Federal Motor Carrier Safety Administration April 2005

NOTES AND SOURCES
[a] FMCSA HOS NPRM estimate, April 2000. This estimate was generated from a number of sources. CDL drivers (required to hold a commercial driver’s license) were estimated from the Federal Highway Administration’s 1996 drug and alcohol testing survey. The ratio of non-CDL to CDL drivers was then used to generate an estimate of non-CDL and total drivers. The estimated number of drivers was then distributed among five driver types outlined in the HOS NPRM, using data from the 1992 Truck Inventory and Use Survey (TIUS), which is conducted every 5 years by the U.S. Census Bureau, using a large sample of trucks.
[b] BLS OES estimate, May 2003. OES is a semi-annual mail survey of 400,000 establishments, which excludes self-employed and farm-employed drivers, owners and partners in unincorporated firms, household workers, and unpaid family workers. It includes only domestic establishments; presumably, foreign-based drivers are excluded. Also, it includes only drivers whose primary occupation is driving; part-time truck or bus drivers are not counted.
[c] BLS CPS estimate, 2003. CPS is a monthly survey of approximately 60,000 households. The survey covers all sectors of the economy, including self-employed workers as well as other employees. Drivers are asked their occupation and categorized accordingly. Although the CPS surveys a large number of individuals, people who drive as a secondary part of their work (such as utility company drivers) are not included as commercial drivers. The CPS is the only source that includes demographic information, such as driver age and gender.
[d] FMCSA MCMIS estimate, October 2004. MCMIS data are generated from Form MCS-150, which carriers are required to file when they apply for authority to operate, and which they must update every 2 years. The MCMIS Census File includes all drivers for interstate carriers and some drivers for intrastate carriers. Driver categories in the file include: interstate and intrastate; 100 air-mile and non-100 air-mile; and CDL and total drivers. The numbers of interstate and intrastate drivers do not sum to the total number of drivers in the file, possibly because some carriers include drivers in more than one category.
[e] FMCSA internal estimate, 2004. This estimate was generated from a number of sources. CDL interstate drivers were estimated from FMCSA’s 2002 drug and alcohol testing survey. Six States appear to require registration of intrastate carriers, and for those States 24 percent of CDL drivers were employed in intrastate operations; accordingly, the 24-percent share was used to estimate the total number of CDL intrastate drivers. Non-CDL interstate drivers were estimated from the MCMIS Census File. To estimate non-CDL intrastate drivers, the total number of non-CDL drivers in the six States above was estimated, and the estimated number of non-CDL interstate drivers was subtracted from the total to determine the share of non-CDL intrastate drivers in those States, which was used to estimate the total number of non-CDL intrastate drivers.
[f] AAMVA CDLIS estimate, February 2004. CDLIS is a distributed database, with specific information residing on State computers. For most of the drivers applying for a CDL, a “master pointer file” is created, whether or not they obtain a CDL. In addition, CDL holders who have had their CDLs revoked or suspended also have master pointer files. Drivers who do not renew their CDLs may also have pointer files in the database. Drivers without CDLs who are convicted of a disqualifying offense while operating a commercial motor vehicle also have pointer files. All such drivers are supposed to remain on CDLIS until their violations or convictions are no longer relevant.
[g] In 2000, the BLS CPS reported approximately 3 million truck drivers. That figure was used in FMCSA’s 2002 Hours-of-Service Final Regulatory Evaluation and Safety Performance History of New Drivers Final Regulatory Evaluation.
[h] The Medical Certificate and Field of Vision NPRMs have not been published.