The predictive powers of the Hang Seng Index

Article

The predictive powers of the Hang Seng Index

February 27, 2012 / By Denis Ma

When speaking with industry pundits about the outlook for the Hong Kong office market, the correlation between the performance of the Hang Seng Index (HSI) and office rentals inevitably comes up – that the HSI is a good leading indicator of office rents.

Indeed, if you plot the HSI against Grade A office rents over a 20-year period, the cyclical turning points of the HSI generally lead rents by six to nine months, something that becomes even more apparent if you compare the year-on-year changes of the two data series. Running a simple causality test confirms the predictive abilities of the HSI.

Play around with the data a little further and an even more interesting relationship emerges: Grade A office rents in Central typically rise when the HSI is higher than one year earlier, and the reverse. This correlation is further improved if we smooth out some of the volatility in the HSI using a three-month moving average (3MMA).

Although the relative lag between the HSI and its equivalent year-on-year growth negates much of the usefulness of this relationship for near/short-term forecasting, it does provide a quick and simple tool for projecting turning points in the office rental cycle.

A similar analysis of Hong Kong’s business cycle (in terms of GDP gap) and office rental cycle would have yielded a very similar result. However, whereas GDP growth is abstract and difficult for the layman to predict (hey Joe, what you think Hong Kong’s GDP is going to come in at this year?), most people have an opinion (rightly or wrongly) on HSI target levels.

So, using this relationship, what can we expect for 2012? Well, based on the movements of the HSI in 2011, office rents are unlikely to return to growth if the HSI (3MMA) does not end the year above 18,700 points; its lowest quarterly end level in 2011. Predicting where the HSI will end up, however, is entirely another matter.