LA Metro’s board is going to address its safety crisis by investigating such costly technologies as facial recognition and body imaging devices to keep known offenders and weapons away from transit vehicles and stations.
Although the state and federal government have pumped billions of dollars into California transit agencies during and after the COVID pandemic, ridership remains well below 2019 levels. One barrier to getting riders back on transit is the perception that riding buses and trains is no longer safe.
Transit safety became a major issue at LA Metro’s April 25th board meeting after a passenger was fatally stabbed on a metro train. Agency staff had originally sought an emergency declaration to accelerate procurement of barriers to protect bus operators, but board members expanded the discussion to include passenger safety as well.
Reports of increased transit violence are not unique to LA Metro: reports of crime on California transit systems extend from San Diegoto multiple Bay Area counties. Like LA, SamTrans in San Mateo County is also planning to install barriers to protect bus operators.
While unions representing bus drivers are well positioned to secure protections for their members, passengers lack comparable advocacy. And, since ridership revenue accounts for a small and declining share of transit agency revenues, passengers have little ability to influence agencies by “voting with their dollars.”
Transit advocates often contend that mass transit is far safer than driving. This seems plausible because fatal car accidents are so frequent in the United States. But since driving is also so much more common than mass transit use, it is less clear that transit is safer on a unit basis.
Here’s some math: there were 42,514 road fatalities in 2022, versus just 223 related to rail transit and 116 related to bus transit. But total vehicle miles travelled on roads in 2022 were 3.17 trillion compared to just 4.16 billion miles travelled by all categories of transit vehicles. So personal vehicle mileage exceeded transit mileage by a factor of over 700.
To make a fairer comparison, it is also necessary to consider the fact that more people travel in a typical bus or train than in a car. Fortunately, the Federal Transit Administration reports total passenger miles travelled on transit and the Department of Energy has estimated that average passenger vehicle occupancy is 1.5.
With these numbers we can roughly estimate fatalities per billion passenger miles for each mode of travel. Using this metric, car travel turns out to be marginally safer: 0.89 car fatalities per 100 million passenger miles versus 0.94 for bus travel and 1.29 for rail transit.
But regardless of whether transit is really more dangerous, the perception that it is risky will continue to deter “choice” riders—those that can afford to drive or pay for rideshare. People generally feel more comfortable in their own vehicles than riding with a group of strangers who often include fare-evaders, homeless individuals, and people engaging in threatening or anti-social behaviors. Efforts to impose and enforce codes of conduct that might keep “unsavory” individuals off buses and trains run into objections from activists citing equity and fairness.
LA Metro’s board is going to address its safety crisis by investigating such costly technologies as facial recognition and body imaging devices to keep known offenders and weapons away from transit vehicles and stations.
But rather than spending ever more money in hopes of making transit more palatable, policymakers should instead consider downsizing the whole enterprise. Reduced transit spending would enable the state government to lower gasoline taxes and vehicle registration fees, reducing the cost of personal vehicles and enabling more Californians to afford them. Zoning liberalization that supports more mixed-use communities will allow more Californians to live closer to where they work, go to school, and engage in other essential activities.
Marc Joffe is a federalism and state policy analyst at the Cato Institute.
