This is an automated rejection. No LLM generated, heavily assisted/co-written, or otherwise reliant work.
Read full explanation
I think some institutional failures that are usually framed as incentive problems are better understood as temporal misalignment problems: capital and decision-making are governed by short, volatile cycles, while the systems they are meant to sustain operate on much longer, more stable cycles.
When capital is structurally tied to short “fragility cycles” (elections, annual budgets, refinancing windows, donor attention), institutions predictably decay—even with competent leadership and good intentions—because renewal and maintenance must occur on longer “mission cycles” (asset lifetimes, replacement windows, capability formation).
This post sketches a simple model for that failure mode.
Why this might be relevant to LessWrong
Many alignment and coordination problems share a common pattern: optimisation pressure is applied on the wrong timescale, producing outcomes that undermine the stated objective.
This shows up in:
Goodhart-style failures,
principal–agent problems,
organisational decay despite capable actors,
systems that look stable until a renewal boundary is missed.
I’m proposing that time—specifically, which cycles govern capital and decisions—is often the hidden variable.
A minimal model: fragility cycles vs mission cycles
I’ll use a very lightweight vocabulary.
Cycles
A cycle is a recurring temporal structure with:
a period (how long it lasts),
a phase (where you are in it),
and an impact on the system.
Institutions usually operate under multiple overlapping cycles.
Fragility cycles
Fragility cycles are short, volatile, and largely exogenous. Across domains, a few show up repeatedly:
Political cycles (elections, ministerial turnover)
Budget cycles (annual appropriations, grant rounds)
Financial cycles (refinancing, liquidity stress)
Attention cycles (donor enthusiasm, public focus)
These cycles fluctuate in ways the institution does not control.
Mission cycles
Mission cycles are longer and intrinsic to what the institution is trying to do:
asset replacement intervals,
maintenance and renewal windows,
capability formation timelines,
intergenerational continuity.
These cycles are usually more predictable and less volatile.
Empirically, the key pattern is:
Fragility cycles tend to be shorter and more volatile than mission cycles.
The failure mode: capital follows the wrong cycle
In many systems, access to capital is effectively governed by fragility cycles. Informally:
Capital availability tracks elections, budgets, or refinancing—not replacement needs.
This creates a structural problem:
renewal moments arrive,
capital is unavailable or deferred,
assets run past end-of-life,
capability degrades,
failure appears “unexpected” despite being predictable.
Importantly, this can happen without bad incentives or corruption. Everyone may be acting reasonably given the constraints.
A concrete example
Consider a hospital with imaging equipment that needs replacement every ~4 years.
The mission cycle is clear: replace at predictable intervals.
The capital cycle in many systems is annual budgets plus episodic grants.
Even with competent management:
the renewal window may coincide with a tight budget year or political reprioritisation,
replacement is deferred,
throughput drops,
risk accumulates invisibly until a crisis occurs.
This looks like mismanagement after the fact, but the failure was structurally induced by cycle mismatch.
How this differs from “just incentives”
A reasonable objection is that this reduces to incentives: actors respond to short-term rewards and constraints.
My disagreement is narrow but important:
Incentives operate within a temporal architecture.
If capital access itself is gated by short cycles, no incentive scheme can reliably enforce long-cycle behaviour.
I would update away from this view if there are strong examples where incentive reform alone preserves capability across multiple renewal cycles without changing how capital timing is governed.
Epistemic status
This is a provisional model.
I’m moderately confident that temporal misalignment explains some observed institutional failure modes, especially those clustered around missed renewal boundaries. I’m less confident about how dominant this mechanism is relative to incentives, information problems, or culture.
I’m particularly uncertain in cases where:
mission cycles are poorly specified,
renewal needs are genuinely ambiguous,
or fragility cycles are unusually stable.
I’m posting this to see whether the framing feels useful, redundant, or wrong.
Why I think this generalises
Although I’ve used a hospital example, the same pattern appears in other domains:
infrastructure maintenance,
scientific equipment renewal,
climate adaptation assets,
long-horizon organisational capability.
In alignment terms, this looks like a general failure of systems to stay aligned with their objective over time when control variables operate on mismatched cycles.
What I’m hoping for from comments
Pointers to prior work that already formalises this better.
Counterexamples where short-cycle capital governance does not produce decay.
Critiques of whether “cycle misalignment” adds explanatory power beyond incentives.
AI assistance disclosure: This post was drafted and edited with the assistance of a large language model. All ideas, claims, and examples are my own; the AI was used for structuring and clarity only.
I think some institutional failures that are usually framed as incentive problems are better understood as temporal misalignment problems: capital and decision-making are governed by short, volatile cycles, while the systems they are meant to sustain operate on much longer, more stable cycles.
When capital is structurally tied to short “fragility cycles” (elections, annual budgets, refinancing windows, donor attention), institutions predictably decay—even with competent leadership and good intentions—because renewal and maintenance must occur on longer “mission cycles” (asset lifetimes, replacement windows, capability formation).
This post sketches a simple model for that failure mode.
Why this might be relevant to LessWrong
Many alignment and coordination problems share a common pattern:
optimisation pressure is applied on the wrong timescale, producing outcomes that undermine the stated objective.
This shows up in:
I’m proposing that time—specifically, which cycles govern capital and decisions—is often the hidden variable.
A minimal model: fragility cycles vs mission cycles
I’ll use a very lightweight vocabulary.
Cycles
A cycle is a recurring temporal structure with:
Institutions usually operate under multiple overlapping cycles.
Fragility cycles
Fragility cycles are short, volatile, and largely exogenous. Across domains, a few show up repeatedly:
These cycles fluctuate in ways the institution does not control.
Mission cycles
Mission cycles are longer and intrinsic to what the institution is trying to do:
These cycles are usually more predictable and less volatile.
Empirically, the key pattern is:
The failure mode: capital follows the wrong cycle
In many systems, access to capital is effectively governed by fragility cycles. Informally:
This creates a structural problem:
Importantly, this can happen without bad incentives or corruption. Everyone may be acting reasonably given the constraints.
A concrete example
Consider a hospital with imaging equipment that needs replacement every ~4 years.
Even with competent management:
This looks like mismanagement after the fact, but the failure was structurally induced by cycle mismatch.
How this differs from “just incentives”
A reasonable objection is that this reduces to incentives: actors respond to short-term rewards and constraints.
My disagreement is narrow but important:
I would update away from this view if there are strong examples where incentive reform alone preserves capability across multiple renewal cycles without changing how capital timing is governed.
Epistemic status
This is a provisional model.
I’m moderately confident that temporal misalignment explains some observed institutional failure modes, especially those clustered around missed renewal boundaries. I’m less confident about how dominant this mechanism is relative to incentives, information problems, or culture.
I’m particularly uncertain in cases where:
I’m posting this to see whether the framing feels useful, redundant, or wrong.
Why I think this generalises
Although I’ve used a hospital example, the same pattern appears in other domains:
In alignment terms, this looks like a general failure of systems to stay aligned with their objective over time when control variables operate on mismatched cycles.
What I’m hoping for from comments
AI assistance disclosure:
This post was drafted and edited with the assistance of a large language model. All ideas, claims, and examples are my own; the AI was used for structuring and clarity only.