Apr 2nd, 2018
The XOR Protocol is a financial system built on a distributed public ledger that supports a unique class of markets and securities, which are more transparent, decentralized, democratic, and general than current systems. Examples include simple loans, mortgage bonds, and derivative markets such as Collateral Debt Obligations (CDO) and Credit Default Swaps (CDS). XOR uses smart contracts to model any type of loan. Many markets can be modeled as debt contracts, thus enabling a universe of potential XOR markets with the aforementioned benefits over their current analogs. All market transactions are recorded on a public ledger using Distributed Ledger Technology (DLT), a more general paradigm of blockchain technologies, in order to ensure end-to-end transparency. An additional benefit of XOR’s decentralized approach is the elimination of parasitic fees, such as fees from third party credit ratings agencies, insurance providers, and centralized banks. Borrowers receive superior borrowing rates and lenders receive higher ROI. Another important aspect of XOR is self governance and modularity. Actors in the system will be able to choose their own credit score implementations, insurance rates, risk models, and any other market variables. We can assume that any given model will not stay relevant for infinite time, but in the future at least one type of basic functioning model will exist. XOR aims to rectify this by allowing democratized modularity in the creation of markets. This flexibility is important in order to ensure that the XOR Protocol can adapt to future changes in market and real-life circumstances. In addition, XOR is designed as a protocol layer rather than as its own system. This makes it possible to build XOR on top of any other currency system, i.e. the Ethereum network or the Bitcoin network. As long as at least one distributed ledger exists, the XOR Protocol can exist. XOR will open doors to fair loans and credit access to people across the world by providing an accessible platform through which anyone in the world can create a market or invest in markets in a way that is transparent, distributed, fair, and flexible.
The 2008 Housing Crisis is well known to have triggered a global financial meltdown. Many factors contributed to the crisis, but one major cause was financial agencies’ lack of transparency. Lenders were allowed to create contracts with borrowers who could not actually afford the mortgage rates over time, creating a market of immense risk. This was compounded by additional liquidity and market exposure from derivative instruments such as CDOs. The risk was covered up by ratings agencies, who had the authority to incorrectly label CDO tranches. Under a system where debt issuers pay ratings agencies to rate the strength and weaknesses of the securities they offered, ratings agencies were incentivized to underrate the riskiness of securities, creating a system where investors were unaware of the true riskiness of investments. When borrowers began to default on the contracts because they could not actually afford them, the housing market collapsed, triggering a global recession.
XOR solves this problem by employing a distributed and self-governed system. All transactions are stored on a distributed public ledger. This makes it impossible to misrepresent the riskiness of contracts because there is no one central party to determine and rate such loans. Anyone can check the ledger and verify the contents of a contract. This automatically allows for a more fair contract system for individual users. Moreover, it becomes possible to create derivative markets (and thus provide important liquidity to markets) in a transparent and scalable way that circumvents the issues faced in 2008. As XOR is decentralized, anyone in the world can make any market or derivative market. With additional governance structures, participants can democratically achieve maximum system efficiency.
Another issue with modern financial systems is that once elements become entrenched in the system, it becomes hard to improve upon them, even if such elements are harmful. For example, the mechanisms for calculating FICO credit scores are entirely opaque, meaning that no credit borrower can actually know why their score is what it is. Additionally, FICO scores reward some unspecified amount of debt exposure, and penalize having zero debt exposure. A system that encourages paying off debts rather than obtaining them would create a healthier financial society. However, it is possible that a future financial system could require a different approach to personal ratings. In this paper, we propose models that are transparent and scaleable. In spite of that, the core premise of XOR is the acknowledgement that no matter how good a model may be, it could become obsolete in the future. Keeping this in mind, our implementation is modular and thus future-proof.
XOR aims to provide a general financial framework that can encompass any market type. Examples include standard cryptocurrencies, stablecoins, physical commodities such as gold, mortgages, car loans, etc. Ultimately, we wish to create a general and open protocol that will stand the test of time and last for as long as the concept of finance itself.
A market can be initialized by anyone with access to the XOR Protocol and can be created on top of any distributed ledger. Because XOR utilizes a smart contract architecture, markets can be based on any asset that can be modeled as a type of monetary debt contract. Any of a market’s parameters can be adjusted modularly, enabling a range of simple hardcoded fixed parameter markets to fully flexible and modular markets.
A market’s interest and insurance rate calculations are carried out by a teller function, which is a piece of arbitrary code that takes as input the parameters of a loan (maturity period, borrower trust score, etc.) and outputs an insurance rate and interest rate. Markets are associated with a teller function at creation time, a process that requires a fee corresponding to the cost of deploying the function as a contract to the blockchain network. Since the function resides on the blockchain, the code is transparent. Currently, a market cannot change its associated teller function once created.
Users are incentivized to create markets with high accuracy teller functions: functions which consistently output insurance and interest rates that justly reflect the underlying risk of the loan. This is accomplished by awarding a fixed percentage of each transaction carried out on a managed market to the market creator. Since users will only be willing to pay this transaction fee if the teller function has been historically accurate, it is expected that markets that eventually become popular on the XOR network will serve as industrial-grade, high-throughput mediums of exchange between lenders and borrowers who have highly variable or even unknown risk profiles. Market definitions themselves are modular; for example, one can swap out the fixed transaction percentage for a variable percentage.
Having programmable teller functions ensures that the XOR Protocol will never become stale as more intelligent implementations are discovered. Rather than accept permanence within a financial market, the XOR Protocol encourages healthy competition in order to find the most intelligent teller functions. For example, in theory this allows markets to not only be traded by artificial intelligences, but also fundamentally created and managed by artificial intelligences. The flexibility of the XOR teller system provides a testbed through which to achieve optimal financial markets.
In order to understand how the XOR Protocol works, we need to first understand how XOR loans work. A loan is essentially a contract describing a debt obligation. This can be used to model many other types of securities. For example, a stock is a trade between cash and equity. With the advent of smart contracts, any type of financial contract can be encapsulated within the XOR Protocol.
In this section, we will walk through the example of a basic loan, and compare it to an XOR loan. Loans are the basic building blocks of financial markets. Most financial instruments can be modeled as debts or derivatives of debts. It is important to note that a simple peer-to-peer loan is a special case of a risk-mitigated market loan.
A simple traditional peer-to-peer loan involves one borrower and one lender. Normally, the basic steps for securing a typical loan would be: A simple traditional peer-to-peer loan involves one borrower and one lender. Normally, the basic steps for securing a typical loan would be:
- The lender assesses the risk profile of the borrower. He normally does this with by requesting a FICO score report from a consumer reporting agency, such as Equifax. This costs money to do. FICO scores are a blackbox and are notoriously difficult to navigate, because FICO does not publish its algorithms.
- If the lender accepts the loan, he has the option of purchasing insurance in case of a default. Insurance providers pay the lender some or all of the borrowed amount back in the event of a default. They charge extra fees on top of initial collateral.
- The lender and borrower agree upon contract terms and proceed with the transaction. The borrower either pays his debt back on time with interest, or he fails to do so and defaults. This can be extended to include annuities as well.
In this system, the two primary third parties are the credit score provider and the insurance provider. Both parties charge extraneous fees to system participants, and both parties are centralized and opaque.
An XOR peer-to-peer loan is different in that it utilizes decentralized approaches to centralized systems. This creates a fee-free structure, enabling better rates for borrowers and higher ROI for lenders by eliminating trickle-down costs versus traditional loans, with no real disadvantages. An XOR loan works as follows:
- The lender asses the borrower’s risk profile by using a distributed rating mechanism (called Trust in XOR). Trust allows for the borrower to have a risk rating by inputting a variety of factors. System participants vote on the Trust protocol and are thus incentivized to make it fair. Trust will be discussed in detail later in this paper.
- Instead of purchasing insurance from an insurance provider, the lender places a collateral fee into a collected pool with all other lenders. Lenders vote upon the parameters of the pool, such as how Trust ratings correspond to different collateral fees, how defaults are handled, etc. This will also be described in detail later in the paper.
- The two parties agree upon a contract, and the anonymized contract is stored on a distributed public ledger. This allows for total transparency of the system, enabling a rich ecosystem of derivative markets.
A natural extension of the peer-to-peer is risk mitigated market loans. Here, an initial borrow request is split amongst multiple lenders, and each lender separately earns interest on their lending amount. The borrower would approach the loan in the same way as he would with a peer-to-peer loan, and lenders would approach their portion of the contract in the same way as a peer-to-peer loan as well. In a risk mitigated market loan, all lenders share responsibilities for loan defaults. Through the creation of a many-to-many relationship between lenders and borrowers in each market, the default exposure for each lender is greatly reduced.
Within a given market, we have designed a procedure for processing loan contracts. It is important to remember that loans are just a model for generic financial contracts. The loan procedure is processed within a round. Rounds include a parameter voting stage and insurance analysis stage. Information for both stages are processed from individual and market level risk profiles. Investors may also process bids with Loan Intents, which are contracts of automated contract selection logic.
All markets operate in rounds, which are composed of a request period followed by settlement. During the request period, borrowers are able to request funds with a certain maturity date and maximum interest rate they are willing to pay. Lenders are also able to file loan intents during this period, which contain a single vote for the risk coefficient used to calculate interest rates in the market (described in more detail below). Lenders and borrowers select markets which best fit their needs by looking at the market parameters. As with other decentralized currencies, multiple lenders can fulfill a single loan request, increasing market liquidity. Market creators are rewarded for the success of their markets. During parameter voting rounds, voters that provide parameters which are close to those voted upon will also receive rewards.
Lenders and borrowers look for markets that best suit their needs. It could be possible that a borrower is looking for the highest amount of cash for the lowest interest rate, and a lender is looking for lending the lowest amount of cash at the highest interest rate. By removing parasitic fees, lenders and borrowers are able to reconcile their interests. Both lenders and borrowers have the flexibility to choose markets however they want. For example, borrowers could query markets for how quickly they can receive cash, rather than for the best financial deal.
Borrowers will always look for the highest amount of cash for the lowest interest rate. On the other hand, lenders will always look for lending the lowest amount of cash at the highest interest rate. While these may seem to be conflicts of interest at first, the XOR Protocol creates incentives for both parties to participate fairly and in a mutually beneficial manner.
A market contains its own risk coefficient (to be voted upon by lenders), and every individual contract has its own risk coefficient as well. Assuming that interest rates are a function of risk, lenders can vote on the mapping of risk coefficients to interest rates. In addition, borrowers can choose to accept that market or to move to a different market.
Lenders are interested in making as high a return on investment as possible, meaning that they want to gain as much as possible in direct interest while paying very little in insurance pool fees. This further drives down the price for borrowers because lenders are incentivized to not vote on high interests at risk of losing customers to another market.
In order to hedge against defaults, lenders have the option of posting a collateral fee to the collected insurance pool. This fee is based on the Trust of any given borrower combined with the risk profile of the market. In general, a higher Trust would correspond with a lower insurance premium (collateral fee) for the lender, because that borrower would be unlikely to default. Similarly, a lower market risk factor would correspond to lower insurance premiums for all lenders. In the event of a default, the lender is paid back based on the programed parameters of the insurance agreement as well as the market. For example, markets may choose to process insurance payouts over time so as to avoid rapidly deteriorating the shared insurance pool. This insurance structure is modular at the market level, thus allowing for total flexibility and control. In many markets, such as in risk-mitigated markets, an insurance pool may not be relevant. XOR simply presents the choice of implementing an insurance model.
When a contract is created, the XOR Open Loan system forces a low interest rate for buyers, and low insurance fees (and thus highest possible ROI on interest) for lenders. It would make no sense for a lender to place a bid for a contract with an unnecessarily high fee amount because that would devalue the bid and increase his own cost to no real benefit. In the XOR Open Loan system, everybody wins by collaborating for mutual benefit.
To supply funds to a certain market, an investor creates a Loan Intent with the specific amount of funds to be supplied, and the risk coefficient that the investor believes characterizes the current risk-reward profile in the market. The Loan Intent is essentially a pre-programmed instruction to automatically invest in a contract that meets certain programmed parameters. If the investor is not selected to participate in the round (i.e. since the supply of credit exceeds the demand for credit, or all contracts are fulfilled), the funds are simply returned to the investor. To remove unnecessary manual labor from the process of supplying funds, the XOR platform could provide a simple utility to reinvest unused funds by recreating rejected Loan Intents until one is accepted. This process closely models a market order in stock market terminology.
Portfolios in XOR, in a manner analogous to mutual funds in the traditional financial system, allow investors to organize their investments. Just as markets aggregate borrower requests, portfolios aggregate a user’s loan intents. This additional layer of abstraction allows XOR to natively support investment diversification; since each market represents a relatively consistent risk profile, holding a portfolio with many different markets constitutes diversification.
Just like a mutual fund, it is possible for many lenders to invest in a single portfolio, which simply means that each investor creates loan intents in the same markets and supplies funds to each market in the same proportions. This allows layman investors who are not capable of analyzing the individual risk-reward profiles of individual markets, as well as investors who do not have the time to construct their own portfolio, to choose a suitable portfolio to invest in out of the current well-performing portfolios. As with everything in XOR, the performance of a portfolio is completely transparent; statistics on past returns and losses can be computed by any user since historical information on the underlying markets is freely available. Finally, as portfolios represent a group of investments in different markets, it follows naturally that portfolios can be combined with other portfolios. This allows portfolios to be iterated upon seamlessly.
A robust Trust score protocol is required in order to allow lenders to verify borrowers without a third party underwriter. The two situations that are possible are that a borrower has a Trust score on the system, and the borrower does not have Trust. In the event that the borrower does not have Trust, the system has to initialize a score. We propose a combination of parameter fitting user input data and a process of Social Staking. Through Social Staking, borrowers are encouraged to verify other borrowers. As with everything else in XOR, the Trust protocol is modular and can be tuned per market. In addition, a given market’s Trust protocol will be transparent, creating a more fair system. Higher Trust directly translates to increased buying power and a better risk profile.
Trust is the decentralized and transparent XOR analog to a credit score. High Trust indicates that a borrower is highly trustable on the network. This means that the borrower’s opinion on vetting low/no credit customers is highly valuable. As a reward for having high Trust, the borrower will have access to more capital and better interest rates from lenders. A high Trust score indicates a low risk profile for that borrower. Borrowers can increase their Trust by paying back loans and by correctly verifying other borrowers through the Social Staking system. Trust can be determined by parameter fitting user inputted data such as existing FICO scores, evidence of education, and any other factor that could have association with an increased likelihood of repaying a debt. In addition, Trust can be supplemented by Social Stakers.
Social Staking is a process through which a borrower can increase his Trust by having other borrowers stake his position. In the event that the original borrower pays the loan back, Trust increases for all participating Stakers, and if a borrower defaults, Trust decreases for Stakers. This creates a system where Stakers are incentivized to accurately assess fellow borrowers.
In the event of a customer without Trust requesting a loan, there are two primary options. The first is that the customer posts a loan request, and the contract is treated as extremely high risk. In this situation, the borrower would receive a low amount of cash and high interest. On the lender side, the lender will post a high collateral fee to the insurance pool, but will also gain high return on investment from the high interest.
Another option for a customer without Trust is to receive initial verification from borrowers who have Trust. In this scheme, having a high number of verifiers as well as verifiers with high Trust Scores will improve the customer’s initial Trust.
The cryptoeconomic model for the XOR Protocol provides a framework for a revenue model and system valuation. We introduce the XOR Value Token (XVT) as the representation of the XOR Protocol’s value. In addition, we introduce the concept of a stablecoin with which to settle the loan contracts themselves. In addition, we present the possibility of using atomic swaps so that lenders and borrowers can use any atomic swappable cryptocurrency. We also describe the interaction between XVT and governance rights within the XOR Protocol. It is important to note that XVT and the proposed contract stablecoin are not necessarily the same token, although loans could be processed with XVT as the currency medium if necessary.
The XOR Value Token (XVT) represents the net value of the entire XOR Protocol. Because the XOR Protocol can be built on top of any distributed ledger, XVT’s value encompasses the value of all of those ledgers’ XOR Protocol layers. This is the model used by most other cryptocurrency companies, ex. Bitcoin, Ethereum, etc.
The Ethereum Foundation provides most of the inspiration for XVT’s model. As the value of XOR itself grows, so too will the value of XVT, which can be used in the XOR Protocol in ways described below. A portion of XVT will be kept aside for the initial members who worked on XOR. A second portion will be kept for running XOR’s company itself, as a coffer. The rest of the tokens, the vast majority, will be purchased by investors and system participants.
Price fluctuations can be extreme in cryptocurrencies. To combat this, some members of the cryptocurrency community are working on stablecoin implementations. Stablecoins are usually aggregates of other coins with low price fluctuations, statistically adjusted so as to provide minimum movement. Makerdao’s stablecoin DAI is pegged against USD at a rate of 1:1. Stablecoins are important in XOR loan contracts because they provide certainty to a loan. For example, if a non-stablecoin is used, its value could increase percentage wise beyond the interest rate of the loan before the lending period is over. In this situation, the lender loses money. Stablecoins can help to prevent this situation.
Atomic swaps allow for the exchange of two cryptocurrencies using the same hashing algorithm without requiring a trusted third party, provided that a bridge between the two currencies has been built. This provides liquidity to cryptocurrency markets and allows for some degree of equivalency between different currencies. We can employ atomic swaps within XOR so as to support contracts that exchange different cryptocurrencies. XOR will support other systems of interoperability beyond atomic swaps.
XVT will be used as the functional currency on XOR. There are several examples of uses of XVT that provide value to XVT. It is important to note that this list does not encompass all use cases for XVT.
- When creating a market, the market creator can receive a portion of transactions or otherwise generate revenue off the implemented revenue model, as discussed above. To encourage participants in the market, the market creator can lock some XVT into the insurance pool as a hedge against his market’s failure. A greater amount of XVT leads to a decreased market riskiness.
- Within any type of market, lenders can lock XVT in the insurance pool so as to increase their voting rights. In general a greater amount of XVT will correspond with greater voting rights, but this model can be nonlinear and decided upon by the market creator or the voters.
- Borrowers can lock XVT into the market or a contract as collateral. This essentially allows for the possibility of secured loans, versus unsecured loans as described in this paper above. In certain markets, borrowers can also have voting rights.
- XVT itself can be used as the currency of choice for contracts.
- For both lenders and borrowers, XVT can be incorporated into the Trust computation. This is because an actor with a high stake in the XOR system is more likely to be trustworthy. XVT can also be used to stake others. In this system, a borrower would essentially use his Social Stakers’ XVT as a staked social collateral for increased Trust.
- When creating Loan Intents, lenders can lock an amount of XVT as the transfer currency. This can be extended to the concept of portfolios as well.
Governance is how choices are made in a system. Examples of governance systems include unweighted voting, weighted voting, ring voting (where a random subsection of a group is chosen at the voting), node/supernode (where supernodes have voting powers but nodes vote for supernodes, much like a Republic), and many other structures. Governance is important to the XOR Protocol because many of the modular parameters need to be governed by the distributed user group. XOR’s governance system itself can be modular so as to reflect the most relevant style for the market and time.
In order to be able to use the protocol, it is necessary to implement a robust Proof of Identity system so as to be able to legally reconcile any possible bad acts. Without Proof of Identity, a financial system is vulnerable to attacks where one user can create multiple identities and increase his borrowing power.
Proof of Identity is a complicated problem that many blockchain companies are attempting to solve. uPort is a company that is developing an open source and decentralized identity system on the Ethereum network. uPort works by storing personal keys on your smartphone and uses a smart contract to lock that information and provide recoverability in the event that a person loses their key. This is an advantage over traditional public/private key paradigms, where losing a private key essentially means losing access to the information protected by that key. Other companies such as Cambridge Blockchain LLC and Civic also have variations on Proof of Identity. For customers in countries which heavily document their citizens such as the US, simply uploading a government ID would potentially suffice as Proof of Identification.
It is again important to note that because XOR is designed to be modular, any identity system can be swapped in or out. For example, in the future, perhaps biometric identifiers will become the new bank-grade standard. We want to make sure that XOR is flexible enough to handle these types of situations.
Contracts have to be legally reconcilable in order for lenders and borrowers to be able to safely participate. For example, in the event that a borrower borrows money and tries to disappear, it has to be possible for lenders to pursue legal action. This requires Proof of Identity as well as a strong legal protocol.
XOR currently plans to use the OpenLaw protocol to handle automation and reconciliation of legal contracts. OpenLaw allows for the decentralized and autonomous generation of legal contracts. It has flexible functionality, including cross-border functionality where lenders and borrowers from different countries can select the appropriate jurisdiction. Combined with blockhain notarizing and Proof of Identity, XOR will be able to provide a framework for customers to assess and reconcile legal issues. The OpenLaw contract can be encapsulated within the lending/borrowing contract itself.
In this section, we briefly discuss several examples of potential XOR Protocol applications. The first two examples are examples of loans, while the rest of the examples are examples of derivative markets.
Typical college loans currently have high interest rates and default rates. The 2017-2018 student loan rates are 4.45 percent for undergraduates, 6 percent for graduate Direct Subsidized Loans, and 7 percent for graduate Direct Plus Loans. As of 2018, the total exposure is 1.48 trillion USD at a default rate of 11.2 percent. The primary issue with high exposure and default rates with college loan borrowers is not necessarily the loans themselves, but the rapidly rising cost of college itself.
An XOR college loan could potentially alleviate part of the problem by attempting to drop interest rates and thus provide cheaper loans. This would be possible with the removal of third party fees. In addition, this would essentially enable a market where lenders can invest in potential college students’ futures. We see this as a popular and interesting alternative to typical government college loans.
A goal for XOR is to enable members of third world countries to participate in the modern financial system. XOR can help with this because of its decentralized and fee-free design. Since anyone can participate in an XOR market provided they meet the preconditions (proof of identity, potential collaterals, etc.) the XOR Protocol is far more accessible as a financial system than traditional banking.
However, it is important to note that Proof of Identity is difficult in countries which do not have strong documentation procedures, because government identities may not exist for everyone.
An interesting use case for XOR would be as a startup fundraising platform. A traditional ICO involves trading some value of tokens (representing equity in the system) for cash. The company running the ICO then uses that cash to further develop their system.
An XOR loan round would be fundamentally different in that instead of trading cash for equity in the startup, the startup would instead provide a contract with a repayment protocol. This contract can be programmed with complex logic to dictate the repayment. A simple example would be to simply pay the cash back with interest. A more complicated example could be to pay the investor back a discounted percentage of the company’s valuation at the time of the settlement date. This type of fundraising round could be used in conjunction with traditional fundraising (i.e. angel investments, venture capital, etc.) and cryptocurrency-specific fundraising styles (public token pre-sales, ICOs, etc.)
When a lender assigns a valuation in XOR Open Loan, he is essentially buying a Credit Default Swap, but with the insurance provider being the collective pool of lenders. This provides a huge opportunity for accurately priced, essentially no-fee, credit default swap contracts. These insurance purchases can then be chained as underlying assets.
An XOR Open Loan Collateral Debt Obligation would also be an incredibly impactful event in finance. The tranches of the CDO can be comprised of loans of different risk tiers. Senior tranches could receive the first payments as the tranches descend to high risk borrowing customers.
More interesting is the potential for synthetic CDOs. A synthetic CDO is essentially a gamble on the performance of another CDO. These gambles can chain with increasing odds, leading to a massive amount of leverage in the market that otherwise didn’t exist. Synthetic CDOs when used correctly are healthy for a market because they introduce liquidity and leverage. However, many synthetic CDOs have opaque underlying assets.
An XOR Open Loan synthetic CDO would be truly revolutionary because of the blockchain architecture. All tranches of the CDO would be fully viewable due to the public ledger of the blockchain recording every step of the creation of the CDO. Thus, XOR can enable a healthy and relatively safe form of CDO leverage and liquidity.
An especially unique facet of XOR Open Loan is that it enables anyone to create any type of market. As explained previously, a debt obligation contract can be used to generally model a myriad of economic systems. Because of this, the XOR Open Loan system allows for the creation of markets by anyone within the framework of fairness, accessibility, and trustworthiness builtin. For example, one could create a sports betting market, or a market betting on the weather, or even some sort of crazy nth order synthetic CDO market where the underlying assets are cryptocurrencies pegged to weather patterns. As long as there are lenders and loaners, any market can exist through XOR Open Loan.