RDP 9106: The Direction of Australian Investment from 1985/86 to 1988/89 4. Investment by Real Estate Operators and Developers

Based on the CAPEX survey, real new private capital expenditure in 1988/89 was $5.7 billion higher than in 1985/86 (1984/85 prices). Real estate operators and developers accounted for 27.5 per cent of this extra $5.7 billion, with their investment rising from $1.1 billion in 1985/86 (just over 7 per cent of total private new capital expenditure) to $2.7 billion in 1988/89 (about 12.5 per cent of total private new capital expenditure).

Most of this industry's growth is, not surprisingly, concentrated in building and structures. This industry is involved in the organisation of land and property developments and not the actual purchase or clearing of land, nor the provision of services to subdivisions or the construction of dwellings. It consists of establishments mainly engaged in renting or leasing their own non-residential properties to others. It also includes establishments mainly engaged in land subdivision or development, but not its construction.

The fact that real estate operators and developers accounted for so much of the growth in real capital expenditure over this period may simply reflect a trend in the commercial property market, away from the end users owning premises they occupy to the end users increasingly renting their premises. For example, a number of the large chain stores and private banks now lease their suburban premises rather than investing capital in their premises. Recall that while the ABS now allocates investment to the user rather than the owner, this does not apply to the rental of commercial property. This is because commercial property is generally rented out under standard or operating leases.

We found evidence consistent with the hypothesis that the spectacular growth in investment by real estate operators and developers was simply replacing investment formerly undertaken by other industries. Graph 4 appears to show a rise above trend following 1987 in the level of building and structures investment by real estate operators and developers. The trend for total building and structures investment appears unchanged while the building and structures investment by other industries falls below trend.[9]

Graph 4: Real Investment in Buildings and Structures
1984/85 prices
Graph 4: Real Investment in Buildings and Structures

Having established that the level of total building and structures investment did not rise above its trend but that this investment was increasingly carried out by real estate operators and developers, it is of interest whether the end use of investment has altered. The ABS carries out a survey of the value of building investment which provides a breakdown of building activity into various structural types (shops, offices, hotels, etc).

Graph 5 shows the breakdown in the real value of completed non-residential private sector construction. The share of building investment going to offices rose from 32 per cent in September 1987 (when building and structures investment by real estate operators and developers begins to grow more rapidly) to an average of 35 per cent in 1988/89. In the September quarter 1989 office construction accounted for 38 per cent of the volume of non-dwelling construction. Office construction made the largest contribution to growth in the real value of non-residential construction over the period 1985/86 to 1988/89 (45 per cent). The 1990 Budget Papers indicate that the real value of office construction in the most recent investment boom as a proportion of GDP exceeded that of the booms in late 1960s and early 1970s. It is perhaps not a coincidence that asset prices for commercial properties appear to have peaked and vacancy rates are increasing (JLW 1990, Treasury 1990, The Australian 1.8.90, BIS Shrapnel 1990).

Graph 5: Non-Dwelling Construction
1984/85 Prices
Graph 5: Non-Dwelling Construction

Footnote

Various econometric tests were carried out by estimating models of the three series in Graph 4. The models were based on linear time trends, lagged dependents and seasonal dummies and were found to be significant, thus supporting these conclusions. Details of the estimated models are available on request. [9]