LayerBank: lending, supply yield, and the Robinhood Chain markets

LayerBank is a permissionless money market for supplying crypto assets and borrowing against them. This is a plain account of layer bank lending, of what people mean by LayerBank staking, and of how markets on Robinhood Chain sit beside Robinhood staking and Robinhood lending.

The short version

LayerBank Finance is the public face of an on-chain bank. Depositors place assets into shared pools. Borrowers post collateral and draw from those pools. Interest flows from borrowers toward depositors, and the rate is set by how much of the pool is already out on loan. The protocol mints a receipt, called an lToken, so a depositor’s claim can grow as interest arrives. The same pattern has been stood up on a long list of networks. In recent tracker readings, two of them have held most of the deposits: Rootstock, which faces Bitcoin, and Robinhood Chain, an Ethereum layer-2 associated with Robinhood and built with tokenized markets in mind.

That is the machine. Everything else is a refinement of collateral rules, a token that pays incentives, or a warning about what fails. Layerbank crypto is self-custodied finance. You hold the wallet. The contracts hold the pool. A familiar brand name on a chain does not turn the pool into a brokerage account.

The searches for this project arrive in a cluster. People type LayerBank, layer bank, and LayerBank Finance. They type layer bank lending, LayerBank staking, and layerbank crypto. A growing share of those searches puts Robinhood in the same sentence: Robinhood staking, Robinhood lending, and layer bank Robinhood. Those phrases point at neighboring ideas. They do not point at one product. This article keeps them apart, and then explains the lending book on its own terms.

Nothing here is an offer to lend, borrow, stake, or buy a token. Rates move. Pool sizes move. Parameters can be changed by whoever holds the admin power on a given deployment. Read the market in front of you, and treat every yield as the price of a risk.

What LayerBank is

Call it LayerBank, layer bank, or LayerBank Finance and you are talking about one project: a decentralized lending protocol. It is software on public blockchains. There is no branch, no credit officer, and no balance that a bank examiner would recognize as a deposit. A person connects a wallet, supplies an asset the market lists, and receives a claim on that pool. Another person, or the same person, can borrow against collateral the market accepts.

The useful picture is a pawn window and a savings window run by a formula. Suppliers are the savers. Borrowers are the clients. Liquidators are the people who step in when a loan is no longer well secured, and they are paid a bonus in collateral for doing it. Nobody at a desk approves the loan. The parameters do: which assets are listed, how much can be borrowed against each, which price feed is trusted, and how steeply the interest rate climbs when the pool runs tight.

LayerBank’s own description of the work is an omni-chain money market. In practice that means the lending contracts are deployed in many places, and a user meets them on the network where the assets already sit. Moving between those deployments is a bridge problem, a wallet problem, and a liquidity problem. It is not a single shared bank account that follows you from chain to chain. A dollar you supplied on Rootstock is a claim on the Rootstock pool. A dollar you supplied on Robinhood Chain is a claim on a different pool, with its own borrowers, its own rate, and its own way of breaking.

Around the plain lend-and-borrow book, the protocol has added the tools that lending markets grow once people want more capital efficiency. Efficiency mode raises the borrowing ceiling for assets that ought to move together. Looping, sometimes wrapped in an automated vault, repeats a supply-and-borrow cycle to magnify a yield spread. Some listed assets are supposed to carry income from outside the chain, or from another protocol, so the collateral works while it sits. Each of those additions is a feature and a new way to be wrong. The base market is easier to understand if you ignore them until the base market is clear.

There is also a protocol token, ULAB, meant to pay participants, to be locked for a boosted position called xULAB, and eventually to steer governance. The token is part of the incentive design. It is not the receipt for your deposit. Confusing the two is how people end up holding a volatile governance claim and believing they still own the dollars they lent.

From LineaBank to a wider name

LayerBank began on Linea, the Ethereum layer-2 associated with ConsenSys, under the name LineaBank. The early market was a lending book for that one network: supply, borrow, interest, liquidation. The project later carried the same idea onto other chains and took the broader name. Linea remains one network in the set. It is no longer the whole story, and on many days it has not been where the deposits sat.

The rename matters for anyone following an old reference. A page that says LineaBank and a page that says LayerBank can describe the same lineage. They can also describe different code, different listed assets, and different admins. An account of the original Linea market is a poor guide to a bitcoin-collateral pool on Rootstock, or to a tokenized-asset pool on Robinhood Chain. The family resemblance is the money-market shape. The risk is local.

Public dashboards have shown the stock of deposits swelling when a campaign or a fashionable chain pulls money in, and shrinking when attention leaves. A large figure from a busy cycle describes that cycle. Lending protocols in this category have printed very large deposit numbers in hot markets and much smaller ones after. LayerBank has lived through that swing. The lesson is dull and useful: the pool you are about to enter has its own size, its own concentration of borrowers, and its own rate. The historical peak is a story about the past. It does not underwrite the exit today.

Market-data sites have also recorded an early financing round, on the order of a couple of million dollars, with ecosystem names such as Taiko and Rootstock among those attached to the reporting. Treat a cap table you did not see in a primary announcement as secondary information. Funding tells you that someone paid for a period of building. It does not tell you that the contracts will behave, or that liquidity will be there when you want it.

Layer bank lending, from supply to borrow

Layer bank lending has two doors, and they face each other.

The first door is supply. You send an accepted asset into a pool. You are lending it to the pool, and the pool lends it onward to borrowers. You do not choose the borrower, and you do not negotiate a term. You take a share of the interest the pool collects, after whatever fraction the protocol keeps as a reserve. Your claim is fungible with other suppliers of the same asset on that deployment. If many people withdraw at once while most of the pool is out on loan, withdrawals wait until borrowers repay or new suppliers arrive. That tightness is utilization, and it is the central fact of this kind of market.

The second door is borrow. You post collateral the market recognizes. You enable that collateral so it counts toward your limit. You draw another asset, up to a fraction of the collateral’s value. The fraction is the maximum loan-to-value. Borrow eighty against a hundred and you are using most of the rope the protocol allows. Borrow forty and you have left room for the price to move and for interest to accrue. The debt grows by the borrow rate for as long as it is open. Plain collateral just sits. Collateral that itself pays a yield can keep paying while the debt grows, which is the seed of the looping strategies described later.

A small example keeps the nouns honest. Suppose a market allows a maximum loan-to-value of eighty percent on a dollar stablecoin, and suppose you supply one hundred units of it. The borrow limit is eighty. You draw fifty. You still hold a claim on the one hundred. You owe fifty, plus the interest that starts accruing immediately. The gap between the limit and the debt is your margin. If the collateral price falls, or the debt compounds long enough, or the asset you borrowed rises in price, the gap thins. The health factor is the protocol’s way of stating that gap as a single number.

Nothing in that example is a rate card to trust without looking. Maximum loan-to-value is set per asset and per network. A stablecoin book can sit high, sometimes near ninety percent in efficiency mode. A volatile asset sits lower. The screen you use should show the ceiling that applies to the asset you are posting. If the interface hides the ceiling, the liquidation threshold, and the oracle, you are flying on a slogan.

There is a reserve between what borrowers pay and what suppliers receive. The supply rate is utilization times the borrow rate, after that reserve. You cannot earn the headline borrow rate merely by depositing. A quoted supply figure that looks unusually high is usually a pool under strain, a temporary incentive, or the two stacked together. Separate them before you compare LayerBank with any other venue.

Some deployments also expose flash loans. A flash loan is an uncollateralized borrow that must be repaid before the transaction finishes. If repayment fails, the transaction reverts, as though the borrow never happened. Builders use the tool to rebalance, to liquidate, or to arbitrage inside one transaction. It does not create a loan you can walk away with. If you are not the one writing the calling contract, the feature can stay in the background. Its existence matters mostly because liquidators and arbitrageurs use it to clear unhealthy positions, which is part of how suppliers get paid back when a borrower fails.

The receipt, called an lToken

When you supply, LayerBank mints a receipt. The project calls these lTokens. An lToken is your share of one pool, on one network, for one underlying asset. Supply a dollar coin on Rootstock and you hold a claim on that dollar pool. The same ticker on Robinhood Chain, or on Movement, is a different pool and a different receipt. Sending the receipt to the wrong network, or trusting a token that merely uses a similar name, is a good way to lose the claim.

At the start, the receipt maps closely onto the underlying asset. As borrowers pay interest, the pool’s assets grow relative to the outstanding receipts. Redeeming later is supposed to return more of the underlying asset than you put in, provided the pool is healthy, the interest was real, and liquidity is available for the exit. You do not file a separate claim each day. The exchange value of the receipt drifts upward as the pool earns. That drift is the interest, made visible as accounting.

Accounting is also a risk surface. The receipt is only as good as the pool behind it. If borrowers default and liquidations fail to cover the hole, if an oracle prints a bad price and lets value leave, or if a bug drains the contracts, the receipt can be worth less than the exchange rate suggests. Sometimes it can be worth nothing. An lToken is a claim on a risky shared pool. It is not a warehouse receipt from a regulated custodian, and it is not a deposit at Robinhood or at any other broker.

Receipts are designed to be usable. Inside the protocol they represent the position. Outside it, another application may choose to accept them as collateral or as a building block. Composability is the polite word for that. The plain word is contagion. A problem in the pool can travel into every application that trusted the receipt. Before you deposit an lToken anywhere else, ask what happens to that second application if the first pool is impaired. If the answer is unclear, the extra yield is unclear as well.

Withdrawing is the reverse of supplying: you return the receipt, and the pool sends the underlying asset, up to your share. If utilization is high, the pool may not have the idle balance to pay everyone at once. Your claim still exists. Your ability to leave on the spot does not. This is the trade at the heart of layer bank lending. You earn more when borrowers are eager, and you are most stuck at the same moment.

Collateral, ceilings, and the health factor

A borrow on LayerBank is over-collateralized. You lock value greater than the amount you draw. The protocol does not underwrite your income, your credit, or your intentions. It underwrites assets it believes it can sell, at a price it believes an oracle can print.

Each accepted collateral has a ceiling, the maximum loan-to-value. Post several assets and the ceiling becomes a blend. The borrow limit is the sum of each collateral’s amount, multiplied by its price, multiplied by its own ceiling. Borrowed value is what you owe, marked at current prices, including interest that has already accrued. The health factor is the borrow limit divided by the borrowed value.

Above one, the position sits inside the line the protocol drew. At one, it is on the line. Below one, liquidation is permitted. A position that opens with a comfortable health factor can decay with no new action from you. Interest on the debt does it slowly. A fall in the collateral price does it quickly. A rise in the borrowed asset’s price does it from the other side, because repayment has become more expensive in the unit you are measuring.

Put numbers on it. You post collateral worth 10,000, and the ceiling is 80 percent, so the borrow limit is 8,000. You draw 5,000. The health factor is 8,000 divided by 5,000, or 1.6. Now the collateral price falls by 30 percent. Collateral value is 7,000. The limit is 5,600. The health factor is 5,600 divided by 5,000, or 1.12. You have not touched the position. One more modest drop, or a stretch of unpaid borrow interest, and you are at the line. If you had drawn 7,500 at the start, the same 30 percent drop would already have put you through it.

People get into trouble by treating the maximum as a target. The maximum is the edge of the cliff as the contract defines the cliff. Volatility, oracle delay, and the time it takes to send a repayment all argue for a wider gap. The protocol’s only hard statement is that one is the threshold. How much room to leave above one is a judgment about the asset. A blue-chip dollar token and a thin wrapper do not deserve the same gap. A quiet afternoon health factor can be gone by evening if the collateral trades in a market that gaps.

There is a second constraint that interfaces sometimes bury. Some assets can be supplied and still be weak collateral, or they can count as collateral only inside a particular efficiency-mode category. Enabling collateral is its own step. Supplying alone earns the supply side of the book. It does not, by itself, give borrowing power until the asset is counted and the market allows it. Borrowers who believe they have posted margin, and have only supplied, discover the difference when the borrow button refuses them, or when a later change in settings alters what counts.

Interest also changes the health factor while you sleep. A borrow rate that looks small in annual terms is a steady leak. On a levered loop, where the debt is a large fraction of the collateral, that leak is the thing you are betting you can out-earn. If utilization jumps and the borrow rate jumps with it, the leak becomes the whole story.

What liquidation does to a position

Liquidation is the protocol defending the suppliers. When health falls below one, a third party is allowed to repay part of the debt and take a slice of collateral at a discount. The discount is the fee for keeping the pool whole. Without that fee, nobody would spend gas to save other people’s deposits, and bad debt would sit in the pool until suppliers absorbed it.

On the ordinary path described for these markets, the repayment is partial. A liquidator clears enough debt to restore a safer health factor, often up to half the debt in a single action, and the remaining collateral stays with the borrower. The bonus varies by market. Correlated books that run in efficiency mode have used a smaller penalty, on the order of a few percent, because those markets assume the two prices will not diverge wildly. The baseline description of liquidation has used a larger incentive, on the order of fifteen percent, where the assets are less tightly bound. The figure that governs your position is the one configured for that market on that chain. A number remembered from a different book is a rumor.

A liquidation is a loss for the borrower. Collateral leaves at a discount, and the position shrinks at a moment the borrower did not choose. It is also the reason a supplier can treat the pool as something other than an honor system. If liquidators do not arrive, because the bonus is too small, the chain is congested, the collateral cannot be sold, or the oracle has frozen, bad debt remains. Suppliers bear that bad debt. The health factor protects deposits only when the liquidation machinery actually clears.

After a partial liquidation you may still have a position. It will be smaller. If you were running a leveraged loop, the liquidation can unwind the loop at the worst price and leave you with less of the asset you meant to accumulate. The bonus paid to the liquidator is value that used to be yours. There is no appeal desk. The transaction is the decision.

Borrowers who want to avoid that outcome repay early, add collateral, or borrow less at the start. They also watch the oracle. A lending market can liquidate you against a price that the broader market does not share, if the feed is stale, manipulated, or pointed at a thin pool. Equally, a feed that fails to update can protect an unhealthy position until it updates all at once. Both failures have emptied money markets elsewhere. LayerBank uses the same family of machinery. The family history is part of the risk.

Why the interest rate moves

LayerBank sets rates from utilization. Utilization is the share of a pool that is currently borrowed. When utilization is low, borrowing is cheap and supplying pays little, because cash in the pool is not scarce. When utilization climbs, both rates rise. Borrowers are nudged to repay. Suppliers are paid to deposit or to stay. The pool is trying to remain liquid enough that withdrawals can clear, without sitting so idle that suppliers earn dust.

The shape of that curve is a policy choice encoded in the interest model: a slope while the pool is comfortable, and a steeper slope once utilization passes the region the designers wanted to defend. The exact kink differs by asset. A stablecoin pool can tolerate higher utilization than a pool of a volatile token, because withdrawals and liquidations behave differently. You do not need the formula memorized. You need the live utilization, the live borrow rate, and the live supply rate, read from the market you are using.

An illustration, labeled as arithmetic and not as LayerBank’s live curve: a pool with 70 percent of deposits borrowed, a borrow rate of 8 percent, and a reserve that keeps 10 percent of borrower interest, pays suppliers about 70 percent times 8 percent times 90 percent, a little over 5 percent. Change any input and the output moves. If a campaign adds an extra reward, that reward sits on top of the 5 percent until the campaign stops. Someone quoting you a single blended number has mixed interest with a subsidy. Ask for the split.

Incentives can be paid in ULAB, once those emissions are live on a market, or in some other asset a campaign chooses. On Robinhood Chain, campaigns have used USDG, the dollar token associated with that chain, as a reward for supplying and, on some markets, for borrowing. A campaign is a decision to rent liquidity for a while. When the budget ends, the headline yield falls back to whatever borrowers are truly paying. Anyone who sized a position for the subsidized number is then holding the unsubsidized one, plus whatever price risk came along for the ride.

Some collateral brings its own yield: a liquid staking receipt for ether, a treasury-linked token, a fund token that accrues income elsewhere. That underlying yield is a third stream, separate from the supply rate and separate from the campaign. Stacking them is the point of several strategies. Double-counting them is the point of several disappointments. If you borrow against a yield-bearing asset, you keep the underlying yield only if the token actually passes it through, and you pay the borrow rate on what you drew. Net yield is the difference, after fees, after any incentive that will expire, and after the chance that a depeg or a liquidation erases the spread you were farming.

Net yield can be negative. Nothing in the contract promises that the supply side, the underlying yield, and the subsidy will exceed the borrow cost. A loop that worked when utilization was calm can bleed when a large borrower arrives or a large supplier leaves. The rate is a live price of money. It is allowed to change against you.

Efficiency mode and looped positions

Efficiency mode is a risk setting for assets that should move together. Dollar tokens against other dollar tokens. Wrapped bitcoin against other bitcoin-linked tokens. Staked ether against ether. Because a wide price gap is less expected, the protocol can allow a higher loan-to-value, and a tighter band before liquidation, than it would allow for unrelated assets. The safety is conditional on the correlation holding. When a wrapper depegs, the assumption fails at the worst time, and the higher ceiling becomes a larger hole.

Published configurations give a sense of the dial. On Rootstock, efficiency-mode books for bitcoin-linked collateral have sat around a loan-to-value in the mid-eighties, with a liquidation threshold a little higher and a penalty around four percent. Stablecoin books there have been set even higher, with ceilings in the high eighties. On Movement, ether-linked assets have used a ceiling around eighty percent, while stablecoin books, including yield-bearing dollar tokens, have been set in the low nineties, with a larger penalty than the Rootstock books. Plume, a network aimed at real-world assets, has carried its own pairings, with ceilings that differ sharply by asset. These are examples of how a category gets parameterized. They are checked on the live market, because an admin or a later governance action can move a number.

The reason anyone accepts the tighter rope is looping. Supply a yield-bearing asset. Borrow the underlying, or a close relative, inside efficiency mode. Convert the borrowed amount back into more of the yield-bearing asset. Supply that too. Each turn adds exposure to the spread between what the collateral earns and what the debt costs. Each turn also multiplies the loss if the spread flips, if the peg slips, or if the liquidity to unwind disappears.

One turn, in words. You hold 10,000 of a token that yields about 5 percent from its own design. In efficiency mode you borrow 7,000 of a related asset at a borrow rate of 3 percent, and you convert that 7,000 into more of the yielding token. On the borrowed portion you are earning the yield and paying the borrow rate. The spread is the point of the trade. You have also raised the chance that a 10 percent gap in the peg, or a spike in the borrow rate, consumes months of spread in a day. A second turn repeats the bet. Automated vaults exist to click these turns for you. A vault is a convenience and a pile of trust: its accounting, its unwind path, the oracles, and the same collateral rules as a manual loop. Leverage does not become gentler because a contract presses the buttons.

LayerBank has described automated leverage vaults as part of the later protocol, and outside reviewers have looked at vault code as its own scope. A review of a vault does not review every asset you can loop, and a review of the base market does not review the vault. If you deposit into a vault, you have added a layer. Know which layer holds the assets, which layer can be upgraded, and how you exit if the vault’s share price and the underlying pool disagree.

A looped rate is sometimes quoted as a multiplied yield. Read that quote as a label on a lever. The result rises with the collateral’s own yield and with any supply rate, and it falls with the borrow rate, scaled by how many times the position has been geared. If you cannot write down the exit as a sequence of repayments and redemptions, the multiplied number is advertising.

LayerBank staking, in the three senses people use

Search for LayerBank staking and you will find a loose word. LayerBank is a lending market. The gesture that feels like staking, for a typical user, is supplying an asset and holding the lToken while interest accrues. Assets leave the wallet, a receipt remains, and a yield is supposed to arrive. People reuse the word because the gesture rhymes with staking on a proof-of-stake network. The machinery underneath is lending.

There is a second, stricter sense. Proof-of-stake networks pay validators, and liquid staking tokens pass some of that payment to holders. Ether has a family of these receipts. If you supply one of them on LayerBank, you may be stacking two yields: the staking yield that belongs to the liquid staking token, and the lending yield that belongs to borrowers on LayerBank. LayerBank did not invent the staking yield. It is a venue where the receipt can also serve as collateral, which is what lets a loop exist. The risks add. A slashing event, a withdrawal queue, or a depeg in the liquid staking token can hit you at the same time as a liquidation on the lending market.

A third sense belongs to ULAB. The design includes pairing the token into a liquidity position and locking that position to mint xULAB, which is meant to increase reward weight and, in time, to act as the base for gauges and votes. That lock is closer to a vote-escrow than to staking ether. It ties up a liquidity position. It exposes you to the price of ULAB and to the other asset in the pair. It should be judged as a decision about the protocol token, separate from the question of whether a dollar pool is a sound place to lend.

When a page promises LayerBank staking, ask which of the three it means. Supply yield. A staking token used inside the money market. Or a lockup of ULAB liquidity. The exit is different in each case. Supply yield leaves when you redeem the lToken, subject to utilization. A liquid staking token may have its own unbonding. An xULAB lock lasts as long as the lock says it lasts. Mixing the three into one annual percentage is how a quiet lending rate gets dressed up as something else.

Robinhood staking is a separate counter

Robinhood staking is a product inside Robinhood Crypto. Where the program is offered, the app lets eligible customers stake ether, solana, and cardano and receive network rewards on a schedule the app controls. Availability depends on the customer’s jurisdiction. The assets sit inside the Robinhood Crypto account relationship, under the disclosures Robinhood publishes for that program. Rewards can change. Staking can involve a wait to unstake. Some networks can penalize a validator, and that penalty can reach stakers. The asset’s price can fall by more than any reward pays. Robinhood’s own notices also make clear that crypto held through Robinhood Crypto is outside FDIC insurance and SIPC protection.

LayerBank does not operate that program. A deposit into a LayerBank pool is a loan to a smart-contract market. The wallet signs it. The health factor governs it. Liquidators can seize collateral when the numbers cross the line. None of that is the button that stakes ether inside the Robinhood app. The shared word “staking” is the trap. So is the shared word “Robinhood,” once Robinhood Chain enters the picture.

Robinhood Chain is a permissionless Ethereum layer-2, built on Arbitrum technology, with ether as the gas token. Its chain id is 4663. Robinhood describes the chain as infrastructure for on-chain markets, with a particular interest in tokenized real-world assets, and as a network where applications can deploy without locking users into a single platform. Because the chain is permissionless, a lending protocol can launch markets there without becoming the brokerage and without replacing the in-app staking product. LayerBank has done that. The deployment puts layer bank Robinhood searches onto a real object: lending pools on that chain. It does not merge LayerBank into Robinhood staking.

If the goal is the in-app stake of ether, solana, or cardano, the path runs through Robinhood’s own screens and Robinhood’s own disclosures. If the goal is to supply or borrow on LayerBank, the path runs through a wallet on the network you select, and every risk later in this article applies. Using one does not open an account in the other. Rewards paid by a proof-of-stake network through a broker are a different cash flow from interest paid by borrowers through an lToken.

Robinhood lending, and layer bank Robinhood

Robinhood lending, as a phrase people type, points down two halls.

The first hall is the broker. Brokerage customers of Robinhood can encounter securities-account features that use the word lending: a program that lends out fully paid shares, or margin borrowing under a margin agreement. Those features live inside a customer agreement. They come with the regulatory status of a broker-dealer relationship. They do not travel with a self-custodied wallet, and they do not follow an asset onto a layer-2. LayerBank is not that desk. A tokenized ticker inside a LayerBank market is not a share enrolled in a brokerage stock-lending program, even when the ticker letters match.

The second hall is on-chain lending on Robinhood Chain. This is where LayerBank enters the Robinhood story. The protocol lists markets on the chain. Markets the project has promoted there have included dollar liquidity in USDG, the Paxos dollar token used on the network, wrapped ether, bitcoin-linked tokens such as cbBTC, and tokenized exposures that follow familiar market names: broad equity funds, single-name stocks, gold, and short-term Treasury products. Names that have appeared in those promotions include SPY, QQQ, NVDA, GLD, and SGOV. Incentive campaigns have paid extra USDG to suppliers of several of those markets, and in some cases to borrowers of USDG. A campaign is yield with an end date, even when the end date is not printed on the button.

A tokenized ticker is a crypto asset whose price is supposed to follow something else, through an issuer, a custodian, a creation and redemption process, and an oracle. Loan-to-value on that collateral is a bet on the tracker and on the oracle together. If redemptions pause, or if the token trades wide of the thing it tracks, the lending market will believe the feed it was given. Borrowers get liquidated against that belief. Suppliers earn interest that may be coming from borrowers who are about to be liquidated into a weak bid. The letters on the token are not the prospectus, and they are not the customer agreement of the brokerage product with the same letters.

Market trackers have placed Robinhood Chain among LayerBank’s larger deployments, with Rootstock ahead of it and a long tail of thinner markets behind. Size is liquidity, which helps you enter and sometimes helps you leave. Size is also a larger pile of consequences when a shared oracle or a popular collateral misbehaves. Robinhood’s own public notes on the chain’s ecosystem have named Morpho as a lending application alongside other infrastructure partners. LayerBank’s pools are an additional deployment on an open network. Read them as a deployment by LayerBank. An open chain allows more than one money market. The presence of a second protocol is a fact about the chain, and it is a reason to confirm which contracts your wallet is about to call.

For a reader who arrived by typing layer bank Robinhood, the practical summary is short. You can lend and borrow on LayerBank’s Robinhood Chain markets by using a wallet on chain id 4663, if those markets are still listed and if you accept smart-contract risk. You stake ether, solana, or cardano inside Robinhood by using Robinhood’s staking program, where you are eligible. You lend securities or borrow on margin inside a brokerage account by using the brokerage product, under its agreement. Three counters. Three sets of rules. The keywords overlap because the brand and the chain overlap. The balances do not.

The map of networks

LayerBank presents itself as a money market across many chains, on the order of seventeen or more, mixing Ethereum-style networks and at least one Move-based network. Names that have appeared across deployments and dashboards include Linea, Scroll, Manta, Taiko, Mode, Morph, zkLink Nova, BOB, Bitlayer, BSquared, Hemi, Nibiru, Mint, BNB Chain, Plume, Movement, Rootstock, and Robinhood Chain. A name on a list means the contracts were pointed at that network at some point. It does not mean the pool is deep today. Several deployments have held only a thin film of assets. Activity has clustered, and the cluster can move.

Rootstock is the Bitcoin-facing book. Collateral there is bitcoin in several wrappers, including RBTC and other bitcoin-linked tokens used in efficiency mode, plus dollar stablecoins. The point of the efficiency-mode categories is to let those correlated bitcoin assets borrow against each other more tightly than a generic risk model would allow. For a reader who wants layer bank lending with bitcoin as the economic exposure, Rootstock is the deployment built for that job. Wrapper risk stays in the room. A token that claims to be bitcoin depends on the bridge, the federation, or the issuer that stands behind the claim. If the wrapper breaks, the lending market is holding the wrapper, not the coin.

Movement is the Move-language deployment. Collateral sets described there have included ether and liquid staking tokens such as rsETH and ezETH, and stablecoins that include yield-bearing dollar tokens. Move contracts are a different codebase from the Ethereum-style deployments. A review of one does not cover the other. LayerBank commissioned separate review work for the Move implementation, which is the right shape of caution and also a reminder that “LayerBank” is a family of codebases wearing one name.

Plume is aimed at real-world assets, and LayerBank’s configuration there has included pairings against its dollar asset, with ceilings that run from conservative to very high depending on the collateral. The point of that network, for this protocol, is collateral that is supposed to carry off-chain or structured income. The point for a lender is to ask who issues the token and what stops redemption.

The long tail matters in two directions. A messaging layer or a canonical token that touches many chains can create shared fate: a failure far from the pool you use can still change what your receipt is worth. And a tiny pool can show a wild rate because a single wallet arrived. Wild rates on empty pools are advertisements until you know who will be on the other side when you try to leave. Depth, borrower concentration, and oracle quality beat a long list of logos.

Bridging into any of these networks adds a step the lending market does not control. The bridge can pause. The bridged asset can depeg from the original. Gas on the destination chain is a separate token, ether in the case of Robinhood Chain, and a position you cannot afford to manage is a position you cannot defend. Arrive with the gas token already in the wallet, and with a clear idea of which bridge you trust. The lending rate assumes you can act. If you cannot act, the rate is a number on a screen.

Yield that starts off the chain

LayerBank has made real-world yield part of the design. Stripped of slogans, the mechanism is plain. An asset that already earns income off a chain, or inside another protocol, is accepted as a deposit or as collateral. Holders keep whatever yield the asset itself pays, if the token is built to pass that yield through. They may also earn or pay the lending rate, depending on which side of the book they occupy.

This is attractive when the upstream yield is steady and the borrow cost sits below it. It is how a Treasury-linked token or a basis token ends up inside a loop, and it is part of why Robinhood Chain, with its emphasis on tokenized funds, equities, gold, and short-term government paper, is a natural place for a lending market to list those assets. The steady part deserves a hard look. Off-chain income depends on an issuer, a legal structure, banking rails, and sometimes a manager. Tokenization does not remove those parties. It gives them a contract address. If redemptions pause, the on-chain price can gap, and a lending market will liquidate borrowers against the gap.

A short-term government bill is a quiet instrument. A token of a short-term government bill, posted as collateral on a new chain, with an oracle and a loan-to-value, is a quiet instrument plus every additional layer. Equity tokens add hours, halts, corporate actions, and the question of what the token holds when the cash market is closed and the chain is open. Gold tokens add custody. Dollar tokens add the reserves and the redemption window of the issuer. USDG, for example, is a dollar token whose credibility is the credibility of its issuer and its reserves, plus the credibility of the market that trades it. Lending USDG on LayerBank stacks the issuer’s risk on the protocol’s risk.

Use these markets with the layers named. What does the token redeem for. Who can pause redemption. Which oracle prints the price, and how stale may it be. How fast does the loan-to-value assume the price can fall. If the answers are missing from the interface, they are still part of the position. Absence from the screen is not absence from the loss.

ULAB, the token that coordinates incentives

ULAB is the protocol token, the Universal LayerBank Token. The stated supply is one hundred million. The published split puts the largest share into incentives, and the rest into the team, investors, airdrops, market liquidity, a small public sale, and a strategic reserve. The weights are the project’s statement of who is paid to build, who is paid to show up with liquidity, and who can sell.

A large incentive budget can support supply and borrow rewards for years. That is why a future ULAB reward sometimes appears in yield discussions even when the live rate on a pool is still just the interest borrowers pay. A large incentive budget is also future sell pressure. Vesting slows it. Vesting does not delete it. Team and investor cliffs tell you when new supply can reach a market. They do not tell you the price at which that supply will clear.

The token is meant to do three jobs. It pays suppliers, borrowers, and liquidity providers. It can be placed into a liquidity pair and locked to mint xULAB, a boosted claim that increases reward weight. And xULAB is intended as the base for emission gauges and protocol governance. The governance job has been described as work still underway. Until votes actually steer risk parameters, treat governance as a plan. Parameters on a live market may sit with admin roles. Admin roles are a risk when the people holding them are competent, and a worse risk when they are not, or when the key is stolen.

ULAB has been described as an ERC-20 on Ethereum-style chains, a fungible asset on Move, and an omnichain token so a single supply can move across networks. Dashboards sometimes label the protocol LAB. The token the project describes is ULAB. A ticker on a dashboard is not a contract address. Confirm the contract in an interface you trust before you buy, bridge, or lock anything. Counterfeit tickers are cheap to deploy, and a lending brand is an attractive thing to counterfeit.

Holding ULAB is a bet on the protocol’s future fees, emissions, and governance design. It is not a claim on the assets inside the lending pools. Suppliers have that claim, through lTokens. If LayerBank grows, ULAB might capture value, if the design routes value to it. If LayerBank shrinks, the token can fall independently of whether your lent dollars are still in the pool. Keep the two wallets, mental and literal, apart.

Security work, and the limit of a review

LayerBank has put core contracts in front of outside reviewers. Early lending markets, including the Linea-era book, were reviewed by PeckShield. A later multi-chain version of the lending market was reviewed by PeckShield again. The Move implementation was reviewed by MoveBit, and that engagement recorded findings for the team to address. A finding is the purpose of a review. Leverage-vault work has been reviewed by AstraSec, as its own scope. Each of these is evidence of effort, tied to a codebase and a period of time. None of them is a warranty, and none of them travels automatically to a new chain, a new collateral, or a new vault.

Read a review for scope before you read it for comfort. A report on vault code says little about a fresh collateral listing. A report on the Move contracts says little about an Ethereum-style pool on Robinhood Chain. Bugs in money markets collect in a few places: the math of interest indexes, the oracle and its stale-price rules, the liquidation bonus, a token that charges a fee or rebases, and the admin function that can change a parameter faster than users can exit. Independent review reduces how many of those survive the first pass. It does not remove the category. It does not watch the oracle next month.

Custody, in the crypto sense, stays with you. A site that wears the LayerBank name can ask for a token approval and then empty everything the approval covers. The protocol cannot unwind a signature you did not understand. Approvals should match the amount you mean to supply. Old approvals should be cleared when you are finished with a market. Bookmarks rot, and search ads are a common place for a counterfeit front end to sit above the real one.

Upgradability deserves a straight sentence. Many lending protocols can replace contract logic. That ability patches bugs. It also lets an admin, or an attacker who becomes the admin, change the rules under an open position. Before you supply size, know whether the deployment you are using can be upgraded, who holds the key, and whether any delay gives you time to leave. A multisig is better than a single key. A delay is better than an instant upgrade. Neither is the same thing as a contract that cannot be changed.

How to read a market before you touch it

The interface wants a decision. A serious reading of a LayerBank market is a short list, taken in order, on the specific pool.

Start with the asset. Identify the contract, the issuer, and the strange behaviors: pause switches, rebasing balances, transfer fees, or a blacklist. Lending pools assume a boring token. A clever token can desynchronize the pool’s internal accounting from the real balance. If you cannot explain how the token moves, you cannot explain your receipt.

Then the oracle. Whose price, how old it may be, and what the market does if the feed stops. Money markets lose funds through prices as often as they lose them through outright theft. A tokenized equity that prints a price while the reference market is closed is a particular version of this question. So is a bitcoin wrapper priced off a single pool.

Then utilization. The supply rate is a symptom. High utilization pays suppliers and traps them. Low utilization pays little and usually lets you leave. A rate without utilization is a billboard.

Then the ceiling. Maximum loan-to-value, liquidation threshold, and the penalty. The gap between the ceiling and the threshold is the last strip of ground before someone else can take collateral. Efficiency mode changes all three. Confirm which mode your account is in. A wallet that opted into a correlated category is living under that category’s rules, including when the correlation breaks.

Then the size, and the concentration inside the size. A pool of pocket change can print an attractive rate that one wallet erases. A deep pool can still be one borrower. If you are the marginal supplier to a single leveraged wallet, you are financing that wallet. Ask whether the collateral would clear if that wallet were liquidated into the real market, not into the oracle’s last print.

Then the subsidy. Any reward beyond borrower interest is a campaign. Note the asset paid, the payer, and the end. Subtract it before you compare LayerBank with another protocol, and before you compare a Robinhood Chain market with a Rootstock market. Campaigns differ by chain on purpose. They are trying to pull deposits toward a launch.

Then the chain itself. Fees, the bridge you used, and which codebase you are touching. A calm experience on one network is not a certificate for the next. Robinhood Chain uses ether for gas and chain id 4663. A wallet pointed at a different id is on a different network, whatever the token names say.

Last, the exit. If you are looping, write the unwind as transactions: repay, withdraw, redeem, bridge. Assume you will do it while prices move and gas is expensive. If you are only supplying, write the condition under which utilization would block you, and decide whether you can live with the wait. A position you cannot describe leaving is a position the market will describe for you, through liquidation or through a long queue.

The risks that stay on the table

Smart-contract risk. A bug in the pool, the token, the oracle adapter, or a vault can take the assets. Reviews lower the odds inside their scope. They do not close the risk.

Oracle risk. A wrong price causes wrongful liquidations, or it lets a borrower extract value against collateral that is not really there. A silent feed delays the correction until it arrives as a shock.

Market risk. Collateral falls, the borrowed asset rises, health collapses. Leverage shortens the clock. Efficiency mode shortens it further, in exchange for a higher ceiling, and the exchange is only wise while prices move together.

Liquidity risk. You cannot withdraw what has been borrowed until it comes back. The best supply rates and the worst exits are neighbors.

Depeg and wrapper risk. Stablecoins, bitcoin wrappers, liquid staking tokens, and tokenized securities can all trade away from the asset they name. The pool holds the token it was given.

Issuer risk. Real-world tokens add lawyers, banks, custodians, and pause buttons. A lending protocol cannot redeem a token the issuer will not redeem.

Admin and governance risk. Keys can change parameters, list strange collateral, or upgrade logic. Governance that is still being built does not yet constrain those keys.

Bridge and chain risk. A failure of the network, or of the messaging that moves canonical tokens, can freeze the practical value of a position even when the lending math is fine. Congestion can also freeze your attempt to repay, which is enough to let a liquidation land.

Incentive risk. Yield that depends on a token emission depends on that token’s price and on the campaign continuing. ULAB rewards, USDG campaigns, and any other subsidy belong in this bucket until they are interest paid by a borrower.

Legal and access risk. Permissionless software is used by people who live under laws. What you may do with a brokerage product, including Robinhood staking or a securities-lending program, and what you may do with a self-custodied wallet on Robinhood Chain, are different questions. This article does not answer them for your jurisdiction. It also does not know your tax treatment of interest, rewards, or liquidations.

Total loss is possible. A higher yield is the market’s way of saying that someone is paying you to hold a risk. Sometimes that someone is a borrower with a legible purpose: they want leverage, they want a short, they want to keep a staking receipt while spending a dollar. Sometimes that someone is an incentive budget trying to attract you before the terms change. Both can be rational. Both can end with the supplier holding the loss. Layer bank lending does not repeal that arithmetic. It publishes it as a rate.

Questions readers ask

What is LayerBank?

LayerBank is a decentralized money market. Suppliers lend assets into pools and earn interest. Borrowers draw against collateral. The protocol runs on many networks under the name LayerBank Finance, and it grew out of the earlier LineaBank deployment on Linea.

What is layer bank lending?

Layer bank lending is the supply and borrow business of that protocol. You lend by supplying an asset to a pool. You borrow by posting collateral up to a stated loan-to-value. Rates move with utilization. Receipts called lTokens track the supplier’s share of the pool, and the value of those receipts is supposed to rise as interest is paid.

What does LayerBank staking mean?

In common use, LayerBank staking means supplying assets and earning the supply yield, or supplying a liquid staking token such as a staked-ether receipt. It can also refer to locking ULAB liquidity to mint xULAB. It does not mean Robinhood’s in-app staking program for ether, solana, and cardano.

Is LayerBank the same thing as Robinhood staking?

Robinhood staking is Robinhood Crypto’s in-app program for ether, solana, and cardano, where that program is offered. LayerBank is a separate lending protocol. The connection is Robinhood Chain, an open Ethereum layer-2 where LayerBank runs markets. A LayerBank deposit is a smart-contract loan book. A Robinhood stake is a brokerage-crypto product, with the disclosures and the account relationship that product uses.

What is Robinhood lending on LayerBank?

On LayerBank, Robinhood lending means the lending markets deployed on Robinhood Chain. That covers supplying and borrowing listed assets, including dollar tokens such as USDG, ether, bitcoin-linked tokens, and tokenized market assets when those markets are open. Brokerage stock lending and margin inside a Robinhood securities account are separate products, with separate agreements. A tokenized ticker is not the brokerage position of the same name.

Which chains does LayerBank use?

Deployments have included Linea, Scroll, Manta, Taiko, Mode, Morph, zkLink Nova, BOB, Bitlayer, BSquared, Hemi, Nibiru, Mint, BNB Chain, Plume, Movement, Rootstock, and Robinhood Chain, among others. Tracked liquidity has concentrated on Rootstock and Robinhood Chain. The list is a map of where contracts have lived. The pool you enter is the only market that can pay you or trap you.

What is ULAB?

ULAB is LayerBank’s protocol token, with a stated supply of 100 million. It is used for incentives, for liquidity, for locking into xULAB, and for a governance design that has been in progress. ULAB is not a receipt for a deposit. The lToken is the receipt. Dashboards sometimes label the protocol LAB. Confirm the contract before you treat a ticker as the token.

Can a LayerBank position be liquidated?

A position can be liquidated when its health factor falls below one. A liquidator may repay part of the debt and take collateral at a bonus. The bonus and the share of debt that can be cleared in one action depend on the market. Correlated books have used smaller penalties. A partial liquidation is still a loss, and it can unwind a loop you meant to keep.

Do audits make LayerBank safe?

Outside reviews by firms including PeckShield, MoveBit, and AstraSec cover specified code at a specified time. They lower the odds of known classes of bugs inside that scope. They leave oracle risk, market risk, admin risk, and every contract outside the scope. Total loss remains possible on a reviewed market.

Where should a new reader start?

Start with one pool, on one chain, in a size you can afford to wait on. Read utilization, the loan-to-value, the oracle, and the split between borrower interest and any campaign reward. If the chain is Robinhood Chain, keep Robinhood staking and the brokerage lending programs in their own categories. If the asset is a wrapper, a liquid staking token, or a tokenized ticker, name the issuer before you name the yield. The protocol is understandable. The position is safe only to the extent those checks come out boring.