Are We Taking Too Much Credit?
Alternatives to Investment Grade credit
Investment grade credit is popular among DB schemes, but with spreads near historic lows, it's time to reassess risks and returns. Our new blog helps trustees and sponsors navigate these challenges in this asset class.
Redington's take on investment grade credit
Jump to a section or flip the page to continue reading.
Introduction
ShorteningDuration
Asymmetry
Active vs PassiveMean Reversion
Reversion in Spreads
But it matches cashflows!
What else can you do
Put Spread Selling
Conclusion
Key Contacts
INVESTMENT GRADE CREDIT – ARE WE TAKING TOO MUCH CREDIT?
Investment grade credit has become very popular, especially among DB schemes, and this is not surprising. It’s a simple, liquid asset, with predictable cashflows, and it’s recommended for long-term allocations in TPR’s guidance. What’s not to like?
The spread over government bonds, pretty much across the board, has tightened to at or near the lowest it’s been this millennium. The term spread, here calculated as the spread on over 10-year bonds minus the spread on under 10-year bonds, is also tight, at less than 30bps extra spread for a tripling in duration.
We can also look further back, albeit limited to BBB/Baa spreads, and again the pricing looks tight.
However, this isn’t the whole picture. In absolute terms, yields have risen markedly since 2021, which is partly why spreads have tightened. Investors focused on absolute returns have become more interested since the major rises in rates over 2022. However, the majority of this yield (c.80%) is just the interest rate on government bonds, which are more liquid and don’t have the same credit risk (especially not US treasuries or UK gilts).
Unsurprisingly, lower spreads tend to lead to lower returns. Looking at US long-dated bonds, there’s a very clear link between spreads and future returns. As an example, below we show spreads against subsequent three-year returns. When spreads are low enough, below around 110bps, subsequent three-year excess returns have so far always been negative. Currently, spreads are at around 100bps.
An easy way to reduce risk here is to shorten the duration of the bonds held. At a minimum, this reduces default risk, as we can see from cumulative default rates. More unexpected things are likely to go wrong over 10 years than over three.
Moreover, most of the historic negative returns have resulted from spread widening. So, a lot can be achieved by simply shortening the duration of the bonds held. This also reduces the risk, because the shorter the duration, the lower the price impact of spreads widening; and when spreads are compressed the extra spread on longer-term bonds tends to narrow too. Looking at the probability of positive returns, we can see the risk is lower generally for short-dated bonds, but especially so at low spreads.
This makes intuitive sense, because when spreads are low, there’s not much scope to fall further – i.e. there’s not much scope for prices to rise. For short-dated bonds, that’s not much of a problem, as they run off anyway – but for long-dated bonds it means there's very little scope for upside in returns, while there's still plenty of downside risk.
Doing more quantitative analysis, we note the absolute size of annual moves in credit spreads is roughly 50% correlated with initial spread levels – i.e. spreads tend to move less when they start low. Does that mean low spreads should be safer?
It turns out the answer is no. If we split spread moves into “widening” and “tightening” events, we find that widening events are only c.25% correlated with starting levels, while tightening events are c.-75% correlated. That is, when spreads are low there’s limited scope for them to fall further, but they can still rise a lot. In other words, if prices start high, they can only go up a little, but they can fall by about as much (especially when adding in convexity) whether they start high or low.
This means there is an asymmetric risk/return profile when spreads are low (and prices are therefore high).
In addition to asymmetry, investment grade credit also tends to exhibit a tendency to revert to the mean. Below we discuss the implications of this.
The trouble with talking about mean reversion is there is no one clear definition, and the term covers a variety of nuances. Because of this, we find it helpful to think of two separate ideas: passive and active reversion.
Passive reversion (or regression) simply means that the average of a large sample will tend to the underlying mean, and this happens basically everywhere. For example, an eight-foot parent’s children are likely be tall, but they’ll probably still be shorter than eight feet. This is often misunderstood, and explains diverse effects such as the “player of the month curse” in the Premier League, where a player performs above their mean level briefly, then reverts to baseline for the next month in what looks like a drop in form. This effect is not causal but statistical, and this is because a low value does not make a higher future value more likely, but on average both high and low values are followed by more average values.
By contrast, active reversion implies an underlying force “pulling” data back to the mean. For example, imagine someone stumbling forwards in a valley. Once they start lurching to one side then going further to the same side (up the hill) becomes harder, and going back to the middle of the path becomes easier. Alternatively, imagine two players play a game with hundreds of rounds. After each round, the winner gets a small (but cumulative) handicap for the next round. If they play enough rounds, the handicaps will even out till the next round is basically a 50-50 toss-up. If the winner gets an advantage, we see the opposite effect, something we often see in nature, sport, and business.
The earnings on a non-defaulting bond paying a fixed amount have active mean reversion, as a fall in price doesn’t change the income or capital expected, so it increases the likely returns from the current price. But that doesn’t mean the spread itself has active mean reversion, just the returns.
How can we tell which is happening? Passive mean reversion happens to everything that has an underlying mean, while active reversion doesn’t. The simplest test for active reversion is whether variance is linear with time (or volatility with the square root of time). What we see is a meaningful reduction in volatility for longer periods.
Period
Annualized Volatility(basis points)
1m
81
1y
2y
72
3y
62
4y
57
5y
53
6y
46
7y
43
8y
41
9y
10y
39
That means that spreads actively mean revert – and do so quite strongly. Which means, broadly, that when spreads are low, they’re more likely to go up than down.
So, when spreads are low, you have all the same risk but less upside, and reversion means spreads are likely to rise. All in all, it may be a good time to diversify away from IG.
One of the most common arguments in favour of keeping corporate bonds, even at low spreads, is that they help to match cashflows, and work in a CDI framework.
The problem is, while liability cashflows are reasonably predictable, other cashflows, such as requirements for collateral to cover interest rate, inflation and FX hedges, can be far larger and are not predictable at all.
Low spreads make this worse, because when spreads are low, you need to hold more bonds to get the same return. That means smaller cash buffers. Corporate bonds have interest rate duration, so can somewhat reduce hedging requirements, but the duration is typically only around half that of the liabilities. A representative gilts + 50 portfolio might be 60% in IG (half UK, half US), 40% in cash and LDI, backing a full hedge, and would have the following cashflow requirements under different shocks (with the one-year cashflow for comparison). Without the credit duration, the rates impact would be roughly 30% larger, but this would be heavily outweighed by having more cash available.
In fact, for any investors in GBP credit specifically there is another reason to be wary, and that’s contagion. UK DB schemes buy a lot of GBP IG debt. UK DB schemes have around £1.2 trillion in assets, while the total GBP IG market is around £400 billion. Even ignoring other investors, such as insurance firms, there’s a substantial supply-demand imbalance. If a tail risk event happens in either market, it’s likely to have a big effect on the other, creating a feedback loop. We saw how powerful these can be in the gilts crisis. For context, the UK gilt market is around £2 trillion, so the contagion problem could be worse in credit.
Fortunately, there are plenty of options. We’ve already looked at the simplest option – shortening the duration on a bond portfolio.
A more drastic option would be to barbell the investment strategy. In the simplest case, to keep the same return this could mean replacing a credit portfolio with around 20-25% of the exposure in equity, and the rest in cash and government bonds (to maintain duration). Equities are harder to time, and less sensitive to lower credit spreads (while the ability to borrow money more cheaply is positive). But this broader approach could also use the whole gamut of alternative assets, systematic risk premia, and so on. The key idea is to replace capital inefficient credit with lower exposure to a higher risk, more capital efficient, diversified portfolio of other assets.
However, that may not appeal to everyone. Is there anything out there that’s more philosophically aligned to credit, but less sensitive to spreads?
Taking a step back, what is a corporate bond? It pays you a known, fixed amount most of the time, and occasionally gives you a large negative return. Conceptually, it’s a lot like a put option (the Merton model actually uses this to price bonds as put options on the enterprise value of a company). Given this, you might expect similar performance. But you don’t get it. Looking at US all stocks IG (ICE C0A0), the S&P500, and the Bloomberg PUT index, credit has performed pretty badly for three and a half decades; even scaling down for risk (to match the IG volatility), put selling has performed far better, and has outperformed equities on a risk adjusted basis.
Now, there are plenty of differences, e.g. in sectors, and plenty of technical reasons why passive investing is not a great way to buy credit. But with such a large gap, the simplest explanation is that lots of investors want to buy bonds, and not many want to sell puts. It’s not an arbitrage, but is this at least an opportunity?
Selling puts has plenty of advantages over IG credit. It offers better historical returns, much better liquidity (based on index options), and it has no contagion risk with the UK DB market. For the same level of return, at current spreads, the exposure would be a lot lower and may be lower risk, but selling puts has a lot of tail risk.
So what if you sold put spreads? Basically, you sell a put (say 1m at 97%), then buy one further out of the money (say 1m at 93%). In this example, the worst month you can possibly have costs you about 3.5% (you get a 4% loss on the payoffs, but you still get the net premium), for a roughly 6-8% p.a. yield. You get a monthly payoff chart like this:
Moreover, once you buy the second put, you can leverage it modestly, say 3-4x, as the option gives you a lot of protection. That means you free up more cash. This means you have an asset which:
Frees up cash
Has far better downside protection than IG credit
Has no contagion risk with UK DB, and
Is extremely liquid and easily realisable
The trade-off is a degree of complexity, and the actual implementation can be far more sophisticated. But the basic idea is relatively straightforward, and it addresses many of the key weaknesses of IG credit, especially when spreads are as compressed as they are now.
At some point, credit ceases to make sense. If credit spreads were negative, for example, other than perhaps a bizarre regulatory advantage there would be little reason to buy credit over government bonds. In fact, the cut-off point must be less extreme than that, as it should allow for defaults, transition losses, and pricing risk too.
There’s an argument that we’re not at that point yet, but we might be, and there are plenty of other investments available. Right now, they look more attractive to us.
Alex WhiteManaging Director, Co-Head of ALM
Find me on LinkedIn
Drop me an email
Pete DrewienkiewiczChief Investment Officer
Patrick O'SullivanHead of Investment Consulting Find me on LinkedIn
Paul AckersSenior Vice President, ALM Find me on LinkedIn
Disclaimer
For professional investors only. Not suitable for private customers. The information herein was obtained from various sources. We do not guarantee every aspect of its accuracy. The information is for your private information and is for discussion purposes only.
The ICE index data contained herein is a product of ICE Data Indices, LLC and is used with permission. ICE® is a registered trademark of ICE Data Indices, LLC or its affiliates. The index data referenced herein is the property of ICE Data Indices, LLC, its affiliates ("ICE Data") and/or its third party suppliers and has been licensed for use by Redington Limited. ICE Data and its third party suppliers accept no liability in connection with the use of such index data or marks.
Redington Ltd is regulated by the Financial Conduct Authority. Redington Ltd do not advise on all implications of the transactions described herein. This information is for discussion purposes and prior to undertaking any trade, you should also discuss with your professional tax, accounting and / or other relevant advisers how such particular trade(s) affect you. All analysis (whether in respect of tax, accounting, law or of any other nature), should be treated as illustrative only and not relied upon as accurate.
©Redington Limited 2025. All rights reserved. No reproduction, copy, transmission or translation in whole or in part of this presentation may be made without permission. Application for permission should be made to Redington Limited at the following address – Floor 6, One Angel Court, London, EC2R 7HJ.
Redington Limited (06660006) is registered in England and Wales. Registered office: One Angel Court, London, EC2R 7HJ
Redington Ltd.
Floor 6 One Angel CourtLondon EC2R 7HJ+44 (0)20 7250 3331
redington.co.uk
