If you’re applying for a technology job in an investment bank, you’ve probably thought of working with the systems that support banks’ sales and trading activities. After all, this is where the highest paid jobs with the greatest exposure to the “front office” are, even though some technologists say they’re best avoided.
Before stepping into an interview for a front office tech job, however, you might want to know exactly what the trading systems do. In this respect, a new report from Greenwich Associates is helpful. Titled, ‘The Technology to Succeed in Fixed Income Trading,” it spends most of its time describing how fixed income trading tech systems function. These are the salient points…
1. As machines take over trading floors, human traders are being supplemented by various types of technology
Human beings are not disappearing from trading floors: they’re just being jazzed up a bit. As “electronification” of trading proceeds apace, Greenwich says there are three key types of technology used to help humans in fixed income trading: autoquoting, real-time pricing systems and low touch hedging tools.
2. Complex algorithms now provide clients with pricing information
Under so-called request for quote trading, clients effectively ask dealing desks in banks to quote prices for particular trades. While traders used to give prices over the telephone, a large portion of RFQs are now handled by so-called “autoquoting systems”.
These systems use mathematical algorithms to generate prices. Before coming up with a price (often in a matter of milliseconds) these algorithms look at public and private data about the security being traded, and add in information from the chart below, including information about the trading counterparty (the client being traded with) and current market conditions. Pricing algorithms also need to take into consideration the dealer’s current exposure to the market to ensure the trade won’t lead to the risk of a big loss. – If a client is a huge buyer, for example, the bank might quote a high price and back off.
As clients demand firm pricing, Greenwich says pricing algorithms are increasingly important source of competitive advantage – and that dealers are increasingly investing in them: “If done properly, such an algorithm connected to the right distribution technology will put the bank ahead of its peers with better pricing, higher margins and reduced risk.”
3. “Internalization engines” are now used to help banks hedge trades
Once upon a time, Greenwich says traders were tasked with hedging (mitigating) the risk associated with their trading activities. To do this, they’d usually either find a customer to take the opposite side of the trade or an inter-dealer broker to pass the risk onto a competitor.
Not any more.
Nowadays, hedging is done via so-called “internalization engines.” These are a sort of internal search engine which looks across a bank’s trades and tries to find ones which will offset the risk that’s been taken on with the new transaction. For example, Greenwich says that if the mortgage desk in New York is looking to buy U.S. Treasuries to adjust the duration of their portfolio, and a corporate customer of the rates desk in London is looking to sell U.S. Treasuries to raise cash, the internalization engine will flag that match and execute the trade.
Greenwich notes that internalization is a big thing for big banks which have all kinds of diverse activities. In future, however, it suggests that smaller banks could club together to create their own internalization pools spread across members of the group (although this would surely raise all sorts of issues about risk sharing).
4. Sometimes internalization engines don’t work though
Although internalization engines are now the go-to hedging systems for big banks, Greenwich notes that they’re not always enough. When an internalization engine can’t find the right matching trade, the bank still has to hedge on the broader market. And this requires another kind of pricing technology.
In this instance, Greenwich says dealers look at “aggregated liquidity streams.” These come both from other peers and so-called “non-bank liquidity providers” like Virtu or Sun Trading. This technology steams prices and puts the banks at the receiving end. If the bank still can’t find an appropriate hedging trade, it will go to anonymous market venues like NEX Group’s BrokerTec and Nasdaq’s eSpeed. Greenwich notes that trading here is, “exchange-like, with anonymous trading and prices on the screen that are immediately executable.” These are seen as a last-resort option.
5. In future, more hedging will take place through derivatives markets
Greenwich notes that today’s internalization engines and aggregate liquidity streams don’t make as much use of futures and options as they ought to. “Market impact, liquidity, fees, and capital impacts should all be used to execute the most efficient offsetting trade, regardless of whether it is a credit default swap, ETF or Eurodollar future at the CME,” says Greenwich. Right now, systems can’t really do this.
6. You don’t need to work in a trading technology team at a bank to work in trading tech
Lastly, Greenwich notes that banks are increasingly buying trading technology systems “off the shelf” from vendor companies. 80% of equity investors now use an execution management system (EMS) provided by a third party. Something similar is coming to the world of fixed income trading.