cover_winter07

Winter 2007


 
 
 

 


The Tax Take is up
Robert Carling
Click here for PDF version

The mantle of ‘highest taxing Treasurer’ may go with the job, but the government’s tax record is still disappointing, argues Robert Carling

The exchange between Andrew Norton and Peter Costello is about the size of government. They have looked at this issue from the expenditure side of the budget, but it can also be looked at from the revenue side. The government diverts potential claims on resources from the private sector to itself through revenue, and uses those claims in its expenditure. In particular years and even over long periods expenditure and revenue can diverge, but ultimately they will tend to converge.

Costello has made much of Norton’s (and Des Moore’s) concession that Commonwealth government expenditure has fallen slightly as a proportion of GDP since the Howard government came to office. But its revenue (and particularly its tax revenue, which makes up more than 90% of the total) can be seen to have risen when proper comparisons are made over time. This is how the government has turned a deficit of 2% of GDP into a surplus of 1% of GDP. It has been more successful in increasing the tax burden than in containing its own spending.

In one sense assuming the mantle of ‘highest taxing Treasurer’ is unexceptional. Historically it has gone with the job. The long-run trend in the tax/GDP ratio has been up. It is only the short-lived Treasurers that might have missed out on the distinction of presiding over a higher tax/GDP ratio than their predecessors. What is disappointing is that those who believe in the benefits of smaller government have been waiting decades for a government to reverse the trend. Each successive government, including the ones whose rhetoric espoused the cause of smaller government, has ratcheted up the tax burden. The Whitlam, Fraser and Hawke/Keating governments left office with the tax burden higher than they found it. The Howard government would also, were the 2007–08 budget year to be its last.

Mr Costello claims to have lowered the tax burden, but only by comparing post-2000 apples with pre-2000 oranges. His Budget papers show a Commonwealth tax/GDP ratio of 22.3% in 1995–96 (the last budget year of the Keating government) falling to 20.9% in 2006–07 and 20.7% in 2007–08.(1) The problem with this is that the Commonwealth tax series after 1999–2000 excludes GST revenue on the grounds that it is all passed on to the states in the form of GST revenue grants, but includes all Commonwealth tax revenue up to 1999–2000 even though some of it was in effect passed on to the states in the form of general purpose grants similar to GST revenue grants. The only difference between current and pre-GST arrangements is that the GST is earmarked for the states, whereas before the GST no Commonwealth tax was specifically earmarked. The Budget papers make no deduction from Commonwealth tax revenue up to 1999–2000 to recognise the pre-GST general purpose grants, and in failing to do so create a serious discontinuity in the series. Comparisons before and after this discontinuity are not comparing like with like.(2)

One way to restore consistency would be simply to add the GST revenue back into the Budget paper series of Commonwealth tax revenue. As GST revenue is 3.8% of GDP, this raises the tax/GDP ratio to 24.7%, compared with 22.3% in 1995–96. This is what the Australian Bureau of Statistics does in its Government Finance Statistics compilations, on the grounds that the GST is constitutionally, legislatively and administratively under Commonwealth control. But for the purposes of like-with-like comparisons over time it is overkill because the Commonwealth introduced the GST partly to pay for the removal of various state taxes. This switch from state to Commonwealth taxation exaggerates the increase in the Commonwealth tax burden.

A better way to restore consistency is to make a deduction from Commonwealth tax revenue up to 1999–2000 for the amount passed on to the states in general purpose grants.(3) For completeness, we should also deduct general purpose grants to local government from Commonwealth tax revenue in all years. The resulting synthetic series portrays Commonwealth tax revenue retained for the Commonwealth’s own purposes. The figures are shown in the following table.

carling_graphic

The Commonwealth own-purpose tax burden has risen from 19.0% of GDP in 1995–96 to 20.7% in 2006–07. It dropped fractionally from 2005–06 to 2006–07 as a result of the income tax cuts in last year’s Budget and is projected to drop a little further in 2007–08, but remains higher than in 1995–96. According to the forward estimates it will go up again slightly from 2007–08 to 2010–11.

This is how the government has turned a budget deficit of 1.9% of GDP in 1995–06 into a surplus of 1.3% of GDP in 2006–07, a turnaround of 3.2% of GDP: by increasing the tax/GDP ratio by 1.7 percentage points, increasing non-tax revenue/GDP by 0.5 percentage points (4) and reducing the expenditure/GDP ratio by 1 percentage point.(5) The same turnaround from deficit to surplus, which is to be applauded, could have been achieved without the increase in the tax burden had the government taken a more disciplined approach to its spending and put its ‘smaller government’ rhetoric into action.

As noted above, ultimately expenditure and revenue will tend to converge, consistent with a balanced budget. As the Commonwealth Budget is now in surplus, convergence could occur through a reduction in the tax burden or an increase in the expenditure/GDP ratio. Which path is taken will determine whether the size of government retreats or advances.

Unfortunately there is no sign that a structural decline in the tax/GDP ratio will occur.(6) The Prime Minister and the Treasurer have made it clear that their first priority in each budget is to attend to expenditure ‘needs’ and that tax reduction, if any, is a residual consistent with holding the Commonwealth tax burden to its current level of around 21% of GDP. Under current policies it is more likely that government expenditure will drift up as a share of GDP. Indeed, all (and then some) of the one percentage point reduction in the expenditure/GDP ratio since 1995–96 occurred in the Howard government’s first two years; in the nine years since, the ratio has again trended upward.

Endnotes
(1) Commonwealth Government, 2007–08 Budget Paper No 1, Statement 13, Historical Australian Government Data, Table 2 (Canberra: Commonwealth Government, 2007). This table draws attention to a break in the series between 1998–99 and 1999–2000 due to the shift from cash to accrual accounting, but the effect is not large and in any case does not stop the government from making comparisons of data before and after the break.
(2) Rory Robertson, Macquarie Bank’s interest rate strategist, has recently made this point with extensive supporting data in Budget/Federalism Watch, Macquarie Research Strategy, 15 May 2007 (unpublished).
(3) Specific purpose payments to the states could also be deducted from Commonwealth tax revenue in all years on the grounds that they are not Commonwealth own-purpose funds, but we have not done so here because the Commonwealth has much more year-to-year discretion in setting the amount of these payments and because their purpose is partly to achieve Commonwealth policy objectives and therefore they are in the nature of Commonwealth own-purpose funds.
(4) The non-tax revenue/GDP ratio actually rose by 0.8 percentage points, but this includes Future Fund earnings in 2006–07, which the Commonwealth excludes from its ‘underlying’ cash surplus.
(5) With expenditure defined as Commonwealth own-purpose payments, which exclude general purpose grants to the states and local government.
(6) There would be a cyclical decline in the tax/GDP ratio in the event of a recession or a pronounced slow-down in economic growth, but this would be reversed in the recovery phase of the cycle.

 
 

 


Policy is the quarterly review of The Centre for Independent Studies. For more information on subscribing to Policy, click HERE

If you are interested in the Centre's activities and publications, why not subscribe to e-PreCIS, our regular email update on the latest news and events.

(e-PreCIS requires html capable email facilities, such as Microsoft Outlook Express or Netscape Messenger)