First Came Swaps
It all started with Swaps. Bespoke contracts to give you the desired risk profile. Trouble is finding someone who wants to be on the other side of that risk. The brokers solve that challenge by finding you a counterparty. Of course, once you find a counterparty, you are not only managing risk of the trade, but of the counterparty themselves. How credit-worthy are they?
Then Came Futures
Enter Futures trades. Commoditized risk instruments with liquidity and a clearinghouse to buffer from catastrophic events to any counterparty (that’s mostly true). We’ve been in this era for much of this century with greater acceleration post-2009 crisis, largely driven by regulations on over the counter trading born from the 2010 G20 Pittsburgh agreement.
The one tradeoff that comes with moving from bespoke products to commoditized ones is being limited in building a risk profile that doesn’t quite line up with available futures products. So what has the market done to help?
Well, ICE as a futures exchange now lets traders create custom products comprised of futures/options legs. In other words, futures and options are lego blocks. Use them to build the structure you need. While this doesn’t perfectly address the limitation above, it does bridge the gap. Whoosh! Just like that the standardization just got complex.
I’m not going to dive into the legal structures and rules built up to run the futures markets, but instead focus on the tech side that accompanied.
Exchange Data and Our Data
Moving from swaps to futures meant the exchanges starting providing data feeds for your activity. You may have heard terms like FIX API, Drop Copy, STP, Trade Capture, end of day feed, etc. The idea is that commoditizing risk management isn’t just about the instruments themselves, but about the end-to-end experience. If it’s a standard instrument with standard info, then exchanges can provide that info to trading firms so humans don’t have to enter data into systems manually all day. Yeah for automation!
What do you do if your older risk system still thinks of swaps and not futures? You need a system to buffer you from the constant changes at the exchange level and to book trades the way your system models them. We have been addressing this problem and others like it with K3 for a decade. But the key point is that you need some type of automation to keep your business running with minimal technical risk. Otherwise every time the exchange makes a technical change, or your firm trades a new product, your real-time risk management will grind to a halt. I kid, but it happens every week. Some executive team says, “Gee I’m really glad we spent $10mm on that sports car that runs like a jalopy!”
Moving from vanilla futures to strategies made up of futures/options also has a major technical change. It means that your system now has to be ready to accept a trade with any number of legs and risk. In the ICE world, it’s called “UDS” or “User Defined Strategies.”
It’s great for traders and markets, but a nightmare for systems and technologists. In a world where systems function based on knowing what to expect, how on earth can we keep things running smoothly and instantly with unanticipated inputs?
Answer? Represent strategies in risk systems as not one thing, but using the underlying lego blocks of futures and options. Simply unite them with metadata so users can view them together.
Know it or not, you may have already dealt with this via spreads and strips. Many risk systems have native support for these composite trades, but underneath they are still simply just futures trades.
A 12 month calendar strip could be represented as a single thing starting on Jan 1 and ending on Dec 31. But it could also be twelve 1 month trades unified by an order id. Traders may have strong opinions about this, but moving towards booking trades in the most granular way means never having to run your sports car in jalopy mode. It’s a buffer from all the things a trader or could do in the market or changes the exchange could make.
Make My Jalopy Go Fast
When advocating this approach of granularity, we often hear the resistance that this would cause an explosion in the amount of data in a company’s risk system. No doubt that can cause batch computations to slow down. So how do we help not overwhelming the jalopies?
The real point of leverage is not in the timeline compression of strips and leg compression of spreads, but actually in compression of fills as received from the exchange. As a trader puts out an order for 100 lots of a particular futures contract, that order will be fulfilled by multiple counterparties. Trader A may want 10 of those lots, Trader B another 30, etc. etc. until all 100 lots of the original order are “filled.” Each of those fills will come through the exchanges data feeds as an individual fill. In today’s market of increasing algorithmic trading, most large orders get filled 1-2 lots at a time. So compressing these fills into a single trade downstream offers the best point of data aggregation to prevent jalopy mode in your systems.
We’re Here to Help
There’s lots of jargon and info to digest here. One of our core principles at BroadPeak is to “Seek knowledge, Give knowledge for free.” So go ahead, email with any questions: firstname.lastname@example.org