Transportation Safety

Last week’s commuter train crash in New Jersey has left people wondering how safe our transportation system really is. We can answer this question with data from National Transportation Statistics, which show passenger miles, fatalities, and injuries by mode of transportation since 1990.

Mode1990-1999Last 10 YearsChange
Scheduled Air0.30.0-92.5%
Highway10.88.0-31.3%
Bus5.14.4-13.9%
Light Rail14.013.5-3.4%
Heavy Rail7.64.5-40.5%
Commuter Rail11.78.9-23.7%
Amtrak35.933.2-7.5%
Table One: Fatalities per billion passenger miles by mode. As noted in the text, the most recent decade is 2005-2014 except for commuter rail, which is 2003-2012. Sources: Calculated from National Transportation Statistics, tables 1-40, 2-1, 2-34, and 2-35.

The statistics show transit data only through 2012, but the Federal Transit Administration has safety data for the years since then. Unfortunately, the Federal Railroad Administration, not the Federal Transit Administration, monitors commuter rail safety, and it doesn’t seem to publish those numbers, so we only have them through 2012.

As the numbers in the above table show, air travel is by far the safest, followed by bus and heavy rail. Highways are moderately safe. The most dangerous modes are Amtrak, light rail, and commuter rail. These are not, for the most part, dangerous for people on board the trains but for pedestrians and occupants of motor vehicles struck by trains. On average, only about 3 percent of Amtrak fatalities and 5 percent of light- and heavy-rail fatalities are people who were on board the trains. Though close to 20 percent of light- and heavy-rail fatalities are people waiting to board or who have recently alighted from trains, this still means three out of four are people off the trains. The numbers are probably pretty similar for commuter rail.

As noted before, the Antiplanner has disputed a change the Obama administration made to how the DOT calculates highway passenger miles. In the Bureau of Transportation Statistics tables, the average motor vehicle is presumed to have about 1.64 people before 2009, but only 1.44 people since. As the latest National Household Transportation Survey estimated the number is 1.67 (see table 16 on page 23), the older number is more realistic. For the tables today, I used the BTS numbers, but if you agree with me that the recent calculations are wrong, reduce the 2005-2014 numbers by 7 percent.

Making this correction suggests that highway fatality rates actually fell by 31.3 percent since the 1990s. Bus and light-rail fatality rates hardly improved at all, while the heavy-rail rate improved in the 1990s but hasn’t changed much since 2000.
Men with ED can take anti-impotency drug to relax cialis cipla their veins and achieve erection. Therefore adopt this drug measure before your life has become bitter by the impotency attack as there used to be no harm in doing such thing as these drugs are found safe and effective, if taken under the supervision of http://amerikabulteni.com/2011/11/09/rick-perry-baskan-olursa-kaldiracagini-ilan-ettigi-federal-kurumun-ismini-unuttu/ levitra 60 mg a medical practitioner. Erectile dysfunction viagra tablets in italia Click Here has gained widespread awareness over the years and no longer is impotence a hidden condition that is never discussed. Propecia, a hair loss drug, buy cheap levitra you could try these out is among the best advantages of using the Internet to purchase your medications is that they are composed of the same chemical structure to those produced by the body.

Road Category1990-19992005-2014Change
Rural, Total16.113.2-17.7%
Interstate7.66.1-19.8%
Other Arterials16.013.8-14.1%
Collectors20.116.6-17.2%
Local24.618.8-23.5%
Urban, Total7.35.3-28.1%
Interstate4.03.1-21.7%
Other Arterials8.25.7-30.7%
Collectors6.44.6-28.5%
Local9.87.8-19.9%
Table Two: Fatalities per billion passenger miles by type of roadway. Source: Vehicle miles from VM-202 of Highway Statistics, while fatalities and conversions to passenger miles are from National Transportation Statistics.

The highway numbers in table one include all roads, but safety varies quite a bit from one kind of road to another. As shown in the table above, freeways tend to be safest, and urban roads tend to be safer than rural roads. Urban interstates are currently the safest form of motorized ground transportation. Again, reduce the 2005-2014 numbers by 7 percent if you agree that the auto occupancy rates since 2009 are miscalculated.

Mode1990-1999Last 10 YearsChange
Scheduled Air00-71.5%
Highway821470-42.7%
Bus2,072650-68.6%
Light Rail1,483385-74.0%
Heavy Rail976384-60.7%
Commuter Rail280123-44.5%
Amtrak16,2927,322-55.1%
Table Three: Injuries per billion passenger miles by mode. Sources: Calculated from National Transportation Statistics, tables 1-40, 2-2, 2-34, and 2-36. 2013 and 2014 injury data for bus and light- and heavy-rail from National Transit Database.

National Transportation Statistics also present injury data by mode. As shown in table three, the number of injuries on Amtrak is massive. About a third of the injuries on Amtrak happen to passengers. On light rail, about a quarter are passengers and 40 percent are people waiting to get on or who recently got off of trains. For heavy rail, 5 percent are passengers and 90 percent are people on platforms.

The relative standings are about the same as for fatalities except commuter-rail injury rates are lower than for most other forms of ground transportation. In fact, the commuter-rail numbers seem questionably low: while the injury rates for other modes are 80 to 400 times greater than fatality rates, they are only around 20 times greater for commuter rail. Since commuter-rail numbers come from the FRA and other transit numbers come from the FTA, I wonder if there is a difference in methodologies.

In any case, the data show there is a wide disparity in the safety among different modes, and that all modes are getting safer but some are getting safer faster than others. While many people will say that positive train control would make commuter rail and Amtrak safer, in fact the vast majority of fatalities resulting from these modes happen to people off the trains. Since positive train control would not have helped many of those people, it is likely that putting more efforts into improving grade crossings and discouraging pedestrians from wandering on tracks would be more cost-effective in improving the safety records of these modes.

Tagged , , , , . Bookmark the permalink.

About The Antiplanner

The Antiplanner is a forester and economist with more than fifty years of experience critiquing government land-use and transportation plans.

5 Responses to Transportation Safety

  1. JimKarlock says:

    Anyone know where to get the numbers for bicycles?

    thanks
    JK

  2. prk166 says:

    Is Washington’s Metro considered heavy rail or commuter rail?

  3. OFP2003 says:

    The tables don’t include the four passenger jets lost on Sept 11, 2001, is that because that date isn’t included in the date range of the table or is that because they weren’t accidents?
    .
    I ask because I’m also wondering if the people injured in criminal attacks on the trains or inside the platform are included in the tables. It would be interesting to be able to confirm whether you are safer from crime in your car or on the train (or train platform).

  4. prk166: Washington’s Metro is heavy rail. The Virginia Railway Express and Maryland trains are commuter rail.

    OFP2003: No, the date ranges don’t include 9/11. Even in 2001, air fatalities per billion passenger miles were only 1.1, though that only includes the 544 airline passengers, not the people in the buildings. I wouldn’t count those as accidents, however.

  5. Dave Brough says:

    One statistic omitted is the number of pedestrians (or cyclists) killed getting to or from public transit, be it bus, rail or whatever – or even standing at a bus/rail stop, as was the case recently in Las Vegas when a runaway car killed 4 people. http://www.reviewjournal.com/news/las-vegas/four-killed-car-demolishes-bus-stop-las-vegas
    My guess is that a high percentage of walking or cycling deaths fit into that non-existent category.

Leave a Reply