Late last week, Mr Salty received a link to an article by a famous Singaporean financial blog. The title was something like, “Will Property Prices in Opposition Wards Drop or Appreciate Less?” After reading the article, Mr Salty was like, “This guy talking cock.”
Using data from data.gov.sg, Mr Salty plotted the graphs of the 4-rm HDB median resale prices from 2011 – 2019 for Serangoon and Hougang, which are assumed to be mostly located in the opposition-held wards. He compared this to the numbers at the national level. All graphs followed a general trend of decline from 2011. However, the drop in prices for Serangoon and Hougang was at least 3.5 times more than that of the national level. This is vastly different from the conclusion made by the author of the reputable finance blog.
What Was The Controversy?
In the article, the Talk Cock Guy (let’s call him TCG for convenience) showed a graph depicting HDB 4-rm resale median prices (which is Mr Salty’s domain, considering all the posts he had written about the topic) and did a comparison between the flats in the opposition wards (a.k.a WP) vs the ones in the incumbent (a.k.a PAP).
TCG said at the start that he wanted to use the case of Aljunied GRC to explain to Sengkang residents whether their property’s prices will drop after WP won the GRC in GE2020. He used data from HDB’s website, which contained median resale prices from 2007. Since Aljunied GRC was won over by WP in 2011, the data from HDB can tell a good story of the situation in Aljunied GRC after the switch.
Logically, Mr Salty looked at the graph plotted by the author from 2011 onwards. While there were ups and downs, the general trend was DOWN.
However, TCG concluded his whole article by proclaiming the prices went up.
Huh?
Mr Salty re-read the whole article 3 times, before finally realising that TCG made the conclusion based on the whole time frame of 2007 – 2018.
To that, Mr Salty thought, “You might as well set the start year as 1819. The home prices of Aljunied would have risen infinitely.”
Frankly speaking, the blog was one of Mr Salty’s favourite references when he was writing his finance and home-related posts. He thought the blog was comprehensive and neutral in its writing. Therefore, he was shocked to see how partial TCG was when drawing the conclusion from the graph.
The Graph In Question
The graph plotted by TCG did not, in itself, tell a very clear picture. In fact, many people who are good at reading graphs would say at the first reaction that more work needs to be done.
It is possible to describe the graph, but definitely not draw a meaningful conclusion from it. Therefore, it is very puzzling how TCG could have made such a rash conclusion.
To avoid “copyright issue”, Mr Salty reproduced the graph published by the blog. He used data from data.gov.sg, which is a central repository of all the data from government agencies. In other words, the data downloaded from data.gov.sg was the same as what TCG downloaded from HDB’s website.
As the data was the same, Mr Salty also faced the same issue when deciding how to categorise the data into the GRCs. Let’s just say that the boundaries of the GRCs are very much different from the boundaries of towns defined by HDB and there is no readily available data regarding GRCs’ boundaries that could help filter the HDB data efficiently.
In human language, Mr Salty also did an agar-ism like TCG.
He assumed that the HDB’s definition of Hougang was referring to the Hougang SMC, while Serangoon town is mostly located within the Aljunied GRC. Mr Salty couldn’t figure out how to identify Potong Pasir SMC from HDB’s data, so that data was just left out.
With that, Mr Salty reproduced a graph that was as close to what TCG had published.
Chart 1: Hougang & Serangoon 4-rm HDB Median Prices, 2007 – 2019
Setting The Right Time Frame – 2011 to 2019
As mentioned, Mr Salty found it absurd that the author had drawn a conclusion based on data from 2007 – 2019. Indeed, a longer time frame helps us see a long term trend, but that was not the purpose of producing the graph.
Our purpose was to answer the question, “Will my HDB price drop after the Opposition takes over?”
Whenever you have such a question trying to figure out the effect of a trigger event, you look at the situation at the point of the event, immediately after the event and some time after the event.
The first 2 serve to observe if a trigger event did result in a change. For example, if Mr Salty slaps Mr Sleepy, he neds to observe Mr Sleepy’s face right before and after the slap. If the face turns red after the slap, we can say the slap most likely caused the redness. If not, we can never be sure slapping someone’s face will cause their face to turn red.
Making observations for some time after the event serves to determine how much of the effect we see after the trigger event was caused by the trigger, and whether other factors were at play at the time of trigger.
In human language, it means if we want to answer the question “Will my HDB price drop after the Opposition takes over?”, we should look at the data between 2011 – 2019 (2019 being the latest, most complete data we can get), instead of starting at 2007.
As everyone who invests in stocks know, when you hold on to a stock for a long enough time, you will definitely see a rise in prices. Therefore, in our analysis, it is meaningless to set the start time 4 years before the trigger event.
Since 2011 was the year Aljunied was taken over by WP, we can observe how the price changes in the 9 years after the take over.
As such, Mr Salty zoomed in to Chart 1 to just focus on the data from 2011 – 2019.
Chart 2: Hougang & Serangoon4-rm HDB Median Prices, 2011 – 2019
Observation of Chart 2
Based on Chart 2, we can see that the HDB median resale prices of 4-rm flats in Hougang followed the same trend as that of Serangoon. The difference between the 2 is that Hougang’s prices are between $20k – $70k lower than that of Serangoon’s in the 9-year period.
We can also observe that after rising for 2 years, the prices started to drop steeply from 2013, before levelling out from 2017.
Even then, this is not a meaningful comparison. We can describe the trend, but we need some calculations and a benchmark to make comparisons to.
Setting Meaning Comparison
We can use the average of the median resale prices of ALL 4-rm HDB flats in Singapore, from 2011 – 2019 as a benchmark for the prices in Serangoon and Hougang.
This makes a better comparison, since there are no 2 towns that are identical. Therefore, we should instead try to see how Serangoon and Hougang is faring, compared to the whole of Singapore (Serangoon and Hougang inclusive).
Chart 3: Hougang & Serangoon4-rm HDB Median Prices Against National Level, 2011 – 2019
Observation of Chart 3
We can see from Chart 3 that the 4-rm HDB median prices at national level follows the same trend as that of Serangoon and Hougang. Serangoon’s prices track very closely to the national level, while Hougang’s prices are markedly lower than both of them.
With the 3 graphs following the same trend, it means that the drop of prices in 2013 – 2017 was not exclusive to Serangoon and Hougang. In other words, we can’t wilfully conclude that the drop in prices in the 2 opposition wards was due to it being managed by opposition MPs.
At the same time, we can conclude that there is something happening at the national level to explain the trend.
Anyone who’s familiar with the property market in the past 10 years will remember how, just 10 years ago, the property prices in Singapore started soaring after recovering from the 2008 recession. From 2011, the government started rolling out several cooling measures to tame the rise in prices. The last most significant cooling measure took place in 2013 (though 2 more, not-so-famous measures were rolled out in 2017 and 2018).
That last cooling measure corresponded to the peak in prices in 2013 as we see in Chart 3.
If anything, the most meaningful conclusion we can draw at this time is that the government has been effective at cooling down property prices.
Drilling Down The Trend
That being said, more can be analysed from the graphs. Although the 3 lines followed the same trend, we can also ask, is one line dropping more than the other?
For that, we plot the trend line in Excel. (Sorry, Mr Salty is not a data analysis, so he does not use fancy data analytics tools)
The trendline, like its name, draws an overall trend for graphs that fluctuate. At the same time, the slope of the graph (the number beside the “x” in the trendline’s equation) tells us the direction of the trend (positive means overall rise and negative means overall declining) and the degree of the rise/drop (the larger the number, the more the rise or drop).
Chart 4: Hougang & Serangoon4-rm HDB Median Prices Against National Level, 2011 – 2019 with Trendlines
Observation of Chart 4
Looking at the x-coefficients of the 3 trendlines, we can see that Hougang had the steepest negative slope (-7238), meaning the prices in Hougang dropped the most since 2011.
Although Serangoon’s prices seemed to track closely to the national level, the trendline suggests otherwise. Serangoon has a steeper negative slope (-2835 vs -817) than the national prices. While the prices across Singapore saw a downward trend from 2011 to 2019, Serangoon and Hougang saw a steeper than national average decline in prices.
The general trend can be explained by the cooling measures mentioned in the last section. However, the trend lines and their correlation to the macro environment in Singapore is up for interpretation.
Interpretation of Chart 4’s Trendlines
On one hand, we can say that houses in Hougang and Serangoon responded very well to the cooling measures set by the Government. A pro-WP commentator can say that the WP MPs had been working very hard on the ground to rein in the runaway prices, so that their residents can cope with the rising cost of living.
On the other hand, an anti-WP commentator can also say that the houses in Hougang and Serangoon are not very high in desirability. When the cooling measures started to kick in, their flats were one of the first to be eout of consideration by potential buyers. In turn, it reflected a disdain for flats in these estates.
Sengkang Residents: Will Your Property Prices Drop After GE2020?
To answer this question, we look at the closest comparison we have: Aljunied GRC after GE2011.
The prices of 4-rm HDB median resale flats of Serangoon, which is located in Aljunied GRC, saw an overall decline from 2011 – 2019. Although the same decline was observed at a national level, the prices of Serangoon fell at a steeper level (about 3.5 times more).
However, hope is not lost.
Like Mr Salty mentioned above, no 2 towns are identical. Serangoon is a more mature (i.e. old) estate than Sengkang, which should be one of the factors driving the steeper drop in prices.
Aljunied was also the first GRC that WP won, so property buyers were understandably more cautious about venturing into the area. If you are worried about the AHTC saga, which had dragged on from 2013 to 2019, and how it reflects the WP MPs’ incompetence in managing Aljunied GRC, be assured that the whole saga was a result of a series of events that took place before 2013.
Since no mishaps have happened since 2013, we can also assume that the MPs have gotten the hang of managing a GRC and can pass on the knowledge to the new MPs taking over Sengkang.
This is what Mr Salty can tell himself to save himself from the verge of breakdown.
And yes, Mr Salty is a Sengkanger.
Bonus
For the fun of it, Mr Salty plotted the median prices of 4-rm HDB flats of all towns in one single chart.The national average is highlighted with a glow. Some super high end areas, like the Central region, did not have data for 4-rm flats, so the total number of towns shown in this chart went from 27 to 23.
Chart 5: 4-rm HDB Flats Median Resale Prices by Town
However, there are simply too many lines to see. Most notably, all but 6 towns were bunched at the lower half of the chart. Therefore, Mr Salty removed the 6 towns that had crazily high prices, so that we can have a good look at the rest of the towns that cost about the same with each other.
Chart 6: 4-rm HDB Flats Median Resale Prices by Town (Minus Top 6)
Serangoon (blue) and Hougang (Brown) were also highlighted, so we can see where they are amidst all the chaos.
We can see that although we saw Hougang had a markedly lower median prices compared to the national average, it is not the town with the cheapest flats. In fact, Hougang is more expensive than flats found in Woodlands, Yishun, Sembawang, Choa Chu Kang and Jurong West – The Far North and Far West.
There were only 7 towns that lie above the national average, and 4 towns that did not have enough data to be shown. In other words, 16 out of 27 (60%) of the prices are below the national average, which goes to show how much the top 7 towns were skewing the graph.
For those who are familiar with statistics, this means that the median of the median prices should be found below the national average, or right about where Tampines and Sengkang is.
Therefore, those who are looking at the national average of prices and thinking that HDB prices are beyond reach should actually look at the median prices. Ideally, you should look at the median prices of individual towns.
Mr Salty will be writing another post that looks at the median prices by towns, and help readers figure out if they can afford the HDB flats in those towns. Before that, please feel free to share this post (and the charts) to argue with your friends who are pro and anti opposition!
Remember, salty is life!
Sources:
Data.gov.sg: https://data.gov.sg/dataset/median-resale-prices-for-registered-applications-by-town-and-flat-type
Disclaimer: Mr Salty, though he has a political preference, did not write this post for political gains. He just hopes that people from both sides of the spectrum will take issue with what he had written in this post and pushes readership for Hello Mr Salty higher. Please help fulfil his wish. Kamsiahamida.