Points programs didn’t disappoint because incentives don’t work; they disappointed because their issuance was hard to model and their rewards were easy to game. Most incentive programs tend to break in predictable ways: they issue rewards without a hard budget, pay for metrics that are easy to game, or rely on volume signals that invite wash behavior. Tapir’s incentives design starts from the opposite premise: we expect & encourage users to game the system.

Every incentive program is a game for users to play. The issue is that most of them are just badly designed. They tended to reward metrics such as “usage” or “user growth” which, even if they could be measured in an anonymous system, didn’t seem to help the protocol in the long run.

In order to provide good UX (low spreads) we need deep enough liquidity and that is exactly what we incentivize. We have clear liquidity targets for each depeg pool and its associated market, and a control mechanism that dynamically adjusts rewards to achieve them.

Emissions are budgeted per pool, AMM rewards are linked to fees earned, and both streams are capped to prevent unwanted token dilution. This article explains the mechanism, why it was chosen, and how to participate without guessing.

Key Takeaways

• Modelable emissions, treating points as underwritable token issuance: 120M TPR points budget, per-pool caps, and clear rules.

• Incentives are organized by Depeg Pool: every pool has a corresponding DP/YB token pair plus an associated AMM that trades that pair. Both the pool and its AMM have concrete targets for each epoch.

• Two types of incentives: TVL rewards are time-bound, and AMM rewards are utility-driven (fee-based).

• Multipliers are the steering wheel to attract activity in a pool.

• Targets are the emission budget (hard cap); scaling enforces the cap if demand is too high.

• Target APRs: ~20–22% APR for passive TVL holders, ~20–22% APR for AMM LPs on fees, and ~40%+ combined APR for LPs earning both streams (before base yield and subject to pool utilization and scaling caps).

• Points convert to TPR at a fixed 1:1 ratio and are time-locked before conversion; you can earn quickly, but you cannot exit instantly. Emissions you can underwrite

It’s broadly recognized in DeFi that most points and incentives programs often pay for TVL and activity without filtering for durability or conviction. Behaving as though emissions were a no-cost lottery, airdrops tend to reward mercenary behavior: “paper TVL” first, post-TGE sell pressure later.

Emissions you can Underwrite

It’s broadly recognized in DeFi that most points and incentives programs often pay for TVL and activity without filtering for durability or conviction. Behaving as though emissions were a no-cost lottery, airdrops tend to reward mercenary behavior: “paper TVL” first, post-TGE sell pressure later.

The track record of points programs is mixed at best: misalignment, dilution, wash trading, opacity, fragile liquidity. Still, they exist for a reason, and Tapir’s design is an attempt to keep the benefits while fixing the failure modes. Protocols need a way to bootstrap liquidity and reward early risk-taking. Tapir is trying to do that in a way that allocators can actually model.

Tapir’s liquidity incentives pay for two things that the protocol can actually measure and the market can actually price. The idea is to optimize for both sticky and useful liquidity. We also cap emissions by design so dilution stays bounded and forecastable.

Tapir treats incentives as an underwriting problem: each pool has a weekly points budget, and if demand exceeds it, everyone takes the same proportional haircut. AMM rewards track fees earned rather than idle liquidity. Multipliers then reprice each pool weekly to keep participation near target. TVL incentives buy time-sticky participation (positions held through full UTC days) while AMM incentives buy useful liquidity (fees earned, i.e., trades actually settled).

We calibrate each pool to target the following incentive APRs (in TPR value):

• ~20–22% annualized on TVL holdings (DP/YB held through full UTC days).

• ~20–22% annualized on AMM liquidity (fees earned).

• ~40%+ combined for LPs earning both streams.

• Maximum target range: up to ~100% APR in exceptional bootstrapping phases.

These targets are pool-specific, and are implemented through:

• α (TVL multiplier) → controls points per dollar-day.

• β (AMM multiplier) → controls points per dollar of fees earned.

Under the weETH launch configuration, for example:

• α = 0.02 → ~21.8% TVL incentive APR at $0.03/TPR.

• β = 60 → ~19.7% AMM incentive APR under baseline turnover assumptions.

These APRs are targeted, not guaranteed. They are also subject to a pool’s demand based on the weekly emissions budget. Finally, these rewards are also independent of the underlying asset’s base yield (which remains fully productive).

Why TVL (Liquidity-based) Incentivization

TVL incentives are the protocol’s budget to onboard the first cohort of users that will make the product usable at scale. The goal is to grow the TVL, as DP and YB holders are the providers of pool’s inventory and risk capital. Designed to reduce “paper TVL” behavior, TVL points only accrue on full UTC days, with transfer resets. This will improve execution and capital efficiency.

Why AMM (Fee-based) Incentivization

Tapir’s AMM is the mechanism that makes the price of depeg protection a market variable. Each depeg pool has a corresponding market with a DP/YB AMM that must clear trades in the background to function. Without enough liquidity depth, users face slippage and cannot reliably move between “protection” (DP) and “yield” (YB).

Every DP or YB buy implies a sale of the other side. When a user chooses one exposure, the “undesired” exposure must be absorbed by the market. In Tapir, that routing happens through the pool’s AMM. Thin liquidity therefore turns the product into a spread problem.

By setting fee-based rewards, Tapir discourages wide, unused ranges and makes wash trading uneconomic. Moreover, LPs can earn both streams, as AMM LP positions count toward TVL accounting as well.

How you earn: two streams, one principle

Tapir ensures users are rewarded for 1) held TVL and 2) fees earned. You earn 1) TVL points by holding DP and/or YB tokens in a given pool, and you earn 2) AMM points by providing liquidity to the pool’s AMM.

For TVL points, Tapir only counts a position if you hold it for a full UTC day. It takes snapshots of user balances at 00:00 UTC. Transfers reset the holding period for the transferred amount. Positions held for less than 24 hours earn zero for that day. Thus, daily TVL points are computed as (DP + YB balance) × asset price × a pool multiplier, but only if you hold through the full day.

Think of TVL points as:

(Your dollar exposure, per day) × (a pool-specific “points per dollar” rate).

Concretely, each day:

1. Tapir looks at your DP + YB balance at the daily snapshot.

2. It multiplies by the asset’s USD price to get a dollar value proxy.

3. It multiplies by a pool TVL multiplier (call it α) to turn dollars into points.

4. If you didn’t hold the full day, the result is forced to 0.

Then Tapir sums those daily values across the days in the epoch to get your raw TVL points for the week.

For AMM points, liquidity providers are rewarded based on fees they actually earn, not on liquidity size. This is a choice to avoid paying for liquidity that never trades.

Think of AMM points as:

(Your fees earned this week) × (a pool-specific “points per fee-dollar” rate).

Concretely, for the epoch:

1. Tapir totals the USD fees you earned as an LP in that pool (call this F).

2. It applies a pool AMM multiplier (call it β) to convert fees into points.

3. It multiplies by the underlying asset’s USD price at epoch end.

Together, the program optimizes for removing the incentive to post wide, unused liquidity ranges, tying rewards to protocol utility. It makes wash trading expensive because fees are real costs, and avoids last-minute time-weight gaming.

Caps and adaptive targets

Each pool has two per-epoch targets set by governance: a TVL point target (max points paid for holding DP/YB), and an AMM point target (max points paid for LP fees). These targets define the pool’s points budget per epoch. If a pool overshoots its target, Tapir scales everyone down to the budget; if it undershoots, it simply pays full (unscaled) incentives to everyone.

• Pool TVL target: 1,000,000 points

• Raw points generated: 1,800,000

• Scaling factor: 1,000,000 / 1,800,000 ≈ 0.556

• Each participant’s final points get multiplied by ~0.556, and the pool distributes approximately 1,000,000.

Therefore, your points depend on your share of activity and how crowded the pool is relative to its target. If three holders had raw points of 800,000 / 600,000 / 400,000, then final points become roughly 444,444 / 333,333 / 222,222, totaling approximately 1,000,000.

Each pool has two multipliers that adjust every epoch:

• α for TVL points; how many points you earn per $ of TVL held (per day).

If you assume a reference value per point, say $0.03/TPR for modeling, you can translate α into an implied APR: α × 7 × 52 × (TPR price). For instance, in Tapir’s weETH pool, where α is set to 0.02, it would be targeting ~20–22% reward APR in token value.

Thus, $1,000,000 held for 7 days earns 0.02 × 1,000,000 × 7 = 140,000, where token value ≈ 140,000 × $0.03 = $4,200 ($4,200 / $1,000,000 ≈ 0.42% weekly ≈ 21.8% APR).

• β for AMM fee points; how many points you earn per $ of AMM fees earned.

Following the example of Tapir’s weETH pool, assuming 30% of TVL in the AMM and 10% daily turnover (volume/TVL) with a 0.3% fee tier, the expected weekly fees can be modeled as: avg AMM TVL × turnover × 7 × fee tier. In the weETH configuration, where β is set to 60, it’s also calibrated to produce ~20–22% reward APR on the AMM portion under those assumptions.

If a pool undershoots its target, the multiplier rises in the next epoch. If it overshoots, the multiplier falls. This allows Tapir to react to true demand, dampening adjustments with γ = 0.5 by default (i.e., the new multiplier moves only halfway toward the fully implied adjustment, smoothing epoch-to-epoch volatility). Governance can set γ anywhere in the 0.1–1.0 range, and multipliers are bounded to prevent extreme outcomes: α ranges 0.005–1.0; β ranges 1.0–1,000.0.

How Big Is The Reward Boost

Multipliers adjust each epoch based on over/undershoot.

• Overshoot → incentives get sparser next epoch.

• Undershoot → incentives get more plentiful.

The system tries to converge on “enough liquidity, without runaway emissions.” Thus, in Tapir, the “boost” can only come from two levers:

1. Higher initial multipliers (α and/or β)

   1. TVL: raw points scale linearly with α.

   2. AMM: raw points scale linearly with β (raw AMM points = β × fees).

2. Higher per-epoch pool targets (the cap)

   1. Targets determine whether participants get haircutted by oversubscription scaling (κ). Higher targets reduce the chance of κ < 1.

If the pool is not oversubscribed (κ ≈ 1), increasing α by 50% increases TVL points by ~50%; increasing β by 50% increases AMM points per fee dollar by ~50%.

On the contrary, if the pool is oversubscribed (κ < 1), raising targets can matter as much as raising multipliers because it pushes κ back toward 1.

Private mainnet launches next week with boosted rewards for early adopters. The boost is implemented by setting higher initial pool parameters for the first epochs: TVL rewards use elevated initial α and AMM rewards use elevated initial β values, alongside higher per-epoch pool targets. Exact boost parameters will be published at mainnet launch.

Practically, this increases points per dollar-day (TVL) and points per dollar of earned fees (AMM), subject to the same pool-level emission caps and scaling if demand crowds in.

You can earn quickly; you cannot exit instantly

Tapir time-locks points before conversion to TPR. This allows it to spread out unlocks over time and reduce one-day supply shocks. It also forces participants to underwrite the protocol’s medium to long-term path, not just its launch.

• Earliest earners face a 90‑day cliff from TGE.

• Points earned before TGE unlock smoothly across TGE+90 to TGE+180 (linear schedule based on earn date).

• Points earned on/after TGE follow a 180‑day lock from earn date.

• Points remain non-transferable until you convert them to TPR.

As an allocator, you can choose your strategy based on your desired exposure: a) TVL (DP/YB holding) fits allocators who want lower operational overhead and can hold through daily cutoffs, b) AMM LPing (fee-based points) fits allocators who can manage ranges, inventory, and rebalancing, or c) for running both in parallel.

LPs earn both TVL points (for tokens in LP position) and AMM points (for cumulative fees earned):

This makes LP provision siginificantly more attractive then passive holding, incentivizing deep liquidity

For each pool you deposit, track:

• TVL and AMM targets (points budget).

• Whether the pool is oversubscribed (scaling factor <1).

• α and β multipliers and their direction of change (undershoot → up; overshoot → down).

• Your own unlock timeline (pre‑TGE vs post‑TGE earn dates).

The Takeaway

Tapir treats points as a controlled distribution system, not a marketing meter. It uses:

• Fee-based rewards to pay for liquidity that clears trades.

• Time-based eligibility to block snapshot games.

• Pool-level targets and scaling to cap emissions.

• Adaptive multipliers to steer participation without blowing out dilution.

If you allocate, you do not need to guess the program’s intent. You can read the budget, watch oversubscription, and size positions based on the same inputs Tapir uses to calculate rewards.

Public testnet is live. Use it to get the opportunity to participate in the private mainnet, get familiar with weekly targets and crowding haircuts.

References

Tapir Protocol Docs

Tapir Protocol Litepaper

Tapir Protocol: Point System & Liquidity Incentive Specification

Tapir Protocol: weETH Pool Configuration v1.1

What is Depeg?
A depeg occurs when a token trades below the price it is designed to track—for example, a USD stablecoin dropping to $0.95.

Tapir is an on-chain structured product that splits a yield-bearing position into senior and junior claims, enabling participants to pay (or earn) an explicit depeg protection without relying on an underwriter’s balance sheet. The goal is to offer counterparty-minimized protection with transparent, pre-defined settlement.

In Traditional Finance, this confidence rests on regulation, capital requirements, supervision, and resolution frameworks. In DeFi (Decentralized Finance), many risk-transfer products operate without comparable ex-ante supervision or ex-post enforcement (stablecoin depegs, LST discounts, etc.). This creates a structural tension between capital efficiency and counterparty risk that existing models struggle to resolve.

This article compares two main risk mitigation models and introduces the Risk Trilemma framework.
It examines how tranching mechanisms can achieve trust-minimized risk protection without sacrificing capital efficiency and why it is much better suitable for the DeFi environment.

Underwriting Models: A Taxonomy

Risk underwriting falls into two categories, each with distinct capital structures and counterparty risk profiles. We focus on (I) balance-sheet underwriting and (II) fully collateralized underwriting. Mutualized risk pools and discretionary-claims models introduce different design constraints (governance, adjudication, operational processes) and are not addressed here.

Balance Sheet Underwriting

Balance sheet underwriting is the dominant model in traditional risk underwriting. The underwriter assumes liability on its own balance sheet without designating specific assets against individual policies. Claims are satisfied from general capital reserves, with no recourse to ring-fenced collateral.

This model is effective when claims are diversified and weakly correlated. A single unit of capital can support multiple policies simultaneously as long as tail events are rare and uncorrelated across the book. When claims become correlated (market-wide stress, shared collateral types, systemic depegs), effective leverage decreases, and solvency becomes the binding constraint.

The principal limitation is counterparty risk. Policyholders have limited visibility into the underwriter’s risk management, asset-liability matching, or exposure concentrations. In correlated-claim scenarios—systemic events, market-wide depegs, or cascading liquidations—the underwriter’s capacity to honor obligations becomes uncertain. The policyholder’s claim is an unsecured liability against the underwriter’s general estate.

Fully-Collateralized Underwriting

Fully-collateralized underwriting inverts this structure. The underwriter posts designated collateral against each policy, granting the policyholder direct recourse to earmarked assets upon a qualifying event. This approach mirrors the funded structures commonly employed in insurance-linked securities and collateralized reinsurance markets.

This model eliminates counterparty risk, since the protection buyer holds a secured claim against identifiable collateral, verifiable on-chain in DeFi contexts. Settlement becomes deterministic rather than dependent on the underwriter’s solvency or willingness to pay.

The cost is capital intensity. Coverage capacity is constrained by posted collateral and associated liquidity terms (e.g., withdrawal and settlement mechanics). While collateral may be yield-bearing, the binding constraint is still funding: protection capacity must be financed up front, and that cost is ultimately reflected in premiums.

The Risk Trilemma

These two models represent opposite ends of a spectrum, each optimizing for different properties at the expense of others. In regulated markets, balance sheet underwriting predominates because regulatory infrastructure provides quite effective mitigation of counterparty risk. Solvency requirements, statutory reserves, and supervisory oversight create accountability mechanisms that substitute for direct collateralization.

Decentralized finance lacks an equivalent enforcement infrastructure. Smart contracts can enforce rules, but they cannot compel an undercapitalized counterparty to produce funds that do not exist. This environment surfaces a fundamental trilemma—three desirable properties that prove difficult to achieve simultaneously:

1. Capital Efficiency: The ability to provide coverage without requiring equivalent collateral lockup, preserving capital productivity

2. Counterparty Risk Elimination: No reliance on underwriter solvency, creditworthiness, or willingness to perform

3. Coverage assurance: Assurance that protection against the claim will be honored. Both underwriters will have the intent and the means to honor their claim

Balance-sheet underwriting is capital efficient but introduces credit exposure to the underwriter. Fully collateralized structures reduce credit exposure but impose higher capital intensity and liquidity constraints.

The question becomes whether an alternative structure can optimize across all three dimensions.

Tranching as a Resolution Mechanism

Structured finance has long employed tranching to redistribute risk across investor classes with heterogeneous risk appetites. The same principle applies to decentralized risk protection markets.

Tapir Protocol implements this approach by bifurcating yield-bearing tokens into two subordinated tranches:

• DP (Depeg Protected): Senior tranche entitled to par redemption, protected against depeg events up to a defined threshold

• YB (Yield Boosted): Junior tranche absorbing first-loss exposure in exchange for enhanced returns

Both tranches maintain claims on identical underlying collateral. Both continue to accrue the underlying yield throughout the protection period. The distinction lies solely in the loss waterfall: depeg losses are allocated first to YB holders until their tranche is exhausted, with DP holders bearing residual losses only thereafter.

The market price differential between DP and YB tokens reflects the implied cost of protection—effectively the premium that risk-averse participants pay to risk-tolerant participants for assuming subordinated exposure.

Mechanism Properties

This tranched structure delivers several properties that address the Risk Trilemma:

Counterparty Risk Elimination. Both DP and YB positions represent claims on actual underlying collateral held within the protocol. No party relies on an external underwriter’s balance sheet or solvency. Settlement is deterministic, governed entirely by smart contract logic and observable price feeds.

Full Collateralization of Senior Tranche. DP holders maintain a secured position backed by the combined underlying of both tranches. In a depeg scenario, their claim is satisfied from YB holders’ collateral allocation before any loss accrues to the senior tranche.

Risk-Adjusted Compensation for Junior Tranche. YB holders receive the underlying yield plus an implicit risk premium derived from the DP/YB price differential. This premium compensates for their first-loss exposure and elevated risk profile.

Capital Efficiency Through Continuous Yield Accrual. Unlike traditional collateralized protection models, where locked capital sits unproductive, both tranches remain fully deployed in the underlying yield-generating strategy. Neither protection buyers nor protection sellers sacrifice capital productivity. The protection mechanism operates as an overlay on existing yield positions rather than requiring dedicated reserve capital.

Algorithmic Claim Resolution. Settlement terms and depeg thresholds are codified at position inception, with claim execution governed entirely by smart contract logic and oracle price feeds. No manual adjudication, governance vote, or counterparty approval is required. This deterministic payout structure eliminates the operational and reputational risks inherent in discretionary claims processes.

Oracle Risk Mitigation. Deterministic settlement depends on price feed integrity. To mitigate oracle manipulation and single-point-of-failure risk, the protocol employs multiple independent oracle sources and applies price sampling across the five days preceding pool maturity. This time-weighted approach reduces exposure to transient price dislocations, flash-loan-driven manipulation, and individual oracle failures.

Structural Limitations

Because DP holders’ claims are secured against YB holders’ allocation, protection capacity is bounded by the junior tranche’s value. The protocol currently employs a symmetric 50/50 split between DP and YB tranches. Under this structure, the YB tranche can absorb depeg losses up to 50% of the high watermark price. Beyond this threshold, the junior tranche is fully depleted and residual losses pass through to DP holders.

The 50/50 ratio balances protection depth with junior tranche attractiveness. A larger junior tranche would extend coverage but dilute YB holders’ risk premium, reducing demand for the subordinated position. A smaller junior tranche would concentrate the premium but narrow the protection buffer. The symmetric split provides a natural equilibrium for the initial protocol version.

For most half-sane yield strategies, 50% buffer is more than enough. Higher losses represent extreme tail events typically associated with a fundamental protocol flaw—this coverage ceiling provides adequate protection for most risk management objectives.

Additionally, while smart contracts serve as the trust-minimisation layer that enables deterministic settlement, they introduce their own risk surface. However, Tapir meaningfully reduces aggregate smart contract exposure rather than adding to it. The smart contract trust assumption collapses into trust in a single codebase: Tapir’s own contracts(the core contract has 314 lines of code). Because the protocol protects against downstream depeg events, the security of the underlying yield platform becomes the hedged risk rather than an unhedged exposure.

This is especially relevant because the highest-yielding opportunities in DeFi are disproportionately available through newly launched protocols—platforms with limited track records and unproven codebases.

Without Tapir, an investor must perform rigorous due diligence on each new protocol, a burden that scales linearly with every opportunity pursued. With Tapir, due diligence on the smart contract dimension needs to be performed only once, rather than repeated for every new deployment.

Tapir’s safety is also independent of token economics, eliminating an entire class of economic exploit vectors common in yield-generating protocols. While smart contract risk cannot be fully eliminated, the protocol’s minimalism, auditability, and isolation from tokenomic complexity substantially reduce the attack surface relative to the multi-protocol exposure it replaces.

Comparative Analysis

The tranched model achieves a previously unavailable combination: counterparty risk elimination comparable to fully-collateralized structures, with improved capital efficiency relative to traditional collateralized models. Unlike balance-sheet underwriting—where capital efficiency derives from leveraging a single reserve across multiple uncorrelated policies—the tranching model preserves capital productivity through continuous yield accrual on all posted collateral. Neither protection buyers nor sellers sacrifice yield exposure to obtain or provide coverage. The cost is a bounded protection range rather than unlimited coverage.

* While we believe the 50% coverage is sufficient for most assets, subsequent protocol iterations are intended to extend protection capacity beyond the current junior-buffer ceiling—ultimately targeting full 100% depeg protection—subject to additional design and risk constraints.

Conclusion

The Risk Trilemma framework illuminates why traditional risk models translate poorly to decentralized environments. Balance sheet underwriting requires a trust infrastructure that does not exist on-chain. Full collateralization imposes opportunity costs that render protection economically unviable for many participants.

Tranching offers a third path. By restructuring claims on existing yield-bearing collateral rather than requiring dedicated capital reserves, protocols can deliver trustless risk transfer without sacrificing capital productivity. Both protection buyers and sellers maintain yield exposure while achieving their respective risk management objectives.

This represents a native DeFi primitive—not an adaptation of traditional models, but a mechanism designed specifically for the constraints and capabilities of programmable, permissionless financial infrastructure.

Subscribe now

Share

Leave a comment

Thanks for reading! Subscribe, Share & Comment.

- Additional documentation available at docs.tapir.money
- for news follow tapir on X