## Notes from 06 March 2026 [[2026-03-05|← Previous note]] ┃ [[2026-03-07|Next note →]] ## The same mistake, thirty years apart Read [[Gabe Menchaca]]'s **"[From Gore to DOGE: The bipartisan history of failed workforce reform](https://hypertext.niskanencenter.org/p/from-gore-to-doge?r=10o1)"** (Nov 2025, [[Niskanen Center]]'s _Hypertext_). Where [[Casey Eilbert]]'s companion piece in the same issue works at the level of intellectual history (paradigms, tensions, accountability theories), Menchaca works at the level of what actually happened to real agencies when the headcount targets hit. A third piece from [[Kevin Hawickhorst]] asks a different question: not what broke, but what lasted. Menchaca's is one of the most operationally honest accounts I've read of how the US federal government ended up in its current workforce crisis. The piece makes me think about a pattern I keep encountering across countries: the gap between what a government _promises_ to stop doing and what it _actually_ stops doing in times of fiscal adjustment. That gap is where the damage happens. And it's where the interesting comparative questions live. ### The Clinton trap The story starts with a correct diagnosis. The National Performance Review, or NPR, (1993) got the problem right: "Washington is filled with organizations designed for an environment that no longer exists". Good people, bad systems. Too much procedure, too much middle management, too little mission focus. This diagnosis still rings true in 2025... Menchaca points out that you could drop lines from the NPR report into a contemporary op-ed about state modernization and nobody would notice the 30-year gap. But a correct diagnosis became a toxic prescription when political incentives took over. Clinton needed something tangible for the 1996 State of the Union. "The era of big government is over" turned a useful _measure_ of transformation (positions that could be saved through reform) into a _target_ (roles to cut). From that moment, the reforms that were supposed to make the cuts possible dropped out of the conversation. Congress mandated faster cuts without providing the flexibility to de-proceduralize the work. Agencies were left to hit headcount numbers while still carrying every mandate, every regulation, every reporting requirement they had before. What Menchaca documents next is the mechanism by which _cuts-without-reform_ destroy capacity: |**WHAT GOT CUT**|**WHAT DIDN'T GET CUT**| |---|---| |426,000 federal positions|Agency mandates| |HR staff (−24%)|Hiring regulations (350-page handbook)| |HR assistants (−43%)|Civil service classification rules| |Procurement support (−59%)|Procedural requirements| |Budget assistants (−27%)|Congressional reporting obligations| |Junior/entry-level staff|Scope of government responsibilities| |Most marketable senior talent| The left column is what the administration could control. The right column is what Congress and the legal framework wouldn't let them touch. The result: agencies had to do the same work with fewer people, less expertise, and a gutted support infrastructure. ### The workforce you can't see Here's where it gets really interesting for me. The government didn't actually shrink. It just became harder to see. Agencies under pressure to maintain low headcounts [[Escape from Public Law|turned to contractors]]. By 2005, the total [[Blended Workforce|blended workforce]] (contractors, grantees, and other non-public employees doing agency work) was _larger_ than the pre-cut federal workforce had been. By 2023, there were 2.1 contractors for every federal employee. Multiple analyses found contractors were generally more expensive than their federal counterparts. So the "era of big government" ended on television, and the era of expensive, invisible government began in procurement offices that had just lost 59% of their support staff. The people who were supposed to manage the contractors were the very people who'd been cut. Menchaca's numbers here are devastating: - Contracting outlays per procurement professional: **$13.8M** (1992) → **$17M** (2024, inflation-adjusted) - Contract actions per specialist: **622** (1992) → **2,718** (2024) This isn't a story about contractors being bad. It's a story about a government that dismantled its own ability to oversee the workforce model it was forced to adopt. Healthcare.gov (2013) and FAFSA (2023) are the visible failures, but the quiet daily breakdowns are worse because nobody outside government sees them. ### The demographic time bomb The other thing Menchaca documents is what blunt workforce tools do to the _shape_ of the workforce over time. Buyouts, early retirement, and layoffs don't cut evenly. Each instrument has a selection bias: - **Buyouts** → the best people leave (they know they can compete privately) - **Early retirement** → the most experienced people leave (institutional memory gone) - **Layoffs (RIFs)** → statutory rules force out the most junior and least-tenured first (the pipeline is severed) The combined effect on the federal age pyramid was dramatic. Under-35s went from 26% of the workforce in 1992 to under 17% by 2000. Over-50s went from 25% to 36%. The median General Schedule grade (a rough proxy for seniority and specialization) climbed from GS-9 in 1992 to GS-11 by 2000 to GS-12 by 2024. The government progressively closed itself off to early-career talent and never reopened. While other large employers moved toward modern apprenticeship models (training people from novice to expert) the federal government kept hiring only at mid-career and senior levels, outsourcing the junior rungs to contractors. ### DOGE walks into the same trap Menchaca's punchline is that DOGE, for all its [[Tech Right|Silicon Valley branding]], has essentially recreated Clinton's playbook: same instruments, same sequence, same structural blind spot: 1. Incentivize departures → best talent and most experienced leave first 2. Congress doesn't reduce scope → agencies still carry every mandate 3. One rescission package of $9B (0.2% of the federal budget) → effectively nothing off the to-do list 4. Agencies scramble to do same work with fewer people → shadow workforce expands again But there's a dimension where DOGE may be doing _deeper_ damage. Clinton at least maintained some pretense of respecting public servants. The current approach deliberately creates a hostile work environment. Menchaca's observation here is just simply true: aspiring public servants now have to assume they're "always just one election away from an arbitrary firing". If that becomes the perception (and it's hard to see how it doesn't) the talent pipeline doesn't just narrow. It poisons. Only the independently wealthy or those with no better options will choose government careers. The professional middle class, the people you actually need running agencies, will go elsewhere. The Clinton administration also tried something DOGE hasn't: genuine experimentation with alternative models. Converting the Patent and Trademark Office, the FAA, and Federal Student Aid into _[[Performance-Based Organizations (PBO)]]_ exempt from many civil service rules. Personnel demonstration projects like [[AcqDemo Project|AcqDemo]] at Defense. These were real attempts to test whether the system could work differently. The telling detail is that many of these pilots have been lingering in "trial" status for over twenty years, while the foundational framework (the 1978 Civil Service Reform Act) remains untouched. Nobody has the appetite for [[2026-02-11#The "_Guerrilla Way_" out (Waissbluth)|comprehensive reform]], so the workarounds become permanent and the core system stays frozen. Hawickhorst's reading of the same history suggests this isn't accidental — it's the predictable result of reforms that bet on culture over institutions. ### What Hawickhorst adds: institutions beat inspiration [[Kevin Hawickhorst]]'s companion piece in the same _Hypertext_ issue (**"[To defeat the bureaucracy, embrace bureaucracy](https://hypertext.niskanencenter.org/p/to-defeat-the-bureaucracy-embrace)"**) works at a different level of analysis, and it's a useful corrective to the pure capacity-destruction narrative. Where Menchaca asks _what broke_, Hawickhorst asks _what lasted_ — and his answer is uncomfortable for anyone who thinks culture change is enough. His core thesis: the NPR reforms that survived across administrations were the _institutional_ ones (the [[Government Performance and Results Modernization Act (GPRAMA)|Government Performance and Results Act]], the President's Management Council, the OPM handbook consolidation) not the cultural initiatives like the Hammer Awards or the FEI curriculum overhaul, which evaporated the moment a new administration had different priorities. The irony he highlights is sharp: the machinery created to implement the layoffs (the PMC) outlasted the layoffs' intended purpose. The bureaucratic infrastructure for cutting bureaucracy became itself a lasting and useful institution. I think that, of all the explanations for why the flexibility reforms underperformed, Hawickhorst's is the most persuasive. It wasn't simply a case of agencies losing capacity, as Menchaca's story suggests. Rather, NPR eliminated poor procedures without providing incentives for agencies to develop better ones. Although agencies were given permission to innovate, they were not provided with any training, templates or new routines. They persisted with their former customs even when they were not required to, which is indicative of a cultural problem. The old Federal Personnel Manual may have disappeared, but the habits it created did not.