Unless you are a skydiver or bungee jumper, getting into a car is almost surely the riskiest thing you regularly do. The data bear this out. The National Highway Traffic Safety Administration (NHTSA) reports that in 2013, people in the United States drove 3 trillion vehicle miles and were involved in accidents causing 32,719 fatalities to occupants and non-occupants, or 1.09 deaths per 100 million vehicle miles driven. That is 300 times higher than the fatality rate in commercial air travel!
There are two ways to make car travel safer: improving driver behavior and improving technology. Both approaches can achieve gains. But the prospective gains from technology are far larger.
Starting with driver behavior, even seemingly irrelevant decisions can have a sizable impact on risk. When people look for a place to live, they frequently consider the cost of their commute both in time and money. Few consider that the risk of an accident declines as the distance traveled gets shorter.
How much does commuting risk fall if you choose to live five miles away from work rather than six? Since most people work about 250 days per year, a round-trip commute that is two miles shorter means driving 500 fewer miles per year. According to the NHTSA data, this will reduce the probability of a fatality from 1 in 30,400 to 1 in 36,500. So, if the heads of all 100 million U.S. households decided to cut their total work commute by two miles per day, the projected number of fatalities would decline by 547 persons a year, or 1.7%.
What about the role of technology? The federal government has been imposing safety standards for automobiles (including pickups, vans, and SUVs) since 1968, when it began requiring seat belts; since then, it has enacted dozens more. Some of these technologies, especially seat belts, had been voluntarily supplied to a limited extent since at least 1960.
What was the impact of these equipment standards and how much did they cost? An NHTSA study covering 1960 to 2012 estimates that these technologies alone saved an escalating number of lives over the period--an average of 11,575 per year, and 27,600 in 2012. When added to the 33,500 actual fatalities in 2012, the lives saved in that year represented a 45% reduction! An earlier study put the cost of these technologies in 2001 at less than $1,000 per vehicle.
Could technology do more? Suppose that everyone had a self-driving vehicle and that--estimating conservatively--this would cut the fatality rate in half. Relative to the 2013 levels, that would mean 16,000 fewer deaths per year and a rate of roughly one 1 fatality per 200 million vehicle miles--still 150 times riskier than flying, but much safer than it is now.
How much would we be willing to pay for that doubling of highway safety?
There are two parts to the answer: the value of fewer fatalities and the reduction of insurance premiums. Starting with the first, we need to compute the monetary benefit of saving 16,000 lives.
For many of us, each life is priceless. But when the government considers the imposition of safety targets, it must estimate the cost and balance that off against the prospective benefit. To do this, officials estimate the "economic value of a statistical life" by examining questions of the following form: How much would a typical person be willing to pay to eliminate a mortality risk that is 1 in 50,000? An answer of $200 implies that the expected value of reducing one fatality is $10 million. Consistent with estimates in academic studies, the Department of Transportation currently uses a value of $9.4 million.
Using that number, saving 16,000 lives per year is worth $150 billion. Because of the efficiencies that would come with self-driving cars--fewer people will need their own cars--let's assume that we need only about one per household in the country, or 100 million. That means the benefit from saving lives is $1,500 per household per year. Since the number of accidents is cut in half, we would expect every household to save half of the average annual $1,000 car insurance premium as well. Adding these together, we find that each of us should be willing to pay $2,000 per year more for a self-driving car than for a conventional one. If a car lasts for 10 years at a discount rate of 2% (adjusted for inflation), the present value of the benefit is about $18,000. That figure is nearly double the current estimates of the additional cost of technology to produce self-driving cars by 2025.
What this means is that a safe, self-driving version of the Ford Focus, which costs about $18,000 today, has an economic value of $36,000. So long as hackers can't take over these cars, owning or leasing one will be worth every penny.
An earlier version of this post appeared on our blog www.moneyandbanking.com.
This version benefited from Gregg Forte's comments.