Decentralized finance has spent half a decade wrestling with a deceptively simple problem. When protocols emit governance tokens and let holders decide where rewards flow, anyone can buy in cheaply, vote in their own short-term interest, and exit before the consequences of those decisions land. The result, repeatedly demonstrated through DeFi’s various boom-and-bust cycles, is a system that attracts speculative capital but struggles to retain stakeholders who genuinely care about the protocol’s future. Vote-escrowed tokens, commonly abbreviated as ve-tokens, were designed as a structural answer to this problem.
The premise is straightforward. Rather than granting governance power based on token ownership alone, ve-token systems require holders to commit their tokens to a time lock, often for years at a stretch, in exchange for voting rights and economic rewards. The longer the lock, the greater the influence. Because locked positions cannot be sold during the commitment period, anyone who votes on a protocol’s direction must live with the consequences of that vote for an extended period. This single design choice converts governance from a low-stakes activity that mercenary capital can game into a higher-stakes commitment that filters for participants with skin in the game.
The model has produced some of decentralized finance’s most enduring revenue-sharing mechanisms alongside some of its most public structural failures. Curve Finance, which pioneered vote-escrow in September 2020, captured roughly 44 percent of all Ethereum decentralized exchange fee revenue by the end of 2025, while veCRV holders received more than 157 million dollars in lifetime distributions. At the same time, the twelve months running into early 2026 saw three major protocols, namely Pendle, PancakeSwap, and Balancer, retire their ve-token systems entirely after each ran into governance attacks, concentrated voting power, or unprofitable market dynamics. Few mechanisms in decentralized finance have generated such divergent outcomes from the same underlying design.
This article examines vote-escrowed tokens from the ground up. It begins with the basic mechanics of how locking and decay work, moves through gauge voting and the bribery markets that emerged around it, and examines three notable implementations spanning success and abandonment. From there it turns to a stakeholder-by-stakeholder analysis of the benefits and challenges that explain why the same design has produced such different results across protocols. The goal is to give beginners a clear understanding of how ve-tokens function and why they matter, while also providing the context needed to evaluate whether the model belongs in the protocols readers may encounter or build. Throughout, the analysis draws on verified data from the implementations themselves rather than the theoretical claims that often accompany discussions of governance design, with the recent wave of abandonments treated as evidence about the model’s structural boundaries rather than as proof of failure.
Understanding the Foundations of Ve-Token Economics
Vote-escrow refers to a tokenomics design in which governance power and economic rights derive from locking a token for a defined period rather than from simply holding it. The mechanism was introduced by Curve Finance in September 2020 as a response to a problem that decentralized exchanges had been facing for years. Liquidity mining programs distributed governance tokens generously, attracting capital quickly, but the same programs created what observers came to call mercenary capital. Users would deposit liquidity, harvest rewards, sell those rewards on the open market, and move on to whatever protocol offered the next attractive yield. The protocols themselves were left with falling token prices and shallow liquidity that disappeared the moment incentives weakened.
Vote-escrow proposed a different philosophical foundation for governance influence. Rather than treating every token holder as an equal voting member regardless of intention or commitment, the model treats time as a credible signal of conviction. A holder willing to lock tokens for four years is demonstrating something that a holder who could sell at any moment cannot: a real economic commitment to the protocol’s long-term health. By scaling governance weight and economic rewards with both quantity locked and duration committed, ve-systems route influence toward participants whose interests genuinely align with the protocol’s future. This is the foundation on which everything else, from gauge voting to bribery markets to wrapper protocols, is built.
The Core Mechanics of Vote Escrow
When a user wishes to participate in a ve-token system, they begin by depositing their governance tokens into a vote-escrow contract and selecting a lock duration. The locking process is irrevocable for the chosen period: the user cannot withdraw, transfer, or sell those tokens until the lock expires. In return, the contract issues vote-escrowed tokens, often denoted by adding a “ve” prefix to the original token name. Someone locking CRV receives veCRV, locking PENDLE produced vePENDLE, locking VELO produces veVELO, and so on. These ve-tokens are non-transferable by design. They cannot be sold on an open market, sent to another address, or otherwise circulated in the way ordinary tokens can.
The maximum lock duration varies meaningfully across protocols, and these choices reflect different philosophies about commitment and accessibility. Curve set its maximum at four years, the longest in the ecosystem, with a minimum lock of one week. Pendle chose a shorter ceiling of two years, partly to reduce the barrier to participation. Balancer’s veBAL system used a one-year maximum and uniquely required users to lock a liquidity pool token (the BAL/WETH 80/20 pool token) rather than the BAL governance token directly. Velodrome and Aerodrome, building on the ve(3,3) variant, also use a four-year maximum but represent locked positions as ERC-721 non-fungible tokens called veNFTs rather than balances tracked at the wallet level. The choice of representation affects what holders can do with their locks, including the ability to transfer the position itself to another wallet.
A concrete example clarifies how this scales. Suppose Alice locks 1,000 governance tokens for the full four-year maximum on a Curve-style system. At the moment of locking, the contract credits her with approximately 1,000 ve-tokens of voting weight, the maximum possible for that token quantity. If Bob locks the same 1,000 tokens for only one year, he receives only about 250 ve-tokens of voting weight, since his commitment is one quarter of Alice’s. If Carol locks 4,000 tokens for one year, she receives the same 1,000 ve-tokens of weight as Alice did, achieving the same influence with four times the capital but one quarter the duration. The system rewards a combination of capital and patience, with patience often weighing as heavily as raw capital in determining who shapes protocol decisions.
Time-Weighted Voting and the Decay Function
What distinguishes ve-tokens from simpler staking arrangements is the decay function. Voting power does not remain constant throughout the lock period. Instead, it decreases linearly over time as the unlock date approaches. Alice’s 1,000 ve-tokens at the start of her four-year lock will be approximately 500 ve-tokens at the two-year midpoint, 250 ve-tokens with one year remaining, and effectively zero immediately before her tokens unlock. The mathematical relationship is simple: at any moment, a holder’s voting weight equals their locked token amount multiplied by the fraction of the maximum lock period that remains. A holder who wants to maintain meaningful influence must therefore actively extend their lock or lock additional tokens.
This design choice creates a fundamentally different governance dynamic than systems where vote weight remains static. In a static system, an early supporter who locked tokens five years ago retains exactly as much influence as the day they locked, regardless of whether they are still paying attention to the protocol or have effectively abandoned it. In a ve-system, that same supporter must continually recommit to keep their influence current. The decay function effectively distinguishes between past commitment and ongoing commitment, with governance power flowing toward the latter. It also creates a natural renewal of the active stakeholder base, since holders who lose interest will see their influence diminish even without selling.
Holders can adjust their position in two ways. They can increase the amount of tokens locked at any point, adding to their existing position, or they can extend the lock duration to push the unlock date further into the future. Both actions increase voting weight. A holder one year into a four-year lock can extend their commitment back out to four years, restoring their voting weight to the maximum. This extension mechanism creates an ongoing relationship between holder and protocol rather than a one-time transaction. Some protocols additionally cap how much voting weight any single address can accumulate to prevent extreme concentration, though these caps are uncommon in the original Curve implementation. The cumulative effect is a system in which time itself becomes the fundamental currency of governance influence.
The combination of irrevocable time-locks, non-transferable ve-balances, and linear decay creates an internal logic that pure ownership cannot replicate. Holders cannot extract governance rights without genuine commitment, cannot influence outcomes without ongoing engagement, and cannot exit during the consequences of their decisions. Whether this logic produces the alignment its designers hoped for depends heavily on what those holders are voting on, a question the next section examines through the lens of emissions and gauge voting.
Gauge Voting, Emissions, and Bribery Markets
The locking and decay mechanics described above produce voting power, but voting power matters only because of what it controls. In most ve-systems, the most economically significant decision that ve-token holders make every week is where the protocol’s token emissions should flow. Protocols like Curve emit new governance tokens continuously, distributing them to liquidity providers across dozens or hundreds of pools, and the question of which pools receive how much emission is decided by ve-token votes. This makes governance influence directly economically valuable in a way that abstract policy votes are not. Every weekly vote translates into actual dollars flowing toward specific assets and trading pairs.
That direct economic value transformed ve-tokens from a governance curiosity into one of decentralized finance’s most actively contested commodities. Once voting power could be cashed in for emission flows, secondary markets emerged in which external protocols pay ve-token holders to vote in particular ways. The result has been a layered ecosystem in which voting rights, emission flows, and bribery payments all interact in ways that have produced both the most resilient liquidity protocols in DeFi and some of its most concerning concentrations of power. The next two subsections trace this layering, beginning with the gauge mechanism itself and then moving to the bribery markets that grew around it.
How Gauges Direct Protocol Emissions
A gauge, in ve-tokenomics terminology, is a smart contract attached to a specific liquidity pool that tracks how much liquidity that pool holds and how much of the protocol’s emissions should flow to it. Each pool eligible for emissions has its own gauge, and ve-token holders vote on the relative weights assigned to those gauges. If a pool’s gauge receives 10 percent of total voting weight in a given epoch, that pool’s liquidity providers receive 10 percent of that epoch’s emissions distribution. This direct mapping between votes and rewards turns gauge weighting into the most consequential weekly decision in many ve-protocols.
Gauge voting typically operates on a seven-day epoch cycle. On Curve, votes can be adjusted or recast continuously, and the cumulative vote weight at the end of each weekly snapshot determines the following week’s emission distribution. This cadence makes gauge voting a high-frequency activity by the standards of on-chain governance, where most protocol-level votes occur monthly or less. The weekly rhythm also means that liquidity providers receive immediate feedback on whether their chosen pools are attracting voting support, and protocols seeking deeper liquidity for their tokens must compete continuously rather than winning a one-time vote.
The Curve implementation provides a concrete illustration of how this works at scale. veCRV holders direct CRV emissions across dozens of stablecoin and like-asset pools, with weekly votes determining how billions of dollars in liquidity get incentivized. A pool that wins significant gauge weight attracts deeper liquidity because liquidity providers earn substantial CRV rewards by depositing there. Deeper liquidity then enables lower-slippage trading, which generates more trading volume and protocol fees, which in turn flow back to veCRV holders. The mechanism creates what observers have called a liquidity flywheel: votes attract liquidity, liquidity generates fees, fees reward voters, and voters direct more emissions. When this flywheel works, it can sustain dominant market positions for years, as evidenced by Curve maintaining over 3 billion dollars in total value locked across multiple market cycles. When it breaks down, as it did for several protocols in 2025, the consequences cascade rapidly through the system as liquidity providers exit pools whose emissions no longer justify the deposit, fees collapse alongside volume, and the value of the underlying ve-position falls accordingly. The self-reinforcing nature of the flywheel works in both directions, which is why the design has produced both unusually persistent winners and unusually rapid collapses among the protocols that adopted it.
Vote Aggregators and the Rise of Bribery
Because gauge weights translate directly into emission flows, and because emissions are economically valuable, external protocols have strong incentives to pay ve-token holders to vote for pools that benefit them. This payment system, universally called “bribery” in DeFi terminology despite the legitimate economics behind it, emerged organically within months of Curve’s launch. The basic mechanic is straightforward. A protocol that wants more CRV emissions flowing to a particular pool, perhaps one that holds the protocol’s own stablecoin paired against a major asset, posts a bribe in some token they hold. Any veCRV holder who votes for that gauge can claim a proportional share of the bribe in addition to their normal voting rewards.
Specialized platforms emerged to coordinate these payments. Votium, Hidden Hand, and Warden each built marketplaces where protocols can post bribes, voters can claim them, and aggregators can route votes toward whichever bribes maximize voter returns. These markets turned governance participation into a meaningful yield-generating activity. Bribe income on ve positions has at times reached 20 to 50 percent annual returns, often substantially exceeding the base protocol fee distributions. For ve-token holders who actively participate, the bribery layer can become the primary source of returns, with bribes from external protocols outweighing fees from the protocol’s own operations.
The most consequential development in the bribery ecosystem was the rise of wrapper protocols, of which Convex Finance is the canonical example. Convex recognized that locking CRV for four years was an unappealing commitment for most users, and built a system in which users could deposit CRV with Convex, receive a liquid wrapper token in return, and earn the same boosted yields that direct veCRV lockers receive. Convex itself locked the aggregated CRV deposits permanently, accumulating ultimately more than half of all veCRV in existence. This made Convex the single most powerful voter on Curve, and quickly attracted similar dynamics on other ve-protocols. The veBAL ecosystem produced Aura, with Aura accumulating over half of total veBAL supply. The vePENDLE ecosystem produced Penpie and Equilibria, which together accumulated roughly half of delegated vePENDLE before Pendle ultimately retired the model. These wrappers solved a real user problem but also concentrated governance power in ways that ve-tokenomics was originally designed to prevent.
The gauge and bribery infrastructure represents the productive layer of ve-tokenomics, the part that converts locked tokens from a passive position into an active economic engine. It also represents the layer where most of the model’s controversies and failures have originated. Whether the bribery market constitutes healthy competition for liquidity or a corruption of governance is a question that protocol designers and observers continue to debate, with the answer often depending on which protocol and which year is being examined.
Notable Implementations Across the DeFi Ecosystem
Curve established the original vote-escrow template in 2020, but the model has since been adapted, modified, and in several recent cases abandoned by dozens of decentralized finance protocols. Each implementation reflects different assumptions about what ve-tokenomics should optimize for, what trade-offs are acceptable, and how the inherent tensions between commitment, liquidity, and accessibility should be resolved. Looking across the implementations that have shaped the model’s evolution offers concrete evidence for which design choices have proven durable and which have not.
The three implementations examined in this section span the full range of outcomes ve-tokenomics has produced. Curve’s veCRV remains the longest-running and most economically significant ve-system, with revenue accruing to lockers for more than five years and the protocol’s market position arguably stronger in 2026 than at any prior point. Velodrome and Aerodrome adopted the ve(3,3) variant that Andre Cronje first introduced through the short-lived Solidly protocol, modifying Curve’s original design to address several of its observed weaknesses and producing what many analysts consider the cleanest current implementation. Pendle’s vePENDLE system, by contrast, was widely admired during 2023 and 2024 but was retired by the protocol in January 2026 in favor of a liquid staking model called sPENDLE. Together, these three cases illustrate both what makes ve-tokenomics work and what makes it fail.
Curve Finance and the Original veCRV Blueprint
Curve Finance launched its CRV governance token in August 2020 and introduced the vote-escrow model the following month. The design was distinctive from the outset. Of all fees generated by trading activity on Curve, 50 percent flow to veCRV holders as protocol revenue, while the other half goes to liquidity providers. Voting power scales with both lock amount and lock duration on a four-year maximum, with a minimum lock of one week. veCRV holders also receive a boost of up to 2.5 times on their own liquidity provision rewards, creating an additional incentive to lock alongside providing liquidity. This combination of fee sharing, governance influence, and yield boosts has anchored the veCRV value proposition for more than five years.
The verified data underlying Curve’s continued success is substantial. By early 2026, the protocol was generating approximately 850,000 dollars per month in revenue distributed to veCRV holders, with lifetime distributions exceeding 157 million dollars. Roughly 40 to 50 percent of circulating CRV supply was locked as veCRV, with an average lock duration of approximately 2.86 years across protocols using comparable models. In August 2025, Curve reduced annual CRV inflation from a previous rate of approximately 6 percent to 5.02 percent, cutting roughly 22 million tokens per year in potential sell pressure. Trading volume grew from approximately 119 billion dollars in 2024 to 126 billion in 2025, and Curve’s share of all Ethereum decentralized exchange fee revenue rose from approximately 1.6 percent at the start of 2025 to roughly 44 percent by December.
Curve has continued expanding the productive surface of veCRV through new products. In September 2025, the Curve DAO approved Yield Basis, a protocol built on top of Curve infrastructure that enables Bitcoin holders to earn DeFi yields through WBTC, cbBTC, and tBTC pools while mitigating impermanent loss. The Yield Basis launch was backed by 60 million dollars in crvUSD, Curve’s native stablecoin, and between 35 and 65 percent of Yield Basis yields flow back to veCRV holders. By December 2025, Yield Basis pools were generating some of the highest fee growth on the platform. The pattern across these expansions is consistent: each new product Curve builds funnels additional revenue streams toward locked veCRV holders, strengthening the economic case for the lock and reinforcing the underlying flywheel that has sustained the protocol since 2020.
Velodrome and Aerodrome’s Ve(3,3) Variant
The ve(3,3) variant of vote-escrow tokenomics was introduced by Andre Cronje in January 2021 through the Solidly protocol on Fantom. The name combines vote-escrow with the (3,3) game theory framework popularized by Olympus DAO, with the basic claim being that aligning all participant incentives toward staking and locking produces better outcomes for the system than allowing free exit. Solidly itself failed quickly after Cronje stepped away from cryptocurrency development, but the ve(3,3) framework was adopted and refined by subsequent protocols. Velodrome launched on Optimism in May 2022 as the most prominent Solidly fork, and Aerodrome launched on Base in August 2023 in partnership with Coinbase shortly after the Base mainnet went live.
Several design choices distinguish the Velodrome and Aerodrome implementations from Curve’s original model. Locked positions are represented as non-fungible ERC-721 tokens called veNFTs rather than address-balance-tracked positions, making transfers of governance positions possible in ways that Curve’s design prevents. Liquidity providers receive 100 percent of trading fees on the pairs they support rather than the 50/50 split used by Curve, with fees on the gauged pairs flowing entirely to whoever votes for those gauges. Critically, neither protocol implements a liquidity provider boost mechanism. This absence was deliberate. Curve’s boost system was the precise feature that allowed Convex to capture more than half the veCRV supply by offering users the boosted yields they could not otherwise achieve. Velodrome and Aerodrome elected to forgo boosts to prevent the same aggregator dynamic from emerging.
The Aerodrome deployment in particular established itself rapidly as the dominant decentralized exchange on Base, in part due to its launch alongside the network itself and its partnership with Coinbase. Velodrome reached an all-time high price of approximately 0.41 dollars in December 2024 amid broader Layer 2 expansion. In a notable consolidation announced in late 2025, Velodrome and Aerodrome announced a merger under a new entity called Dromos Labs to form a multichain protocol called Aero spanning Base, Optimism, and Ethereum while preserving the core ve(3,3) tokenomics. The merger reflects a broader recognition that ve-systems benefit from concentrating liquidity across chains rather than fragmenting it, and represents one of the more ambitious attempts to scale a single ve-implementation across multiple ecosystems while maintaining the alignment properties the design was meant to provide.
Pendle’s Adaptation and Its 2026 Pivot
Pendle Finance built its yield-trading protocol on a vePENDLE system that adapted the Curve template with several modifications. The maximum lock period was set at two years rather than four, which Pendle judged would lower the participation barrier while still enforcing meaningful commitment. Weekly gauge votes directed PENDLE emissions across the protocol’s yield pools, and vePENDLE holders received 80 percent of swap fees from pools they voted for, distributed pro rata among voters of each pool. The vePENDLE system also incorporated bribery-style payments and incentive caps designed to prevent emissions from flowing to low-performing pools that would not generate meaningful fee returns to voters.
For most of 2024, the vePENDLE system performed well by the standards of ve-tokenomics. Active vePENDLE holders earned an average annualized return of approximately 40 percent excluding airdrops, and Pendle generated significant protocol revenue from yield-trading activity, particularly through products tied to liquid staking tokens and the broader restaking narrative. The system also produced its own version of the Curve Wars dynamic, with two aggregator protocols, Penpie and Equilibria, accumulating roughly 50 percent of all delegated vePENDLE combined as users sought liquid alternatives to direct locking. By late 2024, approximately 30 percent of PENDLE supply was locked as vePENDLE, with an average lock duration of approximately 388 days. The model appeared to be working.
The picture shifted in early 2026. Pendle disclosed in January that an internal review had identified several structural problems with the vePENDLE design. Despite generating more than 37 million dollars in protocol revenue during 2025, only approximately 20 percent of total PENDLE supply was actively engaged under vePENDLE, the lowest participation rate among comparable ve-token models. Pool-level data revealed that more than 60 percent of markets were unprofitable, with concentrated voting power and manual gauge allocation directing emissions to pools that did not generate proportional fees. Pendle launched sPENDLE on January 20, 2026, replacing multi-year locks with a 14-day unstaking period and shifting to algorithmic emissions management. The transition cut overall token emissions by approximately 30 percent and redirected up to 80 percent of protocol revenue toward PENDLE buybacks for sPENDLE holders. The Pendle case demonstrates that ve-tokenomics can deliver substantial revenue and still fail to engage broad participation when the underlying product does not generate the kind of organic liquidity competition that the gauge mechanism rewards.
These three implementations span the full outcome distribution that ve-tokenomics has produced. Curve continues to thrive and expand, Velodrome and Aerodrome demonstrate that thoughtful refinements to the model can avoid some of its concentration risks, and Pendle illustrates that even strong revenue generation cannot save a ve-system whose underlying product lacks the external demand that gauge voting requires to function. The pattern is clear enough to inform any subsequent analysis of who benefits from the model and who does not.
Benefits and Stakeholder Alignment
The benefits of vote-escrowed token systems do not accrue uniformly across all participants. Different stakeholder groups derive different value from the same underlying mechanism, and a clear understanding of who gains what is essential for evaluating whether the model serves the broader objectives it claims to advance. This section examines benefits by category, beginning with the token holders who do the locking and moving outward to the protocols, liquidity providers, and external market participants whose interactions with the ve-system shape the resulting incentive structure.
For long-term token holders willing to commit to locks, the benefits stack across multiple dimensions. Governance rights provide direct influence over protocol decisions ranging from emission allocation to fee parameters to treasury management. Fee sharing converts that influence into measurable income, with Curve veCRV holders receiving 50 percent of all protocol swap fees and Pendle’s vePENDLE system historically directing 80 percent of trading fees to voters. Yield boosts can multiply liquidity provision rewards by as much as 2.5 times on Curve, meaning a locker who also provides liquidity earns substantially more on the same deposit than a non-locker would. Bribery income from external protocols seeking gauge support adds another revenue stream, sometimes the largest of all, with active veCRV positions historically generating bribery yields in the range of 20 to 50 percent annualized during periods of intense competition. The compounding effect of these layered rewards meant that vePENDLE holders earned approximately 40 percent annualized in 2024 across all sources combined, demonstrating that the model can produce genuinely substantial returns for committed participants. And because locked tokens cannot be sold, the overall market experiences reduced sell pressure, which benefits all token holders including those who chose not to lock by lowering the supply pressure that would otherwise weigh on token prices.
For the protocols that implement vote-escrow systems, the model addresses the original mercenary capital problem directly. Liquidity becomes sticky because the most influential providers are also the most committed lockers, and the most committed lockers cannot quickly exit. Governance becomes more thoughtful because votes are cast by participants who will live with the consequences. Emission distribution becomes responsive to genuine market signals as gauge voting reveals which assets attract real demand, allowing protocols to adjust their incentive footprint without requiring centralized decisions about which pools deserve support. The protocol gains a credible mechanism for retaining capital through downturns, when mercenary capital would otherwise flee for higher yields elsewhere. Curve’s persistence through multiple market cycles, including a 60 million dollar exploit in July 2023 and prolonged crypto bear conditions, illustrates this stability advantage in practice. The protocol retained billions of dollars in total value locked through periods when comparable decentralized exchanges saw liquidity drain rapidly, and this stickiness allowed it to continue generating fees through the worst of the bear market that other protocols simply could not match.
Liquidity providers occupy a particular position in ve-economics because they can benefit both directly and through participation in the ve-system itself. A simple liquidity provider deposits assets, receives emission rewards proportional to the pool’s gauge weight, and earns trading fees. A liquidity provider who also locks the protocol’s native token gains boosted rewards on their own deposits, often substantial. The lock decision becomes a calculation about expected returns: does the boost on liquidity provision plus governance benefits plus fee sharing plus bribery income outweigh the illiquidity cost of being unable to sell the locked tokens? For larger and more active providers, the math typically favors locking, which is why providers and lockers tend to be the same actors in mature ve-ecosystems. The dual role also produces a useful feedback loop, since providers who lock develop genuine knowledge about which pools their fellow lockers should support, and the votes they cast tend to reflect informed judgments about where emissions will produce the most productive results. This intersection of operational expertise and governance authority is one of the harder-to-replicate properties of ve-systems, and explains why several protocols have continued operating ve-tokenomics even as the broader ecosystem has faced doubts about the model’s general applicability.
External protocols that need deep liquidity for their own tokens or stablecoins benefit from ve-systems in a different way. Rather than building their own incentive infrastructure from scratch, they can purchase governance influence on an existing system through some combination of accumulating the underlying token and locking it themselves, acquiring proxy tokens like CVX that vote large pools of underlying ve-power, or simply paying bribes to existing voters. For stablecoin issuers, this access to ready-made liquidity coordination infrastructure has been transformative. Frax, Abracadabra, Lido, Ethena, and many others used Curve gauge voting to bootstrap deep stablecoin pools without operating their own emission programs. The bribery option in particular often proves cheaper than direct accumulation, since paying voters a portion of the emission value they direct to your pool can be more capital-efficient than buying enough tokens to control the votes outright. Detailed analyses of bribery economics during the peak of the Curve Wars showed bribers earning roughly 1.88 dollars in emissions for every 1 dollar spent on bribes, a multiplier that justified substantial ongoing payments to ve-holders even when accumulation costs would have been prohibitive. The result is that ve-systems have effectively become public utilities for liquidity coordination in DeFi, with external protocols choosing among multiple payment strategies depending on their cost of capital and their projected duration of demand for emission flows.
At the market level, ve-systems produce several emergent benefits that extend beyond individual stakeholders. Locked tokens reduce circulating supply, often dramatically. The combination of locked positions and ongoing fee distributions creates more stable token economics than pure emission-driven models, which depend entirely on continuous demand for newly minted tokens. The transparent on-chain nature of gauge voting and bribery markets provides one of decentralized finance’s clearest examples of price discovery for liquidity coordination, with the cost of directing emissions to particular pools observable in real time. Curve’s structural conclusion, articulated by founder Michael Egorov, that locked scarcity proves deeper than burned scarcity because locking enables ongoing utility while burning simply removes supply, captures the broader point: ve-systems retain capital productively rather than destroying it. The collective effect across the most successful implementations has been to provide DeFi with one of its few mechanisms for sustained, multi-year alignment between protocols and their stakeholders, a property in short supply across the rest of the ecosystem.
Challenges, Risks, and Failure Modes
The same mechanism that produces alignment also produces tension, and the past several years of ve-tokenomics in practice have revealed structural challenges that reasonable participants can disagree about how to weigh. The challenges fall into a few distinct categories, ranging from concentration of voting power among wrapper protocols to direct governance attacks, from the difficulty of attracting broad participation to the recent wave of major protocols abandoning the model entirely. Understanding these failure modes is essential both for evaluating existing ve-systems and for considering whether new protocols should adopt the model in the first place.
The most persistent criticism of ve-tokenomics is that the model concentrates governance power in the same wrapper protocols that exist to make ve-participation more accessible. Convex Finance accumulated more than half of all veCRV by offering users a liquid wrapper that provided most of the locking benefits without the four-year commitment. Aura accumulated a comparable share of veBAL through similar mechanics. Penpie and Equilibria together controlled roughly half of delegated vePENDLE before Pendle retired the system. In each case, the wrapper solved a real user problem, but the resulting concentration meant that protocols that adopted ve-tokenomics specifically to decentralize governance ended up with majority voting power resident in a single secondary contract. Critics observe that this dynamic effectively transferred control from the protocol’s designed governance structure to whoever holds the wrapper’s governance tokens, which may or may not be the same set of actors the original design intended to empower. The phenomenon has even spawned its own vocabulary, with observers describing the accumulation of wrapper-protocol governance tokens as a form of “governance extractable value,” analogous to maximal extractable value in transaction ordering but applied to the longer time horizons of protocol-level decisions. The concentration concern also applies even when no aggregator is involved, as the Balancer ecosystem demonstrated when a single whale identified only as “Humpy” accumulated approximately 34 percent of all veBAL, slightly more than even Aura’s holdings, showing that ve-systems can concentrate power in individuals as readily as in protocols.
Beyond the concentration concern, several protocols experienced direct governance manipulation that exposed weaknesses in ve-system design. PancakeSwap’s veCAKE system faced what Curve founder Michael Egorov characterized as a textbook governance attack in 2025, with on-chain analysis suggesting that insiders erased existing veCAKE holders’ governance rights and may have forced unlocks of their own tokens following the relevant vote. Cakepie DAO, one of the largest CAKE holders, challenged the vote citing procedural irregularities, and PancakeSwap eventually offered up to 1.5 million dollars in CAKE compensation to affected Cakepie users. Balancer experienced its own governance hijacking and security incidents that contributed to the decision to phase out veBAL in favor of redirecting protocol fees entirely to the DAO treasury. These cases illustrate that even when the locking mechanic functions correctly, the surrounding governance infrastructure can fail in ways that undermine the model’s promised benefits.
The bribery layer attracts its own criticism. Detractors argue that the bribery market commodifies governance in ways that may not serve protocol economics. A voter receiving substantial bribes for supporting a particular gauge has an incentive to vote for that gauge regardless of whether the underlying pool generates meaningful trading volume or fees for the protocol itself. Analyses comparing veVELO and veCRV voting behavior have observed that significant ve-holders sometimes vote for pools processing marginal volume relative to their total value locked, either to collect bribes or to support particular tokens, in ways that direct emissions to economically unproductive locations. When this dynamic dominates, the protocol’s emissions effectively subsidize external interests rather than building its own liquidity network. Whether bribery represents healthy competition for liquidity or a corruption of governance depends substantially on whether the bribed votes also correlate with productive economic activity for the host protocol. A separate concern is that bribery markets favor sophisticated participants who can navigate the complexity of multi-protocol claim processes, with smaller holders often leaving rewards unclaimed because the gas costs of claiming exceed the value of the rewards themselves. The result is that the productive value generated by gauge voting flows disproportionately to large operators and aggregator services, further concentrating economic returns in ways that compound the governance concentration discussed above.
For smaller and less sophisticated participants, ve-systems present meaningful barriers to participation. The complexity of choosing lock durations, understanding decay curves, tracking weekly gauge votes, and claiming bribery rewards from multiple sources requires both time and technical understanding that many holders do not have. Wrapper protocols exist precisely to abstract this complexity, but as discussed, wrappers create their own concentration problems. The illiquidity cost of locking is also non-trivial. A holder who locks tokens for four years and then experiences a personal liquidity need has no recourse, and during severe market downturns the inability to sell locked positions can produce significant unrealized losses with no exit. Pendle’s review of its own data concluded that complexity and long lock requirements contributed directly to the low 20 percent supply engagement rate, suggesting that even active DeFi participants frequently chose to opt out of vePENDLE rather than navigate its mechanics.
The most striking development in ve-tokenomics during 2025 and early 2026 was the wave of major protocols abandoning the model. Pendle retired vePENDLE in January 2026 in favor of the liquid sPENDLE staking token with a 14-day unstaking period and a 30 percent emission reduction. PancakeSwap moved to a burn-based deflationary model targeting roughly 4 percent annual supply reduction after the veCAKE governance attack. Balancer began phasing out veBAL with plans to redirect 100 percent of fees to the DAO treasury rather than to ve-holders. Each case had its own specific trigger, but the broader pattern suggests that ve-tokenomics works most reliably when the underlying protocol generates organic liquidity competition that gauge voting can productively coordinate, as Curve does for stablecoin liquidity. When a protocol’s product is yield trading, a multichain general-purpose decentralized exchange, or programmable liquidity pools, the structural reason for external protocols to compete for gauge emissions may not exist with sufficient depth, and the ve-system fails to generate the productive cycle that sustains the model elsewhere.
Final Thoughts
Vote-escrowed token economics represents one of decentralized finance’s most ambitious attempts to solve a problem with no clean parallel in traditional finance: how to align governance influence with long-term commitment in systems where participation is permissionless and exit is normally frictionless. The mechanism has produced some of the most durable revenue-sharing arrangements in the industry, with Curve veCRV holders receiving distributions for more than five continuous years through multiple market cycles. It has also produced spectacular structural failures, with three major protocols abandoning the model within twelve months between 2025 and early 2026. Few innovations in decentralized finance have generated such a divergent set of outcomes from the same underlying design.
The broader implication for decentralized governance is that voluntary time-binding mechanisms can produce real alignment when the underlying protocol generates genuine demand for the influence that locking grants. Curve’s continued success reflects the fact that stablecoin liquidity coordination is a real economic problem benefiting from coordinated voting on emissions, and that external protocols will compete for that coordination through bribes, accumulation, and active participation. The protocols that abandoned the model in 2025 and 2026 generally did so not because the locking mechanic failed in isolation but because the underlying product did not generate the competitive demand that gauge voting requires to function. The lesson is not that vote-escrow is broken but that it solves a specific class of problems, and applying it outside that class produces predictable disappointment.
The intersection between ve-tokenomics and broader questions of financial inclusion deserves more attention than it typically receives. The model provides a credible path for individuals globally to participate in protocol governance and revenue streams that would be inaccessible in most traditional financial systems, where ownership of revenue-generating assets is typically gated by jurisdictional and accreditation requirements. Anyone with an internet connection and a small amount of capital can lock tokens and receive a proportional share of protocol fees, a meaningful democratization compared to historical norms. Yet the complexity of the system, the illiquidity it imposes, and the rise of wrappers that concentrate power among sophisticated operators have meant that practical benefits often flow most heavily to actors who would have benefited from any incentive structure.
The intersection between technology design and stakeholder responsibility runs through every layer of ve-tokenomics. The decay function operationalizes the principle that influence should require ongoing engagement. The non-transferability of locked positions operationalizes the principle that voters should bear the consequences of their votes. The gauge voting system operationalizes the principle that resources should flow toward what stakeholders genuinely value. Each design choice encodes a normative claim about what fair governance looks like, and the failures and successes of various implementations provide real-world evidence about which claims hold up under adversarial conditions.
Looking forward, the model continues to evolve in response to the lessons of recent years. Hybrid liquid staking approaches like Pendle’s sPENDLE attempt to preserve some alignment benefits of locking while removing the long-duration commitments that depressed participation. Algorithmic emissions management seeks to reduce the influence of concentrated voting power on emission distribution. Cross-chain consolidations like the Velodrome and Aerodrome merger under Dromos Labs reflect an understanding that ve-systems work better when liquidity is concentrated rather than fragmented across chains. Whether these adaptations preserve the core insight of vote-escrow or dilute it remains an open question that the next several years of on-chain evidence will answer. What is already clear is that the original problem ve-tokens were designed to solve, the misalignment between governance power and genuine commitment, has not disappeared, and any successor model must address it as directly as vote-escrow attempted to do.
FAQs
- What is a vote-escrowed token in simple terms?
A vote-escrowed token, often shortened to ve-token, is what you receive when you lock a regular governance token into a smart contract for a chosen period. You give up the ability to sell or transfer those tokens temporarily, and in exchange you receive voting power, a share of protocol fees, and often boosted yields. The longer your lock, the more ve-tokens you receive and the larger your governance influence. - How is vote-escrow different from regular staking?
Regular staking typically lets you withdraw your tokens at any time or after a short cooldown, with rewards calculated for however long they were staked. Vote-escrow requires a much longer commitment, often months or years, and the locked tokens cannot be withdrawn early under any condition. Vote-escrow also typically grants governance rights that simple staking does not, including direct control over how protocol emissions are distributed across liquidity pools. - Why can’t I shorten or cancel my lock once it’s set?
The irrevocability of the lock is what makes vote-escrow function. If users could exit at any time, the model would lose its core feature: the credible commitment that filters for participants willing to live with the consequences of their governance decisions. Allowing early exit would also create gaming opportunities where actors could lock briefly to influence votes and then exit before the consequences materialized, defeating the purpose of the design. - What is the maximum lock duration and why do protocols choose different ones?
Curve set its maximum at four years, which has become the most common reference point for the model. Velodrome and Aerodrome also use four years through their ve(3,3) variant. Pendle chose two years before retiring its system, and Balancer used one year. Longer maximums create stronger alignment signals but also create higher barriers to participation, and protocols make this trade-off differently based on their target user base. - How does voting power decrease over time?
Voting power decays linearly toward zero as the unlock date approaches. A four-year lock that starts at 1,000 vote-tokens drops to roughly 500 after two years, 250 after three years, and approaches zero in the final days before unlock. To maintain governance influence, holders must extend their lock period or add additional tokens to the position. This design rewards ongoing commitment rather than one-time locking. - What does gauge voting actually control?
Gauge voting controls how a protocol’s token emissions are distributed across its liquidity pools. Each pool has an associated gauge contract, and ve-token holders vote weekly on the relative weights assigned to those gauges. A pool that wins higher gauge weight attracts deeper liquidity because its liquidity providers receive more emissions, which is why external protocols compete intensely to capture or influence gauge votes on the leading ve-systems. - What are bribery markets and are they problematic?
Bribery markets are platforms like Votium, Hidden Hand, and Warden where external protocols pay ve-token holders to vote for particular gauges. Despite the negative-sounding name, these are legitimate economic incentives where the value of voting power is openly priced and competed for. Whether they are problematic depends on whether the bribed votes correlate with productive trading activity or simply direct emissions toward economically unproductive pools. - Why did protocols like Convex and Penpie become so powerful?
Wrapper protocols solve a real user problem: most holders do not want to lock their tokens for four years. By accepting deposits, locking the aggregated tokens permanently, and issuing liquid wrapper tokens in return, Convex and similar protocols let users access most ve-benefits without the long commitment. Their success accumulating large shares of locked supply, however, concentrated governance power in ways that the original ve-design was meant to prevent. - Why did Pendle abandon vePENDLE in early 2026?
Pendle determined that only approximately 20 percent of PENDLE supply was actively engaged under vePENDLE, the lowest participation among comparable models, and that more than 60 percent of its markets were unprofitable because concentrated voting power directed emissions to pools that did not generate proportional fees. The replacement sPENDLE system uses a 14-day unstaking period, algorithmic emissions management, and a token buyback mechanism funded by protocol revenue. - Is ve-tokenomics still a viable model for new DeFi protocols?
The evidence suggests ve-tokenomics works reliably when the protocol generates organic competition for emission flows, as Curve does for stablecoin liquidity. For protocols where external entities have less structural reason to compete for gauge votes, the model may produce low participation and unprofitable pools regardless of the protocol’s other strengths. New protocols considering vote-escrow should evaluate whether their product naturally generates the demand for liquidity coordination that the model requires.