clock menu more-arrow no yes
Critical care nurse Emily Boucher takes care of Covid-19 patient Hannah Church in Abingdon, Virginia, on June 16.
Katherine Frey/The Washington Post via Getty Images

Covid-19 isn’t the reason that US life expectancy is stagnating

Don’t dwell on the 2020 dip. Worry about the long-term trend.

US life expectancy has declined 1.5 years because of Covid-19, to 77.3 years, the Centers for Disease Control and Prevention (CDC) said on July 21. This is a sobering statistic. But if it makes you worry that your life (or your children’s lives) will be 1.5 years shorter, you can rest easy. As a demographer, I can assure you that’s not what the CDC is saying, and the Covid-19 dip in life expectancy is less surprising and less important than many people might think.

Despite its misleading name, life expectancy does not predict how long anyone should expect to live. Life expectancy is a quick but incomplete measure of health, like gross domestic product for the economy or batting averages for baseball players. Like those numbers, its value does not come from predicting the future, but from explaining the past. It provides a way to track trends over time. In this case, it quantifies what we already knew: The US experienced a lot of deaths last year, more than any other year in recent memory. Another thing we already knew is that Black and Hispanic communities, which experienced a three-year decline in life expectancy, were especially hard hit.

Some news outlets have gotten this wrong. The Associated Press, for example, defined life expectancy as “an estimate of the average number of years a baby born in a given year might expect to live.”

The exact definition is a little more complicated. Life expectancy, the CDC report says, “represents the average number of years a group of infants would live if they were to experience throughout life the age-specific death rates prevailing during a specified period.” The portion I’ve italicized is the important part. It assumes that newborns will face the same health risks, throughout their lives, as people who were adults in the unprecedented year of 2020.

Christina Animashaun/Vox

To be clear, any drop in life expectancy is a bad thing. But this one simply confirms Covid-19’s profound effect on US society, rather than telling us something new. It’s essentially a mathematical way of restating that millions of people have lost loved ones to the coronavirus. Unless the Covid-19 death toll remains in the hundreds of thousands for years to come, we should not expect the decline to be permanent. If people continue to get vaccinated before it mutates more, life expectancy will bounce back to pre-pandemic levels.

What Americans should worry about is the longer-term trend. US life expectancy has been stagnant for the past decade, and had actually declined by 0.1 years before the pandemic started, from 2014 to 2019. Though smaller in magnitude, that drift downward says something much more worrying about health in the US. Our national health has not improved for over a decade, despite the trillions that Americans spend on health care every year.

We won’t know the actual life expectancy of today’s babies until at least 2110

To understand some of the surprising lessons of life expectancy, it’s helpful to consider what the concept does and doesn’t measure. A lot of people think that it tells you how long a child born in a given year can expect to live. Isn’t that literally what “life expectancy” means? Well, not exactly.

Consider the case of babies who were born during 2020, a year when Covid-19 killed hundreds of thousands of people in the US and became the third leading cause of death, after heart disease and cancer. The pandemic was bad news for everyone who was already alive, especially older people who are at higher risk of severe disease. But that doesn’t automatically mean 2020 babies are going to live shorter lives — especially if the world gets the pandemic under control while they’re young.

The only way to know the precise life expectancy of 2020 babies is to wait until most of them die. In 90 or 100 years, between 2110 and 2120, demographers will add up all the years lived by 2020 babies and divide them by the total number of 2020 babies. That’s not very helpful right now.

What life expectancy is and is not

The CDC calculated life expectancy from a simulation of deaths based on what occurred last year. To calculate the 2020 figure, researchers created a fictional cohort of 100,000 babies. They counted how many would live to their first birthday, based on the proportion of last year’s newborns who lived to their first birthday.

Then they did the same thing for every other age, again based on last year’s probabilities. At the end of the exercise, they added up all the simulated years lived by the 100,000 simulated babies. After dividing by the 100,000 simulated babies, researchers got the life expectancy the CDC reported in July.

Given that life expectancy relies on past probabilities of death, it shouldn’t be treated as a projection or model of the future. The calculation doesn’t predict what factors will contribute to deaths. It doesn’t factor in future pandemics or potential medical advances. Life expectancy just summarizes what has already happened.

One might ask why anyone uses such a complicated system. The answer is that unlike other measures of health, like death rates, life expectancy accounts for the specific probabilities of dying at each age. If you don’t do that, your calculations can lead to absurd conclusions. For example, Japan, a country legendary for its longevity, has a higher death rate than the US. Why? Three in 10 adults in Japan are 65 or older, a proportion twice as high as in the US. While Japan may suffer more deaths as a share of its population, the fact that its people have lived such long lives is evidence of its better health.

Since life expectancy accounts for age, it empowers demographers to compare populations across time and geography.

The Covid-19 life expectancy dip tells us what we already knew

Because life expectancy projections are historical, the Covid-19 drop really just confirmed what news stories have been saying for a long time: The US experienced a lot of death last year. It would have been shocking if the number did not fall, given that more than 610,000 US residents have died of the disease.

To put the decline in context, the CDC reported that last year’s decline was the largest that the US has experienced since 1943, in the midst of World War II. But this comparison isn’t perfect. Back in 1943, life expectancy was increasing steadily, and the dip did not interfere with an overall positive trend.

The difference in 2020 is that US life expectancy wasn’t increasing before the pandemic: It was slowly falling. Pre-pandemic life expectancy in 2019 was 78.8 years, compared to 78.9 years in 2014.

Christina Animashaun/Vox

That decline, though much smaller than the 2020 dip, signals something much more ominous about how difficult life has become in the US. Lives have been cut short by addiction and suicide, especially among men — what Princeton economists Anne Case and Angus Deaton call “deaths of despair” — and by long-term, preventable diseases, such as heart disease and kidney failure, especially among women.

The stagnation in life expectancy isn’t due to some natural limit of human lifespans. In 2019, life expectancy was 84.4 in Japan, 83 in France, and 81 in the United Kingdom and Germany. The US, with its life expectancy of 78.8 years, was already lagging before the pandemic.

The US can improve life expectancy by going back to basics

We now have extremely effective tools to reduce Covid-19-related deaths, vaccines chief among them. But the longer-term stagnation and decline in US health can’t be solved with vaccines alone. In fact, it is likely that many solutions need to come from outside of the health care system entirely.

The relatively poor health of the US is rooted in “fundamental causes,” according to epidemiologists Bruce Link and Jo Phelan. These are the social conditions like economic inequality and racial segregation that worsen some illnesses and reduce access to health care. In the US, solutions could also include policies that replace jobs in towns and cities that have been hollowed out by globalization and deindustrialization. The dignity of meaningful work can improve health.

Of course, we should not ignore the gains that can be made within medicine. I don’t mean high-profile technological advances that will make headlines or boost the bottom line of new biotech startups. I mean routine and preventive care that can detect disease early, help get patients into treatment, and provide a trusted source of medical advice.

Rather than wringing our hands about the Covid-19 life-expectancy dip, the US should be passing laws and expanding programs that draw medical workers into primary and preventive care, not least by paying them more. This is especially true in rural areas with aging populations and a shortage of doctors. Training more Black doctors, especially in obstetrics and gynecology, may lead to dramatic improvements in the shamefully bad maternal health outcomes among Black women in the US.

By focusing on one historical measure of years lost to the pandemic, we run the risk of dwelling on what we can’t change and ignoring what we can improve. If you want the next generation to live longer and healthier lives, one of the best things you can do is push for economic and health care policies that reduce economic and racial inequality, and help ensure that every person has access to the kind of world-class, routine health care that saves lives. Let’s give the demographers of 2110 something to celebrate.

Michael Bader is an associate professor of sociology and policy and the associate director of the Metropolitan Policy Center at American University.

Vox Conversations

Elon Musk’s imaginary world

Explainers

The US needs a clear Covid-19 goal now more than ever

Education

Biden’s pre-K plan might not be as “universal” as he hopes

View all stories in Politics & Policy

Sign up for the newsletter Sign up for The Weeds

Get our essential policy newsletter delivered Fridays.