The Texas Transportation Institute examined concerns that red light cameras were being used by cities that had not first exhausted available engineering alternatives such as improving signal timing and visibility. They studied individual police accident reports from 181 intersection approaches across three Texas cities over three years to determine the most effective solutions for problem intersections.

The study found that improving signal visibility reduced violations 25 percent. Other changes could net between 18 and 48 percent reductions. Yet they found when the yellow signal was 1 second shorter than what the standard ITE timing formula specifies as a minimum, red light violations jumped 110%. Adding an additional second to the ITE minimum yellow yielded 53% reduction in violations, producing the greatest benefit of all the factors studied (2-6). When safety is the main concern, preventing crashes is more important than reducing violations. Yellow signal timing again proved most effective in reducing crashes. An extra second yielded a 40 percent collision reduction.

The study also found that the vast majority of red light camera tickets are issued within the first second a light is red -- in fact, the average ticket is issued when the light has been red for half a second or less. Yet right-angle crashes, which account for the majority of red-light related collisions, "with one exception, all of the right-angle crashes occurred after 5 seconds or more of red" (5-16). In other words, tickets are being issued primarily for split-second violations where collisions are not occurring.

Full study available in 1.3mb PDF format file.

Key Statistic:
The data in figure 2-11 indicate that there is a trend toward fewer red-light related crashes when the observed yellow is longer than the computed duration. (2-22)

An increase in yellow duration of 1.0s is associated with an MF [crash frequency] of about 0.6, which corresponds to a 40 percent reduction in crashes. (2-20)

It is also likely that there is a point of diminishing returns where further increases in enforcement effort bring little additional safety benefit. (3-2)

If the frequency of violations is excessive, the violations are most likely caused by congestion, dense traffic streams, or conditions that make it difficult for drivers to stop. (5-6)