Since 2003 I've argued that many academics, economists and consumers confuse two fundamentally different concepts: the risk of a broad-based house price index, and the probability of loss faced by individual home owners. Now this is a profoundly relevant issue for us all. The popular notion of bricks and mortar being a safe investment for owner-occupiers is at least partly based on the observed volatility, or risk, associated with regional house price indices.
The problem here is that a house price index proxies for literally many millions of homes worth billions or trillions of dollars. An index is, therefore, an incredibly well diversified portfolio. In contrast, an individual home owner is making an economic commitment to one highly 'idiosyncratic' asset. This is, by definition, a poorly diversified investment with a much larger prospect of loss. It begs the question as to why Australian policymakers have not sought to foster markets to allow home owners to hedge, or insure away, such hazards. For what it is worth, the company I work with, Rismark, is helping the ASX and RP Data pioneer exactly these opportunities.
A similar analogy can be found when comparing the risk of a micro-cap listed on the ASX with the fluctations observed in the All Ordinaries Index, which captures every company listed on the exchange. As you should know, the micro-cap’s volatility will be much greater than the variations observed in the index. This is why your average super fund's portfolio invests in the index, and not micro-caps.
Since the circa 7-8 million property owners out there do not get the benefit of a national portfolio comprising millions of assets, understanding and managing individual asset-level risk is of profound importance to everyone.
Regrettably there is very little research on the economic hazards faced by single family home owners. Rismark's team have, however, spent a lot of time thinking about this subject. Applying one methodology developed by Professor William Goetzman, who is a leading financial economist at Yale, we found that the total risk of an individual home is about 15-20 per cent per annum depending on the holding horizon. If we take a 7-8 year ownership period, the individual property volatility is about 15 per cent per annum. The first chart below illustrates the results of this research based on all purchases and sales of Australian homes between 1990 and 2009.
Based on this analysis, owning a home outright is slightly less risky than owning every company listed on the ASX, which has exhibited historical volatility of nearly 20 per cent per annum over the last 30 years.
The next diagram shows a simulation of the risk profile of an individual home compared with capital city and national level house price indices between 2004 and 2009. For the purposes of this analysis, the team used RP Data-Rismark’s Hedonic Home Value Index. Importantly, we have also assumed that the individual property owner, which is represented by the red dotted line, has no mortgage debt. You can see that the value of the individual asset varies strikingly through time compared with the capital city and national indices.
To highlight these differences further, I presented approximations of the expected return ‘distributions’ attributable to both a national portfolio and a single family home. Observe in the following chart how the individual property’s return distribution, which is proxied by the blue bars, has far 'fatter' tails and a lower peak. This tells us that the prospect of realising positive or negative capital growth outcomes is much higher than a nationally diversified portfolio, which is represented by the comparatively narrow red distribution. The latter demonstrates that the dispersion in the range of possible outcomes is very tightly wound around the average, or expected, capital growth rate. This is just one way of saying that a national housing portfolio has exceptionally low risk of just 3-4 per cent per annum. If you had this holding, the value of your investment would have fallen only 2.5 per cent in 2008. If, on the other hand, you had a single, million-dollar-plus property located in Sydney or Melbourne, it probably fell in value by 10 to 15 per cent.
The final chart on this topic shows the monthly returns attributable to a simulated Australian home between 2004 and 2009. Critically, we illustrated both 'geared' and 'ungeared' capital growth profiles. The blue line denotes the value changes associated with a home owner with 70 per cent mortgage debt. The red line represents a home owner with no debt (ie, they own the property outright). The impact of leverage is vivid: it amplifies the probability of loss, with the blue line dipping far below the zero per cent (ie, no change) x-axis with disturbing regularity.
So what is the message from all of this? Over and above simple portfolio diversification, one take-away is that households have a profound (latent) desire to deleverage their balance-sheets. This is an obvious statement today—but it was not something that many folks focussed on back in 2003.
The only way to reduce debt is by boosting equity. This brings us to an economic puzzle: equity markets for housing finance don’t exist. While listed and unlisted companies raise debt and equity every day of the week, home owners are stuck with debt. So whereas a company with 90 per cent gearing would be punished by the market for being excessively risky, we happily allow first time buyers to leverage up to the tune of 95 per cent against far riskier underlying assets.
The principal value of equity, as opposed to debt, finance is that the equity provider wears the risk of loss alongside you. When times get tough, equity costs you nothing. In comparison, you always have to service your debt. Equity only has a positive cost when things go well, and you can afford to pay. This is precisely why financial markets and regulators like to see companies with more equity and less debt.
The Bank of England's Andy Haldane recently arrived at this conclusion: policymakers need to revisit the housing finance paradigm and promote the design of smarter instruments that help households better mitigate their risks. Of course, real change is an awfully difficult thing to implement.
In the presentation prior to mine, the RBA’s Dr Luci Ellis touched on similar themes.* While I cannot do justice to her full text, she made several critical points that I have belaboured before.
The first was that housing markets and the household sector have not traditionally been a source of ‘system stability risk’ notwithstanding media commentators’ apparent obsession with convincing us of this. The far more hazardous areas of the economy have been government, corporates and commercial real estate. Or, in Luci’s words:
“Compared with other kinds of credit, lending to households to finance the purchase of housing is relatively low risk. International banking regulations recognise this by applying lower risk weights to home mortgages than to small business loans… Historically, households have not normally been the instigators of financial instability. Aside from the current episode in the United States, financial crises are normally sparked by other sources. These include commercial property, property development, leveraged buyouts, sovereign debt and so on.”
Luci shows that this is borne out by historical mortgage default rates, which, with the exception of the recent crisis in the US, tend to lag rises in unemployment, rather than leading them (see the chart below).
Another important point that Luci makes is that debt levels in the Australian housing market are much lower than equivalent US exposures. This is very likely a function of the tax treatment of Australian housing, as Luci first suggested several years ago.
While commentators like to complain that housing gets concessionary tax treatment, owner-occupiers pay a very big price for the CGT exemption on their homes: unlike every other asset-class, mortgage debt is non-deductible. In the US, however, CGT is levied on owner-occupied housing in exchange for allowing borrowers to deduct their mortgage interest against tax liabilities. Many Australian journalists have argued in favour of the US approach in the context of the Henry Review. But, as Luci thoughtfully notes, the US model has resulted in much larger starting LVRs, longer paydown periods and higher default rates. Her chart below highlights the contrast between the two countries.
Luci also reiterates recent RBA analysis of lending standards and the share of 'at-risk' borrowers. She finds that the share of mortgagors with a conjunction of high LVRs and large repayment burdens remains “very low” (just 2 per cent). And she contends that Australian banks have generally been conservative in their lending practices, which is why our default rates are a fraction of US and UK levels despite much steeper mortgage rates.
In concluding, Luci argues that recent Australian house price growth has been driven by 'fundamental' demand-side factors in concert with supply-side pressures. This is a well-known story. Perhaps more interesting is her ‘final thought’ that “recent data suggest that we do not have a credit-fuelled speculative boom on our hands.” Yes, that’s right: contrary to what media commentators have told you, the RBA does not believe that Australia suffers from a house price 'bubble'. But they are nevertheless focussing on the future, which is why Luci bothers to make the rather tautological statement that “it would not be desirable for the current situation to turn into one.” That’s central banker speak for, Behave!
And just in case there is any residual doubt in our behaviourally-biased brains, Luci explains that the key mitigants are conservatism on the part of lenders in combination with grounded expectations from investors. The RBA would like us to believe that the alternative is an interest rate smack-down.
*The first chart in this speech illustrates real dwelling prices back to the early 1990s. For the nerds out there, Luci employs 'stratified median' and 'simple median' house price indices because RP Data-Rismark's more sophisticated 'hedonic' index only starts in the mid 2000s.
