Jump Crypto: A Comprehensive Analysis of On-Chain Credit, Clarifying Its Value, Use Cases, and Future

Author: Lucas Baker, Jump Crypto

Compiled by: Amber, Foresight News

Since the mid-2020 explosion of DeFi activity and innovation, decentralized lending has become a core pillar of the emerging on-chain financial system. Driven by protocols like Aave and Compound, borrowers can deposit one token as collateral and extract part of its value by borrowing another token, creating an almost frictionless token exchange market and highly flexible trading exposure. However, because most current lending models require over-collateralized debt positions, on-chain borrowers often cannot truly access credit—i.e., unsecured general loans.

While the status quo is very effective for the simple logic of tokenized asset trading, we believe that the long-term potential of DeFi lies in establishing a truly independent on-chain financial system. The infrastructure of this "parallel" system must meet all traditional financial needs, and for this, we view unsecured lending models as an indispensable growth driver. In this article, we will first assess the limitations of over-collateralized loans, and then we will explore two main alternative models for unsecured lending: decentralized prime brokerage and on-chain identity.

Since on-chain entities are untrusted by default, each lending model adopts a different approach to rebuilding trust while reducing credit risk. In the prime brokerage model, the protocol provides credit through contracts or proxy wallets, retaining ultimate control over the use of funds. In the identity-based model, the protocol establishes some mapping from on-chain borrowers to different real-world entities before issuing loans. Both systems exist today, but adoption has been limited compared to the rapid growth of over-collateralized loans. We believe that as the market matures, unsecured loans will be increasingly adopted, and these two approaches represent complementary visions that can together accelerate the growth of on-chain credit.

Credit as the Engine of Growth

Before we delve into alternative methods for credit expansion, we need to define what "credit" means, its forms, and why it is important. In practice, "credit" in DeFi often means something entirely different from what it means in traditional finance.

What is Credit?

In the real world, when we talk about "credit," we usually refer to the ability to obtain money that does not belong to you. In addition to interest rates, two main variables determine the nature of a loan: what the borrowed funds can be used for and how the loan is secured. In the case of a mortgage, a homebuyer might provide 20% of the purchase price as a down payment and obtain the remaining 80% from the bank, resulting in a debt-to-equity ratio of 4:1. This loan is secured or collateralized by a claim on the house itself. If the buyer stops making payments, the bank will repossess the house to recover the loan value. Other loans are not secured by direct collateral but are secured by a priority claim on assets or future income. Credit card purchases and corporate bonds differ significantly in terms and scale, but in both cases, lenders can use the borrower's other assets to partially offset the lack of collateral. In the event of non-payment, outstanding claims will ultimately be resolved through collection (individual) or bankruptcy (corporate) processes.

Direct and indirect secured loans are crucial for the functioning of modern economies because they provide the capital that drives consumption and investment. Given that each loan occupies a point on a spectrum from over-collateralized to unsecured and another point from specific use to general use, it is clear that traditional finance supports all quadrants. (In most cases, the primary variable is collateral, with specific-use over-collateralized loans and general-use unsecured loans, but we will use two axes for greater precision.)

In the world of DeFi, the "credit" offered by decentralized lending protocols often means something entirely different: the primary use case is not to obtain more funds but to access different asset combinations. Common applications include arbitrage trading, borrowing at higher rates and lending at lower rates, and changing portfolio composition without triggering taxable events. Of course, these are also possible in traditional finance, but they are not the typical goals of borrowers.

We all know that DeFi can handle over-collateralized loans through protocols like Aave. What about the other quadrants? It turns out we can go further in expanding trading risk exposure. Various protocols, including margin trading, perpetual contracts (dYdX), leveraged tokens (Tokensets), and exotic options (Opyn), enable DeFi users to trade with leverage and access richer risk-return options. In the future, we may also use over-collateralized loan protocols to simulate leveraged asset purchases like mortgages. However, none of these methods ultimately provide a general, unsecured credit channel. More specifically: if I obtain a personal loan of $50,000 from the bank with only $10,000 in my account, I can use it to buy stocks, purchase a car, or stash it under my mattress, and the lender cannot intervene at all. But if we only use existing over-collateralized loan protocols, it is difficult to "freely dispose" of the borrowed funds.

Making Unsecured Loans Over-Collateralized Again

Before seeking alternatives, let’s first think about whether there are possibilities to solve the problem. The fundamental invariance of any over-collateralized protocol is that if you deposit X, you cannot withdraw more than X. However, there are two ways to turn X into a position worth more than X: recursive deposits and flash loans.

The conceptually simpler method is recursive deposits, or reusing debt as additional collateral. It turns out that as long as we can swap borrowed tokens for collateral tokens, we can achieve leverage this way. If possible, we can simply deposit collateral in one token, borrow another token, swap the debt for collateral tokens, deposit again, and repeat. (Note that this is similar to "re-staking," where the same collateral is staked for two or more unlimited loans, but this could lead to situations where two or more lenders attempt to seize the same collateral. For recursive deposits, this cannot happen even if the entire position is liquidated.)

Assuming the interest rate paid in USDC is higher than that in USDT, and yield traders want to capture the spread by offsetting a long position in USDC with a short position in USDT. If Aave accepts a maximum collateralization rate of 80% for USDC, the trader can combine Aave with Uniswap to form a "leverage loop":

The resulting position will have collateral in USDC offset by debt in USDT, allowing our trader to collect the spread on the leveraged position. How much leverage can they achieve? Ignoring fees and slippage, the collateral ratio r can support leverage of up to 1/(1 - r). For example, with an initial equity of $1 million and a collateral ratio of 80%, our trader can achieve a long exposure of $5 million—that is, Aave will show $5 million borrowed against $4 million borrowed.

While the recursive deposit method has some appeal, transaction costs and slippage limit its practicality in practice. In contrast, while flash loans must be repaid in a single transaction, they can provide any amount of temporary funds, offering a more elegant solution. We can achieve the same leverage as above by borrowing Cr /(1 - r) in a flash loan to an address holding C, depositing all funds as collateral, and borrowing Cr / (1 - r) to repay the flash loan. Then, our debt position is Cr /(1 - r), and collateral is C /(1 - r), with a leverage ratio of 1/(1 - r). Similarly, this method creates a long position of $5 million and a short position of $4 million using just $1 million of initial net worth.

In fact, flash loans not only enhance capital efficiency but fundamentally expand the design space of over-collateralized loans. Specifically, they allow us to purchase not only tokens but also indivisible assets represented by NFTs. Anything that can be tokenized and used as collateral, whether it's a Bored Ape or tokenized real estate, can be "collateralized" in this way. Indeed, if legal and enforceable, anonymous borrowers could use this method to purchase real estate on-chain, which cannot be achieved through recursive deposits, as there is no way to purchase 20% of an NFT, but flash loans allow us to complete the entire process in one step.

Limitations of Over-Collateralized Loans

Despite the clever design of flash loans, there are two issues that severely limit their practicality as a form of credit:

• Cannot be used. Generally, the purpose of obtaining a home mortgage is to live in the house while paying off the mortgage. In the example above, the collateral is trapped in the Aave contract, making it unusable elsewhere. For a Bored Ape, this problem is even worse, as the "owner" will not be able to sign transactions, preventing the Ape from being used to attend yacht parties that require verification of NFT holdings.
• Risk of losing it. If the floor price of the Bored Ape drops significantly, the Ape used as collateral may be liquidated. It is well known that the volatility of NFT floor prices is substantial, so achieving this outcome would pose significant risks. This is not an issue with traditional mortgages—if the mortgage itself remains in good standing, the bank cannot repossess your house when its value declines.

Furthermore, while recursive deposits and flash loans are excellent solutions for margin trading, they are much less effective for asset purchases. In trading cases, borrowers are typically indifferent to risk exposure and ownership; ignoring taxes, index funds or total return swaps are as good as the underlying asset itself. For purchases like homes or NFTs, borrowers are very concerned about complete ownership and usage rights of the asset itself, rather than the flow of funds. Some benefits of re-establishing ownership are possible—imagine Aave having a "restricted transfer wallet" to hold collateral, essentially a MetaMask with a minimum value invariant—but even then, full ownership cannot be achieved.

When assessing volatility, risk exposure versus ownership is also a key issue. For traders, volatility is an inevitable component of a healthy, well-functioning market, and liquidation risk is acceptable and expected. For buyers, the prospect of "liquidation" in the sense of losing ownership is often unacceptable, regardless of expected financial returns.

Why Are Unsecured Loans Important?

Given the issues above, we might ask why "mortgages" (specific-use loans secured by purchased assets) belong on-chain at all, let alone general credit. Why not continue underwriting in the real world?

We believe that if DeFi's true potential lies in establishing independent financial infrastructure, it must ultimately replicate all the major functions of that infrastructure. Among these functions, general credit can be said to be the closest to the reason for the existence of modern finance: adjusting the supply and demand for capital in form and function. In other words, the existence of a financial system is to drive investment and growth by creating liquidity.

From the borrower's perspective, over-collateralization directly leads to cash flow being tied up, which is clearly detrimental to liquidity. That is to say, for every deposit X locked in a platform, at most Xr can be borrowed, where r is the maximum collateralization ratio. To create growth in a decentralized economy, we must find ways to make r greater than 1: that is, to enable borrowers to access more money than they currently have.

Crucially, what drives economic growth is liquidity, not just leverage. Consider a renewable energy company whose products can significantly increase the amount of available energy globally (which would also increase global GDP), but which requires substantial upfront investment. Such a company might take 20 years to generate the necessary funds solely from internal profits, while a mature corporate debt market can provide timely cash flow. But currently, no over-collateralized loan protocol on-chain can underwrite projects because the expected income stream securing the loan does not yet exist. To allow DeFi to function as a truly parallel financial system, we need to establish mechanisms that provide the "kickstart" for transactions, consumption, and investment.

"Witch Lending" and "Default Incentives"

Ideally, honest borrowers could establish on-chain credit for addresses just as in the real world, proving creditworthiness through repeated good behavior. However, this naive approach violates a fundamental economic constraint: the motivation to default. The penalties for ignoring debt in the real world are often so severe that few consider voluntary default. While we may not be able to send all debt defaulters to prison, the deduction of credit scores and the resulting increase in borrowing rates and denial of credit are actually effective and quickly actionable incentives. If you realize that having a low social credit score means you can't take the train, it naturally creates a constraint.

In the cryptocurrency world, this reasoning does not apply because we lack a universal and sufficiently binding social credit score: not due to a lack of attempts, but because the concept of a credit score assumes a clearly defined entity to which the score is assigned. Similarly, this is straightforward in the real world: borrowing entities correspond to known individuals or companies, and if issues arise, they can be traced back to mailing addresses, tax numbers (SSN or EIN), and phone numbers. In the blockchain world, the typical "borrower" is a hexadecimal address that stubbornly provides no information. It may be marked on Nansen (possibly incorrectly), but there is no guarantee that a recognizable individual is behind the address, let alone the possibility of recourse in the event of default. The address could be one of many belonging to an individual, part of a protocol treasury, or a bot. Of course, this versatility has plenty of advantages: it allows DeFi protocols to operate in a permissionless, composable, and decentralized manner. But what happens when address 0x1337 stops paying interest?

This fragmentation of identity incentivizes what we call witch lending: lending on behalf of anonymous or one-time entities, then defaulting without broader repercussions on reputation. Before globalization and modern record-keeping, witch lending borrowers were also common in the real world; in 18th-century Europe, it was quite possible to accumulate unpaid debts in England and then escape to France, starting anew under a different name. Even so, this physical escape ultimately required some sacrifice. But on-chain, especially with the help of Tornado or fake accounts on centralized exchanges, creating and funding a new address anonymously can be as simple as a few clicks.

Due to the existence of witch lending, unsecured lending protocols must strive to avoid being rapidly drained of funds by fraudulent borrowers. Admittedly, it is possible to reward long-term good behavior in some way; anonymous participants would still be incentivized to maintain the reputation of their accounts, just as trading platforms reward numerous participants with lower fees. However, we expect that most anonymous borrowers will not have this level of reputation capital. Most borrowers have every reason to view the protocol as a repeated prisoner's dilemma opponent, "cooperating" through good behavior until the value of the unpaid loan grows large enough to "betray" and abscond with the funds. This is why protocols aimed at providing general loans must address not only credit risk issues but also identity issues.

Each of the three lending models adopts different mechanisms to guard against witch attacks. For over-collateralization, the incentive is economic. In decentralized prime brokerage, it is algorithmic and technical. With on-chain identity, it is social and legal. All three mechanisms create conditions for a viable decentralized lending market, but each mechanism requires different types of sacrifices and entrusts success to different types of market participants.

Security Through Over-Collateralization

The first answer to the witch attack problem, distilled to "Why wouldn't I just run off with the money?", is to use economic incentives as a form of economic security: you won't default because you would lose money. This is the commitment maintained by every fully collateralized debt position: when faced with the choice between giving up collateral or the loan, borrowers are more inclined to retain the use of the collateral because it has greater value. The invariance of over-collateralization—ensuring that collateral is always worth more than the loan—is maintained through liquidation auctions, where participants are rewarded for purchasing debt positions when the loan's value rises to a level that poses a risk of being under-collateralized.

Most activity in protocols like Aave and Compound is single-layer, with deposits and borrowings occurring only once, although, as mentioned, there are some ways to achieve leveraged exposure. Some protocols, like DeFi Saver, create an automated execution (recursive deposit) layer on top of these lending protocols, while others (like Alpaca) offer platform-native leverage. However, since total debt never exceeds total collateral, even if nominal risks increase exponentially, all borrowers' positions remain over-collateralized, ensuring the solvency of the entire protocol.

Establishing a Closed System with Prime Brokerage

The second solution to the witch attack problem is to maintain reliance on "code is law," but from a different perspective: rather than enforcing over-collateralization in an open system, simulate unsecured lending in a closed system by restricting the use of funds. By creating a large but well-defined boundary through an interface or a set of integrations, protocols can offer the benefits of unsecured loans while technically maintaining over-collateralization (retaining ultimate control over the assets). We can view this "borrowing in a bubble" as algorithmic security: the code will solve the problem.

Indirect financial actions are a hallmark of traditional prime brokers, hence the model is referred to as "decentralized prime brokerage." However, the general concept of "prime brokerage" is much broader than banks providing execution for hedge funds. The actual operations involved—providing leverage, market access, and other benefits when trading and executing trades on behalf of clients—also occur every time a client interacts with centralized exchanges like Binance, FTX, or Coinbase. For these exchanges, the system's "boundary" is defined by the exchange's own partnerships and enforced through a combination of technical controls and human review.

Decentralized prime brokerage is the result of defining boundaries and executing trades on-chain. Decentralized prime brokerage protocols can implement necessary restrictions through several types of access control, such as:

• Issuing assets to specific wallets, allowing for a clearly defined set of commands and smart contract interactions
• Issuing assets to proxy contracts, allowing deposits from any wallet but only enabling a specific set of functions
• Being priced in specific tokens, only transferable between protocols with pre-specified interfaces with the lender (we have not yet seen this method, but modified ERC20 could theoretically operate this way)

Restrictions can be specific, such as whitelisting function calls allowed by each smart contract, or general, such as a minimum value invariant for addresses holding debt. Regardless of the method used, the key is to ensure that loans (a) can be fully utilized but (b) cannot be converted into freely tradable tokens in unrestricted self-custody wallets.

Oxygen is an example of decentralized prime brokerage, and other brokerages like DeltaPrime are rapidly building new solutions. Gearbox's credit accounts are another variant of this model, although leverage is currently limited to yield farming and liquidity provision. However, the cumulative trading volume of all decentralized prime brokerage protocols is still several orders of magnitude lower than that of over-collateralized loans.

In the long run, we expect decentralized prime brokerage solutions to exhibit strong network effects because the utility of borrowed funds depends on the number and quality of available integrations on each platform. Network effects often drive consolidation, so we expect to see, if not a winner-takes-all scenario, at most a few major players negotiating and integrating directly with each other. Of course, protocols can also offer permissionless integrations through smart contract interfaces, but for key markets and complementary prime brokerage solutions, we hope protocol teams can drive more customized relationships.

DeltaPrime Prime Brokerage Lending Model

Real-World Mapping with On-Chain Identity

Both over-collateralization and prime brokerage methods impose restrictions on credit, limiting the types of leverage available (over-collateralization) or the available uses of capital (prime brokerage). To achieve fully general loans, where borrowers can truly run off with the money, we need another incentive mechanism to ensure that the net loss from doing so exceeds the direct gain. Ultimately, this requires imposing broader consequences on the borrowers themselves, which we call social legal security: you won't default because you will face penalties off-chain.

Returning to our earlier case of social credit scores, borrowers who default in the real world face two consequences: the prospect of their income or assets being forcibly reallocated, and the impact on their creditworthiness, which will make future loans more expensive or unavailable. As long as either of these conditions holds, borrowers typically cooperate, and in either case, each loan must be traceable to a real-world entity that can impose reputation and/or legal consequences. That is to say, while one party must have enough information about the borrower to impose consequences, it does not necessarily have to be the lender. This is the idea behind executing a minimalist version of KYC through on-chain identity. As long as at least one participant can verify the borrower's identity, providing incentives, and a participant willing to underwrite the loan and provide funds, there is no need to disclose the borrower's identification to any party other than the source of funds.

One of the more complex identity-based approaches is Goldfinch, which aims to serve emerging markets by bringing both underwriters and borrowers on-chain. Using a three-party model of supporters, liquidity providers, and borrowers, Goldfinch divides risk between general LPs holding the "junior" tranche of each loan and supporters providing underwriting services in exchange for higher returns for the "senior" tranche. So far, the platform has provided over $100 million in loans. TrueFi is another form of this model, having borrowed nearly $400 million at the time of writing. Like decentralized prime brokerage, identity-based lending has a long way to go to reach the scale achieved by over-collateralized lending.

Goldfinch's Lending Model

Horizontal Comparison of the Three Approaches

After outlining the three main methods of on-chain credit, we can compare their strengths and weaknesses from several aspects.

Basic Attributes

We will start by summarizing the basics: what are the fundamental attributes of each method, and what types of lending activities do they support? We will define the categories as follows:

• Security: How does the lending mechanism prevent witch lending?
• Impact on Borrower Liquidity: Does the opportunity for borrowers to access liquid funds increase or decrease?
• Borrower Identity: Who or what is ultimately responsible for repaying the loan?
• Use of Credit: How broad is the range of uses for the credit asset obtained?

As mentioned, over-collateralized loans typically reduce borrower liquidity unless amplified by circular or flash loans (in a limited sense). Decentralized prime brokerage occupies a middle ground between over-collateralized and identity-based models, creating liquidity for trading risk and asset ownership but limited to services integrated with the protocol. Identity-based loans successfully unlock true general credit but require a stronger concept of identity in return.

Key Growth Points

From a systems perspective, we can also assess each lending mechanism by looking at the key growth drivers of the protocols using them.

• System Complexity: How many "moving parts" are required to facilitate a loan?
• Network Effects: To what extent does utility depend on the number and quality of participants?
• Value Drivers: What are the key points that determine the value of the service to borrowers?
• Key Groups: What type of participants is most critical for building a strong ecosystem?

The over-collateralized model easily wins in terms of simplicity and low startup costs, which explains why over-collateralized lenders tend to view DEXs as the first complement to each new blockchain ecosystem. Both identity-based brokerage models and prime brokerage models require more effort to overcome cold start issues, as they have relatively strong network effects. However, these two types of networks rely on different core growth drivers: decentralized prime brokerage expands through increased integrations and partnerships, while identity-based systems expand through their underwriters. Given that these network effects drive up the capital costs of identity-based systems and enable broader use of capital in prime brokerage systems, we believe that the network effects in prime brokerage are more fundamental.

Crypto Spirit and the Credit Impossible Triangle

Finally, we will compare these three methods on a philosophical basis, examining the extent to which they align with various aspects of the crypto spirit:

• Decentralization: To what extent can the system be controlled by a few dominant parties?
• Composability: How easy is it to integrate services with other protocols?
• Permissionless: To what extent does the system rely on gatekeepers or third-party institutions?
• Self-sovereignty: How much control do borrowers have over their identity?

Over-collateralized lending embodies the benefits of on-chain approaches: anyone can lend, borrow, or build on top of the protocol without the need for identifying information or third-party gatekeepers (aside from permission pools). Identity-based lending and prime brokerage lending both make sacrifices to achieve broader lending, but the trade-offs come from different domains. Decentralized prime brokerage retains permissionless and self-sovereignty, but at the cost of composability and some decentralization. Driving new use cases and partnerships may require concentrated efforts from core teams, and composability is limited since funds can only flow within the boundaries defined by the protocol. In contrast, identity-based lending retains composability due to general loans but sacrifices permissionlessness, as each loan requires due diligence and approval from off-chain gatekeepers. Self-sovereignty is also affected, as borrowers must voluntarily provide real-world information, even if ultimately only the final proof or encrypted hash is posted on-chain.

In practice, we can view the trade-offs of on-chain lending as another dilemma—the credit impossible triangle. On an ideal platform, loans would be composable, permissionless, and unsecured, but each solution's foundation can only address two issues, while facing challenges on the third:

The End of the Story: What Will the World Ultimately Look Like?

After addressing the pros and cons of each lending method, we may also ask: is there one party that should ultimately "win"?

A credit market driven by prime brokerage will ultimately look somewhat like an archipelago, with each protocol integration serving "islands" linked by mutual agreements between them. As the total capital and scope of use increase, each protocol will benefit from internal network effects and also from "trade" between protocols achieved through reciprocal agreements to fulfill debt positions. This highly interconnected, execution-agnostic final state will be familiar to traditional finance veterans: when a hedge fund buys a stock, its clients neither know nor care whether it is Morgan Stanley or Goldman Sachs making the purchase.

In contrast, an identity-based credit market may tend to reconstruct real-world relationships. Local underwriters will have better risk models tailored to the demographics they know best, and borrowers will be able to obtain better rates from underwriters who understand their identity and credit history. Even in a "decentralized" network with many independent members, these structural factors will be preserved to some extent. However, as lenders learn to output their risk models to similar borrowers in different markets, the importance of geography may diminish. In the long run, competitive on-chain markets with borderless lenders and semi-commoditized risk models could not only drive better expansion of credit channels but also serve as a balancing force in real-world credit markets.

Despite the vastly different scales of these two models, we believe they are complementary, with each model tailored to different core use cases and borrower profiles. You could use a decentralized brokerage to obtain a loan, sending your BAYC to a proxy wallet that allows you to sign ownership. Or you might just need a good borrowing record to access loans using a credit score while maintaining identity protection.

A Bright Future

In this article, we explored three lending mechanisms that make secure and decentralized credit markets possible: over-collateralization, decentralized prime brokerage, and on-chain identity. While the over-collateralization model has driven over $100 billion in lending activity and significant innovation in protocol design, we believe that for achieving comprehensive economic use cases, prime brokerage and identity-based models will be crucial. DeFi has already provided excellent solutions for trading, forex, remittances, and "borderless" capital, but to drive lasting economic value, more general on-chain credit models are needed. Because only based on this can we provide comprehensive lending services and usher in a new breakthrough for the current on-chain lending market.