What we're reading: January 23, 2026
Paperwork Reduction Act, Science funding, and More!
Happy January 23! Here are some interesting things we’ve been reading:
Congress is pushing back on the administration’s proposed science cuts, as the NYT and Science report. The Senate’s bipartisan CJS package largely preserves funding across the board — science budgets at NASA and DOE are stable, and a proposed 55% cut to NSF ended up closer to 3%. Earlier this week, Congress also released a compromise LHHS bill that would raise NIH’s budget slightly, rejecting a proposed 40% cut. These numbers track with a Metaculus report we commissioned back in September, in which pro forecasters predicted relatively steady funding for NIH and NSF despite the turmoil; the proposed NIH and NSF budgets are both just shy of forecasters’ 75th percentile predictions for FY26. — Jordan Dworkin
Market shaping. So hot right now. But for funders used to the simplicity of push funding, it can be difficult to figure out how to appropriately size a pull mechanism such that it sufficiently incentivizes innovation without wasting resources. Thankfully the Market Shaping Accelerator team put out a new guide for sizing prizes and advance market commitments based on expected R&D costs and probabilities of success. The guide also includes an interactive calculator where you can plug in plausible values in your context and get an estimated incentive size in return. — Jordan Dworkin
The Paperwork Reduction Act of 1980 probably wins the prize for “most Kafkaesque US federal law.” Whatever else it does, the PRA dramatically increases the amount of paperwork federal workers have to do. In theory, this is meant to save the public time on paperwork, by requiring that “information collections” by the federal government undergo extensive internal review. In reality, as former CHIPS Program chief of staff Sara Meyers explains in a new piece, the law means that if you want to, say, put up a voluntary survey on a government website, you have to produce “a 20-page document with annotated screenshots to show every permutation of the data entry page; a 7-page Supporting Statement defending the collection” and three more documents besides. Not great! — Dylan Matthews
The Food and Drug Administration has released a draft “Guidance for Industry” on using Bayesian methodology in clinical trials for drugs and biologics, which has been in the works for several years. It lays out practical use cases (largely forms of “borrowing” information across phases, subgroups, ages/pediatrics, rare diseases, and even across diseases) and discusses priors in some detail, including discounting approaches. But as Witold Wieçek points out, the document reads like Bayesianism with frequentist guardrails: by emphasizing calibration to Type I error rates, it effectively reintroduces frequentist control criteria, and it still keeps applications case by case. Time will tell how it plays out in practice. — Saloni Dattani
Climate mitigation debates often pit restrictions on fossil fuel supply (mildly confusingly for abundance readers, called “supply side policy” in the article) against efforts to make clean energy more competitive. Zeke Hausfather makes a convincing case that making clean energy cheap is more tractable, and therefore more effective, than making fossil fuels expensive. Hausfather argues that the 1.5°C target – which represents a probabilistic balancing of risk and achievability, not a tipping point – has become a political snare. The gas-vs-coal lifecycle analysis is worth a read on its own, arguing that you’d need implausibly high methane leakage rates (5%+ on a 20-year timeframe) for gas to be worse than coal. – Willow Latham-Proenca
To what extent is anti-growth zoning really a different problem from stagnant construction sector productivity? An NBER working paper sheds some light on the question. The construction productivity growth collapse after 1970 coincides with the postwar growth control movement that downzoned America’s cities, and the average share of homes built in large-scale projects has fallen by over a third. Growth control regulations directly limit the scale of individual residential projects, literally limiting economies of scale–but also indirectly create barriers to entry for builders between cities. – Alex Armlovich
Abundance as we see it entails a pragmatic disposition towards governance and results that can transcend partisan and factional divides. Mayor Mamdani’s endorsement of SEQRA reform is a genuinely novel demonstration of how an Abundance-coded governing agenda can deliver on populist messaging. – Alex Armlovich
Finally, a few new items from the Abundance and Growth team:
Dylan Matthews launched a substack!
Saloni Dattani wrote about The Golden Age of Vaccine Development (you’re living in it!)
And in case you missed it, earlier this week Jordan Dworkin made the case that replication is underrated on this blog.