20 March 2015

Is Public Housing Really Affordable?

Minister for National Development Khaw Boon Wan tried very hard to persuade us that public housing is more affordable now than it has been in recent years[1].






According to Mr Khaw:


 ■ Between 2009 (after the onset of the global financial crisis) and 2014, prices of HDB resale flats increased 37 per cent while median household income increased 38 per cent.

■ Between 2005 (two years after SARS) and 2014, prices of HDB resale flats increased 87 per cent while median household income increased 72 per cent.

■ Between 2009 and 2014, prices of HDB BTO flats in non-mature estates increased 15 per cent (without grants) or 6 per cent (with grants) while median household income increased 38 per cent.

■ A couple earning $4,000 per month could buy an average 4-room HDB BTO flat in a non-mature estate in 2014 for $295,000 (without grants) or $260,000 (with grants).

■ 90 per cent of 3-room HDB BTO flats were sold below $250,000, 81 per cent of 4-room flats were sold below $350,000, and 89 per cent of 5-room flats were sold below $450,000 in 2014. (Mr Khaw did not say whether he was talking about all BTO flats or BTO flats in non-mature estates.)

■ Families earning $1,000 per month can afford to buy a 2-room flat; $2,000 per month, a 3-room flat and $4,000 per month, a 4-room flat, assuming they service their 25-year mortgage loan "mostly" from their monthly CPF contributions.

Statistics — Resale Price Index
The following table shows how HDB's resale price index[2] and the median resident employed household income from work have increased over the past five to 12 years, ending in 2014:


Mr Khaw appears to have chosen his statistical data to support his case.


The median resident employed household income from work lagged the increase in the resale price index by as much as 30 percentage points for the period from 2007 to 2014.


It should be noted that HDB decided last year that its resale price index wasn't good enough and needed to be refined[3].

Previously, the resale price index was computed using the stratification method, with a representative basket of towns and flat models. Resale prices were stratified into segments based on flat types, models and regions. The average prices for each segment were aggregated using 12-quarter moving average weights to derive the index.


With effect from Q4 2014, HDB has adopted the stratified hedonic regression method to control for variations in attributes of the flat, e.g., proximity to amenities, age and floor level, through a hedonic regression, and derive the general price movements in each segment. These are aggregated using 5-quarter capital value fixed weights to derive the aggregate price change.


Statistics — BTO Flat Prices

It is not clear to me where the public can access the historical information on the statistical distribution of prices of BTO flats, from which Mr Khaw derived his data.

Median Household Income

Whilst Mr Khaw did not make it clear, the median household income that he referred to is the median resident employed household income, including employer CPF.

Comparing BTO flat price or resale price with the median resident employed household income to prove affordability is flawed.


Firstly, the relevant household income is the combined income of the members of the household that is being used to service the mortgage loan. The income of other members of the household e.g., their adult children, cannot be taken into account if it is not used to service the mortgage loan. Mr Khaw probably recognises this because he chose to use the median resident employed household income, rather than the median resident household income (which is the lower of the two because it includes households that do not derive any income from work).


Secondly, the BTO flat price or resale price that is relevant for any household is the price at the time the household buys the flat, not the prevailing price. Comparing the prevailing price with the prevailing median employed household income is misleading and serves no purpose.


Thirdly, 19.3 per cent of resident households in 2014 do not live in HDB flats. Since the average household income from work of employed resident households is significantly higher for non-HDB households than HDB households[4], the median resident employed household income overstates the median resident employed household income for HDB households.


Housing Loans

HDB's criteria for housing loans


 ■ Maximum loan quantum of 90 per cent of the purchase price or, in the case of a resale flat, the lower of 90 per cent of the resale price or 90 per cent of the market value;

■ Maximum loan repayment period of 65 years minus the buyer's age or 25 years (which was shortened from 30 years in August 2013), whichever is lower; and

■ Monthly instalments capped at 30 per cent of the buyer's gross monthly income (which was lowered from 40 per cent to 35 per cent in January 2013 and to 30 per cent in August 2013)

prevents buyers from buying flats beyond what Ministry of National Development considers to be financially prudent. In other words, HDB flats became more affordable because buyers could no longer get mortgage loans that the Government considered to be unaffordable. Affordability (however defined) was forced on buyers.

Retirement Adequacy

A $100,000 25-year mortgage loan at an interest rate of 2.6 per cent requires total payments of $136,000. The interest rate of 2.6 per cent is based on the minimum interest rate of 2.5 per cent for CPF Ordinary Account plus 0.1 per cent. It cannot fall below 2.6 per cent, but can rise above 2.6 per cent.

Even if a household can service its mortgage loan with most or all of their CPF moneys, it is not the end of the story.


There are competing demands for a family's CPF moneys. Every dollar spent on servicing a mortgage is a dollar less for retirement and hospitalisation and other medical expenses.


In addition, the more of a family's CPF moneys that are tied up in its HDB flat, the more the Government cannot allow the housing market to decline.


Even then, the Government cannot fight the inevitable fact that HDB flats have zero value at the end of 99 years, unless it keeps re-housing families through expensive en bloc exercises.


---------------

Notes

1. KHAW BOON WAN Committee of Supply Speech (Ministry of National Development) 10 Mar 2015.


2. There are at least three possible ways of presenting the HDB resale price index:

■ The resale price index for Q2.
■ The resale price index averaged over four quarters ending in Q2. This corresponds to the resident employed household income from work, which is derived from the June Comprehensive Labour Force Surveys. This is the basis of the table presented above.
■ The resale price index averaged over four quarters ending in Q4 i.e., over the calendar year.

These are shown as follows:


It is not clear how the Ministry of National Development computed its data points for Mr Khaw's presentation. It is possible that its annual resale price index is not the arithmetic average of the resale price index over four quarters (whether ending in June or December), but is weighted by the volume of resale transactions in its model.

3. HOUSING AND DEVELOPMENT BOARD Revision of HDB Resale Price Index 9 Dec 2014.

4. DEPARTMENT OF STATISTICS SINGAPORE Key Household Income Trends, 2014.

No comments:

Post a Comment