A good bond trader once required the skills of a good poker player, constantly assessing the odds of a win against other players. Then, with great subtlety, committing huge amounts of capital to finesse a winning trade. Some even relied on an encyclopaedic knowledge of every inquiry in every bond traded, just to gain that edge over their rivals.
The 2008 meltdown blew a hole in this approach with the resulting regulatory pincer play of the Volcker Rule and MiFID II hugely restricting a dealer’s ability to provide capital, and obliging prices on certain bonds to be published immediately to level the playing field. Well that was the plan, but the reality is that these rules have not really made bond markets a more transparent or liquid place to trade for fund managers.
Under MiFID II, there are currently 439 (up from 220 last year) “liquid bonds” out of 70,000+ that need to be made public before and after the trade. The issue here is that those bonds that qualify as “liquid” are hardly in need of further transparency assistance, and don’t even suffer from poor liquidity. It’s the remaining 70,000+ that need help. That leaves an entire universe of bonds that are not designated liquid because they trade so infrequently. And with ESMA’s liquidity criteria still far too limited, there is very little prospect of rule makers increasing transparency, let alone liquidity for these bonds.
So, what is the answer if there is one at all? The answer lies in all the interactions that brokers and hedge fund managers conduct with each other every day. Also known as the encyclopaedic knowledge of every bond inquiry. If it gave the bond trader the winning edge, imagine what combining it with a proper industrial scale neural network of information gathering and analysis could achieve? Never has there been a more pressing need to bring this concept into the bond trading world.
All of this data has existed for years and more is generated every day. Some estimates state over one billion data points per day. It consists of vast amounts of information from trades that fail and most active dealers, to requests for quotes and who is responding most to prices. But that’s the easy part. The hard part is collecting all this insight from an almost infinite number of sources in a range of formats so complex – that it challenges even the most erudite data scientist.
Compare this to what ESMA collects and the outdated analysis and one begins to get a picture of how little impact MiFID II has had on the wider bond market now, and in the future. Understanding how combining this level of data integrity and granularity, combined with dealer capital, can significantly enhance the liquidity of bonds. It can also be of great help to asset managers analysing the liquidity of their portfolios and evaluating the probability of executing a trade without having an impact on the wider market.
Let’s face it, a liquid corporate bond market has never been more important. Businesses across the real economy rely on it for much needed finance, evermore so given the withdrawal of banks from large swathes of direct lending activity. Investors need the most efficient route to transfer asset infrastructure in order to mitigate the ongoing ultra-low interest rate environment. This can only be done in a world where more liquidity is based on the ability to analyse vast swarms of trading information. Call it the new age of “digital capital” which has very much arrived - but not thanks to MiFID II.