Bagels Finance White Paper (as of Nov 16,2021)

Content

1.0 Abstract 3

2.0 Protocol Users 3

2.1 Yield Farmers 3

2.1.1 Up To 10x Leverage 4

2.1.2 Auto LP Staking 4

2.1.3 Hight Debt Ratio Solvency 4

2.1.4 Adjustable Positions 4

2.2 Lenders 5

2.2.1 Interest bearing gTokens 5

2.2.2 Earn Interest 6

2.2.3 Utilization Rate 6

2.3 Liquidators 6

3.0 Market Demands 7

3.1 Farmers Expect Higher APY 7

3.2 Lenders Expect High Interest Rates 7

4.0 Gains 7

5.0 Risks 8

6.0 Dokodoa Cross-Chain Protocol 8

6.1 Dokodoa Architecture 9

6.2 Cross-Chain Asset Mapping 12

6.3 Cross-Chain Yield Farming 13

6.4 Cross-Chain Vault 14

7.0 Token Allocation 14

8.0 Farming Allocation 15

9.0 Revenue Allocation 15

10.0 DAO Governance 16

11.0 Team Background 17

11.1 Team Academics and Experience 18

11.2 Investment Institutions 19

1.0 Abstract

Bagels Finance is the first cross-chain leveraged yield farming protocol with multi-chain deployment. Users can deposit ETH, WBTC, USDT, DAI, BNB, CAKE and other assets into Bagels smart contracts to earn the highest APY in the market; Bagels aims to provide users with 2x-10x leverage to earn 2x-10x APY in yield of leveraged farming and transaction fees of LP providers. It provides better liquidity for BSC, Ethereum and Polygon ecosystems.

Bagels optimizes the trading process to maximize liquidity farming positions and the APY for farmers. In addition to high interest rate, users can also earn lending APY and leveraged farm APY, and receive BAGEL governance token rewards according to the share of deposits and borrowings. Token holders can stake BAGEL to participate in DAO's decentralized governance, become a board member of Bagels Finance, and receive 85% of the dividend from the revenue of Bagels platform.

2.0 Protocol Users and Roles

Users can participate in Bagels Finance as yield farmers, lenders and liquidators.

2.1 Yield Farmers

Users do not need to deposit two assets into the LP pool at 1:1 ratio, they only need to deposit one asset to start yield farming. For instance, if users select the ETH/USDT liquidity pool, users can farm by just supplying ETH. The Bagels system will automatically swap the tokens at the best ratio to make sure users have equal value of both USDT and ETH to supply to ETH/USDT liquidity pool on Dex.

2.1.1 Up to 10x Leverage for Genesis Farming Positions

Users can borrow ETH, WBTC, USDT, DAI, BNB, CAKE with up to 2x to 10x leverage, stake and participate in LP farming in the leveraged yield farming pools, and earn higher trading fees and LP farming APY.

2.1.2 Auto LP Staking

Bagels will automatically stake users' LP token to get farmed tokens for users.

2.1.3 Hight Debt Ratio Solvency

As long as the debt ratio is lower than the "KillFactor", the system will not require the user to close the position value. However, users should closely monitor the value of their positions, and add assets/ close positions when the debt ratio is close to the "KillFactor".

2.1.4 Adjustable Positions

Users can choose to add more ETH, USDT or other assets any time. However, users must close the position before withdrawing part of the collateralized assets.

Example:

Alice farms SUSHI in SushiSwap's ETH/USDT pool with 100 ETH. Usually, Alice will swap 50 ETH into USDT, and then deposit ETH and USDT at 1:1 ratio into the pool. SushiSwap provides liquidity reward SUSHI, assuming that SUSHI’s Farm APY is 50%.

Alice collateralizes 100ETH in Bagels and borrow 200ETH to farm so to increase her APY, which is equal to 3x leverage. Alice collateralizes the 300 ETH into the yield farming pool, and the farming yield will be three times the original, and the APY for farming SUSHI will be as high as 150%. Liquidity farmers borrow from the Bagels vault for leveraged farming, and pay interest to the depositors. At the same time, liquidity farmers can gain leveraged farming rewards “BAGEL”. Of course, high APY are accompanied by high risks. As long as the net value of ETH is not lower than the liquidation line, Alice can continue to farm with 3x leverage and gain high APY; Once the net asset value reaches the liquidation line, liquidation will be triggered by the liquidator..

2.2 Lenders

2.2.1 Interest bearing gTokens

Users who deposit multi-assets to the Bagels vault will receive a proportional amount of gToken, a token that can be traded and acted as a certificate for earning interest, representing the user's shares of the Bagels pool. Similar to cToken in Compound lending platform.

The following are the wrapped token “gTokens” generated by multi-assets: gETH, gWBTC, gUSDT, gDAI, gBNB, gCAKE.

2.2.2 Earn Interest

Lenders can earn deposit interest, and also earn BAGEL as rewards. Taking ETH deposit as an example, the interest paid by borrowers will be distributed to ETH lenders depending on ETH utilization rate. (Borrower’s interest rate model follows Triple-Slope-Curve.)

The-Triple-Slope interest rate curve is as follows:

• At 0-50% utilization rate, the loan interest rate will be fixed at 10%

• At 50%-80% utilization rate, the loan interest rate rises to 30%

• At 80%-100% utilization rate, the loan interest rate rises to 60%

2.2.3 Utilization Rate

The depositor's interest rate is determined by the utilization rate of the USDT pool; The higher the utilization rate, the higher the interest rate.

2.3 Liquidators

For user positions whose debt ratio is higher than the "KillFactor", the liquidator can liquidate these risky positions, and the liquidator can earn 2% of the position value as return.

Example:

• A liquidity farmer leverages 2.5x by borrowing 150ETH with her initial 100ETH to farm SUSHI in the ETH/SUSHI SushiSwap pool, with the total position of 250 ETH at present.

• When the price of ETH increases, the position value of ETH drops from 250 ETH to 175 ETH, leaving his debt ratio at 85% (150ETH/175ETH), which is greater than the "KillFactor" of the ETH/SUSHI pool. In this case, the liquidator can come in to liquidate.

• During the liquidation process, the position value will first be used to pay the debt (in this case 150 ETH).

• Then 2% of the value of the 250ETH position is paid to the liquidator. So in this case the liquidator will earn 5ETH.

3.0 Market Demands

3.1 Farmers Expect High APY

Most assets locked in DeFi ecosystem is for yield farming, followed by decentralized lending. Most of the assets borrowed on lending platforms are used to participate in yield farming, which can increase the position's value and thus yield higher returns. Most of the lending platforms are over-collateralized (borrow without leverage). The value of the assets that can be leveraged is limited and thus can’t provide extra higher returns for yield farmers.

3.2 Lenders Expect High Interest Rates

The low deposit rate of over-collateralized lending platforms in the decentralized market are not enough to attract depositors.For over-collateralized lending platforms like Compound, AAVE, and Maker, the ETH deposit APY is 0.3% ~ 0.8%, but on Bagels Finance, the ETH deposit APY can reach more than 20%.

Bagels Finance provides a very innovative way of liquidity mining, which just meet the rigid market demand and make up for the market weak spots. While yield farmers need to use multiple platforms to borrow assets and then to farm, Bagels platform allows yield farmers to complete the entire lending process within just one click, farming the LP pools of all mainstream DeFi projects on BSC and Ethereum to earn higher APR.

Yield farmers can reach higher APY through leveraged farming on Bagels. Depositors who supply digital assets can earn high interest rates by lending them on Bagels. The high rates are based on the lending interest rate and utilization rate at which yield farmers borrow and lend money when they open leveraged positions.

4.0 User’s Gains

• Provide liquidity and earn Bagels governance token (BAGEL) rewards

• Earn trading fees on leverage and Bagels governance token (BAGEL) rewards from providing liquidity to asset pools with different farming strategies

• Earn farmed token on leverage (e.g. MDX or CAKE)

• Provide deposits for Bagels lending pools to earn interest and Bagels governance token (BAGEL) rewards

5.0 Risks

Bagels Finance can bring users higher APY yield farming, but the protocol involves the use of leverage, and users should participate with caution. Users should only deposit funds when they fully understand the risk of assets liquidation. Therefore, even though the Bagels Finance protocol has been fully audited by audit authorities, this does not mean that the use of the protocol is not risky, and the protocol may still have errors and vulnerabilities.

*** Note: Users must "DO YOUR OWN RESEARCH (DYOR)" before using the Bagels Finance protocol.

Higher returns come with higher risks. For either leveraged yield farmers or liquidity providers, there will be risks, including:

• Impermanent loss: As you're providing liquidity on a decentralized exchange, such as Mdex, Uniswap, SushiSwap or Pancakeswap, you will face the risk of impermanent loss;

• Risk from Short positions: Assuming that the price of ETH increases, it will cause the value of your position to decrease. When leverage is used to establish a position, some borrowed ETH is exchanged for another token, so you borrowed ETH, i.e. shorting ETH.

• Therefore, relative to another token (such as the ETH/DAI pool), when the price of ETH increases against USDT, the value of your position may decrease, which may trigger the liquidation; and vice versa.

• The potential risk of contract vulnerabilities may also lead to hacker attacks.

6.0 Dokodoa Cross-Chain Protocol

The cross-chain protocol is called Dokodoa, which means "Anywhere Door" and implicates that a variety of assets can cross to different public chain ecosystems through Dokodoa Protocol, and complete digital asset conversion and smart contract interaction on different public chains. Dokodoa is a secure and decentralized cross-chain solution, composed of state oracle and cross-chain bridge smart contracts. It can support public chains that include signature algorithms using ECDSA and EDDSA, such as Ethereum, HECO, BSC, OKex Chain, Polkadot, etc. Users are able to complete the deposit and cross-chain swap in one-click without switching between different public chain network settings. The cross-chain bridge node automatically identifies the user's deposit transaction, and accordingly triggers the cross-chain mapping contract to generate 1:1 wrapped mapping assets on the corresponding chain.

Example:

If a user wants to move Ethereum assets to BSC, the user only needs to deploy Dokodoa cross-chain bridge smart contract on Ethereum to initiate token deposit to the custodial account. The smart contract will generate token mapped to BSC according to the state change of the custodial account, and then issue mapping tokens to the user’s account on BSC, so as to complete the cross-chain mapping of tokens.

6.1 Dokodoa Architecture

Dokodoa is a decentralized oracle network that's based on State Oracle to monitor the state and data on different chain and to achieve cross-chain mapping. It supports both isomorphic and heterogeneous cross-chain. It is also a decentralized network of oracle that supports multiple chains.

Assuming after Dokodoa network provides data feed oracle services to the Ethereum and BSC, theoretically, smart contract on Ethereum is able to trigger cross-chain state changes, flowing through Dokodoa client nodes, calling into smart contract on BSC. Thus, Dokodoa network is acting as a connector or bridge between the supported isomorphic or heterogeneous chains. A simple application scenario. For example, looking at the liquidity farming pools on Ethereum, the assets on Ethereum can only be used to farm the liquidity pools on Ethereum, and the same is true for BSC. With Dokodoa, the assets on BSC can be used to farm the liquidity pools on Ethereum, or the assets on Ethereum can farm the liquidity pools on BSC. It thus improves the liquidity of the three chains as well as increase the share of TVL for cross-chain liquidity farming. By deploying Dokodoa oracles and cross-chain bridge smart contracts on Ethereum, BSC, and Polygon, it defines the collaboration functions on those chains, so that the Dokodoa smart contract functions on Ethereum can call the smart contract functions on Polygon or BSC from the Dokodoa cross-chain bridge oracles, and make atomic updates to the status changes of cross-chain orders and account balance. The following shows the application and potential of Dokodoa cross-chain bridge in cross-chain interoperability.

Dokodoa network is chain-agnostic and a scalable layer2 protocol, meaning that it could serve all existing smart contract platforms; it is decentralized, meaning that it has no single point of failure, no centralized trust points are needed, the trust lives in math and code. It is horizontally scalable, meaning that with more nodes running Dokodoa client software the whole network offers more capability and computation power to supported blockchains; it is designed with crypto economic models, meaning that the protocol is resistant to sybil attacks and the network effect is expanded with provable credibility.

Dokodoa network is partitioned into two layers with several key components:

● On-Chain Part：A set of Dokodoa system contracts deployed on supported blockchains, mainly including functionalities such as request handling and response/computation result verification, node registration and staking, stats monitoring, payment processing, etc. Smart contracts and developers on different chains can request cross-chain services through the unified interface provided by system contracts on chain.

● Off-Chain Part: A layer 2 distributed peer-to-peer network of clients that implement the core client protocol and are run by a third party user (that is, the node operator). Client protocol includes several important modules: event monitoring and chain adaptor module, distributed randomness engine module, off-chain group consensus module, and request processing/computation task processing module depending on the type of oracle service the user node provides.

We will discuss components of Dokodoa in details in the following parts, analyzing Dokodoa on-chain architecture and off-chain core protocols. For data feed State Oracles, we will demonstrate how the off-chain Dokodoa clients reach consensus in the open and POS environment by means of unbiased verifiable randomness generation and non-interactive, deterministic threshold signatures. Verification is completed under POS (Proof Of Stake) consensus mechanism. For Dokodoa Oracle, we will elaborate how we achieve the same verifiable computation architecture similar to that of Truebit but in a non-interactive way, powered by zkSNARK and the state-of-art technique of randomness engine we build for the data feed oracle to generate zkSNARK public parameters (called the “setup phase”) in a scalable and trustless way.

We take Ethereum blockchain as an example to discuss the overall process of an on-demand data query initiated by a user contract. It looks similar to the request & response mode, however, it is an asynchronous process from the user contract’s point of view:

①: User contract makes a data query request through a message call to Dokodoa on-chain system (a bunch of smart contracts open sourced and published with well documentation provided to developers), specifically the Dokodoa Proxy Contract;

②: Dokodoa Proxy Contract triggers an event along with query parameters;

③: Dokodoa clients (off-chain part of Dokodoa Network run by users), which keep monitoring the blockchain for the defined event, are notified. Ideally there would be thousands of Dokodoa nodes running, out of which a registered group will be randomly selected, by means of the distributed randomness engine built with verifiable random function (VRF).

④ & ⑤: Members in the selected group do the due diligence, calling a Web Api, performing a computation, or executing a configured script concurrently;

⑥: They will reach "in-group" consensus by the t-out-of-n threshold signature algorithm and report back the agreed result to Dokodoa on-chain system, as long as more than t members in the randomly selected group are honest. The selected group members’ identity and QoS (responsiveness/correctness, etc.) performance will be recorded on-chain, for monitoring and data analysis purposes.

⑦: Dokodoa Proxy Contract notifies the user contract that the result is ready, by calling a callback function provided by user contract.

6.2 Cross-Chain Asset Mapping

The Dokodoa cross-chain bridge will be deployed on Ethereum, BSC and Polygon at first. When the users sends the ERC-20 asset to the BSC or Polygon chain, the users only needs to deposit the token on the Ethereum Dokodoa smart contract, and then redeem the asset on the Dokodoa smart contract on BSC or Polygon. To implement this concept, Dokodoa needs to interact with Curve, Alchemix, Orbits protocols and only supports mainstream assets such as ETH, WBTC, USDT, DAI, USDC, etc in the beginning.

Example:

Alice deposits 100 USDT in Dokodoa on Ethereum, Dokodoa's oracle monitors the account status on BSC or Polygon , and the Dokodoa smart contract on BSC or Polygon immediately maps 100 dUSDT (1 USDT = 1dUSDT) for Alice, which is not the same as the token deposited by Alice on Ethereum, but it can be considered as an equivalent token. When Alice redeems 100 USDT on Dokodoa on BSC or Polygon, the 100 dUSDT will be destroyed simultaneously. The liquidity of dUSDT is provided by: Curve (3pool: USDT+USDC+DAI), Alchemix (alUSD3CRV) and Orbits (Metapool=USDT+HUSD+DAI).

6.3 Cross-Chain Yield Farming

The Dokodoa cross-chain bridge allows cross-chain yield farming. Dokodoa is a fully open sourced network that is open to all vaults in the Ethereum, BSC, and other ecosystems. It embeds Dokodoa protocol to achieve cross-chain vaults. Assets on BSC and Polygon can use Dokodoa to farm the liquidity pools on Ethereum, such as SUSHI, CRV, ALCX, and Badger. Users of BSC or Polygon can pay lower gas fee to farm the liquidity pools on Ethereum. The gas fee generated by smart contract interaction on Ethereum will be paid by triggers on Ethereum, similar to aggregation trigger platforms like KP3R. The trigger pays the gas fee of the liquidity farming pool that has been called on the Ethereum on the KP3R platform and receives a 5% faming tax reward from the Dokodoa cross-chain bridge.

Example:

The APY of Alchemix (ALCX) single-asset alUSD Pool is 75% and the APY of alUSDT3CRV Pool is 72%. Users can farm ALCX on the Bagels platform on BSC, and can increase the position value up to 10x to farm ALCX. Users only need to stake DAI to farm ALCX without the risk of impermanent loss, and get 750% APY from alUSD Pool and 720% APY from alUSD3CRV Pool respectively.

6.4 Cross-Chain Vault

Dokodoa also implements a build-in vault. Vault on Ethereum, BSC, Polygon, etc. which can use Dokodoa's smart contract to achieve cross-chain vault. This provides more farming strategies for vaults on BSC, Polygon and other chain ecosystems. Dokodoa will also be deployed on Ethereum Layer 2 and other chain ecosystems in the future.

Bagels also identify itself as a strategy aggregation platform, which aims to provide higher APY in the DeFi ecosystem. Bagels and Dokodoa make a perfect combination. Bagels is a leveraged yield farming platform. Dokodoa solves the cross-chain asset liquidity problems, thus provides better liquidity and more farming strategies for assets on Bagels. In addition to providing better farming strategies on BSC, Bagels also allows users to farm on Ethereum while spending extremely low gas fees from BSC, Polygon ecology.

7.0 Token Allocation

Total Supply Of BAGEL: 110,000,000

• 60% Farming Pool.

• 18% Early Investors: 10% will be unlocked upon mainnet launch, and 22.5% will be unlocked every quarter with a total lock-up period of 12 months.

• 10% Team Reserve, 5% of which is linearly unlocked within 12 months, and the other 5% is locked in the DAO smart contract for 360 days.

• 7% Community Incentives and Airdrops.

• 5%: Marketing & Strategic Partners.

8.0 Farming Allocation

60% of BAGEL tokens are generated from farming, i.e. user behavior. It specifically includes three types: leveraged yield farming, vault farming, and LP liquidity farming.

1. 32.73% of the farming pool supply, a total of 36,000,000 BAGEL is used for the Vault pool. In the first year, 9,000,000 BAGEL will be farmed, with an annual output reduction of 20%, and it will be depleted in 100 years.

2. 21% of the farming pool supply, a total of 23,100,000 BAGEL for leveraged yield farming, every 8 weeks for each farming phase. The first phase of farming pool has 1,830,901 BAGEL, and it will be depleted in 4 years.

3. 6.27% of the farming pool supply, a total of 6,900,000 BAGEL for LP liquidity farming. Users stake two digital assets with 1:1 ratio to provide liquidity of BAGEL/USDT. After obtaining the LP token, the user can stake it into BAGEL liquidity farming pool to farm BAGEL. Be noted that when BAGEL price fluctuates violently in the process of farming, there may be some volatility losses, but the BAGEL farmed may be enough to compensate for the volatility losses. The number of BAGEL in LP liquidity farming accounts for 6.27% of the total supply. It is a pool of three years, every 10 weeks for each farming phase, the number of BAGEL output in each pool will decrease. Users can stake in the BAGEL LP liquidity farming pool for 7 days, 15 days, 30 days, 45 days and 60 days.The longer you stake, the faster the farming acceleration will get boosted with a multiple up to 3x. Users can redeem liquid assets at any time, and the cumulative farming acceleration time will be adjusted and start from 0.

9.0 Revenue Allocation

Bagels platform revenue comes from two parts - Yield Farming and Interest Income, and it’s allocated as follows:

Yield Farming: 85% to veBAGEL holders in DAO Boardroom, and the Bagels team keeps the rest 15% for team operation and product development.

Interest Income: 50% to DAO Boardroom, and 50% to the Bagels team for team operation and product development.

10.0 DAO Governance

BAGEL is the native governance token of Bagels Finance. Token holders can deposit their BAGEL into Boardroom smart contract and join Bagels DAO governance. In the Boardroom, when making key decisions regarding the future roadmap, product and operations of Bagels, like interest rate, crypto value ratio, liquidation penalty rates, etc, BAGEL holders can vote for their preferable proposals.

Joining the DAO governance is a proof that one has become the boardroom member of Bagels Finance, entitling to 85% platform revenue dividend. The dividends generated are in the form of interest bearing gTokens such as gUSDT, gBUSD, which can be converted 1:1 to USDT or BUSD.

When depositing BAGEL into Boardroom, the user will receive veBAGEL, an interest token of Boardroom, which is used to calculate the platform revenue dividend earned by the holder, payable once a week. The acquisition of veBAGEL will speed up the rate of vault farming up to 2.5x, depending on the number of veBAGEL and the number of assets deposited.Stake BAGEL to Boardroom's Board of directors and the lockup time option provides 1 week, 1 month, 3 months, 6 months, 1 year, and 4 years. The longer BAGEL of the same value is staked to the Boardroom, the higher the voting weight and dividend yield. The number of veBAGEL held by Boardroom members decays with the length of time they are locked. The longer the lockup, the slower the decay.

User will receive veBAGEL, a dividend token to receive payouts. veBAGEL will be used to calculate the share of dividend that will be paid out daily or weekly. Dokodoa cross-chain bridge and Bagels are the perfect combination, in addition to the dividends from the Bagels platform, the user can also receive the Dokodoa cross-chain bridge dividends.

11.0 Team Background

Bagels Finance is incubated by Merkle Labs with technical support and strategic direction. Merkle Labs focuses on the development of DeFi products. Since the beginning of 2020, it has developed more than 40+ DeFi products. The R&D team consists of technical teams from US, Canada and China. There are more than 30 technical R&D and developer talents with the team. The team’s academic background combines technical and financial backgrounds from the University of San Francisco, UC Berkeley, University of Waterloo and other North American universities, holds a bachelor's, master's and doctorate academic background; working background comes from Google, Huawei, Intel and well-known investment banks in North America. The team has a very deep understanding and rich practical experience in finance, mathematics and smart contract development.

Denis Sergeychik (CTO)

Denis has a bachelor's degree from Kennesaw State university. He is proficient in JAVA, C++ and Solidity programming languages and has 9 years’ experience of development. He is a full stack engineer. He has worked for companies like Apple, AT&T and VISA, which are Fortune 500 companies. Dennis is also a UX user experience design expert specializing in UX architecture, product strategy, and visual design. He was a senior Director of UX User Experience Design at Apple, leading a technical team of over 20 people. Dennis developed his love of blockchain in 2015. He is also a developer of the early technology ecology of Ethereum, and has contributed a million-level code base to the main network of Ethereum, which belongs to the hardcore blockchain technology developer.

Alex Xin (Engineer & Operation)

Alex has obtained his bachelor’s degree majoring in both Computer Science and Math from University of California, Berkeley. He is proficient in programming languages like C, Java, Python, and Solidity and specializes in mathematical modeling and statistics. Since entering the blockchain industry in 2017, Alex has worked for various blockchain companies and crypto funds in Silicon Valley, gaining practical development experience as well as building a solid network in crypto space. Alex has been a long-time believer of blockchain and DeFi.

Eric Wang PhD (Financial Advisor)

Eric holds his PhD degree in Computer Science from University of Waterloo, Canada, and is proficient at C/C++, Java, Python, Matlab. Eric has worked in the financial sector for 11 years, gaining solid experience in investment, financial modeling, quantitative trading, risk management and large financial risk system development. He has worked in several major financial institutions such as Royal Bank of Canada, Bank of Montreal and Canada Pension Plan Investment Board. Eric is an enthusiast of cryptocurrencies and DeFi. Since the purchase of the first Bitcoin in 2015, he has been investing and researching blockchain technologies.

Emily Sit (Marketing &BD)

Emily holds a bachelor degree from City University of Hong Kong and won full scholarship while completing her master’s degree in Economics at University of Saskatchewan. Emily has worked in American, Canadian, and Chinese companies. She has also traveled to more than 30 countries and has an excellent international perspective. After diving deeply into the concepts behind Bitcoin in 2016, Emily has built a firm belief in that decentralization business model is the future. She spent more than a year to study all blockchain and crypto currency related materials available at the time then began working full-time in blockchain area. She has managed a 10,000 ETH fund for a leading crypto fund and reviewed more than 50 whitepapers. She also helped global brand operation for Polymath, the STO (Security Tokens Offering) platform and helped to hunt high-quality projects to list for CBX.one, a certified digital currency exchange based in Dubai.

11.1 Team Academics and Experience

Academics & Experience

11.2 Investment Institutions

Copyright © 2021 Bagels.Finance