There was an error in this gadget

Thursday, 9 September 2010

The Property Bubble & Investment Trap Part VI - Land Bid Analysis for DBSS Site & Inherent Weakness of Bidding System

"WAIT-AND-SEE was the order of the day at showflats over the weekend - the first since measures aimed to cool property speculation were introduced by the Government last Monday.

The mass-market segment appeared to be the most affected, with thinner crowds and noticeably fewer sales. At more upmarket properties, there was still some interest among genuine buyers." - Straits Times Online (Sep 6, 2010)

The cautious mood prevails and property sales were lacklustre. This was reported at The Minton, Waterfront Gold, Starlight Suites and Adria. The next weekend may give a better indication of the market, an analyst said - referring to the anticipated launch of NV Residences, a mass-market project in Pasir Ris.

I note certain comments posted in my blog by visitors who could be "speculators". For transparency, I had posted their comments and then responded to test and rebut their intention. They attempted to argue against the cautious current mood of the market by putting up confusing general statements and arguments. This could be very misleading.

The purpose of my postings are to offer an "insight" perspective of the property market to warn would-be genuine "home-occupier" investors about the property investment "trap" and to offer some simple advice in guiding their decisions, especially first time purchasers.

In this and the next posting, let's us take a look at the "land supply" released by the Govt (or SUPPLY side) against bids submitted by Developers and "Develop and Build" Contractors. I will use 2 bids as case studies to illustrate their impact on the final sale price of properties.

These 2 record-high bids involve a HDB DBSS [Tampines Ave 5] and a private 99 Leasehold (LH) condo next to Orchid Country Club at Miltonia Close, on the fringe of Yishun Town Centre. For simplicity, I will just refer to them as Tampines site and Yishun site. [See references in my last posting].

(1) Tampines site (HDB DBSS)

The following are the "land" bids by Developers or "Develop and Build" Contractors [D and B Contractor]. Information are added by me based on my own knowledge for greater clarity.

(a) Bid 1 - S$178.13 M @ $261 psf ppr - Local D and B Contractor

(b) Bid 2 - S$145.77 M @ $213 psf ppr - China D and B Contractor

(c) Bid 3 - S$139.90 M @ $205 psf ppr - Local Developer / M'sian Contractor JV

(d) Bid 4 - S$137.10 M @ $201 psf ppr - Local Developer / Contractor

(e) Bid 5 - S$110.00 M @ $161 psf ppr - Local Developer

An academic analyst had expected the bids to fall between $160 ~ $200 psf ppr. The previous high land bid for a DBSS site was $237 psf ppr in 2008 for a site in Bishan, a much better location.

My comments :-

(a) A "pure" Developer who is not a contractor tends to submit a higher bid. Reason : As DBSS units are sold by them, he must re-coup his costs for - (i) Land bid (ii) Development costs (consultants) (iii) Construction Costs. The higher the land bid, the lesser he has for spending on item (ii) and (iii), but he secures the land. Construction and land costs are the major components to establish the final sale price of the property unit cost (psf), after considering the plot ratio (pr). Hence the land bids are expressed as psf prr (per square foot, per plot ratio). The plot ratio determine the intensity of development or Gross Floor Area (GFA) that could be build in relation to the Gross Plot Area of that land. If there is waiver for GFA calculation in respect of Bay Window and Planters, the Developer can build more intensively but the Bay Window area will be generated as "Saleable Area". [Remember the Bay Window Trap I mentioned before?]

(b) Hence, for a Developer he will "squeeze" his consultants and contractor and push up "land cost" as much as possible in order to "win" the land use from the GOVT tender first. But there is a rock bottom that his contractor can build it depending on the specifications although consultancy cost can be a fixed %. Therefore, in the first round, if a very high land bid is submitted, the real winner is the GOVT. This is evident in the Tampines site bid, Bid 1 that won was 22% higher than the next lower. Bid 2 ~ 4 which are more rational bids were clustered within the $205 +/- $ 5 psf ppr band, which was just $5 higher than the highest benchmark predicted by an academic anaylst. But Bid 1 overshot it by 30%. Is this ridiculous?

Market watcher suggest that since Developer for Bid 1 has a construction arm, it is therefore able to better control costs on "construction". But is this really so? I like you to compare Bid 1 with Bid 3 and Bid 4 which are also "local" Developers with construction arms. They are able to bid about $60 psf (30%) lower. Does it means that they would save from the construction costs later on? I don't think so, because I believe all had tendered based on the same specifications.

Ultimately, this 30% ridiculous land cost "buck" has to be borne by the purchasers of these DBSS flats and buyers would have to arrange their own financing though they are entitled to the same HDB housing grant. So the HDB gives you "wider" choice(s) but you have to pay 30% higher for the land under DBSS even if the contractor decides not to make a profit and pass on savings from construction costs to offer a better sale price of the units. But can any firm sustain itself by not earning profit and just covering cost?

Imagine further if this is priced to chase the next class of property, say EC, which in turn is chasing the higher property class and ultimately the $200 psf benchmark I had mentioned before. Worst, imagine the lower BTO flats are also priced to chase these DBSS flats with land costs which are 30% more expensive. The whole property market will just spiral. This is an inherent fault with the land bidding system, the highest bid wins, but could be ridiculously high. The real winner is the Govt / HDB who offered the lands for bidding and the DBSS Developer will certainly recover this from the end purchasers. We just need to look at the pricing of Bid 2~4. Bid 5 may be too low a bid and we can ignore it. But look at Bid 2 by a China D and B Developer who also has a construction arm. They are are also 22% lower.

Let me take you one more step further into the cost analysis to see how this all ends up with a higher price for end purchasers for these DBSS HDB flats ... and in time to come the future HDB resale market ... and the high property price future generations of Singaporeans has to bear.

The floor areas for DBSS flats and expected launch prices are as follows :-

(a) 3 Rm - 70 sq. m. - $380K~$400 K or avg $5,571 psm or $518 psf

(b) 4 Rm - 91~94 or avg 92.5 sq. m. - $530K~$550K or avg $5,838 psm or $543 psf

(c) 5 Rm - 105~114 or avg 109.5 sq. m. - $640KK~$670 or $ avg $5,982 psm or $556 psf

Actual land cost per sq.m. GFA

(a) Permissible G FA - 63,395 sq.m. less 1,060 sq. m. for social and communal facilities = 62,335 sq. m.

(b) Estimated land cost per sq. m. nett GFA = $178.13 M / 62,335 = $2,858 psm = $266 psf GFA

What is the expected breakeven construction cost (psf GFA) if the units are sold at target price ?

(a) 3 Rm - 70 sq. m. - $380K~$400 K = $518 psf - $266 psf = $252 psf

(b) 4 Rm - 91~94 or avg 92.5 sq. m. - $530K~$550K = $543 psf - $266 = $277 psf

(c) 5 Rm - 105~114 or avg 109.5 sq. m. - $640K~$670K = $556 psf - $266 = $290 psf.

Constructions costs - after staying stagnant for several years following the 911 Event and SARs crisis, and economic downturn - were at its at record levels in 2008.

"Construction cost consultancy Rider Levett Bucknall (RLB) said: Construction prices for medium-quality condominiums indicatively range from $260 psf of GFA to $320 psf of GFA in Q1 2008, and prices have risen further to $280 to $350 psf of GFA for Q2 2008," it said. "High demand and competition for limited resources, the lack of tendering capacity among contractors, sub-contractors and suppliers, and volatile commodity prices have contributed significantly to building tender price escalation," the firm added.

"Construction costs are estimated to have risen 20 to 25 per cent for Q4 2007 compared with the corresponding period in 2006 for average medium quality condominiums (for the upgraders' market).

While the trend of construction costs exceeding land costs has drawn more attention since the recent tender closings of Government Land Sales (GLS) sites, some observers say it surfaced as early as December last year[2007], when Chip Eng Seng bought a plot at Elias Rd in Pasir Ris for $228 psf ppr.

In the same month, Frasers Centrepoint picked up a site at Lakeside Drive for $248 psf ppr – which was probably about equal to construction costs at the time.

Construction costs comprise not just the cost of building materials but also include factors such as workers' wages among others."

So you note, when construction costs were extremely high, the lower "land cost" was chasing after the construction costs which reached its peak in mid-2008.

Now, I want you to focus on the the lower end of the construction cost spectrum for medium-quality condominiums for Q1 ~ Q2 2008. The construction costs then was $260 ~ $280 psf of GFA. Now you look at the bid price of land submitted and won by Bid 1 for Tampines DBSS Site - S$178.13 M @ $261 psf ppr. You take a look to appreciate this whole picture of the costs chasing game! Does what I said now make sense to you?

Next, I want you to focus on the targetted breakeven launch price and construction costs I have worked out per sq m GFA for Tampines DBSS Site for 3Rm, 4Rm and ~ 5Rm - $252 psf GFA, $277 psf GFA and $290 psf GFA respectively. It more or less coincides with the lower end of the construction cost spectrum Q1 ~ Q2 2008 ($260 ~ $280 psf GFA), but for medium-quality condominiums. Mind you not the low end mass market condominiums. I consider DBSS flats as equivalent in quality to the lower-end mass market private condominiums and EC. The only difference is in common facilities and MCST management.

And to analyse the price a little deeper :-

Bid 1 Land Cost @ $261 psf ppr + Lowest Construction Cost Q1 2008 @ $260 psf GFA = $521psf. This is about the targetted launch price for 3 Rm @ Tampines DBSS site.

To explain why I consider that the current launch price is too high and not sustainable, we need to look closer at the construction cost. As I mentioned before in an earlier posting, the only component to note is "steel prices". Other materials and construction related costs are relatively stable. With Euro$ at its low, even Italian marble and tiles prices have fallen. If you look at the Steel - World All Products Price Index, steel price rocketted and reach its peak in August 2008. This explains why construction costs reached its peak then. Since then, steel prices had corrected drastically to about US$600~US$700 per ton now or equivalent to the Mid-2006~2007 prices, which coincided with the 2007 property market peak here. Hence, a construction cost of $260 psf GFA for DBSS at current pricing is still too high just considering both quality of finishes and steel price, other things being equal.

If you look at the URA Private Property Price Index (PPPI) below, you will see the PPPI reached reached its peak in 2007/2008 and then fell in 2009 with the Sub-Prime Mortgage and Mini-Bond Crisis, and in 2010 it shot up sky high again to beat the 1996 record high. That was for Private Property Market Index, but you see the DBSS HDB market followed the same price trend. I believe it is the same for the HDB re-sale market. It is quite evident that prices of HDB flats, particularly DBSS flats are chasing after private property price benchmark.

Comparing Target Sale Price With Historical DBSS Prices

Now, how does the target sale price of $521 psf compare with historical price for the first DBSS project, also in Tampines (Premiere) and was developed and built by the same developer. Most units in that project were 5Rm and was sold between $280~$370 psf with the land bid awarded in Jan 2006. In 2006 property prices was climbing to its peak in 2007. Despite a consolidation in price due to the "sub-prime" crisis, the price has shot up by more than 40%~80% depending on the unit types. The winning land bid for this first DBSS was $113.67 psf ppr. The current bid for the Tampines site is more than double the land price (nearly 230% in a matter of about 4 yrs 8 mths).

Even if we allow an incremental rise in land cost at 10% annually, the compounded land cost after 5 years would have been just $183 psf, if compounded at 15%; at $227 psf. The current bid is extremely excessive.

Hence, even if the DBSS units for this Tampines site are sold at about $521 psf, I expect the D and B Contractor's to be pretty comfortable with a margin of about 15%~22 % with all costs well covered, despite the high land bid. A more reasonable pricing range would be between $450~$480 psf (depending on unit types) with a reasonable profit margin. If it is located very near to an MRT station, a price closer to, but below $500 psf would be justifiable, considering it is just a 103-year leasehold property.

Hence, I would speculate on a price consolidation of at least 8~14% for this class of property, i.e. DBSS flats.If the economy does not improve as expected and / or more control measures are taken, I will not be surprised to see a bigger consolidation of about 10%~20% eating into part of the developer's profit margin. Generally, buyers should be wary of the high land cost they will be contributing as part of the final purchase price. The Chinese saying is "The sheep's hair (wool) is always found growing from the sheep's body".

Honestly, the trick is to drive down "construction costs" to offer a better pricing, while enhancing profit margin. This explains why a developer with a construction arm is better positioned to compete. However, the reality is construction costs are not too transparent, while land cost is driven upwards ridiculously in a hot property market.

Imagine its impact on new BTO flats if they start to chase after DBSS prices. In comparison, BTO new flat launched in March 2010 at Fernvale (Seng Kang) are priced about :-

(a) 3 Rm - $170~$230 psf

(b) 4 Rm - $220~$279 psf

(c) 5 Rm - $240 ~$300 psf.

Hence, you will note from from the BTO prices above, that you are paying twice as much for DBSS flats as compared with BTO flats, but without condo facilities and not the same mid range condo equivalent quality.

So if you buy HDB DBSS property, what are you chasing in this property game?

(a) 2008 high construction costs for medium range condominium which is not justifiable today considering cheaper steel prices and also cheaper finishing materials. This is not reasonable.

(b) 2010 high land cost. This is too excessive considering it is a 103-year lease.

(c) Overall high property price which is equivalent to the lower range mass market condominium or EC at 1996 / 2007 high peak prices.

In my next posting, I will do a similar evaluation for a 99-year LH condominium site at Yishun. Meanwhile, it was reported in Business Times that the NV Residences at Paisr Ris was soft-launched yesterday. The 200 units released yesterday were reportedly priced between $557,000 for a 506 sq ft one bedder on the second floor (working out to $1,100 psf) to $1.9 million ($761 psf) for a penthouse. Anyway, stay CAUTIOUS and your maxim should be "Watch what they do, not what they say" and press reports included.

References :-


  1. There's something I don't quite get here. You said:
    "Bid 1 Land Cost @ $261 psf ppr + Lowest Construction Cost Q1 2008 @ $260 psf GFA = $521psf. This is about the targetted launch price for 3 Rm @ Tampines DBSS site."

    So $521psf is the breakeven price. Then why would the developer sell at $521psf; making $0 profit? I probably mis-understood something...


  2. Hi YS

    If I were a lecturer and you my student, I would think you are the best. Not because you did best in your exams, but because you obviously ask the most intelligent question.

    Land Cost is transparent in a open tender. But construction cost is not really. I am able to understand what this component in the equation means, same with those trained and well versed in construction econs. The ignorant layman can therefore be "conned".

    The second part of the equation is the "variable". I had termed it as "construction cost" to the developer but it is actually a "budget" which you can throw in a "profit mark-up" for both (the developer + the contractor). This being not "transparent" is where a contractor can be marginalised by a developer and he would be forced to give you as a buyer the "cheapest" in quality as long as he can get acrossed the Consultants and developer's project manager.

    From your intelligent question, you can see if you fix the "profit" part for the developer, then the balance can be forced on the "contractor" to give the developer as high a profit as possible.

    I am working from first principle as if a "D&B" developer, and already "generous" with the figures, which is how I can gauge confidently if the "construction cost" component is fair or not to the developer or a contractor. My knowledge of construction costs will help me to see through this missing link in this "magic" formula which is actually not rocket science.

    Hope this explains why I am confident to say $521psf is fair for this particular DBSS. Because at $260 psf I could built you a "condo" with reasonably good materials; the details of which is no great secret to a construction pro handling project costs.

    And if I were a "D&B" dveloper, I have flexibility to go even lower in final pricing since the developer will not take away a bigger share of the profit markup from the equation.

    Your misunderstanding is "intelligent" but not the ignorant layman. And I will not elaborate here, that with the same approach I can fairly and honestly tell if MBT is telling his "truths" about affordability based on "facts".

  3. Yak, so what you're saying is that if the same company is both the developer and the contractor, then usually what is termed as "construction costs" already factors in the profits for the company.

    I guess my initial confusion stems from not knowing "how things work", i.e. what does the developer do, what does the contractor do, and their relationship with each other. Also, if they are separate entities, would I be correct to say target price is then: land costs + construction costs + profit for developer?

    If you're so kind enough, it will be great if you'll do a post to educate us what's a developer, what's a contractor, their roles, their relationship, the typical profit margins they're looking at, and how all these formulates into the target-price formula. To outsiders not in the construction business, like me, you can understand it's fairly hard to know how these things actually work. Will be greatly appreciated!


  4. Yes, I presented it as such from a D&B developer's for simplicity.

    Para 2 - Yes you may split it as such - you can also split the "construction costs' in your formula further into "contractor's cost with mark-up, consultants' costs, legal fees,...I don't think the average person can take so much details ...

    For me it is 2 ways with costs - I can breakdown or built-up to evaluate ... but any comparison must be on a common base.

    Your request - It will take too much of a time. And different project management approaches and contractual models are available. I had worked on various approches. I think the average readers would not be keen.

    More important the point is after land cost is fixed, the construction costs & profits are variable and not so transparent. Developer's profit taking has a great impact on contractors' cost & profits. Contractors are constrained by materials + labour costs , all fluctuates with the level of economic activity.

    My point was land costs once chased construction costs, construction costs had fallen and consolidated, but land costs never do but climbed to new heights instead while Govt restrained supply and then overstretched it by the BTO system in the case of HDB flats while population increased drastically with imported foreigners.

    You read and heard all these analysts' reports and interviews, is always based on their "expert knowledge", then the likely "target price" is such such and in a range ... most of them has a direct interest with developers as they are often the marketing agents unless they are academic analysts. Some academic analysts also collect data from these marketing agents ... hence ... in new launches it is always a high price.

  5. Appreciate your reply, Yak. I've just one comment.

    To a lay-person like me, it seems that comparing the increases in land costs vs. increases in construction costs; over time, land costs will increase more; simply due to land available for construction getting lesser and lesser. I'm assuming that the decrease in world steel supply and increases of worker wages are not as drastic as decrease of land supply.

    However, I do agree that the current land bidding system (accepting the highest bid) is not well controlled by the Govt, and fuels unrealistic increases in land costs.