A Statistical Comparison Of Homicide Rates In The Prairie Provinces And Four American Border States, 1978 - 1992

By L.G. Morrison, M.P.

January 1995


The report "A Statistical Comparison of Homicide Rates in the Prairie Provinces and Three American Border States" was released in October, 1994.

In response to criticism that the three states contained no large cities, Minnesota was added to the mix. The twin cities of Minneapolis and St. Paul have a combined population comparable to that of Calgary, Edmonton or Winnipeg.

The general conclusion of the original study, that homicide rates in the northwest border states are, and have been for many years, significantly lower than in the neighbouring prairie provinces was reinforced by the addition of the fourth state.



Summary and Conclusions

Observations and Discussion

An International Perspective

United Kingdom


Graphical Representations of Recent Rates of Homicide


Fig. 1 Northern Tier Western States, Prairie Provinces and Canada

Fig. 2 Manitoba, Minnesota and North Dakota

Fig. 3 Montana, Saskatchewan and North Dakota

Fig. 4 Montana, Alberta and Idaho

Fig. 5 Minnesota, North Dakota, Montana and Idaho


People living on the prairies of western Canada have long observed that rates of violent crime in neighbouring American states are not significantly different than in their own communities. Since the western border states permit almost unimpeded possession and use of firearms, this contradicts the assumption that crime increases where guns are most readily available.

In order to test the truth of this casual observation, homicide statistics between 1978 and 1992 were compared for the three prairie provinces and four northern tier states.


Of eight jurisdictions (four states, three provinces and Canada as a whole), Montana had the most homicides per capita over the fifteen year period, with an average of 3.8 per 100,000 citizens. Manitoba was second highest at 3.6, followed by Idaho at 3.4, Saskatchewan and Alberta, each at 3.1, Canada at 2.7, Minnesota at 2.4 and North Dakota at 1.3.

The first seven averages are in the anticipated range, but the rate for North Dakota is one of the lowest in the world despite an abundance of guns in the hands of its citizens.

The 15 year per capita homicide rate for the three provinces combined was 3.2 per 100,000 compared to 2.7 per 100,000 in the four states.

Since 1978, homicide rates have been decreasing in seven of the jurisdictions, most notably in Idaho, but there has been a steady increase in Minnesota (Fig. 5). Because Minnesota has a population greater than the other three states combined, this is reflected as an increase for the four states (Fig. 1), for which the 15 year average of 2.7 equalled that of Canada as a whole.

No detailed evaluation was made of other jurisdictions but reference is made to American states with gun controls in place and to other countries with firearms regulations varying from stringent (Japan) to extremely open (Switzerland).

It was concluded that the regulation of possession of personal arms by private citizens has little or no effect on homicide rates.

If this conclusion is valid, it follows that murder has cultural and sociological roots and that the imposition of restrictive regulations on the general population to control it is a futile exercise.


There is a common misconception in Canada that there are no gun controls in the United States. Federal law, except for the recently passed Brady Bill and control of interstate and international trade in firearms, remains constrained by the Second Amendment to the Constitution. However, several states have enacted their own gun control laws, and there is a plethora of local ordinances in those states which permit them.

The gun laws of New York, and especially of New York City are stringent by Canadian standards, and Canadian firearms legislation appears to have been influenced by the older New York statutes, of which the keystone, the Sullivan Law barring the carrying of deadly weapons, dates from 1911.

A permit is required to purchase a handgun anywhere in New York State, and applicants undergo a rigorous screening process which may take up to six months. In New York City, permits are also needed to purchase long guns for which the waiting period is 30 days.

In spite of the restrictions, the state homicide rate is, on average, five times higher than in Canada. In New York City it is ten times the Canadian rate. There is no provable explanation for this anomaly, but it is reasonable to suppose that the presence of organized crime, a flourishing narcotics trade, racial tension, extreme poverty and a collapsing public education system are all contributing factors.

In the District of Columbia, the sale of handguns is prohibited, permits are required to purchase rifles and shotguns, all firearms must be registered and owners must have possession permits. No other jurisdiction in the free world has a more rigid system, yet the homicide rate has been rising for several years and in 1991 reached an astonishing 80 per 100,000 citizens - probably the highest for any jurisdiction in an industrialized nation.

In the United States, the degree of control of firearms is directly proportional to the amount of violence in a particular jurisdiction. Thus Illinois, (especially Chicago) and Michigan are quite restrictive, whereas several western states with relatively peaceful societies are effectively "wide open".

Minnesota, North Dakota, Montana and Idaho are mostly rural and are economically, socially and demographically very similar to western Canada. They were chosen for a comparative evaluation partly for that reason and partly for their minimal legislation relating to firearms.

All four states require permits to carry concealed weapons. North Dakota bars machine guns and fully automatic rifles and Montana permits possession of machine guns only on the owner's premises. Minnesota has a seven day waiting period for the purchase of a handgun, and permits are required to carry them even if not concealed. The age of responsibility for unsupervised use of a rifle or shotgun is 14 years in Montana and 15 in the other states. Beyond that, the only controls are practical, local ordinances with respect to being armed at a public gathering, discharging a firearm within town limits and so on.

Control of long guns was introduced in Canada in 1978. To illustrate the effect of controls, annual homicide rates from 1975 to 1992 in Alberta, Saskatchewan and Manitoba were plotted as line graphs. Rates for Canada, Minnesota, North Dakota, Montana and Idaho were superimposed for comparison. (Fig. 2, 3 and 4)

Because Manitoba, Saskatchewan and the three most westerly American states have small populations, relatively small changes in the number of homicides cause very erratic variations in the rate per 100,000. In Minnesota, with a population of more than 4 million, annual variations are minimal.

The plotted curves were computer averaged to establish long-term trends. It was observed that homicide rates are decreasing in all jurisdictions except Minnesota. A dramatic decrease in Idaho (Fig. 4 and 5) reflects high rates (5.4 per 100,000) in 1978 and 1979, and a very low rate (1.8) in 1991. However, even if these three erratic values are rejected, the trend remains sharply downward.

Of the individual jurisdictions, Montana had the most homicides per capita over the fifteen year period with an average of 3.8 per 100,000 citizens. Manitoba was second highest at 3.6, followed by Idaho at 3.4, Saskatchewan and Alberta, each at 3.1, Canada at 2.7, Minnesota at 2.4 and North Dakota at only 1.3.

The consistently low rate for North Dakota is approximately the same as in Japan, where there is virtually no private ownership of firearms. Among the four states and the three provinces studied, North Dakota is the most rural. It has a slowly declining and presumably aging population, few of the extractive industries that attract unattached young men to Montana and Alberta, and no large cities. The majority of North Dakotans have firearms in their homes.

To nullify the erratic effects of sampling from small populations, the four American states were treated as one single entity, and the three prairie provinces as another. Rates for the two composites were plotted against the values for all of Canada. (Fig. 1)

It was observed that, since the introduction of gun control in Canada, there have been, on average, more murders per capita per year in the prairie provinces (3.2/100,000) than in the four northern tier western states, which had an average of 2.7 per 100,000 - the same as the average for all of Canada during the same period. The rate for the four states combined has been slowly rising. The trend for the prairie provinces and for Canada has been falling.

The foregoing illustrates the absence of a simple cause and effect relationship between crime rates and restrictions on possession of firearms on the civilian population.

No detailed studies have been done by the writer with respect to countries outside of North America, but the following comments, based on readily available public information, are offered as supporting evidence that homicide is a societal problem, unrelated, or at least only marginally related, to public access to firearms.


Switzerland is the country most often cited by opponents of gun control as proof that widespread possession of weapons does not lead to unrestrained criminality. To the contrary, in Switzerland, the knowledge that any household or place of business may contain a firearm and an owner skilled in its use probably restrains criminal activity.

At the municipal level, 15 of the 26 Swiss cantons exercise limited control, mostly with respect to handguns. However, by Canadian standards, the rules are quite permissive throughout the country.

Although the possession of long guns is not restricted, hunting laws are strict, with provision for competence tests, insurance for liability and technical certification of hunters' firearms.

The important statistics are that, among a population of 7 million, there are 2 million firearms, including 600,000 fully automatic military rifles stored in the homes of militia members.

In spite of this massive amount of firepower, the rate of homicides and attempted homicides with firearms in 1991 and 1992 was only 1.4 per 100,000. Total homicide rates for the same period were 2.6 in Switzerland and 2.7 in Canada.

Japan is the model most often used by advocates of the prohibition of firearms. There is no private ownership of handguns, and among 120 million people there are only a little more than half a million privately owned long guns, including air rifles.

Japan's annual homicide rate has been progressively decreasing for a decade and now stands at 1.2 per 100,000. It is reported that 97% of murders are solved - the highest clearance rate in the world.

Japan is one of the most disciplined nations on earth, with an authoritarian and conformist culture that precludes large scale law-breaking. There are few constraints on police powers, especially with respect to search and seizure. Rates of crimes not usually associated with firearms - rape, mugging and assault, are the lowest in the world and are trifling by European and North American standards. Japanese do not kill each other in large numbers because they are, in all respects, extremely law-abiding people. Interestingly, the current Japanese suicide rate of 21 per 100,000 is double the Canadian rate and almost double the rate in the United States.

Norway, like Switzerland, is a small heavily-armed country. It is estimated that 35% of Norwegian homes contain firearms, but homicide rates in 1990 and 1991 were 1.2 and 1.5 per 100,000 - comparable to those in Japan, North Dakota and England.

In the United Kingdom, legal possession of firearms is mostly restricted to the military, police and the gentry. The issuance and renewal of permits to possess firearms is at the discretion of local police, but a refusal may be appealed in court.

Private ownership of semi-automatic rifles and shotguns is prohibited.

Anecdotal information suggests that illegal arms, particularly handguns, are readily available on the black market.

Published homicide rates for the U.K. vary widely, but data obtained from the British High Commission shows a 1992 rate of 1.3 per 100,000 in England and Wales only.


Bartlett, W.C.:Gun Control Legislation in Canada, the United Kingdom and the United States, Library of Parliament, BP-233E (Feb. 1990).

Begin, Patricia:Violent Crime in Canada and Switzerland, Library of Parliament (July 20, 1994)

Government of Japan, National Police Agency: White Paper on Police, (1984 and 1991)

Leiter, Richard A.:National Survey of State Laws, Gale Research Inc., Detroit (1993)

Statistics Canada:Number and Rate of Homicide Offenses, Canada and the Provinces/Territories. (1961-1992)

U.S. Department of Commerce:Statistical Abstract of the United States (1984-93)

U.S. Department of Justice - U.S. Federal Bureau of Investigation: Crime in the United States, Uniform Crime Reports (1978-92)