Over the next two weeks I’ll be writing a few blog posts based on the referendum results data in my data repository – looking at the history of referendums throughout the 20th century.
The Voice referendum will be the first constitutional referendum held since 1999 – that’s 24 years without a referendum.
Throughout the 20th century, referendums were held at least once every decade. The longest gap was between the 1951 referendum on banning the Communist Party and the 1967 referendums. The 2000s and 2010s were the first decades since Federation with no attempt to change the Constitution.
For this blog post, I’ve made a number of charts that all have the same shape, showing a number of characteristics of the 44 referendums held in the 20th century, grouped by the decade they were held in.
Fourty-four referendums were held throughout the 20th century, but they were not evenly spread through the century.
The first referendum was held in 1906 to switch Senate terms from expiring at the end of the year to the middle of the year. It was not controversial and passed easily.
The decade with the most referendums was the 1910s – twelve were held that decade, spread over four referendum days. There was a second spike in the 1940s. The next big spike was in the 1970s, when ten referendums were held across three events.
Eight of these referendums passed. A further five failed despite a majority of voters voting Yes, since they failed to achieve a majority in at least four states. Three of these referendums saw three states vote Yes, and on two occasions just two states voted Yes.
The successful referendums have been spread throughout the century, with three of them won in 1977.
No referendum has been won since 1977, and no referendum has come close since 1984.
Next up, let’s look at which party was in government when each referendum was initiated.
Labor initiated 25 referendums in the twentieth century, with non-Labor governments initiating 19. Labor tended to call a lot of referendums in a short period: eight in 1911-13, five in 1944-48, six in 1973-74, and six in 1984 and 1988. Non-Labor governments tended to hold fewer referendums spread out over a broader timeframe.
It’s particularly interesting to see how the topics of these referendums have changed over the century.
Most referendums can fit into one of two categories: an expansion of Commonwealth powers, or making changes to democratic structures (changes to parliamentary terms, the size of parliament, the terms of judges, the republic, etc). There were also four referendums I classify as “federal finance”, dealing with distribution of funds with the states and local councils, or state debts. I classified the 1988 Rights and Freedoms referendum and the 1999 Preamble referendum as not fitting into those other categories.
Most referendums in the first half of the century focused on expanding Commonwealth powers. Yet in the second half, changes to democratic structures tended to dominate, with a handful of questions of changing powers still hanging around. The 2023 referendum fits this mould, with the Voice a potential new democratic institution.
Quite a lot of the topics of referendums came up again and again, sometimes slightly tweaking the issue.
After the 1911 referendums failed, the Labor government broke up the content into six separate referendums, all of which narrowly failed with over 49% of the national vote and three states voting Yes.
Labor governments attempted to gain powers over prices in both 1948 and 1973.
Four referendums were held to tweak the relationship between House and Senate terms so that Senate terms would end at the same time as House elections in 1974, 1977, 1984 and 1988, with the fourth referendum also extending House terms and shortening Senate terms to four years.
There has also been a shift away from holding referendums alongside federal elections and towards holding them on their own schedule.
The 1911 referendums were held mid-term, but the other ten referendums in the 1910s were held alongside federal elections.
No referendum has been held alongside a federal election since 1984, with a majority of referendums in the 1970s and 1980s held mid-term. That followed the 1951 and 1967 referendums being held mid-term.
Finally, two charts showing how each referendum scored, both in terms of states voting Yes, and the national Yes vote.
The number of referendums losing in a clean sweep have become far more common in recent decades. Just two referendums prior to 1973 failed to win a single state. Both 1973 referendums were defeated in all six states, and a further five referendums suffered the same fate in the 1980s, along with the two 1999 referendums.
Looking at the national vote, there have been more referendums failing to achieve 40% of the vote. While most of the referendums in the 1910s failed, the Yes vote usually came close to half the national vote.
In my next blog post I’m going to explore how the dynamics of the double majority have tended to work – what state has been the “tipping point”, and how did Yes perform in that tipping point state compared to the national vote. I’ll then have another post looking at which states have tended to vote Yes or No, and vote together with other states.
Thanks Ben, a lot to digest there and it possibly doesn’t bode well for the upcoming referendum getting a tick (or should that be a Yes)!
I probably have no great insights into what you have written, other than a few general observations and really you need to delve into the nuts and bolts of each referendum to make any real sense of it all.
Labor probably propose more referendums than the other side, because they tend to be in power less frequently. Thus they tend to take these opportunities more fervently when the opportunity arises.
If I am reading it right, then I am surprised by the number of referendums that fail on the overall majority vote. It seems to be a myth that most referendums get a majority of the popular vote, but fail to cross the 4 states threshold.
My own personal opinion is that we Australians really don’t hold onto / understand our Constitution / Civics as much as say Americans. Therefore many people go into the polling booth from a initial position of ignorance and therefore it is easier to vote No than Yes, purely because they don’t really know or care about what they are being asked to vote on. The current referendum is a good example. The number of vox pops I have heard from people saying they don’t know or understand this referendums, when every person has received a package setting out both the Yes and No case demonstrates to me that people just don’t engage in these things, so No is the easier case to argue for.
I’ve never heard the claim that most referendums get a majority of the popular vote. If people do say that, they’re very wrong.
I have heard on numerous occasions, mainly by talking heads on ABC radio and TV.
“Labor probably propose more referendums than the other side, because they tend to be in power less frequently.”
Its not only that Neil-they are a progressive party so the instinct for change is strong.That means most of what they propose does not get bi partisan support.They don’t seem to have learned any lessons from this over the years.
Comments are closed.