PREDICTIBLE OR NOT? FORECASTING GERMAN AND US OFFICE RENTAL MARKETS

Abstract

This paper presents the results of two variations of an econometric forecasting model for both German and US office rental markets. The main objective of this approach is to elucidate the capability of time-series regression models of basic market indicators to capture and forecast movements in occupancy patterns and rental rates in both countries with their inherently different institutional and economic structures. The first model is a three-stage simultaneous equation model. The first stage incorporates the office space market in terms of occupied space and absorption of new space. The second stage captures the adjustment of office rents to changing market conditions and the third stage specifies the supply response to market signals in terms of construction of new office space. The standard simultaneous model is subsequently modified to account for the specific characteristics using the New York market as a case study The results demonstrate that the market reacts efficiently and predictably to changes in market conditions. The significance of the estimated parameters underscores the general validity and robustness of the simultaneous equation approach in modeling real estate markets. The modifications of the standard model, notably the inclusion of sublet space in the rent equation, contributed considerably to improving the explanatory power of the model. Eventually, the market implications of three exogenously defined employment forecasts are tested with the model. The second model is based on the assumption that the vacancy rate is the sole determining variable of future supply growth. Thus, expected future demand is a function of changes in weighted past vacancy rates using an arctan function. This non-linear function takes into account threshold values at which expected supply decreases drastically. In contrast to univariate time-series approaches, this model is capable of explaining even some of the more idiosyncratic office market developments in Germany with a sufficient statistical goodness of fit such as the crisis of the IT sector in Munich or the development of Eastern German office markets. A further advantage of this model is that it may be incorporated in other, more comprehensive modeling systems. An empirical application to US markets demonstrates the adaptability of the model to various economic and institutional contexts.