Pinned post
100% agree with Adam on this.
The nation's science portfolio should include a small allocation to focused, time-bound research organizations to complement established institutions like universities. The two play different roles and enable different kinds of value creation.
It's useful to think about it the same way as we construct investment portfolios that hold a small set-aside for venture as well as long-duration assets. That's a standard approach in financial portfolio construction and we should be clear that it’s also true for how the federal government should think about allocation.
The point is not to over-invest or concentrate in one type of performer. That means we also should not over-concentrate on new institution types, especially while they are an experiment.
We have 40 years of precedent for this. The SBIR/STTR program is funded at 3.65% of the extramural R&D budget and it has seeded thousands of small businesses creating new innovative products from research. Most of these companies don't survive in the long term and that's a feature not a bug. The turnover that happens through IPOs and mergers and acquisitions, as well as winding down companies, drives creative destruction and cycles of new ideas and talent.
FRO investments, other new institution shapes, and direct awards to individuals are analogous: they complement the permanent research base and create room for experimentation that's hard to do in existing institutional structures. But they in no way should replace it wholesale — just as SBIR shouldn’t replace industry R&D and angel/VC investments shouldn’t replace public markets.
Federal R&D obviously isn't a financial market. But allocation logic analogies are instructive: diversify across time horizons and risk, size new experiments modestly, and let turnover be a generative engine.
And this is critical: You can't diversify your way out of an underfunded, understaffed portfolio.
Universities, national labs, and industry are the core of that portfolio — experiments complement them, not replace them. To get the real benefits out of portfolio diversification, public R&D funding should grow, award flow needs to resume, and science agencies need to be staffed with adequate career staff.
The nation's science portfolio should include a small allocation to focused, time-bound research organizations to complement established institutions like universities. The two play different roles and enable different kinds of value creation.
It's useful to think about it the same way as we construct investment portfolios that hold a small set-aside for venture as well as long-duration assets. That's a standard approach in financial portfolio construction and we should be clear that it’s also true for how the federal government should think about allocation.
The point is not to over-invest or concentrate in one type of performer. That means we also should not over-concentrate on new institution types, especially while they are an experiment.
We have 40 years of precedent for this. The SBIR/STTR program is funded at 3.65% of the extramural R&D budget and it has seeded thousands of small businesses creating new innovative products from research. Most of these companies don't survive in the long term and that's a feature not a bug. The turnover that happens through IPOs and mergers and acquisitions, as well as winding down companies, drives creative destruction and cycles of new ideas and talent.
FRO investments, other new institution shapes, and direct awards to individuals are analogous: they complement the permanent research base and create room for experimentation that's hard to do in existing institutional structures. But they in no way should replace it wholesale — just as SBIR shouldn’t replace industry R&D and angel/VC investments shouldn’t replace public markets.
Federal R&D obviously isn't a financial market. But allocation logic analogies are instructive: diversify across time horizons and risk, size new experiments modestly, and let turnover be a generative engine.
And this is critical: You can't diversify your way out of an underfunded, understaffed portfolio.
Universities, national labs, and industry are the core of that portfolio — experiments complement them, not replace them. To get the real benefits out of portfolio diversification, public R&D funding should grow, award flow needs to resume, and science agencies need to be staffed with adequate career staff.
AtView quoted post on X
@Convergent_FROs
we’ve helped pioneer some of these approaches. We believe in them. That means we’re also clear-eyed that metascience innovations are potentially game-changing additions to America’s scientific core — not a replacement for it.