TradFi vs. DeFi: How is IRS transacted in Traditional Finance vs. on STRIPS
In traditional finance, interest rate swaps are the largest market in the world, trading over $6.5 trillion dollars a day (source), with over $466 trillion dollars in notional outstanding (source). Despite being the largest derivatives market in the world, the rates market is still being traded through decades old infrastructure, through phone calls and chat rooms. Settlement is fraught with many manual processes and prone to human error.
In Trad-Fi, a plain-vanilla IRS transaction’s life cycle:
- On front desk, the buy-side trader will ask for bid / offer from multiple banks, usually 3–4 on Bloomberg for normal size, if the trade size is too big, trader will ask for quote from fewer people to avoid leakage. Sometimes when trader asked for large size, the market will move before the trade is done
- Then traders will check the bid/offer comes back, usually traders tend to write on the paper and circle out the best bid or best offer. This is important when someone traders trade complicated structured products such as asset swap baskets or swaptions. It is never intuitive to find the best “price”, sometimes it takes a while for the most experienced trader to spend couple seconds until they can figure out if they should look for a higher or lower number of the left or right side of the quote they received. Recently in many developed and some of emerging markts, traders can use TradeWeb to get quotes from few banks in standard format, called “live streaming”, instead of using Bloomberg, which also improved the process.
- Sometimes, the trader will not be asking the “correct” banks. Each bank might have large net positions in some markets, and if the macro trader asked those banks, they would not necessarily get the most appealing price in the market, so typically traders will need to ask banks which markets they can show better prices at beginning of the day. At the momne,t there is no effective way yet for banks to show their axes yet (axes means banks can show better price in certain markets/tenors).
- There are also latency issues: sometimes it takes longer for 1 out of the 3 banks to come back on price. Sometimes, after receiving the last quote from Bank C, trader need to refresh the quote from Bank A, because the market might have moved away already, and if Bank A price deteriorated, trader will need to refresh Bank B and Bank C again to see if they can provide better price, which can take much longer, especially in an illiquid market
- Each market has its unique nuances of trading, so experiences in trading one market might not be directly applicable in the other markets.
- All trade confirmation still happens on Bloomberg for the most, or on phone if the trade is sensitive to time and size, so quite often it can cause miscommunication.
- To summarize an IRS on front-end only, it is a highly specialized job and technically there are many challenges from:
- Hard to tell which is the “best” price, especially for structured products
- Latency in quotes, miscommunication on the trades
- Information leakage
- Not asking the right bank due to lack of transaprency on overall positioning
However, these steps above are on the beginning of an IRS transaction. After the trade is confirmed verbally on Bloomberg chat, the trader needs to complete following steps
- Trader will need to copy paste some lines into the chatroom for details of the trade including exact notional, expiry, currency, etc.
- If it is an initiation trade, things will be slightly easier such that the bank will send an email confirmation of the trade based on the details received and the final price confirmed.
- However if it is an unwind trade, which triggers some cash settlements, then banks will need to use their tools to calculate the net cash amount to settle this trade, and send it back on Bloomberg chat, and wait for the macro traders to agree on the cash settlement amount.
- The macro trader will use their own tool, usually excel spreadsheet to calculate what they think is supposed to be the correct net present value, and sometimes there can be big mismatch between bank and the trader. There leads to a technical problem of IRS trading in trad-FI. The best price doesn’t necessarily mean the best cash settlement price, and hence it might take 3 times longer to negotiate on the cash settlement amount more than trading the IRS itself. Hedge funds tend to be more advanced in technology by using proprietary platforms or external vendors that calculate interpolation and net present values, escaping away from spreadsheets. However, “real money” shops, such as most of asset managers who manages your pensions, still suffer from “spreadsheet”.
- After receiving an email confirmation about the cash settlement amount, then the trader will need to manually filling the net cash amount for each trade, line by line, and sometimes one mismatch in line can cause settlement issues on back office later. After manually keying in the cash amount for each trade, trader will wait for the system to reflect and record. There were few times when trading volume was too big, and the system was down and it is extremely hard to recover and figure out which trades have been dropped out in the system. Similarly, hedge funds can be more opertioanlly efficient.
- Not to mention that traders must deal with compression baskets which also takes many manual process to collapse small risks among multiple banks. This market worth 1800 trillion in gross notional, which also indicated some inefficiency in the process.(source) We believe TradeWeb is also working hard to resolve this problem in Trad-fi, and blockchain provides an another possibility.
Once the trade moves from front desk to back office:
- Back office will need to confirm the records in the system against the email confirmation, and if there is any mismatch occurred among
- internal system on which traders record their transactions
- external system from which back office will receive the trade details from banks for matching purpose
- email confirmation
Finally, this transaction will be pushed back to front desk for further amendment.
- Once matched on back office, then switch instructions will be sent out, if needed, and both parties will receive the USD amount after couple days
On block-chain, the first 7 steps are conducted in one-shot on trading page by receiving the quote from AMM and clicking “confirm” button on front end. There is no hassle arising from lack of transparency. The following additional 11 steps will be all accomplished by miners in mempool, finished with few minutes with L2 solution on Ethereum. STRIPS is building the technology to bring traditional finance into the 21st century, we have witnessed how technology re-invented consumer banking, retail investment, payment system, etc. We believe the only stronghold left is institutional trading and STRIPS is trying to create a market place that can trade interest rate as an independent asset, which is catered for crypto native use cases.
While in Tradi-Fi, IRS is mainly used for basis trades such as inflation spreads, even forward vs spot trades, in addition to betting direction on interest rates and hedging. We believe similar demands of these trade strategies are inevitable in Crypto as well. Therefore, STRIPS would be the first decentralized interest rate trading platform to enable all trading strategies mentioned above:
One typical institutional trading user case in crypto is for basis traders. Fr example, institutional traders who find BTC Perp contract at 41967, and BTC-1231 (expiring by Dec2021) is 42720, then traders will try to short BTC-1231 and long BTC-Perp since BTC-1231 will be settled at BTC-Perp price at the end. That indicates a basis trade with annualized return of 7%, with the assumption that the funding cost paid for BTC-Perp is unchanged. If liquidity tightens, which we found happened couple times when interest rate can move more drastically then underlying BTC, then the basis arbitrage’s profit will evaporate. This is only one typical example of how institutional trading demands not fulfilled in crypto space, and we believe IRS will be an inevitably important tool for institutional trading, given interest derivatives is 80% of global OTC financial derivatives market, yielding more than $400 trillion outstanding notional.
This is the first step of STRIPS is to introduce a market place that interest rate, whether from DeFi and CeFi, can be transacted as an independent asset. Traders can speculate on, hedge with, trade on interest rates with any combination of terms, duration and leverage. Interest rate swap is purely a gameboard that empower traders to monetize their trading ideas with creativity and flexibility, in a Capital-efficient way, enabled by robust AMM model, designed by STRIPS.