House Price Prediction – 2016

Dear Sirs & Ma’ams,
I present for your critique my latest thesis on house prices and debt. Some of you know well that I’m a devout sceptic of the ‘House prices double every seven to ten years’ theory, but to date I’ve been unable to sufficiently prove that house prices cannot and will not double over the next decade. This document sets out to change that problem. I believe I have finally found the ‘smoking gun’ if you will that precludes the theory from being anything but a logical and economic fallacy.

Title: Australia’s Debt Mountain – Why house prices cannot double by 2016
Author: foundation
Date: July 2006

The last decade has seen a remarkable and unprecedented boom in house prices across Australia. This incredible growth in house prices has come at a cost. The name of this cost is debt. Housing related debt has grown from 44% annual household income in 1995 to 130% today. While much of this increase has been enabled by lower interest rates which themselves were a product of a historically very low inflation rate, there is more to the story. Housing related debt servicing (the proportion of income required to make interest payments on debt) has grown at a rate far in excess of income growth, to the point that it is at record levels. Interest on housing loans now accounts for 7.5% of household wages, compared to 5.8% in 2002 and 4.5% in 1993. In itself, this is no cause for alarm. Certainly there must be a critical point at which debt-servicing ratios become, well, unserviceable, but the current low levels of loans in default (0.6% compared to 6% in 1991) suggest that most households are well able to service their debts at current interest rates.

In addition, foreign debt levels have risen massively over the last few years to $500 billion in 2006 from around $190 billion in 1996. This fact is not critical to our analysis, but may be discussed later. The cause of this rise is primarily offshore borrowing by commercial banks for the purpose of financing the house price boom.

Some facts and figures
2006 housing debt: $700 billion
2006 population: 20 million
2006 workforce: 10 million
2006 dwellings: 12 million
2006 average new mortgage: $215,000
05/06 dwellings turned over: 600,000
05/06 housing debt growth: $129 billion.
05/06 house price growth: <1%

A startling consideration emerges from these simple figures; that total debt levels have grown substantially despite a very low rate of house price appreciation. The explanation for this is simple. Not all houses are bought and sold every year. The historical average turnover is around 5%annually, a level to which we returned last year following a couple of years peak above 7.5%. This implies that a change to house prices has a long-lasting effect. If house prices rise in one year, then stabilise, total debt levels will increase for twenty years.

Scenario A: House prices stagnate for ten years
What happens if house prices do not appreciate nominally over the next decade? Assuming that turnover continues at the historically unremarkable rate of 5% per year, total new debt will need to be added at a rate of $129 billion per year. This totals to nearly $1.3 trillion dollars, which if added to current debt, amounts to a staggering $2 trillion. Certainly, these are big numbers, but what do they mean? Are they realistic?

If Australia’s total housing debt reaches $2 trillion, this equates to $200 thousand dollars for every working person in the country – an increase of over 200% from the current level of $70 thousand. Assuming the currently prosperous economic times continue, we can perhaps expect wages to rise by 4% per annum. A reasonable ‘best-case’ scenario would be steady interest rates. So how can the debt-burden be represented? We’ll use a simple Debt Servicing Ratio (DSR), representing the proportion of post-tax income required to make minimum interest-only payments. Note that this is not the same as the official (ABS) previously quoted DS/Household income, it is calculated from the above “Facts and figures” (rightly or wrongly!).

2016 DSR would be 34%. Currently this figure is just short of 17%. So if house-prices do not rise, the cost to the economy will be significant, as interest payments on home loans would remove 34 cents from every dollar earned. This means $143 billion dollars (up from $50 billion today) every year would be spent on interest, where it otherwise might have paid for personal debts, been consumed, saved, or invested. Remember, this cannot be considered investment, as regardless the purpose of the loans, this money is only paying interest, it yields no net improvement in total balance.

These are big numbers, but admittedly, they alone might not break the bank. In fact, the average worker earning $57,000 gross might be perfectly content paying $14,000 in interest. But it must be remembered that it takes not ten, but 20 years for the entire housing stock to be turned over. Let’s look further forward.

2026 DSR would have peaked at 37% of net annual income. The burden to the economy would have risen to $236 billion annually. Still, this may not be disastrous… or will it? A recession is considered to be 2 or more consecutive quarters of negative growth in Gross Domestic Product (GDP), and it’s a safe bet that if the total dollar value of debt servicing outstrips GDP growth, we can be sure that the DSR has surpassed its critical sustainable level. While GDP is subject to many variables, and much variation, the average GDP growth of the last decade (at 3.6% per annum) can be considered surprisingly strong, and therefore forms the base of our ‘best-case’ scenario.

Debt servicing at $50 billion currently represents around 5% of GDP. By 2016, debt servicing at $143 billion would represent around 10% of GDP. Probably not enough to bankrupt the country, but the billions of lost dollars must have a depressing effect on GDP, so the proportion may be much higher.

So this scenario is, although highly unlikely, possible, under ideal economic conditions. It should be noted that this relies on 3 volatile variables – sustained 4% wage inflation, sustained 7.2% interest rates, and sustained 3.6% GDP growth, and would be vulnerable to external shocks such as terrorism, pandemics, currency fluctuations and any number of other adverse actions.

Scenario B: House prices double over ten years
Now lets look at the effects of house prices rising by 7% annually. Once again, we’ll assume turnover of 5%, interest rates at 7.2% and wage growth at 4%. Total new debt will begin again at $129 billion per year, but by 2016, it will be over $237 annually. This totals to nearly $1.8 trillion dollars, or when added to current debt, $2.5 trillion. Again, big numbers, but how big? Surely this is just 25% higher than the previous estimate for stagnant house prices?

2016 DSR would be 42%, and climbing. After 20 years, this DSR would take 68 cents from every post-tax dollar earned, and after 30 years… After 30 years, from every take home dollar earned by every worker, the banks would demand one dollar in interest payments. Every single cent would go to maintaining debt for our fabulous houses, which would be valued at a little over $1.7 million each, on average. Repayment of the loan would be impossible.

Conclusion and discussion
Ok, so perhaps this faux-semi-serious analysis has run off the rails and become a little silly. But it can be used to draw a few conclusions:

– House prices cannot double every seven to ten years in a low inflation, low interest rate environment.
– House prices, at least over the long-term can only rise in line with wage inflation.
– A slow unwinding of current house price excesses will have a massively detrimental effect on our economy.

The potential of the last point is predicated on the following requirements; the willingness of foreign investors to provide hundreds of billions of dollars in additional funds, the willingness of future buyers to purchase houses that are not going to appreciate in value for the foreseeable future, the willingness of negatively-geared investors to continue funding loans where expenses are higher than capital appreciation, and ongoing strong economic growth in excess of debt-servicing obligations.

Hopefully some readers will have had an “ah-ha!” moment somewhere along the way, and if not, here’s a prompt. Normally, only 5% of houses are bought and sold in any one year. If increased demand (enabled by low interest rates, government incentives, taxation changes or what-have-you) leads to house price appreciation in excess of wage inflation, we should collectively step back and think about what’s going on. If 7.5% of houses are traded at a 50% premium to the price that most of the remaining 92.5% of the stock were traded for, we should not assume every other house is ‘worth’ this premium. The true total value of every house can only be determined by a 20-year average. If you like, the true value of my house today is somewhere near the middle of its value in 1996 and its value in 2016. So if house prices can only rise at the rate of wage inflation over the long term, and my house has risen in price since 1996 by the magnitude of 20 years wage growth, the prospect of significant further price appreciation over the coming decade is fairly unlikely.

Unfortunately, very few people are capable of even such simple economic thought, and the banks prey on this fact. They are happy to let every single house-owner think that their house is currently worth whatever the market rate is, and take on debt to near that value. This has at least 4 notable effects:
1) a depressing of future economic activity
2) lower potential for sustainable long term house price growth
3) a high potential for further short-term house price inflation at levels unsustainable over the long-term.
4) increased buyer expectations of long-term high returns.

Unfortunately, this can clearly become circular, and guess what? The banks don’t care. In fact, they encourage it, to the detriment of us all. It’s worth noting that the perception of the buyer (effect 4) is in complete contrast to the true situation as prescribed by 1) and 2).

So is there an alternative? Of course, there are a couple. The first is for lending controls, where valuations are either required to be averaged over a few years (thus defusing the circular reference where higher house prices allow higher lending enables higher house prices allow…), or even a cap on the number of loans and refinances allowed in any one year. About 10% would be enough. Both options are unpalatable, both politically and socially. Politicians don’t want to be seen to be controlling financial markets, nor do they want to lose the taxes they rake in from loans and turnover. The public wouldn’t like to think that the government was preventing them from getting rich through house investment.

The other option is for house prices to make up a considerable (around 15-20%) component of the Consumer Price Index (CPI) indexed to regular (pre-boom + wage inflation) levels. The result would be interest rates that reacted to higher house prices by rising. This would increase the attractiveness of regular saving, while increasing the cost of speculative house buying. The whole show could still move up in line with wage increases, which is the only way house prices can sustainably rise over the long term. It would have a more important role in ensuring economic stability and prosperity. Funny enough, this is the whole point of the Reserve Bank and their inflation-targeting control of interest rates.

Pfft. Anyway, I’ve strayed far from the straight-line I intended. Thanks for reading. Please add any corrections below and I’ll incorporate them when I get the time.

Regards, F.[cowboy2]