtimes on Service 851 causing a whole host of safety and welfare problems for staff and passengers!

Back to the JWP and BMP2. If these packages that had weathered the pandemic are any indication (the SYP awarded in 2021 is also facing a serious crunch currently, although they appear to be pulling through well), we should expect capital shortages to plague the operations of these two packages which were won through an intense underbidding exercise, especially as we move closer to the end of their tendered terms, nearer the end of the decade. In particular, much of the fixed sum awarded for BMP2 will likely go into fixing up the financial deficits left behind by BMP1, which means that if SBST doesn’t find some other way to quickly cross-subsidise the remainder of the BMP2 tendered term, it will be in for a wilder ride than anything we’ve seen it do so far, and could well put a premature stop to the 50th anniversary festivities that I don’t see them dialing down the hype on anytime soon.

As for the SMRT bid for JWP, it remains to be seen how they’ll manage to hang on for all 5 years with the meagre sum that they have bidded. This is particularly interesting because if one examines SMRT’s bidding patterns for past tenders, they have consistently bidded in the upper half of all bids submitted, asking for far more than their peer competitors. You’d have to go back all the way to the BCM’s first ever tender (Bulim 1st term) to find an instance of SMRT bidding low enough to be seriously considered as a possible finals candidate to award the tender to. Especially when you consider SMRT’s bidding patterns in 2018 and 2021 for BMP1 and the SYP / Bulim 2nd term: SMRT’s bid came in pretty high on the list, which almost certainly meant an immediate discounting from the decision-making process when selecting the tender awardee. Such a high price requisitioned by SMRT in those bids suggested a financial shortfall that it was unable to make up for had it chosen to lowball like other players had, and corroborated by SMRT’s internal austerity policies which are visible to passengers who frequent bus interchanges — its passenger service offices at non-anchor bus interchanges have been fully emptied out, leaving just a card reader, monitor screen and keypad, for bus captains to tap on and off their duties. What then, causes the financially liberal attitude it demonstrated when underbidding for the JWP? Regardless, I wish them all the best operating a new bus package which they are not permitted to do dual depot operations with, and hope they have truly indeed found a way to magically lean their financial expenditure so significantly, a lowball would not signal financial suicide for SMRT Buses. Even if it did, I would not worry too much, as SMRT has many other alternate sources of income which could potentially be used to cross-subsidise the JWP in the event it financially flops. That is, if the overall economy is well enough to ensure those shops in MRT stations thrive by then.

Tracing cashflows

The sudden shift in SMRT’s bidding patterns is also worthy of further examination. What gives them the gall to take that leap of faith to lowball all their tender options to the extent where usual cost-cutters like TTS are unable to compete with?

We must first recognise two major truths about the BCM framework. One, that aspiring bidders are unable to know their competitor’s bids due to the closed-envelope nature of the process, and hence all price strategy trends observed from the implementation of BCM only become apparent after a few rounds of tendering. Two, that as BCM tenders out a one-off, fixed sum, the amount bidded by the aspiring bidder necessarily reflects the strength of their economic base, or their pre-existing financial reserves, in an inversely proportional manner. That is to say, the stronger your economic base, the lower you’ll be able to afford to bid, since you’re more insured against financial risks that come from operating the contract.

Thus the ability for SMRT to bid low so suddenly, bucking their high-bidding trend of many years, suggests either that SMRT has the ability to foretell the moves of its competitors in the JWP and BMP2 tenders, or that it actually has bolstered its economic base sufficiently to permit it to take the risk of asking so little from LTA for the next 5 years of operations. The former would implicate SMRT of corporate espionage, something I believe a bus operator with more than four decades’ worth of prestige and branding would not resort to just for the sake of winning one or two bus packages. Rather, it’s a greater probability that SMRT has found enough cash from somewhere to somehow be able to support its long term operations in Jurong West (and Bukit Merah). The question then is, where did said cash come from, considering SMRT has not drastically changed operational practices from 2021 to now, in the post-pandemic era?

One possible theory was that SMRT earned a fair bit of cash from “pawning” its OC500LE assets to LTA, hence bringing the above-age OC500LE fleet under the fold of the BCM, which paved the way for them to be transferred to TTS during the SYP handover in 2021, even though similarly-aged assets in the legacy fleets (Scania K230UB, Volvo B9TL CDGE, SMRT-plated Citaros, Batch 1 WEGIIs etc.) continued to remain under ownership of their respective legacy operators despite changes in operator boundaries since 2016. After all, SMRT had previously put up an Expression of Interest (EOI) notice for the sale of its OC500LE and SMRT-plated Citaro buses, which total 185 buses, although nothing came out of said EOI, while a total of 48 OC500LE buses were transferred to TTS under SYP. It should be noted that such a relatively small once-off “pawning” exercise wouldn’t have made much of a dent on SMRT’s financial figures, especially as such events would have taken place around 2021, giving sufficient time for whatever financial benefits earned from it to be exhausted on two years’ worth of daily operations. Also happening around the same time is the “refurbishment” of batch 1 A22 buses initiated by SMRT, which itself is a capital-intensive exercise, especially with the additional cost of repainting the buses to the hideous (bright) Lush Green livery associated with the BCM. Again, selling off the 20 Comet Maxi buses would not have made much of a significant income to put SMRT in a better place to underbid the JWP by a full 10 million dollars. Where does SMRT gain its financial confidence from? It’s hard to tell, and maybe time will, eventually. For now, let’s hope that SMRT knows what they’re doing with such limited resources available to run bus services for the full duration of the tendered term, else passengers of the far west will truly be in for an unpleasant surprise, since the JWP by its conception inherently has a monopoly over all bus routes originating west of Boon Lay (except 99, which belongs to the CLP).

move house lo

The JWP’s handover to SMRT from the incumbent SBST also heralds a major change in bus deployments to be expected. Critically, many cost-saving links that are maintained by the current SLBP under SBST’s charge are expected to be lost, with huge ramifications on bus service that are expected to be felt islandwide, and not just in the west.

It must be understood that SBST’s dual-depot operational practices have enabled great operational savings for it, which is why such a practice is extensively done where possible, especially for long trunk routes that are resource-intensive and span vast distances. Being located at the western end of Singapore, SLBP under the SBST period (as well as its predecessors before SLBP opened in 2001) formed dual-depot links with almost every single legacy and modern SBST depot due to the large number of trunk routes that terminate at SLBP-controlled bus interchanges. Below is just a smattering of the extensive dual depot that SLBP practices:

• With BNDEP: 30, 197 etc.
• With HGDEP: 51, 105, 154 etc.
• With BBDEP: 99
• With UPDEP: 198
• With AMDEP: 52

However, such extensive and interdependent networks of dual-depot operations for long trunk buses is also extremely vulnerable to sudden BCM boundary changes that disrupt such dual-depot connections. It is precisely the number of such operational links that need to be severed during the process of the JWP transition that will bring about immense operational headaches for SBST. On a side note, the lack of an alternative bid submitted by SMRT for any of the three bid options unfortunately means that SLBP will be operated, relative to the other two existing SMRT depots, like a silo, operationally in its own bubble despite dual-depot options being physically available at Boon Lay Int (in the form of services 178, 180 and 187, all of which could use some help in the form of shifting operations to a depot closer by)

Back to SBST. To just give you a sense of how disruptive the transition of the JWP will be, here’s a list of all bus service that either fully or partially carry out garaging operations at SLBP currently. Routes denotes in yellow indicate routes displaced as a result of the JWP transition.

Full Control	Partial Control
174/e 179 181/M 182/M 185 192 193 194 199 240/M 241 242 243G/W 246 247 248/M 249 251 252 253 254 255 257 258 405 502 974	30 51 52 99 105 154 157 197 198 506

Tl;dr: f**k trunks, feeders all the way

Which is to say, how about a big fat f**k you to long trunk routes to the west. Among them, it has been more or less confirmed that the fate of the bus services marked in yellow above is either a return to garaging at the not-so-nearby BBDEP, or full single-depot operations by the other bus depot. Either way, SBST loses big in terms of additional deadhead mileage incurred, but that’s exactly BCM working as intended, a fragmented bus industry split by boundaries defied by immobile capital assets. Where’s the supposed operational efficiency in this, that BCM was promised to bring about?

To be frank, it would have been far better had SMRT won BMP2 (on an alternative bid) and SBST retained the JWP instead, as the positioning of the BMP routes and infrastructure closer to the city allows for far more opportunities for SMRT to conduct interlining and dual depot operations with existing citybound routes such as 190, 75 and 960, all of which could perhaps benefit from having some slack lifted off the shoulders of Kranji (and Woodlands) Depot that currently has to tank the full fleets of all these routes. SBST on the other hand, would have continued to enjoy the benefits of dual-depot operations with long-haul trunk bus routes crossing from east to west, enabling these resource-intensive routes to be operated in a more efficienct manner through reduction of deadhead mileage.

clp getting mad clapped

Services 99, 174/e and 197 in the above list belong to the Clementi Bus Package, which is based out of BBDEP. Some speculation also exists that Service 30 and 506 would also transfer the SLBP portion of dual-depot bus operations to BBDEP after the JWP transition. It sounds like an easy way to get out of the issues that the JWP transition has brought us, until you recall that just like most other legacy SBST depots, BBDEP is no exception from being severely short on parking space.

Placing these resource-heavy services under the direct purview of BBDEP would inevitably put heavy burden on its ability to store buses overnight. With BBDEP already at capacity with existing operations (which is why stabling for some CLP routes had to be outsourced to SLBP in the first place), attempting to cram more buses in as a result of consolidated operations would likely cause a further chain of bus displacements around the other depots, all of which do not bid well for the operational efficiency of our bus routes. In any case, BBDEP’s location (despite its proximity to the PIE) is not the most ideal for supporting bus routes that start from Boon Lay. In coming months, we could potentially see a comeback of a practice from SMRT’s distant past — overnight parking at bus interchanges, to cope with the excess buses unable to be stored in the depot. I suppose if that helps reduce deadhead mileage….

mammoth in the room

Though the BMP2 and JWP (and if you must, drag CLP into this) are west side depots, the ramifications of a major change of hands for SLBP has ramifications far beyond. In particular, we must think of the smallest legacy SBST depot, who now faces a critical capacity shortage as a result of the JWP transitions.

Hougang Depot, located in the Ubi industrial area, near the Kim Chuan MRT depot, is currently a bus depot with an estimated capacity of 350 buses. However, their service catchment area is the widest of all legacy SBST depots — it stretches all the way from Sengkang in the north, to Eunos in the east, and Geylang in the south, all of which are “anchor” interchanges and bus services handled by HGDEP. Currently, HGDEP operates at near capacity, which is why its hands are so tied up with regard to capacity upgrades or frequency improvements on the routes it manages (especially if they’re not dual-depot routes with other bus depots that have more room to spare). With so many routes to manage (easily twice as many distinct routes being operated fully and partially combined compared to SLBP), HGDEP gets barely by by “slacking” in dual-depot operations — they tend to take up the minority of buses contributed in dual-depot operations, to reduce the number of buses they need to put out for each distinct route.

Unfortunately for them, such days are over. With the transition of the JWP to SMRT, the decision has been made internally to convert all HGDEP services currently sharing fleets with SLBP to be fully housed at HGDEP, which refers to the three long trunk services 51, 105 and 154. According to leaks, all buses from these three services will park in HGDEP with effect from early 2024. To give you an idea of how much a disaster this is going to be, note that the full fleet size of service 51 is 32 buses, service 105 has 27 buses, and service 154 has the largest fleet of any bus route in Singapore, coming in at a whopping 43-bus fleet, even larger than that of the cross-border services! Combined, they occupy nearly 1/3 of the entire HGDEP’s parking capacity, which is frankly ridiculous considering the many other routes that HGDEP still must take care of at least until the Sengkang-Hougang and Serangoon-Eunos packages tender out in 2024 and 2025 respectively.

Did a quick check, and apparently service 154 had a few buses cut from it sometime this year. Currently, only 39 buses are needed for this route, but that still puts it in 2nd place for largest fleet of a single route, coming in after service 190 with 42 buses, and tied with the combined fleet of service 170/X. Nonetheless, it’s still a mammoth of a service that SBST will have an amazingly easy time (ahem) shifting over from SLBP to full HGDEP.

With such a massive consolidation going on, estimates for the expected fleet to be handled by HGDEP by early next year vary widely, but range between 600 and 800, a full two times that of the rated capacity of the bus depot itself! Where else have we seen 250% capacity happen, besides the crush loading seen on Japanese trains of old? Amazing work, considering neither the Sengkang West nor the Kim Chuan bus depots that are supposed to succeed HGDEP in future are even operationally or structurally ready to allow HGDEP to offload its intense excesses, at least for the next year or two.

Good luck have fun migrating, bro

We’re not even going into the other operational implications that arise as a result of such a major shift due to pre-existing operational variables caused by previous shenanigans that various SBST depots have had to deal with due to various reasons. In particular, what makes the mass migration of these major trunk services to HGDEP particularly worrying is the already present acute manpower shortages which have been around for quite some time. The resignations of both bus captains and higher management staff allegedly got so bad that even off-peak service standards could not be maintained with the remaining manpower pool, and buses with drivers have had to be pulled from the other depot (UPDEP, SLBP etc.) on a regular basis to make up for the shortfall. Deadly ironic for the depot which also doubles up as SBST’s in-house bus captain training academy, losing manpower at unimaginable speeds. In a gigabrain move, HGDEP transferred a significant number of staff to Seletar Depot in 2021 due to the high turnover rates at the latter, further worsening its own internal shortages. Whether HGDEP is fit for taking over full control over these major trunk services will depend heavily on how willing SLBP BCs are in staying on with SBST and transferring over to HGDEP when the routes are transferred over. In that respect at least, there seems to be some hope, as employee loyalty to SBST appears to be at acceptable levels among SLBP BCs. Of course, that leaves SMRT with the even more pressing issue of sourcing for new BCs to sustain their JWP operations come 2024, and in an environment unfavourable to aspiring bus captains, I wish SMRT all the best, and better luck finding willing workers from across the border.

With such acute resource shortages caused as a result of this groundbreaking operator shift in the west creating rather desperate situations for SBST and SMRT alike, it should not be a surprise when, come 2024, more bus routes are cut to free up storage space in overcongested legacy SBST depots, respond to still-pressing manpower shortages that result in an inability to deliver service levels promised by the original BSRF guidelines (not like LTA has even cared about those ever since the end of the pandemic, honestly speaking). The more likely approach would very likely be to drastically cut service levels on many major trunk bus services in order to reduce the needed inventory require for the time being, until additional storage space in Kim Chuan and Sengkang West comes online. In that respect, expect service 154 in particular to take a huge hit — its mammoth fleet is partially due to the large number of downroutes and short trips operated during the peak periods (its off-peak fleet is quite measly, given the off-peak intervals of 15 minutes) aimed at serving the crunch loads between Clementi and Botanic Gardens stations, connecting students along the Clementi Rd and Bt Timah academic belts to the Downtown Line as well. I can already hear the electric cutters whirring, ready to fleece the route’s bloated fleet. For one, the many downroutes and short trips will most likely go, if not a full removal of those, as necessitated when the parent route migrates to full HGDEP control. It’s absurd that more than 10% of the depot’s entire capacity is being set aside for just one single route, but that’s what dual depot is for. Unfortunately, thank BCM for messing it up.

What’s efficient?

Even if we resolve the manpower issues somehow (as if SBST hasn’t dug a deep enough grave for itself here especially during the pandemic), placing all the buses for these long trunk services in one single depot brings us back to the intrinsic flaw of BCM that necessarily arises out of the fragmentation of the bus turf here: extended deadhead mileage. It’s a problem because every additional unit distance a bus deadheads, as implied by the name, is an additional unit of distance in which the bus is unable to collect revenue, hence leading to increased costs without chance of partial recovery through farebox revenue. This is a serious opportunity cost for the bus operator, which is why dual-depot operations have historically existed for SBST, and SMRT, and in the latter’s case the practice of outsourcing parking to other BCM depots such as Bulim and UPDEP1 closer to the starting points of some long trunk services far away from their operational bases is also an effort aimed at reducing such an opportunity cost from deadheading.

This necessarily implies two things when operations are converted to single-head, tanked by a single depot. Even if we put aside capacity considerations for the bus depots for a moment, single-head operations, like I’ve mentioned with the CRL’s arrangements in an earlier post, create a scheduling headache, especially for operations at the start and end of the revenue service day. Either it means that the non-anchor end will only begin receiving bus services along that particular route much later than the anchor end, or it will come as a great cost to the welfare of bus captains covering first bus duties from the non-anchor end, who must rise and shine much much earlier, at times of the day which are unsustainable for their health over the long term. The same applies for last bus schedules from the non-anchor end: either last buses from the non-anchor end must depart very early, or the BCs covering last bus towards the anchor end won’t be getting sleep till the wee hours of the morning. This holds true even with arrangements to deadhead buses to and fro the non anchor end rather than running all trips back to the anchor end as revenue service trips.

Consider service 154 again. Currently, its first bus timing from Boon Lay is 5.30am on weekdays. This is made possible due to the fact that the first eastbound bus is supplied by SLBP, allowing for the BC to only have to sign-on for duty at 5am or even later due to the proximity of SLBP to Boon Lay Int, and the deadhead time is at most around 15 minutes. Under a full HGDEP arrangement, even if the first eastbound bus was to deadhead all the way to Boon Lay to commence service, the deadhead route would require at least 40 minutes to cover (even with a clear expressway) with our speed-limited buses. Thus to maintain the same departure timing of 5.30am from Boon Lay, the sign-on timing for the first eastbound bus duty must now be shifted to latest 4.45am. Considering that BCs do not live where they work, that means whichever poor soul assigned to this duty must be awake by 4am, which is perhaps unimaginable for most of us. At night, the last westbound bus arrives Boon Lay at about 12.55am. The BC assigned to the last westbound duty can, in today’s dual-depot arrangement shared with SLBP, sign off at SLBP at latest 1.20am, and then maybe with a bit of luck be fast asleep by 2am. In future, this duty’s BC will have to deadhead another 40 minutes back to HGDEP, thus only signing off past 2am. He can basically forget about sleep at that point.

The more likely arrangement would thus be to push back the first eastbound bus timing from Boon Lay to accommodate operational needs (the earliest allowable sign-on timing from depots is 5am, I believe), and bring forward the last westbound bus timing, which necessarily means reduced access for passengers near the ends of the route. In particular, passengers at the western end of service 154 will stand to lose from a later first bus timing, as they tend to use bus 154 to travel to other destinations further east such as Clementi and Bukit Timah, rather than all cramming into the EWL at Boon Lay. With a later first bus timing for eastbound service, 154 may no longer be a viable option for some of them.

Not to mention that when the deadhead from Boon Lay to HGDEP is introduce come early 2024, it will become the longest deadhead route for revenue service buses outside the peak periods, surpassing the record previously set by bus 66 between Bulim Depot and Bedok, when it lost its dual depot option due to its transition to the Bulim package. Given that 154 must operate frequently to match is high demand especially in the morning, we are looking at more buses carrying air on the expressway just to get themselves onto the “wrong” end of the line. I am very sure such a practice over the long term will totally not blow a bigger hole in SBST’s operating expenses, especially when compounded with that of many other routes now converted to single depot as well because of the JWP!

It should also be noted that SLBP’s parking capacity is about 500 buses, due to it being built as Singapore’s first multi-storey bus depot, far greater than what is needed for the actual scope of operations in the JWP itself! Given that SLBP will not be permittted to dual-depot with any of SMRT’s bus operations under SMRT, this lends itself to a scenario where SLBP will most likely remain half-empty (being unable to take in LTA’s storage bus mothballing operations as the facility and land lease still officially belong to SBST) for the entire duration of the JWP’s first term. On one hand, an extra-large multi-storey bus depot will remain underutilised due to some arbitrary regulations inked virtually on GeBIZ. On the other, a tiny depot highly strategic in location will be expected to handle twice as many buses as it can store, with highly inefficient operational practices throughout the day.

Such allocative efficiency! Where the hell is the BCM’s much-touted optimisation capabilities at work??

SBST should take a leaf out of SMRT’s operating playbook and do what’s right for improving the operational efficiency of these long trunk routes — to park part of the fleets of long trunk routes based in HGDEP and BBDEP currently at other locations closer to the bus interchanges served by said trunk routes. For routes in the far west, UPDEP1 and Bulim Depot make for pretty good choices of alternative parking locations to stable parts of, or entire fleets for the time being while waiting for the Serangoon-Eunos and Sengkang-Hougang packages to move out. That being said, UPDEP1 has its own fair share of problems as a potential alternative parking area for bus services controlled by HGDEP.

the sliding block puzzle

What many were unaware of until very recently was that the current UPDEP1 facility was sitting on land that was merely leased temporarily to LTA for a period of 10 years commencing 1st Jan 2016. That is to say, UPDEP1 must be vacated by the end of 2025 to be retaken by the SLA. Around the same time, it was made known that PPDEP was being planned by LTA, to be put into operation starting 2029, while the land lease for BBDEP, currently the operations base for SBST’s CLP, would expire by 2026, 42 years after it was granted in 1984. As can be seen, there will be a delicate sliding block puzzle that the operators and LTA Bus Assets division will have to handle as the decade progresses along, with large-scale transfers of operations here and there as old facilities get phased out and new facilities get introduced. Despite how straightforward mainstream media has framed this transition from UPDEP1 to PPDEP, the actual reality on the ground will be highly complicated, and it will not be just one big shift, but with many intermediate stages in between.

Though UPDEP1 was only meant as a temporary facility from the start, it technically has a replacement just a stone’s throw away: UPDEP2, co-located with the expanded Ulu Pandan train depot, to be located under and around where the reception tracks to Jurong East are currently. Today, this patch of land surrounding the perimeter of the existing Ulu Pandan MRT depot is still overgrown with vegetation, with little sign of construction work in sight, although tenders for planning the rail depot’s extension were awarded as early as 2016 (even before UPDEP1 was built!), hence giving us the nice little artist’s impression of what UPDEP2 could potentially look like when completed. Its capacity, when completed, is reported to be 600 buses, a noticeable increase from the current 468 for UPDEP1. Unfortunately, news about UPDEP2 pretty much stopped after consultancy works were conducted last decade, though with things progressing along on expanding the rail depot there’s a reasonable chance the UPDEP2 part of the project will be revived. Still, even if the construction tenders for UPDEP2 were called today, I’d put the actual completion and commissioning date of UPDEP2 to be somewhere around end-2026 earliest… With PPDEP located far too close to the city, and far away from the near-west nodes in the bus network to act as the anchor for the CLP, it is most likely that the real replacement for BBDEP is actually UPDEP2, which is located within 5km of each other. PPDEP instead will replace UPDEP1 as a more suitable anchor point for the BMP2, being located far closer to the CBD where many of the BMP2’s anchor interchanges and terminals are located.

So UPDEP1 will have to be vacated by end-2025, if no extension for the land parcel is granted. BBDEP’s land lease expires in January 2026, so that has to be emptied out beforehand too. Meanwhile, PPDEP is only coming in 2029. That’s pretty bad news for the BMP2, CLP and every other service that does not wish to be congested in a certain bus depot somewhere else that is 250% over capacity. Sure, some displaced services may be able to seek refuge in UPDEP1 or BBDEP after being evicted from SLBP come 2024, but there will be a three-year period for which both of the abovementioned depots are closed down, and its replacement awaiting completion. On top of that, UPDEP1 is already nearing capacity due to its involvement in the parking of buses from SMRT, used for SMRT-operated bus routes based out of Bukit Merah and Boon Lay. Placing more buses in UPDEP probably means some other buses currently operated under BMP1 and BMP2 will have to go, though that leaves little options since UPDEP1 currently dual depots most extensively with, to great irony, HGDEP. Not a very helpful situation we have there.

It is greatly hoped that an extension of the land lease for UPDEP1 for 5 years be granted to allow for a proper transition of the BMP2 and CLP from UPDEP1 and BBDEP respectively to move into their final homes in PPDEP and UPDEP2 respectively. Having one HGDEP scenario in the north-east is pretty terrible on its own. We don’t need a second such case to pop up in the west, especially in a time of many changes that already mess things up sufficiently. While we are transitioning the financing framework for the bus industry towards the BCM, which is more or less unpopular for industry employees and passengers alike, the less cotroversial transition that the bus industry has also been making at the same time is a hardware on, with the legacy depots under SBST and SMRT from the last century being gradually replaced with new facilities that are hopefully equipped with the necessary features to support bus operations of the future, such as electric bus charging, and enhanced operations control systems.

Oh, and please get working on UPDEP2. It’s quite disappointing that nothing has been heard or seen of it since the consultancy reports came out in 2016. The site still remains an overgrown grass hill…

Maybe the rules of the game should be changed slightly. Let SBST continue to operate some trunk services from SLBP even after the transition, permitting dual-depot operations with the excess capacity available. It’s the lazy way out, but it’s better than the atrocious over- and under-utilisation of bus depot infrastructure around the various depots that result from furthered fragmentation of the bus industry turf.

What are you driving today?

Partially why SMRT’s win for the JWP came as such a huge surprise to many was also due to the great fleet incompatibility between what was operated by SMRT currently and the current fleet of the JWP. Although, like I said in previous posts, this kind of barriers are at best paper tiger boundaries that are pretty easily surmountable with a bit of staff training, though given the inertia of both SMRT and SBST we should examine this to some detail.

Your average SLBP single-deck bus today is a Scania K230UB or a Citaro.

haha funny number go brrr

Your average SLBP double-deck bus today is a Volvo B9TL (of varying bodywork types, but by the time SMRT takes over only the WEGII variant will still be operating):

However, what your average SMRT bus driver is used to driving is probably these instead:

This mainly goes out to the bus enthusiasts who are all the hype over livery and logo changes on the bus fleet, but do expect a lot of movements around the depots before, during, and after the transition to an SMRT-operated JWP, as SMRT sheds the former SLBP fleet in exchange for bus models that its employees are more comfortable working with. To what extent this swap will take place remains to be seen, and I suppose there is much to look forward to. Just don’t flood my area with excess Citaros from SLBP and I’ll be happy, to be honest. Though I’m pretty sure the opposite will be happening instead, welp.

It is through the change in hands of the JWP that exposes the flaws of the BCM, shooting many of its purported benefits right in the foot regardless of what LTA says about it. Hooray, SMRT won a package at last. But this win would result in an entire boatload of operational problems that would resonate far beyond just the west side of Singapore. The franchised model of bus service operations may have lent itself to slack, but it at least enjoyed great economies of scale that enabled long trunk bus routes to be operated more efficiently with greater coordination, at least, allowable in theory. Unfortunately, the fragmentation of the bus industry here erodes this efficient operating framework, and at the same time breeding a culture of price warfare that leave less available to sustain operations… sustainably for even the tendered term itself!

As we wrap up the hype around the announcement of tender results for the JWP and BMP2, let us remember that the BCM is at best a fair-weather friend, with the struggles of the BMP1 and Seletar Package still fresh in our memory, a reminder of what comes from underbidding just for the sake of getting a package. An incredibly myopic system that sacrifices its long-run overall operational efficiency for reduced upfront contract costs, the BCM has quickly fallen prey to typical market failures.

Supposedly, an approach leaning closer to the free market would shake up the local public bus industry by bringing about better outcomes for all. That great assumption aside, it is important to note that BCM is not your typical market of consumers and suppliers. Why is that such a huge delivery gap when it comes to BCM, for us, the passenger?

It’s all about the money!

Because, we the passengers who use such bus services, are sadly not the consumers of such bus services under BCM, if we take the microeconomics definition of a consumer to mean the party that pays for the good or service supplied by the producer. Need you be reminded, the transaction that takes place in a BCM contract is solely between the winning bidder and LTA. Where are the real stakeholders, such as passengers and PTO employees, whose lives and livelihoods are directly affected by such a decision? They are but mere spectators.

What, then, are we? The packets of pretzels in the vending machine, used to generate revenue for primarily LTA! Recognise this, and you’ll never approach BCM the same way ever again.

Accurate depiction of who we are.

Supposedly, LTA would represent the interest of the public that rides buses through its regulatory role, and moderate the excesses of free market capitalism on a public utility such as public transport. Just giving bus packages to the lowest bidder however, has shown that even this safeguard has been given up upon by LTA. The pretzel packets, it seems, have no one but themselves now.

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

PS: Psst psst, STC is also on WhatsApp now! Receive updates from our blog while you text your friends, by clicking on the button below!

STC’s on WhatsApp!!