IPF-sponsored research by Neil Crosby, Steven Devaney and Anupam Nanda examines the drivers of depreciation for office and industrial buildings
The importance of depreciation is well known to investors, advisers and appraisers working in direct real estate markets. It is a key input to property pricing models and has an important effect on real estate performance and management. Some of the previous research in this area has identified causes of depreciation and other research work has concentrated on measuring its impact on rental and capital values, but few studies have attempted to put both strands of this research together and identify the impact of the different causes.
Which of the possible causes has the most impact on the depreciation rate? The Investment Property Forum (IPF) in the UK recently sponsored research to identify just that, focusing on rental depreciation, which has a crucial influence on future asset-level cash flows.
Earlier IPF research established sector average rates for rental depreciation and capital expenditure using IPD data on individual institutionally-owned real estate investments.
These rates are shown for the office and industrial sectors in the table below. Over the past two decades, it is notable that offices have experienced more rental depreciation than industrial properties and had relatively more money spent on them by their owners.
What is also notable, though, is that the standard deviations in the table indicate high levels of dispersion around the average depreciation rates. This implies that sector-level figures may be a poor indicator of the experience of individual buildings and that other factors should be taken into consideration.
What factors could be causing these large variations in rental depreciation across assets? One obvious possible factor is that buildings may be of different ages. Older buildings might be more susceptible to deterioration than newer ones, or more vulnerable to obsolescence ‘shocks’ in terms of changes in technology, user requirements or building regulations.
Obsolescence means that it is difficult to observe a neat pattern in depreciation as buildings age; while deterioration may be relatively predictable, the sources and impact of obsolescence are much harder to estimate. Nonetheless, it is clear that age, in principle, should have an important influence on rental depreciation rates.
However, age alone is insufficient as an explanatory factor for a number of reasons.
Firstly, while buildings of a certain generation may have similarities in design and construction, there will be variations in quality and in the amount of management input (repair and refurbishment) that those buildings have received.
Secondly, in certain locations the rent occupiers pay might be underpinned to a greater extent by the value of the location, which may either appreciate or depreciate over time.
Thirdly, wider market conditions are also very likely to be important as they influence the demand for and supply of buildings. What is the interaction between a strong economy, rents and the development pipeline within a locality, and the impact of these factors on the relative performance of newer and older stock?
This latest research for the IPF sought to identify possible causes from the literature and model rental depreciation rates against these property-specific and market factors. Our study used a dataset of rental depreciation rates for 217 office and 158 industrial buildings located across the UK. We analysed these buildings over 1994-2009, a period in which there are important economic and rental-market cycles. Appropriate panel regression techniques were applied with two different types of model being estimated:
• A model to explore variations between buildings in average depreciation rates;
• A model to explore fluctuations in depreciation rates from year to year.
Explanatory variables included age, expenditure rates, a proxy variable for quality, estimates of land values for different locations, and variables representing wider economic and property market conditions.
The results were as follows. In terms of variations between properties, the modelling suggested that prime properties depreciate faster than lower-quality assets and, in the office sector, newer properties experienced higher depreciation rates than older stock, other things being equal.
Although these results may seem odd, they suggest that once buildings reach a certain age or quality level, additional years make little difference to their relative value. There was no evidence in our dataset for an S-curve profile to rental depreciation (where depreciation rates are low for the first few years, then accelerate during the building’s ‘mid-life’ before reducing again in old age) or any alternative profile.
Meanwhile, properties in locations where rental growth was stronger had higher rental depreciation rates, but the higher land component to values in London and the South East of England appears to offer some protection to investors, with this appearing to reduce depreciation rates. In the office sector, capital expenditure also mitigated the severity of depreciation, but the same result could not be confirmed for our sample of industrial assets.
Finally, in terms of variation through time, the research found that economic growth in different regions was negatively related to rental depreciation rates. Thus, as the economy improves and demand for space increases, depreciation is reduced. Yet, after controlling for economic growth, strong rental growth in different areas was associated with higher rental depreciation rates. This may reflect the introduction of new developments in such areas, which push up prime rents, but not the rental values of existing assets. However, a lack of sufficient local-level data on new office and industrial development prevented the study from testing this idea further at the present time.
In summary, the research sought to explain the rental depreciation rates of individual, institutionally-owned real estate investments. By doing so, it sought to shed light on a number of outstanding issues such as the shape of depreciation, the impact of market state, the influence of quality and location, and the effect of land values on depreciation.
Nevertheless, the earlier discussion of obsolescence highlights that rental depreciation can never be completely predicted. For instance, the drive towards greater energy efficiency and environmental performance from buildings could constitute a future source of depreciation and a motivation for expenditure to retrofit buildings to comply with new regulatory standards. The extent of its impact on values, though, has yet to unfold.
Neil Crosby is a professor of real estate, Steven Devaney is director of studies for real estate, and Anupam Nanda is associate professor at Henley Business School, University of Reading