Eximchain Whitepaper

Friday, June 15, 2018
Download document
Save for later
Add to list
Whitepaper.io is a project of OpenBook

Eximchain: Supply Chain Finance solutions on a secured public, permissioned blockchain hybrid Juan Huertas , Hope Liu and Sarah Robinson Date of Whitepaper: March 13th 2018 V6 ABSTRACT ships. SCF enables buyers to optimize working capital and Eximchain brings visibility to global supply chain finance suppliers to generate additional operating cash flow while si- (SCF) through smart contracts. We build a public, permis- multaneously minimizing risk across the entire supply chain. sioned chain for small-and-medium enterprise (SME) buyers More specifically, SCF enables buyers and sellers to shrink and suppliers to create supply chain optimization tools and their inventories, collect money from customers faster and gain access to affordable capital sources to grow their busi- delay payments to their suppliers. Citi recently completed nesses. Our smart contract-based ecosystem allows SMEs to a working capital study that showed companies managing quickly implement customized SCF solutions or issue digi- working capital through SCF were able to reduce these as- tal tokens on a permissioned fork of Ethereum supporting sets by 30%, resulting in EPS (earning per share) increases data privacy. We adopt a consensus protocol and quadratic in the range of 1% to 4%. Furthermore, the top 10% of com- voting based governance model to provide practical, finite panies that reduced their CCC (cash conversion cycle) were time security guarantees on our public, permissioned block- rewarded with a 30% stock price appreciation[2]. chain hybrid. From financing to procure-to-pay, our products utilize smart contracts to optimize the global supply chain 3. THE PROBLEM for buyers, suppliers, and financiers. Eximchain is an official According to the International Finance Corporation (IFC), candidate of the Blockchain Regulatory Sandbox Program small to medium-sized enterprises (SMEs) in developing coun- in Guiyang, China and a member of EEA (Enterprise Ethe- tries face a financing gap totaling over $2 trillion[3]. In China reum Alliance). alone, the vast majority of the 40 million SMEs remain un- served by existing financial resources[4]. This financing gap is particularly concerning because, as cited by the World 1. ABOUT EXIMCHAIN Bank, SMEs contribute up to 60% of total employment and Eximchain project was kicked off in 2015 at MIT (Massa- up to 40% of GDP in developing markets[5]. Innovative SCF chusetts Institute of Technology) by a team with decades of solutions offer a powerful tool for SME buyers, suppliers and academic and industry experience in computer science, ban- financiers to overcome this funding gap[6]. Innovations in king and global supply chain across the world. Our vision this space have traditionally been driven by large corpora- is to use blockchain technology to bring the global supply tions and banks in the field; however, cutting edge financi- chain into the digital era and lower financing barriers for al technologies offer new opportunities to extend the bene- SMEs. The project has received valuable mentorship and fits of SCF to businesses of all sizes. According to a 2015 guidance from the MIT Media Lab Digital Currency Initia- McKinsey Report, of the $20 billion potential revenue pool tive, MIT Center for Transportation and Logistics, and Plug that exists for implemented SCF programs, only $2 billion And Play Fintech Accelerator. The project was chosen as the is being captured today[7]. One primary issue inhibiting the Grand Champion of the 2016 Boston Seagull Entrepreneur- widespread adoption of SCF programs among SMEs is that ship Contest, the Engine of Innovation Prize of the 2016 they currently lack a trusted tool that provides transparency Rice Business Plan Competition, and a Finalist of the 2017 to eliminate information asymmetry problems. Beyond this, MIT $100K Business Plan Competition Accelerate. there are quite a number of difficulties for adoption of SCF programs on a global scale: 2. SUPPLY CHAIN FINANCE OVERVIEW First of all, any supply chain strategy cannot be determined Traditionally, supply chain management has focused on in isolation. The market for SCF programs is inherently dy- the material flow of physical goods from manufacturers to namic: exogenous factors, such as interest rates, and endo- end consumers. However, the recent global economic down- genous factors, such as inventory decisions and capital cons- turn demonstrated that managing financial flows within the traints, influence the evolution of the system. supply chain can be as important as managing physical flows of goods and services[1]. Supply chain finance (SCF), one Furthermore, the complexity of implementing SCF across of the most exciting and promising new products emerging the global supply chain limits its adoption. It takes tremen- in the banking industry, is a set of technology-based busi- dous effort to onboard suppliers and integrate the process ness and financing processes that allow financiers to fund an with the existing operational flow by involving finance, pro- organization’s operations through its supply chain relation- curement, and IT departments inside the organization. 1

ted information displayed in contracts, it is hard for banks Last but not least, it is difficult to align incentives among to understand the market expectation and real dynamic bet- different participants. Suppliers, buyers and financiers are ween buyers and suppliers to mitigate risk. In 2014, a Chine- independent decision makers attempting to maximize profits se trading company, Dezheng Resources, and its subsidiaries in the face of asymmetric cost structures and uncertainties. were alleged to have used duplicate receipts to pledge metal Such independent profit maximizations often lead to poor as collateral for loans. The result was a flurry of lawsuits, performance of the entire supply chain. including the UK High Court case between Mercuria and Citi, over exposure to a $270 million financing deal[10]. The The Eximchain platform solves these challenges by offering Eximchain platform provides access to smart contract histo- seamless integration into the existing workflow of SMEs and ry between two parties, which grants banks more visibility enabling developers to create customized tools for specific into the supply chain operations and enables them to bet- businesses and industry use cases. ter estimate the risk of a transaction. Furthermore, access to this smart contract information will provide other financiers, 4. OUR SOLUTION beyond traditional banks and lenders, with an opportunity to fund and generate value off of these transactions. Eximchain offers a platform to implement smart contract based SCF solutions on a permissioned fork of Ethereum Eximchain allows developers to create variations on supply supporting data privacy. Our ecosystem will enable SMEs to contracts secured by the network and build solutions that are gain access to affordable capital sources by giving financiers customized to their global supply chain. Currently, supply visibility into the supply chain cash flow. The first game chain partners use a variety of solutions to align incentives theory-based application we plan to build, Multi-Party and mitigate inventory risk, including: buybacks, quantity Dynamic Contracting*, (details to be described under discounts, revenue-sharing and two-part tariffs contracts. On Appendix- Use Case Example) will be designed in such a the Eximchain platform, developers can build customized way that no partner can improve his profit by deciding to solutions based on specific industry needs, user dynamics deviate from the optimal set of decisions. That is, there is and market competition. Banks and anchor buyers can easi- no incentive for the buyer, the supplier, nor the financier to ly scale the solution on the global supply chain by involving deviate from the set of actions that will achieve the global- upstream and downstream players. Although these solutions ly optimal solution. The implementation is executed by the will be varied, the contracting process will be highly stan- consensus of the network in a standard, automated, priva- dardized and can be seamlessly integrated into the bidding te, and auditable fashion. From the user’s perspective, after process and other supply chain management tools. making the initial input and setting up the negotiation rule, they can receive the negotiation result from a “black box” Customized negotiation rules: where the system automatically executes an optimization engine through multi-stage coordination until Nash Equili- brium is reached and each participant’s incentive is aligned. Each graph represents one state transaction. The negotia- tion will end after all parties reach Nash Equilibrium based Buyers and suppliers can automate contracting processes on the negotiation rules set by each party. Grey arrows re- with the Eximchain network. Suppliers and buyers normal- present locked status while green arrows represent unlocked ly have different and conflicting objectives. For instance, status. suppliers want buyers to commit themselves to purchasing large quantities of products in stable volumes with flexi- ble delivery dates. Conversely, buyers need to be flexible to their customers’ needs and changing demands. The difficulty with global optimization is that it requires firms to surren- der decision-making power to an unbiased decision maker. Additionally, establishing trust between the supplier and the buyer is difficult to achieve. For example ,in January 2013, the Walt Disney Company sued Blockbuster, accusing them of cheating its video unit of approximately $120 milli- Private channel coordination: on under a four-year revenue-sharing agreement[8]. Through the Eximchain platform, buyers and suppliers can carefully design supply contracts to maximize profit on the supply Each communication channel is private where only the in- chain with more visibility into demand, inventory and ups- volved parties can see the information provided during each tream/downstream operations through a trusted and secu- negotiation. The system runs automatically based on pre-set red network. Additionally, interactions and agreements are inputs and only updates state of each negotiation cycle for saved automatically with a traceable record. involved parties. Financiers will be onboarded to the network with aligned incentives. In the real marketplace, suppliers and buyers fre- quently finance their working capital from a variety of credit sources, such as banks, and incur significant financing costs. Studies have shown that under the assumption of positive inventory financing costs, traditional supply contracts fail to achieve joint profit maximization[9]. Based on the limi- 2

5. VALUE FOR OUR STAKEHOLDERS up to 15% of annual working capital loan for SMEs can be unlocked either by paying less to cover margin deposits or by eliminating the fee of the third-party guarantee company. Financiers: On the buyer side, a company with $10 million in revenue Risk Mitigation and a 60% cost of goods sold (COGS) ratio can experience We enable dynamic, real-time monitoring of supply chain a cash flow increase of $16.7k per day of extended paya- financing processes to provide visibility into the whole sup- bles ($10m * 60%/360 days=$16.7k). Parlaying that into a ply chain’s operations. Lenders can now better understand 60-day term extension on trade-credit would result in a po- the actions taken by end purchasers and upstream suppliers tential annual working capital/cash flow benefit of $1 million of each deal to better assess risk. With the combination of for the buyer ($16.7k * 60=$1m). a “Purchase Money Security Agreement” (PMSA)[11], al- so known as purchase money security interest, on the loan, Credit Rate Arbitration which makes the lender the first claimant of the collateral The Eximchain platform enables anchor buyers to take ad- related to the loan in the case of default, the financiers gains vantage of their comparatively better credit ratings to offer a much better estimation on the risk and return of the in- SCF programs to fund their suppliers in return for extended vestment despite all external uncertainties. terms, lower prices, and improved long-term relationships. In this “buyer-driven” model, the anchor buyer becomes the Operational Efficiency financier in the ecosystem. With more information about the trade, the financier could more accurately set the rating and risk factor. Increased For example, if a supplier’s access to capital is 5 percen- transparency over trade transactions tends to lower the risk tage points higher than the buyer’s 2.25% financing rate, a factor and thus risk equivalent, which improves efficiency simple calculation will show the buyer that a 30-day invoice in operation and management of the loan. By digitalizing for $500k can be extended to 96 days without adding costs to and standardizing the contracting process, financiers save the supplier. On the other hand, if the buyer were to main- the overhead effort of processing paperwork, understanding tain the 30-day payment terms, the supplier could achieve a different terms, and going through multiple levels of manual reduced carrying cost of approximately $2,083( representing review to make a final decision about the credit they are 0.4% of the $500k contract). Armed with this knowledge, issuing. the buyer could negotiate a price reduction based on all, or some portion, of these savings. For a company with $10 billi- Investment Opportunity on in revenues and $600 million in COGS, even a 0.4% drop In the traditional corporate financing scheme, the finan- would equate to $2.4 million in savings[14]. ciers would be banks, focused on assessing the borrower’s historical credit, capital position, collaterals and guarantees of the companies in the supply chain that they are providing credit to. However, in the SCF view, financiers could be any investor or anchor buyer who is capable of assessing and ta- king the credit risk of the SCF product. Our solution will lower the barrier to entry for financiers in the global supply chain, and will allow institutional investors and accredited individual investors who are looking for short-duration, low risk and highly collateralized credit products to offer alter- native funding sources to SMEs in the future. Buyers and Suppliers: Working Capital Management By adopting our solutions, suppliers facing difficulties ob- taining affordable credit will now be able to access capital through banks or alternative sources. Capital-constrained buyers who rely on direct financing from a financial instituti- on can obtain trade-credit in addition to standard contracts to subsidize their costs of inventory financing and improve their working capital. Supply Chain Optimization Given the lack of transparency of the supply chain, we see The Eximchain platform also helps participants achieve that most SMEs rely on margin money deposits for bank global optimization, without the need for an unbiased deci- guarantees[12] or factoring houses to get loans. For example, sion maker, by allowing them to share the supply chain’s risk in China, banks normally require 5-20% of the loan amount and the potential benefit. Below is an example to quantify as a margin deposit depending on the credit risk of a SME. the value created for the supply chain based on a dynamic Fees charged by factoring houses can run as high as 15% contract between a supplier and buyer: of the loan amount every year[13]. With the combination of a PMSA and improved clarity on each deal, potentially Assume the price of a product for the end customer is 3

$125 per unit, the wholesale price paid by the retailer to the problems. We recognize that partnerships and a minimum manufacturer is $80 per unit, the fixed production cost for viable ecosystem (MVE), rather than a minimum viable manufacturer is $100k, the variable production cost per unit product, are crucial to successfully bringing a new block- is $35, and any item unsold in the end has a salvage value chain solution to market. of $20. This implies that the retailer’s marginal profit for selling a unit: $45, is smaller than the marginal loss: $60*. MVE: The smallest configuration of elements that can be The optimal order quantity depends on marginal profit and brought together and still create unique commercial value marginal loss. So the retailer’s optimal policy is to order 12k units for an average profit of $470k as shown on below MVE graph Profit vs. Order Quantity. If the retailer places this Combining a Blockchain, SDK, and Platform layer is ne- order, the manufacturer’s profit is $400k ((12k*($80-$35))- cessary to bring enough players to bear for a healthy mini- $100k=$440k). Note that in this case, the retailer assumes mum viable network. Eventually, others will be able to de- all of the risk of having more inventory than sales. velop solutions on our platform and make them immediately available to parties on the chain. In order for a blockchain to By using a Buy-Back Contract where the manufacturer work, the computers on the system must validate and agree offers to buy unsold items from the retailer for $55, the re- that a transaction was completed. This is accomplished by tailer’s marginal profit: $45, is now greater than its marginal agreeing on a sequence of ’blocks’ which each contain a set loss: $35, thus motivating the retailer to order more than of transactions. BTC, for example, incentivizes participati- average demand. In this case, the retailer has an incentive on in consensus by rewarding the first user to propose and to increase its order quantity to 14k units, for a profit of validate a new block at each canonical height. Important- $513.8k, while the manufacturer’s average profit increases ly, our design draws a distinction between nodes capable of to $471.9k. Thus, the total average profit for the two parties proposing and validating blocks. increases from $910.7k, (=$470,7k+$440k) in the sequenti- al supply chain to $985.7k (=$513.8k + $471.9k) when a Blockchain Buy-Back Contract is used. The Eximchain platform allows Technical challenges with blockchain technology have kept the retailer and manufacturer (i.e. the buyer and supplier) corporate players at bay. At the time of this writing, fur- to share the additional $75k in profit generated from this ther research and development for the full potential of this contract based on their consensus[8]. technology to be realized are required on the Bitcoin and Ethereum main nets. These are: Scalability- All smart contract code must be executed by all participants on the network. Privacy- All information on a public blockchain is visible to all participants to read. Faced with the choice of waiting for Ethereum to end the Ice Age[15] and release Casper in the coming years, or buil- *– Marginal profit/loss is the profit/loss of a firm or individual when one ding on recently produced viable solutions in this area, we additional unit is produced and sold. Marginal profit and loss determine the optimal ordering quantity- the ordering quantity that gives the maximized have chosen to leapfrog the Ethereum Roadmap by lever- profit for a firm or individual. aging Quorum, an enterprise focused version of Ethereum released by JP Morgan. Quorum allows us to leverage Con- Supply Chain Provenance stellation[16] and a minimal Ethereum client fork to offer: Using smart contracts and a token system, Eximchain can expand the supplier-buyer-financier model and help buyers Confidentiality- private transactions secured by the net- to gain visibility over the whole supply chain by interacting work with upstream suppliers. Performance- performance increases to handle the load Process Automation that complex private transactions place on the network. Eximchain helps to automate the negotiation and contrac- ting process through private communication channels. Both Consensus- permissioned consensus and governance ru- buyers and suppliers can stop wasting time on tracking the les to drive new economic incentives at the protocol level. supply chain process flow manually and will, instead, be ab- le to maintain an audit of the final agreed upon contract in Quorum, and therefore Eximchain, is designed to develop real-time through Eximchain’s trustless network and secured and evolve alongside Ethereum. Because it only minimally protocol. modifies Ethereum’s core, we are able to incorporate the majority of Ethereum updates quickly and seamlessly[17] 6. ECOSYSTEM . At the lowest level, we are modifying how the ethere- um client protocol enables consensus and permissions the Introduction blockchain. Our blockchain distinguishes between partici- Eximchain streamlines complex multi-party transactions pants that can validate blocks and participants that can in SCF by using blockchain to solve information asymmetry propose blocks. The latter allows us to both secure priva- 4

te transactions on the network and incentivize developers to participate in the evolution of our environment. Maker Nodes Maker Nodes are responsible for proposing blocks and their addresses are registered in the BlockVoting contract. The initial set of Maker Nodes is pre-configured in the gene- sis block and will be comprised of our initial token holders who we expect will add strategic value to the project, howe- ver once the network is established, this will begin to change as described below. Maker Nodes will have the responsibility of voting in a new Maker Node every governance cycle. Nodes can opt to become Maker Nodes if they are KYC approved and voted in by an agreed threshold of existing Maker Nodes according to the Quadratic Voting Governance Mechanism. Validator Nodes Consensus Validator Nodes help to secure the network and they are Rather than proof-of-work (mining), Eximchain uses a also registered in the BlockVoting contract. They are re- vote-based consensus algorithm that forks QuorumChain to sponsible for voting to determine which block will be the add a few governance rules. The governance mechanism is canonical hash at a particular height. Like Maker Nodes, included in the appendix for completeness; implementation the initial set of Validator Nodes is pre-configured in the ge- details can be found here. nesis block.[18] Settlement Finality Registered Voter Nodes The recent development on Istanbul BFT (Byzantine Fault Registered Voter Nodes tie real world identities of nodes. Tolerance), which will be incorporated into Quorum, ensures All nodes (including Maker Nodes, Validator Nodes, and settlement finality and speeds up settlement time to under 1 other network participants) are required to register to vote second from the current tens of seconds to tens of minutes. in order to participate in network governance mechanism. Registered Voter Nodes will have the responsibility of voting 7. GOVERNANCE out a Maker Node every governance cycle and may be voted The reason why Eximchain chose to use Quorum is becau- in as Maker Nodes once the network is established. se the current proof of work-based blockchains like Bitcoin and Ethereum are not ready for enterprise-grade solutions Governance Analysis (details to be described under Appendix- Motivation). We It is important to consider the evolution of the system and believe the incentives of developer pools, clients, and corpo- stakeholders in the face of dishonest nodes (traitors) and in- rates contributing value to the ecosystem must be aligned centive driven collusion. at the protocol level. This governance model can be applied to other public, permissioned blockchains that are looking Recent developments in permissioned distributed ledger to address similar issues. technology (DLT) have enabled us to design novel consensus rules that allow for a fundamental change in the underlying Our proposed quadratic voting-based governance model incentive mechanism and actors that underlie the consensus allows for a series of checks and balances between nodes. We protocol of permissioned, public chains. analyze how a voting system-based consensus model draws a clear line between safety and liveness: the percent vote re- The desired outcome is to keep a cohort of dishonest no- quired for consensus trades safety for speed. We analyze the des from gaining control of the block proposal mechanism counter balance of power between actors in the system, pro- or forking the chain, while bootstrapping a cohort of honest posing a model that discourages collusion and analyzes the nodes with vested interest to protect the network and punish robustness of the protocol to keep block proposers and vali- potential traitors. In the face of an undefeatable adversary dators in check under: independent choice, coordinated leading to mechanism failure we analyze the time, collusion, choice, and a bribing adversary. We conclude by ana- and resources required for such an attack. lyzing several failure modes of the system and subsequent recovery. While not strictly necessary for the purposes of this ana- lysis, it is important to keep in mind the real world identities Voting Smart Contract[17] behind these nodes will largely represent corporate incenti- ves in the global supply chain and development communi- Quorum Consensus Process Flow [17] ties. Ultimately, we believe the incentives of these stakehol- ders must be aligned at the protocol level. It follows that Quorum Consensus Block Creation [17] developers have a vested interest and duty to secure the in- frastructure their applications, clients and corporates rely Quorum Consensus Block Voting [17] on. Similarly, clients and end-users have a vested interest 5

in ensuring honest development that improves the ecosys- years and coordination of 683 parties. Each must be ca- tem, as well as a vested interest in punishing freeloaders on pable of convincing honest nodes representing the developer the network. The latter reflects our view that Maker No- pools, clients, and corporates that they are contributing va- des should represent developer pools, motivating clients and lue to the system throughout the duration of the attack to end-users to become Registered Voter Nodes to participate keep any of them from being voted out at each governance in governance. cycle and extending the minimum attack period. In all cases, while attacks on the network yield several fai- Undefeatable Bribing Adversary lure cases, the confidentiality of private data stored on the Assume an undefeatable bribing adversary who by a set of blockchain is preserved. That is to say, protocol or gover- highly unlikely events, knows exactly which nodes will pro- nance failure does not necessarily entail breach of security pose the next sequence of n blocks and the price required to private data as long as private keys remain secure. to bribe each one to propose a tampered block. Assume all nodes will commit treason for a price, the undefeatable bri- Adversary under independent choice bing adversary has resources k at his disposal, and is driven The consensus algorithm and governance is not fully open by profit incentive. We analyze the cost of such an attack, to any client issuing commands. Under independent choice, m, over a sequence of n blocks. stakeholders have no incentive to deviate from the protocol as their attack would be defeated by the network. Nodes Let b[1], b[2], . . . , b[n] be nonnegative integers that deno- cannot single-handedly confirm the proposed block at each te the price required to bribe each block-proposer to confirm canonical height. Maker Nodes who deviate from the pro- the proposed block at each canonical height, over n blocks. tocol would reveal their intentions to the network, placing A k-strategy is a sequence that allows the adversary to con- themselves at risk of being voted off the Maker Node pool in trol block proposal over a sequence of n blocks for a price m. the next governance cycle. Both Maker Nodes and Validator Nodes have a history of who voted them into the system, by 1 ≤ b1 < s1 < b2 < s2 < .... < bm < sm ≤ n, a majority Maker Node vote and by invitation respective- ly; in this way the reputation of known traitors is linked to where m is a positive integer less than or equal to k. The other nodes on the system (and real world entities) visible profit corresponding to the strategy is the total reward ob- to anyone in the voting smart contract. tained if you bribe the network for b1, and profit s1 from controlling a fixed number of transactions for one block, bri- Adversary under coordinated choice bing the network for 2 blocks at b2, profiting s2 from con- Assume an undefeatable, coordinating, racket of dishonest trolling a fixed number of transactions for two blocks, and so nodes who by a series of governance cycles gain control of forth. If the adversary can calculate the profit from control- a majority of block maker nodes that would allow them to ling each proposed block p[1], . . . , p[n] it is trivial to design control block proposal to deviate from the protocol. The an algorithm for the adversary to compute a k-strategy with consensus algorithm is not fully open to any client issuing maximum profit. commands; taking control of block maker governance requi- res traitors to be voted into the block maker pool. Governance Level Attack A similar k strategy could be designed to control gover- Assume all nodes in the initial block maker pool of size k, nance over n vote cycles at cost m to an undefeatable bribing under irrational choice, consecutively vote into the pool: an adversary. undefeatable coordinating racket of dishonest block maker nodes of size, m, over a sequence of n governance cycles. We Our governance model is unique in that nodes have the analyze the cost, coordination and time required to launch responsibility of periodically voting out Block Maker nodes an optimal attack and fool the network. counterbalanced by block maker nodes being able to appoint a replacement. We make an analogy to Block Makers as Let b[1], b[2], . . . , b[n] denote the nodes added to the generals and other nodes registered to vote as soldiers to block maker pool and s[1], s[2], . . . , s[n] the nodes removed reason in laymen terms. You can say that governance level from the block maker pool over n governance cycles length attacks are part of the protocol and built into the system T. A k-strategy is a sequence that allows the coordinated to discourage majority collusion. Under similar assumptions adversaries to control block proposal after a sequence of n as above, we expect generals (Block Makers) to appeal to governance cycles by voting in m nodes without replacement. soldiers (registered voters) to defend their status. A coalition that controls over 21 of tokens could vote out any general (b1, s1) → (b2, s2) → ..... (bm, sm) ; m ≤ n, they chose, we expect this to incentivize generals to spin up soldiers in their control to defend their status the only way where worst case, m=n and m > 13 k, assuming a traitor the protocol allows: by voting out another general not in is added each cycle and an honest node is removed. If the their coalition, simultaneously improving network security. adversaries can assure the above, it is left up to the reader to see coordination of the attack requires collusion of 13 k Governance Hardening nodes over a minimum attack duration, P = 31 kT before Assuming an initial token supply of X tokens, we analyze our strong assumption of k honest initial nodes fails. the case where an adversary owns X-1 tokens and is att- empting to gain control of the governance mechanism. We For a concrete example, k= 2048, T = 2 weeks , a perfectly analyze the minimum cost of controlling a sequence of n coordinated attack to control block proposal would take 26 governance cycles with length T of an undefeatable attack 6

strategy. We analyze the cost and time required from an be realized in the infancy of the network. undefeatable adversary to maintain control of the network. Our governance model implements quadratic voting ma- Quadratic Voting king the cost of each v votes for a governance decision cost v 2 Nodes can make v votes for a governance decision by pay- tokens, all tokens paid are then redistributed equally among ing v 2 tokens, from there it’s just a majority vote. All funds the voting pool. To assure control of the first governance raised are returned by giving back to each individual in the cycle, the adversary must place 2 votes at a cost v[1]2 = 4 voting pool,1/(voting pool-1) of the funds paid by the vo- tokens, controlling X- 5 of the token supply in the second ter. The theory is that if someone gains x from a decision cycle, to assure control of the next cycle, the adversary must being made, and each vote has a probability p of being pi- place 6 votes at cost v[2]2 = 36 to beat the possible 5 votes votal, then they have the incentive to keep buying votes for against his governance decision; this can continue to scale as long as the price of the next vote is less than px. Because this way until v[k] where the marginal cost of k nodes vo- the total price of v votes is v 2 , and we know from calculus ting against the adversary increases: that the derivative of v 2 is 2v, users will have the incenti- ve to keep spending tokens until 2v > px; hence, they will (votes1, cost1) → (votes2, cost2) ... → (votesn, costn) spend v = px/2 tokens. You can see from this math that the number of tokens that a voter buys should be proportional Optimal Defense : (1, 1) → (4 , 4) → (25, 25) → (676, to x, ie. the amount that they gain from the decision being 676) → (k-1 , k-1) made. Hence, the number of votes that a voter makes should actually reflect the strength of their preference, and not just Optimal Attack: (2, 4) → (5 , 25) → (26, 676) → (677, which option they prefer.[19] 458329) → (k, k2 ) Finite Attack Window Sequence of optimal voting decisions for an attacker would It would be irresponsible of us to claim undefeatable ad- cost: versaries are unlikely in practice* or claim that the proto- col will defend the network indefinitely, rather we choose to 1 + 4 + 25 + 676 + 458329 + 210066388900 + ... + v[k]2 < err toward practical guarantees that ensure network safety X − 1/2 by attempting to bound the minimum attack period and cost of an undefeatable adversary. It follows shortly that By correctly parameterizing the system it is possible to given an unbounded amount of time a determined enough prove an adversary can control a bounded number of gover- undefeatable adversary would eventually be able to launch nance cycles that can be defeated by k honest nodes. a successful attack. The only way to hedge against this is to Counter Measures ensure a finite possible attack period in a similar time scale to the minimum attack period of an undefeatable adversary. The premise of attack from a bribing adversary is a valid In 100 years, the voting contract will suicide halting all fur- concern if we are to expect the value of the network to scale ther change of state in the network, triggering open-source as the square of connected users [Metcalf’s Law]. Perhaps licences on any remaining proprietary portions of our soft- unsurprisingly, distributed concentrations of wealth and a ware. smaller block size discourage bribing people to deviate from the protocol for profit incentive. In practice, the non deter- ministic choice of block proposers using random timeouts limits the calculation of such a k-strategy by an undefeata- ble bribing adversary. The cost of the k-strategy to launch a governance or protocol level attack scales inversely to the percentage of validators required for block confirmation and number of nodes on the network. Our proposed governance model allows for a series of checks and balances between no- des. We are experimenting on mechanism design trade offs on our testnet and will make our results public in the coming year. *For example a single blockchain enabling Export/Import (or any other Bootstrap Mechanisms value exchange for that matter) could become a military or terrorist target. QVEC- Quadratic Voting on Ethereum Chain 8. APPLICATION Peacekeepers At a higher level, we are creating a smart contract SDK. These are k = safety factor * initial token holders partici- The smart contract SDK layer will allow developers to build pant nodes originally launched as block makers to defend the applications from basic components, accelerating develop- network and are registered in the voting contract as part of ment of future SCF solutions. the bootstrap mechanism; these nodes participate in con- sensus but are forced to abstain at each governance cycle. We want to enable the engineering of monetary systems They will be first to be voted off the block proposer pool and make implementation of monetary theory, economics to make room for new Maker nodes, allowing the ecosystem and law accessible to developers. The first step is enab- to become fully decentralized over time while ensuring the ling the implementation of systems that can be reduced minimum attack period of an undefeatable adversary cannot to modeling the strategic interaction between two or mo- 7

re players,specifically in situations containing a set of ru- Strategy Computation Machine les, outcomes, and desired objectives. Participants can build We consider the design of Computational Game-Theoretic supply chain applications on top of the Eximchain smart Frameworks - machine games, where we replace strategies by contract SDK: Turing machines (smart contracts)[20]. For example, given the state of a game, a strategy computation machine may Our aim is to show developers how our SDK easily sup- return a distribution over possible actions, the complexity of ports complex multi party transactions, enabling them to the computation depends on implementation of the machine build applications and other supply chain finance products but the interface can be clearly defined. on top of our ecosystem. Smart Contract SDK Platform development status and plan Financing: Through instant and transparent proof of or- der verification on blockchain, suppliers seize early payment Q3-Q4 2017 discounts to improve working capital and buyers seize credit arbitrage opportunity. Signed letter of intents from 9 companies including SMEs, a listed company in China and a cross-border e-commerce Procure-to-Pay: Participatants can maintain an audit of platform. the final agreed upon contract in each step of the procure- to-pay process in real-time through a trustless network and A Strategic Cooperation Agreement with Guiyang High- secured protocol. tech Industrial Development Zone Management Commission and Guiyang Big Data Development and Management Com- Sourcing: Sourcing platforms or rating agents can use the mission, China. Proof of Existence (PoE) functionality- document time stam- ping, verifying document integrity, demonstrating data ow- Current Stage (Q4 2017- Q1 2018) nership without revealing data, to provide independently ve- rifiable reputation data. Governance hardening Inventory Management: By sharing real-time demand in- First POC in Supply Chain Sourcing formation cross the supply chain, participants can make bet- ter inventory plans and automate the reconciliation process Large scale network testing across different ledgers and systems to save time and reduce cost. Token Generation Event Supply Chain Optimization: Using Multi-Party Dynamic In the next few months, Eximchain will launch several Contracts - a combination of private channels and a confi- testnets and onboard SMEs to execute smart contract ba- gurable state machine to coordinate a confidential bi-level sed solutions through the network. stackelberg SCF negotiation (as described in Appendix). Release 2.0 (Q2 to Q3 2018) Supplier Management: Using smart contracts and a token system, Eximchain can help buyers to gain visibility over the Native Token Swap whole supply chain by interacting with upstream suppliers but still maintain privacy on each transaction level. Mainnet Launch Risk Mitigation: Eximchain enables dynamic, real-time First Governance Cycle monitoring of supply chain processes to provide visibility into the whole supply chain’s operations. Participants can Second POC in Supply Chain Financing or Inventory/Logistics manage risks in one place. Management First SDK Release Private channels Release 3.0 (Q4 2018 to Q1 2019) Series of confidential two way channels coordinated by the FSM (Finite State Machine). Second SDK release Second Governance Cycle FSM Coordinates rules and implements mechanisms modeling Third POC in Supply Chain Financing or Inventory/Logistics the strategic interaction between two or more players. Takes Management the shape of coordinating information exchange between a number of two way private channels through a semaphore Over the next 15-18 months, our platform will enable par- notion and embedded mechanism design. ticipants in the global supply chain to launch their own sup- ply chain management tools using our smart contract SDK. 8

9. TEAM AND ADVISORS Mainland China. Can Kisagun is a co-founder at Enigma- a data-driven Team crypto investment platform that raised $45m during Token Hope Liu, CEO of Eximchain, B.A. from Peking Universi- Generation Event in 2017. Previously Can worked at McKin- ty and MBA from MIT, handled cross-boarder transactions sey & Company for 3 years focusing primarily on finance and at UBS Beijing, Hong Kong and Singapore for 6.5 years. She banking engagements. Can holds an MBA from MIT Sloan is the Lab Lead of the North America Blockchain Associa- and BS in Industrial Engineering from Northwestern Univer- tion and has been working on Eximchain project from MIT sity. Prior to Enigma, Can was a co-founder of Eximchain. Media Lab since 2015. She led the team to win the grand Peter Missine established himself as an independent stra- champion of Boston Seagull Entrepreneurship Competition tegy consultant after a few years at McKinsey & Co. Prior and has been featured by mainstream blockchain medias in to his MBA at MIT Sloan, Peter spent five years as an in- U.S. ternational commodity trader on three continents for Louis Juan Huertas, CTO of Eximchain, B.S.in Computer Science Dreyfus Commodities (LDC). At LDC he also worked with from MIT. He started coding at age 13 and has been a Tech- international trade flows, supply chains, transfer pricing and nology Consultant for startups since he was 18 years old. He anti-dumping. built a cryptocurrency enabled game to play and distribute Manmeet Singh is currently the managing partner of Block- cryptocurrency anonymously during his junior year in colle- seed Venture, Expert-in-Residence at Chinaaccelerator and ge at MIT. the Investment Advisor for Nanjing Municipal Government. James Xu, Architect of Eximchain. He worked in IBM for He previously run an Asia focused financial advisory firm 14 years holding various positions across the globe as enter- and was the China Representative of ICIC Bank. prise packaged software offering manager, Delivery Project William Peckham is the Managing Partner of Proteus Executive of a team 100+ across 14 time zones, and Associa- Growth, a cross border strategy consulting firm based out te Partner managing key account in China. He has extensive of Beijing, as well as China Operating Partner of Higher- exposure in supply chain management, retail, CPG and glo- Order VC, a crypto currency fund. He has served as an ad- bal trade domains. He also co-founded a startup, building a visor to numerous foreign tech startups in China including EV car sharing platform. Robomed, Enigma, Rogue Initiative, INS Ecosystem, and Jia Zhang, Business Analyst of Eximchain. Jia has been Quintype. in supply chain field since 1994. She spent 7 years in ICBC managing global trade finance and international settlement. 10. TOKEN AND PARTICIPATION After that, she acted as the Chief Representative of MS Tex- Eximchain Platform Native Tokens (Eximcoins) will be tiles in China for almost 10 years, managing local supplier used to pay network fees, validate state changes, and execu- relationships, goods inspection, and logistics arrangement. te governance. Eximcoins will also be used to access appli- She speaks Chinese, English and French. cations built on the Eximchain network developed with the Louis Lamia, Director of Engineering and Infrastructure Eximchain SDK. at Eximchain, B.S.in Computer Science from MIT. He was a Software Development Engineer at Amazon Web Services Token Structure on the Elastic File System team for over 2 years involved in various projects, most notably encryption-at-rest. Because Eximcoins will be native to the permissioned fork Douglas Sanchez, Director of Product at Eximchain, B.S.in of the Quorum blockchain which will be integrated into the Computer Science from MIT. Before joining Eximchain, he Eximchain platform and will only be available when the was leading industrial, product and brand design for Tulip, platform is deployed for commercial use (Mainnet Launch) a manufacturing app platform startup in Boston. (which is expected to occur in Q2 2018), Eximchain plans to sell a digital token native to the Etherium blockchain that Advisors will be a precursor to Eximcoins (EXC Tokens) (the sale of Ramble Lan is the president of NABA (North America EXC Tokens, ”Token Sale”). EXC Tokens will be ERC-20 Blockchain Association), Chief Architect of the Regulatory compatible tokens distributed on the Ethereum blockchain Sandbox in Guiyang, China, and the Chairman of Supply pursuant to a related ERC-20 smart contract. Shortly befo- Chain Blockchain Association in Fujian, China. He is also re Mainnet Launch, EXC Tokens will be permanently frozen the CEO of SwftCoin (www.swftcoin.com). and not capable of being used for any purpose, in prepara- Tiger Zhong is the CEO of Trademanger (www.Tradema- tion for being converted into Eximcoins. Each EXC Token nager.com)- a platform that provides big data services for will be automatically converted to one Eximcoin at Mainnet small and medium sized business in global trade with over Launch. 50k active users. He has experience providing services to Total supply of EXC Tokens: 75,000,000 small and medium size business in China for over 16 years. Fundraising target: US$20m equivalent Catherine Dai is the Founder and Owner of BoaoTech and HongKong Boao, Co-Founder and Shareholder of Gibo- Star Int’l. She plays Angel/VC role for 16 companies ran- Token Distribution ging from TMT, Biopharmaceutical, Heavy industry, to Gar- Token distribution: ment, Catering, and Entertainment Industry in China. Sold during Token Sale: 60,000,000 Daniel Wang is a China based, U.S. trained executive with Bonus Account (including airdrop): 15,000,000 extensive transactional background. He is currently the di- rector of investment of a Fortune 500 company, and used to Sold during Token Sale: No more than 60,000,000 EXC are be an attorney practicing in Silicon Valley, Hong Kong and to be sold to buyers. Each token will be sold for the ether 9

equivalent of approximately US$0.33. There will be no bo- nus or discount available to contributors during the Token The Eximcoins distributed to the reserve account will be Sale. locked and will be unlocked at the end of the 2-year period beginning at Mainnet Launch. Bonus Account: 15,000,000 EXC Tokens from the total EXC Token supply will be allocated to a bonus account. Approximately 10% of the EXC Tokens in the bonus ac- Token Sale Timeline count will be allocated for an Airdrop, as described below. Token Sale: Q1 2018 The remaining EXC Tokens in the bonus account will be Mainnet Launch and start token listing process: Q1-Q2 2018 reserved for distribution by Eximchain in its discretion to List tokens on multiple exchanges: Q2-Q3 2018 marketing and community partners during the EXC Token generation event and participants in the Eximchain Bounty Program. Use of Funds Eximchain intends to use the funds from the sale of EXC Airdrop: In conjunction with the private sale, Eximchain Tokens at the Token Sale for the following primary purposes: plans to conduct an airdrop of EXC Tokens to persons who were placed on a whitelist of persons who expressed an inte- Platform Development: This includes maintaining compe- rest in purchasing EXC Tokens but who did not participate titive salaries for top talent in China and the U.S. paying for in the private sale. Eximchain will deliver a small quantity software hosting, accelerated development of legacy system of EXC Tokens (valued at about US$200) to each such per- integrations and acquisition of hardware. The hardware and son for free. hosting will be offered to anchor buyers and that will need stand- alone computers to manage their respective supply chains on a global scale. Conversion of EXC Tokens to Eximcoins Shortly before Mainnet Launch, the EXC Tokens will be Build the Supply Chain Ecosystem: This includes effort to permanently frozen and not capable of being used or trans- onboard SMEs through incentivizing early stage users and ferred for any purpose, in preparation for being converted on-going business development to onboard suppliers, buyers into Eximcoins. At Mainnet Launch, all EXC Tokens will and financiers. Eximchain will reserve a proportion of tokens be automatically converted to Eximcoins on a one-for-one for this purpose in a secured wallet to ensure fair distribu- basis. An additional 75,000,000 Eximcoins will also be crea- tion. ted and distributed at Mainnet Launch, as follows: Operational Expenses: This includes basic operational costs o 30,000,000 Eximcoins will be allocated to a promotional such as office space, legal costs to pursuing and obtaining re- account that Eximchain will manage to promote the use of quired licenses, and security measures. the platform by SMEs and early adopters to be selected by Eximchain in its discretion over a 10-year period. It is im- However, Eximchain may in its discretion use those pro- portant that Eximchain token holders plan to use our sup- ceeds for any purpose, whether or not consistent with the ply chain solutions and participate in the early stages of the foregoing. Eximchain makes no undertaking, representation Eximchain governance mechanism. Eximchain will distribu- or warranty in respect of its use of such proceeds. te Eximcoins from the promotional account as an incentive for early stage participants who will use the Eximchain plat- Participation form to conduct contract negotiations, build supply chain Eximcoins will only be minted in the genesis block, asso- products and participate in network governance. SMEs in- ciating them with the wallet addresses of the holders of EXC terested in applying for tokens to cover development and Tokens. EXC Tokens will only be minted while the contract prototype phases will need to sign a Commitment Letter or is active. Letter of Intent to qualify. The private sale registration was closed on Jan 11th 2018. o 30,000,000 Eximcoins will be distributed to the Exim- In conjunction with the private sale, a total US$500k worth chain founding team and advisers. of EXC Tokens are available for airdrop. Only participants on the whitelist can join the airdrop process. The Token Sa- o 15,000,000 will be allocated to a reserve account that le KYC process is scheduled to continue through to March Eximchain will manage in its discretion to assure there are 2018. All participants in the Token Sale, including airdrop sufficient Eximcoins in circulation for use on the platform. participants, must complete our KYC-AML process. The whitelist registration was closed on Jan 16th 2018. The Eximcoins allocated to the promotional account will be locked (incapable of transfer or use) and will be unlocked We invite engagement and dialog on our business model during the 10-year period beginning at Mainnet Launch in and our ecosystem design. The opportunity that these tech- equal monthly allotments. nologies will unlock for businesses of all sizes is remarkable and we need public involvement to fulfill that mission. The Eximcoins distributed to Eximchain’s team and ad- visers will be locked and will be unlocked during the 4-year To become a part of our community, give us feedback on period beginning at Mainnet Launch in equal monthly al- the whitepaper or just to find out more about Eximchain: lotments. 10

Visit our website at www.eximchain.com . acknowledge that this Whitepaper, the undertaking and/or the completion of the Eximchain Initial Token Sale, or future Join our Telegram channels. trading of the EXC tokens on any cryptocurrency exchange, shall not be construed, interpreted or deemed by you as an Follow us on Twitter. indication of the merits of the Eximchain , the EXC tokens, the Eximchain Initial Token Sale and the Eximchain Wallet Or email us at [email protected] (each as referred to in this Whitepaper); (e) the distributi- on or dissemination of this Whitepaper, any part thereof or any copy thereof, or acceptance of the same by you, is not 11. LEGAL CLARIFICATION prohibited or restricted by the applicable laws, regulations Eximcoins are utility tokens whose entire value derives or rules in your jurisdiction, and where any restrictions in from the services provided by the Eximchain platform in relation to possession are applicable, you have observed and exchange for holding or consuming the tokens, as detailed complied with all such restrictions at your own expense and above. They are not intended for speculation and hold no without liability to Eximchain ; (f) you agree and acknow- claim to intellectual or other property or cash flows. They ledge that in the case where you wish to purchase any EXC grant no right to participation in the company, and no claim tokens, the EXC tokens are not to be construed, interpre- in decision making over company assets or strategy. There ted, classified or treated as (i) any kind of currency other is no promise of value or claim on revenue associated with than cryptocurrency; (ii) debentures, stocks or shares issu- EXC other than that derived from platform usage. In short, ed by any person or entity (whether Eximchain ) (i) rights, EXC are not securities. Also, the estimations under Value options or derivatives in respect of such debentures, stocks for our Stakeholders are based on assumptions and there can or shares; (ii) rights under a contract for differences or un- be no guarantee that they will be achieved. Actual results der any other contract the purpose or pretended purpose of may vary substantially from the figures shown. which is to secure a profit or avoid a loss; (iii) units in a collective investment scheme; (iv) units in a business trust; (v) derivatives of units in a business trust; or (vi) any other DISCLAIMER OF LIABILITY security or class of securities. (g) you are fully aware of and To the maximum extent permitted by the applicable laws, understand that you are not eligible to purchase any EXC regulations and rules, Eximchain shall not be liable for any tokens if you are a citizen of China, or a citizen, resident indirect, special, incidental, consequential or other losses of (tax or otherwise) or green card holder of the United States any kind, in tort, contract or otherwise (including but not of America; (h) you have a basic degree of understanding limited to loss of revenue, income or profits, and loss of use of the operation, functionality, usage, storage, transmission or data), arising out of or in connection with any acceptance mechanisms and other material characteristics of cryptocur- of or reliance on this Whitepaper or any part thereof by you. rencies, blockchain-based software systems, cryptocurrency wallets or other related token storage mechanisms, block- NO REPRESENTATIONS AND WARRANTIES chain technology and smart contract technology; (i) you are fully aware and understand that in the case where you wish Eximchain does not make or purport to make, and hereby to purchase any EXC tokens, there are risks associated with disclaims, any representation, warranty or undertaking in Eximchain and the Distributor and their respective business any form whatsoever to any entity or person, including any and operations, the EXC tokens, the Eximchain Initial To- representation, warranty or undertaking in relation to the ken Sale and the Eximchain Wallet (each as referred to in truth, accuracy and completeness of any of the information the Whitepaper); (j) you agree and acknowledge that neit- set out in this Whitepaper. her Eximchain nor the Distributor is liable for any indirect, REPRESENTATIONS AND WARRANTIES BY special, incidental, consequential or other losses of any kind, YOU in tort, contract or otherwise (including but not limited to loss of revenue, income or profits, and loss of use or data), By accessing and/or accepting possession of any informa- arising out of or in connection with any acceptance of or tion in this Whitepaper or such part thereof (as the case reliance on this Whitepaper or any part thereof by you; and may be), you represent and warrant to Eximchain as follows: (k) all of the above representations and warranties are true, (a) you agree and acknowledge that the EXC tokens do not complete, accurate and not misleading from the time of your constitute securities in any form in any jurisdiction; (b) you access to and/or acceptance of possession this Whitepaper agree and acknowledge that this Whitepaper does not con- or such part thereof (as the case may be). stitute a prospectus or offer document of any sort and is not intended to constitute an offer of securities in any jurisdicti- on or a solicitation for investment in securities and you are CAUTIONARY NOTE ON FORWARD LOO- not bound to enter into any contract or binding legal com- KING STATEMENTS mitment and no cryptocurrency or other form of payment is All statements contained in this Whitepaper, statements to be accepted on the basis of this Whitepaper; (c) you agree made in press releases or in any place accessible by the pu- and acknowledge that no regulatory authority has examined blic and oral statements that may be made by Eximchain or approved of the information set out in this Whitepaper, or their respective directors, executive officers or employees no action has been or will be taken under the laws, regulato- acting on behalf of Eximchain or the Distributor (as the ry requirements or rules of any jurisdiction and the publica- case may be), that are not statements of historical fact, con- tion, distribution or dissemination of this Whitepaper to you stitute ”forward-looking statements”. Some of these state- does not imply that the applicable laws, regulatory require- ments can be identified by forward looking terms such as ments or rules have been complied with; (d) you agree and äim”, ”target”, änticipate”, ”believe”, ”could”, ëstimate”, ëx- 11

pect”, ı̈f”, ı̈ntend”, ”may”, ”plan”, ”possible”, ”probable”, ”pro- ther, Eximchain disclaim any responsibility to update any ject”, ßhould”, ”would”, ”willör other similar terms. Howe- of those forward-looking statements or publicly announce ver, these terms are not the exclusive means of identifying any revisions to those forward-looking statements to reflect forward-looking statements. All statements regarding Exim- future developments, events or circumstances, even if new chain’s financial position, business strategies, plans and pro- information becomes available or other events occur in the spects and the future prospects of the industry which Exim- future. chain is in are forward-looking statements. These forward- looking statements, including but not limited to statements NO ADVICE as to Eximchain’s revenue and profitability, prospects, fu- No information in this Whitepaper should be considered ture plans, other expected industry trends and other mat- to be business, legal, financial or tax advice regarding Exim- ters discussed in this Whitepaper regarding Eximchain are chain, the EXC tokens, the Eximchain token sale and the matters that are not historic facts, but only predictions. Eximchain business (each as referred to in the Whitepaper). These forward-looking statements involve known and un- You should consult your own legal, financial, tax or other known risks, uncertainties and other factors that may cau- professional adviser regarding Eximchain and their respec- se the actual future results, performance or achievements tive businesses and operations, the EXC tokens, the Exim- of Eximchain to be materially different from any future re- chain token sale and the Eximchain business (each as refer- sults, performance or achievements expected, expressed or red to in the Whitepaper). You should be aware that you implied by such forward-looking statements. These factors may be required to bear the financial risk of any purchase include, amongst others: (a) changes in political, social, eco- of EXC tokens for an indefinite period of time. nomic and stock or cryptocurrency market conditions, and the regulatory environment in the countries in which Exim- chain conducts its respective businesses and operations; (b) NO FURTHER INFORMATION OR UPDATE the risk that Eximchain may be unable or execute or imple- No person has been or is authorized to give any informa- ment their respective business strategies and future plans; tion or representation not contained in this Whitepaper in (c) changes in interest rates and exchange rates of fiat cur- connection with Eximchain and its respective businesses and rencies and cryptocurrencies; (d) changes in the anticipated operations, the EXC tokens, the Eximchain token sale and growth strategies and expected internal growth of Exim- the Eximchain business (each as referred to in the Whitepa- chain ; (e) changes in the availability and fees payable to per) and, if given, such information or representation must Eximchain in connection with their respective businesses not be relied upon as having been authorized by or on be- and operations; (f) changes in the availability and salari- half of Eximchain token sale and the Eximchain business (as es of employees who are required by Eximchain to operate referred to in the Whitepaper) shall not, under any circum- their respective businesses and operations; (g) changes in stances, constitute a continuing representation or create any preferences of customers of Eximchain ; (h) changes in com- suggestion or implication that there has been no change, or petitive conditions under which Eximchain operate, and the development reasonably likely to involve a material change ability of Eximchain to compete under such conditions; (i) in the affairs, conditions and prospects of Eximchain or in changes in the future capital needs of Eximchain and the any statement of fact or information contained in this Whi- availability of financing and capital to fund such needs; (j) tepaper since the date hereof. war or acts of international or domestic terrorism; (k) occur- rences of catastrophic events, natural disasters and acts of NO OFFER OF SECURITIES OR REGISTRA- God that affect the businesses and/or operations of Exim- TION chain ; (l) other factors beyond the control of Eximchain ; This Whitepaper does not constitute a prospectus or offer and (m) any risk and uncertainties associated with Exim- document of any sort and is not intended to constitute an chain and their businesses and operations, the EXC tokens, offer of securities or a solicitation for investment in securi- the Eximchain token sale and the Eximchain business(each ties in any jurisdiction. No person is bound to enter into any as referred to in the Whitepaper). All forward-looking state- contract or binding legal commitment and no cryptocurren- ments made by or attributable to Eximchain or persons ac- cy or other form of payment is to be accepted on the basis of ting on behalf of Eximchain are expressly qualified in their this Whitepaper. Any agreement in relation to any sale and entirety by such factors. Given that risks and uncertain- purchase of EXC tokens (as referred to in this Whitepaper) ties that may cause the actual future results, performan- is to be governed by only the terms and conditions (T&Cs) ce or achievements of Eximchain to be materially different of such agreement and no other document. In the event of from that expected, expressed or implied by the forward- any inconsistencies between the T&Cs and this Whitepaper, looking statements in this Whitepaper, undue reliance must the former shall prevail. not be placed on these statements. These forward-looking You are not eligible to purchase any EXC tokens in the statements are applicable only as of the date of this White- Eximcoin Initial Token Sale (as referred to in this Whitepa- paper. Neither Eximchain nor any other person represents, per) if you are a citizen of China, or a citizen, resident (tax warrants and/or undertakes that the actual future results, or otherwise) or green card holder of the United States of performance or achievements of Eximchain will be as dis- America. No regulatory authority has examined or approved cussed in those forward-looking statements. The actual re- of any of the information set out in this Whitepaper. No such sults, performance or achievements of Eximchain may differ action has been or will be taken under the laws, regulatory materially from those anticipated in these forward-looking requirements or rules of any jurisdiction. The publication, statements. Nothing contained in this Whitepaper is or may distribution or dissemination of this Whitepaper does not be relied upon as a promise, representation or undertaking imply that the applicable laws, regulatory requirements or as to the future performance or policies of Eximchain. Fur- rules have been complied with. 12

RISKS AND UNCERTAINTIES Rr, and Manufacture’s wholesale price. To pursue the result Prospective purchasers of EXC tokens (as referred to in of the Stackelberg equilibrium, Retailer and Manufacturer this Whitepaper) should carefully consider and evaluate all select a best-reply policy, denoted by (4) and (5). risks and uncertainties associated with Eximchain, the EXC tokens, the Eximchain token sale and the Eximchain (each The model can be expressed as below: as referred to in the Whitepaper), all information set out in this Whitepaper and the T&Cs prior to any purchase of EXC tokens. If any of such risks and uncertainties develops into actual events, the business, financial condition, results of operations and prospects of Eximchain could be materially and adversely affected. In such cases, you may lose all or part of the value of the EXC tokens. 12. ACKNOWLEDGMENT We would like to express our gratitude to the many people who supported us as we wrote this paper: Nina Yan, Zach The Eximchain ecosystem involves a single shared block- Chen, Jessie Liu, Malak Alyousef, Louis Lamia, Steven Ri- chain and a smart contract SDK. Besides the model mentio- vera, John O’sullivan and Sawan Jindal. Thank you. ned above, our software architecture is tailored to implement many SCF products from basic components on our ecosys- 13. APPENDIX tem. Ultimately, developers can use our software to create variations on Stackelberg Strategies or their own multi party SCF products. Use Case Example *Multi-Party Dynamic Contracting MOTIVATION Inspired by the system proposed in “Optimal bi-level Sta- As of writing, large mining pools control and secure proof ckelberg strategies for supply chain financing with both capital- of work based blockchains like Bitcoin and Ethereum, as the constrained buyers and sellers”[21] and “A partial credit gua- systems shift toward transaction based economies, and proof rantee contract in a capital-constrained supply chain: Finan- of stake respectively, the financial incentive provided to de- cing equilibrium and coordinating strategy,”[22] below we vote a large amount of hash power or coin holdings to secure provide an example of a Dynamic SCF system that can be the network will drive decentralized networks into the palms created through the Eximchain platform. of a handful of mining pools. We believe it is important to distinguish the value these pools contribute to the chain by In the supply chain financing system shown, we formulate devoting resources to add security to the network, from the the interactions between the capital-constrained buyer (Re- value developers contribute by adding novel functionality to tailer), the capital-constrained supplier (Manufacturer) and the ecosystem. the financier (Bank) as privately executed smart contracts that coordinate a bi-level Stackelberg game. The Bank acts Recent developments in permissioned-DLT have enabled as the leader in setting the financing rate for the Manufac- us to design novel consensus rules that allow for a funda- turer and Retailer. The Manufacturer acts as the subleader mental change in the underlying incentive mechanism and to make the best response according to Bank’s decision and actors that underlie the consensus protocol of permissioned, to set the wholesale price for the Retailer at the same ti- public chains. me. The Retailer acts as the follower to respond to both the Bank and Manufacturer. We propose a system that grants developers, and mem- bers of our Token Generation Event a privileged position in the consensus protocol of the underlying blockchain, incen- tivizing and funding developer pools to add functionality to the system by allowing them to exclusively collect network rewards at the protocol level from state changes driven by Dapps on the network. Today*, Ethereum and Bitcoin stakeholders (as a whole) keep mining-rigs humming to the tune of $5M and $6M per Firstly, the leader (Bank) will evaluate the Retailer’s and day respectively. Bitcoin, touted as the prime example of the Manufacturer’s financing conditions (e.g., initial capital, successful dPoW is majority controlled by a handful of mi- bankruptcy risks and procurement/production quantities) ning pools, with no roadmap to alleviate the centralization and make optimal decisions to announce the interest rate of hash power. While the case is similar for Ethereum at Rr and Rm, respectively. Then, in response to the Bank, present, we will have to wait and see how their transition the subleader (Manufacturer) will simultaneously update his to security deposit based dPoS will affect the distribution decision and decide how much to charge the buyer for the of block proposers and whether token holders begin to bond wholesale price w. Acting as the follower, the Retailer de- their ETH (which brings up peripheral concerns about ille- cides how much to order according to Bank’s interest rate, gitimate initial token holders who will have a huge incentive 13

to raise as much ETH as possible). Hashrate Distribution - BTC March 2018 [1] Lotta Lind, Miia Pirttilä, Sari Viskari, Florian Schupp, and Timo Kärri. Working capital management in the automotive industry: Financial value chain analysis. Journal of purchasing and supply management, 18(2):92–100, 2012. [2] Benefits Beyond Treasury: How Supply Chain Finance Impact the Bottom Line. Citi Transaction Services. http://www.citibank.com/transactionservices/ home/about_us/articles/docs/citi_insights_scf_ updated.pdf/, 2012. [3] Peer Stein, Oya Pinar Ardic, and Martin Hommes. Hashrate Distribution - ETH March 2018 Closing the credit gap for formal and informal micro, small, and medium enterprises. 2013. [4] Blockchain Platform for Supply Chain Finance Launches in China, CFO Innovation Asia, 2017. [5] Small and Medium Enterprises (SMEs) Finance. The World Bank. http://www.worldbank.org/en/topic/ financialsector/brief/smes-finance/, 2015. [6] How Supply Chain Finance Can Close the Funding Gab for SMEs., 2016. [7] Mckinsey on Payments, Supply-chain finance: The emergence of a new competitive landscape, 2015. [8] David Simchi-Levi, Edith Simchi-Levi, and Philip Kaminsky. Designing and managing the supply chain: Considering a 66% attack is possible in both networks un- Concepts, strategies, and cases. McGraw-Hill New der collusion of less than 10 parties, it becomes clearer that York, 1999. the security of these ecosystems are largely hinged on the [9] Chang Hwan Lee and Byong-Duk Rhee. Coordination bounties available to these miners each day, none of whom contracts in the presence of positive inventory contribute to the codebase of the underlying system that financing costs. International Journal of Production enables the creation of this value. The result is that mining Economics, 124(2):331–339, 2010. pools have complete control over how the system evolves; in these networks, developers are second class citizens, con- [10] Henry Sanderson and Neil Hume (2014), Qingdao stantly at risk of forks in a network that their applications, fraud case taints commodity financing. Financial clients, and corporates rely on, forks that are ultimately de- Times. https://www.ft.com/content/ cided by the mining groups profiting from their innovation b3cc4dc4-86ba-11e4-9c2d-00144feabdc0?mhq5j= and market speculation. Perhaps due to the latter, these e3/, 2014. networks look more like security rackets every day, where [11] https://www.investopedia.com/terms/p/ each additional dollar of network incentive gives you less se- purchase-money-security-interest-pmsi.asp//. curity than the last dollar*. A new protocol is necessary to [12] http://www.investopedia.com/terms/b/ align network incentives with those of developers, clients, bankguarantee.asp//. and corporates. [13] Based on the prevailing rates on margin money *consider that only 5 months ago Ethereum miners were getting paid deposit for bank guarantee and guarantee fee offered $500k/day to secure the network compared to today’s $5M/day in China . [14] Bob Dyckman. Integrating supply chain finance into Consensus + Token FAQ the payables process. Journal of Payments Strategy & The token functions to keep the network incentivized to Systems, 3(4):311–319, 2009. participate in consensus and governance. Our native token’s [15] Ethereum to end the Ice Age. https: primary use is to pay gas, a mechanism used to incentivi- //github.com/ethereum/go-ethereum/commit/ ze Block makers to process your transactions. Further all 71d32f54f70917c53fd3a691cface3bc73ffa1b7/. nodes registered to vote can make ’passive income’ (don’t [16] Quorum allows leverage Constellation. https: particularly love the term) by participating in governance //github.com/jpmorganchase/constellation/. system via quadratic voting. 1000 EXC min required to par- [17] https://www.jpmorgan.com/country/US/EN/Quorum/. ticipate in consensus mechanism, as a masternode, but note [18] https://github.com/jpmorganchase/quorum/wiki/ that this is not exactly PoS: nodes can be voted off the con- QuorumChain-Consensus/. sensus mechanism irrespective of their wealth via quadratic [19] https://www.reddit.com/r/ethereum/comments/ voting governance. Masternodes are required to have 1000 453sid/empirical_cryptoeconomics//. EXC collateral, a dedicated IP address, and be able to run [20] Joseph Y Halpern and Rafael Pass. Game theory with 24 hours a day without a more than a 1 hr connection loss, costly computation. arXiv preprint arXiv:0809.0024, they claim 100% of the transaction fees in blocks they make. 2008. [21] Nina Yan, Hongyan Dai, and Baowen Sun. Optimal 14. REFERENCES bi-level stackelberg strategies for supply chain 14

financing with both capital-constrained buyers and sellers. Applied Stochastic Models in Business and Industry, 30(6):783–796, 2014. [22] Nina Yan, Baowen Sun, Hui Zhang, and Chongqing Liu. A partial credit guarantee contract in a capital-constrained supply chain: Financing equilibrium and coordinating strategy. International Journal of Production Economics, 173:122–133, 2016. 15