Wednesday, April 28, 2010

Statistics on House Prices

Another graph from the Minister for National Development to show that house prices have not outstripped income growth. Another careful selection of base year. This time it is 1995. The graph is presented despite the absence of 1996 income data, resulting in an incomplete and rather odd looking chart. [ Cannot find another appropriate base year is it? ]

Sigh. I could of course as someone suggested show the charts of all the different base years, most of which would show the Resale Price Index (RPI) increasing more than the median household income(MHI) , but I fear that would make me appear rather juvenile. Many netizens already know the truth. As others have already pointed out, using all the others years from 2000 to 2008 as base years would show RPI increasing faster than MHI. This can be easily proven if challenged.

And, judging from the fact that the chart presented by M for ND is a since-1995-but-minus-data-for-1996-due-to-lack-of-info chart, I will hazard a guess that using 1998 and 1997 as base years would similarly show RPI outstripping MHI. Reasonable guess you think?

And for those who clamour for longer years of data, Lucky Tan has a chart going back to 1990.

But enough of charts. My intention when I blogged about the misleading chart was to show that there are many ways of interpreting and presenting statistics. Charts for 2000, 2001 and 2006 were chosen as examples to illustrate that point, not to advocate them as suitable base years. We need to look at data more holistically. Statistics should be used to understand, not to justify.

In fact, there are more problems with the comparison of RPI vs MHI apart from the base year:

1) Other netizens have pointed out that the RPI does not take into account the differences in sizes of the same flat type over the years. As flats get smaller over the years, the price in terms of per-square-foot (psf) basis increases by more than that indicated by the RPI.

2) According to the Department of Statistics:
“Household income from work refers to the sum of income received by all working members of the household from employment and business but excludes the income of domestic helpers. For statistical purpose, a household refers to a group of persons living in the same dwelling unit and sharing common living arrangements. A household may comprise related or unrelated members. Resident households are households headed by Singapore citizens or Permanent Residents. This category includes employed households and households with no working person.”

This would mean that household income would increase without any increase in individual wage levels under the following circumstances:
a) More households with both husband and wife working
b) More working children staying with parents for longer periods due to later marriages or no homes of their own
c) Later retirement
d) Relatives without homes of their own moving in
e) Renting out of rooms to other working adults. The incomes of the tenants are included in the household income as well according to the above definition.

All of these make it difficult to understand the people’s pain if we merely look at RPI vs MHI. In fact, this creates a vicious cycle whereby if prices increase, making homes more unaffordable, we will have more people crammed in each household therefore pushing up the household income, which then seemingly justify the higher prices!

Hopefully DOS will start to release / collect data on median income for individuals and use that for comparison instead. DOS currently publishes data on average income of individuals, but average values can be skewed by extremes and hence would not be ideal.

I am happy to read that HDB will be looking at shortening the waiting time for new flats by moving away from the BTO scheme. Much of the frustrations of home buyers arose from the fact that waiting time for new flats is too long and prices of resale flats are too high, making them feel sandwiched between a rock and a hard place. The long waiting time also channels demand towards resale flats, applying upward pressure on prices. If this can be done, it will make the lives of many young couples easier.


  1. Try MOM for employment income:

  2. Hazel

    First I admire your courage for standing as an opposition. You come across as a compose & confident & intelligent person.

    Smart people are everywhere. Smart and brave ones are rare. I hope you succeed to instill changes to our system, for our next generations sake.

    May you stand firm on your ground. May you & your family receive all support needed.

    May God favours you.

  3. Mah trivialises the concerns of the people, dismissing it repeatedly until cornered with facts, and then continues to wiggle out with another graph attempt.

    A pictorial representation of how these graphs actually paint a different story except graphs of only two particular years (mah's graphs) is a juvenile representation, which makes it easy to understand and have a strong visual impact, how Mah sees the intelligence of the audience by showing us his 2 graphs. What is juvenile is not the plotting of multiple graphs but the reflection that all this was triggered by a minister who looks upon the population as 3-year-olds to even attempt to show his 2 graphs.

    a way to multiply-plot all the years in a single graph would be to have one anchor line in strong colour - the resale price index,
    then from that line, at different years, branch out faint lines of income.

    you will see that only two faint lines end up higher than the strong line - mah's years (base year 1997 and 1999).

    all other lines do not behave like that.

    mah's base year was 1997 (the year the two lines intersect), but his failure to understand the base year concept led him to say the base year was 1995, where data was not revealed and claimed unavailable. or he could blame the reporter for misquoting him.

  4. Here are more views on public housing and its sustainability from a singaporean.

  5. re: futurehdb

    'The great divide…one side prefers using a longer period while the other side prefers to look at recent history.'

    incorrect. the difference is not in the length chosen (longer versus shorter) because on one side, you have 1997, 1999, on the other side, you have 1998 2000 2001 2002 2003 2004 2005 2006 2007 2008.

    Notice that 1998 and 2000 are there too, therefore it is not about length.

    Instead, it is about how, out of 12 years, only 2 get chosen in an attempt to distort the facts. if it was purely accidental and unmeditated, Mah might have "accidentally" chosen 1998, or 2000, or 2001. In fact, Mah said he wanted to compare "one decade" which meant that he should have chosen 2000, not 1999. But "magically" it turned out that his two choices of years, 1997 and 1999, coincided with the fact that only these two years can produce the kind of graph he wants.

    Therefore, the facts are plain for all to see, when a person manipulates with statistics, the motive behind this manipulation.

  6. 敬佩!敬佩!佩服你的勇气,你的真知灼见,一针见血地刺穿了上级的谎言。

    那匹高高在上的“马”,如果仍然一意孤行,只为主君服务,不为人民服务,可能即将成为新加坡的第二匹“跑马“, 落选的“跑马”。


    PAP已经不是以前的PAP了。现在的PAP,是想尽办法向人民牟利,花言巧语,什么“市场津贴”,(MARKET SUBSIDY ) 根本就是一种手段,要人民终身负债!!

    政府要人民负担沉重的债务,用意何在??我们是在计划建造航空母舰,成为本区的超级强国,还是政府早已在海外投资失利巨大, 但仍然好大喜功,要人民分担损失,连我们的养老金也要抽干,向外国老嚣张,是吗??太可恶了!!!


  7. Hello there,

    Your point 1) is valid. This has been a similar problem in Australia with the use of median prices to gauge affordability for a long time, since the definition of a "median" home in an area has changed greatly over the years.

    Point 2) about the changing nature of household incomes is also valid; in fact, the affordability situation here is such that women are being encouraged to join the workforce to boost household incomes.

    Basically, as you have mentioned, statistics are meaningless if we do not understand exactly how they are derived and the underlying implications.

  8. HDB will be looking at shortening the waiting time for new flats by moving away from the BTO scheme. Much of the frustrations of home buyers arose from the fact that waiting time for new flats is too long and prices of resale flats are too high, making them feel sandwiched between a rock and a hard place. The long waiting time also channels demand towards resale flats, applying upward pressure on prices.

    It sure took a long time to discover the pitfalls, even for an average waged officer

  9. Dear Hazel Poa Koon Koon,

    I am a student at the university of Amsterdam. At this moment I am in the end phase of my study and have started to conduct research for my thesis to complete my master in political science. In my thesis and related research I aim at analyzing blogs in Singapore that focus on Singaporean politics, including for instance political issues, political system critics etc.

    Is it possible that you would cooperate with me and that you answer some questions I will send to you? The questions will be about blogging in Singapore.
    It is very important for me that enough people respond, because otherwise my data are not complete and I can’t write a good conclusion based on enough respondents.

    Thank you very much for your help!

    Yours sincerely,

    Sarah Nienke van Voorthuisen

  10. Additionally, any numbers presented are back-up by certain assumptions and premises. Affordability is always argued and defended that the total mortgage debt should not exceed certain proportion of household income.

    That is only partially correct to some extent base on a very important premise that all other things are held constant throughout the entire mortgage servicing period, ie, constant interest rate regime and meaningfuly employed throughout the entire mortgage period.

    At present economic environment, interest rate is low and housing price is high and continues to rise. Base on current low interest rate regime and if we can hold this low interest regime constant, then, the affordability risk issue would be low and the argument is correct. What if the interest rate were to rise by 2% to 3%, which it will very soon be because global economy cannot sustain a borrowing cost than is lower than the official inflation and at near zero rate. Today, Sibor is around 0.3% to 0.5% while inflation rate is more than 2% in Singapore??.

    In our parents' generation where interest rate regime was high and household income was low, mortgage loan was paid off fully within 5 years. Now, we have a low interest rate regime and more household income as compared to our parents' times but mortgage period is being dragged outward to 30 years. How do we reconcile this?

    To sum up, there are two major housing risk that will affect the future affordability issue and not the present affordability issue. When interest rate rises, servicing burden increases and many people may look to sell off the property and downgrade. This will contribute to the decline in property price as a result of a forced and unintended rise in housing supply in the market versus demand. Who will buy house when mortgage interest rate is at 5% and at current high housing price? Second, If interest rate remains incredibly low, then, we will expect more frequent boom-burst economic cycles because the economy is driven much by excess liquidity rather than real sector growth and innovation. Either way, we are caught in a double whammy web and can we say that affordability is not an issue then?

    Just my viewpoints.