The Big Yellow Taxi model of liquidity

IFR 2018 1 February to 7 February 2014
6 min read

I DON’T KNOW how often I have had the conversation with traders, investors, journalists and all other comers with respect to liquidity in the bond markets – or, as the case may be, its absence.

There doesn’t seem to be anybody left out there who doesn’t have some story to tell of how even some of the most liquid issues were only trading by appointment and that screen prices should be put forward as candidates for the Man Booker Prize for fiction.

Just recently I was chatting to a senior portfolio manager – himself a former bond trader – who was trying to square the circle of transparency and liquidity and who, after a modest bit of ear-bashing from me, had to concede that the two made uncomfortable bed-fellows in the fixed income market.

I had earlier had a similar conversation over dinner with another former bond and rates trader who is now a nearly full-time pension fund trustee. The latter, not unreasonably, thought that it might be advantageous if traders would be committed to making and standing by two-way prices in a defined number of bond issues.

This would be, for the older ones among us, something to smile at, not least because my friend (he was the best man at my wedding) had traded Eurobonds in those heady days of knock-for-knock markets.

For younger readers, that might not mean very much, even under its other name of head-to-head. But under those rules, every trader would formally declare which bonds he made committal two-way prices in and he was obliged to do so to both clients and other declared market-makers in a minimum and maximum size.

The rule between traders was that if Trader A asked Trader B for a committal two-way in an issue, irrespective of whether it was traded on or not, then Trader B could do the same to Trader A in return. Through this process, price finding took place all the time and as near as is humanly possible in a transparent fashion. “I’m not axed… ” did not exist. Traders knew exactly what the market price was in the issues quoted on their book and woe betide those who didn’t.

This cartel assured transparent pricing and a modicum of liquidity in most issues

SECTORAL TRADING BOOKS – say autos or industrials, but also supras and sovereigns – were split into actives and illiquids. The actives traded head-to-head and the illiquids didn’t. Customer knew which was which and announcements were made when an issue was about to be moved from one to the other, thus giving investors time to sell, should they fear the lesser liquidity provided by the Street on the inactives.

It might now sound as though the Street had a neat little cartel arranged and, to be frank, it did. But this cartel assured transparent pricing and a modicum of liquidity in most issues.

But accurate market pricing is not just a matter of efficient tradability, it is also of significant importance in the sphere of portfolio valuation and audit. It has long been established that, irrespective of what listing a bond issue might sport, the exchange price is usually about as useful as a chocolate teapot.

The “market price” needs to be used for pretty significant purposes, the most important of which is to gauge the liquidation value of a portfolio. Yet, we all know by now that finding a generic price that works, in whatever size, is often impossible.

I have ranted and raved for years over the uselessness of generically pricing bonds over either CDS or asset swap levels – each bond issue has a life story of its own.

SO WHAT OF my chum’s idea of simply moving away from broker-driven markets to return to market maker-driven head-to-head trading? This would guarantee that all bonds defined as active either trade or are at least tested at known prices on a regular basis, even if only within the dealer community, assuring a visible, correct and executable price to investors.

The more I think about it, the more I come to the conclusion that knock-for-knock, even in the closed world of reporting dealers, would be less manipulable than the platform system such as TRACE or whatever its proposed European counterpart will be called, once all acronyms have been tested.

Trading via head-to-head markets is hard work. It requires traders to call each other every morning and again every afternoon to check prices against each other – and if you get out of line and you get hit or lifted.

I have seen short positions get handed around from one dealer to the other getting more and more expensive at each trade until either an investor is motivated to fill the Street short or someone gets bought in. That might not be perfect but is better than having thousands of bonds quoted on “runs” that trade neither on the bid nor the offer.

What, I ask you, is the value of showing a two-way price to the market, neither side of which works? Imagine going into a supermarket only to be faced with empty racks with labels attached telling you what all the products would have cost, had there been any.

Looking back, the head-to-head or knock-for-knock markets worked remarkably well and should truly be reconsidered as a trading model even if it does mean committing the bank’s capital and, far more valuably, the trader’s word.

As Joni Mitchell sings in Big Yellow Taxi: “Don’t it always seem to go, that you don’t know what you’ve got till it’s gone… ”