<p>We can define a cosmos network as a <em><em>social galaxy</em></em> with various entities and different types of participants who are fully self-responsible for decisions they make. To make such system as healthy as possible, minimize cheating and other fraudulent behavior that cause loss of confidence, it should contain a set of rules, instruments and other incentives which will determine the right direction together with moral ethics.</p><p>Slashing is an event, which results in a loss of stake percentage, depending on the type of network violation and jeopardizing the safety of other participants. It represents not only a financial incentive to act properly but also is a measure to prevent nothing at stake problem.</p><p>Cosmos is a complex ecosystem where atom act not only as an economic incentive but also represent a governance unit playing a crucial role in ecosystem security. In that way, slashing becomes a tool that influences voting power distribution.</p><p>Besides, it affects the authority of caught fraudlent participant, motivates validators to improve their infrastructure and in case of delegators, to provide a deeper due-diligence and diversification amongst validators. Slashing also act as a decentralization mechanism motivating re-delegate atoms to more reliable or even smaller validators with equal level of security and infrastructure set up.</p><p>For now this motivation can be not so obvious but after enabling Inter Blockchain Communication (IBC) and feature of shared security when validators will be slashable on multiple validated chains slashing risk will be different for all validators depending on conditions and number of chains they operate.</p><h1 id="slashing-events"><strong>Slashing events</strong></h1><p>There are two types of events when stake liquidation happens:</p><ul><li><strong><strong>Downtime</strong></strong> occurs when validator is offline and do not participate in block commitment signing less than <code>5%</code> of the blocks in a row of <code>10 000</code>. This situation leads to loss of <code>0,01%</code> stake not only for validator but for bonded to him delegators as well. In addition, validator drops out of the consensus and do not earn block rewards for at least <code>10 minutes</code>. After fixing the issues validator can re-join validators set by sending un-jail transaction.</li></ul><figure class="kg-card kg-image-card"><img src="https://p2p.org/economy/content/images/2020/09/1-14.png" class="kg-image" alt loading="lazy" width="647" height="151" srcset="https://p2p.org/economy/content/images/size/w600/2020/09/1-14.png 600w, https://p2p.org/economy/content/images/2020/09/1-14.png 647w"></figure><ul><li><strong><strong>Double-sign</strong></strong> can lead to more harmful consequences than the previous one. It can cause double-spend or chain fork. The wrong setup of the validator’s infrastructure or key compromise are the most common occasions that cause this type of slashing. In this case, stake penalizes by <code>5%</code> and validator loses the right to propose blocks and earn rewards without an ability to un-jail. All delegators of this validator enter the unbonding period, which lasts <code>21 days</code>.    </li></ul><figure class="kg-card kg-image-card"><img src="https://p2p.org/economy/content/images/2020/09/2-17.png" class="kg-image" alt loading="lazy" width="652" height="195" srcset="https://p2p.org/economy/content/images/size/w600/2020/09/2-17.png 600w, https://p2p.org/economy/content/images/2020/09/2-17.png 652w"></figure><p><strong><strong>Slashing also affect atoms, which were in unbonding phase</strong></strong> at the moment when one of these events happened. If a validator have low self-bonded ratio (low self-delegated amount) and large amount bonded then, in theory, it could have economic incentive to double-sign. This behaviour will lead to a loss of confidence in this validator and as a consequence inability to earn transaction fees and atom provisions in future missing opportunity of the long term ecosystem adoption and development.</p><p>Validators with low self-delegated amount should be able or will have to find the way to maintain resilent infrastructure with low costs in order to increase self-delegation and/or commision rate while bonded atoms to them are increasing.</p><p>If self-bonded ratio is decreasing or low in some cases (for instance, validator bonded to other validators in order to diversify holdings or increase network decentralization), if validator charges fair commission long term incentives should overcome short term gains. Commission rate should be reasonable and cover existing expences. If a validator with high stake is not earning to maintain infrastructure and operations (by self-bonded amount that generate rewards and/or commission rate) it is at least concerning.</p><h1 id="how-slashing-works-in-theory"><strong>How slashing works in theory</strong></h1><p>Assume that we have three validators:</p><ul><li><code>V1</code> with a commission of <code>5%</code></li><li><code>V2</code> with a commission of <code>8%</code></li><li><code>V3</code> with a commission of <code>9%</code></li></ul><p>If delegator bond to <code>V1</code> with an annual return on staking (RoS) around <code>10,2%</code> for <strong><strong>5 years</strong></strong> and <em><em>without taking advantage of compounding</em></em>, then his cumulative interest for five years nominated in atoms will be <code>48,5%</code>. Let’s have a look at how <em><em>monthly compounding</em></em> with slashing will affect this number. To simplify calculation we assume that: <strong><strong>In the case of downtime</strong></strong>, it happened <em><em>three times</em></em> and delegators stake passively without re-delegating after the first event:</p><ol><li>End of the 2^nd month and <code>2 days</code> passed before un-jail</li><li>End of the 7^th month and <code>1 day</code> passed before un-jail</li><li>End of the 11^th month and <code>1 day</code> passed before un-jail</li></ol><p><strong><strong>In the case of double-sign</strong></strong>, slashing and unbonding period occurred once at the end of the 12^th month. After <code>21 day</code> of unbonding delegator bonded to another validator with the same commission rate. I will use the same conditions for other comparisons in this article. Overall result for delegator will look like:</p><figure class="kg-card kg-image-card"><img src="https://p2p.org/economy/content/images/2020/09/3-16.png" class="kg-image" alt loading="lazy" width="1062" height="535" srcset="https://p2p.org/economy/content/images/size/w600/2020/09/3-16.png 600w, https://p2p.org/economy/content/images/size/w1000/2020/09/3-16.png 1000w, https://p2p.org/economy/content/images/2020/09/3-16.png 1062w" sizes="(min-width: 720px) 720px"></figure><p>We can notice that:</p><ul><li>Delegator who took advantage of monthly compounding even with a double-sign event outperformed another one who just delegated once and forgot. The magic of this feature I will cover in the next article.</li><li>Downtime has not much influence on the result for delegator even if it happens quite often and validator re-join validator set much later than current jail time for downtime (10 minutes). It has a bigger influence on the validator in the long term. His delegators will lose confidence and will immediately re-delegate their holdings to others. Until validator un-jail, the staked ratio will be lowered by the number of atoms delegated to him. This will increase voting power of other validators resulting in higher probability of proposing a block and may lead to higher transaction fee gais in comparison with validators who often go offline.</li><li>Double-sign has the most harmful event on RoS and the difference is about <code>9%</code> in comparison with delegator who have chosen an honest and secure validator</li></ul><p>If we will compare the performance of delegators who bonded to different validators with a various commission rate, we will see that RoS for <code>V3</code> is higher than RoS for <code>V1</code> and <code>V2</code> if double-sign occurred. For a taken period of 5 years this will be correct even if the commission of <code>V3</code> will be <code>16%</code> that is more than three times higher than the <code>5%</code> commission of <code>V1</code>.</p><figure class="kg-card kg-image-card"><img src="https://p2p.org/economy/content/images/2020/09/4-11.png" class="kg-image" alt loading="lazy" width="1062" height="535" srcset="https://p2p.org/economy/content/images/size/w600/2020/09/4-11.png 600w, https://p2p.org/economy/content/images/size/w1000/2020/09/4-11.png 1000w, https://p2p.org/economy/content/images/2020/09/4-11.png 1062w" sizes="(min-width: 720px) 720px"></figure><p>You can notice that in the longer term (in our example &gt;5 years) current double-sign slashing do not cause huge effect on the performance and there still exist high incentive for delegators to choose validators basing predominantly on the commission rate. In theory, this may cause weaker decentralization level of the network.</p><p>Downtime slashing has even less voting and economic influence. Current slashing conditions should be considered as a starting point for further discussion on that topic and may be changed in future via governance mechanism.</p><p>For example, every repeating downtime event over the period of <code>X</code> could cause atom slashing equivalent to <code>prev_slashing_percentage * 2</code>. If a validator constantly goes offline this will cost more for him and his delegators thus increasing incentive to properly maintain the state of own infrastructure and for delegators to re-delegate to others. One of concerns about changing initial parameters is a lack of empirical data so as the network evolve we will see more experiments in this field.</p><h1 id="smart-ideas-for-delegators-to-protect-from-slashing-consequences"><strong>Smart ideas for delegators to protect from slashing consequences</strong></h1><p>No one can predict the future and <strong><strong>one of the best ways for delegators to protect themselves from misbehavior is diversification</strong></strong>. Suppose that delegator bonded all his atoms to <code>V1</code> with the lowest commission possible, <code>5%</code> in our case. Another delegator diversified amongst all three validators equally - <code>33%</code> for each. If <code>V1</code> will be caught on double-sign, the second delegator will get <code>2,5%</code> higher RoS than the first one who put all atoms in one basket even if <code>V2</code> &amp; <code>V3</code> went offline for some reason.</p><p>Another idea is responsible behavior. Bonding to a validator is not a blind step and simple way to earn passive income. To be up to date delegators should continue to monitor validator uptime. Frequent downtimes may indicate unreliable infrastructure.</p><ul><li>What actions validator take in order to prevent slashing conditions?</li><li>Does valiadator disclose an infrastructure setup?</li><li>Is it secure?</li><li>What upgrades and improvements are in the roadmap?</li><li>Is the commission rate sufficient to support validator activities and maintain reliable infrastructure?</li><li>What is the responsibility level of a validator and how valuable is it's contribution to the ecosystem development?</li></ul><p>Answers to these questions can help delegators to diversify amongst the most remarkable validators.</p><p>The most prominent validators who set up well-protected infrastructure and have a high level of confidence can offer refunds for their delegators in case of slashing event. In this case reserve funds or the idea of developing slashing insurance for delegators make sense. For some delegators who have no ability to follow up with the state of their atom performance this could be a reasonable solution.</p><blockquote><em><em>The first rule – do not lose your money, the second rule – remember of the first one.</em></em><br><em><em>"Warren Buffet"</em></em></blockquote><p>In the cosmos ecosystem, your atoms are your assets, which can generate additional income for you. Take care of your holdings and be responsible for the decisions you make.</p><hr><p><strong><strong>P2P Validator</strong></strong> offers high-quality staking facilities and provides up to date information for educational purposes. Stay tuned for updates and new blog posts.</p><hr><p><strong><strong>Web:</strong></strong><a href="https://p2p.org/?ref=p2p.org"> https://p2p.org</a></p><p><strong><strong>Stake ATOMs with us:</strong></strong><a href="https://p2p.org/cosmos?ref=p2p.org"> https://p2p.org/cosmos</a></p><p><strong><strong>Twitter:</strong></strong><a href="https://twitter.com/p2pvalidator?ref=p2p.org"> @p2pvalidator</a></p><p><strong><strong>Telegram:</strong></strong><a href="https://t.me/p2pvalidator?ref=p2p.org"> https://t.me/p2pvalidator</a></p>

<p>To understand economical structure of Cosmos we should look closely at the key principles of the ecosystem, basic incentives for all different participants and possible influence of these principles on their behavior. Overall network purpose (mission) is <em><em>satisfying the needs of ecosystem users by giving them an opportunity to provide their services in a decentralized manner with the ability to interoperate without centralized entities.</em></em></p><p>Cosmos network consists of application-specific blockchains (Zones) which can be designed differently depending on the utility purpose. All blockchains are interoperable within a single ecosystem connected through intermediary blockchains (hubs) that in fact replace centralized organizations by set of validators. As a separate blockchain each zone can have its own token to govern the private or public network and have its own set of validators but governance can decide that validators of the Cosmos hub will be required to validate additional zones.</p><p>Cosmos ecosystem utilizes Tendermint - practical byzantine fault tolerance (PBFT) consensus mechanism. It means that finalizing blocks depends on 2/3 plus one quorum of all validators agreed on the current state of the network in order to reach the consensus. There are three key groups of participants in Cosmos ecosystem. Each group has its own incentives and impact on the state of ecosystem.</p><h3 id="validators"><strong>Validators</strong></h3><p>For Cosmos hub there are <code>100</code> possible validators, who are responsible for proposing new blocks and validating transactions. This number will rise by <code>13%</code> a year until it reaches three hundred maximum possible validators. For other hubs and zones this number is not mandatory and will depend on the particular use case and required level of security. If there are more validators then the right to participate in consensus will have participants with the higher amount of ATOMs bonded.<br></p><figure class="kg-card kg-image-card"><img src="https://p2p.org/economy/content/images/2020/09/1-15.png" class="kg-image" alt loading="lazy" width="558" height="79"></figure><p>Cosmos hub validators has the highest impact on network security and provide intercommunication between zones. They must actively participate in governance and are required to vote on every proposal otherwise their ATOMs are at the risk of being slashed (currently this feature is not active).</p><blockquote><em><em>Tell me who are the validators and I will tell you if the network is safe</em></em></blockquote><p>This group have enough knowledge and resources to maintain infrastructure and are interested in generating maximum long-term gains from ATOM inflation and transaction fees (about this later). That is why validators care about healthy and sustainable ecosystem development. They should act in interests of their delegators if they want to keep them loyal and increase the voting power in the long term.</p><h3 id="delegators"><strong>Delegators</strong></h3><p>This group consists of ATOM token holders who have not enough skills or resources to run and maintain the infrastructure but still want to increase the network security and earn a share of the transaction fees and inflationary reward by bonding tokens to the validators. It boosts the voting power of validator and frequency of being chosen as block proposer. In fact, <em><em>by bonding ATOM to validator delegators choose one of the pillars of the ecosystem so their choice is very important and affects the level of decentralization.</em></em><br>Delegators are eligible for transaction fees and inflation reward as well but they have to pay commission, which vary within existing validators. There is no way for validators to steal bonded tokens but there are still other risks related with choosing the validator, which we will discuss later. Decisions based solely on a low commission rate is not always the best decision for delegators.</p><h3 id="users"><strong>Users</strong></h3><p>If we compare Cosmos network with the market, users are consumers and service providers. Developers, entrepreneurs and buyers who want to utilize the advantages of Cosmos ecosystem. Many different hubs can co-exist. To go further we can compare the network with a nation where there is a national hub with cities and zones each acting as smaller hubs within it and the possibility to join with the secure hub (Cosmos) if needed. Activity of this group measures the overall value of the network and can have a significant impact on demand for the token, transaction fees, workload and so on.</p><h1 id="cosmos-hub-economy-and-reward-distribution"><strong>Cosmos hub economy and reward distribution</strong></h1><p>Cosmos hub economy relies on the inflationary approach. The target annual inflation rate represents the percentage from total supply that is changing each block. If the total bonded stake is less than <code>66%</code> of the total ATOM supply, the inflation rate will slightly increase until it reaches a maximum of <code>20%</code> or the total bonded stake climbs higher than <code>66%</code>. In this case, the annual inflation will decrease to <code>7%</code> depending on ATOMs participating in bonding. New tokens incentivize participants to secure the network. The more tokens locked via staking the higher the threshold for initiating a successful attack.</p><p>There are two fundamental streams of revenue for validators:</p><figure class="kg-card kg-image-card"><img src="https://p2p.org/economy/content/images/2020/09/2-18.png" class="kg-image" alt loading="lazy" width="705" height="560" srcset="https://p2p.org/economy/content/images/size/w600/2020/09/2-18.png 600w, https://p2p.org/economy/content/images/2020/09/2-18.png 705w"></figure><ul><li><strong><strong>Block reward</strong></strong>, which distributing amongst all validators proportionally to their voting power. This reward is paid in ATOMs and depends on the current annual inflation rate, which varies each block with the frequency of average block time. In the future, this reward can be split between ATOMs and Photons that will play the role of a secondary token with the purpose of transaction fee payment only, decreasing the velocity and liquidity of ATOMS and making a 1/3 attack more expensive. Inflation of Photons is expected to be fixed and equivalent to <code>500</code> tokens per hour. <em><em>Currently Photons are not available but could be activated with specific implementation and distribution method by the community via governance.</em></em></li><li><strong><strong>Transaction fees</strong></strong> form in each block and proposer gets <code>1%</code> or <code>5%</code> depending on the number of precommits included. The frequency of proposing blocks is proportional to the voting power of a validator. Before fees distribution, <code>2%</code> goes to a reserve pool. These funds can be spent on network development or other activities by voting.<br>Currently transaction fees are close to zero as the network is still in an early stage of adoption. In the future, it will be possible to receive fees in other whitelisted tokens than ATOMs in Cosmos ecosystem. The amount of fees will depend on gas prices and usage of the network.</li><li>Validator’s <strong><strong>commission rate</strong></strong> represents a percentage from both streams of revenue that delegators pay to a validator. Delegators reward streams are the same as for validators less the commission percentage.</li></ul><h1 id="penalties"><strong>Penalties</strong></h1><p>In some circumstances occurs slashing of bonded ATOMs. Penalties should increase the responsibility level of participants who are directly involved in decisions associated with network security. Validators have no control over delegator’s stake but if such an event happens both parties lose a percentage of their tokens. This is in order to prevent misbehavior and negligence from validators and bring incentives to delegators to diversify amongst them, perform proper due diligence and choose wisely.</p><ul><li>The first reason a stake can be slashed is double-signing a block. This means that a malicious node is broadcasting two blocks with different content for the same height. The penalty for that is currently set at <code>5%</code> and the validator who is responsible for that drops out of validators set. All ATOMs enter an unbonding (process of undelegating ATOMs from validator) period that lasts for <code>21 days</code> and <em><em>within this period the stake will not earn provisions and transaction fees.</em></em></li><li>If a validator fails to sign more than <code>95%</code> blocks in a row of <code>10000</code> due to inactivity, <code>0.01%</code> of the bonded ATOMs will be lost and the validator will be <em><em>jailed for <code>10 minutes</code> without allowance to participate in consensus and be eligible for rewards.</em></em></li></ul><p>If slashing happens, it decreases stake and leads to fewer ATOMs paid as a reward.</p><h3 id="example"><strong>Example</strong></h3><p>Let’s compare three imaginary validators. Assume that delegator bonded equal amount of ATOM to each. Slashing decreases the amount of stake thus meaning a proportional decrease in ATOM provisions since the event has taken place.</p><p>For example, <em><em>validator 1</em></em> <code>V1</code> was caught on a double sign and slashing occurred on the 60^th day. After unbonding, the rest was staked with the same conditions to another one. <em><em>Validator 2</em></em> <code>V2</code> had three liveness slashes on the 30^th day with a 2-day recovery period and an inability to fix the issue, on the 60^thand 180^th days being offline for one day each. <em><em>Validator 3</em></em> <code>V3</code> had no such events in place.</p><figure class="kg-card kg-image-card"><img src="https://p2p.org/economy/content/images/2020/09/3-17.png" class="kg-image" alt loading="lazy" width="779" height="265" srcset="https://p2p.org/economy/content/images/size/w600/2020/09/3-17.png 600w, https://p2p.org/economy/content/images/2020/09/3-17.png 779w" sizes="(min-width: 720px) 720px"></figure><p><strong><strong>Overall results show that validator with a higher commission and honest behavior performed better than validators with stated slashing events.</strong></strong> In this example we have not taken compounding into consideration. Every slashing event may decrease the confidence of delegators and may lead to immediate re-delegation to another validator. That will cause fewer commission rewards in ATOM for validator and can lead to inability to maintain secure infrastructure in future.</p><h1 id="recommendations-for-delegators"><strong>Recommendations for delegators</strong></h1><p>Every delegator is self-responsible for the financial decisions made. To choose the proper validator and understand misbehavior risks it is important to read the full conditions of the delegator agreement and find out the validator’s policy on this topic.</p><ul><li><strong><strong>Security audit</strong></strong> that proves the stated level of security that should decrease the probability of a key compromise that may lead to double-sign.</li><li>Check for an <strong><strong>available roadmap</strong></strong> of future steps to increase the security or transparency of annual expenses on upgrades and improvements.</li><li><strong><strong>Skin in the game</strong></strong> proves that the validator puts their own funds at risk as well. The higher the percentage of their stake the more responsible the validator should be and if slashing does occur he will lose his own ATOMs and .</li><li><strong><strong>Compensation policy.</strong></strong> Are there any payouts or bonuses from the validator’s side in case of a slashing event?</li><li><strong><strong>Diversification amongst 3-7 validators</strong></strong> who operate in different countries and utilize different hardware. It sounds like a good idea, as it decreases amount of stake in case of slashing, increases overall network security and brings voting power diversification as well.</li></ul><p>Do not forget to <strong><strong>re-delegate your ATOM rewards</strong></strong> in order to maximize profits and take advantage of the compounding.</p><hr><p><strong><strong>P2P Validator</strong></strong> offers high-quality staking facilities and provides up to date information for educational purposes. Stay tuned for updates and new blog posts.</p><p><strong><strong>Web:</strong></strong><a href="https://p2p.org/?ref=p2p.org"> https://p2p.org</a></p><p><strong><strong>Stake ATOMs with us:</strong></strong><a href="https://p2p.org/cosmos?ref=p2p.org"> https://p2p.org/cosmos</a></p><p><strong><strong>Twitter:</strong></strong><a href="https://twitter.com/p2pvalidator?ref=p2p.org"> @p2pvalidator</a></p><p><strong><strong>Telegram:</strong></strong> <a href="https://t.me/p2pvalidator?ref=p2p.org">https://t.me/p2pvalidator</a></p>