All the fuss about Big Data...

By The Folks @PropTalk - April 2, 2015 3 Comments

The wife and I have been following a series of exchanges that occurred over the course of the past two weeks about the Singapore property price data. More specifically, who has the "better" data.

It all started on March 17 with a column penned by Mr Sam Baker, the co-founder of SRX Property. In his article, he claimed that a lack of pricing transparency is the reason for the property market reaching a point where cooling measures currently in place have cost home owners an estimated $21 billion in lost market value (and counting).

When participants have insufficient information to make pricing decisions, markets fail - he said.

Mr Baker proceeded to expound on the virtues of Big Data and how such will make a transparent pricing possible.

So what exactly is Big Data? This is where SRX Property's X-Value comes into play. Supposedly developed with government agencies, academics and valuers, it sources from the nation's most comprehensive property database and instantaneously calculates a single value for every home using best practices methodologies, including comparable market analysis.

As a computer-driven price mechanism, X-Value factors in all the comparables and adjusts for important variables like location, size, floor and age.

It also factors in macro trends from its proprietary price indices, and important on-the-ground information from thousands of market participants interacting with SRX Property apps and pricing data on a daily basis.

It is impossible for a human, even if he had access to all SRX Property's raw property and geospatial data, to do what a computerised price mechanism does in seconds in terms of accuracy, data completeness, relevancy and objectivity.

And to reinforce the value of Big Data, Mr Baker claimed that 93.8% of HDB homes were transacted within 10% of the X-Value. And SRX Property arrived at the figure of a market loss of $21 billion by using the X-Value to calculate the value of all transacted flats (recorded) at the market’s price peak and then compared it with their values at Dec 31 last year.

Similarly, researchers and analysts can use X-Value to track market movements and estimate gains and losses. Consumers can also use the tool to check on a valuation price of properties they are interested in. Government agencies and research institutions can use X-Value and associated data and analytics to forecast the market and solve problems before they get too big.

So Mr Baker concluded that with Big Data and technology, productivity, efficiency and decision-making will improve. And a property market with a universally accepted and transparent pricing mechanism would make the market more transparent, keeping it on an even keel that reduces the need for cooling measures.

With the article’s innuendo of “my property price data is better than everything that’s out there, including the Government’s”, the wife and I were pretty certain that the other vested parties would not take this lying down. We did not have to wait long.

The Council for Estate Agencies (CEA), Housing Development Board (HDB) and Urban Redevelopment Authority (URA) wrote a combined-response that was published in the Forum page of The Strait Times yesterday (April 1). In their response, they refuted some points raised in Mr Baker’s article:

1.      Government agencies that deal with housing issues, such as HDB, URA and CEA did not collaborate with SRX to develop X-Value.

2.      There is NO lack of transparency in Singapore’s housing market - both HDB and URA have been releasing data on the transacted prices of individual private and public housing units on their respective websites free of charge for many years to help home buyers make informed decisions. In fact, SRX has been relying on such open-source property data from government agencies for its work.

3.      It is untrue that the Government is unable to monitor the prices of properties in different market segments. Government agencies have up-to-date and comprehensive information on individual sale and rental transactions of housing units, and use them to monitor property market trends and formulate policies.

4.      The movement in property prices is due to a complex mix of internal and external economic factors, such as the economic outlook, interest rates, liquidity, market sentiments, and business and investment strategies of industry players. It does not hinge on the availability of a market value for each home alone.

In other words, your (SRX) price data is not any better than ours (HDB/URA) and definitely not more useful!

And in the latest twist to the tale, Mr Barker wrote a reply to CEA/HDB/URA today. In his response (again published in the Forum page of The Straits Times), he acknowledged that raw data is available to the Government and the public. However, raw data does not equal a transparent pricing mechanism for the property market.

Mr Baker akin the free transaction data on the HDB and URA websites as rice – it must be processed and combined with other ingredients to form something meaningful, like sushi. The problem with raw data is that it must be processed by users with different levels of expertise, or no expertise at all. Furthermore, even experts can misinterpret or manipulate raw data.

Big Data, on the other hand, IS sushi. Again referencing to SRX Property’s X-value, this standardizes market pricing by applying best-practices methodologies to process raw data from multiple public and private data sources, and instantaneously computes and disseminates a single value for each unit. With X-Value, all Singaporeans start from the same page – meaning the property market joins other markets in having a transparent pricing mechanism.

So in the words of Mr Baker: We don’t have to settle for rice. We can eat sushi!

The wife and I wondered if the latest development will signal an end to the whole debate (probably not). We also wondered if yesterday’s announcement by URA to revise their Property Price Index to take into account attributes such as a property’s proximity to MRT, size and age was (at least in some small ways) prompted by Mr Baker’s original commentary. But from what has transpired thus far, the wife and I will like to give our two-cents on what we can only deem as a "pissing contest":

1. While any attempt to understand the complexity associated with deriving the X-Value may be too much for our simple minds, we are quite certain that this involves some kind of formula/algorithms that require weightages to be assigned to the variables used. And how/why these weightages are assigned will surely introduce some level of subjectivity to the X-Value.

2. The wife and I do not believe that there is one single set of property price data/index that is the best or should be used as "the standard" for all references. Based on our experiences, each set of data produced by the various institutions/agencies has their own merits (and deficiencies). Only when compared together instead of relying on any single set of data/index will enable the user to make the most informed judgment on how the market is behaving.

3. Property price data/index generally provide more of a macro picture. While these are useful as a gauge on how the property market is moving, nobody (at least not us anyway) will base their buying decisions solely on the price data/index. And many a times, sentiments do come into play during property purchase - you may like that condo unit so much (and it has nothing to do with the unit facing, proximity to MRT etc.) that you are prepared to pay way above what the X-Value had deemed as "fair value".  

And finally, on the subject of rice versus sushi: the wife and I reckon that this is really a case of "to each his/her own". While the wife and our little son are huge fans of sushi, I will take my bowl of rice any day! I will probably complement it with some chicken, small helping of vegetables and probably a fried egg. But I will not have everything combined into a single rice-ball and then claimed that this is the one-and-only type of food that anyone will need. 

Then again, that's just me. We can always agree to disagree! 

3 comments to ''All the fuss about Big Data..."

  1. At any rate I would imagine this would be good publicity for SRX that can only have the effect of attracting more customers to them? I'm assuming their price data is not free..

  2. SRX is just trumpeting that they have a better system just like any retailer with a new product would. It is still the consumer's prerogative whether to follow the data, or more specifically the X-value when making the decision to buy a property. Logically speaking, just like when u check valuation for any property from the bank, you don't just rely on one to give u the figure but a basket of valuers/banks to derive a figure for yourself. So it is to the benefit of the property buyer to have many sources such as X-value, URA, squarefoot research, etc to help the buyer conclude the correct figure for himself. He should be his own judge, just my 2 cents worth.

  3. SG Floorplans: Thank you very much for your "2 cents". The wife and I would like to think that we are not just logical but realistic beings as well. So realistically speaking, your might not be able to obtain as many valuations as you would have wanted due to costs and other practical constraints. And if you are looking to take a housing loan, you would probably be stuck with whatever valuation figures that the chosen bank provides.

    But we do agree that in a perfect world, you should be getting all available valuations to make comparisons. And even in our not-so-perfect world, we should make the best use of whatever info that we could obtain within our means (and not just the one that claimed to be the best in the market) to make an informed judgement. Cheers!