Interview: Paul Benson, Head of Systematic Fixed Income at Insight Investment

Spreads on high yield debt are currently tight compared to historical averages. Does that make high yield expensive?

Valuations is the main concern we’re hearing from clients right now. They're tight compared to history, but it is important to understand that the Bloomberg US High Yield Index is different to 20 years ago. It’s a higher quality universe now; you have less CCCs, more BBs and bigger companies giving a larger overall market cap. So, of course it's going to look rich because you're comparing a higher quality index today versus the lower quality index of history.

One reason for this change is that a lot of the lower-quality companies have left the high yield space to the loan/ leveraged loan market and private debt/private equity space where there's a huge infusion of capital, so they don't need the high yield public debt market. Although returns cannot be guaranteed, high yield even today yields almost 8%¹ that's offering potential for equity-like returns with a fraction of equity volatility.

Another reason spreads are not wider is that they simply reflect robust corporate balance sheets. Leverage is contained, interest coverage ratios remain strong, and defaults have yet to rise meaningfully

What will happen to high yield debt if there is a recession?

We have had the most telegraphed recession we've seen for a long time; everyone keeps waiting for it to happen, but the economy keeps surprising on the side of resilience. This is probably hurting investors, as they’re again too cautious about just the spread relative to history and not thinking about calculating expected returns from each asset class.

One thing that is under-evaluated is the power of carry. So, you could ask: “What is the cost to sit and wait for the worst thing to happen?”. In equities, if you get it wrong, it may take longer to climb out of; but by contrast bonds revert to par. When yields are low, investors can wait patiently for an exogenous event to drive spreads wider. When bonds are yielding 8%, sitting on one’s hands for a lengthy period of time may become painful.

How are you positioned in terms of fears of an uptick in high yield defaults?

We think that there is way too much fear built into the marketplace right now. Due to this market-wide misunderstanding and aversion to default risk, most managers tend to have low beta exposure to the broad high yield index. Managers are typically defensive and tend to avoid CCCs, which may lead to them excluding 10% to 15% of the universe. If you do that, you're likely to underperform, in our view.

High yield has had better realized returns historically than expected returns, in our experience; it has been a very positive excess-return producing segment, so in years when the markets are doing well, which is most years, managers tend to underperform.

Our basic philosophy is our clients are paying us to get exposure to this asset class, which has done phenomenally well historically, so we're going to make sure that we get full risk exposure and still help protect the portfolio based on the models that we use. Also, in our view, getting the full exposure to the high yield risk premium is going to give you potential for a much better return and risk-adjusted return profile over the long run.

What is the average default experience for high yield bonds?

In our experience, the default metrics discussed in the financial news have almost nothing to do with what institutional investors are investing in. We generally see that the typical investor thinks the historical high yield default rate is at 4% to 5%; in reality, it is much less.

Much of this disconnect is due to the use of default numbers from ratings agencies which tend to calculate issuer-weighted metrics from a very broad universe of high yield. However, the high yield universe is much narrower. By our estimates 90% of managers invest in two main universes that consist of market-cap weighted exposure to 2,000 bonds and just over 1,000 issuers: the broad diversified Bloomberg US High Yield Corporate Index and the ICE BofA US High Yield Index. For the Bloomberg US High Yield Corporate Index, we calculate the annualized default rate since 2005 to be roughly 2.5%², by contrast, we observe that rating agencies can quote anywhere between 3.5% to 5%. And while high yield investors should rightly be focused on estimating defaults, we believe a diversified high yield investor should be focusing more on the expected loss given-default, i.e. what is my actual loss after a recovery rate has been calculated? Does my expected default rate include more benign distressed debt exchanges? What am I expecting to earn on top of raw yields given high yield issuers’ wonderful propensity to call bonds prior to maturity?

How do you overcome the difficulties around illiquidity in high yield bonds?

In our experience, liquidity has never been better. We trade through brokers in the ETF sector where ease of trading has continued to evolve over the past five to 10 years. We have found that with the right technology, tools and relationships, we can trade custom baskets of up to 1,000 bonds worth as much as $500m within hours if not minutes; we can sell this way too.

The asset class is only illiquid if you’re trading in traditional ways, by calling brokers and trading one bond at a time which can take days or a week to implement. This can cost 60 to 80 basis points for each trade, in our experience, whereas when trading large custom baskets, the fee is in the range of 10 to 20 basis points for each security.

High yield is a beautifully inefficient asset class due to the untapped exposure that some managers are not buying into. These are smaller companies, smaller issues, illiquid bonds, fallen angels, private companies. It's beautiful, because if you can get exposure to it and manage the risk on the margin, you potentially enjoy very strong returns.

How does your trading system impact the portfolios you create?

One of the advantages of our trading system is that we can increase the size and diversification of our portfolios. So, rather than choosing a concentrated portfolio of 200 to 300 bonds recommended by a credit analyst team, we look at the whole universe of 15,000 bonds and invest in say 2,000 to 4,000 of those. So, we get a really diversified exposure to that universe.

One of the advantages of this extreme diversification is that if we own a bond that defaults or is downgraded, it is much less costly than in a more concentrated portfolio. And recall from our discussion on defaults that high yield investors tend to be compensated for the default risk.

Can you describe the systematic models Insight uses?

Our investment process uses systematic and quantitative models. In the fixed income world this tends to represent a narrow segment of structural alpha opportunities, as opposed to equities where there can be a wide ice cream parlor selection of all sorts of factor flavors. While the academic research can show highly seductive back-test results, we believe it is critical to consider implementation, and ensuring models are representative of the real world.

In terms of a quantitative approach, we find best results in a bottom-up approach, where we're looking at each company or bond on a relative basis. We avoid making decisions where we don’t have strong empirical evidence of predictive power. One perspective is that from a top-down basis we can adhere closely to the client’s desired asset class and sector exposure.

What are one of the advantages of a systematic approach?

One of the ways our systematic approach works well is protecting the portfolio from downgrades and defaults in a cost-effective manner. The model scrapes balance-sheet information like leverage, profitability, and earnings metrics on a daily basis and so can compare each company on an equal basis, providing us with accurate results for understanding which companies are at greater risk relative to peers.

This is very cost-effective downside protection, whereas discretionary managers are likely to pay through the nose either through a defensive low-beta profile or explicit downside protection.

Our time-tested models have tended to produce alpha during sell-offs due to this tilt to higher-quality issuer profiles; however, importantly when the market reverses, we may fully participate in the upside with a beta 1 exposure. This convexity profile and consistency of performance is a key value proposition to our clients.

What is special about your investment team?

Traditionally, large high yield managers have a star manager with a credit analyst team. We hire people into our Systematic Fixed Income Team who are collaborative and not in it for ego, but are in it for developing something unique and differentiated. It has now been 15 years with the same core team here.

When we started, we combined skill sets in managing index funds, ETFs and active systematic strategies to create the team we have today. Then about a decade ago, we decided to put our researchers, portfolio managers and traders in one room to work closely together on developing our innovative approaches.We're really passionate about the value of systematic approaches in fixed income, and in doing something that no one’s done before.

As a team, we manage almost $10 billion of pure quantitative, fixed income strategies.