Major biomedical journals tend to publish papers on the same scientific topics that their editors studied, but this effect likely plays a very small role in the publishing process -- increasing the churn of editors (e.g., via shorter tenures at journals) wouldn't change things much.

The spillovers from public R&D grants are large and reach far across geographic and technological space, and focusing only on firms that directly receive grants causes you to miss a lot of what's going on.

The COVID-19 pandemic led to a significant decline in the number of "new" research projects initiated by academic researchers.

The adjustment costs of science -- getting scientists to study what you want them to -- are very large.

Scientists experienced a wide range of disruptions to their work in the early months of the COVID-19 pandemic, with the largest declines amongst those in the "bench sciences," female scientists, and researchers with young children at home.

When people want more new drugs, firms are happy to invest in ideas that cost more. And as they run out of "low hanging fruit" while demand keeps growing, R&D costs will naturally grow.