Does Gun Control Reduce Crime? Looking at the Statistics

After Newtown, Connecticut

In the wake of the 2012 Sandy Hook Elementary tragedy, there was been tremendous debate in the United States about increasing gun control. It took only hours to find hundreds or thousands of forum conversations, calls from our politicians for more gun control and a general demand that something, anything, be done to prevent such an occurrence in the future.

Many of the forums were simply cries to "Eliminate gun ownership," "Get rid of assault weapons" (whatever an assault weapon is, no one seems able to give a definition), or "Over my dead body!" None of which are particularly productive—all the old arguments, like "Guns don't kill people, people kill people" were raised again (with predictable results) but there were precious few actual facts to be found.

The debates have brought out the question of will gun control actually prevent, or even help prevent, homicides or mass killings, though. No one seems to know, or even care—we just have to do something—so I went looking for answers and preferably answers from someone without an axe to grind.

I've seen the opinions and looked at some options and controls but wanted some hard data to make a decision with—something beyond a simple opinion based on fear or dismay at what happened that sad day in Newtown.

To make it very clear, I am a gun owner. I don't carry a concealed weapon for self-defense, I don't hunt, and very rarely even shoot my guns, so the loss of them would affect me very little. On the other hand, I am concerned about any loss of personal freedom, and feel that the government needs to tread very carefully there; that it is important to keep what freedoms we have.

The days after the Sandy Hook incident were sad ones for me as I watched my grandchildren in their school and feared for their lives. I want a solution to the killings and murders in America, but I very much want one that will work, not just a sop to soothe our conscience until the next time it happens.

Here are the statistics and data that I found. It is presented in graph form for easier examination, but the hard numbers are also given at the end of this article. Make up your own mind about the effectiveness of gun control, just as I have.

Gun Ownership vs. Gun Homicides

I found lots of numbers being bandied about concerning gun-related homicides so let's start there. Reliability of data is always a primary concern and the information contained in the graph above comes from UNODC, the United Nations Office on Drugs and Crime and from Small Arms Survey (link removed, see new link in update below), a highly respected source that the UN often uses itself.

Data was chosen from 42 countries out of the 170+ that were available, and is primarily from 2007 as the most recent year with data from most countries. Data was not always available in or near that year for the areas needed, and a few of the data points come from the closest year to 2007 where it was available.

Middle East countries were rejected in total for lack of data, as well as for not being representative culturally of industrialized nations and for being very violence-prone.

The bar graph has a few odd "spikes" in it that are almost certainly the result of factors other than gun ownership.

Mexico, for instance, has a very large problem with drug lords and in some areas is almost a military operation. South Africa and Brazil were both deleted from this graph because of enormous homicide rates that clearly have to little to do with the question of gun control, although that data is available in the charts at the bottom of the article.

Looking at this graph, there is no firm equation between gun ownership and gun homicide rates, but we can see a rough correlation. In general, as gun ownership rises, so does the gun homicide rate. More guns means more gun deaths. There are some problems with making the connection, though, that need to be carefully considered.

There are those odd spikes out of nowhere that should probably be disregarded as being caused by some factor other than gun ownership rates. We can see a correlation, but that is not indicative of cause.

As an example of this type of statistics problem, speed limits on American highways were reduced in the 1970s to 55 MPH everywhere in the country. Not surprisingly, car deaths went down right with the speed limit, but when the limit was raised again a few years later they did not climb back to where they were.

Can we then say that higher speed limits don't have a major effect on car related deaths? No, because in those intervening years seat belts, airbags, antilock brakes and other safety improvements all grew in popularity and were an additional, primary, factor in preventing deaths.

In just that manner, the deaths from the drug war in Mexico will skew the figures from that country; extraneous causes must always be looked for and considered.

In addition, even though we find a correlation between the number of guns and the number of gun deaths, that correlation is worthless in deciding whether or not reducing the number of guns will save lives. We need to look at the total, overall homicide rate instead. The premise is that if guns are not available, the killers will use another tool, but is it true?

Scroll to Continue

What is a "LOWES #00907 - 866-483-7521 NC" Charge on My Credit Card Statement?

Should I Watch..? 'Sonic The Hedgehog 3' (2024)

A List of Dams in Punjab

Gun Ownership vs. General Homicide Rate

In this graph, we find some of the same huge spikes, but what we can't find is much correlation between gun ownership rates and overall homicide rates.

Yes, as we move from the left (low ownership) to right (increasing ownership) the rate of homicide rises. But then it drops back to near zero, rises again, drops once more and finally rises slightly near the end with a blip (the United States) as the last point on the graph.

If we disregard those tall single spikes (that are scattered throughout the graph) as anomalies with a different cause, we have nearly a straight line with little variance. It wavers a bit up and down, but not significantly so.

So this graph cannot really show that gun control will affect the homicide rate. What about one particular country that instituted strong controls? One that we have data for both before and after the controls were instituted? Australia is such a country; let's look at that.

Homicide Rates in Australia, Over Time

The data for this graph was compiled from the same source as earlier graphs, UNODC, as well as Wikipedia. The information from the UNODC only went back to 1995 so the years from 1990–1994 were taken from Wikipedia. It's worth noting, though, that Wikipedia agreed with the UNODC data for later years, so there is no real reason to disbelieve it for the earlier years.

Australia instituted strong gun controls in 1996, making it one of the most restrictive nations in the world. Did it help?

Prior to 1990, the homicide rate in Australia was in a very gradual decline, and after that date, basically continued the same decline as the graph shows. Yes, there was a downward blip in 1998, followed by an even greater upwards blip in 1999. The graph pretty much levels out after that for several years, once more starting a gradual decline in about 2003.

This conclusion is born out by data from the Australian government as well. Graphs of homicide victims (observed by author earlier but link removed) show the same gradual decrease in the number of victims with no significant change in or shortly after 1996.

Nor is this conclusion unheard of. Joyce Lee Malcom, professor of law at George Mason University, has looked at both the UK and Australian experience with very similar conclusions. Her report in the Wall Street Journal on the effectiveness of strong gun controls is instructive.

It seems unlikely, then, that strong gun controls had much effect on the homicide rate over the years in Australia. Other factors are most likely causing the slow decline seen over decades.

But if the homicide rate didn't fall as a result of the new law, what about mass murders like Sandy Hook? Well, Wikipedia gives a list of mass murders, or massacres, in Australia. It shows that since the new law, there have not only been more such incidents, but more people are being killed as well.

When guns were not available the killers turned to matches; the arson rate went up, with more people being killed per incident. It would seem that taking guns away, in this case, resulted in more deaths, not fewer, even though the number of deaths by gun fell.

So Are Strong Gun Controls Effective?

While the first graph does show that decreasing the number of guns corresponds to a decreasing number of gun-related homicides, that's a no-brainer. Take away the guns and killers won't kill with guns.

Far more interesting is that we find no correlation between gun ownership rates and general homicide rates. Take away the guns and killers still kill; they just don't use guns to do so. This is true whether looking at a variety of countries with a variety of gun ownership rates or at a single country that instituted strong controls.

You will make your own decision as to the effectiveness of gun controls in preventing homicides, but it appears plain to me that it does not do so. If we want to avoid another Newtown incident we need to spend our time and resources looking elsewhere, because it wasn't caused by guns and removing guns won't help. Given that, there is no reason to further infringe on citizens' rights to own weapons.

We might look hard at how we view and treat mental illness, we might look at stopping gang activities or drug-related violence. We can put some thought into the violence our kids see on TV or in life; about how to change the violence our culture produces. There is something wrong in America, but it isn't coming from guns. They are only tools being used to express that violence.

Is the above video of somber actors and actresses in black, promoting gun control and earning a living from gratuitous violence onscreen, a part of the problem?

We pay lip service, but will not consider changing our culture to one less infatuated with violence, less filled with sights and sounds of killing every day. I can't answer definitively, but it certainly sounds reasonable to me. We glorify violence and killing while wondering why we see so much of it.

Data in Numeric Form

Update 2022

It has been 10 years since this article was written, and it seems time to update the data and take another look. We've had 10 years to see if anything has changed—we've changed some laws, and we've watched as other countries have struggled with their own growing problem of mass murders and murders in general.

I used the same sources for data that were used in 2012; UNODC and the Small Arms Survey. In addition, while the data from the Australian government that was used in that section of the article is no longer available, other studies from that government are.

The homicide rates, in deaths per 100,000 people in the graph below, comes from UNODC, which in turn collected it from several sources. Most of the numbers are from 2019 and 2020, although a handful are older.

Gun ownership rates, in guns per 100 people, come from the same small arms survey as used before, except from 2017 and 2018. The information from the Australian government is from 1993 to 2019.

I did not try to compare gun ownership to homicides by gun; it was and remains apparent that if we remove guns, killers will not kill with guns. Granted, this is only true to some extent as neither U.S. law enforcement nor that of any other country is able to remove all guns from their society; some criminals (and killers) will always have guns and we will always have some gun violence.

Below is the graph of gun ownership rates for a wide range of countries, and the homicide rates for those countries. As before, I tried to choose countries similar to the United States and that did not have a war or some other reason for large numbers of homicides.

A few of those countries in the data list below were not included as being anomalies with other reasons for high murder rates. In addition, a couple from the first list 10 years ago did not have current numbers and were also left out.

Once more, just as in the past, there is no significant correlation between gun ownership and homicide rates. As before, there are some spikes but in general more guns in a society does not mean a higher homicide rate. There may or may not be more people killed by guns, but there are not more people killed by all means.

I submit that the only thing that really matters is that people, whether one person or 50, have died as a result of violence in our country—whether they have bullet holes in them or not is irrelevant. Neither the dead nor the survivors care whether the bodies have bullet holes or not.

The Australian Experience

As mentioned, the data used 10 years ago has been buried too deep in Australian archives for me to find, but another study is available. The graph below is reproduced from an Australian government study.

This time around the data from Australia differs somewhat from what I found in 2012, with one of the big differences being that it is a record of "homicides and related offenses," including manslaughter, murder, etc.

Keeping in mind that the gun legislation implemented there was in 1996 and it was a buyback of all semi-automatic weapons, the number of homicides in Australia not only didn't fall after that date—it went up!

Those guns were confiscated in the first year of the law; it took just one year to get them out of circulation, but it was 2004 before the number of homicides fell to what it was in 1995, the year before the law went into effect—10 years of additional work on crime before the death toll fell back to what it was before the law was passed.

It seems rather obvious that the law had little to no effect, unless you want to count the increase in homicides in the nine years after it, as an indication that taking guns away will result in more, not fewer, murders.

The bottom line then is that there is still no correlation between gun ownership and homicide rates. Without a correlation we cannot even say that more guns means more deaths, let alone say that having more guns will cause more deaths.

Yet we as a nation continue down the same path, always and forever trying to disarm Americans because it is "common sense" that more guns does equal more deaths. Facts be damned; if they don't support the popular mantra, then we'll ignore them.

We'll go so far as to exaggerate dangers, insinuate that scary looking guns that we also (falsely) insinuate are used by the military, are more dangerous than other weapons. We'll apply labels that have nothing to do with reality ("assault" rifles) in order to frighten a public generally ignorant about guns.

All with every experience and every bit of data showing that more guns does not equal more deaths. Is there a clearer indication that the goal is not to prevent deaths but instead to disarm the American public? Will we never actually address the problem of violence in America, instead wasting our resources on a pretend solution that we know will not produce the result we claim we want?

2022 Data

Comments

Katharine L Sparrow from Massachusetts, USA on July 12, 2018:

Thank-you, Dan! You are absolutely right to base opinions on facts and figures rather than emotion, which most gun control advocates do. Guns are not the problem. In past decades, there were many more guns per household than today. Most families had at least one gun, it was like having a rake, it was a tool you just had. Yet mass shootings were almost unheard of until recent years. What was the difference? It wasn't the availability of guns, so it is something else. In my opinion we need to spend more time thinking on that than about banning weapons. Great hub with tons of info! Congratulations!

Dan Harmon (author) from Boise, Idaho on April 20, 2018:

Brad

Mexico was not included in the discussion here (although the stats are given) because of that high homicide rate (half again ours). It seemed to me, and still does, that a great deal of that rate is the almost uncontrolled activity of drug lords. It would be like including Syria, knowing there is war going on, and putting their violent death stats against those of the US.

Any time there is a big anomaly in the statistical data a reason must be searched for or the data discarded. In this case I felt the anomaly was due to an active drug war and did not use the data in drawing a conclusion. Of course the US is a massive anomaly in itself, but if I discard that the whole purpose of the study disappears!

If you follow the forums here, I have repeatedly offered that, given a country with low gun ownership and low homicide rate (or the opposite) I would provide two more with lower guns and higher homicides (or the opposite). That this can be done (or the opposite) plainly says there is no correlation between gun ownership rates and homicide rates. At least it does to me; others have said that it can't be true because everyone KNOWS that taking guns away will save lives.

An interesting point along these lines is that in looking at the US homicide rate through history, there was a huge spike for a short time in the 20's. At almost exactly the period that prohibition was in effect. And another that that coincides almost exactly with our "war on drugs"; as marijuana laws relax the homicide rate has fallen. Now, two data points are not enough to prove causality, but it surely looks like if we take something away that people want enough, they will kill to get it.

Brad on April 20, 2018:

Dan

I thought that the stat on Mexico was interesting.

With 17% of the gun ownership of the US

They have gun homicide almost the same as our. 3.7 to 3.8

and a general homicide rate of 8.1 to our 5.7.

Less gun owners, but almost the same gun homicide rate?

FYI

Mexico’s gun controls are strict and, when enumerated, read like a wish list for U.S. Senate Democrats. Think about it: For a Mexican citizen to legally acquire a gun, he or she must obtain a license, a process which requires them to pass a background check. That background check looks at criminal history, mental history, physical health and any past drug additions.

Making the background check even more onerous, CBS News reports, is the requirement that applicants submit six pieces of documentation: A birth certificate, a letter confirming employment, proof of a clean criminal record from the attorney general’s office in the applicant’s home state, a utility bill with current address, a copy of a government-issued ID and a federal social security number.

Brad on April 20, 2018:

Dan

Agreed

and the I in FBI seems today to mean incompetence.and not Investigation. When you don't even satisfy your name FBI then that should be at least one of the problems.

I doubt if you are going to hear from Mike or ME on this article.

Dan Harmon (author) from Boise, Idaho on April 20, 2018:

Bradmaster

We can agree that there were numerous failures at Parkland. It is my hope that they were errors and not conscious, intentional actions taken to increase the death toll. Though I will sat that deputies standing outside with shots ringing out makes that hard to accept even as we saw much the same in another recent shooting when it took far too long to break in. The fear was that it would cause additional loss of life, but I can't imagine how that could be a concern at the Parkland school.

But even if that is true, it is obvious that there was a great deal of political action that was more in line with money and politics than with safety, and you are right in that no one cares. It may or may not be actively concealed, but it surely is not given the attention it deserves and requires...that attention is lost in the cries for more gun controls.

Brad on April 20, 2018:

Dan

The most important point was that student lives were lost while sheriff deputies listened to the shots that Cruz fired inside the school, but they stayed outside.

While they didn't go inside the school lives were lost.

This was a big law enforcement failure, yet it wasn't an issue for the gun control people.

Second

The sheriff had a deal with the Parkland school district to not file criminal charges against the students, as the school would lose federal money offered by Obama.

Cruz had numerous incidents that were criminal but not filed.

The police had numerous complaints about Cruz, even from his mother before she died.

After the shooting the FBI said we are not sure if this was the same Cruz. They would have known if they would have followed up the leads they were given.

That to me is something unlike gun control that is straight forward to address and fix. Gun control because it already exists, and fails cannot be directly assessed as to the value in lives it may save by adding to it.

In the case of Broward their sheriff failed to protect or serve, and yet no one seemed to care!

Dan Harmon (author) from Boise, Idaho on April 20, 2018:

Yes, I got that, and I pretty much agree with it. I'm not convinced that we had enough evidence on the killer, before he took action, to arrest him, but he certainly should have been under near constant surveillance.

That's another problem - what we need to arrest someone, and when should we do it? Neighbor comments that "He's weird" or even really questionable facebook posts without an actual threat, are not a crime and we cannot arrest based on them. No grand jury would ever indict on that slim evidence, as is proper, that a person MIGHT misbehave in the future. But law enforcement can certainly keep watch. Whether we have the manpower to mount a 24 hour surveillance of every suspect is questionable, so priorities will have to be made, and the public will always second guess those priorities. In this case it appears that the FBI dropped the ball entirely, but in others they won't and may STILL be wrong and putting their resources in the wrong place.

And of course the deputies waiting outside until the shooting stopped in inexcusable. I understand their lives and safety are important, but their job is to risk both when necessary, and an active killer shooting inside an occupied building certainly makes that necessary.

Brad on April 20, 2018:

Dan

I am sorry to be persistent, but I think it is an issue bigger than gun control.

I have copied what I am talking about.

"IIf the existing gun control is not sufficient to solve the problem, then maybe that is the evidence that more gun control is not the answer. And the answer may be the person.

In the Parkland shooting, the government already had the necessary info, but didn't follow it. In addition, they already had an armed sheriff deputy assigned to the school, but while he and his other 3 deputies could have saved lives by confronting the shooting. They were in safety outside the school while they listened to the deadly shots being fired by Cruz inside.

Yet, the gun control people don't even think that is an issue.

Why, because that doesn't help their political agenda."

Dan Harmon (author) from Boise, Idaho on April 20, 2018:

Bradmaster:

I fully agree that (most) of those laws desperately need better enforcement. Those convicted of violent crimes that still carry a gun, for instance. Better background checks, and on all sales. Better record-keeping.

As you say, we have plenty of laws to do the controlling we need to - we just don't enforce them well enough.

Brad on April 20, 2018:

Dan

What about the law enforcement aspect that I mentioned in my previous comment?

Dan Harmon (author) from Boise, Idaho on April 20, 2018:

bradmaster: First, I would never delete a comment just because I disagree with it.

But I do think these stats are valid even though taken worldwide rather than just the US. People are people everywhere, and I did make an effort to choose countries closest to ours in culture. These stats also point out that old adage that "Guns don't kill people; people kill people" is too true, and it is the same conclusion that you come to.

You blame it on politics (correctly, IMO) but that does not invalidate the concept. Using politics to ban the most popular gun in the country, that is used to murder less than 3% of the people killed with violence, can be nothing but political activism and cannot possibly have anything to do with reality.

This is a fight that will eventually be won; the American public will be disarmed one day. The loud, ignorant, voices that are using fear mongering rather than reality as their tool will win, for fear will always be a more potent weapon than reason. You are entirely correct in that we have enough laws already on the books to provide the controls that are needed, but those loud voices aren't concerned with control - their goal will not be met until there are no guns to confiscate.

We see politics being used in their fight, we see court attempts to force manufacturers to stop making guns of any kind and we now see the massive financial clout of Big Business being brought into play to stop manufacturing. The fight will not end until we lose our rights completely, but it has nothing to do with reality or reason.