Independent publication · Not affiliated with, endorsed by, or representing Victoria Police. About us
Crime

Victoria’s homicide rate, and why it tells you less than you think

Every year, when the Crime Statistics Agency publishes its annual update, a story or two will lead with a headline about Victoria’s homicide rate. “Highest in a decade” is a common framing. “Lowest in a generation” turns up less often but does appear. Both have been true, in different recent years, and both are usually less informative than they sound.

Our newsroom looks at the homicide series alongside everything else in the annual release. Jack Renton has put together a plain-English explainer of what the numbers mean, why year-on-year movement is mostly statistical noise, and what does and does not show up when you compare Victoria to its peers.

What the official series actually contains

“Homicide” in the Crime Statistics Agency series is a category that bundles together several distinct offences: murder, manslaughter, attempted murder, conspiracy to murder, child homicide, infanticide, defensive homicide (where the offence still appears in records, although it was abolished in 2014) and a small number of related offences. The headline number most often quoted is the count of recorded homicide and related offences, not just murder convictions, and not just deaths.

The Australian Bureau of Statistics’ Recorded Crime — Victims series provides an alternative count, structured around the victim rather than the offence, and the Australian Institute of Criminology’s National Homicide Monitoring Program provides a third, more analytic dataset that drills into circumstance, relationship between victim and offender, weapons used and motive.

The three series do not always agree to the unit because they count different things. None of them is wrong; they are designed for different purposes.

The number that is closest to what a reader probably means by “Victoria’s homicide rate” is the count of murder and manslaughter victims per 100,000 residents, drawn from the ABS series. In Victoria that number has, in recent years, run somewhere in the order of 1.0 to 1.6 per 100,000.

Why year-on-year movement is mostly noise

Victoria has a population of about 6.9 million. At a rate of around 1.2 per 100,000, that produces somewhere in the order of 80 to 100 homicide victims a year.

That sounds like a lot, and in human terms each one is a catastrophe. But for statistical purposes, 80 to 100 is a small number. A move from 85 to 105 — a 23 per cent year-on-year increase, the kind of change that produces a “highest in a decade” headline — is exactly the kind of swing you would expect from random variation in a count of that size, even if the underlying rate has not changed at all.

Statisticians have a precise way of expressing this: in a Poisson process with an expected count of 90, the standard deviation is the square root of 90, or about 9.5. A single year coming in at 105 is roughly one and a half standard deviations above the expectation. That is unremarkable. To establish a real change in the underlying rate you need to look at trends across multiple years, ideally five or more, and at the rate per 100,000 rather than the raw count, because population is itself moving.

This is why criminologists and the Crime Statistics Agency itself caution against drawing conclusions from a single year’s homicide figure. The statistical signal is genuinely weak. The headline that goes with it is, often, much louder than the data warrants.

The state-by-state comparison

Australia as a whole has a homicide rate of around 0.8 to 1.0 per 100,000, which is among the lower rates in the OECD. The United States runs in the order of 5 to 6 per 100,000. The United Kingdom, France and Germany run between 0.6 and 1.2.

Within Australia, the Northern Territory has consistently the highest homicide rate of any jurisdiction, several times the national average, driven by a combination of remote-community vulnerability, alcohol-related violence and the long-standing failures of public services in those communities. The other state and territory rates cluster more tightly, with Victoria typically running close to the national average — sometimes a touch above, sometimes a touch below, depending on the year.

None of that detail tends to make it into a headline. “Victoria’s rate is the highest in a decade” is a more attention-grabbing story than “Victoria’s rate continues to sit in the middle of the Australian range and well below most OECD peers”.

What is genuinely changing

Strip out the noise and look at the trend. The picture, across the last twenty years in Victoria, has three real features.

The first is that the overall rate has trended down. The 1990s and early 2000s saw rates closer to 1.8 to 2.2 per 100,000 in Victoria. The decline since then is statistically real and is consistent with the broader decline in violent crime across most developed economies in the same period.

The second is that the composition of homicide has shifted. The proportion of homicides occurring in the context of family-violence relationships has remained substantial, and reducing family-violence homicide is one of the explicit ambitions of the post-Royal Commission reforms. The proportion of homicides connected to organised crime has risen and fallen with the visibility of the underlying networks. The proportion of stranger-on-stranger street homicide has remained low — Victoria, like Australia generally, has very little of this category compared with international peers.

The third is that the recent period has seen periods of elevated organised-crime-related violence. Where these run in clusters, the headline figure for the year can move in ways that are real but not representative of the underlying rate of homicide more broadly. The taskforce response — Victoria Police’s Echo, Lunar and other named taskforces over the years — has reflected that.

What does not show up in the data

There are at least three things the headline series does not capture.

The first is near-misses. An attempted murder that succeeds and an attempted murder that fails differ in the data, but in human and policy terms they may be very similar events. Looking at attempted-homicide data alongside completed-homicide data gives a more stable signal, particularly in the family-violence context where serious assault precedes lethal assault in a high proportion of cases.

The second is the long tail. Some homicides are not recognised as such for years — historical disappearances later confirmed as homicide, deaths originally classified as misadventure later reclassified, cold-case investigations producing belated charges. The official series gets revised. The revisions almost never make headlines.

The third is unrecognised homicide entirely. The Coroners Court of Victoria reviews unexplained deaths. A small number of deaths are reclassified each year. The total population of unrecognised homicide is, by definition, not knowable, but most criminologists assume it is non-zero and is concentrated in vulnerable populations — homeless people, people with substance dependence, isolated older people in residential aged care.

How to read the next set of numbers

When the Crime Statistics Agency next publishes its annual release and the headlines arrive, three habits help.

One: look at the rate per 100,000, not the raw count, particularly in growing population centres. Two: look at the five-year trend, not the single-year change. Three: look at the composition — family violence, organised crime, stranger violence — rather than the headline total alone.

The honest summary, as our newsroom reads it, is that Victoria’s homicide rate is low by international standards, has trended down over twenty years, fluctuates within a narrow range from year to year, and contains some real internal patterns that are worth understanding on their own terms. That is a less satisfying story than a single shock headline. It is the more accurate one.

Support contacts

Homicide leaves long shadows. The Victims of Crime Helpline operates on 1800 819 817 and can refer you to dedicated homicide-bereavement services. Lifeline is on 13 11 14 for immediate emotional support. The Coroners Court of Victoria provides ongoing support to families through its Coronial Counselling Service.

If you have information about an unsolved homicide, the line is 1800 333 000. Victoria Police maintains a Cold Case Unit and a published list of homicide rewards.

Jack Renton

Jack Renton covers crime, policing and major incidents for Victoria Crime News. He has reported on organised crime, drug trafficking and major operations across metropolitan Melbourne and the western suburbs. Outside the newsroom he sits on the board of a regional volunteer surf rescue club.

Leave a Reply

Your email address will not be published. Required fields are marked *

Related Articles

Back to top button

Stay informed

Get our weekly Victorian crime & courts brief

A short, independent weekly summary of what's actually moving across Victoria's crime, courts, road-safety and integrity beats — written by our newsroom. No spam. Unsubscribe in one click.

By subscribing you agree to our privacy policy.

Important notice. Victoria Crime News is an independent news and commentary publication. We are not Victoria Police, are not affiliated with Victoria Police, and do not represent the views of Victoria Police, the Victorian Government, or any law-enforcement agency. For official information, statements or operational matters please visit police.vic.gov.au. In an emergency call 000. To report a crime confidentially call Crime Stoppers on 1800 333 000.

About Editorial standards Contact Privacy Disclaimer