Will loans help?
S$90 billion is a lot, but it can be all snatched up pretty quickly.
There are many things to be concerned about. New MRT lines may be the biggest issue. There is also the DTSS, S$6.5 billion. And of course, expansion of electricity-generating capacity, which already has had its effects. Don’t forget the S$100 billion climate mitigation plan, though some of that may still overlap with the previously-mentioned projects.
And the government wants to pay for all of it with S$90 billion under SINGA? Ambitious. On further reflection though, it might not have to, with only part of the cost of the projects being covered by loans and the rest still coming from tax income, allowing the same amount of income to be spread out across more projects. Whatever it is, though, it is a better position than the German way of only investing in infrastructure when there’s leftovers from recurrent expenses, which is what we’ve been doing.
What lies ahead
In order to stretch the dollar and make the S$90 billion last longer, the first priority is that we need to fix the sleaze in the construction industry, if you want to call it that.
But one thing is for sure that cleaning up the sector will cost money, between providing a living wage for even lower-skilled locals, and the investment needed in automation. In fact, the most recent MOT Budget has S$87 billion alone earmarked for rail expansion, and one can only expect this to go up with the aid given to the industry to tide them over, and associated reinvention programmes. They may have not used all of it in the past (from Budget book records), but they may need to use all of it now.
I do not think Lawrence Wong will be very amused, though, if asked to tolerate an environment where costs go up for the same outcomes, even if the money is already set aside. I may have thought previously that throwing more resources at the problem may help; but will it now? It’ll probably have to get worse before it gets better, with the amount of technological development necessary to get to a more sustainable position.
Though if we don’t know where to start, we can’t even start. I’d suggest the JRL, which as an aboveground line is best placed to benefit from many of BCA’s technological productivity initiatives. From a quick survey at time of writing, even the Stage 1 stations have not really yet begun foundation works or are in the midst of it.
While that takes place, and it is likely to proceed at a pace slower than expected, perhaps several reengineering initiatives can be undertaken in order to optimize the station structures for more efficient building. If they can even get out quicker, it might be a net benefit to the residents too — one can look at historical records on Google Street View along Pioneer Road to see what might be in store for the JRL in the next few years.
From bad to worse
Yes, between manpower and resource shortages, the construction sector has had a very hard time this year.
The thing is that underground civil engineering works do not really lend themselves well to a large amount of precasting works, which may make it difficult to really improve the productivity of building underground stations, such as that of the CRL. Perhaps a solution could be found to use precast slabs and wall pieces for the interior construction, but short of robotic mining machines to aid excavation I’m not sure how much more can be done.
Yet, all this comes before any potential legislative remedies to the foreign workers issue, which may impose higher costs on the various projects; cost sharing is anyway mandated under the current relief measures, which means that contract sums paid out will go up. Decreased quotas and higher levy rates in order to force the industry to kick the habit are also a thing to consider in the medium to long term, which may make contractors feel it necessary to claim for additional costs not expected at the time of tender. All the more so with the 10+ years it takes to complete an MRT project, and which you never know what can happen in the meantime.
The problem may just have been time previously, and the answer at that time might have been to throw more money at it, but now it’s all the more important that the manpower dimension is taken into account, especially in order not to further derail the sorry state of infrastructure expansion.
So how can we recoup some of the losses attributed to the above? The way I see it, something may eventually have to be cut. Furthermore, even if the entire project cost is not going to be paid for by government bonds, cost control is still necessary in order to get the most out of whatever portion of that S$90 billion that can be used by rail projects. It is likely to happen at the same time, but don’t forget there is at least still one more MRT line coming up after the CRL. Financial bandwidth will have to be left for that too.
The scissors should be out
Among others, the Levyist literature always likes to point at South Korea, given apparently low construction costs there. What is the secret to their success? Well, we have Korean companies being given the largest (by dollar value) packages such as CR108 (Pasir Ris, Daewoo-Dongah JV) and CR112 (Hougang, Samsung C&T). In the case of the former, tender records put almost everyone else’s bids at over $1.1 billion — the successful tenderer squeaked under $1 billion, and barely. Maybe we’ll be able to find out, and even if it can even be put in place here.
At least when I cover this topic, I’ve long been an advocate of designing for efficiency in stations. You’re going to be in the place for 10 minutes at most, there is really not that much point to form over function that poses inconvenience to the passenger. In fact, it might even serve as an access penalty; dissuading the use of multimodal journeys because so much time is spent getting to the train platform. This is all the more important when it comes to CRL, given its role as a “funnel” between the east and west sides of the island.
But what can be realistically cut? It is unavoidable that several CRL stations have to be quite deep in order to pass by neighbouring buildings, especially in the east in order to go around the underground NEL and TEL. With its relatively suburban route, it is also unavoidable that several interchange stations will have considerable vertical distance between existing lines and the CRL, since most of its catchment in the west means either existing elevated lines or the elevated JRL.
And I personally don’t know if this will blow up in our faces, but some quarters may think that ridership projections can perhaps be relooked in order to consider the possibility of removing the future provisions for eight-car trains and keeping only a six-car long platform, or pushing more of the enabling work to some time in the future. This may be possible if CRL, like other lines, proves capable of running headways below 2 minutes or something. Though with the underbuilding problem we face elsewhere, this could potentially be foolish as well should CRL be far more popular then expected.
A question of space
The overall takeaway, I would say, is that space needs to be better used. It may be alright to have a long passenger concourse so long as escalators, stairs and lifts are adequately spaced out along the concourse and platform to prevent crowding hotspots; hard to do on the 3-car lines, but probably easier with a longer train. Another concern is void space; plenty of such double or triple-volume space in DTL3 stations, for example. Not only does cooling them require higher energy usage (with the whole energy efficiency schtick), the space they take up could potentially be used to house back of house supporting rooms.
In this case, reducing the footprint of the stations, to me, is a no-brainer — platform-end equipment rooms and other staff rooms can then be placed on such intermediate levels between the platform and concourse. After all, mining may be out of the question anyway due to soil conditions, so cut and cover may be the only choice, which means this space is created anyway. That way, a greater percentage of deeper level (typically the platforms) is occupied by either tracks or usable platform, for either surge load or Civil Defense purposes.
An example of this, though not a very good one, is CCL6 Cantonment station. But instead of siting the fareline at B3 and requiring at least two changes of lift from street level to platform, B3 could have been used to house the spare mechanical rooms that take up so much space at B4. Creative escalator routing might then allow for the same lifts to run all the way from B2 to B4; also increasing accessibility for the disabled. Maybe, even, the existing railway station could have been reused as part of the CCL6 station, which to me sounds like a more meaningful tribute than the design that has been used here.
Something else that can also be considered may be to increase the permitted gradient at which tracks can be built. Increasing gradients may allow for shallower stations to be built while still allowing tunnels to pass under buildings safely. As an example of this, the Shatin to Central Link has something that looks like a 4% gradient — steeper than ours at 3%, on show at Jurong East. Of course, it bears pointing out that this may backfire, since the cost savings from shallower stations may be outweighed by the additional costs of more powerful train motors to climb the steeper gradients; though developments in power electronics may also render this moot.
Small things that matter
It bears remembering too that all that is cut and cover is not stations; crossing tracks also need to be built for service flexibility — the lack of them along the NSL and TEL have, in the case of the former, affected the service that can be run should there be any form of outage, and there is a chance it may cause havoc with the latter too.
So crossing tracks are definitely important. But they are also built by cut and cover for the most part, which still means surface disruption. Thus, typically, they’re built with stations, increasing the size of the station itself. However, a thing that is happening with CRL is the use of large diameter tunnels which allow both tracks of the line to be placed within a single tunnel. That way, crossing tracks can be built within the single tunnel, avoiding the need to excavate from ground level. In Barcelona, even, both tracks are stacked within the single bore tunnel, creating space for a third track that can be used as a siding or at least a crossover — eliminating the need to cut and cover for that, too.
Another thing that may need to be relooked is the usage of diamond crossings at non-terminal stations. With the use of a single bore tunnel this may be a moot point, but under current practices, in order to provide similar operational benefits, it would be necessary to provide dual crossovers (like the one to be replaced at Pasir Ris), which means a larger cut and cover worksite. This may place engineers in a tough spot — eat more land, or reduce the use of the crossover? In any case, it would then be important to again, make better use of space at platform level. Placing crossovers near to the platform also allows better headways to be run, should there be such a serious incident that requires short trips to be performed to maintain service on the rest of the line.
Take, for example, the previous failure at Bukit Panjang station requiring all DTL service for the day to operate on a single track. Had this happened anywhere else with only a single crossover, the service would have fared far worse; and placing the crossovers so far from the platform at Bukit Panjang tied their hands with the amount of trains they can run.
Additionally, unlike CCL4/5, extended exits can now be easily built by more productive methods such as the rectangular TBM. This allows the service areas of each station to be expanded, allowing a single station to serve more places. The price tag may also be reduced as deep cut and cover excavation can be avoided, with a TBM-bored tunnel below a mined underpass like what’s done at Havelock station.
One can also consider CCL 4/5 — generally similar materials, palette, and potentially layout too. Find a place to put the station core and the rest should be able to sort itself out. However, I’m not sure how well-placed CRL is to benefit from standard station design, though, given its lower station count and thus higher station spacing; and the many interchange stations that may need some form of special treatment. But it may be of use to the planners of the new “Seletar Line” or whatever they eventually call it. Especially in an environment of resource shortages and precast works being encouraged, where standardization may allow contractors to pool orders and reap economies of scale.
Another consideration may also be the size of the work packages, with each CCL4/5 work package consisting of a now-unheard of 4–5 stations per package. From the 2011 Budget, CCL4/5 was actually cheaper on a per km basis compared to the first three stages; while an argument can be made that stages 1 and 2 are more complex because of the CBD and the depot respectively, stage 3 was built on an individual basis and that still cost more.
From my point of view, apart from the largely standard station designs, large contracts such as C854, C855 and C856 might not only have allowed contractors to take advantage of economies of scale in terms of purchasing, but also retain the flexibility to redeploy manpower to locations where they are needed. Though this can also be seen with the Alpine bankruptcy, where the contractors nominated to replace Alpine were all from neighbouring contracts and able to quickly redeploy specialist manpower for mining and related works that had finished works on their stations.
On JRL, larger work packages might also have helped to drive an environment of innovation, whereby whatever investment made by individual contractors could have been paid off across a larger scope of work. However, it might have worked against the goal of getting more people into the market — lesser bidders for larger projects may still mean a cost increase as bidders feel more willing to charge more.
One thing is for sure though — doing something about the rising cost of rail projects has to become a priority if we are to make the most out of the funding made available through SINGA. More importantly, though, we also must ensure that increasing manpower costs don’t impact projects too much and basically drink up the whole S$90 billion, be it either through reducing manpower use in the first place, or by reducing costs elsewhere to provide such a cushion.