RDP 8704: The Role and Consequences of Investment in Recent Australian Economic Growth Appendix C: Adjusting Capital Expenditure Data for Leasing^[1]

Expenditure on capital is allocated to industry of ownership of capital rather than the industry of its use in both the National Accounts (sources (a), (b), (c) and (d)) and the Capital Expenditure Survey (source (e)). As a result, the leasing of buildings, structures, plant and equipment from the finance sector by the non-finance sector can distort investment by industry data. In aggregate private sector investment is not affected unless a private financial institution has purchased a capital good and leased it to the public sector.^[2]

Over the 1970s and 1980s the rapid growth of leasing has made data on capital expenditure difficult to interpret. For example, our estimates (which may understate the total) show that leasing grew from less than 0.5 per cent of GDP in 1967/68 to over 2.3 per cent in 1984/85. This appendix outlines the methodology and sources used to adjust the data presented in Figure 2.6 for leasing.

Whilst adequate data is available on new lease commitments undertaken by industry for the period January 1985 to June 1986 (source (i)), no data is available for leasing by industry prior to this time. However, aggregate data is available for leasing undertaken by finance companies from September 1966 to December 1985 (source (j)). Although the corporations classified as finance companies vary over time, to our knowledge, this data is to be the best historical series currently available.

C.1 The Method

Finance company leasing by industry has been allocated on an annual basis using the relationship between each industry's own GDP and total GDP excluding defence, dwellings, import duties and the imputed bank service charge (hereafter referred to as total adjusted GDP). All of which are available in current prices by industry on an annual basis.

In the first instance we assumed that the proportion of leasing undertaken by each industry was the same as that industry's contribution to total adjusted GDP. Whilst it is difficult to check the validity of this assumption, the table below shows that it appears to hold for finance, wholesale and retail and the other industry groups in 1985 (source (i)).

However, it appears not to hold for mining, manufacturing and other services. Since some industries are characterised by high/low levels of leasing in relation to their proportion of GDP, this is not surprising. To account for this, an arbitrary adjustment has been made to the assumption for these Industries. It was assumed that the relationship between their leasing and their own GDP remains the same in earlier years as was evident in 1985 (the category other services picks up any remaining leasing).

Figure 2.6 shows private fixed capital expenditure by industry using published data and adjusted for leasing using allocators calculated annually for each industry using the method outlined above. The allocations are then applied to the published constant price data. Both graphs a and b in figure 2.6 are shown as an index with 1966/67. Consistent with Hall (1984) we have only adjusted the published data for leasing since 1973/74 as it is suspected that the smaller quantity of leasing undertaken before 1973/74 was captured by the ABS sampling method used at this time.

C.2 Some Limitations

The first limitation relates mainly to the commitments data used for 1985/86. The national accounts only includes gross fixed capital expenditure on an accruals basis (i.e. when physical goods are received). Leasing commitments data (includes any… “firm offer to provide finance which has been or is normally expected to be accepted …”^[3]). This data could not be expected to line up with the national accounts exactly due to cancellations and lags between commitment and accrual.

Secondly, in both the commitments series and the earlier finance company data some leased items included may not be fixed capital. Their inclusion may overstate investment by particular industries since they were not included in aggregate gross fixed capital expenditure in the first place. Also, the ABS Capital Expenditure Survey (upon which data for gross fixed capital expenditure by mining, manufacturing and finance property and business sectors are based) includes only new fixed capital expenditure. Leasing data, on the other hand, includes the capital cost of new goods, the written down value of goods re-leased and the purchase price of the second-hand goods. The latter two categories are only included in the “all other private fixed capital expenditure” (see Appendix I, series (13)).

Additionally, the data includes leasing provided to government and public authorities which are not included in the measures of private fixed capital expenditure used in figure 2.6. However, since there is no way to distinguish between leasing of new or second-hand assets in any period or leasing to the private or public sector, all assets leased are assumed to be new and leased by the private sector.

A further limitation relates to the scope of the data in both series. It includes only plant and equipment on lease – not buildings. To the extent that leasing of buildings still remains attributable to the finance property and business sector and not to the industry of use, this category may still be overstated.

Clearly, it appears that there are limitations which may overstate and/or understate the extent of leasing undertaken making the series derived an approximation at best. However, it is possible that these limitations are offsetting, at least to some extent.

C.3 An Alternative Method

Another method for adjusting for leasing was devised by Hall^[4] in which 92.5 per cent of gross fixed capital expenditure on equipment by the Finance, Property and Business Services sector is allocated to other industries in proportion to their reported non-leasehold investment. However, the Hall method does not incorporate the information available from the leasing commitments data (source (i) (which was introduced after Hall's series was published) or allow for the Finance Property and Business services sector leasing to itself. Unlike Hall, our allocator is based on the total output (GDP) of each industry rather than reported non-leasehold equipment investment, on the assumption that the share of leasing undertaken by each industry is likely to be closely related to each industry's share of output in each period. Also, our method allocates all identifiable leasing to the four industry groups (including finance) while Hall's method allocates a fixed proportion of finance company equipment investment (92.5 per cent) to the other three groups.

While we have chosen to remain with our measure for this analysis, the extent to which the resulting data differ to those derived by Hall highlights the need to interpret the derived data with caution. Indeed, our method is just one plausible interpretation of the known data, and the data derived should thus be regarded strictly as estimates.

Post Script

Owing to changes in accounting standards it is expected that the ABS will provide more accurate estimates of gross fixed capital expenditure adjusted for leasing in 1987.

Footnotes

A finance lease refers to “… the leasing or hiring of tangible assets under an agreement, other than a hire purchase agreement, which transfers from the lessor to the leasee substantially all the risks and benefits incident to ownership of the asset without transferring the legal ownership” (ABS Form LF1 p.3). [1]

These transactions are identified in the Quarterly Estimates (source (c)) and The Round-up (source (m)). That is, sales from the public to the private sector which are leased-back to the public sector. The data shown in this paper has been adjusted for these transactions since September 1981. [2]

ASS Form LF1 p.3. [3]

A. Hall (1984), Leasing Finance and the Industry Composition of Investment. Business Council Bulletin No. 7. August, p.13-15. [4]