But just like the customer market, wholesale liquidity is constrained.
KCG, a firm formed from the merger of Knight Capital and Getco, two of the big electronic equity market makers, sees an opportunity in fixed income to act not as an inter-dealer broker, but as an inter-dealer market maker, providing principal liquidity in on-the-run US treasuries through direct connections to the leading banks.
|Isaac Chang, KCG|
This approach may have much to commend it given all the current noise around high-frequency trading. The problem with relying on high-frequency traders as liquidity providers – quite aside from all the concerns over spoofing as a set up for outright market manipulation and the suspicion of firms reportedly investing in microwave towers to boost their latency advantage – is that that liquidity can be easily withdrawn, invariably just when it is most needed.
A bank dealing through a direct link with a specialist market maker can at least measure the market impact of executing orders with that firm over time and gauge the depth, size, cost and reliability of its liquidity provision, as well as saving on brokerage commission.
Chang reasons that this better fits the specialist provider. “No matter how expensive capital is for banks, they will always have more of it than us and cheaper funding, which means they will have an advantage in warehousing less liquid securities” he says. “They also have a much larger franchise, more end customers through which to source flow and offset risk. So we decided to approach this by focusing on capabilities which are trying to be complementary to the banks. We can let the banks outsource to us the operational complexity and technology investment to connect to and execute on the exchange-like venues for on-the-run government bonds.”
One of the reasons why banks might welcome this and why it is so hard for them to find such complementary capacity is that the more tech-savvy of the large banks – think Goldman Sachs, for example – are doing a much better job of automating the internalization of customer orders. That leaves less trading to be done between dealers, each now with almost a worryingly large share of that shrunken secondary market.
Consolidation has come fast. “Three years ago there were many more dealers in US treasuries trading into the platforms and through the brokers,” says Chang. “But today the top five banks have maybe 60% of the market between them, say 12% each, while banks 11 through 20 have maybe 1% each.”
It seems the biggest market of all, US treasury bonds, is falling increasingly dark.
Surely if you’re a central bank or the debt management office inside a government treasury department, you want some insight as to where your bonds are actually trading in real time. US regulators remain hung up on the sheer impossibility of working out what happened in the US treasury bond market during the 12-minute flash crash in October 2014. But if anyone tells you they know exactly how the market infrastructure will evolve from here, don’t believe them.