America is in the midst of an extraordinary startup boom

Although America has a deserved reputation as a country at the cutting-edge of innovation, fuelled by entrepreneurial vim, in recent years some economists have worried that this reputation no longer holds true. Startups have formed a smaller and smaller portion of the business landscape: in 1982 some 38% of American firms were less than five years old; by 2018, 29% were that young. The share of Americans working for startups likewise fell. Silicon Valley sizzled with high-tech wizardry, but its giant companies hoarded the best researchers, leading to a slower spread of new ideas throughout the country. Researchers, including at the Federal Reserve, pointed to this decline in dynamism as a cause of weaker productivity growth.

Suddenly, what was old appears to be new. An array of data indicate that Americans are rediscovering their go-getting spirit. The most striking evidence comes from applications to form businesses, a proxy for startup activity. These soared in mid-2020, when America was still in the grip of covid-19. The initial surge was easy to dismiss: some of the new firms were scams, trying to profit from the government’s financial assistance for small businesses; others reflected the strangeness of the moment, with companies set up to import face masks or flog hand sanitiser.

But now, well after the pandemic has faded away, the surge continues (see chart 1). Last year applications to form businesses reached 5.5m, a record. Although they have slowed a touch this year, the monthly average is still about 80% higher than during the decade prior to covid, compared with just a 20% rise in Europe.

Startups normally play an outsized role in creating employment in America, as elsewhere. By definition, every startup job counts as new, whereas mature companies have more churn. That difference has become even starker. In the four years before the pandemic, established firms added one net job for every four created by startups; in the four years since the pandemic, established firms have actually lost one job for every four created by startups (see chart 2).

Perhaps even more important than the numbers is the kind of ventures that are being created. In 2020 and 2021 many startups catered to the working-from-home revolution. These included online retailers, small trucking firms and landscapers. Since mid-2022, however, the baton has been passed to technology firms, according to Ryan Decker of the Fed and John Haltiwanger of the University of Maryland. A paper published in March by the Census Bureau found a particularly sharp increase last year in business applications based around artificial intelligence (AI). For researchers, this carries echoes of the 1990s, when computers and the internet took off. “It feels like a step-change increase across the economy in entrepreneurial potential," says Kenan Fikri of Economic Innovation Group, a think-tank. “You never know which firm is going to be the next growth firm. So the more shots on goal you have, the better."

What has fuelled the boom? The pandemic got things going, as millions lost their jobs and more shifted to remote work. “People realised that they do like being around their families, and it gave many a sense of freedom," says Jeanette Brewster of Village Launch, a nonprofit in Greenville that supports black entrepreneurs. Most of the new firms are tiny and destined to stay that way. Startups in Ms Brewster’s network include food trucks, handicraft makers and paralegals. Still, these can be important steps towards greater wealth. Researchers at the Brookings Institution, a think-tank, found that in 2019 just 5% of business-owning families were black and 4% Hispanic. By 2022 their shares had risen to 8% and 7%, respectively.

The strength of the economy has also helped. When the job market is tight, it is easier for a potential startup founder to take risks, knowing that they can fall back on paid work if need be. The advent of new technologies, especially AI, also feeds into it. Entrepreneurs are creating AI-powered tools to interact with customers, prepare taxes, sift through court records and more. “The causality isn’t necessarily running from startups to innovation. It runs both ways," says Mr Haltiwanger. “Innovation attracts startups, particularly when there are rapid changes that have potentially large market opportunities."

A striking feature of the boom is its geographic spread. Traditionally, American innovation has been focused in California’s Bay Area and urban dynamos such as Austin and New York. The new wave has fanned out to smaller cities around the country, from Boise to Raleigh. Greenville is another example. Better known for its genteel pace and walkable downtown than a striving business culture, it is an unlikely candidate to be a cradle of entrepreneurship. In the past few years, though, its liveability has been a selling-point, as remote work has taken off. Smallness can also be an advantage.

At the start of the pandemic, John Barnett, a serial app creator, moved to the city, which is where his wife grew up. In 2022, within days of being sacked by Twitter, he got together with tech friends to build Supermoon, an app that uses AI to help small firms manage overcrowded inboxes and reply to customers. In Silicon Valley, Mr Barnett says his team might have done formal research sessions in rooms with one-sided mirrors to watch users play around with the app. In Greenville the process was more organic. He knew from local businesses that they were struggling to stay on top of their inboxes, so he asked them to try the tool out. “It’s so easy just to connect with folks. It is like a testbed for research," he says. Incora, the firm making health-tracking earrings, got its clinical trial off the ground because of buy-in from the University of South Carolina’s local medical school. “In bigger cities it would take much longer to establish these kinds of opportunities," says Ms Gevaert.

Owing to the proliferation of remote work, even scrappy startups can tap into big talent pools. Mr Barnett’s team includes engineers in the Bay Area and London. Ms Gevaert says her firm “rented the brains" of product designers in Silicon Valley and strategy consultants in New York. Funding remains a challenge in America’s south-eastern states, where there is not much of a tradition of venture capital. “We’re in our early adolescence in building up a vibrant startup ecosystem. There’s still maturing that will take place," says John Osborne of Good Growth Capital, a VC firm in Charleston, South Carolina. Although fundraising by American VC firms plunged by 60% in 2023 to a six-year low, amid higher interest rates, many are still sitting on unused capital raised in earlier years, so the slowdown may not bite for a while yet.

The big economic unknown is whether the startup boom will translate into productivity gains. In theory, the arrival of new companies should breathe vitality into the economy: entrepreneurs tend to make use of new technologies and create novel business models, in the process keeping incumbents on their toes and propelling growth. There is not much yet in the economic numbers to indicate that this is happening. Labour productivity shot up last year, but that merely made up for a decline in 2022.

One possibility is that the new generation of startups may provide less of a boost to growth than their predecessors, because many reflect changes in where and how people are working rather than a true increase in efficiency. A more promising possibility is that America may be experiencing a repeat of the Solow paradox. In 1987, Robert Solow, an economist who won the Nobel prize that year, quipped that you could “see the computer age everywhere but in the productivity statistics". These days, you can see the return of startups everywhere in America—even in Greenville—but not yet in the productivity data. Eventually, though, the Solow paradox was resolved. By the mid-1990s it was evident that productivity had increased. Give the latest startups a few years to make a mark.

© 2024, The Economist Newspaper Ltd. All rights reserved. 

From The Economist, published under licence. The original content can be found on www.economist.com