Have the wheels fallen off traffic enforcement?

By: Lawrence P. Lonero

Date: 1999-09-09

Larry Lonero is an expert on crash causes and a principal of Northport Associates, a consulting company based in Ontario, Canada.

Abstract/Summary

A recent multi-fatality crash in Ontario, Canada has temporarily penetrated public and media complacency about road safety. Given the concern about driver actions and government programs, it seems timely to probe for some answers to tough questions about what has been happening to traffic enforcement in this and other jurisdictions. In Ontario, the number of traffic convictions dropped by about half in a decade, while the numbers of drivers, vehicles, kilometers traveled, and reported collisions all increased. The crash rate per kilometer stayed constant, and fatalities declined during this time. Possible reasons for these surprising observations are discussed.

If we are concerned about driver behaviour, some tough questions about enforcement, past and future, have to be answered. What accounts for the drop in convictions? What have been the costs in lives and dollars of reduced enforcement? Is the drop in enforcement unique to Ontario, or has it been going on in other jurisdictions as well?

Our earlier comprehensive review of the evidence on what influences driver behaviour makes it clear that a visible, credible enforcement presence is a critical piece of the puzzle of safer roads. Positive influence of driver behaviour requires both "the carrot and the stick," and enforcement threat is the sharp end of the stick. Research has shown that the chances of getting caught are much more important than the size of the penalty in shaping driver behaviour.

Introduction: safety news and politics

As reflected in recent Ontario experience, highway carnage occasionally focuses the attention of the public, news media, and politicians. Most media consumers are drivers and exposed to real risk on the roads, and the personal connection makes newsworthy crashes seem even more fascinating than other sources of death, mayhem, and disaster. Indeed, research has verified a very strong public interest in crashes and related stories in the media. Recent media interest in Ontario has surrounded a neglected and supposedly lethal "killer strip" section of the Queen's Highway 401, which is the main east-west freeway across the southern heartland of the province.

Just as the killer road issue seemed about to fade, a foggy morning in the same region produced a massive, 80 vehicle chain-reaction pileup with the tragic loss of seven lives. (See another Drivers.com article entitled " Killer roads, let's look closer ".) Some victims were burned alive in the most horrific manner imaginable. Predictably, media attention shifted to the chain-reaction crash. Police and politicians have, so far, been unable to ascertain what happened, but they expressed certainty that fog and bad, aggressive driving were involved. The first is obvious, the second predictable, since aggressive drivers have taken over from drunk drivers as road safety's favorite bad guys.

What's been going on out there?

Given the intensity of interest and the flood of questions and opinions, it struck me as odd that nobody in the media asked any tough questions. What is really going on the freeways? How about the numbers, not just vague impressions, and longer-term changes, not just one year over the last. What really are the trends in traffic, crashes, driver actions, speeds, and enforcement? What's the trend in average speed on the freeways? Sometimes these simple questions can produce a surprise. But even in the recent media pressure cooker, nobody seems to have thought to question the police or the government about the details of what's been happening out there over the years.

It is possible to get an idea of what has been happening, even for the media and other non-bureaucrat outsiders. Since 1985, Ministry of Transportation's comprehensive Ontario Road Safety Annual Report (ORSAR) includes crash and other statistics, such all the convictions registered against Ontario drivers' records. The most recent report available is for 1996, which may raise some questions in its own right. The numbers are, of course, kept secret until the book is published. While browsing the Ontario statistics, I noticed a surprising trend in the number of convictions in the province.

Drastic drop in convictions

I had a look at the 1996 convictions and compared them to a decade earlier to see how total traffic enforcement efforts in the province may have changed. I thought this might be interesting, because of the coming and going of photo radar, the recent public concern about "aggressive" driving, and the government's motto "Highway Safety Through Enforcement." I was expecting to see the conviction numbers increasing gradually. To my utter amazement, I saw that convictions have declined drastically in the 1990s. In 1996 there were just over 850,000 convictions of all types. A full decade earlier, 1986's total was over 1.3 million. Despite widespread complaints about increased bad driving, the conviction data seem to show that Ontario drivers have become much more law-abiding.

The ten-year comparison shows a 39% drop in total convictions registered; a massive change. The total convictions for all violations combined in '96 were less than speeding convictions alone in '88. In all years, of course, the majority of convictions are for speeding, which accounts for about half of the total. In 1996, there were under 450,000 speeding convictions, compared to 786,000 in 1986 (down 43%).

The apparent drop in convictions seemed so strange that I thought at first there must be something odd about either '86 or '96. Having a few other old ORSARs around, I looked at convictions in other years. The graph below shows total convictions, Highway Traffic Act convictions, and HTA speeding convictions for the available years between 1985 and the present. It seems that '86 was not an unusually high year for convictions; it fits in with the trend of the late '80s. Convictions peaked in 1988, and compared to '88, the '96 drop is about 50%! It also seems '96 is not just an unusually low-conviction year either, since '94 and '95 were about the same. After 1989, you could say the wheels started falling off Ontario's traffic enforcement. The decline seems to have leveled off abruptly in 1994.

Ontario Traffic Convictions 1985-1996
Click to enlarge

What else changed?

When I noticed the surprising trend in convictions, I went looking for comparable changes in the other figures reported in the book. Economic times were still tough in '96, I figured, so maybe there were fewer drivers or vehicles than in the late-'80s boom times. I figured wrong, not all that many drivers moved out to British Columbia or stateside. In '96 there were 1.4 million more licensed drivers than in '86 (up 25%). In '96 there were over a million more registered vehicles (up 19%), and the estimate of total kilometers driven was up 16%, compared to '86.

If driving had improved, maybe there were fewer collisions and less need for enforcement. Wrong again, reported collisions in '96 were 215,000, up 15% from '86. This increase is just about in-line with the distance driven, so the collision rate (per kilometer of driving) was virtually the same in '96 as in '86. Apparently the risk of crashing stayed about the same over the decade, suggesting driving had not improved too much.

Missing 600,000 convictions?

If the violation and conviction rates had stayed the same over the decade, it seems there would have been over 600,000 to 700,000 more convictions in 1996 than there actually were. That's a lot of convictions. If nothing else, it represents a big loss of fine revenue for the government: $60 to 70 million a year, assuming a modest $100 for the average fine.

What could account for such a drastic decline in the province's enforcement effort? Certainly, police budgets were severely constrained for many years, starting even before the late 80s, but perhaps being even tighter in the deficit-busting 90s. Many police forces in the province have moved over to a "community policing" model, which may also reduce the effort available for traffic enforcement. No-fault insurance seems to lead to fewer charges for crash-involved drivers. Recently, there has been some diversion of ticketed drivers into local educational programs, with charges being dropped by the local police, who collect a fee for the training. Photo radar tickets were not registered against driver records. Court backlogs may reduce the number of convictions in the more serious, criminal code offenses. So, while there are additional factors that account for a small fraction of the reduction the number of convictions being registered, it seems unlikely that there has not been a real reduction of enforcement effort.

To be fair, the government has put some resources back into policing in the last couple of years, and an official I spoke to thought '97 conviction numbers might be higher than '96. But how long will it take to get the enforcement effort back to where it was 10 years ago?

Why did fatalities drop?

So, was there any good news on the roads as enforcement declined? Happily, fatal and non-fatal injuries did decline, in terms of rate per kilometer, rate per head of population, and even in absolute numbers of casualties. Only 929 persons were killed in '96, down 16% from '86. So how is it we were crashing at the same rate but doing less human damage?

There are, of course, a number of possible explanations for fewer deaths per crash. The passenger protection in cars improved in the 1990s. However, as well as fewer vehicle occupant deaths, there were also fewer pedestrian deaths, so the decline in deaths is not only the result of airbags and more crashworthy cars. Modern aerodynamic cars, with their low noses, are probably a little friendlier when striking a pedestrian. The increase in light trucks and sport-utilities in the fleet could offset either of these improvements to some extent.

Fewer are killed in tough times

A likely explanation for at least part of the general decline in deaths, is the recession in the '90s, which was still lurking around in '96. It is known that annual variation in fatalities is closely related to the business cycle, with bad times reliably leading to fewer deaths. Slightly lower average traveling speeds can produce a reduction in average crash severity, and fewer casualties, without necessarily producing fewer crashes. This may be one of the ways that bad times reduce deaths: time is always money, but in bad times it may be less money.

Should we care about reduced enforcement?

What are the consequences of the decline in enforcement? The province seems to have "gotten away with it", in the sense that there has been no public outcry and fatalities, for whatever reason, went down. Some may think a flat collision rate and 929 fatalities is good enough. Indeed, when I asked a senior safety official about the conviction drop, he immediately pointed out that the fatalities were looking good, so maybe it is not a problem. We might, however, want to consider how many of the 929 fatalities might be alive, and how much other cost in lives and dollars might have been saved, if enforcement effort had remained at the late eighties levels. If enforcement is really as important as we think it is, why has the province not seen drastic consequences in the crash rate as enforcement dropped in half? The evaluations of naturally occurring "quasi-experiments," such as police strikes and enforcement blitzes, suggest that the effects are not always obvious or immediate. To have short -term effects, changes in traffic enforcement have to be very substantial, with patrol intensity being doubled or tripled before drivers seem to notice. However, short-term studies also cannot assess the long-term effects that enforcement, or its lack, may have on the development of drivers' expectations and habits. The enforcement change in Ontario was not abrupt but took place over 4 or 5 years. We would expect the ultimate effects to take some time to set in as well. If driver behaviour really has deteriorated, as many people seem to think, maybe that is the effect of the earlier reduction in enforcement. This is hard to prove one way or the other. We have no ongoing study of drivers' perceptions of enforcement threat.

How important is enforcement?

We have the background of developing a comprehensive review of the scientific literature on influencing driver behaviour. From this perspective, enforcement clearly is a critical part of an effective, well-managed road safety program. The resources must be provided to support a visible and credible enforcement presence on the roads.

The "driving culture" that determines how we as individuals choose to drive is very much influenced by the deterrent effects that result from a real chance of getting caught for violations. To produce a positive influence on driver behaviour we need both "the carrot and the stick." A visible, credible enforcement threat is the sharp end of that stick. We know from the research literature that the chances of getting caught are much more important than the size of the penalty in shaping driver behaviour. Massive penalties are meaningless if there is little chance of them being delivered against a violator.

There are ways of increasing enforcement cost-effectiveness, such as STEP-type selective enforcement, but the basic infrastructure and resources need to be there. Automated enforcement provides an economical way to increase surveillance for some violations. However, if the program is weak or resources saved are diverted away from traffic enforcement, the net benefit of enforcement automation is likely to be low. Certainly, Ontario's recent experiment with photo radar was a weak implementation of the technology and ended badly. For a complete literature review and discussion of how enforcement works, see our books on changing road user behaviour listed below.

In good safety management, the costs of not enforcing have to be considered along with the costs of doing it. Of course, this consideration requires some serious coordination across different departmental budgets and responsibilities. As always, this coordination, and the loss of bureaucratic autonomy that it requires, remains the biggest roadblock to making the roads safer. Most places, nobody is in charge of the bureaucratic head banging and budget trade-offs needed for coordinating programs. Fixing the bureaucratic coordination problem was a main recommendation of The Ontario Select Committee on Highway Safety in 1977. Some things never change.

Some answers needed

Is the decline in convictions a real reflection the traffic enforcement effort in the province?
If so, what caused enforcement to drop in half?
Was the cause the result of a deliberate policy decision?
If so, whose decision was it and how was it justified?
Who is accountable for the level of enforcement in the province?
What have been the trends speeding and other violations?
What have been the costs in lives and dollars of the reduction in enforcement?
How many conviction have been registered in the years since 1996?
Is there a plan to return to past levels of enforcement surveillance?
How long will it take to bring enforcement levels back up?
Is the drop in enforcement unique to Ontario, or has it been going on in other jurisdictions as well?

References

Lonero, L. and Clinton, K. (1998). Changing Road User Behaviour: What Works and What Doesn't. Toronto, PDE Publications, (416) 767-4885.

Lonero, L., Clinton, K., Wilde, G., Roach, K., McKnight, J., Maclean, H., Guastello, S., and Lamble, R. (1994). The Roles of Legislation, Education, and Reinforcement in Changing Road User Behaviour. Ontario Ministry of Transportation, Toronto. Ontario

Ministry of Transportation (1996). Road Safety Annual Report (ORSAR). Safety Policy Branch, Ministry of Transportation, Downsview, Ontario, (416) 235-3585.