Token in Real-Life: and Incentives Design for Insolar’s Network

Henry M. Kim Marek Laskowski Michael Zargham Hjalmar Turesson Schulich School of Business Schulich School of Business BlockScience Schulich School of Business York University York University Oakland, USA York University Toronto, Canada Toronto, Canada [email protected] Toronto, Canada [email protected] [email protected] [email protected]

Matt Barlin Danil Kabanov BlockScience Insolar Oakland, USA Zug, Switzerland [email protected] [email protected]

Abstract— The study of how to set up cryptocurrency incentive a decentralized way, an alternative scheme for incentives is mechanisms and to operationalize governance is token economics. needed. Given the $250 billion market cap for , there is compelling need to investigate this topic. In this paper, we present For instance, the network design mitigates the facets of the token engineering process for a real-life 80-person “double spend” problem where a sender simultaneously sends Swiss blockchain startup, Insolar. We show how Insolar used money that they cannot fully cover to two different recipients systems modeling and simulation combined with cryptocurrency [2]. The network ensures that one recipient is designated to be expertise to design a mechanism to incentivize enterprises and the rightful recipient and nullifies the money transfer to the individual users to use their new MainNet public blockchain other. A third party that is neither the sender nor recipient network. The study showed subsidy pools that incentivize verifies this designation. That third party must be incentivized application developers to develop on the network does indeed have to verify accurately and not collude with the sender or the the desired positive effect on MainNet adoption. For a startup like recipient. Roughly every ten minutes, the network selects a Insolar whose success hinge upon how well their model “miner” to add the next block to the Bitcoin blockchain, where incentivizes various stakeholders to participate on their MainNet the block is comprised of all transactions logged since the last network versus that of numerous alternatives, this token block formation that are verified by the miner to not be “double economics simulation analysis provides invaluable insights. spent.” Currently, a miner receives 12.5 , or roughly $125,000 each time a new block is added. Keywords—Token economics, token engineering, cryptocurrency, blockchain, agent-based modeling and simulation We have thus simplistically described Bitcoin’s token (or crypto) economic model. The model lays out the mechanism by I. INTRODUCTION which the Bitcoin economy is sustained: Senders and recipients The core premise of blockchain is straightforward: rather are assured that money cannot be transferred from insufficient than relying upon a trusted intermediary to proprietarily funds and third-party miners are incentivized by the promise of maintain one ledger of transactions between members of a receiving (“mining for”) bitcoins to provide this assurance. network, allow all members to maintain their own copies of the Bitcoin constitutes an example where token economics is “in same ledger and ensure that all copies are synchronized. This the protocol.” That is, bitcoins – the network’s tokens – are obviates the need for the intermediary, who often exploits its automatically transferred by programmatic rules, and in so information asymmetry advantage to act in extractive, doing, the network properly incentivizes stakeholders to inefficient, or corrupt ways. A key challenge of this collectively behave to sustain its operations. Contrastingly, decentralized design is incentives. Even when not behaving platforms such as Hyperledger Fabric and R3’s Corda do not use exploitatively, the intermediary is well-incentivized to provide cryptocurrencies to effect stakeholder behavior. Rather, these centralized services in the form of transaction fees (e.g. Visa or platforms are meant for use in private networks and use Mastercard), fees charged for data analysis (e.g. Google), and traditional group policies “outside the protocol.” Private finder’s fees (e.g. Uber) [1]. When such services are provided in network partners are usually familiar with each other and transact on an ongoing basis; they are inclined to adhere to

policies that will sustain their relationships in and outside of the In particular, the predominant simulation technique private blockchain. employed for token engineering, in addition to game theoretically based models [19], [20], is Agent-Based Modeling In this paper, we present a real-life study conducted to (ABM) [18]. For instance, ABM of network dynamics develop an “in the protocol” token economics model for Insolar, analytically shows that Bitcoin network’s specific token model an 80-person blockchain startup headquartered in Switzerland leads the network to a stable, self-sustaining state [21] and other with offices in five countries. Insolar conducted an Initial Coin efforts simulate the rich nuances of the Bitcoin trading market Offering (ICO) of their cryptocurrency in December 2017 and [22], [23]. For extensions to other cryptocurrency and raised $42 million [3]. As they are about to launch their public, blockchain design, many efforts have produced generalized permissionless, blockchain – the Insolar MainNet – they sought frameworks and simulation environments. They include to update their model using applied systems dynamics modeling LUNES-blockchain [24], BlockSci [25], Blocksim [26], and simulation. The simulation study led to insights, which were Simblock [27], MIT’s BASIC [28], and cadCAD [29]. For non- incorporated into a multi-stakeholder token economics model Bitcoin contexts, we discover that these efforts generally focus that complements their updated business model. on ABM to simulate technical features of blockchain consensus We believe that the opportunity to investigate a novel real- mechanisms and throughput and assume an underlying token world application such as token economics modeling provides economics model. Our work is unique insofar as it uses rare and contributory visibility to the academic community. To simulation to design different features of an economic model that end, our paper is presented as follows. In the next section, and provides details of such simulations in an academic we provide related literature, and briefly describe how a token publication. We could speculate that crypto startups do not wish economics model is engineered. Then, we further describe to divulge such information or are not interested in publicly Insolar and detail their specific requirements for token presenting such work even if they have undertaken it. economics modeling. We then present an excerpt of the Regardless, providing a worked through example as we do in simulation study and results related to mechanisms for a this paper appears then to be a novel contribution. “subsidy pool,” an incentive program to subsidize third parties For our work of designing Insolar’s token economic system, who provide applications for user to execute on the MainNet. we use cadCAD [29], which entails conceptualizing the token Finally, we provide concluding remarks about how the design problem as a differential game problem in which state simulation informed Insolar’s design of the token economics variable values evolve over time according to some differential mechanism and provide general insights about cryptocurrencies equations. Since mathematical modeling for a complex scenario and token economics. – as is described for Insolar in Figure 1 – is intractable, the II. BACKGROUND conceptualization must first be modeled graphically to especially show feedback mechanisms between key We put forth that there are two distinct streams of academic components. literature regarding designing micro-economies of blockchain- based cryptographic tokens. The first is an economist’s perspective as seminally presented in [4]; the primary author, Christian Catalini, is the current Chief Economist for the Facebook-sponsored Libra cryptocurrency system. Some works study incentives specifically for cryptocurrency miners [5]–[7]. Others have studied incentives for various stakeholders in token- based platforms like speculative [8],[9] and utilitarian [10] cryptocurrency investors, platform users [11], and distributed application developers [12]. As would be expected, closed-form economics models underlie this perspective. Even those that apply more nuanced dynamical models [13], [14] generally eschew complex multi-stakeholder analysis in favor of closed- form economics models. Other complementary works include a bibliometric survey of token economics [15] and token economics applied to specific industries [16]. A contrasting perspective that focuses on the systematic development of token-based economies [17][18] is sometimes referred to as token engineering. Note that this is the process for Figure 1: All feedback mechanisms in Insolar MainNet token modeling, not the larger software engineering process that leads to the blockchain system itself. The outputs of the token III. MODELING INSOLAR’S TOKEN ECONOMY engineering process are models, results, and documentation that Similar to the Ethereum model, individuals and small are implemented in the blockchain system. A key difference vis- enterprises can use the Insolar MainNet as a public, à-vis the economics perspective is that multi-stakeholder models permissionless blockchain for a per transaction fee. Insolar also can be produced via token engineering through use of simulation has a complementary model that does not involve tokens: For a models that handle complexity albeit without elegant closed- fee in fiat currency, Insolar can design and operate a private form solutions. blockchain for enterprises in a business network. For instance, Insolar is involved in other projects in renewable energy, supply bandwidth, or databases, such Resource Providers are chain management, and mining and natural resources. paid in the MainNet. Miners do expend exorbitant amount of resources but that constitutes more of an However, in this paper, our focus is the MainNet. In the indirect provision, where contribution is not necessarily following table, we summarize the stakeholders and their roles proportional to the reward. in the MainNet as modeled in Figure 1 and describe what characteristics the right token economics model would achieve • Coin holders provide their XNS’s to maintain consensus for these roles [30]. in the network via staking. Whereas in Bitcoin and Ethereum networks, miners receive tokens to maintain TABLE I. ROLES AND MOTIVATIONS WITHIN THE INSOLAR TOKEN consensus throughout the blockchain (Proof-of-Work), ECONOMICS MODEL TYPE STYLES coin holders that have “skin in the game” vote to ensure The token economics model consensus (Proof-of-Stake). By ensuring that there is a Stakeholder Role in token economic model should: trustworthy consensus mechanism, coin holders/stakers SME provide confidence to SME Consumers, Application • Provide predictable fee Consumer structure Developers, and Resource Providers, in turn ensuring (Individual • Pays application user fees to stability for their coin investments. The coins that are and Application Developers • Assure predictable quality enterprise of application staked serve as a source of insurance for the MainNet: If users) performance Application Developers or Resource Providers violate • Converts fiat currency to Service Level Agreements (SLA’s) with each other, XNS Coins (MainNet’s coins from the pool of staked funds can be used to pay • Ensures “money supply” Broker(s) cryptocurrency) and pays for MainNet the rightfully aggrieved party. Stakers in turn are Application Developers on behalf of the SME Consumer rewarded a staking fee – an investment income, akin to interest, earned from leaving their staked funds • Develop and deploy Apps unwithdrawn. (smart contracts) and receive fees for their usage • Provide sufficient • Not only should the operation of the MainNet be • Pays “rent” for executing consumer base and decentralized, so should its governance. A MainNet applications and using resource capacities hardware resources to Insolar governed exclusively and in perpetuity by Insolar is not Application • Assure performance Developers Foundation as Application desirable. Instead, a governance foundation, similar to Platform and Resource quality of Resource Platform fees, respectively Providers consistent with those that underlie the Bitcoin and Ethereum networks Service Level • Also pays Resources Fees to Agreements (SLA’s) and one in which in the long run Insolar is one of many Resource Providers, which in members, is needed. part is used to pay Stake Rewards to XNS holders IV. SIMULATING INSOLAR’S TOKEN MODEL: APPLICATION • Provide hardware capacities DEVELOPER SUBSIDY (telecom, CPU, database) as a Node on the MainNet for A. System Simulation Model Resource running applications and • Provide stable and Providers receive fees for this predictable income for Resource Providers The token modeling process starts with a specification of the provision overall system model, which is shown in Figure 2. • Pay Staking Rewards to XNS Holders

• Provide XNS coins for staking as collateral behind the SLA’s and receive Staking Rewards. • Provide sufficient staking Coin holders • Provides stakes to Resource demand from Resource Providers, who in turn Providers extend insurance to Application Developers •

• Create a large • Collects Application Application Developer Platform and Resource community • Create a large hardware Insolar Platform Fees and also resource market Foundation provides subsidies to Application Developers and • Attract enterprises to use Resource Providers to Insolar MainNet incentivize participation • Maintain and improve the platform

Here are some further clarifications. Figure 2: Overall System Simulation Model for Insolar • Whereas Bitcoin and Ethereum are peer-to-peer networks that do not pay those that provide CPU cycles, B. State Transition Model • 훽(푡 + 푗) = 푓훽(퐼(푡)). (3) Although Insolar conducted an extensive study addressing 1 App Dev The Application Developer Income (in XNS) at time t, I(t), all multiple parties , we highlight one facet to bound the scope Income is the sum of App Subsidy received by time t and the number of this paper: how to incentivize pioneering Application of applications * c, the average cost of an app. Developers to contribute applications when the MainNet is • 퐼(푡) = 퐴0 − 퐴(푡) + 푐 ∙ 푈(푡). (4) nascent and network effect has not been achieved yet. The black- Found- The size of the Foundation Treasury at time t, T(t), is outlined excerpt in Figure 1 is a diagram for this excerpt and ation dependent on App Subsidy Pool as well as other state transitions of states within this excerpt are modeled below in Treasury variables such as Capacity Subsidy and Platform Fees that Figure 3. Size are outside the scope of the subsidy pool excerpt. Moreover, just as how App Dev Income at t is used to update parameters for Number of Apps, T(t) is used to calculate various parameters for variables such as Capacity Subsidy and Platform Fees. f(t) is an arbitrary function of these out- of-the scope variables and g(t) is an arbitrary function of A(t) and f(t). Parameters for other state variables at t+j are used in functions of T(t). • 푇(푡) = 푔(퐴(푡), 푓(푡)). (5)

D. Simulation Runs

App Usage Parameter No and β(0), and App Subsidy Parameters Ao and  from Table II were the input variables into the simulation. Of these, the key decision variable is Ao as that reflects a policy decision: How much should Insolar initially Figure 3: Transition of State Variables for Insolar (Application commit to subsidize Application Developers to develop apps on Subsidy Excerpt)2 their platform?

C. Specification of State Variables TABLE III. SIMULATION PARAMETERS The table below provides a simplified formal specification Initial Reward Pool Decay Time Monte Carlo of the excerpted state variables. 3 4 Size in XNS (Ao) Rate () Steps Runs

TABLE II. FORMAL SPECIFICATION OF STATE VARIABLES FOR INSOLAR (APPLICATION SUBSIDY EXCERPT) 250 x 10e6 0.0005 3652 100

Variable Formal specification 500 x 10e6 0.0005 3652 100 App The number of XNS’s distributed to developers at time t, Subsidy A(t), through the App Subsidy is governed by the initial 750 x 10e6 0.0005 3652 100 Pool reward pool size A0, and the exponential decay rate λ. Thus, • 푑퐴 = −휆 ∙ 퐴, (1) 푑푡 1000 x 10e6 0.0005 3652 100

Number Number of apps at time t, U(t), grows at an exponential The simulations were carried out in a Monte Carlo fashion of Apps growth rate β(t) from an initial count N0. The growth rate is using four possible scenarios for the starting reward pool size (App parametrized to t, β(t), whereas the App Subsidy decay rate λ and means of affected state variables – Number of Apps (App Usage) is constant. We recognize that number of apps is dependent Usage), App Dev Income, and Foundation Treasury Size – were minimally on numbers of users and developers and other computed. The decay rate was kept constant throughout these factors. However, we simplify this complex interaction by updating β(t) that captures this interaction every time step j experiments, as were the numbers of simulated time steps, and as a function solely of app income. This parameter is tuned Monte Carlo runs. at the discretion of the modeler, and not necessarily based on formal rules. E. Simulation Results • 푑푈 = 훽(푡) ∙ 푈, (2) Depletion from the Application Subsidy Pool (top panel, 푑푡 Figure 4) confirms the expected exponential decay in token U(t+j) is calculated using β(t+j), which in turns is modeled distribution as well as the pattern resulting from the change in as some function fβ of the Developer Income I(t) at time t: initial starting reward pool size. The resulting XNS token

1 MoE in the diagram stands for Medium of Exchange and refers to any arbitrary currency, fiat or crypto, other than XNS tokens 3 Additional experiments were run during the Insolar economic parameter 2 As with Foundation Treasury size, XNS price movements are a function of design and validation research, including but not limited to exploring the complex interaction between state variables. As again with Foundation effects of varying the decay rate. Treasury size, most of the variables that affect XNS prices are outside of the 4 Steps in the simulation are in 1 day increments up to 3,652 days or 10 years scope of the subsidy excerpt, and so we chose not to discuss simulation of price movements in this paper. distribution to the Application Developers (bottom panel, Figure 4) can be seen in the figure as well. This again is intuitively reasonable given the described model and varying initial reward pool size.

Figure 5: Subsidy Effect on Application Developer Income and Foundation Treasury

F. Simulation Insights The key insight from this simulation is that the Application

Subsidy has a positive effect on Foundation Treasury; that is, the Figure 4: Subsidy – From Pool and To Developers larger the initial pool of funds for the Application Subsidy, the larger the eventual size of the Foundation Treasury. If we Interestingly, the net effect on Application Developer assume that the growth of the treasury is a reasonable by-product Income (top panel, Figure 5) is less intuitive. It shows that in the of the success of the platform, the simulation results point to the long run Application Developers receive similar incomes Application Subsidy as having the desired effect of helping regardless of the initial size of the subsidy pool. There appears platform growth. The results warrant recommending a policy of to be a complex and non-linear relationship between the initial seeding the Subsidy Pool with significant amount of XNS subsidy pool size and reward value delivered to Application tokens. Developers. The direct overall benefit to Application Developers is less The initial reward pool size also has an interesting effect on clear. Even if more XNS tokens are granted to Application the value of the tokens held by the Insolar Foundation. Developers through the subsidy, the net effect on them is not Predictably, the Foundation’s holdings of XNS tokens vary guaranteed at every point in time. Application Developers pay inversely with initial application subsidy pool size. However, Application Platform and Resource Platform fees in XNS tokens due to complex behaviors (from simulation runs related to but and these grow at some rate, yet the rate at which they receive outside the scope of the subsidy excerpt), the value of tokens subsidies decays. Ultimately, the Insolar Foundation reaps the held by the Foundation does increase as initial subsidy pool size reward from this mismatch of accelerating fee revenues and is increased (bottom panel, Figure 5). decelerating subsidy outlays.

V. CONCLUDING REMARKS To summarize, given the increasing role of cryptocurrencies – especially in light of recent announcements of corporate cryptocurrencies from Facebook, JP Morgan, and Walmart [31] – we determine that the topic of token economics is under- researched. This paper presents excerpts of the token engineering process to demonstrate how simulation modelling 2019, Accessed: Feb. 04, 2020. [Online]. Available: informed policy design for Insolar’s subsidy pool – their https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3025604. mechanism to incentivize application providers to develop more [7] L. W. Cong and Z. He, “Blockchain Disruption and Smart Contracts,” Rev. Financ. Stud., vol. 32, no. 5, pp. 1754–1797, May 2019, doi: applications onto the public MainNet. Simulation results point 10.1093/rfs/hhz007. to investment into the subsidy pool as having a desirable positive [8] S. Krueckeberg and P. Scholz, “Cryptocurrencies as an Asset Class?,” effect that helps MainNet growth. The insights from modeling in Cryptocurrencies as an Asset Class?, S. Gouette, G. Khaled, and S. will be important as Insolar tries to foster a vibrant ecosystem of Saadi, Eds. Springer, 2019. application developers and computing resource providers that [9] G. Dorfleitner and C. Lung, “Cryptocurrencies from the perspective of euro investors: a re-examination of diversification benefits and a new will entice individual and enterprise users to join the MainNet day-of-the-week effect,” J. Asset Manag., vol. 19, no. 7, pp. 472–494, after launch. Dec. 2018, doi: 10.1057/s41260-018-0093-8. [10] K. Malinova and A. Park, “Tokenomics: When Tokens Beat Equity,” For this brief paper, a fuller exploration of theoretical issues SSRN Electron. J., Nov. 2018, Accessed: Feb. 04, 2020. [Online]. was beyond our scope. We did not delve much into the formal Available: https://papers.ssrn.com/abstract=3286825. control theory underlying these simulation models, nor did we [11] E. Pagnotta and A. Buraschi, “An Equilibrium Valuation of Bitcoin relate much to traditional literature on market design from and Decentralized Network Assets,” SSRN Electron. J., Jul. 2018, Economics and Operations Research. Moreover, though always Accessed: Feb. 04, 2020. [Online]. Available: https://papers.ssrn.com/abstract=3142022. a pressing issue for blockchain-based systems, we did not [12] C. Catalini and J. S. Gans, “Initial Coin Offerings and the Value of address security issues as bounding of our scope entailed Crypto Tokens,” SSRN Electron. J., Mar. 2019, Accessed: Feb. 04, assuming security breaches to the MainNet to be exogenous 2020. [Online]. Available: https://papers.ssrn.com/abstract=3137213. events that our simulation does not factor in. Regardless of these [13] B. Biais, C. Bisière, M. Bouvard, and C. Casamatta, “The Blockchain limitations, for a fledgling startup like Insolar, their future Folk Theorem,” Rev. Financ. Stud., vol. 32, no. 5, pp. 1662–1715, May 2019, doi: 10.1093/rfs/hhy095. success will in no small part hinge upon how well their model [14] L. W. Cong, Y. Li, and N. Wang, “Tokenomics: Dynamic Adoption incentivizes various stakeholders to participate on their MainNet and Valuation,” SSRN Electron. J., p. 43, May 2019. network versus that of numerous alternatives. We believe that [15] A. F. Bariviera and I. Merediz-Solà, “Where do we stand in providing visibility to such a strategic and academically novel cryptocurrencies economic research? A survey based on hybrid exercise serves a contribution to the research community. analysis,” ArXiv200309723 Q-Fin, Mar. 2020, Accessed: May 03, 2020. [Online]. Available: http://arxiv.org/abs/2003.09723. Practically, Insolar is also interested in future work that more [16] J. Zhang, F.-Y. Wang, and S. Chen, “Token Economics in Energy granularly examines application subsidies, namely the mix of Systems: Concept, Functionality and Applications,” ArXiv180801261 spending between these subsidy types: Q-Fin, Aug. 2018, Accessed: May 03, 2020. [Online]. Available: http://arxiv.org/abs/1808.01261. • App Seeding entails incentivizing application creation [17] A. Pazaitis, P. De Filippi, and V. Kostakis, “Blockchain and value systems in the sharing economy: The illustrative case of Backfeed,” events such as “code jams” or “hackathons.” Technol. Forecast. Soc. Change, vol. 125, pp. 105–115, Dec. 2017, doi: 10.1016/J.TECHFORE.2017.05.025. • New Application Subsidy is offered for newly deployed [18] E. Dhaliwal, Z. Gurguc, A. Machoko, G. Le Fevre, and J. Burke, applications, with the subsidy decreasing to 0 the “Token Ecosystem Creation A strategic process to architect and developer after an “introductory discount” time. engineer viable token economies,” 2018. [19] J. Barreiro-Gomez and H. Tembine, “Blockchain Token Economics: • A Volume Resource Subsidy is meant to ease the A Mean-Field-Type Game Perspective,” IEEE Access, vol. 7, pp. resources costs incurred by developers of popular 64603–64613, 2019, doi: 10.1109/ACCESS.2019.2917517. applications. [20] S. Bartolucci and A. Kirilenko, “A Model of the Optimal Selection of Crypto Assets,” ArXiv190609632 Cs Q-Fin, Jun. 2019, Accessed: • A Success Reward is awarded to Application May 03, 2020. [Online]. Available: http://arxiv.org/abs/1906.09632. Developers for meeting strategic Key Performance [21] M. Zargham, Z. Zhang, and Preciado. V., “A State-Space Modeling Indicators, as determined by the Insolar Foundation. Framework for Engineering Blockchain-Enabled Economic Systems,” in IX International Conference on Complex Systems, Cambridge, MA The simulation model required for determining this mix USA: STEM Academic Press, 2018, pp. 4–21. needs to build upon and extend the work presented in this paper. [22] L. Cocco, R. Tonelli, and M. Marchesi, “An Agent-Based Artificial Market Model for Studying the Bitcoin Trading,” IEEE Access, vol. 7, pp. 42908–42920, 2019, doi: 10.1109/ACCESS.2019.2907880. VI. REFERENCES [23] K. Lee, S. Ulkuatam, P. Beling, and W. Scherer, “Generating [1] M. Swan, “Blockchain Thinking : The Brain as a Decentralized Synthetic Bitcoin Transactions and Predicting Market Price Autonomous Corporation [Commentary],” IEEE Technol. Soc. Mag., Movement Via Inverse Reinforcement Learning and Agent-Based vol. 34, no. 4, pp. 41–52, Dec. 2015. Modeling,” J. Artif. Soc. Soc. Simul., vol. 21, no. 3, p. 5, 2018. [2] S. Nakamoto, “Bitcoin: A Peer-to-Peer Electronic Cash System,” [24] E. Rosa, G. D’Angelo, and S. Ferretti, “Agent-Based Simulation of Bitcoin.org, 2008. ,” in Methods and Applications for Modeling and [3] G. Moore, “INS Closes $42M ICO for Decentralized Grocery Simulation of Complex Systems, Singapore, 2019, pp. 115–126, doi: Delivery,” Blockchain News, Dec. 27, 2017. https://www.the- 10.1007/978-981-15-1078-6_10. blockchain.com/2017/12/27/ins-closes-42m-ico-decentralized- [25] H. Kalodner, S. Goldfeder, A. Chator, M. Möser, and A. Narayanan, grocery-delivery/. “BlockSci: Design and applications of a blockchain analysis [4] C. Catalini and J. Gans, “Some Simple Economics of the Blockchain,” platform,” ArXiv170902489 Cs, Sep. 2017, Accessed: Feb. 04, 2020. Commun. ACM, vol. Forthcoming. [Online]. Available: http://arxiv.org/abs/1709.02489. [5] D. Easley, M. O’Hara, and S. Basu, “From mining to markets: The [26] M. Alharby and A. van Moorsel, “BlockSim: A Simulation evolution of bitcoin transaction fees | Elsevier Enhanced Reader,” J. Framework for Blockchain Systems,” ACM SIGMETRICS Perform. Financ. Econ., vol. 134, no. 1, pp. 91–109, Oct. 2019. Eval. Rev., vol. 46, no. 3, pp. 135–138, Jan. 2019, doi: [6] G. Huberman, J. Leshno, and C. C. Moallemi, “An Economic 10.1145/3308897.3308956. Analysis of the Bitcoin Payment System,” SSRN Electron. J., Mar. [27] Y. Aoki, K. Otsuki, T. Kaneko, R. Banno, and K. Shudo, “SimBlock: startups as well as UN/CEFACT. He holds a Ph.D. in Computer A Blockchain Network Simulator,” in IEEE INFOCOM 2019 - IEEE Engineering from the University of Manitoba, where his work Conference on Computer Communications Workshops (INFOCOM on multi-agent modelling of complex healthcare systems was WKSHPS), Apr. 2019, pp. 325–329, doi: 10.1109/INFCOMW.2019.8845253. cited by the US Centers for Disease Control (CDC). [28] L. Marrocco et al., “BASIC: Towards a Blockchained Agent-Based SImulator for Cities,” in Massively Multi-Agent Systems II, Cham, Michael Zargham is the founder and CEO of BlockScience, 2019, pp. 144–162, doi: 10.1007/978-3-030-20937-7_10. an engineering, R&D, and analytics firm combining systems [29] Z. Zhang, “Engineering Token Economy with System Modeling,” theory, computational social science to design economic ArXiv190700899 Cs, Jun. 2019, Accessed: Feb. 04, 2020. [Online]. systems. He holds a PhD in Systems Engineering from the Available: http://arxiv.org/abs/1907.00899. University of Pennsylvania where he worked on dynamic [30] Insolar, “Insolar Economic Paper,” Jun. 2019. resource allocation in decentralized networks. https://insolar.io/uploads/Insolar Economic Paper.pdf. [31] M. del Castillo, “Blockchain Goes To Work At Walmart, Amazon, Hjalmar Turesson was born in Helsingborg, Sweden. He JPMorgan, Cargill and 46 Other Enterprises,” Forbes, Apr. 16, 2019. received a B.Sc. in Biology from Lund University, Sweden, a M.Sc. from the University of Tübingen, Germany, and a Ph.D. VII. AUTHOR BIOS from Princeton University. Currently, he is teaching AI at Schulich School of Business, York University, Canada. Henry Kim is an Associate Professor at the Schulich School of Business and the Director of the blockchain.lab at York Matthew Barlin is Lead Systems Engineer at BlockScience University. Prof. Kim has written extensively on blockchain, with research interests in Model Based Systems Engineering and authoring more than 25 blockchain-related papers in venues blockchain. He holds an SM from MIT in Ocean Systems such as IEEE journals and Frontiers in Blockchain. He serves as Management and is a member of INFORMS. a senior research advisor for several blockchain startups and co- nd Danil Kabanov is the Head of Development Center at organized the 2 IEEE Conference on Blockchain and Insolar. His research interest are in Enterprise Blockchain Cryptocurrencies. Prof. Kim received his PhD from the Systems. He received a Master of Mathematics (Crypto . He is a member of the IEEE. analysis) from Tomsk State University and a Master of Marek Laskowski serves as Senior Research Scientist with Economics from St. Petersburg Polytechnic University. BlockScience and is a Co-Founder of blockchain.lab and Toronto Blockchain Week. He also advises several blockchain