Some states already have primary offense laws. A survey of seat belt use among the fatally injured suggests that seat belt use in that group was 15 percent higher in states with primary offense enforcement laws. In 1996, states that treated seat belt use as a primary offense reported that seat belts were used 74 percent of the time. States that treated seat belt use as a secondary offense reported usage rates of 61 percent.

Unfortunately, data like these fail to show that making seat belt usage a primary offense decreases traffic injuries and fatalities. In fact, no jurisdiction that has passed a seat belt law has shown evidence of a reduction in road accident deaths. To explore this odd but highly robust finding, experimenters asked volunteers to drive five horsepower go-karts with and without seat belts. They found that those wearing seat belts drove their karts faster. While this does not prove that car drivers do the same, it points in that direction.

A similar study was done with real drivers on public roads. When subjects who normally did not wear seat belts were asked to do so, they were observed to drive faster, followed more closely, and braked later. In other words, people who are naturally cautious voluntarily choose to wear seat belts, and voluntarily drive safely. When reckless people are forced to wear seat belts, they "compensate" for the increased safety by driving more recklessly.

Nor is it clear that making seat belt use a primary offense will significantly change either usage or motor vehicle injury and death rates. It is important to keep in mind that some people wear seat belts whether there are laws requiring it or not. States with more risk averse populations may also have populations that are more likely to both drive carefully and buckle up. They may also be more likely to pass primary seat belt laws. New York passed a primary seat belt law in 1984. In 1996, its observed seat belt usage rate was 74%, and a large fraction, 46%, of its fatally injured car occupants were wearing seat belts. Its fatality rate per 100,000,000 vehicle miles traveled (VMT) was 1.3. But Iowa, which has had a primary law since 1986, had a fatality rate per VMT of 1.7 in 1996 despite the fact that its observed seat belt use rate was 75% and fully 50% of its fatally injured car occupants were wearing seat belts.

Colorados fatality rate per VMT, 1.7, is the same as Iowas. This is in spite of the fact that Colorados observed rate of seat belt use was just 59%. Furthermore, high observed usage does not guarantee a lower death rate. Wyoming had no primary law and an observed usage rate of 72%. However, its fatality rate was higher, at 1.9 per VMT, and only 28.8% of its fatally injured car occupants were wearing seat belts.

The point is that there is more to highway safety than seat belt use. The age of the population, the condition of the roads, the speed at which people habitually travel, their affinity for drink, and a great many other factors all make a difference. Making failure to wear a seat belt a class B infraction will probably not do much to changes behavior, let alone accident results, if only because relatively few motorists will even know that the change has occurred, let alone what it means.

Seat belt laws differ from traffic laws in that they attempt to regulate behavior that poses no danger to others. A person who refuses to wear a seat belt increases his own risk of injury or death, but not necessarily anyone elses. Traffic laws have historically sought to regulate driver behavior that poses an obvious risk to others. Everyone can see that running a stop sign endangers others. Since the law makes obvious sense, most people obey it. A person who refuses to wear a seat belt increases his own risk of injury or death, but not necessarily anyone elses. Traffic laws have historically sought to regulate driver behavior that poses an obvious risk to others. Everyone can see that running a stop sign endangers others. Since the law makes obvious sense, most people obey it.

As the 20-year experiment with artificially low speed limits demonstrated, laws designed to regulate individual risk do not necessarily enjoy high rates of compliance, and low compliance with one law may erode general respect for all laws. This seems to be a particular problem when government tries to regulate the risk involved in routine activities that generally end without incident.

When people see no reason to change the risk they are exposed to, they do not change their behavior. Frustrated government officials then proceed to ratchet penalties higher and higher in an effort to save face and force compliance. Since not enough people were thought to use seat belts in 1987, Colorado passed a law requiring front seat passengers to buckle up. Now "enough" people still are not using seat belts and the response is to increase police powers and to make the legal penalty harsher. What if that doesnt work, either? Just how much should otherwise law-abiding citizens have to pay for failing to wear a seat belt when that activity poses no danger to others? Should it cost them their license?

Moreover, granting police increased ability to stop people for something which does not endanger other people is an invitation to selective enforcement and abuse. Making seat belt laws a primary offense gives unscrupulous police officers a pretext to pull over minorities, young people, people with pro-gun bumper stickers, or any other type of person the police officer may not like. In the context of enforcement of laws against speeding or running a red light, the danger of abusive enforcement is much less; the drivers conduct has provided objective evidence that he is a danger.

Laws against dangerous driving can be easily enforced just by observing how the car moves. In contrast, seat belt laws can only be enforced by looking into the interior of the car. Will photo radar stations eventually be photographing the occupants of a car, to see who is wearing seat belts?

The real question is not about seat belts. The real question is whether, and how much, government can and should regulate the risk that a free adult chooses to incur. Trying to convince people to wear their seat belts is one thing. Requiring them to do so when educational efforts fail is another.

Have the Amendments fixed the Bill?  The House transportation Committee added an amendment to sunset the "primary offense" portion of the bill. But the sunset does not go into effect if a study to be performed by the Commissioner of Insurance finds that the bill reduced auto insurance rates by 5% or more.  It is rather unusual for government studies ever to find that less government is needed. Accordingly, the sunset provision is unlikely to have any effect.

The full House made two other notable changes. First, the "primary offense" portion would not go into effect until December 2, 1999.   But the police would still be allowed to stop vehicles in order to issue warnings.   Thus, the risk of the seat belt stop being used as a pretext for additional police intrusion wold remain.

The other change from the House attempts to deal with the problem of seat belt stops as pretext stops, but fails.  The Amendment states that a seat belt violation is not justification for arrest (it wasn't anyway), and that a seat belt violation is not probable cause for a search of the car (it wasn't anyway).

The problem with allowing seat belt stops was never that the seat belt violation itself allowed an arrest or a search. Rather, once the police officer has initiated an encounter based on the seat belt violation, he can simply   ask for consent to search the car. Such "consent searches" do not require probable cause, since they are based on the "consent" of the driver. but unless the driver is a lawyer, it is unlikely that the driver will understand that he has a real right to refuse consent.  The inherently coercive atmosphere of a traffic stop, along with the presumed authority of the policeman, means that almost all drivers who are asked for "consent" will allow a search.

In the rare case where the driver does not consent to a search, the police officer can use the seat belt stop to develop probable cause.  For example, "When I asked for consent to search the car, the driver seemed nervous.   I thought his explanation of where he was going was suspicious.  I  saw him make a furtive movement as I approached the car.  Based on my training as  a   police officer, I knew that the driver was traveling along a road commonly used for drug deliveries (e.g., I-70 or Colfax Avenue)."