Consider two Americans, one born in 1824 and the other in 1924. Both are fortunate to enjoy a 100-year lifeFootnote 1 in a small or midsize city. Over the course of the first American’s lifespan, it is likely that she would see her independence increase on at least one dimension: her practical ability to access valued destinations.Footnote 2 The advent of steam locomotives and steamboats made long-distance travel feasible, and horsecars and rail did the same for local travel. Roads became more common and improved dramatically in connectivity and quality.Footnote 3 As the twilight of her life approached, the automobile was becoming a mass-market product that opened up still other avenues of possibility. In her nineties, she likely could not take full advantage of this innovation, but she didn’t need to. All the other options still existed. In very old age, she continued to enjoy high levels of independence for as long as she could board a streetcar.
In terms of independence during the golden years, the American born in 1924 isn’t so lucky. As she progressed from early adulthood to middle age, cities dismantled their streetcar networks, shrank their sidewalks to widen roads, and routed highways right into the heart of their downtowns, dividing communities and gutting Main Streets in the process. Viewed narrowly, this might not have seemed all bad: In her middle years, it perhaps improved her independence on net, if she was in the fortunate majority able to take advantage of the car’s promise in that era even as other forms of transportation languished or were discontinued. But by the time she reached her eighties, when her ability to drive safely was deteriorating, reliable rapid transit no longer existed. In her final two decades, the fulfillment of her basic needs – buying groceries, accessing medical care, seeing family – became increasingly laborious, expensive, or infeasible. The ability to travel by air did little to offset these daily difficulties. Progress in technology, medicine, civil rights, and the economy left her much better off overall than her counterpart a century earlier, but her independence diminished more in old age than it would have in a different transportation paradigm. The driving life expectancy of Americans – the period during which they are able to drive safely – is about seven to ten years shorter than their life expectancy,Footnote 4 and for those who live 100-year lives, this gap is far larger.
In the US, the natural challenges of aging are magnified by a land-use and transportation system that compels reliance on the car. Alongside the rise of mass motoring in the middle third of the twentieth century, regulations that inhibit the growth of population density – zoning, parking requirements, minimum lot sizes, maximum lot coverage ratios, and so forth – came to dominate at the local level across the country. These ensured that different uses of land, like housing, schools, grocery stores, and job centers, were more separated from one another than they had been when their location was dictated by the usual pressures of supply and demand, and that they were arranged along a road network that required a car to navigate. This generates a variety of challenges for society as a whole, ranging from climate change to traffic deaths, as well as for particular groups. Older Americans are one such group: Many cannot operate a vehicle safely but could ride a bus, light rail, or train if a reliable and accessible network existed.
These policies, and the laws that implement them,Footnote 5 create two barriers to realizing the promise of the 100-year life. The first is that they shorten Americans’ lifespans. Compared to their peers in other high-income countries, the life expectancy of Americans is brief. Paradoxically, however, boosting American longevity would aggravate a second problem: In their golden years, Americans enjoy less independence than their peers. These problems cannot be reduced to a single cause, but they do share a common, vexing, and perhaps unexpected input: a built environment that requires driving as the price of first-class citizenship.
Addressing the bundle of laws and policies that require universal automobility is a colossal undertaking. Viewing these matters through the narrower aperture of the 100-year life, however, suggests cause for optimism: Perhaps more than in any other area, policy governing transportation can both increase life expectancy and broadly enhance quality of life in old age.
Section 20.1 considers the impacts of America’s automobile-first transportation system on the US’s life expectancy. It urges the adoption of proven, low-cost, and high-value automobile and roadway safety interventions in response. Section 20.2, which addresses quality of life in the golden years, proposes efforts to catalyze a competitive market to attract older Americans to locations that enhance their independence.
The concept of a competitive locational marketplace is well grounded in urban economics, but its potential as a solution to the challenges of sustaining independence in old age, described in Section 20.2, is relatively underexplored. The basic problem is that those fortunate to enjoy long lifespans can expect to outlive their driving life expectancy – the age when they can no longer drive safely – yet, outside of a handful of costly cities, America’s automobile paradigm offers them no viable alternative.Footnote 6 Section 20.2 makes the case for differentiated forms of living arrangements in old age that increase independence – the ability to access valued destinations – in diverse ways. Specifically, drawing on the stylized consumer-voter model of residential sorting developed by Charles Tiebout,Footnote 7 it observes that local governments can (and to an extent already do) join forces with developers to compete for retiree population share, allowing seniors to sort into communities that foster greater access to valued amenities in their golden years. This decentralized, instrumental approach will not appeal to everyone, but has the potential, through competition, to generate benefits beyond the places where it is embraced directly.
20.1 Car Crashes and Life Expectancy
Unintentional injuries – accidents, colloquially – are the leading cause of premature death for Americans aged one to forty-four.Footnote 8 In 2021, they accounted for more such deaths than the other top five causes combined.Footnote 9 In 2020 – a year when the COVID-19 pandemic caused approximately 350,000 deaths in the USFootnote 10 – unintentional injuries, which disproportionately claim young victims, cost more potential years of life than any other cause.Footnote 11 Among such accidental deaths, traffic fatalities are either the leading or second leading cause of death for all age groups above one year.Footnote 12 Because of their intensely negative impact on lifespan, these early-in-life deaths have a disproportionate impact on life expectancy in the US.
20.1.1 Car-Crash Deaths Reduce US Life Expectancy
Life expectancy represents a prediction about how many years of life a given person has remaining from a given age (at birth, at age eighteen, etc.).Footnote 13 A focus on life expectancy brings out the distinction between, say, drug overdoses and heart attacks: both cut lifespan, but the former cause of death, which skews young, reduces life expectancy more than the latter.
Life expectancy is positively associated with national income: Those who live in high-income countries can generally expect to live longer than those in middle- and low-income countries.Footnote 14 However, the US is a partial exception to this rule. While average life expectancy in the US was once comparable to that in Germany and the UK, beginning in the 1990s the US began to lag behind its peers.Footnote 15 In 2019, the last full year that excludes the effects of the COVID-19 pandemic, Americans’ life expectancy at birth was just shy of seventy-nine years, lower by over three years than a peer group of twenty-one “high-income advanced democracies”Footnote 16 despite the fact that the US is far wealthier on all or nearly all economic measures than each of those countries. This gap – long a focus of demographers and public health scholars – both widened notably and garnered increased attention during the pandemic.Footnote 17 It has since reverted somewhat as the virus has waned,Footnote 18 but this only returns the US to roughly the same uncompetitive position that it previously occupied.
Explanations for American weakness in life expectancy abound. Commonly identified causes include infant mortality, roadway mortality, gun homicides and suicides, drug overdoses, and cardiac events – all of which are more common or more deadly in the US.Footnote 19 Several of these claim young lives disproportionately, causing the US to experience a very high number of potential years of life lost (PYLL)Footnote 20 compared to other industrialized countries.Footnote 21 Indeed, among members of the Organisation for Economic Co-operation and Development (OECD), plus a few other middle-income countries, the US suffered one of the highest levels of PYLL; only countries that have a far lower national income, such as Mexico and Latvia, perform worse.Footnote 22 In Canada and Australia, for example, premature death erases half as much or less early-life potential as compared to the US.Footnote 23 Road traffic deaths, as well as gun violence and drug overdoses, contribute substantially to America’s distinctively low average life expectancy.
It is worth unpacking why the roadway mortality contribution to US life expectancy is so negative relative to that of America’s peers. For starters, a high number of Americans – on average, about 40,000 a year – die on roadways. On a per-population basis, the US has more than twice the number of deaths as the European Union.Footnote 24 These deaths are overrepresented at the low end of the age distribution. The situation is deteriorating in the US, especially for certain subgroups; during the 2010s, for example, deaths of pedestrians in peer nations dropped 20 percent on average, while in the US they increased 46 percent.Footnote 25
One causal inference conundrum is that deaths per capita from a given activity are to an extent endogenous with the pervasiveness and intensity of the activity. Accordingly, researchers also sometimes consider deaths as a function of driving intensity, reported as vehicle miles traveled (VMT) per capita. Each year, by that measure, Americans travel in cars roughly two to four times the distance that residents of other high-income OECD countries do,Footnote 26 so – all else equal – one would expect more Americans to die in car crashes than their counterparts in peer countries. Historically this relationship explained a high percentage of elevated US roadway mortality (and, by extension, some of the life expectancy gap), but in this century the relationship has weakened somewhat as peers have cut deaths drastically while the US’s VMT has grown only modestly and its traffic deaths remained elevated. In a recent three-year window, America’s rate of per-capita deaths per VMT was 300 percent that of Norway, more than 200 percent that of the UK, Ireland, Switzerland, and Sweden, and about 165 percent that of Germany.Footnote 27 In other words, Americans do not simply die more on the roads because they drive more. The US road transportation system – a system that includes drivers, vehicles, and the roads themselves – claims more lives even on a VMT per-capita basis. The disproportionately young skew of roadway deaths converts this into a disproportionately high volume of PYLL.
20.1.2 Maximizing the Value of Road Safety Interventions for Life Expectancy
Organized national efforts to improve road safety in the US are at least a century old,Footnote 28 but success remains elusive for a variety of reasons, many of them legal and institutional.Footnote 29 A small package of high-value solutions would go a long way to reducing the drag that roadway mortality exerts on American life expectancy.
Especially if the overall objective is to add years to US life expectancy, a major goal of safety measures must be to increase seatbelt use. In 2020, nearly half of all fatalities among US vehicle occupants were of people who were not wearing seat belts, and among the young these types of deaths are particularly prevalent.Footnote 30 After years of progress, restraint use has stalled, and a mix of enforcement (switching from secondary to primary enforcement, for example),Footnote 31 education, technology, and insurance incentives is probably needed to make further headway.
After decades of emphasizing personal responsibility as a route to improved safety, transportation agencies have in the past decade settled on a “safe system” approach that urges safer road designs – narrower or fewer vehicle lanes, protected bike lanes, traffic circles instead of conventional intersections – as a mechanism for improving outcomes. This approach, while important, is incomplete, and not only because physical design change is costly and requires navigating NIMBY (“not in my backyard”) roadblocks:Footnote 32 It ignores the behavior of outliers – especially reckless drivers. It also overlooks the opportunity costs of forgoing incremental targeted changes in favor of wholesale physical transformation of the street. Given limited budgets and the imperative to save lives quickly, planners should reject design essentialism in favor of interventions that maximize the expected value of life-years saved per unit of time and money spent.
Speeding, for example, is associated with nearly one-third of traffic fatalities, about the same share as alcohol involvement.Footnote 33 While the latter generates considerable regulatory attention, speeding and other reckless behaviors do not attract the same opprobrium or official response. Most Americans don’t drive double the speed limit or run red lights regularly, but there is reason to believe that the small minority who do are responsible for an outsize share of traffic fatalities. Their conduct should be a special focus of regulation.
Despite the obvious dangers connected to speed, populist opposition to enforcement of the speed laws is as old as speeding.Footnote 34 Creativity is needed to help defuse or sidestep this. Drivers could be required each year to buy a modestly priced bond (say $100) that is refundable conditional on an absence of grave speeding violations. (They could even be paid an above-market interest rate, subsidized by forfeited bonds.) Or a portion of the revenue from speeding fines could be awarded to nonspeeding drivers – an experiment in Sweden doing just that yielded promising results by encouraging motorists to see safe driving as a possible economic windfall to them. Automated enforcement of speed limits and red lights, which have been shown to be effective at reducing serious crashes, should also be increased. Because they will help give back large numbers of life-years, prioritizing these and other reforms that reduce roadway mortality will have a disproportionately positive impact on life expectancy and open up the 100-year life to more Americans.
20.2 “Gray Tiebout Sorting”: A Transportation-Based Retiree Location Market
Although roadway fatalities and other causes of premature death restrict access to the 100-year life for many Americans, millions do enjoy extended longevity. Some projections suggest their ranks will swell in the coming years. Even among this fortunate group, however, many will not enjoy the full promise of their golden years because, nearly everywhere in the US, the built environment requires driving during a stage of life when driving is no longer safe. In fact, the season of life when those who live to 100 still have most of their faculties but can no longer drive safely is likely quite long – perhaps twenty years or more. To the extent that life expectancy climbs, the duration of this period will too.Footnote 35
At present, the needs of Americans who outlive their ability to safely drive are underserved. A recent study concluded that “up to 60% of older adult drivers with mild cognitive impairment, and up to 30% with dementia, continue to drive,”Footnote 36 and fatal crash rates on a VMT basis “increase noticeably starting at age 70–74 and are highest among drivers 85 and older.”Footnote 37 Because a fatal crash in old age does not take as many life-years as it does in youth, increasing older-driver safety, while imperative, will yield fewer dividends in terms of life expectancy. However, the prospect of extending and enhancing their independence will appeal to many seniors. Some would presumably be interested in ways of preserving their ability to access valued destinations later in life.
Recognition by local governments and developers of the benefits of increasing personal independence for seniors offers a way forward. This can mean many things. States like Florida, for example, famously offer master-planned retirement communitiesFootnote 38 that cater to the demands of retirees, including for independence.Footnote 39 Other localities and developers provide late-in-life independence through walkable dense areas, golf carts and shuttles (perhaps autonomous ones in the futureFootnote 40), and hyperlocal public or club amenities. So-called naturally occurring retirement communities in college towns, small cities, and some big-city neighborhoods provide access to a cluster of amenities without requiring a high degree of mobility. For municipalities, states, and developers, the financial benefits of attracting retirees are great, so in principle the incentives are already in place for what might be called transportation-based “gray Tiebout sorting.”
The standard urban economics model of voting for local government services with one’s feet is identified with an early exponent, Charles Tiebout.Footnote 41 Tiebout sorting posits the existence of a location market where individuals “move to the local government that best fits their preferences for public policies” through its particular mix of taxation and services.Footnote 42 Local governments offer different products – different bundles of services at different prices – that appeal to different location-consumers.Footnote 43 Locations that provide consistent access to desirable amenities in old age could (and to some extent already do) compete for a bigger slice of the retiree population pie. Although it targets movers, standard economic theory predicts that on the margin this process also benefits the great many retirees who prefer to stay put by exposing their localities to competition.
When it comes to transportation and land use, the challenge is not merely one of mobility – physical movement across space – but rather one known in the urban planning literature as accessibility,Footnote 44 which refers to connecting people with valued destinations. In the context of aging, this would allow them to continue to enjoy high quality of life and independence. These are high priorities for many retirees. As one indication, the AARP has taken up the cause of rating localities for their ability to “provide safe, walkable streets; age-friendly housing and transportation options; access to needed services; and opportunities for residents of all ages to participate in community life” and introduced a Livability IndexFootnote 45 to facilitate what might be called gray transportation-based Tiebout sorting.
The variety of ways to operationalize an independent lifestyle – mixed-use development, subsidized private shuttles, public transportation, greater access to delivery services, and so on – underscores the promise of the market metaphor. So does the potential of increased fiscal capacity for local governments and profits for private developers: For these players, retirees are in many ways an ideal group. They have assets and bring revenue in the form of property, consumption, and income taxesFootnote 46 but tend to underconsume widely subscribed local resources relative to their numbers. They do not attend or send their children to public schools. Meanwhile the cost of the services that they draw on disproportionately – physicians and hospitals, assisted living facilities – is borne primarily by the users themselves or by third-party payors like the federal government or private insurers. Others, like public libraries and parks, are provided at a relatively fixed cost such that increasing their use intensity is fiscally neutral.
Independence is only one component of the retiree location decision.Footnote 47 There is reason to believe, however, that seniors place an unusually high value on sustaining independent mobility in their final years (or, increasingly, decades) – and that doing so is necessary for maintaining access to essentials like medical appointments and food. Making changes to the transportation–land use bundle requires reform of both law and policy,Footnote 48 which runs into a thicket of Catch-22s. It is hard to make the case for public transit in an area that lacks the density to support it, but it is hard to grow the necessary density in an area that outlaws it through restrictive zoning or excessive parking requirements.Footnote 49 The market for retiree location offers a plausible mechanism to help resolve this.
Economists have observed that the quality of local amenities explains some of the housing price premium in high-amenity areas over and above differences in the opportunity to increase earnings.Footnote 50 A market for retiree location might be able to harness the same types of forces that render certain neighborhoods desirable to a particular demographic cohort. Just as some areas appeal to young professionals or families, others appeal to retirees. A deep market that offers a variety of ways of providing older residents ready access to amenities could stimulate intralocal or within-metropolitan competition for retiree market share. It might even be able to do this to a degree that generates spillover benefits to the general population. A catalyst of this sort might help areas that have struggled to overcome the inertia of the twentieth-century automobile planning paradigm to up their game.
A retiree location market already exists, and growing attention to longevity can provide a framework within which it can be cultivated further by profit-driven and public-minded forces alike. Not many places offer meaningful alternatives to automobility, and even doing so within a concentrated area requires combating decades of entrenched policy.Footnote 51 This is what makes the promise of the retiree location market intriguing: the potential for an efficient alignment of incentives. A well-defined cohort and geographic area lend themselves to narrower matching of services to needs. In the simplest case, a single owner, such as a private developer, stands to internalize the benefits of investment in accessibility-driven independence. In a more typical case, a small number of owners, some private and others public, may be able to coalesce around a common goal of improving accessibility within a given area. Many ingredients of success are present.
20.3 Conclusion
Broadly, extended longevity poses two policy challenges: making it feasible in the first place and enabling those who achieve it to thrive. Transportation policy is central to meeting both challenges.
In order to increase access to the 100-year life, it should be the policy of the US at all levels of government to grow life expectancy. There are many worthy ways to do that; this chapter emphasizes road safety interventions that target high-mortality crash factors – including failure to wear seat belts (associated with about half of crash deaths), speeding (about a third), and drunk driving (same). Each of these factors claims a disproportionate number of young lives, which means improving outcomes will reap outsize gains in life expectancy. They can also be rolled out in relatively quick and low-cost ways, using proven and experimental interventions that maximize gains in life expectancy by targeting those roadway behaviors that rob Americans of the most life-years. This focus may sound like common sense, but it rejects the fad of design essentialism that has swept roadway reformers in the past decade.
At the other end of the age distribution, a focus on helping seniors maintain access to valued destinations can enhance quality of life in the golden years. If local governments and property developers approach hyperlocal goods with the challenge of golden-years mobility in mind, they can stimulate transportation-based gray Tiebout sorting: residential self-selection by older Americans into locations that maximize mobility and independence. This, in turn, can create pressure on other places to increase their offerings. Thus, while skepticism is warranted about the potential for transportation-based Tiebout sorting in general, there are good reasons to expect one aspect of it – sustained independence past the age when one can drive safely – to hold more promise for some retirees.