Abstract
Until the recent introduction of real estate futures on the Chicago Mercantile Exchange (CME), there have been few opportunities to manage house price risk. This paper examines whether house price risk can be effectively hedged in Las Vegas, one of the CME contract cities. The analysis considers hedging from the viewpoint of real estate investment groups, mortgage portfolio investors, builder/developers and individual homeowners. For investment groups and mortgage holders holding a mix of new and existing home assets, CME futures would have reduced house price risk by more than 88% over the 1994–2006 period. Similarly, homeowners implicitly hedging price volatility of existing homes also would have fared well over the sample period. However, builder/developers worried about new home price appreciation would have been much less successful in managing their risk. One important caveat, minimum variance hedge ratios change over time and may cause hedge performance to suffer.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
The CME Real Estate Futures contracts are not the first example of house price securities. Stocks and Bonds traded on the New York Real Estate Securities Exchange (NYRESE) beginning in 1929. However, in 1941, with the plunge in real estate and securities prices, the SEC decertified the NYRESE as a national market. Later, residential and commercial real estate futures contracts briefly traded in London. Case et al. (1993) blame their failure on the public’s lack of understanding of the derivative real estate market.
The S&P/Case–Shiller® Index also adjusts for outliers (Standard and Poor’s 2007, page 6).
We adjusted for outliers in both directions. Thus, our analysis which examines the representative house as measured by the median, should not be greatly affected by the filter rules. To ensure that our results are insensitive to the filter process, we compared the median house from the full sample of 308,183 to the median house in our sample. The correlation coefficient for quarterly returns between these two samples equals 0.9825.
We consider a home as “new” if the tax record shows the year built is equal to or after the sales year. The latter may occur if there is a sellout period that precedes completion of the home or completion occurs at the end of a calendar year.
Hedging effectiveness can be shown to equal the returns correlation coefficient squared. Thus, a correlation of 0.8322 implies that using the S&P/Case–ShillerÒ Index to hedge, on average, reduces 69% of the price risk for the representative house in Las Vegas.
Ultimately the minimum variance hedge ratio depends upon basis risk, the change in the spot price relative to the futures. If futures prices are unbiased, Castelino (1992) shows that the hedge ratio, h, may be written as:\(h = {\text{1}} + \rho _{BF} \sigma _B {{\left( {t,T} \right)} \mathord{\left/ {\vphantom {{\left( {t,T} \right)} {\sigma _F \left( {t,T} \right)}}} \right. \kern-\nulldelimiterspace} {\sigma _F \left( {t,T} \right)}}\)where ρ BF is the correlation between the basis and the futures price, and σ B and σ F equal, respectively, the standard deviation of the basis and futures price at time t, the end of the hedge period. As t approaches T, the expiration of the futures contract, the futures price converges to the underlying index, I. Thus, the hedge ratio may be rewritten as:\(h = 1 + \rho _{BI} \sigma _B {{\left( {t,T} \right)} \mathord{\left/ {\vphantom {{\left( {t,T} \right)} {I\left( {t,T} \right)}}} \right. \kern-\nulldelimiterspace} {I\left( {t,T} \right)}}{\text{ }}\)and is equivalent to b estimated in Eq. 1. Following Castelino, it can also be shown that whether the nearby futures price or the underlying index is used, the measure of residual risk of the hedged portfolio is approximately the same, and therefore, both lead to a similar estimate of hedging effectiveness.
While the subsequent analysis presumes risk minimization is the hedger’s objective function, we might consider other measures of hedging effectiveness. Moosa (2003a) points out that investors may be concerned with both risk and return. Hedging effectiveness might then be examined by looking at certainty equivalent returns (Cecchetti et al. 1988). Alternatively, Lien and Tse (2001) considers the possibility of loss aversion on the part of investors. Along those lines, Swidler et al. (1999) show how semi-variance can be used to measure hedging effectiveness when only downside risk is important. In all cases, however, hedging effectiveness ultimately depends upon the correlation between the asset and the derivative instrument so that the analysis in this paper would also apply to these other situations.
These values likely overstate the true hedging effectiveness of the real estate contract since they are based on in-sample estimates of β. The results implicitly assume that hedgers correctly forecast the risk minimizing hedge ratio. On the other hand, if β is non-stationary and there are unanticipated changes over time, hedging effectiveness will be less than the in-sample estimates. Subsequent work in Figs. 2, 3 and 4 illustrates this point and shows that hedging effectiveness is uniformly lower when you assume a (constant and) naïve hedge ratio equal to 1.
To verify that our minimum variance regression results are not sensitive to model specification, we also estimate Eq. 1 using first difference of prices. We find similar beta estimates to those listed in Table 4. In all cases, the estimated betas lie within the 95% confidence bands reported in the table. We also test for first order serial correlation. For the all equation in Table 4, the Durbin–Watson statistic equals 1.77 and implies no autocorrelation. On the other hand, if we estimate the all equation with first differences, D–W equals 1.18 and suggests positive autocorrelation of the error terms. Given these time series properties, we only report minimum variance regressions estimated from returns.
Correspondingly, the low R 2 implies a large confidence interval for beta. The appropriate hedge ratio may be as large as 2.39 or as small as −0.37. If the latter value, the developer/builder would be long housing and long the futures contract.
We measure hedge effectiveness as \({\text{1}} - \left( {{{\sigma ^{\text{2}} _{\text{h}} } \mathord{\left/ {\vphantom {{\sigma ^{\text{2}} _{\text{h}} } {{\text{ }}\sigma ^{\text{2}} _{\text{u}} }}} \right. \kern-\nulldelimiterspace} {{\text{ }}\sigma ^{\text{2}} _{\text{u}} }}{\text{ }}} \right)\) where \(\sigma _{\text{h}}^{\text{2}} \) is the variance of returns of the hedged portfolio and \(\sigma _{\text{u}}^{\text{2}} \) is the variance of returns of the unhedged portfolio. This is equivalent to R 2 in our hedge regressions.
References
Case K., Shiller R., Weiss A. (1993). Index-based futures and options markets in real estate. Journal of Portfolio Management, 19, 83-94.
Castelino M. (1992). Hedge effectiveness: Basis risk and minimum-variance hedging. Journal of Futures Markets, 12(2):187–201.
Cecchetti S., Cumby R., Figlewski S. (1988). Estimation of the optimal futures hedge. The Review of Economics and Statistics, 70, 623–630.
Ederington L. (1979). The hedging performance of the new futures market. Journal of Finance, 34, 157-170.
Global Insights and National City Corporation. (2007). House prices in America: 2nd quarter 2007 update. [Online]. Available at www.nationalcity.com/content/corporate/EconomicInsight/HousingValuation/documents/2Q2007report.pdf.
Herbst A. (1985). Hedging against price index inflation with futures contracts. Journal of Futures Markets, 5, 489-504.
Hinkelmann, C., & Swidler, S. (2006). Trading house price risk with existing futures contracts. Paper presented at the 2006 Cambridge-UNCC Symposium on Property Derivatives and Risk Management.
Kolb R., Overdahl J. (2003). Financial derivatives, (3rd ed.). Wiley, New York.
Labuszewski, J. (2006). Introduction to CME housing futures and options. Chicago Mercantile Exchange Strategy Paper, 1–28.
Lien D., Tse Y. K. (2001). Hedging downside risk: Futures vs. options. International Review of Economics and Finance, 10, 159–169.
Moosa I. (2003a). International financial operations: Arbitrage, hedging, speculation, finance and investment. Palgrave, London, 156–182, Chapter 6.
Moosa I. (2003b). The sensitivity of the optimal hedge ratio to model specification. Finance Letters, 1, 15-20.
Standard and Poor’s. (2007). S&P/Case–Shiller® Home Price Indices: Index Methodology. macromarkets.com/csi_housing/documents/tech_discussion.pdf.
Swidler S., Buttimer R., Shaw R. (1999). Government hedging: Motivation, implementation and evaluation. Public Budgeting and Finance, 19, 75–90.
Acknowledgements
The authors would like to thank the editors, Richard Buttimer and Kanak Patel for their encouragement. We also appreciate the support and constructive comments of João Duque, Inês Pinto and an anonymous reviewer.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Bertus, M., Hollans, H. & Swidler, S. Hedging House Price Risk with CME Futures Contracts: The Case of Las Vegas Residential Real Estate. J Real Estate Finance Econ 37, 265–279 (2008). https://doi.org/10.1007/s11146-008-9129-z
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11146-008-9129-z