The Red Light Running Crisis
Is it Intentional?

Office of the Majority Leader
U.S. House of Representatives
Red Light Running Scam
May 2001

5. Changes in the Safety Codes

The Report
1. Something Funny
2. Camera Revenue
3. The Theory
4. The Fact
5. Code Changes
6. Cameras Ineffective
7. Conclusion
8. References

Printed Copy
(200k, PDF format)

View the Documents

Where do the problem intersections come from? We've seen that experience tells us that if there's a red light running problem, yellow light times should be increased. And the theory tells us the same. So why have yellow signal times decreased? The answer is that the organizations responsible for maintaining our intersection safety codes have altered the regulations specifically to accommodate camera enforcement and decrease yellow times.


30' 50' 70' 90' 110'
20 3.8 4.4 5.6 5.7 6.4
30 3.6 4.1 4.5 5.0 5.5
40 3.9 4.2 4.5 4.9 5.2
50 4.1 4.4 4.7 5.0 5.2
60 4.5 4.7 4.9 5.1 5.4
Source: 1976 Transportation & Traffic Engineering Handbook
The chart below provides the theoretical minimum yellow clearance signal times based on speed and intersection width from the 1976 edition of the Institute of Transportation Engineers (ITE) handbook. Note that the 100-foot intersection in Beaverton, Oregon had a 3.1 second yellow signal time in a 30MPH zone, as mentioned in Chapter 1 above. As one can see from the chart, that time would be inadequate for any condition. But it's quite profitable for the red light camera installed at that location.

To understand more fully the extent of changes to the signal timing codes, one must first examine the prior formula used for calculating yellow times.

The 1976 ITE Handbook

In 1976, yellow time was known as the "yellow clearance interval." This was the theoretical minimum amount of time needed for an automobile to clear the far side of the intersection from a given distance away, or come to a safe stop. This was calculated by adding three variables:
  1. Reaction time: How long it takes, on average, to recognize the situation and decide whether to stop or continue through the intersection. Usually this is 1.0 seconds.
  2. Stopping time: This figure is calculated based upon the length of the intersection and the average deceleration rate for automobiles.
  3. Time needed to clear the intersection: Based on the approach speed, how long it would take an automobile to traverse the length of the intersection.

The 1985 ITE Proposed Recommended Practice

By 1985, ITE had begun to change the way signal times were calculated in the past. The first modifications were published in their "Proposed Recommended Practice" a mere three years after New York City began researching how it would implement the first red light cameras in the United States.

These changes were further explained in the 1989 ITE Journal article, "Determining Vehicle Signal Change Intervals." This report begins by clearly stating that the ITE's intent is to change laws across the country because, "adopting a uniform method cannot precede adoption of uniform laws" (page 27). In other words, for red light cameras to be adopted nationwide, the laws must change nationwide. And they provide at least three methods that have as their result a reduction, in most cases, of yellow signal time as well as easy adoption of camera enforcement.

1. Their goals are not entirely safety related

The goals and objectives of the 1985 and 1989 documents are clearly related to red light camera enforcement. Consider:
Goal: Recommend legal definitions for the various aspects of the change interval and a defensible methodology for calculating and evaluating change intervals. (1985, page 5; 1989 page 27.)

And the second signal timing objective listed:
Allow easy identification of violators by law enforcement agents. (1985, page 5; 1989, page 28.)

This is a strange goal for someone who wants to design safer intersections. Yet it is a perfect goal for one whose true intent is not safety but rather the convenient installation of a red light camera.

2. Reduced Yellow I: Ignore the Actual Speed of Traffic

The first method for reducing yellow time is found on page 29 (1989) where the document states, "It may be possible to use the posted speed as the approach speed."

What that means is that signal times would be determined by the speed limit rather than the actual speed 85 percent of traffic is traveling, known as the "85th percentile speed." The result of this change in practice would be an underestimate of the actual speed of vehicles at the intersection. And this factor alone can result in yellow time shortfalls of 20 percent or more.

The laws of physics dictate that the distance required to stop your car is based entirely on the speed at which you are traveling, not what is printed on a sign on the side of the road. No rational safety consideration would lead one to choose posted speed over actual speed. But it does allow for a reduction in yellow light time.

3. Reduced Yellow II: Replace yellow time with "all-red clearance"

A comparison of old and new calculations
1976 & earlier Yellow time:
Clearance time:
The result:
Reaction time + Stopping time + Clearance time
Included in yellow time, plus all-red of 1-2 seconds at the option of the engineer.
Entries on red happen, but are rare.
1982-1985 Yellow time:
Clearance time:
The result:
Reaction time + Stopping time
Has changed from yellow to all-red.
Yellow is reduced by a third from '76 values, and more red entries occur.
1999-present Yellow time:
Clearance time:
The result:
Reaction time + Stopping time
All-red clearance time is now optional.
Yellow time is the same as '85, but opposing traffic gets the green while people going the other way are entering and clearing the intersection against their own red signal.

Take the traditional definition and formula for calculating the duration of the yellow light signal. You might need three seconds of yellow to warn approaching motorists that they need to stop, and two more seconds of yellow on top of that to allow vehicles enough time to clear before opposing traffic is given the green light. The total yellow time for such an intersection would be five seconds.

On page 30 of the 1989 report, the ITE proposes to take that five seconds of yellow in the hypothetical intersection above and reduce it to three seconds of yellow, and two seconds in which all sides of the intersection are given the red light (this is known as the "all-red period"). Eliminating that much yellow time, again, is of questionable safety value. But there is no question that in practice this method would yield an increase in the number of vehicles that enter the intersection on red, given the two second reduction in the amount of time one would have to clear the intersection legally. Again, it is unlikely that a rational safety consideration would lead you to choose this method. But it does allow for a reduction in yellow light time. And it will increase red light running. Why? Because the light turns red faster.

Changes were made to the code specifically for camera enforcement

These changes are significant. But if it was not clear enough in the above documents that ITE had cameras in mind in 1985, they make it explicit a few years later. The 1994 ITE "Determining Vehicle Signal Change and Clearance Interval" states:
When the percentage of vehicles that entered on a red indication exceeds that which is locally acceptable, the yellow change interval may be lengthened (or shortened) until the percentage conforms to local standards, or enforcement can be used instead. (Page 5, emphasis added).

In other words, if too many people are running red lights, jurisdictions need not address deficiencies in intersection design or signal timing. Instead, they can simply "use enforcement" by putting up a red light camera. They are suggesting creation of an intersection that will have a perpetually high level of red light runners by design. Since enforcement by police officers wouldn't be 24-hours a day, it is hard to conceive that they had anything other than 24-hour red light cameras in mind.

Changes in the yellow light formula linked to red light running
Yellow Times Compared
Posted speed: 35 MPH. Width: 80 feet. Grade: 2.6% downhill.
1976 ITE Formula 1999 ITE Formula
4.64 seconds
(round up to 5)
3.8 seconds
(round up to 4)

The changes in the yellow signal timing regulations have resulted in the inadequate yellow times. And these inadequate yellow times are the likely cause of almost 80 percent of red light entries, as discussed above.

If we look closely at one of the intersections Retting studied, the signal at Columbia Pike at Greenbrier in Arlington, Virginia, we find that it has a measured yellow time of 4.0 seconds. This location was the second site studied in his "Red Light Running and Sensible Countermeasures." Using the 1999 formula results in a one second (20 percent) decrease in the yellow time compared to the 1976 formula. And, as mentioned above, according to Retting's study, 77 percent of red light entries happened in that first second the light was red instead of yellow.

Thus, if the old formula had been employed, the red light entry problem Retting studied would have been substantially reduced.

Elimination of the vehicle change interval, a chronology

It may be useful to consider the following excerpts from signal timing regulations that, when presented in chronological order, show a clear progression toward lowering yellow times to accommodate red light cameras:

1985 -- ITE, "Determining Vehicle Change Intervals: A Proposed Recommended Practice," states, "When the percent of vehicles that are last through the intersection which enter on red exceeds that which is locally acceptable (many agencies use a value of one to three percent), the yellow interval should be lengthened until the percentage conforms to local standards."

1988 -- Federal Highway Administration, "Manual on Traffic Control Devices" (MUTCD) states, "Signal Operation Must Relate To Traffic Flow" (Section 4B-20). Note that red light camera promoters use the opposite principle: they wish to use signals to modify traffic flow.  

1994 -- ITE, "Determining Vehicle Signal Change and Clearance Intervals" states, "When the percentage of vehicles that enter on a red indication exceeds that which is locally acceptable, the yellow change interval may be lengthened (or shortened) until the percentage conforms to local standards, or enforcement can be used instead."  

1999 -- ITE, "Traffic Engineering Handbook: Fifth Edition" states, "The red clearance interval is an optional interval that follows a Yellow Change Interval and precedes the next conflicting green interval. The red clearance interval is used to provide additional time following the Yellow Change Interval before conflicting traffic is released" (page 482).

2000/2001 -- Federal Highway Administration, "Manual on Traffic Control Devices" (MUTCD) states, "47. Red Clearance Interval: an optional interval that follows a yellow change interval and precedes the next conflicting green interval" (page 4A-5, Part 4, Highway Traffic Signals). Yellow time is calculated from "E. The posted speed or statutory speed limit or the 85th percentile speed on the uncontrolled approaches to the intersection" (page 4C-3).  

In all the above citations, emphasis is added to the key changes. The words in italics mark the differences between the old and new codes. Namely:
  1. The "should" in 1985 was changed to "may" in 1994.  
  2. "Or shortened" was added to the formulation in 1994.  
  3. "Or enforcement can be used instead" was added in 1994.  
  4. "Optional" was added to the definition of red clearance interval in 1999.  
  5. Finally, the Federal Highway Administration endorses all these changes in the December 2000 edition of the MUTCD.

Previous Page Next Page

Front Page | Get Updates | Site Map | About Us | Search | RSS Feed Driving politics