Reforming government in the Trump era

A little-noticed report provides an apolitical blueprint.

AI-generated image from Playground via Perplexity Pro

Happy new year! With the presidential inauguration just three days away, let’s look at something you would expect—indeed, hope—the Trump administration cares about: how to overcome bureaucratic bloat, fraud, delays, cost overruns, and general incompetence in the federal government.

Few things are as emblematic of government brokenness as the Paperwork Reduction Act. The act says that any time a government agency wants to collect information from members of the public—even something as simple as a short online survey—it must prove it’s not collecting any more than necessary. This well-meaning attempt to reduce bureaucracy in fact creates a whole lot more of it, since getting approval for each new information collection takes an average of nine months. In an era when you might want to, say, trial-run a new government webpage or service, get user feedback, tweak the product in response, get more feedback, and so on in an iterative way, the PRA makes it essentially impossible.

That’s just one of the examples in a recent 50-page report from the Niskanen Center, “The how we need now: a capacity agenda for 2025.” The thrilling title, and the fact that it dropped just before Christmas, probably explain why it got exactly zero pick-up in the press. That’s unfortunate, because it’s worth a closer look.

---

The TL;DR: This is the nearest thing I’ve seen to a thorough, completely non-partisan agenda for making government less awful. And most of its recommendations should, at least in theory, be music to the ears of the incoming administration. I just hope someone there reads it.

---

The report, by Jen Pahlka (whom I’ve often quoted here) and Andrew Greenway, is about why government is so bad at implementing its own policies. I won’t attempt a comprehensive summary, though in the Appendix at the bottom of this post I’ve pasted the decent if dull two-pager created by Google’s NotebookLM. But I think the key points are:

• Risk aversion and fear of lawsuits lead to ass-covering. Bureaucrats put compliance with rules and procedures over achievement of outcomes.

• There’s no feedback loop between policy and implementation—no way to iterate when (as inevitably happens) things turn out not to work as intended. As a result, laws and policies are translated into detailed, prescriptive, rigid rules that try (and inevitably fail) to anticipate everything in advance, instead of flexible frameworks that can adapt to change and to the unforeseen.

• Lawmakers refuse to make trade-offs, but instead layer on requirement after requirement with little thought to the cost of implementation.

• Relatedly, a successful project is a completed project, regardless of time and cost overruns. Instead, a successful project should be one that stays within time and budget, even if that means not getting everything done. (In the tech industry, this is called product management, as distinct from project management; the product manager has the power to identify and ditch the least important requirements in order to get the job done in time.)

• The dogma that government is bad at building technology has made it so completely reliant on outsourcing that it longer has the skills to even outsource properly: it doesn’t know what good technology looks like and how to specify requirements. Contractors are held responsible when things go wrong, but suffer few consequences because contracting rules are so onerous that the government has little real choice of vendors. Meanwhile, because contractors take the heat, nobody in government is held accountable for their own part in the failures.

From there the report goes into some recommendations. These include several things that a Trump administration ought to love, such as

• make it easier to fire incompetent bureaucrats (the report cites the case of an EPA employee who admitted to watching hours of porn a day on his work computer but still hadn’t lost his job four months later)

• reduce the burden of legal compliance and make it harder to sue government agencies for small breaches

• reduce paperwork and slim down or eliminate unnecessary procedures (the Paperwork Reduction Act being a prime example)

• cut away policies and rules that have accreted over time, perhaps using generative AI to identify outdated, conflicting, or unnecessary ones

There are some other things in there that could be harder for a new administration to swallow, but not because they have a left-leaning political bias; rather, because they would be hard, such as radically changing the way tech outsourcing works, or forcing lawmakers who introduce any new mandate to cut an existing one out in return.

It’s surely no accident that not once in the report’s 50 pages do the words “democracy” or “democratic” appear (except in two references to the Democratic and Republican parties). The Niskanen center bills itself as “transpartisan,” and the report is clearly designed to be as politically uncontroversial as possible.

That, of course, doesn’t mean it will necessarily gain any traction. Aside from the fact that someone influential in the new administration would have to read the report, it’s not yet clear whether making government more effective is actually Trump’s goal, as opposed to simply clearing out “wokeness” and regulations that get in the way of big business.

But I genuinely think it’s too early to say. The dozens of bright young things that Elon Musk and Vivek Ramaswamy have reportedly recruited to spend stints working in government agencies might conceivably find some real problems to work on. And this report could help them. So if you happen to know any of them, send this their way.

This week’s links

“Billionaire-proofing” social media. What if, when your favorite social media platform is suddenly overrun with people you don’t like, you could leave it without losing all your connections there, thus robbing the billionaire owner of his or her leverage? That’s the promise of the “Fediverse”, where users own their networks and can move them from one service to another. It’s already true for Mastodon; “Free our Feeds” is a new project to make it possible for BlueSky users. (Pluralistic and The Connector)

Implementing proportional representation in the US. It could break the stranglehold of the two-party system and gives voters parties that represent their actual beliefs. This piece co-authored by Lee Drutman, a long-time proponent of PR, is an easy-to-understand graphical representation of how it might look. (New York Times)

Misinformation isn’t a technology problem, it’s a political one. Specifically, it’s driven by radical right populism. Which means fact-checking, digital literacy, or content moderation won’t help. Tackling it as a political problem is essential, even if that’s hard and messy. (Adrian Monck)

Liberal democracy is not under threat. It’s easy to lump the news from various countries together as evidence of democratic backsliding, but a closer look, as well as some historical perspective, suggests what’s going on is a “patchwork of storms” rather than a “global extinction event,” argues Serge Schmemann. (New York Times)

Governments shouldn’t own their citizens’ data. Decentralized, blockchain-based national ID systems, where governments issue credentials and safeguard the security of the system but don’t actually control the data, can reduce fraud and identity theft and make data interoperable. (GovInsider)

Starlink is undercutting African ISPs. Elon Musk’s satellite internet service is now cheaper than some local fixed-line providers—another small notch in his growing global influence. (Rest of World)

America’s civic renewal is bottom-up, not top-down. Efforts to reinvigorate or reinvent democracy in the US have to come from local-level efforts in citizens’ “associational lives” (i.e. groups they belong to), and probably can’t rely on big philanthropic funders, Daniel Stid argues, in a post with many links to studies and reports about how to make civic culture flourish. (The Art of Association)

Appendix: The Niskanen report summary

Note: the following is an AI-generated summary prepared by Google’s NotebookLM. I haven’t checked it for accuracy against the original document.

This briefing document summarizes the Niskanen Center paper, "The How We Need Now: A Capacity Agenda for 2025 and Beyond," which diagnoses the current challenges to US government effectiveness, or "state capacity," and proposes a four-pronged approach for reform. The authors argue that the US government is struggling to meet its policy goals due to a combination of structural and cultural issues, and that rebuilding capacity is a critical political and leadership choice, not merely a technical one. The paper advocates for a shift from a compliance-focused bureaucracy to one that is outcome-driven, agile, and adaptable.

Key Themes and Arguments

The Problem: A Broken System

• The Pacing Problem: Government is too slow and cannot keep up with the rapid pace of societal and technological change. This is evidenced by long delays in projects like connecting new power sources to electrical grids. "Where speed is actually most needed, we are least likely to get it."

• The Stuck Brake: The government is burdened by excessive and ossified procedures, a "procedure fetish," that slow down progress. Examples include the bloated 5000 series procurement framework at the Department of Defense. These procedures have grown due to the "tyranny of tiny decisions," where many individually sensible rules combine to create an unworkable system.

• Loose Steering: There is a gap between the intentions of policy and the outcomes achieved. This is driven by a "cascade of rigidity," where high-level principles become increasingly rigid and prescriptive as they descend through the layers of bureaucracy. The authors note, "The principles at the top and the mechanics at the bottom are in conflict."

• Trying to go Everywhere at Once: The government is trying to achieve too many policy goals at once ("everything bagel liberalism"), leading to a lack of focus and trade-off denial. San Francisco's ill-fated business ban is given as an example. This stems from a lack of product management in government, where the art of deciding "what to do" is missing. The paper notes, "If something is a true priority, it means that it will happen at the expense of something else."

• Abdicated Navigation: The government has outsourced too much of its work to private contractors without developing the internal competencies and capacities to manage these partnerships effectively. "We lack the internal core competence to outsource well." The paper stresses that if the government is on the hook for the outcomes, "public servants must make the critical calls."

The Historical Context

• The paper highlights the Progressive era as a time of significant governmental reform, focusing on professionalizing the civil service and standardizing procedures. The authors suggest, "Now would be a good time for this historical cycle to return."

• The UK’s reforms of the 1860s and 1870s, inspired by the colonial Indian civil service, is cited as another example of historical reforms which can provide inspiration today.

Four Key Pillars for Rebuilding State Capacity

1. Hire the Right People and Fire the Wrong Ones:

• Reform the Civil Service: The current civil service hiring process is deeply flawed, relying on self-assessments and keyword scans, failing to identify true competencies. The paper asserts that without bold civil service reform, we will not have the state capacity to face the challenges of our decade and beyond.

• Rebalance the Brake and the Gas: The government needs to shift its focus from compliance to results, empowering those closer to the work. "We were six people trying to deliver the product. There were easily 60 people telling us everything we couldn’t do.”

2. Reduce Procedural Bloat:

• Tackle Policy, Procedure, and Regulatory Accretion: Leaders must engage in large-scale inquiry into mandated procedures, asking if the intended protections are actually met and if they are worth the cost.

• Principles for De-Proceduralization: Shift emphasis from compliance to mission, accept judgment as essential for public servants, make change top-down and bottom-up, default to "yes" rather than "no", and recognize that complexity itself is the problem.

• Reduce Policy Accretion: Use generative AI to remove vestigial or conflicting regulations.

• Reduce Legal Attack Surface: Reconsider how the right to sue the state has crippled it through inaction and bureaucratic red tape. The current statutes allow for many procedural objections, creating a situation where fear of litigation drives bureaucratic behavior.

• Resist Trade-off Denial: Policy makers must recognize the necessity of trade-offs, adding by subtracting. For every new mandate or constraint, they should remove an old one.

3. Invest in Digital Infrastructure:

• Transition to Product Model Funding: Shift from project-based funding to product-based funding, allowing for continuous development, iteration, and user feedback. "Software is never done." The product model is less expensive and results in better outcomes because it allows for continuous improvement.

• Invest in Centralized and Decentralized Teams: Support specialized teams like the US Digital Service (USDS) and encourage the development of digital expertise throughout government agencies.

• Invest in Platforms: Develop common digital platforms (like the UK's GOV.UK Pay and Notify) to avoid repetitive development, improve consistency, and reduce costs.

• Right-Size AI Guardrails: Avoid overly rigid regulation and recognize that a lack of understanding of new technology drives a "better safe than sorry" approach.

4. Close the Loop Around Outcomes:

• Align Funding to Support Feedback Loops: Move away from rigid budget requests towards more flexible funding streams that allow agencies to adapt based on experience.

• Fund Internal Competence First: Ensure that government has the in-house expertise to understand its own needs, craft a product vision, and manage both internal and external resources.

• Align Oversight to New Models: Oversight must shift from focusing on compliance with process towards evaluating outcomes. This includes convening IGs to educate them about downstream effects of their actions.

• Focus on Enablement over Mandates and Constraints: Legislators need to move away from primarily issuing mandates and constraints, and explore new ways to enable agencies to achieve their goals.

The Importance of Leadership

• Rebuilding state capacity is not a technocratic choice, but a political one requiring strong leadership.

• The public has lost trust in the government’s ability to deliver on its promises. Restoring that trust is a moral and political imperative.

Key Quotes

• "Yet a question all too rarely asked is: How well do our institutions actually work in addressing these and other questions? Can they translate the political intent – of any party – into public impact?"

• "The brake is stuck...The steering is loose...We’re trying to go everywhere at once...We’ve abdicated responsibility for navigating."

• "With all these feet on the brake, it is not surprising that on some occasions, the government doesn’t appear to move at all."

• “'I’m in year 34 of Federal civilian service. I’ve never before seen the kind of bureaucratic squeamishness I’ve seen in the last 5 years. There’s so much fear of making a mistake and getting embroiled in a lawsuit that people are maximally risk-averse. This is murdering the mission.'”

• "Project management is the art of getting things done. Product management is the art of deciding what to do."

• "Software is never done"

• "Government knows how to acquire technology. What we need to acquire are capabilities."

• "The worst budget request is one with round numbers."

• "We put in an order for 100 sandbags, we get 100 sandbags.”

Conclusion

The Niskanen Center paper provides a compelling diagnosis of the challenges facing the US government and outlines a practical and ambitious agenda for reform. Rebuilding state capacity is presented as a critical task for the current administration and Congress. The authors argue that a shift towards a more agile, outcome-focused, and digitally-competent government is not only achievable, but essential for addressing the complex challenges of the 21st century. They emphasize the need for both cultural and structural changes, requiring a new mindset and a willingness to rethink how government operates. It emphasizes that the "how" of government is just as important as the "what."