Investors have the tools to assess the investment and financial risks relating to listed real estate, but operational risk remains opaque. Dirk Brounen and Wendy Verschoor explore the implications this has for accurate pricing

Why invest in real estate? This is the question often asked at the start of real estate investment courses taught around the globe. Depending on the individual financial text book, the answer could be one of four: competitive returns; real estate returns protect against the risks of inflation; real estate offers solid diversification opportunities in a stock and bond portfolio, or real estate risks are low.

There is some truth in each of these, but we might want to reconsider the low-risk profile of real estate.

Ever since the start of the credit crisis, investors have been shying away from real estate investments as they were suddenly perceived as risky bets in overpriced (and levered) markets. In recent literature, we learned that losses stick with us longer than the merry days of high returns. Investors are very sensitive to the occasions of so-called ‘black swans'; rare events that before their occurrence were perceived as improbable or even impossible. The weak performance of real estate investments during the credit crisis can be regarded as such a black swan.

Figure 1 explains why investors care so much about these exceptionally poor return days. The black line depicts the price return index of the European EPRA index of listed property. The blue line redraws the same history after excluding the twenty most disastrous return days. In other words, for a period of 20 years, we exclude only 20 days, and find a total return that is four times as much (blue line). This is why investors care about risk - the most dramatic events can easily shape the end result.

But what is causing real estate risk? And how well is it understood by the markets?
Like in most investment markets, the risk of real estate is the sum total result of a wide variation of uncertainties. There are many different sources of risk and multiple decisions that need to be made that can go wrong. Overall, we could categorise the vast majority of real estate risks into three main areas:

• Market risk: This is the changing state of the overall markets. A key determinant of the performance of real estate markets is the condition of the overall economy. The success and performance of real estate investments tend to hover along with the overall business cycle. Market risks, however, differ across real estate markets, and especially across the different property sectors;

• Financial risk: As in any firm the capital structure can dampen or enhance the turbulence of corporate results. Firms with a high level of leverage that face floating rates and short-term liabilities suffer more when the going gets tough. A prudent CFO can limit these odds and several accounting ratios can assist in assessing this type of risk;

• Operational risk: This is the most ignored and concealed type of risk. Operational risks are the losses that results from internal processes, in other words, the decisions made while working hard to achieve success. In every firm operational risk occurs as a result of the internal decision-making process, but these risks are anecdotal and typically not shared with others.

Both the market and financial risks can be assessed in a straightforward manner. Multiple-risk measures are available today, ranging from equity beta to earnings semi-variance, and from debt-service ratios to the Altman z-score. In other words, we know how to measure the market and financial risks in real estate investments. But the operational risks remain opaque. For major price-sensitive events, most entrepreneurs are not keen on sharing their mistakes and failures, which means that outside investors can often only guess about this third risk category.

To illustrate the complexities of disentangling real estate risks, we study a unique sample of 116 bond issues of listed real estate investment companies. These bonds were issued in the period 1986-2011 and were carefully documented by the European Public Real Estate Association (EPRA). The price of risk, measured here as the spread of the real estate bond over a 10-year government bond, will differ for various reasons. Besides the bond specifics like the maturity and size of the loan, we should find traces of the three clusters of real estate risks. On average, this spread equals 122bps and runs up to as high as 491bps for a €65m bond issue on May 2010. We performed a multi-variate regression analysis on the bond spreads in this EPRA sample, which allowed us to isolate the different risk factors and measure their relative importance.

In figure 2, we graphically highlight the main results of this analysis, showing the movement in basis points in the observed bond spread if the risk factor involved should increase by one standard deviation. In other words, if bond maturity would increase with one standard deviation (11 years) than the bond rate would increase with 20.9bps.

Overall, we can conclude that there are some clear patterns in the variation of bond spreads. Larger and longer bond loans are issued at higher spreads, which is what we would expect. Extending this analysis with a set of risk factors that capture the market and financial risks of the underlying firms yields some useful insights. High-market risk (high betas) comes at a higher price (spread), and we also find that bond issues of residential funds are priced at a lower spread compared to office and retail bond issues. The systematic market risk-profile of the underlying investment portfolios is important when setting the adequate spread. In the same cross-section, we also find that highly leveraged real estate firms need to offer higher spreads for their next bond issues.

But perhaps the most important result from this analysis is the one that is missing. All in all, we find that only 13% of the cross-section of bond spreads is explained by the combined analysis of bond characteristics, and the market and financial risk of the firm. This leaves 87% as unresolved, a gap that could be narrowed if we were to know more about the third risk component of the firms involved: the operational risk.

The risk of real estate investment has gained importance in recent years, but a lot is still not known about this risk. The operational risks of real estate investment firms remain especially opaque. A study of real estate bond issues shows that spreads are set according to fundamental risk factors. This means that communicating all the relevant risks involved is key when seeking an adequate price in the market.

Dirk Brounen is professor of real estate economics at Tilburg University, and Wendy Verschoor is director at PwC Netherlands