Beware of lowering the bar
Researchers with the Federal Motor Carrier Safety Administration's Field Systems Team recently completed a 12-month study that examined the relationship between drivers' traffic convictions and their employer's overall safety performance. (Go to http://infosys.fmcsa.dot.gov.)
The purpose of the study was to determine if a “driver history measure,” as calculated by severity-weighted traffic convictions, could be used to identify high-risk motor carriers. In fact, the study showed that carriers with poorer driver history measures were also likely to have poorer safety performance.
The study included data from state and national driver and carrier safety databases such as the Commercial Driver License Information System (CDLIS) and the Motor Carrier Management Information System (MCMIS).
First, the researchers created a hypothetical driver history measure by attaching severity weights to various traffic convictions. A disqualifying offense (driving while under the influence, for example), was assigned a severity weight of three, while a serious offense (e.g., more than 15 mph over posted speed limit) was assigned a severity weight of two.
Next, the researchers sampled the September 1999-September 2000 federal truck safety database (MCMIS) to identify approximately 75,000 drivers who were positively linked to 16,000 employing motor carriers by either roadside inspection or accident reports. Measures were taken to remove statistical bias from the sample data.
The researchers next computed driver history measures using three years of the CDLIS traffic conviction history for the 75,000 identified drivers. Individual driver history measures ranged from a low of 0 (no violations) to a high of 49. A driver history measure for each carrier was computed by averaging each company's individual driver history measures. Average carrier driver history measures ranged from a low of 0 to a high of 27.
Finally, each company's driver history measure was correlated with safety metrics such as accident frequency, driver and vehicle out-of-service rates, and the four Safestat Safety Evaluation Area (SEA) indicators — Accident SEA, Driver SEA, Vehicle SEA and Safety Management SEA.
The comparisons revealed that driver history measures were most strongly correlated with the Accident SEA indicator. Remember, the Accident SEA is a national ranking of all carriers with respect to their time and severity-weighted crash involvement rates. The study found carriers with high (poor) driver history measures were also likely to have high (poor) Accident SEAs.
Strong correlations were also demonstrated between driver quality and driver out-of-service rates and the SafeStat Driver SEA.
The implications are far-reaching. First of all, the findings support long-held anecdotal evidence of the relationship between moving violation history and crash involvement. The findings also confirm 1998 research, conducted for the states of Michigan and Indiana, which showed that high driver citation rates were associated with high carrier crash rates.
Second, the research provides information for enhancing the current SafeStat carrier ranking system. For example, the research supports the development of a driver history measure, which could be incorporated into the Safety Management SEA. This measure might provide even more information about high-risk carriers or possibly identify others that have gone undetected.
We have repeatedly remarked that many carriers hire their own problems. Some carriers lower driver standards during periods of expansion or high turnover so they can fill empty seats. Others opt to permanently lower the bar. Frequently, these actions are followed by dramatic increases in claim frequency and severity.
How do your criteria stack up against best industry practices? My guess is that any savings realized by lowering driver standards are more than offset by increased crash and insurance costs.
Jim York is the manager of Zurich North America's Risk Engineering Team, based in Schaumburg, IL.
