PDA

View Full Version : Understanding property indices



reporter2
01-12-14, 23:03
http://www.businesstimes.com.sg/real-estate/understanding-property-indices

Understanding property indices

Different indices can paint a conflicting picture of property prices.

By Lynette Khoo

[email protected]@LynetteKhooBT

1 Dec


Keeping tabs on property price movements is crucial on several fronts.

For many, residential property is their single largest purchase, for which they require accurate and reliable price information. Property is also a major component on households' balance sheets.

Policymakers use property price movements as one indicator of macroeconomic activity, as well as to measure risks in the financial system; similarly, banks and insurers use property values to underwrite mortgages and risks.

In Singapore, government agencies such as Urban Redevelopment Authority (URA) and the Housing & Development Board (HDB) have been providing quarterly updates on private and public home prices respectively.

But with the need for timelier information, some private sector players have created newer indices that are updated on a monthly basis. While this should be good news for everyone, since we now have more price barometers to refer to, the irony is that investors might not be the wiser for it.

This is because different indices use different data sets and methodologies. As a result, they can differ from period to period, leading to a confusing picture on what is happening to property prices.

Between 2008 till now, there were at least seven quarters of conflicting price signals between URA's non-landed residential price index and a similar index by the Singapore Real Estate Exchange (SRX) that tracks the same market.

According to the European Union's statistical office, Eurostats, the use of monthly data may lead to rather "noisy figures", whichever method is used to compile the index.

Comparing property prices is not easy. This is because no two properties are identical. Each has different attributes such as location, floor levels, and lease tenure. But these characteristics should ideally be kept constant to measure the actual change in prices over time.

Collecting data is also challenging. Property prices are negotiated and only known after the units are sold. Sales of certain property types are infrequent, making it hard to derive their market value.

Besides the need for a robust mechanism to gather data, the next key thing is the choice of methodology to create a representative index. Each model will have its pros and cons.

It is hence important for users to understand the indices better, know how to interpret them and recognise their limitations.

Here, we'll look at some of the key methodologies that are currently being used.

URA'S STRATIFICATION APPROACH

The URA private residential price indices are based on the stratification or the mix adjustment method.

Simply put, the transacted housing units are sorted into different baskets based on their key attributes such as location, lease tenure (freehold or 99-year), property type (landed or non-landed) and completion status (completed or uncompleted).

The median prices of each basket is calculated. Each basket is then weighted based on the transacted values of the past 12 quarters in relation to the overall market.

To illustrate, let's assume that housing units in Bishan accounted for 10 per cent of the overall market based on the past 12 quarters of transacted values. The Bishan basket will have a 10 per cent weighting on the index. This weighted scheme ensures that baskets of higher transaction values have bigger weightings on the overall index.

URA publishes the overall private residential index as well as a slate of sub-indices covering landed properties, non-landed properties and their sub-types.

The indices capture all new sales by developers as URA collects the data from a developers' survey, and more than 80 per cent of sub-sales and resales. This is because data on sub-sales and resales are derived from caveats that purchasers lodged with the Singapore Land Authority and these are non-mandatory legal documents.

SRX'S 'HEDONIC' APPROACH

Since May last year, SRX - a digital platform of StreetSine Technology Group - started publishing indices that are based on a method called hedonic regression. The holding company recently became a 60 per cent subsidiary of the Singapore Press Holdings.

The conventional hedonic approach tries to figure out how much value buyers will place on each characteristic of a property. Simply put, hedonic regression uses an algorithm formula that takes into account all relevant property characteristics.

StreetSine says this index is the first to take into account uniquely Singaporean factors such as a property's distance to a top primary school or an MRT station, in addition to the standard factors such as location, age of property, size, floor levels and lease.

The SRX Property Indices (SPI) track non-landed private residential sales, resales, non-landed private rental, and HDB resale transactions.

According to its founders, Streetsine's data is more comprehensive as it uses pre-caveat information from 14 property agencies to supplement the officially lodged URA sales data. The risk of pre-caveat transactions being aborted is also at a low 1 per cent. Outliers are filtered from the overall data.

While this method makes the most efficient and timely use of available data, it faces the same disadvantages as other hedonic price indices. Because the index runs on a formula, previous index levels may shift when new data is added over new time periods.

This could be a key bugbear for official statistical agencies, which may be reluctant to accept continuous revisions of previously published figures.

NUS'S 'BASKET PLUS HEDONIC' APPROACH

Four years ago, the NUS Institute of Real Estate Studies came up with an index suite to track monthly movements in private residential prices using a hybrid model, which it flagged as an improvement over conventional methods.

Unlike the traditional repeat-sales method that tracks only resale units, the NUS SRPI (Singapore Residential Price Index) tracks a fixed, pre-determined basket over time - hence taking into account first-time sales too.

The NUS SRPI also uses hedonic regression to estimate prices. Since it tracks a fixed basket, it is not affected by any change in the mix of properties traded.

Its basket comprises private condo projects that are completed (received their temporary occupation permit, or TOP status) in the past 10 years, with projects with fewer than 40 units left out since they are less liquid. This means the index tends to track more resales than new sales by developers.

The index is computed based on the market value of this fixed basket of properties. Each month, the values of these properties are marked to market according to transacted prices of units observed. For untraded properties, a mark-to-market algorithm is used to estimate their prices based on sales signals observed in their vicinity.

NUS also has a sub-index to track shoebox units, which it defines as having a size of 47 square metres (506 square feet) or less. This allows for an additional assessment of how this segment, which typically has higher per square foot pricing, can affect the overall index.

Every two years, the basket is created anew to comprise private condo projects that received TOP status in the preceding 10 years.

NUS hopes that this index will serve as a reference index for the development of property derivatives in Singapore, though it will probably take a while before these financial instruments spring up here.

WHAT'S NEXT?

So what exactly is the best method to measure price movements in Singapore's residential market?

This may be a difficult question to answer.

Some larger markets elsewhere are using the repeat sales method which tracks resale transactions: the most well-known repeat-sales index is the Standard and Poor's/Case-Shiller Home Price Indices in the US that are computed for 20 cities. Residex and the UK Land Registry also compute repeat sales indices for Australian cities and for the UK, respectively.

However, the volumes of transactions in minuscule Singapore may make the repeat-sales method untenable in the local context.

In 2011, three NUS dons recommended that "a matching approach" be used for private residential price indices here. This involves pairing each property in the base period with a comparable unit in the subsequent period.

Professors Deng Yongsheng, Daniel McMillen and Sing Tien Foo argued in a paper that this approach produces a much richer set of results than is available using either the repeat sales or standard hedonic approach.

Compared to a repeat-sales method, the sample sizes in the matching approach are much larger and compared to a standard hedonic method, the matching approach has no statistical complexity and is less sensitive to outliers.

Meanwhile, Eurostats and Bank for International Settlements have been promoting the hedonic method. It is unclear if URA will eventually go down the hedonic route for its property price indices and how it will mitigate the downsides of this method.

BACK TO MICRO-DATA

At the end of the day, no matter how robust the index is as claimed by its creators, its broad average figure may deviate from the actual experiences of particular individuals at some points in time.

The price dynamics could vary widely across locations or projects. Suppose you are eyeing a condo in a particular location where prices are still holding up and where sellers refuse to lower their prices, you'll probably be scratching your head over the falling price indices.

Buying a property can also be a complex and emotional affair. Suppose you need to buy a home because you are getting married soon, these indices will have little influence on your considerations.

For investors, it is tough to time the market where property purchases are concerned given that buying and selling of homes, unlike the trading of stocks, takes time. Inflection points also tend to be visible in hindsight.

Property price indices still serve as a good guide in negotiating prices. But any serious investor will drill deeper into the micro-data, such as prices and rental yields in specific locations and projects for specific-size units before deciding whether it makes sense to buy.

There are other market indicators that property buyers should also pay attention to.

In particular, the volume of transactions relative to the stock in housing markets is a crucial indication of demand and has proven to strongly correlate with house price indices.

Some say that even the number of advertisements on properties for sale in the SPH classified ads can be a good reference, going by how property ads tend to shoot up in a hot market.

Property buyers should also be looking at forward-looking indicators such as sentiment surveys on the property market.

The Real Estate Sentiment Index (RESI) - jointly developed by the Real Estate Developers Association (REDAS) and NUS - measures the perceptions and expectations of real estate development and market conditions in Singapore. A further weakening of the future sentiment index, for instance, will reflect a gloomy outlook among developers in the next six months, which may translate to their pricing strategies.

Lastly, I would add that it pays to keep abreast of economic conditions to understand if current price levels are consistent with fundamentals. In 1997, the Asian Financial Crisis caused property prices here to take a dive. It was not until 2010 before they recovered back to their 1996 highs. Taking a longer-term view offers a different perspective.

AdoLee
04-03-16, 04:35
well this is one year old post to which I am replying but I want to know if I should buy the house before this financial year ends or later as if would there be any effect on the price of home loans, and what is the best method of payment, CPF or Loans.

AdoLee
24-03-16, 05:22
Is there any forum for real estate on which people are active???
I don't think people use forums much in Singapore.

Arcachon
25-03-16, 22:24
Is there any forum for real estate on which people are active???
I don't think people use forums much in Singapore.

Use to have a number of Guru at skyscrapercity forum. http://www.skyscrapercity.com/forumdisplay.php?f=333

Arcachon
12-07-16, 21:46
well this is one year old post to which I am replying but I want to know if I should buy the house before this financial year ends or later as if would there be any effect on the price of home loans, and what is the best method of payment, CPF or Loans.

financial year ends or later - if you are young then no different, if you are old every year is one year lesser loan for you.

If got lot of CPF than using CPF better than loans, but when the interest of the loan is less than CPF than loan is better.

Property Index after all the Control Measure is no longer correct Property Index, it should be Property after Control Measure index.