Surprisingly little research has been performed on the impact of interest rate changes on the value of properties, especially when considering the facts that (1) real estate plays an important role for many financial investors, (2) interest rate risk is difficult to diversify and (3) traditional estimates of the interest rate sensitivity, i.e. Macaulay and Modified Duration, are not applicable to real estate. The present study gives a better understanding of the real estate interest rate sensitivity. This is achieved by modelling the whole life of a typical but simplified (office) investment property, based on a representative and exclusive data set for the Swiss investment real estate market. The interdependencies between interest rates, inflation, office market rents, current rent paid and expenses are defined and modelled empirically. As opposed to the simplistic analytical formulae of past studies, making use of the Monte Carlo Simulation allows to incorporate explicitly the uncertainty of the underlying stochastic processes, of their interdependencies and of the modelling uncertainties in general. The proposed approach allows not only to find a reliable and empirically defined estimate of the real estate interest rate sensitivity, but also to indicate the detailed uncertainty of the finale estimate as well as the precise impact of changes in major model parameters. Results show that the interest rate sensitivity of a typical office property is 13.1 %, whereas this estimate is associated with a standard deviation of 5.96 %. The risk premium and the degree of rotation of the interest curve are found to be the two prime determinants of the interest rate sensitivity, while the vacancy rate, the term of lease and the years between rent reviews influence the sensitivity, too, but to a much lesser extent.