There are many possible functions of a budget. Australia's leading scholar of budgeting, John Wanna, in Controlling Public Expenditure (Edward Elgar, 2003) listed 13 different purposes of budgeting, including planning, priority setting, forecasting, fiscal management, authorising, exchequer (the accounting function).

From a treasurer's and government's perspective, one of the most important purposes of the budget is political - it shapes public opinion. A well-received budget can boost a government's re-election chances. If a budget does not go down well, past political consequences have included shuffling the responsible treasurer out of office, other ministerial changes, or changes in how cabinet considers budget strategy.

The political function is exemplified by leaks. It used to be that the contents of the budget were kept secret until budget night. These days, governments have realised they can milk the budget for days in advance with selected leaks of snippets of information. By the time of the budget not much is left unknown, but it can still be announced again - so government gets twice the publicity bang for its policy buck.

The political imperative means that the annual budget has a lot of theatre around it - the ritual of the budget lockup, where the treasurer and senior officials wander around selected journalists to sell their vision; a full House of Representatives to hear the budget speech - any government backbencher who does not turn up faces a severe talking to from the Whip; a round of carefully scripted interviews the next morning; an address to the National Press Club.

The political purpose is however more than just the electoral calculus. Through the budget a government sends messages about its priorities. Most budgets don't change very much by way of government tax or spending - tonight's will be an exception, with far higher new spending than usual due to the COVID-19 crisis. Even so, there is a general consensus around the basic structure of the budget. It is the decisions at the margins that signal what a government does or does not support. Some signals are with revenue - which groups benefit from tax cuts, or are expected to pay more. On the spending side, the myriad of small decisions at the margins about who gains and who loses tells a story about where the government wants the country to head.

The Parliamentary Library identifies three purposes for the budget: statement of the Commonwealth's economic and fiscal position, to outline policy priorities and announce new policies, and to outline the estimated resourcing of Australian government agencies. This is all true, but there's more. The budget discusses not only the Commonwealth economic and fiscal position but the health of the whole economy (including other levels of government, households and the private sector). It is the key statement of how the Australian government sees our economic future. It is also a vital accountability document. Through the budget the government makes itself accountable to the parliament for economic management and fiscal policy.

It is not like that in other countries. Budgeting in other countries fulfils a variety of functions different from ours. Of the OECD member countries the only one with a budget process roughly comparable to ours is New Zealand. The United Kingdom's budget concentrates on revenue projections and tax proposals; these are the focus of their Chancellor of the Exchequer's budget statements. Among the peculiar UK budget traditions is that the chancellor is allowed to drink alcohol - normally banned inside Parliament - while delivering the budget speech. Controls on government spending come through Treasury setting departmental spending limits. Although Australia is sometimes referred to as having a "Westminster" style of government, in practice our budgeting is completely different to that in Westminster.

In Canada, the budget is a broad statement of the government's economic intentions, sometimes more aspirational than realistic. Its equivalent of our spending budget is called main estimates, and may not necessarily occur in conjunction with the budget. In Canada, as in many other countries, the equivalent of our treasurer is known as the finance minister (departmentally, Treasury is a lower level function than Finance). In the USA, spending is multiple and various, with the presentation of a presidential budget proposal only one of many aspects of how the Congress decides to spend money or amend taxes. The budget tussle between the President and the Congress starts with proposals in February that generally by the time they are through the process in September are unrecognizable. Budgeting is not so much an annual event as an ongoing war between the legislative and executive branches of government with multiple skirmishes along the way.

In many developing countries the budget is more likely to be vague promises to comply with World Bank lending rules than a realistic statement of projected revenue and spending. Many still draw a distinction between a major capital (or "development") budget, often funded from foreign aid, and remaining components of domestic expenditure which are vague and indeterminate.

So what we know is budgeting looks completely different elsewhere. It can be annual only, or include forward years (what the OECD calls a medium term expenditure framework). It might be on an accruals basis, better practice, or simple annual cash allocations. It could be a comprehensive statement of policy, or just a few highlights. What we take for granted is by no means common practice.

Another feature we take for granted is the forward estimates. Our budget provides revenue and spending estimates for not just the coming budget year but three years thereafter. Forward estimates have become a vital government policy tool. We did not always have them, and most countries don't. Australia had patchy and unreliable spending projections in the 1970s, but our system of forward estimates was not cemented in place until the Hawke government's 1984 budget reform white paper.

The Finance Department developed forward estimates for spending; Treasury joined a few years later with forward estimates for revenue. Their reluctance to provide revenue numbers reflected the higher volatility and unpredictability of tax receipts - and it is still the case that revenue estimates are less certain. But on the spending side, forward estimates are highly reliable - so much so that if in any one year the government decided no new policy or savings decisions were required, the previous year's forward estimates (with technical adjustments for economic changes) would become that year's budget. Few countries have that level of reliability.

As we have discovered this year, however, a budget is not essential for the continuing functions of government. Although agencies cannot spend money without an appropriation (section 83 of the Constitution says "no money shall be drawn from the Treasury of the Commonwealth except under appropriation made by law") - the necessary appropriations can be passed by the Parliament separately. Having appropriations accompany the annual budget is tradition, not a legal requirement.

Unlike in the US, the Australian Parliament has few opportunities to change budget priorities. The party with a majority in the House of Representatives forms the government, so that House always agrees with the government. The Senate is limited by section 53 of the constitution, it cannot amend revenue bills or laws appropriating moneys for the ordinary annual services of the government. So in reality, the role of the Parliament in controlling government expenditure is limited. On budget night the Parliament is mainly theatre, providing the backdrop and bit players; but it comes into its own through estimates hearings, when accountability becomes the primary function. Public servants now face many happy days of answering estimates questions.