assembly-economics/models/GADM-0.3-Candidates.md

GADM-0.3 Candidates — Life-Cycle and Power Extensions

Status: outline. Nothing here is implemented.

GADM-0.1 § 26 plans for GADM-0.3: age cohorts, household distribution, wealth separated from power, common versus private ownership, and adoption modeled separately from attention. GADM-0.2 already added attention headroom A. This note outlines the two extensions INTENT.md asks for, plus the items GADM-0.2 deferred.

---

1. Life-cycle extension

1.1 Cohorts

Replace the single population with four cohorts c \in {children, working-age, seniors, very old/dependent}. Each cohort carries its own H_c, G_c and A_c. Commons, productive capacity, P and E stay society-wide.

1.2 Flows between cohorts

Flow	From → to	Mechanism
Ageing	c → c+1	demographic transition rates
Capability formation	society → children	education and care. Care is generative infrastructure (concept § 9).
Intergenerational transfer T_g	seniors → working-age, children	mentoring, tacit knowledge, archives (GADM-0.1 § 18)
Dependency	very old/dependent ← society	care demand draws on working-age attention A
Organism mediation	K_Ω → transfer	organisms can preserve tacit knowledge (inheritance) or displace mentoring (deskilling)

1.3 Hypotheses to test

• L1. Strong T_g lets an ageing society keep H constant (GADM-0.1 § 11 interpretation).
• L2. Care demand competes with other uses of attention headroom. Without recognized care participation (C_p), A falls fastest in the working-age cohort.
• L3. Children's capability formation responds with a lag of 15–20 periods. Effects of policy on H appear only beyond the current 100-period horizon unless the time scale is fixed.

1.4 Prerequisite

Fix the time unit: one period = one year is the natural choice once cohorts exist.

---

2. Power extension

2.1 Separate wealth from power

Introduce wealth concentration W_c as a stock distinct from P:

[ \dot W_c = \text{private return concentration} - \text{taxation/diffusion} ]

P then gains a conversion term from wealth into control, with a coefficient that depends on institutional checks:

[ \dot P \mathrel{+}= \kappa(Q, S),W_c(1-P) ]

This allows regimes with high wealth inequality but bounded power (strong checks), and with moderate wealth but extreme power (platform or membrane control). GADM-0.2's sensitivity results suggest the second kind matters most for information organisms.

2.2 Channels of P as sub-stocks

Split P into channel sub-stocks, each with its own drivers and counters:

Channel	Drivers	Counters
Market	network effects, scale	competition, entry
Infrastructure	control of essential facilities	interoperability, public provision
Discovery / attention	ranking, recommendation	plural discovery, attention protection
Membrane	operator control of organism membranes	user/community control (m), portability
Lineage	ownership of high-use organism lineages	lineage diffusion (L), open inheritance
Political / institutional	lobbying, capture	transparency, checks, revisability

Aggregate P = 1-\prod_k (1-P_k), so any single dominant channel suffices for high P.

2.3 Hypotheses to test

• P1. The success → control loop (R3) becomes self-sustaining above a threshold of channel concentration. Find that threshold per channel.
• P2. Membrane and lineage channels interact with R7 (learning with use) to form a natural-monopoly regime unless L exceeds a critical value (mapping note, open question 1).
• P3. Wealth taxation alone does not bound P when the membrane channel dominates.

---

3. Items deferred from GADM-0.2

Item	Source	Note
Compute C as a stock	GADM-0.2.md § 1	Needed once energy classes are modeled (GADM-0.4).
Participation claims tied to contributions	GADM-0.2.md § 6	Requires the WS-14 participation model; currently a fixed share.
Global sensitivity analysis	GADM-0.2.md § 5	Morris screening first, then Sobol indices on the top parameters.
Stronger attention dynamics	GADM-0.2.md § 6	aext appears underestimated; calibrate against time-use evidence.
Organism-aware scenario grid	InformationOrganismsInGADM.md § 5	Full 2 × 2 × 2 grid of lineage, membrane and attention regime. GADM-0.2 runs two corners.

---

4. Suggested order

1. Fix the time unit, and add global sensitivity analysis to the existing model (cheap, and it improves every later step).
2. Power extension § 2.2 (channel sub-stocks), because GADM-0.2's sensitivity points there.
3. Wealth separated from power (§ 2.1).
4. Cohorts (§ 1), the largest structural change.