This week we had a look at the Build More Agenda from the Build America Caucus. The agenda proposes 62 solutions across housing, healthcare, energy, innovation, transportation, infrastructure, childcare, education, workforce, and state capacity. We thought we would each spotlight one (or two) recommendations from the report we’re excited about.
Metascience units and the 1949 classification system — Matt Clancy
Small fixes, big unlocks — Nisha Austin
Have you tried taking it for something else? — Saloni Dattani
Making CatEx work for transit — Willow Latham-Proenca
The housing twin of the CatEx problem — Alex Armlovich
Metascience units and the 1949 classification system — Matt Clancy
There are a lot of ideas for how we could improve the distribution of federal R&D funding, either to target breakthrough science more effectively, or to reduce the time cost for researchers to apply for funding. We could, for example, do two-stage grant review, golden tickets, distributed peer review, use more program officer discretion, implement AI peer review, use partial lotteries, and so on. But which ones are actually effective? To find out, we need a way to systematically pilot, test, and gather evidence. I’m excited therefore, about the Build More Agenda’s call to establish metascience units at each major scientific agency. Such a structure has existed in the UK for a few years, and was launched at NSF in September. They exist to perform precisely this function of incrementally improving our major funders. To learn more about metascience units, check out Eight Questions About Metascience Units from IFP’s Andrew Gerard and Jenn Gustetic.
We care a lot about federal science, but ultimately it’s a tiny share of what the federal government does. Improving the rest of government typically requires granular and program-specific policy changes, but there are a few cross-cutting initiatives that can improve government across the board. One such area relates to how the government manages its workforce. The Build America Agenda calls for the government to modernize how it classifies the federal job classification system, which is still based on a system first adopted in 1949. Gabe Menchaca at the Niskanen Center has written a comprehensive case for how this system poorly serves the government today, and has all sorts of downstream effects on who the government hires, how well it retains them, and the general culture of what being a good performer means in government. Updating the classification system wouldn’t solve everything, but it’s an important first step.
Small fixes, big unlocks — Nisha Austin
In addition to the classification reforms Matt mentioned, another part of the state capacity section that I’m excited about is the plan to stop applying the Paperwork Reduction Act to voluntary surveys. That might sound minor, but Sara Meyers at Factory Settings wrote about what it took to get a five-minute, voluntary meeting-request form through PRA review at the CHIPS office (spoiler alert, it’s a lot: a Federal Register notice, a 20-page document, a 7-page supporting statement, and an emergency justification letter).
The agenda’s workforce section is worth a look too. It’s baffling that an EMT needs about 36 days of training while a cosmetologist needs 342, and a security alarm installer needs 747(!). The agenda proposes incentivizing states to match licensing requirements to the actual risks of the work. Third Way sketched one possible approach last month: a competitive Department of Labor grant program that pays licensing boards to review whether each requirement is actually needed for safe practice.
I also liked the plan’s push for interstate compacts, so a license earned in one state counts in another. States have already built compacts for 11 professions, including teaching. Portability helps all kinds of workers move to wherever the better opportunity is, whether that’s a job, housing, childcare, or anything else that goes into building a better life. (ICYMI, Matt Clancy, former daycare worker, wrote about another way to unlock more childcare earlier this year.)
Have you tried taking it for something else? — Saloni Dattani
One of the healthcare abundance ideas that stood out to me was to ‘create a prize for proving inexpensive generic medicines can successfully treat new diseases.’ Roughly a third of all FDA-approved drugs are later approved for a second use, because many drugs are ‘promiscuous’, with multiple effects in the body. One example is omalizumab, which was originally approved as an asthma drug in 2003, but turned out, in 2024, to also be effective against food allergies including peanut and milk allergies, thanks to an NIH-sponsored trial that tested its effect on these conditions. The missed opportunity for patients during that time was large, but it’s even worse to imagine if it wasn’t tested for them at all. It’s interesting to think about which other drugs might have more effects we don’t yet know about, and how best to find out.
If there are lots of competing hypotheses about which drugs might work for a particular condition, a prize for success could help especially if companies or researchers already have reasons to suspect their own candidate will succeed and could fund the trials to prove it. But you’d also want a benchmark to define success at the end. That could mean having sufficient evidence to apply for and secure an FDA approval for another condition, or perhaps a shared protocol (and maybe even a platform trial?) to define how such drugs would be tested in clinical trials to secure the prize. I would love to find out which other drugs have transferable skills.
Making CatEx work for transit — Willow Latham-Proenca
A lot of the Build More Energy agenda was, in the best way possible, scooped by BAAJA. To avoid re-treading similar policy ground for a third week, I’m turning to the Transportation & Infrastructure section for this week’s spotlight. There’s a lot to like here, including value-for-money procurement (which lets public project managers leverage their expertise to select the bid that will actually offer the best value over a project’s lifetime, not just the cheapest bid). The first one, though, is a bit of a BAAJA call-back – aligning treatment for transit and highway projects so that “categorical exclusions” (streamlined NEPA processes for actions without significant environmental impacts) actually work the same across both. Currently, states can take on highway approvals themselves, but transit agencies – who are more likely than states to actually handle most transit projects – aren’t able to take on the same delegated authority. This desk fix is a starting point rather than a final fix (even on the highway side, states have been reluctant to take on CatEx authority, likely due to litigation exposure), but it’s a no-regrets one.
The housing twin of the CatEx problem — Alex Armlovich
Willow's transit CatEx point above has a housing twin: A Federal Housing Administration (FHA) multifamily construction loan takes nine months to a year to close, versus two to three months for conventional Fannie/Freddie GSE financing, partly because HUD’s FHA reviews them on outdated systems but also because loan-level NEPA review is a unique burden for multifamily housing: for example, FHA money can also trigger federal environmental review even for apartments on an already-paved urban parking lot. ROAD raised and indexed FHA's multifamily loan limits, but the Build America agenda would modernize HUD underwriting and implement new infill categorical exclusions. (That's important, because a cheaper loan that takes a year to close can lose to a pricier one that closes in a quarter on a net present value basis: time is money even before you count carrying costs.)