Walter Torous is a leading American scholar and researcher in real estate finance and holds a number of professorial positions. He presently lectures at the Centre for Real Estate and the Sloan School of Management at MIT. His consulting clients have included government departments, banks and other financial institutions. He recently participated in a real estate forum at UNSW Business School and spoke with Julian Lorkin for BusinessThink.

An edited transcript of the interview follows.

BusinessThink: There's nothing more solid than bricks and mortar, or possibly an apartment block development – it's "way safer than shares", or so we used to believe. But commercial property returns are highly variable and can go down as well as up – and down by a long way. One man who's been studying this phenomenon is Walter Torous. He's been analysing commercial property returns. Walter, it's great to have you with us.

Walter Torous: Thank you, Julian.

BT: When we are thinking about investing in either real estate or shares, are they both inherently unpredictable?

Torous: Yes they are, because whether the financial [investment is] a share, or a commercial property, or even one's own home, it is an asset. And what financial economists have discovered during the past 20 or so years of research is that all assets' expected returns vary over time and it's that time variation which is a source of unpredictability, on the one hand, but there are definite movements that can be exploited.

Real estate is like any other asset. It obviously has different attributes which make it desirable for diversification purposes for a fund to hold, or a super fund to hold – by diversifying, say, share risk – but at the end of the day it is an asset and its expected returns, just like shares' expected returns, do vary over time and therefore do have this element of predictability associated with them.

BT: In your research, you also say that people should be taking advantage of "underlying macroeconomic conditions". Are you talking about, say, the cash rate, or what's happening in the wider economy?

Torous: We know that expected returns to all assets, be they shares or property valuations, vary with the business cycle. We know that when we are in the trough of the business cycle, for example, expected returns tend to be higher. Why? Well, again firms and managers need to entice individuals to buy their shares or to buy their properties and they do that by offering higher expected returns.

Whereas if we are at the peak of the business cycle expansion, the opposite holds. One doesn't need high expected returns to attract investors. So, knowing where we are in the business cycle  – and we do know that macroeconomists and governments bodies and central banks do collect very good information on the business cycle – that information should be exploited in one's investment decisions and portfolio decisions.

In the US, it's quite interesting that with residential real estate, home prices fell – beginning in the middle of 2006 – by 30% to 40% and it took maybe one year for that to move to the commercial property market, where in later 2007 commercial property prices fell by approximately 40%. So again, there is a link because clearly as individuals' home prices depreciated in value, they felt less wealthy. Their permanent income was lower so they didn't consume as much and that fed into retail, obviously employment opportunities dried up, and there was less demand for office space. So eventually commercial real estate fell in concert.

BT: And it has a long-term effect as well. If the value of somebody's house goes down then they understand that they don't have the money they never had anyway. But if commercial property goes down, that actually impacts on the super funds which invested a huge amount of cash into those properties to pay for people's retirement, to pay for people's pensions. Is there a long time-lag there? I'm thinking in terms of decades where people are not getting the returns on their super fund because commercial real estate has performed very badly.

Torous: The data shows that with the high-end properties – or the so called "classy properties" – in very desirable cities like Manhattan or San Francisco, the valuation today is higher than it was at the peak prior to the crash.

So, again the super funds that invested in these very desirable buildings are probably doing quite well. But on the other hand, had they invested in the less desirable properties they would not be doing so well, so again it all boils down to portfolio allocation. How did the super fund allocate the allotment of its funds in commercial real estate across these various segments? It's very important.

BT: Your research is mainly based on the US market. But are there lessons for Australia, which of course never suffered the crash that the rest of the world had?

Torous: Australia, and similarly Canada, among other countries, did not suffer this particular devastating decline in real estate values and subsequent recovery in some aspects. I think my research would still show that even in the absence of a crash, portfolio allocation is important simply because if one has done well in the past it does not guarantee that you'll do well going off into the future. And to make sure one does well going into the future, keeping track and being aware of the characteristics of the portfolio – be it shares or commercial real estate – does very much matter.