March 2001

[Previous] [Contents] [Next]

Paradise Lost & Regained: Crime in Canada and Around the World

by Stephen T. Easton

For Canadians, one of the most striking statistics of the past decade is the dramatic fall in the crime rate. Since 1991, crime in Canada has fallen by 25 percent. It has fallen across the board, from homicide to vandalism (except for motor vehicle theft which is up about 6 percent.) There are many possible explanations for such a decline, one of which will be discussed below, and one of which will be discussed in the article "The Parent Trap," by Chris Schlegel, later in this issue. More broadly, though, the purpose of this article is to remind us of the magnitude of the changes that are upon us.

Crime in Canada

Crime rates in Canada are displayed in figure 1, which chronicles Canadian crime patterns since 1962, the first year in which the "new" measuring system for crime was in place. This measurement is of crimes that are "known to the police." Prior to 1962, crime rate data were based on convictions for particular crimes. In the figure, the rates are per 100,000 of the general population to ensure that simple increases or decreases in the population do not lead to the perception that crime is increasing or decreasing solely on the basis of changes in the number of people.

From 1962 to 1980 there were undulations, but the general tendency in each of the series was determinedly upward. The 1980s saw increasing crime rates, but the fluctuations were less clearly upward until the mid to late 1980s when there was a great leap. The upward rise in crime reached a peak in 1991. This peak is not uniform as many of the sub-series that constitute the crime rate reach their maximum values just before or just after that year, but 1991 marks a sea change in many of the categories of crime.

From 1991 to 1999 the crime rate in Canada has fallen from a peak of 10,342 crimes (per 100,000) to 7,733. These are substantial numbers. The last time we observed a crime rate of 7,700 (per 100,000) was in 1979! To put these numbers in perspective, if crime were to be evenly spread across the entire Canadian population, then each individual would have had just less than one chance in 10 to be a victim of a crime in 1991. That probability fell to one chance in 13 by 1999.

The surprise in the numbers is related both to the change in direction, from a generally rising crime rate for the previous 40 years to the falling rate now, and to the sharpness of the decline. It took 12 years for crime rates to peak, but only 8 years for them to fall back to the same levels.

The fall in the crime rate takes place in all areas: violent crime, property crime, and other crime. Property crime has fallen more quickly than violent crime. In 1981, there were about 8.8 property crimes for every violent crime; by 1999, that proportion had fallen to 4.5 property crimes for every violent crime.

Lest we think that we are unique in our good fortune, we can look south for even more dramatic evidence of declining crime rates. In the United States, as in Canada, the crime rate most recently peaked in 1991. Since then, the US crime index has declined by even more than ours: measuring the 1991 peak in both countries as 100, the US crime index has declined to 72, while Canada’s has fallen to 75.

Around the world

Around the world, although many developed countries are seeing falling rates of crime, generally the fall is not as quick as it has been in the US and Canada. In a sample of 27 countries¹ (those for which data were available), between 1988 and 1998, the United States had the most rapidly falling crime rate, and Canada’s was nearly as fast. The percentage change in the crime rates over the decade are displayed in the first column of figures in table 1. In the table, the countries are ordered by the decline in their crime rates since 1988. Since the series of crime rates for each country is variable, the second column, "Fall from the peak crime rate," gives the fall in the crime rate since the highest level of crime was reached. In those countries in the table for which there is a zero in the cell, that simply means that the peak crime rate was reached (thus far) in 1998.

A look down the "fall from peak" column shows that crime rates in the UK and Ireland have also fallen very rapidly from their peaks. Together with the US and Canada, they are the countries whose crime rates have fallen the most. Most of the rates in the other countries in the sample have fallen, but not by nearly as much.

The fall in crime in England and Wales illustrates the point. Figure 2 shows three series: theft of and from motor vehicles, burglary from dwellings, and violent crime. In each case, the series have begun to fall, and in each case they look remarkably similar to the pattern seen in Canada.

Incarceration and crime during the past decade

There is no commonly accepted explanation of the data on falling crime rates. We do not have any better understanding of why crime rates are declining now than why they were increasing previously. There are a number of theories, but none has been compellingly successful in explaining the data across countries and time.

This article will now focus on observations about rates of crime and rates of incarceration. Incarceration does two things to the crime rate. First, it removes proven offenders from the street. Second, it provides some element of deterrence, both for those who would re-offend and for those primed to offend who witness the outcome of offending. The effect of incapacitation and example on the potential criminal’s margins of choice is to reduce the crime rate.

Critics argue that prisons make offenders more likely to offend in the future and that the deterrence feature is minimal. Column 3 in table 1 may provide some evidence. This column reports the percentage change in the rate (per 100,000 of the general population) of incarceration. The numbers reveal several things. First, of the countries that have declining crime rates over the 1988-98 period, all but Denmark show an increase in incarceration over the same decade. The UK does not fit this exact pattern, but it has seen a very sharp decline in crime since 1991, and also has had, as column 3 shows, one of the larger increases in the rate of incarceration.

But the table reveals many anomalies with this analysis as well. Several countries have higher rates of incarceration and higher crime rates. Clearly, a more subtle assessment is required for a complete explanation for the changing pattern of crime.

But what is the simple relationship between the crime rate and the rate of incarceration? To answer this we must ask, what is the impact of changes in the rate of incarceration on the crime rate? To get a clear answer, we need to account for crime rates that measure different things in different countries, and rates of incarceration that are also measured differently across countries. Once we make these adjustments, an association emerges between the changes in the crime rate and changes in the incarceration rate. In fact, preliminary statistical evidence suggests that a 10 percentage point change in the incarceration rate leads to a 7.5 percentage point reduction in the crime rate.²

Although there is much more to say about this issue, it is interesting that the international data tend to support the negative effect of incarceration on the crime rate. But because incarceration only explains 13 percent of the variation in crime rates, this suggests that there is plenty of room left for additional explanations.³

Notes

¹Austria, Australia, Belgium, Canada, Czech Republic, Denmark, France, Finland, Greece, Germany, Hungary, Ireland, Italy, Japan, Luxembourg, Netherlands, New Zealand, Norway, Poland, Portugal, Russia, South Africa, Spain, Sweden, Switzerland, the United Kingdom, and the United States.

²Easton and Brantingham (2001), The Cost of Crime: Who Pays and How Much? (forthcoming).

³This is not as negative a result as it may appear at first blush. The estimation rates change, not the levels, and link the rate of change of crime to both the rate of change of incarceration and the level of both incarceration and crime. Consequently we would anticipate a far lower explained variation that is usually seen in simple levels regressions.

Stephen Easton, Ph.D., is Professor of Economics at Simon Fraser University and Senior Fellow at The Fraser Institute in Vancouver.

[Previous] [Contents] [Next]

info@fraserinstitute.ca
4th Floor, 1770 Burrard Street, Vancouver, BC, Canada, V6J 3G7
Tel: (604) 688-0221 Fax: (604) 688-8539 Book Orders: 1-800-665-3558 ext. 580

You can contact us at the above email address for any comments or information requests. Please report any dead links or technical problems.