> For the complete documentation index, see [llms.txt](https://docs.contributionism.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.contributionism.io/worked-examples-and-evaluation/sensitivity-to-different-parameters.md).

# Sensitivity to different parameters

An arrangement may appear fair under one set of assumptions and behave very differently when those assumptions change. Sensitivity analysis examines how outcomes respond to different parameters.

## **Temporary rights duration**

Compare several durations, such as three, five and ten years. Longer participation can:

* Improve recognition of delayed impact.
* Increase historical claims on current surplus.
* Slow changes in voting weight where governance is linked.
* Make administration more complex.

Shorter participation can:

* Increase the relative weight of recent contributions.
* Reduce historical recognition.
* Encourage attention to near-term distributions.
* Leave long-development projects poorly supported.

No duration is universally correct.

## **Risk adjustments**

Test outcomes with and without risk multipliers. Examine:

* Total expected compensation.
* Time required to satisfy claims.
* Founder concentration.
* Later contributor participation.
* Default exposure.

Large multipliers can preserve early dominance even where rights are technically temporary. The analysis should also test whether risk is being rewarded more than once through multiple mechanisms.

## **Distribution and reserve policies**

Compare policies that retain different proportions of available resources. Higher retention may support resilience and growth, but it can reduce payments before expiry. Lower retention may benefit current rights holders while weakening the organisation. The model should not evaluate distributions without considering the resulting cash position and investment capacity.

## **Workforce growth**

Test different scenarios such as no growth, gradual growth, rapid growth and workforce decline. Hiring can materially change participation percentages. A system that works for a stable five-person organisation may produce very different outcomes when hundreds of new contributors join.

## **Compensation levels**

Test different combinations of:

* Current pay.
* Deferred obligations.
* Contingent bonuses.
* Shared benefits.
* Compensation rights.

An arrangement with high contingent rewards may be inaccessible to people who need reliable income. Financial inclusion should therefore be considered alongside average returns.

## **Voting rules**

Compare:

* Equal voting.
* Contribution weighting.
* Hybrid weighting.
* Voting caps.
* Separate representation.

Measures may include:

* Largest individual voting share.
* Concentration among the largest holders.
* Representation of recent contributors.
* Ability of affected minorities to block adverse changes.
* Administrative effort.

A more equal distribution is not the only objective; decision quality and accountability also matter.

## **Valuation uncertainty**

Contribution values should be varied where assessments are uncertain. For example, a model might compare outcomes if a major contribution is valued:

* At the lower end of a reasonable range.
* At the central estimate.
* At the upper end.

If small valuation changes repeatedly reverse leadership control or produce very large payment differences, the system may be too sensitive for its measurement accuracy.

## **Combined scenarios**

Changing one parameter at a time helps isolate effects. However, real conditions often change together. Revenue may fall while financing costs rise and key workers leave. The organisation should therefore also test combined downside scenarios.

## **Reporting sensitivity honestly**

A useful report identifies:

* Which outcomes are robust.
* Which depend heavily on assumptions.
* Which parameters create the largest changes.
* Which groups bear the downside.
* Which choices remain matters of preference.

The objective is not to identify one mathematically perfect configuration. It is to understand the consequences of the available choices.