APY starts with emissions, not with a landing page headline
A staking APY that is chosen for marketing usually ends up being paid for with future sell pressure. The major proof-of-stake networks all encode this in their reward logic. Solana states directly that staking yield depends on inflation, the total amount of SOL staked, and validator uptime and commission. Cosmos Hub targets a 67% bonded ratio and adjusts inflation between 7% and 20% as participation moves around that target. Polkadot uses an ideal staking rate and routes a defined share of annual DOT expansion to stakers rather than promising a static yield.
That is why the number on the landing page is the last thing you decide in token economy design, not the first. First decide how many tokens can be issued. Then decide how much of that budget should go to stakers instead of the treasury, ecosystem incentives, or other sinks. Then estimate how much supply will actually stake. Only after that do you get to a defensible headline APY.
Staking rewards design is an engineering problem disguised as a marketing decision. In token economy work, the hard part is not picking an attractive percentage. The hard part is making that percentage survive contact with real participation, real unlocks, real validator fees, and real secondary-market liquidity.
The sustainable APY equation is simple, but the inputs are not
Model APR first. Market APY later. The base equation is straightforward: annual reward budget to stakers divided by expected staked supply. The difficulty is choosing realistic inputs and keeping the budget constant when behavior changes.
| Step | Formula | What it means |
|---|---|---|
| 1. Define emissions budget | Annual token emissions = E | Total new token issuance the protocol can afford in a year. |
| 2. Carve out staking allocation | Reward budget to stakers = E × s | s is the share of emissions reserved for staking rewards. |
| 3. Estimate participation | Base staking APR = (E × s) ÷ r | r is the expected staking ratio, measured against the relevant eligible supply. |
| 4. Add lockup premium correctly | APR for tier i = Reward budget × multiplier i ÷ weighted stake | Longer lockups should get more weight inside the same pool, not a separate uncapped emissions stream. |
| 5. Convert to user-facing yield | Net yield = gross APR - fees - expected penalties | Validator commission, operator fees, and slashing risk matter. |
This structure matches how live networks already work. Solana distributes inflationary issuance to delegated stake accounts and validators, and makes clear that yield is driven by the stake fraction plus validator commission and uptime. Cosmos distributes both newly created tokens and transaction fees to validators and delegators. Ethereum’s per-validator base reward falls as total active balance rises, even though aggregate issuance grows with the square root of active stake.
A simple example makes the point. Assume annual emissions equal 5% of supply. Assume 80% of that goes to stakers. The annual staking budget is therefore 4% of supply. If you expect 50% of supply to stake, base gross APR is 8%. If participation rises to 70%, the same budget only supports 5.71%. Nothing broke in the contract. The model is doing exactly what it should. The team just advertised a number as if the denominator could not move.
Lockup premiums should be layered on top of that base rate through weighted accounting, not wishful thinking. If flexible staking has a 1.0x weight and a 12-month fixed lock has a 1.3x weight, the protocol should distribute the same reward pool over weighted stake. That keeps total emissions constant. It also makes the trade-off visible: higher yield for longer duration, lower immediate float for the market, and lower liquidity for the user.
APR and APY are not interchangeable here. Solana auto-redelegates rewards into active stake each epoch. Polkadot lets rewards be bonded immediately or paid out as transferable balance. Ethereum sends reward balances above 32 ETH out through automatic partial withdrawals every few days, which changes the compounding path for solo validators. If the product page says “APY,” the protocol should know exactly how compounding is supposed to happen.
Staking-ratio drift is where most APY models break
The most common staking-design mistake is holding APY constant while letting participation float. That is backwards. In practice, participation moves first and yield adjusts second. Solana says this plainly: the annualized figure varies each epoch because inflation and total active stake continually change. Polkadot’s reward curve is explicitly tied to the gap between actual staking and the ideal staking rate. Cosmos changes inflation when the bonded ratio moves away from two-thirds.
This is not just a math issue. It is a market-structure issue. When rewards are automatically restaked or frictionlessly re-bonded, staking participation can ratchet up in quiet markets and compress yield for everyone. When yield compresses enough, marginal stakers exit. That creates a second-order liquidity event because unstaking converts security collateral back into float. If those rewards were already being sold to cover taxes, costs, or treasury needs, the extra liquid supply arrives exactly when the yield narrative is weakening.
Dynamic reward curves exist to manage that feedback loop. Cosmos changes inflation when bonded ratio moves away from two-thirds and pushes it down toward 7% when bonded stake is too high. Polkadot’s ideal staking rate can vary between 45% and 75%, and the reward rate falls when staked supply exceeds that target. Ethereum reaches the same conclusion through a different mechanism: each validator’s base reward is inversely related to the square root of total active balance.
The liquidity point matters more than the headline percentage. A late-2025 Solana governance proposal argued that excessive emissions create persistent downward price pressure and raise the opportunity cost of using SOL in DeFi because native staking yield behaves like a risk-free rate for capital allocation. Even if a team disagrees on the exact level, the framing is correct. Reward emissions are not abstract dilution. They are recurring supply flows that have to clear in actual markets.
Lockups, penalties, and restaking are liquidity controls before they are product features
Lockup design determines how fast staking rewards can become tradable supply. Ethereum full exits require a voluntary exit and then variable waiting time, while partial withdrawals are processed through a protocol queue. Cosmos makes delegators wait three weeks to retrieve ATOM through standard unbonding. Polkadot keeps stake locked for the unbonding period and ties that delay to slashing accountability.
Flexible staking should pay the base rate because it creates the shortest liquidity duration. Fixed-term staking can pay more, but only if the premium is funded inside the same reward budget. The right question is not “How high can we make the 12-month APY?” The right question is “How much extra duration does a 12-month lock buy us, and what weight should that duration get inside the pool?”
Penalty structures matter because they define whether the protocol is paying for actual security or just subsidizing passive inventory. Ethereum applies penalties and slashing for validators that fail to behave correctly. Cosmos explicitly warns that delegators share slashing risk with the validator they choose. Polkadot keeps tokens locked during unbonding in part to preserve accountability for past offenses.
Restaking deserves separate treatment because it adds a new risk budget. EigenLayer’s whitepaper says delegated stake can be subjected to slashing from modules an operator joins, and those operators can earn fees from both the Ethereum beacon chain and EigenLayer modules. That means restaking yield is not a free uplift to native staking APY. It is a distinct reward stream with distinct slashing domains and operator-selection risk.
Liquid staking changes the same equation from the other side by shortening effective duration. Cosmos Hub’s Liquid Staking Module allows already-staked ATOM to be instantly liquid-staked without waiting through the full 21-day unbonding period, while governance-controlled limits and validator-bond requirements are used to contain that added flexibility. From a market microstructure standpoint, that means liquidity shock transmission is faster, even if user experience improves.
What sustainable APY ranges actually look like
Most sustainable ranges are lower than the market prefers to advertise. The broad pattern is consistent across tokenomics design work: native-security staking usually settles in the mid-single digits, DeFi staking can support double digits when it is tied to fees or explicit duration, and bootstrapping programs can go much higher only if the market is told up front that the rate is temporary.
| Protocol type | Typical gross APY range | What makes it sustainable | What usually breaks it |
|---|---|---|---|
| L1 native staking | 4% to 8% | Explicit inflation curve, known staking target, validator economics covered, bounded unlock mechanics | Promising a fixed rate while staking participation rises or emissions decay faster than modeled |
| DeFi staking or governance locking | 10% to 30% | Time-boxed emissions, fee support, lockup-weighted rewards, real penalty or forfeiture mechanics | Paying everything in newly minted token emissions with no demand sink |
| New launch bootstrapping | Higher, but time-limited | Hard decay schedule, finite runway, clear disclosure that the rate is not steady state | Marketing the launch subsidy as a durable equilibrium |
The L1 range is defensible because major networks already anchor yield to inflation logic rather than pure marketing. Solana’s schedule starts at 8% inflation and disinflates by 15% per year toward 1.5%, while realized staking yield still varies with stake participation and validator commission. Ethereum’s validator reward function also compresses individual returns as active stake rises.
DeFi can support higher rates, but the sustainable versions usually pay for duration, governance utility, or fee share rather than pretending idle tokens deserve venture-style yield forever. Curve is the clean example. Its vote-escrow design lets CRV be locked for up to four years, makes voting power proportional to lock time, and uses that locked weight to boost reward streams. That is a duration market. It is not magic yield.
New launches can go higher because they are buying bootstrapping. That can be rational. It is still temporary. If the schedule is not explicitly time-limited, the protocol is usually borrowing demand from the future and paying for it with emissions that the secondary market will later have to absorb.
The failure mode is obvious once you write the model down
The failure mode is APY set to attract TVL with no inflation model. The symptoms are always the same. Participation assumptions are missing. Flexible and fixed terms are both overpaid. Reward compounding is hand-waved. Sell-through is ignored. Liquid staking is added later without re-running the float model. Then the treasury or governance process is asked to solve a problem that was created at launch.
When that happens, teams usually end up with one of two outcomes. The first is emergency parameter changes. The second is a staking-rewards death spiral where emissions increase circulating supply, market absorption weakens, price falls, the real value of rewards falls with it, and the protocol raises nominal APY again to keep TVL from leaving. Protocols do revise reward systems in production. Polkadot’s documentation notes that the DOT model changed in November 2024 from one inflation structure to another through governance. Solana governance has also debated faster disinflation because issuance and validator economics pull in opposite directions. Those debates are normal. They are not substitutes for an initial model.
A disciplined staking design process is short to describe and hard to execute. It is one part of proper tokenomics:
- Choose an annual emissions budget that the token can survive.
- Choose the staking share of that budget.
- Model staking-ratio scenarios, not one staking ratio.
- Specify whether the public number is gross APR, net APR, or APY with defined compounding.
- Fund fixed-lock premiums with weighted accounting inside the same pool.
- Model unlock cadence, claim cadence, and likely sell-through by holder type.
- Treat liquid staking and restaking as changes to liquidity and risk, not as free product extensions.
- Decide governance powers up front so “temporary” emergency edits do not become policy by habit.
At FinDaS Tokenomics, this is where tokenomics consulting stops being narrative work and becomes engineering. Sustainable token economy design means running stake-ratio bands, lock-mix scenarios, commission assumptions, and post-reward sell-flow cases before any headline yield goes live. Getting this wrong is rarely recoverable without a token migration. That is why staking APY design belongs with a tokenomics expert or token economy advisor, not with a quick prompt and a pretty dashboard.
