Why Boomers Are a Boon to the U.S. Economy
The widely-used "dependency ratio" is inaccurate, this analyst says
(This article previously appeared on the PBS NewsHour website.)
Is America really about to suffer an economic crisis at the hands of its aging population? The magnitude of the coming demographic challenges — namely the prospect of a burgeoning retired population being economically dependent on a smaller workforce — depends on how you define “dependent,” says journalist and Yale Law School visiting lecturer Lincoln Caplan.
The standard measurement of that economic dependency comes from what’s called the “old-age dependency ratio.” It’s first and foremost a demographic tool, says Caplan, not an economic one, and its misuse as the latter has led to warnings about our nation’s fiscal viability, encouraging cuts to federal programs for children and the poor.
The following essay is an excerpt from Caplan’s original piece on the dependency ratio, first published as “The Fear Factor” — the cover story of the summer issue of The American Scholar, on whose editorial board Caplan serves. – Simone Pathe, PBS NewsHour Making Sen$e Editor
When the first wave of baby boomers turned 65 in 2011, the rest of the country started paying attention to what demographers had been warning about for a generation: The wave was turning gray and driving the United States toward fundamental shifts in social policy.
Demographers use a measure called the dependency ratio to prove their point. They add the number of Americans who are regarded as not in the workforce (traditionally, those 14 and younger and 65 and older) and divide that total by the number of people who are regarded as in the workforce (those 15 to 64). Then they multiply by 100.
An increase in the ratio is understood to mean a growing burden on each person in the workforce to support the economically dependent.
Misuse of the Dependency Ratio
The calculation is often modified to yield what’s known as the old-age dependency ratio, comparing the numbers of Americans 65 and older and those of working age. For the past four decades, that ratio has been relatively steady, but between 2010 and 2030, it is projected to shoot up from 13 to 22 — an increase of close to 60 percent.
The addition of “old-age” as the modifier to the starkness of the dividing line between the dependent and the productive bluntly separates older Americans from everyone else. It presents them as being unable to care for themselves.
With an objective formula, it seems to document that those who make up the wave of gray are a large and growing liability, undermining the country’s assets.
The old-age dependency ratio appears to justify the view of alarmists like former Federal Reserve Board Chairman Alan Greenspan, who testified before the Senate that this outsized group of aging Americans “makes our Social Security and Medicare programs unsustainable in the long run.”
How The Ratio Shapes Policy
The dependency ratio is only occasionally mentioned in debates about public policy, but its premise — that the growth in the ratio indicates how greatly baby boomers will burden the rest of society — is shaping some of the most consequential debates in the United States today: about the size of the federal government, about how government expenditures should be allocated, and about the nation’s financial viability in the next generation.
A demographic tool has become an economic one, treating a demographic challenge as both an economic crisis and a basis for pessimism justifying drastic reductions in bedrock government programs, including those supporting children and the poor. Even at state and local levels, the aging boomer demographic is repeatedly blamed for our economic difficulties.
That is a lamentable mistake. The United States has serious economic problems, and the aging population poses significant challenges, but those challenges are not the main cause of the problems. They should not be treated that way.
What the Dependency Ratio Doesn’t Say
The dependency ratio does not justify the solutions that the alarmists propose. Just as important, perhaps, it fails to account for the striking benefits accruing from the dramatic increase in life expectancy in the United States during the 20th century — what the MacArthur Foundation’s Research Network on an Aging Society called “one of the greatest cultural and scientific advances in our history.”
Although the concept of the dependency ratio dates back to Adam Smith’s 1776 book, The Wealth of Nations, remarkably there seems to be no published history of the concept as it is used today.
The inclusion of young people over the age of 14 in the productive segment reflected in the traditional ratio suggests that it was developed in the 19th century, when America’s farm economy still required their help. But as late as 1933, in Recent Social Trends, (a vast and definitive statistical portrait of the United States in that era), the dependency ratio was not referred to or used.
By the 1940s, however, the ratio became a regular tool among the different measures the government employed to describe the state of the nation and to project what it would look like in the future. Although no group was spared in the Great Depression, older Americans were especially hurt.
Social Security was the first national program to ensure that Americans 65 and older would have at least a minimum of income to pay for food, clothing and shelter. That program, which became law in 1935, started to make monthly payments in 1940. Social Security’s large scale required the government to anticipate how many Americans it would cover, which made the dependency ratio, and the old-age ratio in particular, an important demographic measure.
A Somewhat Better Yardstick
The U.S. government has been calculating what it calls the economic dependency ratio, which includes in the productive segment people who are still working after the presumed retirement age of 65. In 2010, it was 22 percent of men that age and older and 14 percent of women and, by 2020, it is projected to be 27 percent of men and 19 percent of women.
Even that ratio is crude, but it is less crude and more accurate than the traditional old-age dependency ratio.
Still, the revised calculation does not address fundamental concerns that scholars have raised in the recent past. They have criticized the dependency ratio for its blatant oversimplification of reality and for its ideological bias.
In 1986, sociologists Toni M. Calasanti and Alessandro Bonanno described the bias as the “social construction of the elderly’s obsolescence.” They meant that our society has chosen to regard older people as a burden when age alone does not make them so.
Age is accompanied by decline, but different people decline in different ways. A person’s race, ethnicity, wealth, and level of education are often better predictors of that decline than age.
Sociologist Donald E. Gibson wrote in 1989 that “it has become commonplace to predict or assume that demographic trends will lead to an economic crisis in the third or fourth decade of the next century.” He went on: “All current versions of the dependency ratio, however, share one important deficiency. They fail to take account of economic productivity, and how improvements in productivity lead to progress through increases in income and in “the country’s capacity to support people.”
Why There’s No Aging Crisis
Projections in the 1980s of an aging crisis rested on the assumption that, in the subsequent half century, the American economy would perform poorly and produce little improvement in real income.
In fact, between then and now, real income has grown somewhat more than was predicted, with dramatically unequal distribution. (The rise in real income for nine out of 10 Americans has been slight. The rise for the top 10 percent has been much higher — and higher still for the top one percent.)
Gibson’s point was more fundamental, however. “Very often this assumption is not explicated,” he wrote. “Not to do so is to present an economic problem as a demographic problem.”
In a similar argument last year, economists Ronald Lee at The University of California-Berkeley and Andrew Mason at the University of Hawaii criticized the dependency ratio for being “incomplete and misleading” and for exaggerating “the adverse impact on the macro-economy of population aging,” because it does not reflect that in the United States the elderly “rely heavily” on income from their own private wealth to support them — in economic terms, to pay for their consumption.
In general, Lee and Mason reported, “Net transfers from the working age population, mostly through the public sector in the form of Social Security benefits, Medicare, and Medicaid, make up only about 40 percent or less of funding for consumption.”
What Will Really Happen in America
As a result, the aging of the U.S. population will mean an increase in the number of older people who are only partially dependent on the government, not wholly dependent, as the dependency ratio assumes.
By contrast, Lee and Mason assume that the increasing number of elderly boomers in the United States will arrive at old age having accumulated assets similar in amount on average to those who are currently elderly. More older people per capita will mean more assets per capita. These assets will generate national income, boost productivity, and contribute to a future justifying a more optimistic outlook rather than the pessimistic one that the ratio is used to justify.
In 2010, a respected international team published a study called Remeasuring Aging in the September 2010 issue of Science magazine finding that old age generally arrives later than the dependency ratio assumes — if old age is defined as the point at which older people need permanent care, that is, when they are disabled.
The demographers Warren C. Sanderson and Sergei Scherbov wrote in the same study in Science magazine, “Alternative measures that account for life-expectancy changes” — improvements in health and longevity — “show slower rates of aging than their conventional counterparts,” based on “fixed chronological ages.”
They wrote that chronological age is less useful than life expectancies in predicting national health costs because “most of those costs occur in the last few years of life.”
The True Cost of Caring for the Disabled
Sanderson and Scherbov developed a measure they called the adult disability dependency ratio, defined as the number of adults 20 and over with disabilities, divided by the number of adults 20 and over without them. In the United States, this measure will likely remain flat for the next generation, meaning that the cost of caring for the disabled is not likely to skyrocket as a result of a major increase in the number of disabled people.
John Shoven, a Stanford economist, takes that idea a step further: in a scholarly paper called New Age Thinking, he argues that age should be defined differently from the universal convention of years since birth.
“The measurement of age with different measures is not like choosing between measuring temperature on a Fahrenheit or Centigrade scale,” he warned. The reason to change how age is measured is that the connection between the universal definition of age and the alternatives he proposes is constantly changing. Because of advances in nutrition, sanitation, and other factors, as well as health care, someone who has lived a long time is no longer as old as his or her numerical age once indicated.
A man born in 1900 was expected to live until he was 51 ½ and had less than a 50 percent chance of living until he reached 65. A man born in 2000 is expected to live until he is 80 and has an 86 percent chance of reaching 65. That dramatic advance in longevity indicates that knowing how many years a person has been alive tells only so much about the person’s risk of dying.
We’re Measuring Age In the Wrong Direction
Shoven proposes that instead of measuring age backward, as in years since birth, we measure it forward, as in years until projected death.
One option is to measure age by mortality risk. A 51-year-old man in 1970 had the same mortality risk (a one percent chance that he would die) as a 58-year-old man in 2000: in one generation, longevity advanced by seven years for that level of risk.
Another option is to measure age by remaining life expectancy, a more accessible measure because it is computed in years rather than as a percentage. In 1900, a man who reached 65 had a remaining life expectancy of about 13 years. In 2000, a man who reached 65 had a life expectancy of about 21 years.
Measuring backward yields starkly different results from measuring forward.
“Consider two alternative definitions of who is elderly in the population,” Shoven writes, “those who are currently 65 or older and those who have a mortality rate of 1.5 percent or worse.” In 2007, when he wrote this paper, the two definitions were equal: the average mortality rate was 1.5 percent or worse for 65-year-olds.
According to the U.S. Census, the population of those who are 65 or older will increase from about 12.5 percent of the population in 2035 to about 20.5 percent in 2050. But “the percent of the population with mortality risks higher than 1.5 percent (currently also 12.5 percent of the population) never gets above 16.5 percent,” because of what James Fries of the Stanford School of Medicine called “the compression of morbidity”— the tendency of illnesses to occur during a short period before death if the first serious illness can be postponed. That number “is projected to be just slightly below 15 percent and declining by 2050.”
By the conventional measure of years since birth, the population considered elderly is expected to grow by 64 percent. By Shoven’s measure, on the other hand, it is expected to grow by just 32 percent.
“The point,” he says, “is the great aging of our society is partly a straightforward consequence of how we measure age.”
Lincoln Caplan is a senior advisor to Encore.org and the Truman Capote Visiting Lecturer in Law at Yale Law School. He has been a journalist with and contributor to, among others: The New Yorker, The New York Times and U.S. News and World Report.