Janus Henderson Investors: Trading bonds during a time of ferocity and velocity

Here, he reflects on his many decades of experience trading through various crises to explain what a 'liquidity collapse' means and describes what it was like being a trader during March’s periods of peak market volatility.

A colleague of mine recently admitted he needed a thesaurus to find the right words to describe what happened in the bond markets in the early weeks of March 2020. It was so historic, he said, we needed new words. For me, two come to mind: ferocity and velocity. The velocity of the collapse in bond prices during March was ferocious.

I was a bond trader in 1998 when Russia defaulted, through the 2015-16 energy crisis and, of course, the Global Financial Crisis (GFC). The GFC is the best analogy for what happened in March 2020 insofar as it reached so many different bond markets, pushing each to extremes. But what makes March unique is that everything happened so incredibly quickly. Those previous crises played out over a long period of time. Some even took years. This time, liquidity collapsed in most bond markets in a matter of days.

Liquidity is a slippery term

Corporate bond prices usually rise and fall because of the day’s news. Markets are designed to rapidly adjust prices to bad or good news, so while you may not like the price of a particular bond on a particular day, it is the ‘market price.’ A bond can also seem ‘illiquid', which in normal times means it trades infrequently and has a larger bid/offer spread – meaning that brokers will only commit to buying or bidding the bond for much cheaper than what they would commit to selling or offering it. A 2018 Citigroup study found that only 18% of US dollar corporate bonds traded on any given day. So a bond can be ‘illiquid’ until someone wants to sell, at which point buyers appear. But what happens if there are no buyers because nobody shows up to the market?

Usually, the price drops more – and keeps dropping until a buyer is found. You could say that the bond, after falling, has found its new clearing price, albeit one you do not like (unless you are the buyer). But there are two different effects at play. On the one hand, the bond may be re-pricing lower because of a change in, or lack of, information. On the other hand, the bond may be re-pricing lower due to a lack of participation in the market; a lack of information about where buyers are happy to buy and sellers are happy to sell results in caution, and caution results in lower bids and higher offers. When whole markets become illiquid in this way, when prices are falling because of a lack of participants, you have a liquidity crisis on your hands.

The role of the broker

Brokers are the middlemen of the market, the people who connect the buyers and sellers, making the transactions happen and keeping the market functioning smoothly. They are, in a way, a lot like car dealers.

If you want to sell your car, you can take it to a car dealer. But you are probably not going to get the best price because the dealer has to hold the car for a while, using up capital, and is taking a risk that the car is eventually worth less than they paid you for it. You could skip the dealer, do some research and sell it yourself. This is a lot of work, but skipping the middleman should save you some money. The absolute best option is to find a person who wants to buy your specific car today. If you can find that one person, you will get the best price.

In March 2020 – to continue the analogy – your local car dealer did not want to buy cars because they did not know at what price they could sell them. The car auction market, which is the buyer of last resort for car dealers, was probably closed. And nobody is walking onto the lot anyway, whether they are selling or buying. The market just stopped functioning. Not because cars were worthless, but because nobody knew what they should be worth, and the dealers were not going to buy all the cars people might want to sell and hold them till the storm passed. In fact, they would probably be unable to because they do not have unlimited cash. They needed to sell some of their cars to raise the cash to buy new ones.

Bond brokers are not much different. Even if they know that a bond is fairly priced, they cannot buy all the bonds the market may want to sell. It is too much risk. The market could go lower before they could find a buyer and – in March of 2020 – their risk management department may have told them that because the price volatility was so high, they can buy fewer bonds now than they could before. So the brokers drop their bids and raise their offers, discouraging you from selling to them or buying from them. Like a snowball rolling down hill, the bid/offer spread in credit markets grew.

From a spark to a flame

The spread of COVID-19 was an exogenous shock to the market. What happened in March of 2020 was not a correction in a broadly healthy market; it was a rapid, self-inflicted shuttering of the economy. The velocity of the sell-off is important: When people do not have time to do proper evaluations of price changes or to digest the effect the day’s news has on their fundamental long-term outlook, they stop moving, or only transact if the prices look like they provide plenty of cushion.

Rapidly, the entire web of the market is affected, and each part reacts with increasingly more caution. The commercial paper market – which consists of very short-term loans, typically a few days to three months – seized up as each participant acted rationally: Buyers were reluctant to take on risk (i.e. demand falls), sellers wanted to get money quickly (i.e. supply rose) and dealers did not want to get stuck in the middle, holding lots of debt that could fall in price. They are brokers, not long-term holders. At one point, the web got so taut that buyers were reluctant to lend (that is, purchase commercial paper) beyond one day.

The ripple went along the yield curve, with one-week paper following the overnight market, then on to the one-year market, and so on all the way to the 30-year US Treasury. These, the most liquid bonds in the world, at one point were quoted with a 3% bid/offer spread. The effect was worse as you went down the credit curve, with illiquidity in Aaa rated US government bonds setting a bad example for single A, BBB and high yield bonds.

Between weaker demand, increasing supply and understandably reluctant brokers, the velocity and ferocity spiraled. Of course, markets rose and fell on the news during the GFC as well – but there was more time to evaluate the risks and formulate a price. In March 2020, the sell-off was simply too fast for that to happen.

Trading in ferocious markets

In normal times, a lot of bond pricing is done with algorithms as the bid/offer of one bond is priced off all the others. A vibrant, active market provides many reference points to triangulate the right price for one particular bond. Computers are good at that. But in March it became a more manual process: human traders would look at physical screens while they answered actual telephones. COVID-19 has made us all more aware of human relationships, of how, when and why we interact with people. It changed how we did our jobs.

I come in each morning to the blotter of trades our portfolio managers want me to execute. In normal times, some would be quick as they were highly liquid. Others would take some time and some work to get a good price. The best price – just like selling a car – would come from finding a natural buyer or seller, somebody who wants to transact in that particular security. If you could find that person, you could collapse the bid/offer spread – we call it trading “inside the touch” – and everybody benefits. But it is a time-consuming grind.

During those weeks in March, every trade was a grind. Even high quality, large securities took a lot of work. Many of the electronic trading platforms were turned off or unresponsive because with so little information, so little price transparency, there was too much risk leaving a bid or offer out there. Markets were gapping down and up. Some dealers just did not want to, or could not, take more risk. US Treasury bonds, typically moving a few ticks a day (1/32 of a dollar is a tick), were moving two to three points (dollars) a day. So you call. You get on the phone and see if you can trade inside the bid/offer spread by finding the person who wants to trade in that particular bond. And, you often lowered the size of your trade. Do a bit here. A bit there. We did it; we found the liquidity we needed, we got what we felt were reasonable prices to buy and sell, but we had to grind it out.

The long road to normal

It is often said that all financial crises have one thing in common: they end. How, and when, is always different, but even in these aspects the current environment shares similarities to the Global Financial Crisis of 2008. Policy makers, particularly the US Federal Reserve, acted to restart dysfunctional markets by improving their liquidity.

This time, the Fed acted swifter and more forcefully then they did in 2008. The result has been well received by the market, with liquidity first returning in the commercial paper market and money markets generally. As time went by and the Fed added or expanded programmes, liquidity started to return to the mortgage market, the investment grade credit market – even the high yield market. Highly rated companies announced new bond issues. Generally, they were oversubscribed and traded up after coming to market. But it is not over. While new issues have traded better, many existing bonds are often still illiquid. Too much uncertainty remains. Caution still prevails.

While the road back to normal is likely to be long, we believe the Fed will continue to intervene as necessary to ensure that liquidity in the bond markets continues to improve. While liquidity can, and will, dry up in a particular industry or company as a result of news that affects those particular bonds – the energy sector today struggling with a dramatically lower oil price is a good example – we do not think it will dry up because the mechanics of the market stop working like they are supposed to. In our view, the Fed will make sure of that.

Yield curve: this is a line that plots the relationship between a bond’s yield (interest rate) and its time to maturity. The curve’s slope may indicate future interest rate and economic trends.

Fixed income securities are subject to interest rate, inflation, credit and default risk. The bond market is volatile. As interest rates rise, bond prices usually fall, and vice versa. The return of principal is not guaranteed, and prices may decline if an issuer fails to make timely payments or its credit strength weakens.