Three radical new shifts in bond-market structure

By:
Peter Lee
Published on:

Algorithmic trading comes to the government bond market, while large virtual networks compensate for reduced dealer balance-sheet holdings of corporate bonds.

In its May edition, Euromoney reports on three radical new initiatives to address illiquidity that highlight the emergence of new structures in the fast-changing secondary markets for government and corporate bonds.

Quantitative Brokers (QB), an agency broker in the interest-rate futures markets, is bringing trade execution algorithms to the cash market in US treasuries in a radical development for the bond markets.

The initiative comes just as the IMF highlighted in its April global financial stability report the growing problem in the bond market of poor liquidity caused by reduced dealer inventories. The IMF warns that: "Rising liquidity risks could amplify shocks and complicate the exit from extraordinary monetary policies."

In response to higher capital charges on retained bond inventory, most large dealers have sought to internalize customer order flow. When that takes place across a large bank’s own treasury as well as its entire customer sales and trading franchise in government bonds and also credit and mortgage bonds, where trades might often include a government-bond leg, these can be quite sizeable dark pools.

QB has just gone live with three large bond-market dealers, running its trade execution algorithms for clients across each of these banks’ internal pools for Treasury bonds.

Robert Almgren, who co-founded QB in 2008, says: "Depending on the client’s time constraints for executing the order and our systems’ reading of market micro-structures and likely movement of prices, there will be times when our algos will aggressively cross the spread at the best price available provided by those dealers, but there will also be occasions when we will passively provide liquidity or wait for other buyers or sellers to accept our pricing, and so our clients will collect spread."

Allowing asset managers to capture spread is a radical shift in the bond markets, coming as investors look to reduce trade costs and slippage that can otherwise eat into investment alpha in a low-rate world.

Ryan Sheftel, global head of fixed-income e-commerce at Credit Suisse, puts the decision to open up the firm’s US Treasury pool to the QB trade-execution algorithms in the context of a bigger shift in rates markets towards electronic trading.

"Electronic trading doesn’t suddenly make illiquid markets more liquid," he says. "But where markets are already sizeable – and the US Treasury market sees $500 billion a day in turnover – electronic trading can make them much more efficient.

"What the QB trade-execution algorithms in treasuries show is that it is in the larger, more liquid government bond and interest-rate derivatives markets that the biggest innovations are taking place."

In the corporate bond market, dealers are trying to capture data on missed client queries that might help salespeople broke the markets better, and put buyers and sellers in touch.

Algomi, a firm founded in 2012 by Stu Taylor, Robert Howes and Usman Khan, who had previously worked together on the UBS PIN network for matching client orders, is now beta testing with 20 buy-side accounts a new product to link them into the internal networks of certain banks that grant permission.

So a French client trying to buy VW bonds might leave an indication of interest with the biggest bank dealers. If some time later an Asian salesperson at one of those dealers gets a client interest to sell those VWs, an alert will now transmit through the bank in question and back to the French investor looking to buy. As such, it massively improves the odds for both end-clients and the bank in the middle.

Codestreet, like Algomi, builds systems for dealers in the corporate bond market to discover and exploit opportunities to cross orders internally, allowing them to conserve balance sheet.

It is now launching Codestreet Dealer Pool, to link dealers and allow that crossing of client orders to take place across a much larger network of inter-connected sell-side firms.

Howard Pein, chief executive of Codestreet, points out: "The beauty of the Codestreet Dealer Pool is that the liquidity needed to crystallize trades is already latent on our platform.

"Dealers are shown round-lot opportunities, while still preserving their clients’ anonymity and preventing the information leakage that can be so damaging, especially with illiquid bonds."