Cardano Research: Big Reserve no go alone secure Proof-of-Stake

One peer-reviewed paper by Cardano Foundation researcher Manvir Schneider and Paolo Penna of Input Output Global (IOG) comot on October 2 for di official AFT 2026 proceedings, for Schloss Dagstuhl LIPIcs series. Di Foundation present am for one post on October 13. Di paper use Cardano as example of network wey dey fund rewards from fees and one reserve wey dem dey spend small small. Di authors conclude say make people no dey judge reserve spending policy only by di date wey e go nominally finish.

Economic Security and di Reserve Threshold for di AFT 2026 Paper

Di study, "Reserve Depletion and Security Runway in Proof-of-Stake Systems," dem present am for di 8th Conference on Advances in Financial Technologies for London. E focus on economic security, using one economic model to check how much secure capacity validators fit provide when their rewards dey come from fees and di reserve while token price and demand dey change. Di Cardano Foundation na Swiss nonprofit wey dey support Cardano development and adoption, and IOG na di engineering and research company behind di Ouroboros protocol.

Di model describe how network fit move from reserve funding go fee funding. When network still young, fees no dey cover di cost of di target security level, so token reserve go pay di difference until usage grow reach point wey fees fit take over. Security runway na di time wey di reserve, at current token price and demand, still fit sustain dat target. According to di paper, dat period no dey end when reserve don empty, but when e first become too small to cover di shortfall.

Di authors call di line between reserve wey enough and reserve wey no enough di reserve threshold: di smallest amount wey still fit fund di target under given market conditions. Di threshold divide possible network states into three zones. When demand weak too much, no amount fit sustain di target. For di middle zone, security depend on subsidy from di reserve, while for di third zone di threshold na zero because fees alone dey fund security.

Cardano Reserve, di ρ Parameter and Stake Pool Rewards

According to Cardano monetary policy documentation, fees from all blocks for one epoch dey go into one reward pot, wey also dey receive fixed percentage of di remaining ADA reserve. Dat percentage, di ρ parameter, dey set at 0.3 percent per epoch. At dat rate, di documentation talk say, half of di remaining amount go spend within four to five years. From di reward pot, 20 percent (di τ parameter) dey go treasury, and di rest dem dey distribute as rewards in ADA.

Di study no model Cardano rules directly but one general case with two parameters: di share of fees wey return go di reserve and di share of di reserve wey dem dey pay out each period. Validators for di model dey pay their costs for external currency, like dollars, while di reserve dey held for di native token. Because of dat, di authors note say di security level wey reserve fit fund depend on di number of tokens multiplied by their price, no be only di token count.

To reach di zone wey fees alone go fund security na different logic. According to di paper, e depend on demand and no be token price, because di fees wey users dey pay don already dey measured for di external currency.

Stress Test for Protocol Governance and di Model Limits

Theorem 15, wey di authors talk say dem intend am for protocol governance, dey translate pessimistic assumptions about price and demand into minimum reserve size for one given time horizon. According to di theorem, if di reserve meet dat condition, security go maintain for every price and demand path wey stay above di chosen floors. Na conservative estimate because di test ignore future fee inflows into di reserve, and di guarantee hold within di assumptions of di theorem.

Di authors still outline di model limits. Price and demand dem take as given, so di paper no explain token valuation or adoption. Validators equal for di baseline version, na why dem talk make people read di threshold as benchmark. Di numerical examples use illustrative parameters, including five validators, and no be calculation for Cardano. Community members ask for exactly dis kind calculation for Cardano current state for replies to di Foundation post.

As next step, di authors propose one adaptive policy wey payouts and fee retention go follow di state of di network, for example by cutting payouts after price drop or retaining more fees when reserve low. Protocol governance go then choose rule wey dey adjust to reserve size, price and demand, subject to incentive and credibility constraints. Di authors describe dis as move from evaluating existing policy to designing new one and name am as topic for their future research.