The State of Onchain Yield: From Stablecoins to DeFi and Beyond

Introduction

One of the most compelling innovations of decentralized finance (DeFi) is onchain yield, the income generated from blockchain-based financial activity without reliance on traditional intermediaries. Initially, DeFi yield was largely driven by straightforward mechanisms such as lending and liquidity provision. The ecosystem has since evolved into a rich landscape of complex and highly diverse strategies that span a wide range of risk profiles, economic models, and asset types.

This paper presents a framework for understanding and comparing the diverse ways yield is generated in the DeFi ecosystem. It considers several dimensions of classification, including the variability of returns, the type of underlying assets involved, the mechanisms used to generate income, and the risks associated with each strategy.

We start with non-yield-bearing stablecoins, which are pegged to the U.S. dollar and pay no interest. These are engineered for stability and fast settlement rather than generating income for the holder. Held in self-custody or at platforms without rewards, these balances earn nothing. Some platforms advertise rewards on custodial balances, but we do not count those here. For these assets, our focus is whether the issuer pays cash yield to holders, either directly or through a platform partner.

With zero return as our ground floor, we then climb the ladder to instruments and strategies that enable holders to capture yield, with complexity and risk increasing at each rung.

Next, we discuss platform-dependent yield-bearing stablecoins. These tokens do not pay native interest. Cash yield appears only when a participating platform passes yield to users’ custodial accounts through an arrangement with the issuer. Yields apply only to holders’ assets held on that platform. In self-custody, or on platforms without such an arrangement, the yield is still zero. We carve out this category to prevent confusion of issuer and platform pass-through arrangements, on the one hand, with platform promotions or points on the other. For contrast, there is also a set of decentralized non-yield-bearing stablecoins that pay zero yield until a holder opts into a separate product like sDAI or another staked protocol that offers return.

From there the paper moves into debt-based strategies, the first tier where yield is routed to the holder by design. Staked stablecoins direct a portion of the fixed-income coupon to token holders instead of solely to issuer reserves. Centralized fiat-collateralized wrappers hold short-duration government securities and automatically pass through fund income.

Next, we examine protocol-based yield, which is generated by onchain mechanisms instead of an issuer’s reserves. This category includes native staking of proof-of-stake assets, restaking and liquid restaking mechanisms that extend validator rewards, and lending and borrowing markets. In these markets, rates float with utilization and credit demand and sit above the risk-free floor.

After the protocol primitives, we turn to structured and managed yield. These are actively run or engineered strategies, including Pendle V2, Boros, and Euler V2, that splice simpler cash flows, term them, or apply leverage. The final layer is activity-generated yield, where income comes from market microstructure itself through automated market maker (AMM) liquidity provision fees, incentive programs, and yield farming.

Lastly, we introduce a stablecoin risk framework and widen the lens to capture the broader ecosystem by comparing return versus risk and return versus complexity. Arranged from cash-like stability to more engineered income streams, this progression lets readers weigh each extra percentage point of return against the additional risk and headaches it requires.

Non-Yield-Bearing Stablecoins – Centralized

Stablecoins are the primary liquidity conduit for exchanges, wallets, and protocols. Non-yield-bearing stablecoins are dollar-pegged settlement assets backed by high-quality liquid reserves such as short-dated Treasuries, overnight repos, cash at regulated custodians, and in some cases government money market funds. In this paper we classify USDT, FDUSD, and USD1 as non-yield-bearing because none of these issuers pay native interest to holders. Some platforms advertise returns on these balances, but those payouts are venue programs funded by lending or promotions. We exclude such venue yield here because it is not issuer-funded.

As the largest stablecoin, USDT anchors exchange order books and onchain routing. Tether, the issuer, reports reserves concentrated in cash equivalents and other short-term assets and discloses a capital buffer called excess reserves, which support peg stability and redemptions. USDT does not pass reserve income to holders; any return advertised on exchanges is funded by the platform’s own programs rather than an issuer pass-through.

First Digital’s FDUSD follows a fiat backed model with reserves held in Treasuries and cash and distribution focused on major centralized venues. It does not pay native interest. Headline APRs you see on affiliated platforms are platform programs and sit outside this issuer-funded yield lens.

USD1 is a newer fiat-backed stablecoin with reserves in cash equivalents and, in part, government money market funds. Growth today is driven by exchange distribution and issuer or venue points programs. There is no native interest paid to holders, so we classify USD1 as non-yield-bearing; promotional rewards do not change that classification.

Keeping yield for the issuer keeps these tokens simple. Redemptions are designed to clear at par (1 token ≈ $1), balances are instantly spendable across venues, and issuers can keep transfer costs low while supporting deep liquidity.

For holders, the opportunity cost is considerable. As a scale marker, applying the average short-term Treasury yield to Tether’s treasury holdings implies the issuer has earned about $12.6 billion of interest income over the past three years. The result is a stable, highly liquid settlement layer that favors ubiquity and simplicity over paying yield to holders.

Platform-Dependent Yield-Bearing Stablecoins

USDC and PYUSD are fiat-backed, dollar-pegged stablecoins supported by high-quality liquid assets. USDC holds cash at reserve banks and invests the remainder through a government money market fund that holds short-dated Treasury bills and reverse repurchase agreements. PYUSD is issued by Paxos Trust Company under the oversight of the New York Department of Financial Services, with reserves held in short-dated Treasuries and cash.

Where does the yield come from?

Neither token pays native interest; in self-custody or on venues without rewards, balances earn bupkis. For USDC, Circle shares a portion of reserve income with Coinbase under a commercial partnership, and Coinbase uses that share to pay USDC rewards to customers. For PYUSD, PayPal and Venmo may pay balance rewards inside their apps under a partnership with the issuer Paxos. These programs are discretionary, often capped or tiered, and they apply only to balances held on the participating platforms. The GENIUS Act, signed into law by President Trump in July, may limit indirect pass through of interest, depending on how regulators write the detailed rules putting the law into action. If rules tighten, expect a shift from cash yield to points or rebates. Only a slice of supply sits on any single platform and as a result, most reserve income still accrues to issuers, which is why platforms can advertise attractive rates while issuers continue to earn substantial profits.

Takeaways

Low Caps Fighting for Market Share: USDT and USDC account for the overwhelming majority of supply, while smaller entrants (e.g., FDUSD, USD1, PYUSD) compete for a thin remainder. USDT pays zero no matter where it’s held. USDC shares yield only on Coinbase and only for custodial balances, which are a small slice of the total supply. As a result, most balances in both tokens still earn 0% for holders while issuers retain the bulk of T-bill income. Although PYUSD, USD1 and FDUSD have gained attention as the newest entrants, their combined supply is still only a sliver of the non-yield stablecoin market, leaving the landscape firmly dominated by longstanding incumbents.

Money on the Table: The foregone cumulative retail interest for retail holders of USDC and USDT is staggering. The missed yield has snowballed into tens of billions of dollars, evidence that keeping stablecoins idle while short-term Treasuries pay over 4% compounds into a massive opportunity cost over just a few years.

Dollars parked in “cash-only” stablecoins are doing little more than changing custody onchain. That sets the stage for the next tier of decentralized assets, non-yield-bearing but fully onchain mint-and-burn designs such as USDS, DAI, USDf, and USDe. These tokens preserve the same spend-anywhere convenience of USDT and USDC yet rely on transparent collateral baskets or real-time hedging rather than centralized treasuries. Understanding how they safeguard a $1 claim without paying the holder is essential before we explore their yield-enabled counterparts, where staking wrappers unlock the income that these base layers leave on the table.

Non-Yield-Bearing Stablecoins – Decentralized

In the same camp as non-yield-bearing centralized stablecoins are decentralized non-yield-bearing stablecoins. They achieve their $1 peg without distributing interest to holders, and instead of relying on cash and Treasuries they achieve their peg through other mechanisms like over-collateralization or synthetic hedging plus internal surplus mechanisms.

USDe is a delta-neutral synthetic dollar. For each dollar of long spot collateral (e.g., ETH, stETH, or BTC), the system opens a matching short perpetual-futures position on selected venues so that spot gains are offset by perp losses and vice versa. It maintains a small stablecoin reserve (≈1% in USDC/USDT) for redemptions and margin top-ups and holds a small ETH balance to post margin for short positions in ETH-M (ether-margined) perps. The peg is sustained by this continuous hedge plus the liquidity buffer. Plain USDe pays 0% until staked into sUSDe, which passes through perp-funding, and any underlying staking yield if liquid staking tokens (LST) collateral is used.

USDS and DAI are over-collateralized stablecoins minted from onchain vaults. Borrowers deposit assets (e.g., ETH, staked ETH, WBTC, or approved tokenized RWAs) and draw USDS/DAI up to a risk-based limit, with typical collateralization ratios in the ~115–175% range depending on the asset. If a vault falls below its liquidation threshold, the protocol auctions collateral to repay the debt. Borrowers pay a variable stability fee on outstanding debt and a penalty if liquidated; these revenues accrue to the protocol’s surplus buffer or treasury to backstop the system or fund buybacks, not to passive coin holders. Simply holding USDS or DAI pays 0%; yield appears only if the holder opts into the separate savings modules (e.g., DAI → DSR via sDAI, USDS → sUSDS).

USDf is minted against an onchain collateral basket and must remain at least 116% collateralized. Eligible collateral includes stablecoins, BTC, ETH, and select large-cap tokens. The protocol does not keep a separate reserve fund; the extra 16% of collateral serves as the redemption buffer and liquidation backstop. Simply holding USDf pays 0%; yield is available only if the holder stakes into the designated savings module.

Taken together, these designs illustrate that decentralization alone does not guarantee a yield; it merely shifts the peg mechanism from fiat reserves to onchain engineering. Whether collateral is parked in an over-collateralized vault or deployed in a delta-neutral hedge, the income it generates is retained by the protocol, not by the end user. That foregone return creates the same incentive we mentioned in the centralized realm: once a base stablecoin proves its stability and liquidity, markets quickly demand a companion “staked” wrapper that passes the underlying cash flows through to holders. The next section of this paper turns to those yield-bearing derivatives, the first step in transforming otherwise idle stablecoins into productive onchain assets.

Takeaways

Diverse designs, common result: All four tokens hold their pegs without paying interest, but do so with very different mechanics: heavy over-collateralization for USDS (~270 %); moderate cushions for DAI (~150 %) and USDf (≥116 %); and a delta-neutral hedge plus a 1% reserve for USDe. Just because they’re in the same category doesn’t mean they have the same risk profile.

Adoption split: USDe has become the clear growth driver among decentralized non-yield-bearing stables, now the largest by market cap and deployed on 22 chains, illustrating how new collateral designs can quickly reshape market share. Meanwhile, DAI and USDS occupy the mid-tier and USDf remains a niche product.

Opportunity cost persists despite differing structures. Every token on this list still routes the bulk of reserve or strategy income to protocol treasuries rather than holders, underscoring the multi-billion-dollar yield gap that the next section’s staked wrappers are designed to capture.

Issuer policy steers yields: sDAI and sUSDS yields fluctuate and often track the stability fee, but Sky is actively pushing migration to sUSDS with more favorable parameters and incentives, so the spread between sUSDS and sDAI reflects issuer policy as much as market conditions.

Decentralized non-yield-bearing stablecoins dwarf the low-cap centralized stablecoins, with a total combined market cap of top coins just under $20 billion, compared to the $4 billion in the low-cap centralized stablecoins.

With the zero-yield baseline established, we can pivot to yield-bearing wrappers – sUSDS, sDAI, sUSDf, and sUSDe – that transform otherwise inert stablecoins into automatically compounding assets. The gap between what centralized and decentralized non-yield-bearing tokens deliver (nothing) and what their reserves could earn highlights a clear market inefficiency; these staking vaults close that gap by letting users keep the same stable, onchain dollars while unlocking a stream of income without sacrificing liquidity or composability.

Debt-Based Strategies

Staked Stablecoins

Staked stablecoins are an evolutionary layer that lets onchain dollars earn yield while preserving the familiar, spend-anywhere form factor of their base tokens. When a user deposits USDe, DAI, USDS, or USDf into the protocol’s vault, they receive a wrapper (sUSDe, sDAI, sUSDS, or sUSDf, respectively) whose redemption value rises automatically as protocol revenue flows in. No new wrapper tokens are printed to account for the yield; instead, each wrapper represents an ever-growing claim on the underlying stablecoins, so wallet balances stay constant while purchasing power compounds in the background.

Although the wrappers share this auto-compounding mechanic, the cash-flow engines beneath them are distinct. sDAI and sUSDS pass through T-bill and stability fee income from over-collateralized debt vaults; sUSDf routes fixed-rate credit interest from prime brokerage FalconX’s institutional loan book; and sUSDe channels a blend of perp-funding spreads and liquid-staking rewards harvested by Ethena’s delta-neutral hedge. Together they convert what was once an opportunity-cost line item, idle stablecoin balances, into a flexible, dollar-denominated savings vehicle that can still move freely across DeFi.

Regulatory treatment varies by wrapper and jurisdiction. In some markets these products may be deemed securities or collective-investment schemes, which is why several issuers geofence U.S. users or offer access under exemptions, such as those for accredited investors.

Wrappers plug directly into lending markets, liquidity provider (LP) pools, and leverage loops, letting users choose when plain stability is enough and when to convert the same dollar into an interest-bearing asset.

Takeaways

Ethena’s sUSDe has become the clear heavyweight: it now accounts for the majority of total value locked (TVL), daily yield, and cumulative payouts across the staked-stablecoin sector, dwarfing sDAI, sUSDS, and sUSDf in absolute dollars.

In early 2024, yields shot above 50% because Ethena was earning unusually rich perpetual-futures funding and basis spreads; staking income on its ETH collateral added a few extra percentage points. At launch, the amount of USDe staked was small, so those dollars of revenue were divided by a tight sUSDe supply, inflating the APY. As more capital poured in and arbitrage tightened, the same funding pool had to be shared across a much larger base, pushing returns lower. Funding rates themselves also cooled from mid-2024 highs, further compressing yield. Today the combination of a broader deposit base, calmer funding markets and a steadier token distribution schedule has brought the 30-day trailing APY into a roughly 7% to 12% band, a level that still reflects positive funding but no longer the launch-phase frenzy.

Why the recent APY spike in sUSDe yield?

Ethena’s reward stream is driven by how much funding-rate and incentive income it can harvest each day, divided by the outstanding supply of sUSDe. Three recent developments have pushed that numerator up while holding the denominator roughly flat:

1. SPAC-funded ENA buyback has juiced the basis trade

   1. In July, TLGY, a publicly traded special-purpose acquisition vehicle (SPAC), announced a proposed merger with StablecoinX (an Ethena-ecosystem treasury company) alongside a ~$360 million PIPE. The deal consists of $260 million in cash and $100 million in ENA. The cash was used to buy locked ENA from an Ethena Foundation subsidiary at announcement, and the foundation is redeploying those proceeds into roughly $5 million per day of open-market ENA purchases for about six weeks. The merger is slated to close in Q4, after which StablecoinX would continue accumulating ENA under a multiyear agreement. Market makers hedged that flow via the same long-spot/short-perp basis trade Ethena already runs, lifting funding where the protocol earns yield for sUDe.

2. Treasury redemptions shrank plain USDe but left sUSDe flat

   1. APY rose when revenue rose. Ethena’s treasury has been buying USDe in the secondary market and redeeming it. That reduces non-staked USDe outstanding while sUSDe supply barely changed. Because only sUSDe receives protocol revenue, keeping the denominator (sUSDe supply) flat while the numerator (funding/incentive income) moved higher resulted in a larger per-token payout to stakers. Put simply: fewer USDe tokens available to stake means fewer new sUSDe tokens to dilute rising revenue.

3. “Liquid Leverage” on Aave protocol adds layered incentives

   1. On Aave’s high-efficiency (E-Mode) setting, users can loop a 50% USDe/50% sUSDe position and farm an additional ~12% promotional rate on top of native sUSDe yield with promo tokens paying out daily.

Rates have converged: after the funding-rate fireworks of late 2024, the 30-day trailing APYs on all four staked stablecoins have compressed; sUSDe’s once-extreme swings have cooled and clustered near the 10% level, roughly in line with the upper bound of peers.

Yield dispersion reflects engine design: sDAI and sUSDS show smooth, stepwise APY adjustments dictated by governance, whereas sUSDe’s and sUSDf’s curves remain more jagged, mirroring their reliance on perp-funding spreads and institutional credit flows. Each has settled into a similar mid-single-digit to low-teens band.

Centralized Fiat-Collateralized

Centralized fiat-collateralized yield-bearing stablecoins are tokenized claims on regulated cash-equivalent reserves, typically U.S. dollar deposits and short dated T-bills, held by a single trust or asset manager. Issuers operate under strict regulatory frameworks, require know-your customer (KYC) and anti-money-laundering (AML) procedures (and, often, accredited investor status), and publish regular attestations or audits to prove that every token is backed 1:1.

So far, these sound a lot like USDC and USDT. But unlike the two leading stablecoins, these coins deliver yield to holders. They do so through built-in distributions mechanisms without any extra steps once the holder’s address is whitelisted and as long as their balance remains in the issuer’s ecosystem. The trade-off is clear: the investor gains professional custody, transparent audits, and broad institutional on-ramps, but sacrifices the permissionless mint/burn option, composability, and open-ended liquidity of fully decentralized models.

Unlike payment stablecoins covered by the GENIUS Act’s interest ban, tokenized cash-equivalent funds such as BlackRock’s BUIDL and Franklin Templeton’s BENJI are securities that continue to distribute fund yield subject to securities law regimes. Ondo’s USDY is likewise offered as a security, so U.S. retail availability is constrained by securities law rather than GENIUS. In short, GENIUS curbs interest paid on payment stablecoins; it does not prevent yield distributions from tokenized money funds or similar securities.

Takeaways

Parabolic growth in tokenized Treasuries: The sector’s market value has ballooned by roughly $4.7 billion, a 545% surge that set a new record at $5.6 billion by April 2025.

Treasury-bill trackers in tokens’ clothing. These tokens channel roughly 80–100% of reserves into short-dated U.S. T-bills and overnight repo, so the quoted 4–5% yields are essentially pass-throughs of the front-end Treasury curve. Yields are set by the market and reflect supply/demand and expectations for the Fed’s policy path. As a result, these tokens’ APYs closely track 3-month T-bill yields, with small basis differences from fees, cash drag, and the mix of bills vs. repo.

Gated access, narrow distribution: BUIDL, BENJI, and USDY restrict primary issuance to qualified investors, which explains their minuscule onchain holder counts: fewer than 1k addresses for BENJI and under 100 for BUIDL despite multi-billion-dollar market caps.

Concentration and target audience: BUIDL alone commands almost half the segment’s market value and is clearly aimed at institutional treasuries looking for tokenized money-market exposure, while USDY and BENJI court similar but smaller RWA-focused investors.

Liquidity still thin: exchange coverage ranges from five to 10 venues, and the largest pools cluster on permissioned institutional rails rather than public AMMs. Until broader listings and onchain liquidity deepen, these tokens will function more like tokenized fund shares than truly fungible, composable stablecoins.

Up to this point, yield has mostly come from structures rooted in traditional finance debt markets. Returns trace back to offchain collateral and interest rates, and results depend on issuers, banks, and market makers. Protocol-based yield works differently. Rewards are paid in the native token and funded by block issuance, transaction fees, and other onchain revenues. Native restaking takes the same collateral and pledges it to additional services for extra yield, which shifts risk toward smart contracts, slashing, and tighter cross-protocol correlation.

Protocol-Based Yield

Native Staking

Staking turns a proof-of-stake token into a productive asset by granting the holder a share of the revenue that rewards validators for securing and running the network (net of any commission the validator keeps). Most of the payout comes in the form of newly issued tokens that compensate validators for proposing and attesting blocks, while a smaller but growing slice is derived from transaction fees and other onchain earnings streams, such as maximal extractable value (MEV). Because inflation drops over time and fee capture rises or falls with network activity, effective yield is a moving target, shaped by token economics, throughput, and validator behavior, and should always be weighed against the liquidity cost of lockups and the tail risk of slashing events.

Where does the yield come from?

For this example, we will concentrate on Ethereum, because it is the biggest staking protocol in terms of volume.

New ETH issuance (protocol rewards):

• Ethereum mints new ETH as part of its proof-of-stake reward mechanism.

• These rewards are distributed to validators for:

  • Proposing blocks (when selected as the block proposer)

  • Attesting to blocks (voting on the validity of proposed blocks)

  • Participating in sync committees (used for light client support)

• These are the base rewards and are algorithmically adjusted depending on the amount of ETH staked across the network.

Priority fees (a.k.a. tips):

• Users submitting transactions can attach priority fees to incentivize validators to include their transactions faster.

• These tips are not burned (unlike base fees) and go directly to the validator that proposes the block.

• Priority fees are variable and tend to rise during periods of network congestion or DeFi activity.

MEV (Maximal Extractable Value):

• Validators that integrate with MEV relays or block builders can extract additional value by reordering, inserting, or censoring transactions within a block.

• MEV strategies include:

  • Sandwich attacks

  • Arbitrage across DEXs

  • Front-running liquidation events

• This source of yield is optional and non-guaranteed; only validators that opt into MEV relays (such as Flashbots) capture this additional revenue.

• MEV revenue is volatile and can vary widely depending on market activity and validator participation.

Ways to stake natively on Ethereum

Running Your Own Validator (32 ETH minimum)

• Deposit exactly 32 ETH into the official staking contract.

• Run your own validator node, manage uptime, and pay penalties if node goes offline.

• Receive all rewards directly into validator balance.

• To withdraw, initiate an exit and wait in the validator queue.

Yield source: Directly from block proposals, attestations, and priority fees collected through validator participation.

Staking via an Exchange (e.g., Coinbase, Kraken) - Custodial Native Staking

• Deposit ETH into an exchange account and choose to “stake.”

• Exchange stakes ETH on your behalf using its own validator infrastructure.

• You do not receive a liquid token (unless you explicitly opt into a wrapper like cbETH).

• Rewards appear in your staking balance within the platform.

• The exchange typically takes a fee from the staking rewards.

Yield source: Same as above but shared between the validator (exchange) and the user. The user sees net rewards after platform fees.

Takeaways

Yield source mix - Rewards come from two main streams. Most chains mint new tokens that are paid to validators or delegators, while an increasing slice is funded by transaction fees and other onchain earnings such as maximal extractable value. The split matters because minted rewards are dilutive, whereas fee-driven rewards are not.

Real yield lens - Nominal reward needs to be weighed against token supply expansion and the commission that node operators keep. A high stated APR can fall below zero in purchasing power terms once inflation and fees are removed, while deflationary or fee-burn models such as Ethereum’s can push real yield above the headline rate.

Participation and security link - A higher staking ratio generally means more economic weight is helping to secure the network, but it can also compress yield and reduce circulating float, which may amplify price swings when large holders exit.

Liquidity spectrum - Unstaking times range from instant withdrawals on Hedera’s HBAR or Cardano’s ADA to weeks on Stellar’s XLM and Polkadot’s DOT, and they are not fixed everywhere. Ethereum shows that even “flexible” systems can jam when many validators want out at once.

Risks

Exit queue congestion - In late July, the Ethereum validator exit line swelled past 600,000 ETH worth roughly $2.3 billion, stretching the wait to more than eight days, the longest on record, as discussed in the Galaxy report “Why Ethereum Unstaking Queue Has Suddenly Swelled.” A backlog of this size can trap capital during market stress and create a price overhang when the unstaked coins finally hit exchanges.

Slashing and performance penalties - Validators that double-sign or go offline can lose a slice of their stake. Delegators share that tail risk and must trust the operational discipline of the node they back.

Reward variability - Monetary schedules change (XLM recently cut its maximum inflation to 10%) and fee volumes fluctuate with network usage, so sustained yield is not guaranteed.

Smart-contract and bridge exposure - Even when staking directly, validator software and the underlying protocol rely on complex smart contracts and code that may contain bugs or vulnerabilities. While native staking avoids the bridge and oracle risks tied to liquid staking, it is still exposed to potential flaws in client implementations or consensus rules.

Regulatory and tax uncertainty - Jurisdictions differ on whether staking income is taxable at the time of receipt or sale, and policy shifts can alter the economics overnight.

Restaking

As we just learned, staking is a foundational DeFi practice after Ethereum’s 2022 transition from proof-of-work to proof-of-stake. By locking ETH (or SOL on Solana) in network contracts, token holders help secure consensus and earn rewards. While traditional staking underpins network validation, it ties up assets until the protocol’s exit or unbonding process completes, which can constrain liquidity and sideline capital that could otherwise be deployed.

Liquid staking addresses this limitation by minting fungible liquid staking tokens (LSTs) to represent any staked ETH or SOL. Holders receive an LST when they deposit tokens, allowing them to trade, lend, or use it as collateral across DeFi while still accruing base staking rewards.

Put simply, liquid staking is native staking delegated to a third party in exchange for a composable token.

Where Does the Yield Come From?

The yield sources for liquid staking are the same as native staking. The difference is operational: the depositor outsources validator operations (keys, uptime, proposing/attesting, fee recipient/MEV setup) to a third-party operator and, in return, receives a liquid staking token (LST) that stays composable across DeFi. We will explore those opportunities more in the following sections.

Risks and Takeaways

Peg Instability (Depeg Risk)

LSTs such as stETH are designed to track ETH, but the link can slip. In a recent episode, stETH traded at a discount when three forces hit at once: ETH borrow costs on the Aave protocol spiked, leverage built up in looped positions, and the validator exit queue slowed redemptions. In that setting, arbitrage weakens and pool balances skew, so the LST can diverge from ETH. If large holders unwind into thin liquidity, even small imbalances can persist, making depegs hard to resolve without deeper, steadier liquidity or faster withdrawals.

Liquidity Crises

The ability to enter and exit LST positions smoothly depends heavily on the health of secondary markets. Key asset pools including Curve Finance’s stETH/ETH pair serve as the main liquidity hubs for stETH, but they are vulnerable to imbalances during market volatility. If significant amounts of LSTs are dumped into these pools, they can become one-sided, leading to high slippage and poor execution for users looking to exit. Despite Lido’s dominance, with over 9 million ETH staked, recent reductions in its market share and changes to Curve pool parameters have exposed vulnerabilities in LST liquidity infrastructure. If liquidity thins and exit queues back up, users may be left holding discounted assets without immediate redemption paths.

Leverage Amplification

Many users loop LSTs such as stETH or rETH by posting the LST as collateral, borrowing ETH against it, swapping that ETH into more LST, and restaking. While this can amplify returns during calm markets, it introduces systemic risk. When the peg slips or interest rates spike, these positions become vulnerable to liquidation. The result can be a cascading sell-off of LSTs that further pushes the price below the peg, drains liquidity pools, and triggers additional liquidations in a feedback loop. Leverage enhances both upside and downside, but in volatile markets, it accelerates stress across the system.

Centralization and Counterparty Exposure

While LST protocols aim to decentralize staking, many still rely on a narrow set of validator operators and governance structures. Lido, for example, manages a large share of all liquid-staked ETH, which introduces systemic concentration risk. Any governance issue, validator misbehavior, or smart contract bugs could affect the protocol’s integrity and trigger a market-wide reaction. Additionally, the DeFi ecosystem’s reliance on just a few liquidity pools and wrapper tokens means that design choices, such as Curve’s amplification factors or withdrawal mechanics, can have an outsized impact on user experience and market health.

Redemption Delays and Market Fragility

Even after Ethereum’s 2023 Shapella upgrade enabling withdrawals of staked ETH, the redemption process for LSTs still depends on the blockchain’s validator exit queue. During high-demand periods, this queue can back up significantly, leading to delayed redemptions. If secondary markets are strained at the same time, users may find themselves unable to sell their (belatedly) unstaked ETH at par. This gap between redemption logic and secondary market liquidity makes LSTs vulnerable to liquidity shortages, particularly during periods of high volatility or structural market stress.

Liquid Restaking

Liquid restaking, the next evolution in onchain capital efficiency, builds on liquid staking. Through protocols such as EigenLayer, users can deposit their LSTs into Actively Validated Services (AVSs), decentralized systems that require trustless validator security. In exchange, users receive a liquid restaking token (LRT). These LRTs, such as eETH, ezETH, or rsETH, retain the composability and liquidity of the original LST, while layering on new forms of yield.

What makes liquid restaking powerful is its ability to stack incentive streams.

• Users continue to earn native staking rewards from Ethereum.

• They gain access to AVS-driven incentives, including points (crypto’s answer to frequent flier miles), airdrops, and early token emissions.

• Many LRT protocols also offer DeFi integrations, liquidity mining, or lending opportunities, amplifying potential return without compromising liquidity.

This layered model transforms what was once passive, locked collateral into a yield-generating, DeFi-active instrument. It turns ETH or SOL into multi-purpose capital, able to secure multiple networks simultaneously, and get paid for each.

As the restaking economy matures, liquid restaking is poised to become a key building block in Ethereum’s and Solana’s modular infrastructure stacks.

Where Does the Yield Come From?

Liquid restaking tokens (LRTs) boost returns by unlocking additional yield sources on top of standard staking. Here’s where the extra return comes from:

Restaking incentives from AVSs LRTs participate in securing new decentralized services (AVSs) through EigenLayer or similar networks. These services offer points, early token rewards, or future airdrops to restakers as payment for security.

Protocol incentive programs LRT issuers including Renzo, ether.fi, and Kelp DAO offer loyalty points, reward multipliers, and liquidity mining campaigns to attract users. These incentives can meaningfully boost net yield, especially early on.

DeFi integrations LRTs are often deposited into lending markets, decentralized exchange (DEX) pools, or automated vaults, where users can earn additional yield on top of restaking rewards.

Takeaways

LRTs extend staking into a new reward layer Liquid restaking tokens take staked ETH or SOL and redeploy it to secure additional decentralized services, unlocking new streams of incentives beyond standard staking rewards.

Restaking yield is powered by AVSs, not just Ethereum While LSTs earn yield from Ethereum’s native Proof of Stake (PoS) system, LRTs layer in upside from AVSs, including points, airdrops, and early token rewards from protocols including EigenLayer and its ecosystem.

LRTs stay liquid and composable Just like LSTs, LRTs are ERC-20 tokens (or SPL tokens on Solana) that can be freely traded, used in DeFi, and plugged into lending markets, vaults, and AMMs. Yield remains accessible without lockups.

Not all yield is paid in ETH or SOL Much of today’s LRT yield comes in the form of points, future protocol tokens, and speculative ecosystem rewards. These are not always liquid or guaranteed but are central to current yield strategies.

Protocols compete with layered incentives Top LRT issuers (including Renzo, ether.fi, and Kelp DAO) actively boost returns through loyalty points, reward multipliers, and integrations with EigenLayer and partner AVSs, making LRTs a dynamic part of DeFi yield stacking.

LRTs are still evolving Restaking is a young sector. Risks around slashing, AVS reliability, smart contract complexity, and incentive dilution remain salient.

Risks

LRTs’ composability unlocks powerful new strategies but also intertwines risks. If an AVS is exploited, or a node operator violates any slashing rule, the same event can slash multiple positions at once, forcing LRT holders, LST holders, and native stakers to absorb correlated losses. The result is a yield stack that looks attractive on the surface yet hinges on prudent operator behavior, transparent governance, and the market’s ability to absorb liquidity shocks when large positions race to exit.

Lending and Borrowing

Onchain lending turns idle tokens into productive capital; depositors supply assets to a money market contract and receive interest-bearing receipts. Borrowers draw against that pooled liquidity at variable rates that rise as utilization climbs, and all loans remain overcollateralized. When a position’s margin falls below a safety threshold, the smart contract’s liquidator steps in automatically. Interest paid by borrowers flows, block by block, to suppliers, so yield tracks real-time demand for leverage.

Unlike non-yield-bearing stablecoins that simply sit in wallets as digital cash, DeFi lending applications turn those same dollars into productive capital by routing them through shared liquidity pools. Anyone can deposit a stablecoin, instantly becoming a lender who earns whatever variable rate the protocol’s algorithm sets. Borrowers, in turn, post separate collateral positions, draw from that common pool, and pay interest back into it. The mechanism creates a live marketplace where supply and demand for credit constantly rebalance. Yields rise when borrowing demand spikes and fall when liquidity is abundant; liquidations are handled automatically whenever collateral values slip below the safety threshold.

In effect, non-yield-bearing stablecoins monetize collateral for a single issuer, providing dollar exposure but no income, whereas DeFi lending markets recycle that collateral across many participants, converting idle tokens into system-wide credit while distributing yield and risk among all suppliers and borrowers.

Where Does the Yield Come From?

Lending and borrowing protocols generate yield through a combination of mechanisms that work together to create stackable income streams for both sides of the market.

Interest Paid by Borrowers

At the core, lenders earn variable interest rates paid by borrowers who take out loans from the protocol. These rates fluctuate depending on utilization; as demand for borrowing rises, so do rates. Borrowers may be hedging exposure, leveraging positions, or unlocking liquidity without selling.

Protocol Incentives

Many lending protocols, especially in their early stages or during growth campaigns, distribute native governance tokens (e.g., COMP, AAVE, MORPHO) to lenders and borrowers. These incentives act as subsidies, enhancing the net APY for lenders and offsetting the borrowing cost for users. In some cases, incentive emissions can exceed the base interest rate, making “yield farming” possible even in low-rate environments.

Underlying Token Rewards

When users lend yield-bearing assets, such as stETH, rETH, or ezETH, they continue to earn the native staking or restaking yield embedded in those tokens. These rewards accrue passively in the background while the asset is deployed in the lending pool. This creates dual yield streams:

• One from the staking or restaking layer

• Another from the lending interest or incentives on top

In some cases, this can be extended further via leverage loops (e.g., borrowing ETH, then re-staking or re-lending it), compounding both yield sources.

The Power of Stacking Yield

Let’s take a closer look at the power of stacking yield, one of DeFi’s most compelling innovations. In earlier sections, we saw how users can stake ETH or SOL into liquid staking tokens (LSTs) like stETH, rETH, or JitoSOL, and restake them into liquid restaking tokens (LRTs) like eETH or ezETH. These base assets already generate yield through protocol staking and restaking mechanisms.

But that’s just the first layer.

On platforms like Aave, users can deposit their LSTs or LRTs into lending markets, enabling them to earn additional yield on top of their staking rewards. Lending markets like Aave pay depositors a variable APY, funded by borrowers who post collateral to access liquidity. In effect, this creates a dual-reward stream: staking income + lending income, without having to unstake or sell the original asset.

Yield Stacking with stETH

Imagine a user holds $10,000 worth of stETH, which represents ETH that’s already staked through Lido and earning native Ethereum staking rewards. Instead of simply holding the stETH in a wallet, the user deposits it into Aave V3, the protocol’s most advanced market, as collateral.

By doing so, the user continues to earn the underlying staking rewards from stETH, while also becoming eligible to earn additional lending income from Aave’s protocol. On Aave, that stETH can be borrowed by other users, and the lender (in this case, the stETH holder) earns a share of the interest paid on those loans.

This approach unlocks stacked yield streams from a single asset, combining passive staking income with protocol-level lending rewards, all while maintaining collateral flexibility to access liquidity if needed.

Stacking with Stablecoins

This concept of stacking yield extends naturally to stablecoins. For instance, a user could deposit stETH into Aave as collateral and borrow a stablecoin like USDe or DAI against it.

The borrowed stablecoin can then be

• Re-deposited into Aave or another protocol to earn lending yield

• Deployed into yield farms or automated vaults

• Swapped for more stETH or another LST to loop the position and boost exposure

Whether used passively or actively, this borrowed capital becomes a second layer in a broader yield strategy, turning leverage into a tool for composable, capital-efficient returns across multiple protocols.

Yield Looping on Aave

Aave enables users to “loop” yield by borrowing against collateral and reinvesting the borrowed funds into the same or similar asset. For instance, a user could deposit stETH as collateral, borrow ETH against it, convert that ETH into more stETH, and deposit it back into Aave. Each cycle increases the user’s exposure to stETH and compounds the associated staking and lending yields. This technique is often referred to as leveraged staking.

Looping magnifies potential returns, but it also introduces leverage risk. Because the position is debt-backed, changes in market conditions, such as stETH trading below ETH or borrow rates rising, can quickly increase the loan-to-value (LTV) ratio. If LTV exceeds the protocol’s liquidation threshold, the user’s collateral may be partially or fully liquidated. This makes looping inherently more volatile and demands active monitoring, especially in fast-moving or thin liquidity markets. For most users, it’s a strategy that offers high reward but requires precision and risk tolerance.

The stack of colored bands in the chart above shows capital scattering across dozens of chain-specific deployments. While this diversifies tail risk, it also dilutes onchain liquidity and forces traders to bridge or pay up for cross-chain liquidity, an operational headache that did not exist when lending was an almost-pure Ethereum story in 2020-2021.

Together the charts confirm that the lending sector has recovered in size but not in efficiency: abundant deposits chase episodic borrowing booms, rates gyrate accordingly, and liquidity is now spread over far more venues and chains than in the last cycle, a landscape that rewards active rate-shopping and careful collateral management.

Takeaways

Activity is back near all-time highs, but the mix has shifted. Dollar value supplied to money markets has climbed past $80 billion while outstanding loans approach $35 billion, levels last seen in the 2021–2022 bull run. Ethereum mainnet (black in the charts above) is still the anchor, yet the fastest growth now comes from Optimism-style L2s and newer alt-L1s, echoing the migration trend highlighted in Galaxy Research’s April State of Crypto Lending report.

Utilization remains soft, signaling structural excess liquidity. The gap between supplied and borrowed capital has widened with each up-cycle; only ~40% of deposits are lent out today. That slack keeps blue-chip lending rates anchored in mid-single digits even when speculative demand surges.

Borrow costs are cyclical, not secular. Stablecoin APRs on mainnet have swung from sub-2% in late 2022 to brief 15% spikes during funding-rate frenzies in Q1 2024; the 7-day moving average now hovers around 5%–6 %. The pattern matches the April report’s observation that lending yields remain tightly coupled to leverage trades (basis, restaking loops) rather than to sustainable credit demand.

Risks of DeFi Lending

Liquidity-mismatch and utilization risk - Deposits have surged to ~$85 billion while outstanding borrows sit near $35 billion, leaving ~60 % of supplied capital idle. Galaxy’s April lending report noted that “open DeFi borrows were only $19 billion across 20 protocols at end-2024, even after a 9.6x rebound from the bear-market low.” A sudden flight to safer yields could force protocols to slash rates or incentivize withdrawals, stressing smaller chains where depth is thin.

Fragmented collateral, concentrated liquidations - More than a dozen L1 and L2 networks now host lending pools, yet liquidation engines for most assets still route through a handful of bots and back-running keepers. A correlated price shock could trigger cross-chain liquidation cascades faster than bridges or keepers can arbitrage collateral, amplifying the drawdowns visible at each 2024 funding-rate spike.

Rate volatility tied to speculative leverage, not sustainable credit - Borrow APRs on Ethereum lend markets have whipsawed between 2% and 16% since 2021, tracking funding-rate trades and restaking loops rather than organic demand for working capital. Galaxy’s April lending report noted the cyclicality: “lending yields remain tightly coupled to leverage trades rather than lasting credit demand.” Investors should recognize that juicy moments (12%+ spikes) are short-lived and quickly revert to the mean.

Access asymmetry - While anyone can supply liquidity, many high-yield opportunities (e.g., isolated pools on smaller L2s) require bridging, custom oracles, or governance whitelists. Retail users who keep funds on mainnet may face lower returns yet the same market risk, illustrating a “risk-to-availability” gap not captured by headline APRs.

These factors mean that, despite headline growth, today’s DeFi credit markets still carry liquidity and cyclicality risks that non-yield-bearing collateralized debt position (CDP) stablecoins do not, underscoring the need to weigh yield against the structural vulnerabilities that underpin it.

Structured/Managed Yield

As DeFi matures beyond simple lending and staking, new protocols are emerging that allow users to repackage and customize yield itself. This is the domain of structured yield, platforms that separate, automate, or tokenize future cash flows in ways that mirror more sophisticated traditional finance instruments.

Two of the most innovative players in this space are Pendle and Euler v2. While both enable users to engage with more advanced yield mechanics, they approach the challenge from different angles, one through yield tokenization, and the other through modular strategy vaults.

Pendle V2

Max Boosted APY: Indicative upper‑bound rate shown on Pendle assuming full vePENDLE boost, optimal pool selection, uninterrupted reward capture, and current incentive schedules. This yield is variable, incentive‑driven, and not guaranteed; it typically requires locking PENDLE and/or actively managing LP/YT positions. Actual realized returns may be materially lower and exclude gas, slippage, and fees.

Pendle is a DeFi protocol that enables users to split a yield-bearing asset into two separate tokens, unlocking more precise control over time-based yield exposure. This mechanism, known as yield tokenization, allows users to isolate, trade, and structure yield in a way that's not possible with standard staking or lending protocols.

When a user deposits a supported asset such as stETH, ezETH, sDAI, or USDe, Pendle locks the token and issues:

• Principal Token (PT) - Represents the right to redeem the original asset at a set maturity date. PTs do not earn yield but typically trade at a discount. Holding a PT is equivalent to locking in a fixed return, regardless of what happens to yield markets during the holding period.

• Yield Token (YT) - Captures all variable yield generated by the underlying asset from when you invest until maturity. YT holders receive this yield as it accrues, and the market price of YT reflects expectations about future rewards, making it a tool for speculating on or hedging interest rate movement.

This structure, which echoes the principal-only and interest-only “strips” of mortgage-backed securities pioneered on Wall Street in the 1980s, lets users customize their exposure. Risk-averse users can buy PTs to lock in a fixed return, while yield-seeking or directional traders can buy YTs to capture upside from variable rewards like staking APY or protocol incentives. These tokens are freely tradable and can be combined or sold separately at any time.

While it’s possible to manually loop Pendle positions by depositing Principal Tokens (PTs) as collateral on Aave or Morpho, doing so means actively managing the position. You need to track borrowing limits, watch liquidation risk, and rebalance as markets move. Lenders generally accept PTs, not YTs, because PTs act like zero-coupon claims that pay out at par on maturity, which makes them easier to price and haircut. Yield Tokens (YTs) are the variable, time-decaying yield strip and are rarely eligible as collateral.

Pendle separates principal and yield, and ecosystem tools automate how those pieces are used. Boros builds on Pendle and runs a PT carry loop for investors: borrow against PTs, buy more PTs, redeposit, and repeat within preset risk limits. The target is the PT’s implied fixed yield minus the floating borrow rate, after fees and slippage. Looping exists across DeFi, but this fixed-carry profile is specific to Pendle’s PT design. Loops on other assets, such as LSTs, mainly amplify exposure rather than lock in a carry spread.

Boros

Boros is a platform in the Pendle ecosystem that automates a specific type of yield strategy; leveraged looping of Principal Tokens (PTs) based on funding rate differentials. Rather than interacting with Pendle manually, Boros wraps the process into a streamlined vault, enabling users to earn amplified fixed yield without managing the mechanics themselves.

At a high level, Boros vaults execute the following loop

1. Buy discounted PTs on Pendle (e.g., PT-USDe or PT-sDAI), which are redeemable at full face value at maturity

2. Pledge those PTs as collateral on a lending platform like Morpho or Aave

3. Borrow a stablecoin (like USDC or USDe) against the PT collateral

4. Use the borrowed stablecoin to purchase more PTs

5. Repeat the cycle, compounding exposure

This loop amplifies the return from Pendle’s fixed-yield positions by recycling capital, effectively creating leveraged exposure to fixed income. The key to the strategy is the positive spread between the fixed return embedded in the PT and the borrow rate on the lending protocol.

As long as the borrowing cost remains lower than the PT’s implied yield, the strategy delivers a net positive carry.

Boros automates this loop through smart contract vaults that manage collateral ratios, borrowing limits, and position health, freeing the user from needing to monitor liquidations or manually rebalance. Vaults are designed with predefined parameters and are publicly auditable, allowing users to choose risk levels and maturities that align with their preferences.

By simplifying a complex, multi-step process into a single deposit, Boros transforms an advanced interest rate arbitrage into a turnkey DeFi strategy, giving users a capital-efficient way to capture structured yield using the Pendle ecosystem’s fixed-income instruments.

Where Does the Yield Come From?

Pendle

Setup: In Pendle, stETH is split into two tokens

• YT-stETH (Yield Token): priced at 0.04 stETH

• PT-stETH (Principal Token): priced at ~0.96 stETH

• Maturity: ~1 year

• Underlying APY (Lido staking): ~ 5%

• Market Implied APY of the PT: ~ 4.2%

When you buy 1 YT-stETH, you gain rights to all staking yield from 1 stETH until maturity. If staking rewards are 0.05 stETH over the year and you only paid 0.04 for the YT, your net profit is 0.01 stETH. That’s a 25% return on your capital, no leverage or borrowing required.

Yield Token (YT)

YT gives you access to the full yield stream from the underlying asset, in this case, Lido staking rewards on 1 stETH. Your return is simply the yield earned minus what you paid for the YT. Because YT trades at a deep discount to the full principal, even a modest yield becomes a high-percentage return on capital. That’s why YT is often described as offering “leveraged” yield exposure, without any borrowing or liquidation risk. If the realized yield exceeds the market’s implied APY, your YT position will outperform.

Principal Token (PT)

PT is the flipside: it represents the principal only. You’re buying 1 unit of stETH at a discount, say 0.96, and it will redeem at maturity for exactly 1 stETH. That discount is your fixed return. PT behaves like a zero-coupon bond: its price gradually pulls to par as maturity approaches. The “Fixed APY” you see in the app reflects that yield-to-maturity. It’s passive, predictable, and has no exposure to staking rewards or points.

When You Can Sell

Both PT and YT are freely tradable before maturity on Pendle’s AMM.

• YT: You can sell anytime, either after collecting some yield or when market rates shift in your favor. A common strategy is to sell YT if implied APY rises and the price appreciates, locking in a profit before full maturity.

• PT: You can sell at any time. While you can hold to maturity to capture the full discount (par redemption), you can also sell early if its price appreciates, say if rates drop or demand for fixed yield increases.

Boros

Boros generates yield by automating a looping strategy built on Pendle’s Principal Tokens (PTs), leveraging the difference between fixed yield available in PTs and the cost of borrowing stablecoins on integrated money markets like Aave or Morpho.

Here’s how it works in practice

1. Fixed Yield from Pendle PTs - Boros vaults begin by purchasing Pendle Principal Tokens (PTs), such as PT-USDe or PT-sDAI, which are typically priced at a discount to their redemption value at maturity.

2. Leverage via Borrowing - Once PTs are acquired, Boros uses them as collateral to borrow stablecoins (e.g., USDC or USDe) on platforms such as Morpho Blue. These borrowed funds are then recycled into more PT purchases, expanding the position size. This loop can be executed multiple times within acceptable risk thresholds, amplifying the yield earned on the original capital.

3. Yield Spread as Net Return - The strategy earns the spread between the fixed PT yield and the floating borrow rate on the lending venue. Net carry per unit is approximately PT APY minus borrow APY minus fees and slippage. The vault then loops collateral and borrows up to risk limits, so effective return is roughly net carry multiplied by the achieved leverage factor. Bigger positive spreads produce higher returns; if the borrow rate rises above the PT yield, the carry turns negative.

   1. Loop efficiency means how much exposure the vault can achieve per dollar after slippage, fees, and LTV caps. Higher efficiency allows more loops before hitting limits and therefore a larger leverage factor; lower efficiency means more drag and less amplification.

4. Optional Incentive Layers

   1. Some vaults may layer in Pendle incentives (e.g., PENDLE emissions or LP rewards), further boosting returns. While not guaranteed, these rewards can materially increase the net APY, depending on market conditions and vault configuration.

By automating the loop and abstracting away execution complexity, Boros transforms a sophisticated fixed-income arbitrage into a one-click DeFi vault, letting users earn structured, leveraged yield while managing position health through onchain logic.

Risks and Takeaways

Yield Gap Risk (YT)

If the APY implied by the market price of a Yield Token exceeds the staking rate paid on the underlying collateral, the trade could end up unprofitable unless conditions improve. Even if you enter with implied < underlying, future APY can drop below your entry’s implied level.

Time effect on YT price

As Pendle tokens approach maturity, the YT’s price tends to drift lower because it’s only the vestigial unpaid yield, unless some offset comes from APY moves in the underlying collateral.

Rate/mark-to-market (PT)

The PT’s market price moves with rate expectations; selling before maturity can realize more or less than the implied carry, whereas holding to maturity redeems at par.

No liquidation ≠ no risk

The levered feel of YT comes from pricing, not borrowing, so there’s no liquidation/oracle risk. Your outcome hinges on future yield materializing.

Pendle Yield Curve (Inverted)

Shorter maturities offer higher yields than longer ones. This happens because:

• PTs maturing soon tend to trade at deeper discounts

• Market demand often concentrates around shorter-term trades

• Traders may be pricing in uncertainty or volatility over longer durations

Treasury Curve Proxy (Normal)

• Longer-term bonds typically offer higher yields to compensate for time and inflation risk. This is the expected structure in traditional interest rate markets like U.S. Treasuries.

• TradFi yields reference the U.S. Treasury constant-maturity series (1M/3M/6M bills; 1Y/2Y/5Y notes).

Euler V2

The Euler v2 protocol offers foundational yield opportunities through its decentralized lending and borrowing markets, where users can supply assets to isolated ERC-4626 vaults to earn interest or use them as collateral to borrow other tokens at customizable rates. But the platform's core innovation is its managed strategy vaults. These vaults, often overseen by risk curators who dynamically adjust parameters for optimal risk management, automate sophisticated DeFi maneuvers, enabling users to deploy capital into structured products with minimal manual intervention. By tokenizing these strategies as transferable shares, Euler v2 transforms passive lending into actively managed positions, fostering composability across the ecosystem.

Strategy 1 - Leverage Looping on LSTs and LRTs

What It Is

This is a recursive lending strategy designed to amplify exposure to yield-bearing tokens. A user deposits LSTs or LRTs into a vault. The vault then pledges that asset as collateral to borrow a related asset (typically wrapped ether, or WETH), swaps the borrowed WETH for fresh supplies of the original LST/LRT, and redeposits it into the vault. This process, known as "looping" or "folding," is repeated to achieve leverage, magnifying the user's underlying yield.

Assets Involved

The primary assets are high-quality LSTs (like wstETH) and LRTs (like eETH). The borrowed asset is typically WETH.

Source of Yield

The yield comes from two streams. First, the LST/LRT’s intrinsic APY from Ethereum staking or restaking. Second, the Euler vault’s supply APY paid on the deposited asset. Because the position is levered, both the staking/restaking yield and the supply APY accrue on a larger notional amount, amplifying the total return.

Strategy 2 - Leveraged LP Positions

What It Is

This strategy allows liquidity providers (LPs) to earn yield on their LP tokens while simultaneously using them to gain leveraged exposure. A user deposits LP tokens from an Automated Market Maker (e.g. Uniswap) into a Euler vault. The vault uses these tokens as collateral to borrow one of the underlying assets from the pair. This borrowed asset can be used to buy more of the original LP tokens or to free up capital for other uses, all without having to sell the LP token.

Assets Involved

The collateral consists of LP tokens (e.g., WETH/USDC LP). The borrowed assets are the underlying tokens that make up the liquidity pair (e.g., WETH or USDC).

Source of Yield

The primary source of yield is the underlying return from the LP position, which comes from trading fees generated by the AMM pool. By using leverage, the user amplifies their exposure to these fees. Additionally, they earn the lending APY from Euler for supplying the LP token as collateral.

Strategy 3 - Delta-Neutral Farming

What It Is

A sophisticated strategy designed to generate yield while minimizing exposure to the price volatility of an asset. The user deposits a stablecoin (e.g., USDC) as collateral into a vault. The vault then borrows a volatile asset (e.g., ETH) against this collateral and immediately sells it for more stablecoins. This creates a delta-neutral position where the user profits from the interest rate differential, not the price movement of the volatile asset.

Assets Involved

Stablecoins (USDC, DAI) are used as collateral, while volatile assets (ETH, WBTC) are borrowed.

Source of Yield

The yield is derived purely from the net interest rate spread. The user earns the lending APY on the stablecoins they supplied, while paying the borrow rate for the volatile asset. The strategy is profitable as long as the yield earned on the stablecoin collateral is greater than the cost to borrow the volatile asset.

Strategy 4 - Cross-Protocol Yield Arbitrage

What It Is

This is an opportunistic strategy that exploits interest rate differences between DeFi protocols. A vault on Euler can be programmed to borrow an asset at Euler's relatively low interest rate and then deposit or "farm" with that asset on another protocol (such as Pendle, Yearn, or a new, incentivized money market) that is temporarily offering a higher APY. This is an advanced strategy that effectively performs an onchain carry trade.

Assets Involved

This can involve any asset with a significant interest rate differential, but it is most commonly performed with stablecoins or WETH due to their deep liquidity across the DeFi ecosystem.

Source of Yield

The profit comes directly from the interest rate spread between the two protocols. The yield is the APY earned on the external protocol minus the borrowing cost paid to Euler. This type of yield is highly variable and depends on fleeting market inefficiencies and promotional incentives.

Risks and Takeaways

Euler v2 introduces a modular lending architecture with key risk management improvements, but several risks remain that affect suppliers and borrowers alike.

Euler V1 Exploit

In March 2023, Euler Finance v1 was exploited and about $190 million worth of cryptocurrencies, including DAI, wBTC, stETH, and USDC, was drained from markets. The attacker took out a flash loan (one that is repaid in the same transaction as it is made), deposited the proceeds into Euler, and minted a large number of collateral tokens. They then called a vulnerable function without proper checks, pushed their account into liquidation on favorable terms, and captured the proceeds. By manipulating the health score and liquidating their own position, they were able to siphon assets from Euler’s reserves.

Although most of the stolen funds were eventually returned through prolonged negotiations and investigation, the incident highlights critical vulnerabilities around collateral integrity, liquidation logic, and vulnerability to flash loan-enabled attacks. It serves as a reminder that even audited protocols can harbor complex logic flaws, especially when combining novel leverage mechanics with automated financial operations.

Interest rate volatility

Euler uses utilization-based interest rate models (IRMs), meaning supply APY can fluctuate depending on borrow demand. When utilization is low, your yield may drop to near zero; when it's high, returns can improve but also signal liquidity constraints. These rate changes can happen quickly, especially in vaults with reactive IRMs.

Liquidity constraints

You can only withdraw from a vault if there’s idle liquidity available. At high utilization it may not be possible to exit your position instantly. In that case, you’ll have to wait for borrowers to repay or for new supply to arrive. This risk is particularly relevant in volatile markets or for assets with narrow borrower bases.

Underlying asset exposure

When you supply a vault, you're exposed to the asset you deposit. If that token depegs, suffers an exploit, or underperforms structurally (due to a failing yield strategy), your token value will drop. Some vaults wrap yield-bearing assets like sDAI or wstETH, which stack yield but also pass through protocol-specific risks.

Reward sustainability

Certain markets show enhanced APYs because of external incentives, distributed either onchain (via reward streams) or off-chain (via incentive and point management platform Merkl). These incentives are temporary and discretionary, and when they end, the effective APY can drop significantly, catching users who entered based on inflated yield off-guard.

Vault governance dynamics

Every vault is governed by a specific entity that can modify parameters like reserve factor, rate curves, caps, and collateral settings. These changes affect how interest accrues and how risky the vault becomes. Suppliers must implicitly trust that vault governors will act rationally and transparently, which isn't guaranteed.

Smart contract vulnerabilities

Euler v2 has been audited and rearchitected after the v1 exploit, but it remains a complex system with custom rate models, governance controls, and token integrations. Any bug in its logic, especially in the core vault contracts or interest-rate calculations, could lead to fund loss or mispriced risk. Smart contract risk is reduced but never fully eliminated.

Oracle and pricing dependencies

Euler relies on external oracles like Chainlink or Pyth to value collateral and, in some vaults, borrowed assets themselves. Malfunctioning, lagging, or manipulated oracles could result in incorrect rate adjustments, mispriced assets, or systemic vault shutdowns. While oracle infrastructure is battle-tested, it's still a critical dependency.

Protocol governance and systemic risk

The broader Euler protocol is governed by the Euler DAO, which can authorize upgrades, pause mechanisms, or change vault policies. A stalled or captured DAO poses risks to the protocol’s responsiveness and safety mechanisms. Like all DeFi protocols, Euler v2 also inherits systemic risk from the broader Ethereum ecosystem.

Activity-Generated

Liquidity Provider

Providing liquidity (LPing) is when you deposit equal amounts of tokens into a decentralized exchange (DEX) pool so traders can swap against your funds. In return, you earn a share of the trading fees that pool collects. Unlike lending, there is no borrower; your assets sit in an automated market maker (AMM) contract that quotes prices and executes trades. For example, if you were to LP the wETH/USDC pool on Uniswap v3, you would have to supply an equal amount of both wETH and USDC.

Understanding LPing

What is an AMM?

In a traditional brokerage or order book you submit an order and a counterparty fills it. That counterparty is often a market maker who quotes a bid and an ask, takes inventory risk, and earns the spread and exchange rebates. Price comes from active quotes and matching between buyers and sellers. Your trade executes only if someone meets you at your price.

In an AMM, the counterparty is a pool of tokens held in a smart contract. The contract always quotes a price that is computed from a formula that depends on the pool’s token balances. You trade against the pool rather than a firm on the other side. Anyone can deposit tokens into the pool and earn a share of the swap fees that traders pay. The pool’s price moves as its balances change, and arbitrage brings that price back in line with the wider market.

A simple way to remember it: an order book is a meeting place for buyers and sellers, while an AMM is a vending machine for asset pairs. The vending machine always offers a quote based on what is left on its shelves, and the fee you pay is shared with the people who stocked the machine.

Why there are different pairs and pools

If an AMM is a vending machine for assets, the next question is what sits on the shelves. AMMs are organized into pairs, and each pair holds exactly two assets. For example, wETH against USDC or USDC against USDT. Pairs exist because traders need direct routes between specific assets, for example given them the ability to trade wETH for USDC.

Highly correlated assets such as stablecoins benefit from low slippage formulas and low fees. More volatile assets benefit from wider pricing flexibility and often higher fees. Within a given pair there can be multiple pools that differ by fee rate so traders and liquidity providers can choose the economics that fit current market conditions.

Pools and fee rates in Uniswap v3

Each pair offer separate pools that charge different fee rates, often called fee tiers. wETH/USDC is most active at the 0.05% and 0.30% fee tiers. Liquidity is not shared across these fee tiers, so your returns depend on the specific tier you choose and the price range where you post your liquidity.

How to choose a fee rate and the tradeoffs

• 0.05% pool - Better when realized volatility is low and spreads are tight. This pool tends to route more volume, especially on L2s where gas is cheaper, so you earn many small fees. The downside is that each trade pays less, so you rely on sustained volume to cover inventory risk and management costs.

• 0.30% pool - Better when volatility and spreads are higher or when the 0.05% pool is crowded and your share there is small. Each trade pays more, which can help offset inventory risk. The tradeoff is usually less routed flow, so fee income is volatile and more path-dependent.

Price range and ticks

This chart shows how liquidity from liquidity providers is distributed across ticks in the Uniswap V3 ETH/USDC pool at the 0.05% fee tier, with a marker at the current tick.

After you choose a pair and a fee rate, you set the price range where your liquidity will be active. Uniswap v3 divides price into tiny steps called ticks. Your lower and upper bounds must sit on tick multiples, and your position earns fees only while the market price is inside those ticks. As trades move price across ticks, your inventory shifts between the two tokens and fees are redistributed to the LPs whose ranges include the current tick.

• Narrow range: higher fee density while you are in range, more monitoring, greater chance of sitting out of range.

• Wide range: fewer fees per dollar, more time in range, less maintenance.

• Outside your range: you hold mostly one asset and earn no fees until you reposition or until the price action comes back into your tick range.

Supplying Liquidity to an AMM Pool and Fee Tier

Once you’ve chosen the pair, the pool with a fee rate, and the tick range where your liquidity will be active, you deposit equal dollar value of both tokens, and the contract mints an LP non-fungible token (NFT) that records your share and the fees it earns. While price trades inside your range you accrue fees, and the pool automatically shifts your inventory between wETH and USDC. If price moves outside your range, you stop earning until you reposition or the price action comes back into your tick range. You can exit at any time by removing liquidity and collecting any unclaimed fees.

To summarize:

• Choose the pool and fee rate – On Uniswap v3 wETH/USDC is most active at 0.05% and 0.30%. Higher fee pools pay more per trade but often route less volume. Lower fee pools usually route more flow.

• Set a price range - If ETH is $3,200, you might set $2,800 to $3,600. Narrow ranges earn more while you are in range but fall out of range faster. Wider ranges earn less per dollar but stay active longer.

• Deposit equal value of both tokens for that range - The contract mints an NFT that represents your position.

• Earn fees while trades occur inside your range - Fees accrue in the contract until you collect them.

• If price leaves your range, you hold mostly one asset and earn no fees until A) you burn the NFT, claim fees, and redeposit with a new range, or B) the price action comes back into your tick range.

• Exit any time by removing liquidity to withdraw your tokens plus any unclaimed fees.

Impermanent Loss

It is important to explain a key concept and risk of LPing, referred to as impermanent loss (IL). We will continue with the example of supplying wETH/USDC into Uniswap v3 to explain the mechanics of LPing and how impermeant loss can affect your profits.

Impermanent loss is simply the gap between what your LP position is worth and what you’d have if you just held the same tokens after prices move. Fees you collect can offset, sometimes fully cover, that gap.

How the pool auto-rebalances you

In a 50/50 pool like wETH/USDC, the AMM keeps price by trading your tokens with swappers. When wETH rises, the contract sells some of your wETH for USDC; when wETH falls, it buys wETH with your USDC. On Uniswap v3 this happens whenever the price is inside your chosen tick range; if the price moves outside that range, you stop earning fees and sit mostly in one asset until you adjust your range or price comes back into your range.

When IL is realized

It’s called impermanent because it’s just a mark-to-market effect; if the price returns to your entry, the gap disappears. You realize IL only when you close or change the position (e.g., burn the LP tokens to withdraw, or burn/re-mint to reposition your range). Collecting fees doesn’t realize IL by itself; realization happens when you exit or rebalance, and the pool’s current token mix becomes your actual holdings.

LPing Investment Choices

Concentrated liquidity AMMs (for example Uniswap v3) let you choose a price band where your capital makes a market. Tight bands give high fee density while you are in range, but you need to monitor and rebalance more often. Wide bands earn less per dollar but stay active longer.

Stable swap AMMs (for example Curve’s stable pools) target assets that trade near parity. These typically deliver steadier fee income with lower impermanent loss, with returns driven by volume rather than price swings.

Weighted and multi-asset pools (for example Balancer) behave more like portfolio exposures. Fees add to or subtract from that baseline depending on volume and incentives.

High-throughput venues on BNB Chain (created by crypto exchange Binance) or Solana can have lower fees per trade and high volume, but economics still hinge on the same ingredients: time in range, share of active liquidity, volatility, and incentives.

Where does the yield come from?

Swap Fees

Every trade that crosses your ticks pays the pool fee, and you earn a pro rata share only while the market price sits inside your chosen range. In back-of-envelope terms, your fee yield scales with the fee rate, the volume that trades near your ticks, and your share of the active liquidity at those ticks. Crowding can dilute your share, and narrow ranges raise fee density while you are in range but increase the time you fall out of range and raise price risk. Gas and repositioning costs reduce what you keep, especially on mainnet. Headline APYs often annualize short lookbacks and can be noisy, and volume can migrate between pools or chains. Swap fees are a real source of yield, but they are variable and path-dependent, so results hinge on flow through your ticks and how much of that flow you capture.

Incentives and why many LPs chase them

Some venues pay extra token rewards to attract liquidity. These programs lift the headline APR and pull in capital, but they are temporary and frequently change by vote or budget. When emissions are reduced or halted, APYs can fall quickly. If fee income alone is thin, positions that looked profitable with rewards can flip to unprofitable once incentives stop.

Incentives can make LPing work for a time, but they are not durable income. Always measure base APY first, then treat rewards as a temporary boost that can vanish.

LP profit is not a single number; it is the net of fee capture while your liquidity is active and is affected by inventory effects as price moves, which create or reduce impermanent loss. Profits are also driven by management choices and costs, such as how often you reposition, how often gas is paid, and any hedges you run. Small changes to any of these can turn your profits into losses.

Is LPing Profitable?

Why this is hard to measure

LP profit is fee income minus inventory drift and management costs, and each of those moves with market conditions. The same position can look positive over a short horizon and negative over a longer one, or vice versa, depending on how much volume crosses your ticks, how long you stay in range, and how price trends while you hold inventory.

In stablecoin pairs, fees are steadier and impermanent loss is smaller, so outcomes are often mildly positive when volume is healthy, and the pool is not crowded. Net returns can still compress once you include gas, rebalancing, and any dilution from new liquidity. In blue-chip volatile pairs, short horizons during busy, mean-reverting markets can look good because fees are high relative to small price moves near your range. As horizons lengthen or markets trend, unhedged inventory drift tends to overwhelm fees and push results into the red unless you actively manage or hedge.

The odds improve when your base fee yield is acceptable without incentives, you post ranges that keep you in market most of the time, you operate where gas is low, and you respond to regime changes in volume and volatility. They worsen when pools are crowded, gas is high, your range is narrow and left unattended, or price trends one way for an extended period.

Bottom line. LPing can be profitable, but it is not a passive yield instrument. Results are conditional on realized volume near your ticks, time in range, price path, and costs. Small changes in those inputs or in how you measure returns can flip the sign.

Conservative Yield Estimates

Note: The APY figures below reflect swap fees and, when relevant, reward emissions over a 30-day window. They do not include impermanent loss, time spent out of range, gas or rebalancing costs, so they are not a full profit and loss. These APYs are rolling snapshots that move with volume, liquidity, and incentive schedules. Where possible we separate base fee APY from reward APY and include the snapshot date.

For stablecoin pools, fee-only yields typically sit in the 0.5–2% APY range on main venues

• Uniswap v3 DAI/USDC showed 0.81% (30-day APY, Feb. 14, 2025)

• Uniswap v3 USDC/USDT showed 1.65% (Feb. 25, 2025)

• When incentives are active, totals can step up into the 3-8% band

• Curve’s USDC/USDf displayed 9.4% (Aug. 8, 2025) with most of that coming from rewards

• FRAX/USDe showed 5.8% (Jan. 24, 2025) driven largely by incentives

• Curve 3pool APY hovered around 1%, underscoring how low base fee yields can be without rewards

For blue-chip volatile pairs (e.g., wETH/USDC, wBTC/USDC), fee APYs are much more flow and volatility-sensitive. Quiet weeks can be low single-digits, but busy months often print mid-single to low-double digits, with occasional spikes.

Additional Risks

Toxic Order Flow

When the real price moves on outside markets, fast traders or bots rush to hit your pool before it updates. They buy from you right after an upward move and sell to you right after a downward move, so you end up selling low and buying high. Fees help but often do not cover this cost, especially in volatile markets or thin pools where a few informed trades dominate. The problem is smaller when there is plenty of everyday, uninformed trading that lets you earn fees without taking the worst side of the move.

Pool and fee tier choice

Some pairs attract steady everyday trading while others have thin volume and sharp moves. If the fee tier does not match the pair’s volatility and typical trade size, you may earn too few fees or get little volume at all.

Price path and market regime

What happens during your holding period matters, not just the start and end prices. Calm and rangebound periods tend to help; long one-way trends tend to hurt as the pool keeps rebalancing into the move.

Tight ranges near spot

Very narrow bands behave like limit orders. Small fees rarely cover the loss when the price runs through your band, especially once slippage and less-than-perfect entries are counted.

Just-in-time liquidity and MEV

Trying to appear only for big trades is hard. Faster bots and MEV, which is the profit from reordering or inserting trades, can take the edge and leave you with worse fills and higher gas costs.

Hedging and payoff shape

Opening an LP position is like selling insurance against big price swings. Hedges cost money, can lag in fast markets, and may still leave you with large losses in a jump.

Measurement and benchmark risk

Results can flip depending on how you score them. Profits in dollars can hide losses in the coins you deposited, and onchain data can miss offchain hedges or context.

Yield Farming and Incentive Pools

Yield farming means taking the LP token you receive for providing liquidity and staking it in a rewards contract to earn extra tokens. Incentive pools are the venues that pay those extra rewards to attract liquidity.

Example: farming CRV on Curve using DAI

1. Deposit DAI into a Curve DAI pool to earn trading fees.

2. Receive the pool’s LP token.

3. Stake that LP token in the Curve pool to earn CRV incentives while you keep earning swap fees.

4. Claim CRV at any time and choose what to do with it. You can sell it, hold it, lock it for boost, or deploy it in other protocols such as lending.

Where does the yield come from?

• Swap fees - Traders pay a fee on each swap. Your share is pro rata as long as your liquidity is active in the pool.

• CRV emissions - The pool’s gauge distributes CRV to staked LP tokens. Rewards vary with gauge weights, emissions, and your boost.

• Optional boosts and wrappers - Locking CRV for veCRV can increase rewards. Staking via Convex can pass boosted CRV and add CVX or partner incentives.

• Net result depends on costs and risk - Headline APYs do not include impermanent loss, time out of range, gas, or price risk of reward tokens. Emissions and TVL change over time, which moves yields.

You can also route through an optimizer such as Convex. Convex stakes the LP tokens for you, passes through CRV and sometimes extra tokens, applies a boost so your share of emissions is larger, and auto-compounds on a set cadence. This can raise net yield if gas is reasonable and the optimizer’s fees are modest.

Conservative Yield Estimates

Note: The APY figures below reflect swap fees and, when relevant, reward emissions over a 30-day window. They do not include impermanent loss, time spent out of range, gas or rebalancing costs, so they are not a full profit and loss. These APYs are rolling snapshots that move with volume, liquidity, and incentive schedules. Where possible we separate base fee APY from reward APY and include the snapshot date.

For Convex wrapped Curve LP tokens, a conservative baseline is 2-10% APY across pool types depending on incentives and market conditions.

• Stablecoin pools via Convex - Fee only returns are often low single digits. With CRV and CVX rewards active, a conservative band is about 2-6% APY.

  • FRAX sDAI at 3% 30-day APY in January 2025 with 0.7% base and 2.3% rewards.

  • Incentivized outliers exist, for example PYUSD/crvUSD at about 10.2% 30-day APY in July 2025, driven mostly by rewards.

• Blue chip mixed pools via Convex - Returns are more volatile and reward-dependent. A conservative band is about 4%-10% APY when programs are active.

  • Convex TriCryptoFRAX printed about 5.8% 30-day APY on a prior snapshot, with rewards as the main driver.

• LST or ETH pairs via Convex - These tend to sit in the 2%-4% APY range in quiet conditions, with roughly half from fees and half from rewards.

  • stETH/frxETH at about 2.6%-3.4% 30-day APY on Aug. 8, 2025, split near-even between base and rewards.

• Governance token and Curve ecosystem pairs - These can screen a bit higher when incentives are pointed at them.

  • CRV/cvxCRV showed about 9.8% 30-day APY in May 2025, mostly rewards. Treat these as higher betas.

Risks

Stablecoin pools still carry peg risk and smart contract risk. Incentives can change by governance, so CRV emissions and your APR can fall. If you choose a volatile pair instead of a stable pool you introduce impermanent loss, which can outweigh fees and incentives during strong price moves.

Insights

TVL by Category

DeFi has matured from a DEX and yield farming story into a stacked yield economy. The chart above shows TVL broadening across liquid staking (now a dominant source of onchain capital), lending/borrowing (cyclical with market activity), and DEXs (steady core infrastructure), with newer wedges from restaking, LRTs and RWAs climbing from a small base. Compared with 2020–2022 incentive-driven spikes, today’s TVL looks more diversified and durable, anchored by protocol revenues.

DeFi’s growth reflects multiple engines operating in parallel, not a single narrative. As interest-rate regimes and L2 adoption evolve, TVL has rotated, and may continue to rotate, across segments from base-rate RWAs to protocol-security staking to flow-driven DEX and lending markets, each with distinct mechanics and risk profiles.

Yield at a Crossroads: Banks and Blockchains

Now that we’ve mapped how onchain yield works end-to-end, let’s look at its spillovers to traditional markets, banks, money markets, and other off-chain yield vehicles.

Onchain stablecoins and yield-bearing cash instruments are increasingly intertwined with traditional money markets, as stablecoin reserves cycle into T-bills and repos while tokenized cash competes with bank deposits and retail money funds for the same short-term dollars. If these parallel money markets scale, balances could migrate, potentially reshaping short-term funding, collateral flows, and rate transmission. In many developing economies, they could become everyday money for savings and payments, pulling deposits from local banks and influencing exchange rates.

GENIUS Act

A fundamental question follows from the onchain yield opportunity: what impact might it have on yields offered by traditional bank savings accounts?

The U.S. GENIUS Act, signed into law by President Trump in July, tries to keep payment stablecoins “money-like,” and prevent them from becoming a savings product. Among other guardrails, it prohibits regulated stablecoin issuers from paying any yield or interest to holders, discouraging them from marketing stablecoins as deposit substitutes. However, USDC issuer Circle has a longstanding arrangement with Coinbase that circumvents the prohibition. Instead of funneling yield directly to USDC holders, Circle shares a portion of the USDC reserve income with Coinbase, and Coinbase in turn funds USDC rewards it pays to the holder. Economically, holders receive yield; legally, it’s Coinbase, not the issuer, making the payment. Doing it this way does not violate the issuer-side prohibition, because Coinbase and Circle publicly describe this as a revenue-share model. Will other issues follow suit and look for ways around the GENIUS Act’s guidelines?

The Bank Policy Institute, a lobbying group, argues that letting distribution partners pass reverse income to users indirectly defeats the statute’s purpose. It has urged Congress and regulators to close the “payment-of-interest” workaround by banning interest paid indirectly via affiliates or agents as well as directly. U.S. policy is converging on no interest paid by stablecoin issuers to keep payment tokens from looking like bank deposits, yet for now revenue-sharing remains a live channel for yield to reach retail. Whether that persists will hinge on how the final rules interpret “paying interest” and whether they lump the intermediaries in with issuers.

In April, the Treasury Department speculated that allowing interest-bearing stablecoins could pull as much as $6.6 trillion from bank deposits into stablecoins, depending on how broadly yield is offered. That estimate suggests a real funding risk for banks; large outflows would shrink their low-cost deposit base, push them toward more expensive wholesale funding, and ultimately tighten credit and raise borrowing costs for households and businesses. For users, the math is the opposite: onchain wrappers that pass through T-bill yields or DeFi rewards become a compelling alternative to paltry bank savings. Policymakers are weighing the trade-offs as yield-bearing stablecoins stand to drain deposits and alter money-market dynamics even as they boost T-bill demand.

High-Inflation Countries Seeking Stablecoins

As onchain finance matures, yield-bearing stablecoins and tokenized cash accounts give savers reasons to move balances off traditional rails and into programmable dollars. Capital follows utility: always-on settlement plus the prospect of a modest return can pull day-to-day money onto open networks, especially where bank access is limited or fragile. Across much of the world, inflation is a daily reality. In dozens of countries running double-digit and even >25% annual inflation, households are fleeing local currency into dollar-pegged stablecoins as a practical hedge against debasement. Mobile wallets and P2P rails let people move value 24/7, bypass capital controls, and hold a unit of account that behaves like dollars, all without needing a bank account. In high-inflation environments, stablecoins increasingly function as the working money, used for savings, invoices, remittances, and cross-border commerce, because they combine familiar dollar pricing with global, near-instant settlement.

This migration isn’t risk-free. Users still face issuer/custodial risk, potential freezes, and the difficulty of cashing in and out when on-ramps are restricted. But the pull is strong. Predictable purchasing power and spendability beat rapidly eroding local cash. As onchain rails spread and compliance-friendly wallets improve, expect continued adoption in the very places where money is most fragile. For many, stablecoins aren’t a means of speculation; they’re a means of survival.

In the chart above we look at USDT market capitalization by holder type.

• Services are business wallets, typically exchanges or other merchant/fintech companies, whose addresses were identified by onchain sleuthing firm Chainanalysis.

• Savers are wallets that on average retain 2/3 of the USDT they receive.

• Senders are wallets that hold less than 2/3 of the USDT they receive.

Across the tracked networks (Ethereum, TRON, BSC, Solana, Avalanche, TON, Polygon, Arbitrum, Celo, Optimist, Kaia) the share of USDT market cap held by savers has recently surpassed those of services and senders, suggesting a shift toward using stablecoins as a longer-term store of value rather than purely for exchange balances or rapid payments, consistent with the view that users in high-inflation environments are turning to stablecoins as a hedge against local-currency debasement.

If that saver share keeps rising, balances could migrate out of bank deposits and retail money funds abroad and into onchain cash, pushing banks and funds to compete harder on rate and intraday liquidity. At scale, stablecoin treasuries and tokenized cash accounts would buy T-bills and repos directly, rerouting demand away from traditional money funds, fragmenting global money markets, and subtly changing how policy rates transmit across borders.

Takeaways

Snapshot of the landscape

Yield arises in two broad venues. Offchain cash tools such as high-yield savings, money market funds, and Treasury bills pass through policy rates with large distribution and deep liquidity. They appear here as a baseline because they anchor the prevailing cash rate that onchain yields reference. Onchain instruments add protocol mechanics and market microstructure. These include yield-bearing or staked stablecoins, pooled lending, basis and funding strategies, automated market maker fees, protocol staking and restaking, and structured wrappers that package or lever these cash flows. As designs incorporate more market activity or multiple protocols, the variability of returns and the set of relevant risks expand. Return potential rises as you move from cash-plus toward activity-driven and structured strategies, while risk, monitoring, and operational work rise as well.

Stablecoin risk framework

Stablecoins vary in how they are backed, governed, and redeemed, which means surface similarities can hide very different tail risks. The framework below scores each token across categories that capture where value loss or loss of access can occur, and it consolidates those signals into a simple total for comparison. These scores are not ratings or recommendations; they summarize information so readers can see which risks dominate for a given use case and so risk-adjusted yield can be computed in a consistent way. (For another, highly granular perspective on assessing risk in DeFi, read "A Risk Rating Framework for DeFi and Crypto Investors" by Galaxy Research's Thaddeus Pinakiewicz.)

Stablecoin Risk Categories

Reserve/ Backing Quality– The quality, liquidity, and transparency of collateral, and whether there are genuine overcollateralization or excess reserves.

Peg Stability Mechanism – Historical and structural likelihood the token trades materially away from $1.

Oracle Risk – Dependence on external price feeds and the chance a bad oracle misprices collateral.

Censorability/Centralization – Degree to which an issuer or contract administrator can freeze, blacklist, or otherwise halt specific transfers. Regulated issuers have to maintain this capability to remain in business; nonetheless, it exposes holders to unilateral seizure or transactional censorship.

Redemption – Ease, speed, and conditions under which holders can convert the token back to dollars (or base assets), including any KYC gates, minimum sizes, fees, or cooldown periods that could impede timely exit.

Liquidation – Probability that forced sales or onchain liquidations cascade in stressed markets.

Counterparty– Fragility of the off-ramp/on-ramp venues that users rely on to move the token.

Regulatory – Clarity, rigor, and jurisdiction of the legal regime supervising issuance and reserves.

Smart Contract – Technical security of the contract code, audits, upgradability controls, and pause/blacklist powers.

Governance - Concentration and accountability of decision making over upgrades, parameters, reserves and emergency actions, including key management, quorum and transparency.

Risk-Adjusted Yield

Risk-adjusted yield compares a token’s payout to the amount of risk you assume to earn it. For each stablecoin we divide its 30-day trailing APY by the total risk score according to our matrix above, then convert the result into basis points (bps) per risk point. A value of 18bps, for example, means the coin has been delivering 0.18 percentage points of annual yield for every point of risk in the matrix. Higher numbers signal a more efficient “return per unit of risk,” while lower numbers indicate that investors are accepting sizable risk without being compensated by commensurate yield.

The chart above compares headline APY to a risk-adjusted figure that discounts yield for riskier assets. Centralized cash stables such as USDC and BUIDL cluster near a risk-adjusted 2%-2.5%, which reflects modest risk with steady reserve income. PYUSD lands in the same range. Among decentralized yield stables, the spread widens. sUSDf screens highest at roughly 3.4% adjusted because it started with a double-digit APY before the risk haircut. sUSDe and sUSDS settle near 2%, while sDAI is lowest at about 0.7% due to smaller headline APYs. Tokenized yield products like BENJI and USDY come in just under 2%. The takeaway is that higher raw APY does not always survive the risk penalty and, in this snapshot, only sUSDf clearly tops the centralized cash tier on a risk-adjusted basis.

Returns vs risk, and returns vs complexity

Two complementary lenses place these instruments on an efficiency spectrum. Together they separate the economics of the payoff from the complexity of running it, so readers can see which designs deliver return for the risks they take and how much of that return persists after accounting for operational complexity.

Return, Risk and Complexity Score

The chart below places each yield design on three simple scales so you can see the tradeoffs at a glance. Scores range from 1 (low) to 5 (high). The point is comparison rather than precision. Look for patterns in the bar heights to see which instruments deliver more return relative to their risk and how much day-to-day work they tend to require. Scores reflect current design choices and recent market behavior and can change as liquidity, incentives, or protocol parameters evolve.

Returns Adjusted for Risk and Complexity

The next chart translates return into two efficiency indices. The black bars show return per unit of measured risk. The orange bars apply an additional penalty for operational complexity. The gap between the two bars indicates how much of the efficiency survives once custody, workflows, and monitoring are considered. Scores are comparative snapshots based on design features and recent market behavior.

Insights

The efficiency view shows a clear pattern. Cash-like tools score highest on return per unit of risk and lose little after the complexity penalty. Tokenized cash and fixed-income-style instruments like Pendle PT retain most of their efficiency. Native staking sits mid-pack, while yields for restaking and LRTs fall more once complexity is applied due to validator, incentive, and integration dependencies. Structured and incentive-driven designs look strong on risk-adjusted return but less so after considering complexity, and lending markets land around the middle. Active AMM LPing shows the lowest efficiency and the largest drop after complexity, consistent with path dependence and inventory effects.

Conclusion

This paper maps where onchain yield originates across five engines: policy rate pass through in tokenized cash; protocol rewards from staking and restaking; credit and basis spreads in pooled lending; market microstructure fees from automated market makers; and structured wrappers that package or lever these flows. When designs are combined, these engines, return variability, and potential failure points change.

We compare instruments on three axes. The return profile covers payout magnitude, steadiness, and reliance on incentives. The risk set spans backing and redemption, technical and oracle dependencies, liquidity and exit, governance and censorability, and any leverage. Operational complexity captures custody, cross-app and cross-chain workflows, monitoring, and reporting. Offchain cash tools appear as a baseline because they anchor the cash rate that many onchain designs reference.

Measurement and context matter. Rolling realized yield can differ from headline snapshots. Outcomes for activity-based designs depend on time in market, price path, and market regime. Incentive-driven components are temporary. Efficiency improves when more of the payout persists without them.

Dependencies are part of the product. Many designs stack a chain, a bridge, an oracle, a money market, and a staking layer. Because the same components are often reused across products and protocols (e.g., the same L2, bridge, oracle, or LST), a problem in one service can hit many positions at once, creating correlated stress. Liquidity and exit sit on equal footing with return. Market depth, redemption terms, queues, and gates determine how positions behave when conditions tighten.

Onchain yield is not magic. It is policy rates wrapped, protocol rewards metered, credit priced, and fees shared. The labels change; the engines do not. Read every headline number through two lenses: what you earn for the risk, and what remains after the work involved.