Global and local economic uncertainties and office vacancy in Australia: a sub-class analysis

ABSTRACT

Motivated by uncertainties caused by the COVID-19 pandemic, this study examines how office vacancy ratios (for premium, A, B, C and D-grade spaces) in Australia respond to shocks in global and local economic uncertainties. Using semiannual data from 1998 to 2020 and applying the vector autoregression (VAR) model, our results suggest that office vacancy ratios respond positively to economic uncertainty shocks in general, and especially to local economic uncertainty. Moreover, our sub-class analyses show that vacancy ratios in various office spaces respond differently to uncertainty shocks. The office vacancy ratio in premium-grade office responds to both shocks after one year with its responses fading out in year three. B-grade and C-grade vacancy ratios respond to both shocks within the first year and responses last for about 4 years. Additionally, sub-lease vacancy ratios in the aggregate office space show quick responses to uncertainty shocks (dying out in 2 years), while direct vacancy ratios respond after about 1 year (but fading out in 3 years).

KEYWORDS:

• Uncertainty
• office market
• vacancy
• Australia

JEL CLASSIFICATION:

• C32
• E60
• R33

Acknowledgement

We appreciate the helpful comments of the editor (David Peel) and two anonymous reviewers. We also thank Jhoana Ocampo for her research assistance.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 According to EPRA (2020), the size of total listed real estate companies (in $ billion) as a percentage of total stock market was 6.82% in December 2020, which is much higher than the average of the developed markets (3.27%) and global markets (3.51%).

2 We perform the Augmented Dickey-Fuller (ADF) unit root test (Dickey and Fuller 1979) to examine the stationarity of all variables. Table A2 in the Appendix presents the ADF test results for aggregate office space. Results indicate that total vacancy ratio, direct vacancy ratio and local EPU are stationary in level. However, sub-lease vacancy ratio, global EPU, total office stock and service employment ratio are stationary in the/its first difference.

3 The results are available upon request.

4 Table A3 in the Appendix shows the number of optimal lags used in each VAR estimation.