Malaysian Property: Bubbles AND Gluts?

Via the Malaysian Insider, Dr Dzulkefly Ahmad of opposition PAS talks about the property bubble (emphasis added):

Of Property Overhang and Mounting Household Debt

Property Overhang as reported by Napic (National Property Informational Centre) for the third quarter of last year was supposedly a cause for alarm. Then there was an overhang of 20,286 residential, 5,450 shop and 619 industrial units worth a whopping RM5.3 billion.

Of the 6,401 new residential units launched during the third quarter, which was a far cry from the 14,588 units launched in the previous corresponding quarter, only 20.2% found buyers…

…The Finance Ministry also reported another 44,954 residential, 4,605 shop and 794 industrial units were under construction as of the third quarter. Projects approved but yet to be implemented comprised another 14,993 residential, 1,011 shop and 872 industrial units…

…More recently, Napic’s Property Overhang reports show that unsold properties in Malaysia rose to 22.6 per cent of new launches in the second quarter of this year, from 19.5 per cent in the fourth quarter of last year. For Kuala Lumpur, unsold properties rose to 16.1 per cent from 15.8 per cent, while for Selangor it rose to 14.6 per cent from 12.4 per cent.

Quite ironically while a glut is emerging, prices of residential property have surged by as much as 35 per cent in the past year, far above income growth and giving rise to concerns that the market is becoming unsustainable.

Checks on developments completed this year also show that vacancy rates remain at 50 per cent or higher…

…On the back of the emerging property bubble, the revelation of a ballooning household debt, comprising mainly house mortgages, cars loans and personal financing such as credit cards, debit cards which stood at an all time high of RM560 billion as at Aug 31, 2010 from Bank Negara Malaysia data is surely a cause for alarm…

…It is worth noting that our household debt to GDP ratio shot up to 76% between 2004 and 2009 and is the highest in Asia, except for Japan. But as a reminder, Japan’s per capital income of US$32,700 or about RM8000 per month in 2009, while Malaysia’s average income per capital is less than RM2000 per month.

According to a note by CIMB Research, the ratio of household debt to personal disposal income hit 140.4 % in 2009- higher than Singapore 105.3% and the US 123.3%. This means Malaysians owe double the amount they earn…

…Since Malaysians tend to have short memories and have a penchant for the denial syndrome, it’s pertinent to remind our political leaders and planners of the costly deflation of the property bubble in the aftermath of the 1997/98 regional financial crisis.

When the property bubble burst, the banking system was left with RM51.8 billion worth of non-performing loans (NPLs), forcing the government of Tun Dr Mahathir Mohamad to form Danaharta Bhd to assume the NPLs and Danamodal to help recapitalise the banks.

…Going by the orgy of real estate developments in recent years, it is clear that both the regulators and the developers have forgotten the 1997/98 lessons or have not learnt much. Notably the government is discouraged to put a higher real property gains tax (RPGT) and restrictive loan-to-value (LTV) caps as it will be a deterrent to foreign direct investments and high net worth individuals. Besides, a higher LTV for second and third home buyers and in specific hotspots areas, of say 70:30, may not be complied with after all, by the financial institutions…

I wish more politicians would take the time to cut the data this way – even if it turns out they’re wrong, or mostly wrong as in this case.

Dr Dzul starts out with a thesis and proceeds to try and prove it, but it doesn’t quite hang together. It also resulted in this hilarious exchange in parliament:

Property prices rose only 3.3%

RESIDENTIAL property prices in the country rose only by 3.3% this year, said Deputy Housing and Local Government Minister Datuk Seri Lajim Ukin.

Citing a survey conducted by the Global Property Guide, Lajim said the price appreciation of residential properties in Malaysia was the fourth lowest in the Asian region.

“Among the Asian countries, Singapore recorded the highest increase of 38.1%, followed by Hong Kong at 24.53% and Taiwan at 12.74%,” he told Dr Dzulkefly Ahmad (PAS-Kuala Selangor).

However, Dr Dzulkefly felt the figure was not right.

“The statistics must be wrong. It does not make sense,” he told the deputy minister.

Lajim insisted his information was right and said: “It is not wrong. The Opposition is wrong. We are the Government, and these are the official figures”...

So on the one hand, Dr Dzulkefly is accepting the official statistics with respect to the property overhang, but doesn’t accept the price statistics (which actually come from the same source). As for the Honourable Deputy Minister, I’d refer him to Greece; government statistics; fraudulent. Saying that statistics must be correct because it comes from the government (bearing in mind that data can and have been revised extensively), is rather silly.

Property

But getting back to the issue at hand, there are a number of things going on, which I’ll try my best to explain such as why there’s a “glut” in the midst of a price “bubble”.

First the insight that is missing here is that the property “market” is really a collection of markets, with different demand and supply conditions and intrinsic characteristics. The aggregation of prices under the Malaysian House Price Index hides a lot of heterogeneity (Annual index numbers and log annual changes; 1988-2010; 2000=100):

You’ll notice right off the bat that comparisons with the 1997-98 crash aren’t valid – there hasn’t been a generalised rise in house prices such as happened in 1995-97. The reason is fairly simple – we haven’t had the kind of abrogation of monetary responsibility that occurred in September 1995.

Prior to that date, BNM was intervening heavily in the foreign exchange market, and largely sterilising that intervention. In September however, Jaafar Hussein was replaced by Ahmad Don as Governor, and Anwar Ibrahim became Finance Minister. Intervention continued, but this time without sterilisation. As a result, money supply and loan growth exploded into the 20%+ range, feeding directly into real estate prices. Who you want to assign blame for 1997-98 depends on your point of view, but I for one am not blaming Soros and hedge funds.

In any case, nothing like that is happening now (log annual changes; 2000:01-2010:09):

As for current residential property prices, how to reconcile the overall slow rate of increase with the anecdotal evidence of double-digit price increases?

We know that most of the price increases have been confined to Kuala Lumpur and Penang and largely to prices of high rises. But since these are just a fraction of the overall property market, it should be no surprise that the overall price level isn’t fully reflecting the price bubbles in these markets (Quarterly MHPI indices; log annual changes; 2000=100):

Note that price increases peaked in 4Q 2009, and have decelerated since – a circumstance that better fits the observation of higher unsold property stocks. Even if prices double in say the Mont Kiara enclave, it’s not going to push prices across the entire market; you can’t meet the demand in KL by building supply in Perlis. That also helps to explain the concurrent observation of rising prices with higher stocks of unsold property – prices are rising in one area, but unsold stocks are higher in others, something the good Doctor notes but somehow fails to grasp the significance of.

More to the point, we need to put the property overhang in its proper context. The growth numbers look bad – until you look at the actual time series (units completed but unsold; quarterly data; 2004:01-2010:02):

…and if you consider the unsold stock in proportion to the total stock (quarterly data; percentage terms; 2004:01-2010:02):

Current stocks of unsold property represent just 0.5% of the entire housing stock, and is slowly coming down in absolute terms. To me, the rise in unsold stocks over the past year look more like seasonal variation and the typical lags that affect residential development, rather than a structural change in the market.

Some more perspective (quarterly data; units; 2004:01-2010:02):

Note the left-hand scale: when completions and sales in any given quarter is in the tens of thousands, a one or two thousand change in unsold properties shouldn’t be unusual.

Just as important, the chart above would tend to disprove the contention that the doctor and some of others hold that property developers are indiscriminately launching new projects – if anything, they’ve been producing progressively less. That fits the anecdotal evidence that developers have become more selective in their project planning since the recession hit. And if you want further proof (quarterly data; log annual changes):

Growth in the total stock of housing has been falling continuously over the past decade. Part of the reason is that, right at this moment, the market at a whole appears to be saturated (quarterly data; ratio of housing units to total population):

There are a couple more things that need to be taken into account here:

First is that there is long term trend of migration from rural to urban areas – the ETP is forecasting the urbanisation ratio (the proportion of the population in urban areas) to rise from 70% to 75% by 2020. That will increase demand for housing in urban areas, while leaving rural areas with a glut of housing. This may be part of what we’re seeing now.

Secondly, there is the demographic transition. As the population ages, demand for housing will increase pretty dramatically as those currently in school enter the workforce. I haven’t formed an estimate of how much, but eventually it will lead to the pop/housing ratio to drop again.

But taking all these disparate numbers together, my opinion is that there’s very little fundamental reason for house prices to rise in aggregate over the short term. Foreign demand can only impact certain urban addresses, but we’re' not Singapore or Hong Kong. Malaysia’s property market is geographically larger, more diverse, and less vulnerable to isolated short term demand spikes.

Which also implies that BNM’s selected measures is a better policy move than a general rise in interest rates, which would be too blunt an instrument for the purpose.

Household Debt

On this issue, I don’t know where Dr Dulkefly is getting his data. Based on BNM data, household borrowing from the banking system is just RM475 billion, not RM560 billion:

Is it increasing? Yes. Is it a problem? Potentially, but not yet. Again I don’t know where Dr Dzul got his figures, but they don’t fit the official data (annual data; 1996-2009):

You’ll only approximate Dr Dzul’s figure of 76% if you make the tyro’s mistake of using real GDP instead of nominal GDP. Now there’s no doubt the ratio has increased, but this is a little misleading, as nominal (and real) GDP crashed in 2009. Data for this year indicates that the ratio has flattened out (quarterly data; 2001:01-2010:03):

Again, this looks like a non-issue to me. The numbers of bankruptcies and people seeking debt counselling have increased in absolute terms, but as a proportion to the banking systems total loan portfolio, there hasn’t been a big increase.

Going back to the demographic story, given the low median age of Malaysia’s population, a rise in debt should in fact be expected. You’d really only see the debt ratio falling once the median age passes 40, which isn’t expected to happen until after 2050.

So to sum up:

1. There’s been no overall increase in residential market prices, except in certain areas;
2. The property overhang is larger this year compared to last year, but the trend is still firmly down;
3. Sales and project launches are coming down as well, not going up;
4. The market is saturated, and this won’t change for a few years;
5. Household debt has risen, but not out of hand.

Technical Notes:

1. Property stock and price data from NAPIC
2. Loan data from BNM’s Monthly Statistical Bulletin