Benchmarking für Eigentümerkosten von schweizerischen Renditeliegenschaften

Abstract

Purpose

Real Estate as a capital investment has become ever more important around the world during recent years. In Switzerland in particular, rapidly rising market values have led to it being increasingly demanding for property owners to obtain the return on investment that they require. This challenging environment demands a more distinguished and rigorous preparation of investment decisions. A concept seeking to achieve this is the use of benchmarking. Benchmarking has proven to be highly useful in other industries and offers considerable potentials for the real estate industry. The aim of this thesis was therefore to first develop a periodically repeatable benchmarking-process for the owner-costs of Swiss investment properties and then to perform the process afterwards.

Methodology

In a first step, a benchmarking-process is developed that meets all the requirements predefined by the company «Wüest & Partner» from Zurich. Subsequently, this process will then be implemented for the first time. First, correlation and regression analyses are conducted in order to analyse which variables have a significant influence on the owner-costs of properties and should therefore be taken into account when calculating corresponding benchmarks. On this basis, a user-friendly benchmarking-tool is then developed and programmed. This tool enables the automatic calculation of benchmarks, which are tailored to the individual circumstances of a property. Using this tool, it is possible to generate the desired benchmarks and to subsequently conduct in-depth analyses in a fourth step.

Results

This thesis demonstrates what the most important demands for benchmarking are and how they can be implemented into a systematic and periodically repeatable process. The execution of the process shows that the owner-costs of a property are significantly influenced by the variables «Property Type», «Location», «Size», «Age», «Average Floor Size», «Condition» and «Standard». Finally more than 3,400 individual benchmarks for ten different property types are generated, based on more than 10,600 actual cost statements. The subsequent indepth analyses of these benchmarks prove expected relations and uncover surprising and highly interesting new dependencies.