Land Markets and Business Cycles
in the United Kingdom and Australia
Estimates of the Public Equity in Land in Australia
The land price aggregates estimated in Appendix 1 are based on market
evidence of land transfer deals struck between willing buyers and
willing sellers. The deals take place in full knowledge of the fact
that the prices agreed become the tax base for local government rates
and State land taxes, which are levied at fairly predictable rates.
The annual interest costs buyers are willing and able to bear in
order to purchase sites are not increased by the existence of ad
valorem taxes on sites. The taxes must be included within the budgeted
annual cost of land ownership, leaving purchasers less able to bear,
by the same amount, interest costs on loans for site purchase. The
prices which they offer are thus reduced by the capital amounts which
correspond to those interest cost parings. This is expressed in the
classic tax capitalisation equation:
V = a ÷ (i + t) e.g,
1,214.3 = 85 ÷ (0.07 + 0.00) and
1000 = 85 ÷ (0.07 + 0.015)
where V = selling price of land;
a = annual net income of land before land tax (but after other
i = property market yield;
t = annual rate of tax on the market value of land.
A tax-induced reduction of land prices does not mean that financial
equity in land disappears. It is transferred into public ownership -
all other aspects of ownership remaining with private purchasers. In
the simple numerical example, which has tax rates and yields typical
of Australia in the late 1980s, 214.3 of equity is transferred to the
public sector by the imposition of a 1.5% land value tax rate. Thus,
P = tV ÷ i e.g,
P = 214.3 ÷ 0.07 = 15
where P = public equity in land; and
tV = the revenue from the tax on the assessed selling
price of land (0.015 x 1000).
Land taxes are, in effect, rental payments on the portions of the
financial equity in sites which state and local governments
lay claim to. A complete estimate of the value - equity value - of
Australian land must therefore include public equity as well as
The yield of the public equity is realized through annual taxes, and
the same yield percentage applies to that income as is evident in the
market for private equity. The public equity may be calculated by
capitalising the taxes at the same yield as in the private sector.
As, V = a ÷ (i + t) is algebraicly the same as V = (a - tV) ÷
then ad valorem taxes on land (tV) plus private net-of-tax land rents
(a - tV) form the annual value of land. Those same taxes plus
net-of-tax rents capitalised at property market rates of yield form
the complete capital value - equity value - of Australian land. That
V + P = (a - tV) ÷ i + (tV ÷ i )
e.g., 1000 + 214.3 = (85 - 15) ÷ 0.07 + (15 div 0.07)
Appendix 1 calculated V for the whole of Australia. To complete the
valuation it is necessary to calculate P; that is, to add up all the
real property tax revenues collected in Australia.
The source for the tax revenues is given in Appendix 1. That source
does not give the revenues obtained by Net Annual Value (NAV) rating,
and other types of rates levied on improvement values as well as land.
In later studies the Land Values Research Group (LVRG) included in
their calculations proportions of the revenues under these rating
systems equal to the proportions of assessed site values to improved
values (sites plus improvements), assuming that only those portions of
the taxes captured land rents. However, economic theory suggests that
the whole of the incidence of real property taxes is on land rents
alone, so that they reduce net land rents by the full amount of the
taxes. In fact, they reduce land rents by more (see the author's other
three chapters in thse two volumes). The assumption made in the
following calculations is that all the revenue raised by real property
taxes other than site value taxes consists of land rent.
For 1937/38, the revenue of rating systems other than site value
rating is extrapolated back from later figures. Between 1957/58 and
1976/77 SVR appears to have taken 56 to 61% of rates revenue, while it
has covered 63 to 65% of local authorities. It undoubtedly covered
less local authorities in 1937/38, so 55% has been chosen as the
proportion of rates revenue contributed.
For the sake of completeness in Table 1 the rates revenue has been
split 65:35 - in line with the data for later years - between that
raised by local authorities for general government purposes and that
raised by public utilities.
LVRG's 1961 booklet is the source for all the property tax revenues,
apart from Canberra's rates (for which, see Appendix 1). The data
shows local rates revenues split 64:36 between general government and
utilities, and 58:42 between UCV and NAV tax bases.
Hutchinson (c.1967: 9, 28-31) provided the property tax revenue data.
The split between municipality rates revenue and utility rates revenue
was 67:33, and between UCV and NAV tax bases 56:44.
Hutchinson (1981: 12-13, 23-24, 41-46) provided the property tax
revenue data. The splits between municipality rates revenue and
utility rates revenue and between UCV and NAV tax bases were both
61:39. The leasehold system in the ACT was altered in 1971,
effectively converting leasehold rents into local government rates.
The distinction between leasehold rents and local government rates is
Hutchinson's data may now be checked against two widely available
annual publications: the International Monetary Fund's
Government Finance Statistics Yearbook (GFSY) and the OECD's
Revenue Statistics. Both treat all the ACT public revenue as federal
(Commonwealth) government tax revenue. Both show local government
general purpose rates revenue. Only the OECD distinguishes recurring
State taxes on immovable property (i.e., the State land taxes). Rates
revenue raised by utilities, however, is not distinguished by either.
1970/71 - 1989/90
The IMF and OECD publications cover the whole of this period (except
the subdivisions for the first two years, which are partly estimated).
Rates revenues for the utilities have to be estimated. It is assumed
that they remained at the same real level after 1976/77, as they did
not rise as fast as taxation. So the figure for that year is raised
thereafter in line with inflation, as measured by the Gross Domestic
Product deflator. Before that year, however, an assumption of real
growth in utility costs must be made to raise the revenue growth rate
towards the growth rate implied by the difference between the 1964/65
and 1976/77 figures. The growth rate before 1976/77 is assumed to have
been in line with the growth of the national income at current prices.
This still leaves growth rate for the second half of the 1960s greater
than the national income, so the figures for the early 1970s may be
Mining and forestry royalties and levies
Hutchinson recorded mining and forestry royalties paid to the States
amounting to $0.241bn in 1976/77 (1981: 15). He also noted that
$1,137bn was paid to the Federal Government as "crude oil levy
and other mineral levies" (p.35). Taken together with the land
tax revenues in column H of Table 1, these raise the tax revenue on
land rents for 1976/77 to $2.663bn. Adding utility rates (not part of
"taxation") raises the public revenue total to $3.333bn.
Land taxes and royalties/levies (but excluding utility rates) as a
proportion of total taxation become 10.8%. As a proportion of the
national income, total rents (private and public, including utilities)
become 11.4%. Were all the financial equity in Australian land
publicly owned, as it is with most mineral deposits, the rental income
would have defrayed 31.5% of the national tax bill (in addition to
continuing to provide revenue for public utilities).
Royalties tripled by 1983/84. The crude oil level tripled. The
figure in column H of Table 1 may, therefore, be raised to $6.822bn,
which was 12.2% of national taxes. Total rents become $26.443bn, or
17% of national income. They could have defrayed 45% of the national
tax bill in addition to providing revenue for public utilities.
- Richard Braddock, "Land
Value Taxation: Some Australian Experience," annual Macquarie
Lecture in Honour of Henry George, Macquarie University, Sydney,
- Ibid., p.9.