| |
Not So Small Government
Andrew Norton
Click here for PDF version
A growing economy can conceal an expanding government, argues Andrew Norton
Big compared to what? According to the Treasurer, deciding whether a government is big requires comparing its spending to the economy’s total size. Judged against other countries now and Australia at the tail end of Paul Keating’s government, he argues, small-government liberals should be enjoying a feeling of success that the government has managed to reduce its spending as a share of GDP.
This is a widely-cited indicator when evaluating size of government trends. But judgment requires more than comparing the headline indicator over time; it needs to be seen in its time. Government can look bigger or smaller depending on where a country is at in its business cycle. During recessions, governments tend to consume a greater share of GDP; expenses increase as more people become reliant on welfare while GDP shrinks. Without policy change, this is the inevitable result of a welfare state and a tax system based on income and consumption. It is what happened in Australia during the last recession. According to the OECD, government spending as a share of the Australian economy grew 4 percentage points from 1989 to 1993.
This is cyclical growth in government, which reverses itself as the economy recovers. Economic growth expands the economy, so without spending increases the size of government relative to GDP declines automatically. Economic recovery also reduces government expenditure, as unemployed people move off benefits into work. When growth-generated increases in tax and other revenue produce surpluses, government debt can be paid off and interest costs decreased. This happened under the Howard government, and was the largest source of reduced spending. While precisely quantifying the impact of cyclical factors is difficult, with continuous economic growth over the Howard government’s term it would have been remarkable for government spending not to have shrunk as a proportion of GDP.
But given favourable economic circumstances Australia has not experienced as significant a drop in welfare spending, as a proportion of GDP, as we might have expected. A small reduction in transfer payments relative to the total economy has been outweighed by a large increase in spending on ‘welfare services’. This disappointing result is partly because the welfare state continues to spread to people who are not poor, especially families (the welfare services category includes childcare). The latest Budget papers show a spending forecast of nearly $30 billion over 2007–08 in the category ‘assistance to families with children’. The Treasurer’s Intergenerational Report 2007 recorded that since the first intergenerational report the cost of family tax benefits alone had increased by 0.3% of GDP. Are families suffering so much with higher real wages and the lowest unemployment in a generation that they need all this help?
Though welfare demands from working-age people have a relationship with GDP growth rates, there is no intrinsic reason why other key Budget items should rise or fall in line with GDP. As I noted in my ‘Big Government Conservatism’ article—and which the Treasurer avoids discussing in his comment—economic growth doesn’t of itself increase the number of sick people, school children or old-age pensioners. Indeed, data published in the 2007–08 Budget papers (and partially reproduced in figure 1) shows that the Howard years coincided with an unusually low proportion of the population aged 0–14 or 65+, the groups whose needs trigger most government spending. Demographic as well as economic factors were working in the government’s favour.
To better incorporate demographic factors into the analysis, while still using a unit of measure that can be applied across the Budget, in my original Policy article I used spending per person measures. Expenditure per person helps show the year-to-year spending increases concealed by proportion-of-GDP analysis when the economy grows at an equal or faster rate.
Figure 2 contains updated calculations of real per person spending increases in key welfare-related areas. To many people, this figure would look more like the record of a Labor than Liberal government. Looked at in detail, there is a Liberal flavour to some of this spending: encouraging private schools, subsidising private health insurance, and fostering the family.(1) But another aspect of the Liberal tradition was missing—keeping taxation low enough for people to help themselves without government. On a per person basis there have been some compensating expenditure cuts, mostly (as expected) in public debt interest and on universities—though higher education outlays are now increasing again. But the overall trend is upwards. This is where the ‘big government’ label comes from. The government is big relative to the needs it must meet.
The Budget handed down in May again showed few signs of spending discipline. Most apparent reductions in spending in 2007–08 compared to 2006–07 are due to one-off payments in the earlier year, not to structural changes that would lead to sustained decreases in recurrent outlays. The exceptions are less spending on veterans (presumably as their numbers decline), a continued decline in interest expense, and a drop (for unspecified reasons) in ‘other health services’. Overall, however, though the Budget papers assume good GDP growth of 3.75%, government expenditure is predicted to increase by almost the same amount. For 2008–09 the Budget papers forecast spending to grow even more quickly, by 3.9% in real terms, and for the government’s own expenses to increase as a proportion of GDP.
This analysis of spending is, in one important way, too generous to the government’s record. As Robert Carling’s article to follow notes, the government’s record on tax is worse than its record on spending, with the difference seen in large Budget surpluses. Taxpayers over the last decade have paid for increased current spending, and also, as is usual in an economic recovery, for past deficit spending. What’s unusual is that taxpayers are also paying for future spending now. The Future Fund will be used to finance the superannuation liabilities of public servants, and perhaps Labor’s broadband proposals. The Higher Education Endowment Fund will be spent on capital works at universities, lessening the need for direct subsidy in later years.
The optimistic conclusion to draw from this fiscal innovation is that saving tax revenue now will lead to lower taxes later on, or at least lower taxes than might otherwise be levied, as the two surplus-financed Funds lessen demands on future taxation receipts. But if past experience is a guide, governments find a way of spending most of the revenue that they can get away with collecting. The Funds seem more likely to further increase long-term spending than to reduce long-term taxation.
Nobody pretends that smaller government is politically painless. It is much easier to give away than to take away. But the electoral hazards involved in decreasing outlays are reasons for being far more cautious than this government has been in setting up or expanding spending programmes. A legacy of entitlements will fuel growth in government well into the future.
Endnotes
(1) There is not space here to discuss all the reasons why spending may increase in these areas. My original article noted population ageing as a factor. Another reason is that some health costs are increasing more quickly than inflation for reasons unrelated to government policy.
|
