Nicholas Bloom on Management, Productivity, and Scientific Progress

Episode Summary

Executive Summary: Tyler Cowen and Nick Bloom discuss Bloom’s research on slowing scientific and productivity growth, the role of government R&D, why IT and working from home have not yet translated into proportional productivity gains, and how management practices, legal systems, and adjustment costs shape firm performance. The conversation also turns to Bloom’s own career path and the practical sources of research ideas and innovation.

Main Topics: Slowing scientific and productivity growth (Priority: 5/5): Bloom argues that many fields experience diminishing returns over time, and that the U.S. has not seen enough new fields emerge to offset declines in older ones. He cites examples from Moore’s Law, cancer research, and crop yields. Why technology hasn’t boosted productivity more (Priority: 5/5): The discussion centers on the paradox that computers, digital tools, and IT are ubiquitous yet aggregate productivity growth remains subdued. Bloom emphasizes that organizational change, not just technology, is often the bottleneck. Government R&D, public goods, and institutions (Priority: 5/5): Bloom suggests declining public funding for basic research is a major contributor to slower scientific progress, while also noting that globally the picture is more complicated because spillovers are localized and other countries’ R&D may not substitute well for U.S. public R&D. Management practices and consulting (Priority: 4/5): Bloom explains the value of management consulting as helping firms adopt underappreciated operational improvements such as Lean and quality-control systems, especially when legal and organizational constraints prevent firms from discovering or implementing them on their own. Working from home and long-run innovation (Priority: 4/5): He sees short-run productivity gains from remote work for many workers, but worries about long-run costs to idea generation, seminars, spontaneous collaboration, and patenting if remote work persists at full scale. Trust, law, and firm organization (Priority: 4/5): Bloom links trust and effective management to contract enforcement and legal systems. He argues weak legal institutions encourage family hiring, reduce professional management, and limit scaling, especially in developing countries. Bloom’s career path and adjustment costs (Priority: 3/5): The final section reflects on his unconventional training and career choices, his early focus on adjustment costs, and how those ideas shaped his work on uncertainty, management, and innovation.

Key Arguments: Scientific progress slows in mature fields, and new fields arise too slowly to fully offset the decline in older ones. Measured productivity per researcher has fallen even as society has poured more scientists, universities, and R&D spending into innovation. A major reason IT has not delivered more productivity is that firms and institutions adjust slowly; technology adoption requires complementary organizational change. Government R&D matters because it is more likely to fund basic research with broad spillovers, whereas private R&D is more development-oriented. Management consulting can create value by identifying process improvements that are not obvious ex ante, such as Lean production and defect reduction. Weak legal systems reduce trust, encourage reliance on family members, and hinder the scaling of firms with professional management. Working from home likely helps short-run efficiency but may hurt long-run creativity if it weakens seminars, informal interactions, and coauthor formation. Bloom’s own research agenda was strongly shaped by data availability, coauthor networks, and the economics of adjustment costs.

Data Points: Decline in scientific productivity per researcher: about 5% per year - Bloom describes the estimated fall in efficacy per researcher over time Productivity growth peak in the U.S.: about 3% to 4% per year - Bloom says technological progress peaked in the 1950s Historical U.K. productivity growth: about 0.1% per year - Bloom cites very low growth from roughly 100 A.D. to 1750 before the Industrial Revolution Moore’s Law scientists needed: 18 times more scientists since the 1970s - Bloom says more researchers were needed to maintain chip performance gains PhD length in economics: 4 years to 6 years plus pre-docs - He uses this as an example of longer training times before reaching the frontier Government share of U.S. R&D funding: roughly two thirds in the 1960s; now roughly one third - Bloom argues government has pulled back while private firms now fund more R&D Remote-work feasibility: about 40% of Americans can work from home - Bloom notes this group accounts for a disproportionate share of GDP GDP share represented by WFH-eligible workers: about 50% to 60% of GDP - Higher-earning workers are more likely able to work remotely C-trip/WFH duration in China study: about 9 months - Bloom cites rising loneliness and fatigue over extended remote work Management survey scale: several million organizations - Bloom describes the data collection effort behind management research Swiss/other high-trust countries: Norway, Sweden, North America - Used as examples of places where trust correlates with stronger legal systems

Pivotal Quotes: "the wheel was a fantastic innovation. But at some point, you know, progress slows down." — Nick Bloom: Explaining why individual technologies and fields naturally hit diminishing returns "we in the New York Times, he wrote, I think, in 86, you see computers everywhere except in the productivity figures." — Tyler Cowen quoting Robert Solow: Introducing the puzzle of IT’s weak measured productivity impact "the initial spark or idea is much more effectively generated in person." — Nick Bloom: On why fully remote work may impede innovation and new collaborations

Implications: The conversation suggests that productivity policy should focus less on devices alone and more on institutions: public R&D, legal enforcement, management quality, and in-person collaboration. Remote work and IT can help, but only if organizations adapt in ways that preserve innovation.