Red light cameras, redux

[unixronin]

schneier reports on the latest on red-light cameras.  I'm pretty sure no-one here will be surprised by the conclusions:

"The rigorous studies clearly show red-light cameras don't work," said lead author Barbara Langland-Orban, professor and chair of health policy and management at the USF College of Public Health.  "Instead, they increase crashes and injuries as drivers attempt to abruptly stop at camera intersections."

Comprehensive studies from North Carolina, Virginia, and Ontario have all reported cameras are associated with increases in crashes.  The study by the Virginia Transportation Research Council also found that cameras were linked to increased crash costs.  The only studies that conclude cameras reduced crashes or injuries contained "major research design flaws," such as incomplete data or inadequate analyses, and were always conducted by researchers with links to the Insurance Institute for Highway Safety.  The IIHS, funded by automobile insurance companies, is the leading advocate for red-light cameras since insurance companies can profit from red-light cameras by way of higher premiums due to increased crashes and citations.

The report also mentions that not only do cities and states that use red-light cameras "tamper with" yellow lights, making them shorter to increase the number of inadvertent red-light violations, but six cities — Union City, CA; Dallas, TX; Lubbock, TX; Nashville, TN; Chattanooga, TN; and Springfield, MO — have been found guilty of shortening the yellow light cycles below what is allowed by law in order to increase red-light camera profits.