Resale prices for public housing flats in Hougang single-member constituency (SMC) are not clearly lower than nearby flats of the same design, judging by an analysis done by Bernard Leong and his Singapore Angle friends.
The analysis was based on a total of 798 resale transactions of 4-room flats in Hougang Town over a thirteen-month period between 1 April 2010 and 30 April 2011. Of these, 103 flats were located within Hougang SMC, of which 79 were of the ‘New Generation’ design and 25 were of the ‘Model A’ design.
In the adjacent parts of Hougang Town that were assigned to other constituencies and held by the People’s Action Party (PAP), there were 49 ‘New Generation’ flats and 453 ‘Model A’.
There were a further 192 flats of other designs in the PAP-held areas, but since there were no equivalent ones in the Workers’ Party-held Hougang constituency, these were omitted from the analysis.
The raw data, including floor area and age of the building came from the Housing and Development Board (HDB). Source. Prices post-analysis are shown on a per square metre basis to remove any bias in flat sizes. On the right-most column the difference in average ages of the flats is also shown so that you can make a mental adjustment.
The price difference between Workers’ Party-held areas and the People’s Action Party-held areas in the same town does not look very great.
* * * * *
This differs from figures released by housing minister Mah Bow Tan and published in the Straits Times 4 May 2011, which shows an “Aljunied premium” of 16 percent for 4-room flats.
Housing Board resale flats in the Hougang single-seat ward have fetched lower prices than those in Aljunied GRC on average, said National Development Minister Mah Bow Tan yesterday.
He released figures aimed at refuting Workers’ Party chief Low Thia Khiang’s claim that flats in his former ward were valued no less than similar resale flats in wards held by the ruling People’s Action Party (PAP).
— Straits Times, 4 May 2011, HDB data shows Hougang flats fetch lower prices than Aljunied units: Mah
Below are the figures from the minister published alongside the newspaper article:
Assuming that Mah Bow Tan obtained his figures from the HDB just like Bernard Leong did, it may strike you as strange that the difference in prices is so much greater in Mah’s figures. I can’t explain how Mah crunched his figures but it is perhaps useful to note the differences between the two tables.
1. The upper table is drawn from a much larger sample (606 versus 54), but all were still flats in Hougang township. Why Mah chose to use such a small sample is unknown.
2. The upper table includes ‘Model A’ flats whereas the lower table does not.
3. The upper table uses price per square metre, not price per flat.
4. The upper table indicates age difference.
5. The price difference between Hougang SMC flats and others is much less in the upper table.
Mah also gave figures of 3-room and 5-room flats, but since Bernard Leong’s analysis did not include them, there is no discussion about these types of flats here.
* * * * *
When Bernard Leong gave me those figures and the comparison with Mah Bow Tan’s, it seemed to be at odds with what I recalled. I recalled that it was not Mah Bow Tan, but Lee Kuan Yew who made the assertion that flats in opposition party-held wards were lower in value. So I did a search of the Straits Times, and found that we were both right. Both Lee Kuan Yew and Mah Bow Tan made the same assertion. However, what I found interesting was that Lee’s assertion came four days BEFORE Mah released his data.
Homes in Hougang cost less than similar ones in neighbouring areas because they belong to an opposition ward, said Minister Mentor Lee Kuan Yew last night.
‘If you have the wrong government, your property prices go right down,’ he warned.
‘Ask why in Hougang the property prices are not as high as their neighbours’.’
He added: ‘If Singaporeans are that fickle, they will have to pay a price, the hard way.’
— Straits Times, 30 April 2011, Why Hougang flats’ value is lower
It’s a bit of Alice-in-Wonderland, that. Assertions are first made, then data supporting them are released.
And then we have nosey, itchy-fingered people who go dig up the same data and recalculate everything.
Yawning Bread 
We need to be nosey and itchy finger. Singaporeans need to, as Puah Cho Kang said “use your blain”, and not just believe everything we read in the newspapers. Well done, sir.
Mah produced his data only after MM’s assertion, probably in order to confirm it? And you wonder why those data are questionable?
It might have been another instance of such a phenomenon. Remember Devan Nair was sent to US for second opninion about his “alcoholism”? Someone who accompanied him told me that they were to get only a confirmation and no other! I hope he was joking.
Anyhow, it’s like this Deputy Director of the Institution I worked in, who had been asked to write papers for the then Minister. He told me, I hope jokingly too, that the strategy was to anticipate what the Minister wanted to see, and then produce data to prove it.
An alternative scenario – could it be partial statistics were channeled up, which then became the premise for the Assertion, which in turn had to be propped by the same partial information Provider?
Could it be also that the building shortage has caused Hougang prices to catch up (the sample period is last 13 months)? A knock-on effect. A policy lapse that ironically comes back to undermine a past Assertion. Not that price is everything.
Have you seen a slightly more rigorous analysis of this that’s floating round on facebook?
http://www.facebook.com/notes/giam-xingli/no-evidence-that-flats-in-opposition-held-wards-are-worth-less/183135021738371
Using the HDB e-service, you can derive the official statistics using the criteria:
4 Room, Hougang, May 2010 to Apr 2011
Even for Model A, Aljunied wins Hougang
Average resale Aljunied, 395822.7143, N = 77
Average resale Hougang, 374726.6667, N = 15
Premium = 5.6%
(Readers are free to check that for the premium for 3-Rm New Generation falls to 2%.)
Yet, that there is a premium for PAP-held wards over opposition-held wards should not come as a surprise. After all, the monarchy has been giving more carrots to their faithful adherents.
See ‘government grants for the town council’ chart in
http://geraldgiam.sg/2010/07/opposition-wards-achieve-more-with-less/
Indeed, it would be a grave scandal if there is no or low premium. The question should be framed in terms of the bang for the extra buck.
It occurs to me as I look at the data from the facebook and NUS analysis, pricing of a long-term asset such as housing usually has to take into account expected future discounted returns. In other words, the probability of the ward falling in opposition hands (or returning to the monarchy) matter. If you expect Aljunied to fall into opposition (and the monarchy will make the rebellious citizens ‘regret and repent’), you as a buyer will take this into consideration in pricing. There is no point paying the premium if the opposition is not going to upkeep those impressive statues of the monarchy.
Indeed, the market has been known to be a better predictor of election outcomes than polls. You vote with your money.
Looking at pricing trends since 2006, Aljunied resale prices are not growing as fast as Hougang resale prices. This points to the market losing confidence in PAP’s hold on Aljunied. (Conversely, it can suggest that the market thinks that Hougang will return to the monarchy’s embrace.)
However, insufficient data preceding opposition days makes housing prices at best an interesting conjecture at election predictor. Maybe whoever who works at HDB can dig up the ancient data to run the analysis. It will help the monarchy a lot to pre-punish those rebellious voters.
If you have worked in the civil service you wouldn’t be surprised. This is probably one of the most common occurences – massaging the numbers, to suit your boss’s requirement.
Surely you recall that other MBT HDB stats. which NSP’s Hazel Poa commented on!
So HDB not only build, it cooks too!
I’m just curious. Although comparing price per square meter allows a direct comparison between square meters, wouldn’t the total size of the flat factor in a bit when a buying price is offered?