REFERENDUMS – THEN AND NOW

Constitutional change does not come easily in Australia by way of referendum. To change the constitution, Australians must vote a majority, for any proposal, in a majority of states. Since Federation, there have been 44 proposals put to a referendum and only 8 have been approved. So where does this leave the referendum on “The Voice” for Indigenous Australians to have constitutional change? Professor George Williams is Deputy Vice-Chancellor of Planning and Assurance at the University of New South Wales and an expert on Australian Constitutional law. He addressed The Sydney Institute on Wednesday 3 August 2022 to give an assessment of Australian referendums, their weaknesses and reform needed.

REFERENDUMS – THEN AND NOW

GEORGE WILLIAMS

This topic is very much in the news given the commitment of the Albanese government to change the Constitution to insert an Indigenous Voice to Parliament. The only way of achieving this is via a referendum of the Australian people.

Despite the optimism of those who support reform, it must be recognised that the task will be a formidable challenge. Some of Australia’s most persuasive and successful leaders have failed to win a referendum. Bob Hawke is a good example, with failures in 1984 and 1988, the first coming at the height of his popularity as a new prime minister.

Robert Menzies is another. Despite being Australia’s most electorally successful prime minister, he was left bitter and disillusioned after his unsuccessful attempt in 1951 to give Parliament the power to ban communism. He failed despite early polling showing 80 per cent of Australians supported his reform. Menzies described winning a referendum as “one of the labours of Hercules”. He did not make another attempt at constitutional reform.

Hawke and Menzies’ experience reflects Australia’s long record of referendum failure. Since 1901, 44 proposals have been put to the Australian people with only eight of those succeeding. No referendum has been passed by the people since 1977 when Australia voted, among other things, to set a retirement age of 70 years for High Court judges.

Forty-five years have now passed since Australia changed its Constitution. At more than one-third of the life of the nation, this is by far the longest period that Australia has gone without amending its Constitution (the next longest period was 21 years between the 1946 and 1967 referendums).

The Australian Labor Party has the lion’s share of failed referendums: 25 of the 44 proposals (57 per cent) have been put by Labor governments. Just one – the 1946 (Social Services) referendum put by the Chifley government – passed, a success rate of four per cent. By contrast, 7 of 19 non-Labor proposals, or 37 per cent, have passed.

The fact that the Australian people vote “No” to a referendum is not necessarily a problem. Australians will understandably vote “No” to poor proposals, and the record may accordingly reflect the good sense of the community.

However, as an aggregate, the record is problematic, and reveals a long record of complacency and political mismanagement. It begs important questions such as: why have Australian governments put so many proposals that have proved to be unpopular? Why have Australian governments failed to identify and champion more proposals that benefit the community and have popular support?

The record overall is also problematic because it is in the national interest that some constitutional change does occur. As the Constitution ages, the need for reform in some areas increases, yet the pace of reform has never been slower. One prominent example is our dysfunctional federation, which affects everything from the quality of our hospitals and schools through to how much tax we pay.

A Business Council of Australia study found that problems with our broken federal system means that every Australian family pays an unnecessary $1,100 in tax each year. This is wasted money. Overall, we are taxed a pointless $9 billion. This is the amount being used to prop up the Australian federal system.

The figure greatly underestimates the true cost which is how much the community pays extra tax for the duplication of services, buck-passing and inefficiency that bedevils the relationship between our federal and state governments. It does not include the money lost to businesses in having to comply with unnecessary red tape and extra regulation from multiple layers of government.

Taking some of these other costs into account, it has been estimated that the duplication and extra co-ordination costs in Australia’s federation are $20 billion a year, or 9 per cent of all general government expenses or 3 per cent of GDP.

As someone who has carefully studied the Constitution and its operation for decades, I see a document that has served us well but that, equally, needs updating. Any document of this kind needs to keep up with the times, lest it inhibits efficient government and economic growth.

Unfortunately, we have had very few people on either side of politics who have been able to navigate the referendum process. Overall, the record reflects the fact that our political leaders make the same mistakes time and time again. The running of poorly conceived proposals with inadequate political strategies has cost the community many hundreds of millions of dollars.

It is important to look at how referendums are run. It is 23 years since we had such a national poll, and so it is important to remind ourselves of the basics. I shall focus on two aspects of referendums. First, the need to modernise the rules for conducting referendums. Second, what we can learn from the referendum record about how to win an elusive Yes vote from the Australian community.

The basics

The Australian Constitution can be changed by a process that reflects its democratic and federal characters. Section 128 of the Constitution states that it “shall not be altered except” by way of a proposed law for its alteration being 1) passed by an absolute majority of both Houses of the federal parliament, or by one House twice; and 2) passed at a referendum by a majority of the people as a whole, and by a majority of the people in a majority of the states.

The requirement that, in addition to a national majority, the electors in a majority of States must vote “Yes” reflects the framers’ conception of the Constitution as establishing a “federal Commonwealth”. This dimension is further reflected in the requirement that some changes receive the consent of a particular state. This prevents the Constitution from being altered at the behest of a majority of states to single out one or two states for negative treatment. For example, a majority of voters in the affected state is required to approve changes “diminishing, or otherwise altering the limits of the State”.

Once a proposal for constitutional change has been passed by both Houses of the federal parliament, section 128 states that it “shall be submitted” to the people. This constitutional imperative has on occasion been ignored, including on the last occasion that Australia seriously considered holding a referendum.

In 2013, the federal parliament passed an Act to alter the Constitution to authorise the Commonwealth to make grants of financial assistance directly to local government bodies. The referendum was scheduled to be held with the federal election of that year. It did not proceed after Julia Gillard was replaced by Kevin Rudd, and the federal election was brought forward.

Conducting a referendum

The Constitution sets out what is required for a referendum to succeed, but not how it should be run. It is silent on this, except to say that the referendum “shall be taken in such manner as the Parliament prescribes”.

Accordingly, the referendum procedure is regulated by the Referendum (Machinery Provisions) Act 1984 (Cth). It is an electoral act for the holding of referendums, and regulates all aspects of the poll. For example, section 45 imposes a system of compulsory voting: “It is the duty of every elector to vote at a referendum”, with a penalty for failing to do so.

These rules were first set down by parliament soon after Federation in 1901. They have been amended from time to time, including in 1912 to provide voters with a written Yes/No pamphlet setting out 2000 word statements in favour of each case. These must be approved by the parliamentarians who voted for and against the change.

Decades have passed since these rules were reformed to reflect changes in Australian society and new campaigning techniques. The result is a referendum rulebook that is woefully inadequate. It is a major barrier to holding a fair and well run referendum, let alone to winning such a poll.

The referendum machinery act shows its age in having key provisions that precede the introduction of radio in Australia, let alone television or the internet. As a result, the primary method for disseminating the Yes/No pamphlet is still for it to be printed and then posted to each voter. The rules also prohibit disorderly conduct at public meetings, but fail to account for the existence of social media.

Other rules have been shown over time to be problematic. For example, the Commonwealth cannot spend money arguing for or against the change. There are merits to federal neutrality, but this makes little sense when there are no limits on any other organisation or government. As a result, past referendums have seen public debate monopolised by groups seeking to oppos­e reform. State governments have also used their finances to undertake blanket advertising against change, with no capacity for the Commonwealth to respond.

These and other aspects of our referendum machinery legislation are long overdue for reform. Importantly, this needs to occur now before we embark on the next referendum campaign. Experience shows that it is wise to do so well before the campaign itself. Without this, rational debate over what amounts to a sensible and fair process of holding a referendum risks being overtaken by political assessments of whether reform would favour the yes or no case.

This happened on the last occasion Australia came close to holding a referendum. In 2013, the federal Parliament agreed to put a proposal to the people on whether to permit the Commonwealth to make direct financial grants to local government. The idea won overwhelming support across the major parties. However, when Parliament came to debate the rules for the poll, and in particular how much money should be provided to each side, agreement broke down. In the end, the proposal foundered and was never put to the people.

These problems were recognised by the Inquiry into Constitutional Reform and Referendums conducted in 2021 by the House of Representatives Standing Committee on Social Policy and Legal Affairs chaired (until his appointment as Speaker) by Coalition MP Andrew Wallace. It produced a valuable, bipartisan report that suggested sensible improvements including:

• a public awareness and education campaign on the Constitution and Australia’s democratic system;
• a new joint standing committee of the Australian Parliament on constitutional matters with a broad mandate to inquire into and make recommendations on constitutional reform;
• conducting community conventions to debate constitutional matters;
• enabling the Electoral Commissioner to distribute the Yes/No pamphlet to electors using any method thought appropriate;
• permitting the Australian government to fund referendum education campaigns and the arguments for and against a proposal; and
• establishing an independent expert panel to advise Parliament in the lead up to a referendum on the wording of the question and neutral, explanatory wording to be included in the Yes/No pamphlet.

These bipartisan recommendations should be adopted now.

I would also like to see these support a regular, popular constitutional convention, convened once each decade or “half-generation”. It should consider proposals put to it by:

• the federal Parliament,
• a majority of the states, or
• the petition of a large number of Australians.

The Convention should debate reform ideas and recommend proposals to the federal Parliament for submission to a referendum.

It is also past time for a holistic review of the Constitution. These have been held at roughly 30-year intervals, reporting in 1929, 1959 and 1988. This means that the last review should have reported in 2018 or 2019. These processes typically take a number of years, meaning that it is well overdue.

Winning a referendum

In People Power: The History and Future of the Referendum in Australia, David Hume and I examine all of Australia’s 44 referendums in detail, and how this experience might be applied to hold referendums with greater prospects of success.

We conclude that Australia must avoid repeating, yet again, the same past mistakes, and that there are realistic prospects that the Australian people will vote “Yes” if a referendum is approached in the right way.

To maximise the chances of winning a referendum, a political strategy should be based upon the following principles:

Bipartisanship

To date, bipartisan support has proven to be essential to referendum success. Referendums have needed support from the major parties at the Commonwealth level. They have also needed broad support from the major parties at the state level. The history of referendums in Australia provides many examples of proposals defeated by committed opposition from a major party at either level. This has been a particular feature of failed referendums put by the Australian Labor Party.

This has been recognised across the political spectrum. After losing the 1951 referendum, Menzies saw the result as “further proof of the fact, demonstrated in every previous referendum in our history, that no amendment to the Commonwealth Constitution can be carried if the Parliamentary Opposition is against it.”

Menzies’ observation that a referendum is doomed unless it has bipartisan support proved true in 1951 but may not be the case today. Much has changed in the intervening years, and more than two decades has passed since Australia voted in its last national referendum in 1999. It is dangerous to make assumptions based upon this past experience.

What is clear though is that the chances of a proposal being supported at a referendum will be lower if the opposition mounts a “No” campaign. Some Australians will take their lead from this to vote “No”. Others will be influenced by the uncertainty that such a political contest will create. Past opposition members, such as Peter Reith in 1988, have proved to be formidable opponents because of their capacity to sow doubt. Many Australians have voted “No” as a result even where they felt well disposed towards the change. The 1999 referendum on the republic is a good example of this, with the majority voting No despite polls showing most Australians supported the idea of a republic.

Popular ownership

Just as deadly as partisan opposition is to constitutional reform is the perception that a reform idea is a “politicians’ proposal”. From the 1967 nexus proposal, which was felled by the cry of “no more politicians”, to the Republic referendum, which was killed off by the claim that it was the “politicians’ republic”, Australians have consistently voted “No” when they believe a proposal is motivated by politicians’ self-interest.

This reflects a well-known undercurrent of distrust of Australian politicians. The constitutional design of Australia’s reform process exacerbates this problem. Politicians, and only politicians, can initiate constitutional reform through the federal parliament. This renders every referendum proposal at risk of being perceived as self-serving, especially of those interests aligned with the Commonwealth.

This must be countered by the campaign and drafting process that maintains an ongoing connection between the community and the referendum proposal. Popular conventions are one way of doing so.

Popular education

Surveys of the Australian public show a disturbing lack of knowledge about the Constitution and Australian government. Rather than being engaged and active citizens, many Australians are know little of even the most basic aspects of government. This is often a reflection of the fact that disengaged citizens tend to have less knowledge about their system of government and any reform being proposed. The problem has been demonstrated over many years. For example:

• one survey found that almost half the population did not realise Australia had a written Constitution, with the figure being nearly 70 per cent of Australians aged between 18 and 24.
• recent polls show that majority of Australians believe we have a national Bill of Rights.

These problems can be telling during a referendum campaign. A lack of knowledge, or false knowledge, on the part of the voter, can translate into a misunderstanding of a proposal, a potential to be manipulated by the “Yes” or “No” cases and even an unwillingness to consider change on the basis that “don’t know, vote No” is the best policy.

4      Sound and sensible proposal

As important as it is to get the process of generating proposals right, it is equally important to get the proposals themselves right. A major weakness in Australia’s referendum record to date is that attempts at reform have been dominated by what have been (often rightly) perceived by the population to be grabs for extra federal power. Other proposals, like that on the republic in 1999, have had flaws that have been sufficient to many dissuade people from voting “Yes”.

Referendum changes can shift the nation in the new direction and alter the Constitution for 50 or 100 years. Australians understandably want to support a proposal that they know is watertight.

Conclusion

Australia’s record of past failed attempts at constitutional reform do not mean that winning a referendum is “mission impossible”. Referendum success is achievable, and indeed necessary, when it has been built on a modern, fair process.

Even with such a process in place, we should expect that referendums may fail when there is high level political disagreement, or when poor management means that the Australian people feel left out or confused about what is being changed. People will also vote “No” to a proposal that is dangerous or has been poorly thought out.

This of course is common sense, yet the referendum record displays a tendency to repeat these same mistakes time after time. Australia’s referendum history contains few successes for good reason.