By: Conor Gowder & Mark Skinner

A widely held tenet of most economic-development-focused regional competitiveness policy is that small businesses drive innovation. However, by at least one of the most commonly applied metrics for most of the U.S. economy—money invested—that doesn’t hold water and hasn’t for a long time, based on the latest Business Enterprise R&D survey (BERD) data released by the National Science Foundation.  

Small businesses may represent an inordinate share of the number of firms and may have more ideas, but the real money for R&D is spent and conducted by companies with more than 250 employees. The difference climbs significantly for businesses with 1,000 or more workers.  

The table below presents domestic R&D expenditures for businesses by employment size across ten years, 2015-2024, the most recent decade available. Sliding the bar reveals each year, allowing the viewer to see changes over time.  

 

Should regional innovation policymakers abandon their efforts to grow science- and technology-focused startups? Of course not. The localized churn of innovation entrepreneurs, STEM workers and investment feeds the creation of cycles of vibrant opportunity within their regions, attracting more of the same in successfully sustaining geographies.  

Larger local firms play an important role, however, in sustaining that churn and innovation vibrance. Acquisitions and mergers are the chief success route for the vast majority of startups securing risk capital A smaller percentage reach the 1,000-employee threshold or higher on their own.  

For TBED policy makers and practitioners, adding some intentionality to the types of startups supported, based on their match to the mix of existing academic and corporate R&D strengths, paired with the local companies best positioned to accelerate successful tech to the marketplace through their production and distribution channels, may lead to a healthier, more competitive overall regional economy.  

As the latest NSF BERD data reconfirms, the large firms have a lot of money to spend on R&D, and they can do it pretty much anywhere in the country. Why not make sure they recognize the opportunity to do so in their own backyards? 

The 2024 BERD data is available from NSF here.