Repository navigation
Referral Network Rationale
The Referral Network combines two choices: fixed commissions based on descendant asset balances, and discretionary marketing for competing over children. Neither enlarges the global emission budget. Admission prices and marketing policies are participant choices within those rules.
Every earning ancestor receives the same fraction
Each commission is calculated from the source's original attribution. With
A shared emission budget still creates dilution. Suppose Alice is the only asset holder, with 100 eligible units, and initially receives the entire lane's allocation. Another 100 units joining directly under her gives her 110 core units out of 200, or 55%. If that capital instead joined a different top-level branch, she would have 100 out of 200, or 50%. Recruiting improves her outcome relative to that alternative; it does not preserve her original 100% share.
These are fractions of the lane's available budget. The global emission rate also decreases continuously, so an unchanged percentage does not promise an unchanged credit income per unit of time.
A deeper node has both a lower own-asset retention fraction and less recruiting
room. The finite bound
Finite depth alone does not create scarce top positions. That scarcity comes from protocol-authorized auctions of depth-zero seats under a non-earning root. Winning payments burn credits. Ordinary parents separately price admission to their branches and may offer free entry to attract future participation.
Auctions limit the number of seats, not the number one beneficial owner may control. Seat scarcity does not itself prevent concentrated ownership or a monopoly over top-level positions.
Consider one owner's connected region, whose highest owned member has
Consequently, the group retains
This is a limited property, not proof of human identity or resistance to every
coalition strategy. For example, with
The example earns more from outside capital, so it does not contradict own-capital splitting neutrality. It does rule out a stronger claim that extra owned accounts can never improve an owner's referral income. Pooling capital above an outside intermediary or acquiring a different family's offer is also a different arrangement. Permanent edges prevent inserting an intermediary above an existing member; they do not identify which new accounts share an owner.
Permanent edges do not make the economic position of an intermediary permanent.
Suppose A is a top-level seat, B is A's child, C is B's child, and only C has
assets. At
This requires custody and redistribution that participants find acceptable; it does not require duplicated balances or a changed parent edge. Efficient pools at top-level seats could therefore undercut deeper positions. The model allows this competition, but cannot promise that multilevel intermediaries remain economically necessary or continue receiving capital from old recruits.
Equal rewards per child would reward manufacturing accounts. Weighting children by escrowed credits instead requires a real, exclusively held resource. Splitting 100 credits among ten sibling accounts gives the same combined share as staking 100 through one account. An empty account receives no family allocation.
Credits also provide one unit across asset lanes. The parent first earns its overrides in credits under the asset and allocation rules. It then divides its chosen marketing budget by immediate-child credit escrow, without comparing raw BTC, ETH, or SOL quantities again.
This budget gives siblings exposure to each other's recruiting. A nephew's assets can increase the grandparent's gross overrides without increasing the uncle's competing escrow denominator. That makes expected family growth relevant when shopping for a link.
Credit escrow does not itself generate additional commissions or emission. It competes for a fraction of someone else's already allocated income. Borrowing credits over real elapsed time is capital rental, not evidence of independent people or unleveraged wealth. Correct historical settlement prevents a temporary new stake from collecting an earlier pool.
Mutualization is a marketing budget. A parent may raise it to attract children, lower it when expected participation changes, or stop it. The payment rule determines how each elapsed interval is allocated; it does not guarantee a future percentage or a profitable campaign.
A parent can stake credits through child accounts it controls. That can
recapture some of its offered pool, just as reducing
Future admission revenue and reputation can make a generous offer worthwhile. They do not automatically make it permanent or prove that every parent will honor today's rate indefinitely. Children choose how much credit principal to commit and can withdraw it; their existing parent connections remain fixed.
An ordinary join fee is a price paid to a parent. A winning top-seat payment is burned. Escrow remains the child's withdrawable property. Core credits are earned from eligible assets, and family allocations reallocate gross overrides. These roles must not be counted twice.
This separation supports a simple market without account-count bonuses, permanent family shares, or mandatory marketing promises. It also makes the limits visible: global dilution, competing escrow, changing gauge allocations, and discretionary offers all affect what a member earns. See Mutualization for the mechanics and Emission Accounting for the chronology required to implement them.
AVM
AVM features
Linear AVM
Referral network
- Overview
- Rationale
- Specification
- Parameters and authority
- Family marketing
- Choosing a parent
- Diagrams
Credits
- Mayflower Credits
- Earning credits
- Channel weighting
- Credits and revenue clearing
- Long-term economics
Credit emissions
- Overview
- Emission rate and cumulative budget
- Lane allocation
- Participation and referral rewards
- Accrual, collection, and conservation
Revenue clearing