6-minute read
Core Question
Why do public systems and government-adjacent institutions grow larger over time without producing better results?
Executive Gist
Many public systems expand year after year while outcomes stagnate or worsen. Budgets grow, staffing increases, and new programs are layered on—yet the underlying problems remain strikingly persistent.
This pattern is usually explained away as a shortage of resources, compassion, or political will. But the deeper issue is not external. It is structural.
Systems that are insulated from failure are also insulated from learning. When meaningful feedback mechanisms are absent, improvement cannot occur. Institutional inertia sets in, activity is mistaken for progress, and persistence becomes the primary indicator of success.
This article examines why systems without feedback persist without improving and what conditions are required for institutions to learn.
The Pattern: Growth Without Improvement
Across many areas of public policy, the same pattern appears: budgets increase, staffing expands, and programs multiply yet the underlying problems remain.
When outcomes disappoint, the explanation is usually external: not enough funding, not enough political will, or too much opposition. What is rarely questioned is the design of the system itself.
Growth without improvement is not accidental. It is the predictable result of systems that lack feedback and clear exit conditions.
Accountability Is Feedback, Not Punishment
Accountability is often framed as blame, cruelty, or political attack. In morally consensus-driven systems, this framing makes accountability feel taboo.
But accountability is not punishment, it’s information.
Feedback tells a system what worked, what didn’t, and what should change. Without it, errors persist, success cannot be distinguished from failure, continuation becomes the default definition of success and systems lose the ability to learn.
When Feedback Is Missing, Oversight Becomes Illusory
What appears as “feedback failure” is often more fundamental: feedback is absent altogether.
Most systems rely on compliance audits, process reviews, and reporting requirements. These measure activity, not outcomes.
What is missing are mechanisms that evaluate whether a mission succeeded, trigger correction when it didn’t, and define when a program should end.
Existing oversight does not fill this gap. Auditors focus on compliance, not results. Legislative oversight depends on the same departments it is meant to evaluate and is often incentivized toward affirmation rather than interrogation. Voters cannot evaluate outcomes that are never clearly measured and are instead offered narratives in place of comparisons.
The result is simple: no institution is clearly responsible for asking, Did this work—and what happens if it didn’t?
Institutional Inertia: Motion Without Correction
When feedback is absent, institutional inertia takes its place. Systems continue without clear benchmarks, without scheduled reassessment, and without meaningful termination thresholds. There is little reward for candid evaluation and significant risk in surfacing failure.
In mayor–council systems, this inertia is structurally reinforced. Department heads are appointed and managed by the executive, while legislative oversight depends on the same information channels. The incentives become clear: alignment is safer than evaluation, and affirmation is safer than candor.
Leadership may change, but the underlying assumptions remain fixed.
Government and Its Intermediaries: Diffused Accountability
This pattern is not limited to formal government agencies. It extends to government-adjacent organizations that carry out public functions through distributed responsibility.
In these systems, evaluation shifts toward proxies rather than outcomes. The focus becomes whether procedures were followed, whether activity can be documented, and whether reporting requirements are satisfied.
What is less often examined is whether the underlying condition has improved, whether the problem has been reduced, or whether alternative approaches perform better.
This is not a failure of intent. It is a consequence of design. When responsibility is diffuse, accountability becomes diffuse. Measurement gravitates toward what is easiest to observe rather than what is most meaningful.
Government authorizes and funds. Intermediaries execute. Evaluation centers on compliance. Outcomes become secondary—and persistence becomes success.
When failure is impossible, improvement becomes optional.
When systems cannot fail, improvement is no longer required. Expansion becomes the default response, evaluation becomes destabilizing, and continuation becomes synonymous with success.
Over time, incentives adjust accordingly. Job security ties to persistence. Funding aligns with narrative coherence. Correction becomes political risk.
The outcome is not driven by coordinated intent, but by the incentives embedded in the system itself.
Why This Matters: Civic Impact
When systems do not improve, problems persist across budget cycles while costs continue to rise. Institutions grow more complex but less effective, and public confidence declines as outcomes stagnate.
Debate gradually shifts away from measurable results and toward competing narratives. Substance gives way to slogans.
Over time, systems designed to persist rather than learn become increasingly expensive and increasingly difficult to correct.
Path Forward: Guardrails That Enable Learning
Improvement does not require a new ideology or a comprehensive reform agenda. It requires structural guardrails that allow systems to learn.
At a minimum, systems capable of improvement share several features:
- Clear goals stated in advance
Success must be defined before implementation, not assumed through continuation. - Outcome-based evaluation
Assessment should focus on whether objectives are achieved, not whether activities occur. - Pre-scheduled review and reassessment
Evaluation should happen at defined intervals and include real options: continue, adjust, or end. - Authority to correct course
Feedback must be actionable. Information without the ability to change policy or funding is inert. - Real exit conditions
Programs that do not meet their goals should be revised, replaced, or ended. Permanence should not be the default.
These guardrails do not guarantee success, but they make learning possible.
Without the possibility of failure, systems cannot improve.
Conclusion
Systems improve only when they are allowed to learn. Learning requires feedback. Feedback requires accountability, including the possibility of correction.
When feedback is absent, institutional inertia fills the vacuum. Activity substitutes for progress, and persistence replaces performance.
If failure is impossible, improvement is optional.
Systems that cannot learn create an environment where persistence replaces performance and structure replaces accountability. Without feedback, correction becomes unlikely—and drift becomes direction.
Counterarguments Considered
1. “These systems deal with complex human problems that can’t be measured.”
Partly true. Not all outcomes are easily quantified. But complexity does not justify the absence of evaluation. Even imperfect measures allow for comparison—and comparison is the basis of learning.
2. “Accountability will discourage innovation and compassion.”
Accountability does not suppress innovation; it enables it. Compassion without feedback risks becoming neglect, as systems persist regardless of whether they reduce harm.
3. “Oversight already exists through audits, councils, and elections.”
Oversight exists in form, but often not in function. Compliance audits, aligned legislatures, and narrative-driven elections do not substitute for outcome-based evaluation or correction mechanisms
