Search results

Aims to re‐examine the role that can be played by real estate in Swiss mixed‐asset portfolios. For this purpose, constructs a constant quality price index for apartment buildings in Geneva. Also uses data pertaining to real estate mutual funds. In real terms, and after taking into account the illiquidity of real estate, the results suggest that the optimal portfolio composition is 20 per cent for stocks, 53 per cent for bonds and 27 per cent for real estate. Real estate mutual funds could also be included in a small proportion in the portfolio.

Swiss institutional investors hold approximately 19 per cent of their wealth in property, and the bulk of the allocation to property is housing. The financial reasons which are often given to motivate this investment strategy are twofold. First, property returns are hypothesized to be lowly correlated with the returns on stocks and bonds, and the inclusion of property in portfolios of financial assets should lead to diversification benefits. Second, property is viewed as acting as an effective hedge against inflation. Empirically investigates these two assumptions on the basis of hedonic price indices for Geneva and Zurich apartment buildings. Examines whether an investor who already holds Geneva (Zurich) housing should invest in the other canton. Investigates whether real estate mutual funds should be included in the portfolio in addition to direct real estate holdings. Results suggest that housing is an effective portfolio diversifier but does not provide any better short‐term inflation‐hedging effectiveness than financial assets.

Tests the inflation‐hedging effectiveness of Swiss real estate. Four proxies for expected inflation are used, two of them being based on autoregressive conditional heteroscedasticity models (ARCH‐M and QTARCH). The series used to proxy for real estate returns is a transactions‐based series adjusted for quality of the buildings by means of a hedonic model. Swiss stocks, bonds, real estate and real estate mutual funds are usually positively related to expected inflation and negatively related to unexpected inflation. Many of these coefficients, however, do not exhibit statistical significance.