The US economy is changing — and not always in the ways people expect. Fewer people are working in the manufacturing sector, yet we're producing more manufactured goods than ever. In many ways, the US economy is less dynamic than is commonly believed — the number of startups is dropping, people are changing jobs less often, and worker productivity is growing at its slowest pace in decades.
Meanwhile, American cities are enjoying a renaissance, with job growth and home prices soaring in the biggest cities. Read on for 27 charts that show the surprising evolution of the US economy.
A massive shift from goods to services
Yes, America still makes things
Declining manufacturing employment over the past 30 years has given a lot of people the impression that America's manufacturing sector is in decline. But that's actually wrong, as this chart shows. Since 1987, US manufacturers have increased their output by 80 percent at the same time as they have reduced their workforce by about 17 percent. In other words, American factories are about twice as efficient today as they were three decades ago. So we're producing more and more stuff, even as we use fewer and fewer people to do it.
Manufacturing employment is dwindling, while more and more people provide services
In 1965, 20.6 million people worked in industries that produced goods like clothing, cars, or airplanes. Now, 50 years later, there are a million fewer people working in these same industries — despite the fact that the US population is about 60 percent larger. Meanwhile, jobs in the service sector — like nursing, teaching, waiting tables, and selling real estate — have exploded, with more than three times as many people doing them today as in 1965.
Services keep getting more expensive while manufactured goods get cheaper
If you ask a middle-class American family about the biggest sources of financial stress in their lives, they're unlikely to name basics like food, clothing, or even transportation. Instead, American families worry about the soaring costs of college tuition, health insurance, and child care. And this isn't a coincidence — in fact, it's the flip side of the trends shown in the previous two charts. The US manufacturing sector has enjoyed big productivity gains, but the inherently labor-intensive service sector hasn't seen the same efficiency gains. As a result, the cost of these services — relative to the overall inflation rate — has zoomed upward, consuming more and more of our spending. The cost of housing has also risen, but for different reasons having to do with the scarcity of land and regulations limiting high-density development.
Image credit: Timothy B. Lee / Vox
A very long-term perspective on interest rates
The yield on a 10-year Treasury bond — that is, the interest rate the government pays when it borrows money for 10 years — is below 2 percent. Interest rates tend to move together, so this also means that the interest rate you pay to buy a house — as well as the interest rates you can get from a savings account — are also at record lows. That freaks out a lot of people because they remember the much higher interest rates of the late 20th century and worry that today's very low rates are a sign that something has gone badly wrong with the economy.
But taking a much longer view suggests a different interpretation. From the nation's founding until about 1950, interest rates were on a clear downward trend. This makes sense: As society gets richer, there's more money available to lend, and so the price to borrow money naturally goes down. From this perspective, the high interest rates of the 1970s, '80s, and '90s look like an aberration driven by the unusually high inflation of that era. And today's low rates look more like the resumption of a 200-year trend toward ever-lower borrowing costs. So don't panic — enjoy your 3.5 percent mortgage.
A disappointing recovery
A big economic mystery: What happened to the missing output?
Most recessions are followed by a period of faster-than-usual catch-up growth. This means that in the absence of a catastrophic event like a war or an epidemic, there's no reason for the US economy to suffer a big, permanent decline in output. Yet as this chart shows, that's exactly what happened in the wake of the 2008 financial crisis. Between mid-2007 and mid-2009, inflation-adjusted national output fell by 3.5 percent. Since the economy normally grows at about 3 percent per year (illustrated by the dotted red line), this means the economy fell about 9 percent behind its pre-2007 growth trend. And because the economy never experienced a period of rapid catch-up growth, that lost output appears to be gone for good.
Image credit: FRED
Productivity is growing at its slowest pace in decades
Economic progress ultimately occurs because we figure out ways to produce more stuff with every hour of work. In the long run, if workers don't become more productive, then they can't earn higher wages. So there's reason to worry about the fact that productivity per hour grew at just 1 percent per year between 2010 and 2015 — that's the slowest pace of the postwar period. If this slowdown continues in the coming years, it will mean much slower improvement in living standards than occurred during the 20th century.
Image credit: Javier Zarracina / Vox.com
Fears of out-of-control inflation have proven unfounded
From the 1960s through the 1990s, the main concern of the Federal Reserve, America's central bank, was to keep inflation rates down. The problem reached its apex with an inflation rate of 13 percent in 1980. So when the Fed cut interest rates to zero in 2008 — and then kept printing more money — a lot of people warned that inflation would come roaring back. But these inflation hawks were wrong. After eight years of near-zero interest rates, inflation has remained stuck below the Fed's 2 percent target. And while that might seem like something to celebrate, too little inflation can be a sign that the Federal Reserve isn't doing enough to support the economy.
America is getting older — and that matters
As life expectancy has grown and fertility rates have declined, the fraction of the population over age 65 has steadily increased. And thanks to the demographic bulge of the baby boom, the over-65 share is growing faster during the current decade than ever before. This has serious economic consequences. Research suggests that the drag from an aging population could be making a significant dent in the overall rate of economic growth. And of course it's also going to require adjustments in programs like Medicare and Social Security to allow a smaller share of workers to support a rapidly growing retired population.
Image credit: Rcragun / Wikimedia Commons
Don't blame robots for job losses
There's a lot of concern about the possibility of robots displacing human workers and causing mass unemployment. But this chart from the International Federation of Robotics helps to put those concerns into perspective. It shows that the global supply of robots in 2014 was 229,000. That's a big jump from the 178,000 robots sold in 2013. But in absolute terms, it's a tiny number in a world where workers number in the billions. Even if robot sales continue to grow at a rapid pace, it will take many years before there are enough robots to pose a significant competitive threat to human workers.
Image credit: International Federation of Robotics
Big changes in how we live and work
Average workers' wages have been falling far behind productivity growth
For the first few decades after World War II, productivity growth and wage growth went hand in hand. Workers produced more and more per hour, and they were rewarded with higher and higher wages. But in the 1970s, this tight connection between productivity and the wages of the median worker started to break down.
Workers have continued to get more productive over the past 40 years. But median compensation — which includes cash income as well as benefits like health insurance — hasn't kept up. Some of the gap may reflect technical issues with the way these figures are adjusted for inflation. And a large portion of the gap reflects growing inequality among workers. The highest-earning workers have seen their incomes rise along with productivity growth, which isn't reflected by median income figures. Still, it appears that the traditional link between productivity growth and wage growth is broken.
A college education keeps getting more valuable
Wage growth has been weak over the past 40 years, but the trend has been much worse for those without a college degree. In 1979, the typical worker with a college degree earned 38 percent more than a worker with just a high school diploma. By 2014, this college wage premium had risen to 78 percent.
People have been dropping out of the labor force since about 2000
The labor force participation rate measures the fraction of the civilian, non-institutionalized population, age 16 or over, that is either working or looking for work. At the peak of the dot-com boom, this figure reached 67 percent. That represented an all-time high — a high that we haven't come close to regaining over the past 15 years. Participation in the labor force declined during the 2001 recession and then declined even more dramatically between 2008 and 2010. The result: Labor force participation is now at its lowest level since the 1970s.
Men have been drifting out of the labor force for decades
That last fact — that labor force participation is at its lowest level since the 1970s — sounds pretty alarming. But it's helpful to look separately at labor force participation for men and women. The big increase in labor force participation we saw between 1965 and 2000 was driven by women entering the workforce. Meanwhile, the participation rate among men has been slowly but steadily declining for as long as the Labor Department has been keeping statistics. And over the past decade, women have started to behave the same way, slowly reducing their participation in the workforce. So the declining labor force participation of the past 15 years may simply be a natural consequence of a society that keeps getting wealthier.
People do less job hopping than you think
Lots of people think that America has undergone massive shift in its work culture. We used to devote our careers to a single company, the story goes, but now we're all free agents switching jobs every couple of years. But this chart, taken from an excellent report by the Employee Benefit Research Institute, shows that this story is greatly overstated. It is true that fewer men are spending their careers at a single company than they did a few decades ago. But few women have ever had that opportunity, and even among men it's always been atypical.
Image credit: Employee Benefit Research Institute, based on BLS data
The teen summer job is going out of style
As late as 1989, almost 60 percent of teens ages 16 to 19 had a summer job. But teenagers are working less, especially during the summer months. Teen summer employment fell to an all-time low of 31 percent in July 2011. It's rebounded a bit since then, to 36 percent, but there are still way fewer working teens than there were a generation ago.
This is partly because teens are working less year-round: The number of 16- to 19-year-olds working in February fell from 39 percent in 1994 to 28 percent in 2016. But the fall in July employment — from 54 percent in 1994 to 36 percent in 2016 — has been much bigger.
More and more women are having children before getting married
In 1970, a typical American woman would get married around age 21 and then have her first child a year or two later. Since then, women have been waiting longer for both of these milestones, but the change has been larger for marriage. As a result, the average woman in 2011 was having her first baby a year before getting hitched. Of course, these averages hide a lot of individual variation. In 1970, women without a high school diploma were already having kids slightly earlier, on average, than they got married — but the gap grew to five years by 2010. On the other hand, while college-educated women have been getting married later, the gap between marriage and childbirth has stayed about the same — at three years — over the past half-century.
Image credit: National Marriage Project
Married, two-earner families continue to enjoy income gains
The growing polarization of people's marriage patterns seems to be connected to greater economic polarization. Obviously, it's inherently difficult — financially and otherwise — to raise a child as a single parent. On the other hand, things have gone pretty well for married couples with both spouses working. Their earnings have continued to rise steadily over the past 50 years, surpassing $100,000 per year by 2015. Of course, the causation could run in both directions: Getting married could provide financial and emotional support that helps people succeed at work, while a challenging financial situation may make it more difficult for people to become and stay married.
Image credit: American Enterprise Institute
The urban recovery
The current recovery is an urban recovery
In the economic boom of the 1990s, most of the job gains came in less populated areas — like outer-ring suburbs and smaller cities. This seemed to make perfect sense because big cities were already densely populated, leaving little room for further development. But as this chart from the Economic Innovation Group shows, the current economic recovery has turned this logic on its head, with the biggest cities enjoying the strongest job gains.
Image credit: Javier Zarracina / Vox
Was the housing bubble really a bubble?
When the housing market crashed in 2008, a lot of people became interested in this data from Yale economist Robert Shiller. The index is computed by comparing sale prices of the same home at different times, providing an estimate of how housing costs have changed holding the quality and size of homes constant. It shows that for much of American history, home prices were fairly constant, with two big outliers. One was a big drop in housing prices between the world wars. The other was a 70 percent rise between 1996 and 2006. Between 2009 and 2012, a lot of people projected that home prices would return to the "normal" levels we saw between 1946 and 1996. But that didn't happen. Instead, average prices hit bottom in 2012 (still well above 1996 levels) and resumed their upward march.
Housing prices have grown a lot faster than construction costs
In a normal, competitive market, the price of a product should rise and fall along with the cost of producing that product. But this chart, created by White House adviser Jason Furman using research by Joseph Gyourko and Raven Molloy, shows that this relationship has broken down in the housing market. Since 1980, there's been little overall change in average, inflation-adjusted construction costs. Yet housing prices — adjusted for the size and quality of homes — have risen about 60 percent. This likely reflects the scarcity of land in desirable areas as well as housing regulations that have limited the amount of housing that can be constructed.
Image credit: Jason Furman
Apartment buildings are the new McMansions
The housing boom of the 2000s was mostly a boom in the construction of single-family homes — mostly at the edge of urban areas. But demand for these exurban housing developments crashed after 2007, and since then we've seen a big shift in housing construction. Today, we're building fewer than half as many new single-family homes per year as we did at the peak of the housing boom in 2006. On the other hand, construction of multi-family housing has fully rebounded to the levels of the 2000s, and could go even higher if demand for housing in high-density areas continues to grow.
Image credit: Jason Furman
Booms in smartphones and oil
People are now buying a lot more smartphones than PCs
The PC revolution that began in the 1980s was a big deal. But the smartphone revolution Apple launched with the iPhone in 2007 has the potential to dwarf the PC — at least if you measure by the number of users directly affected. PC adoption was always concentrated among more affluent users in richer countries. In contrast, the smartphone revolution is on track to be truly universal, with many poor people skipping PC ownership and getting a smartphone instead.
In the US, smartphones have already reached about 80 percent of the population, and globally they're expected to reach billions more over the coming decade. That has helped make Apple and Google two of the world's most valuable companies, and it's paved the way for smartphone-based services like Uber and Lyft.
Image credit: Benedict Evans / Andreessen Horowitz
America is becoming less and less dependent on foreign oil
Throughout the 1990s and early 2000s, America was importing more and more oil and other petroleum products, prompting concerns that the US was becoming "dependent" on foreign sources of oil. These concerns never made a ton of sense given that US allies like Canada, Mexico, and the United Kingdom are major oil producers — and because major oil exporters are as dependent on American money as we are on their oil. But in any event, technological advances such as fracking have dramatically reduced America's dependence on foreign oil over the past decade. America's net imports of crude oil and petroleum products declined from 12.5 million barrels per day in 2005 to 4.7 million in 2015.
Image credit: US Energy Information Administration
A changing investment climate
The way Americans retire is changing
In the mid-20th century, it was relatively common for employers to offer a defined benefit pension. Under this system, retired workers received a fixed monthly payment determined by factors like years of service and final salary at retirement. But in the 1980s, employers started to shift their employees toward "defined contribution" retirement plans like the 401(k). Under this system, employee savings are invested in stocks or bonds, and an employee's retirement income depends on how well these investments perform. The new system has some big advantages — it's more portable between employers, and employees don't have to worry about having their benefits reduced if a former employer goes out of business. But it also shifts a lot more risk and responsibility onto employees. If an employee fails to contribute consistently to his account — or chooses investment funds that have excessive risks or high fees — he could wind up without much of a nest egg during his retirement years.
Image credit: Employee Benefit Research Institute
The racial savings gap is growing
White workers earn more, on average, than black workers, and this chart shows the cumulative result of that earnings gap over the course of a career. It also shows that — despite laws banning discriminatory practices like mortgage redlining — the wealth gap between white and black families has increased over the past 30 years.
It's also important to note that things aren't so rosy even for many white families. Whites' average savings of $130,000 is only enough to generate retirement income of around $500 per month. And this average is skewed upward by a minority of white families with much more than $130,000 saved, putting them on track for a comfortable retirement. On the other hand, the median white family in 2013 had just $5,000 saved for retirement — meaning that half of white (and nonwhite) families have even less than that. Most of these families will be entirely dependent on Social Security for their retirement incomes.
Image credit: Urban Institute
The decline of the tech IPO
In the 1980s and '90s, the goal of most technology startups was to offer their shares on the public stock market. This process, known as an initial public offering (IPO), made it possible for anyone to participate in the technology boom by buying shares in companies like Apple, Microsoft, Amazon, and Yahoo. It allowed a startup's founders and early employees to cash out by selling some of their shares, and it made it easier for companies to raise money by issuing shares. But since the dot-com boom ended in 2000, companies have been waiting longer and longer to go public. Amazon went public in 1997, less than three years after its founding, when it was worth just $400 million. In contrast, Facebook waited until 2012 — eight years after it was founded — when it was already worth more than $100 billion. Uber has raised $11 billion from private sources and has no plans to go public. The result: Non-wealthy investors now rarely have the chance to invest in small tech startups the way they could before 2000.
Image credit: Benedict Evans, Andreessen Horowitz
Americans are starting fewer startups than ever
High-flying technology startups like Uber and Airbnb get a lot of attention in the media, but as this chart shows, startups are becoming more and more rare. Of course, most of these startups are not — and never have been — efforts to create the next billion-dollar technology company. Many are small-scale, local affairs like restaurants, auto repair shops, or IT consulting firms. And to some extent, the decline reflects slowing population growth — with more new people in general, there's less need for new businesses to serve them. Still, new firms are an important source of new ideas and innovations, so declining entrepreneurship could be contributing to America's growth slowdown.
Image credit: Inc
In moments like this — as people grapple to understand variants and vaccines, and kids head back to school — many outlets take their paywalls down. Vox’s content is always free, in part because of financial support from our readers. We’ve been covering the Covid-19 pandemic for more than a year and a half. From the beginning, our goal was to bring clarity to chaos. To empower people with the information they needed to stay safe. And we’re not stopping.
To our delight, you, our readers, helped us hit our goal of adding 2,500 financial contributions in September in just 9 days. So we’re setting a new goal: to add 4,500 contributions by the end of the month. Reader support helps keep our coverage free, and is a critical part of sustaining our resource-intensive work. Will you help us reach our goal by making a contribution to Vox with as little as $3?