Structural changes in the credit markets could bring to an end a prolonged period of strong correlations. What are the implications for fixed-income investors?
4 minute read
A curious phenomenon is being observed in the financial markets. Since the second half of 2016, equity correlations have crashed to multi-year lows, prompting analysts to hail the advent of a stock-picker’s market. In credit, however, the story is very different.
Corporate bonds remain highly correlated, moving in lockstep in response to technical factors rather than the underlying fundamentals of individual issuers. As figure one illustrates, credit dispersion remains only slightly higher than it was in the aftermath of the financial crisis, even as equities have become vastly more differentiated.
Figure 1: CDX investment-grade credit dispersion (inverted) vs S&P 500 realised correlation
The main cause of this disparity is well-known, with central bank activity still exerting a disproportionate influence on credit. By flooding the market with liquidity, quantitative easing policies have dampened volatility and tightened spreads within and across different sectors. But as central banks begin to withdraw stimulus, this dynamic could finally change, bringing risks and opportunities for investors.
“We’re probably past the point of peak liquidity,” says Joubeen Hurren, fixed income portfolio manager at Aviva Investors. “Although the Bank of Japan and the European Central Bank (ECB) are still implementing quantitative easing, policy is gradually normalising across the globe. As a result, we expect correlations to fall and credit to slowly start trading on the basis of fundamentals again.”
A tale of two dispersions
In credit markets, there has historically been a positive relationship between credit dispersion and overall spread levels. Simply put, when spreads are compressed, the market tends to be more strongly correlated, and this is currently the case in investment-grade credit.
James McAlevey, senior portfolio manager, fixed income at Aviva Investors, says strong correlations can pose significant challenges for bond investors. “When the market is highly correlated, it becomes more difficult to achieve diversification within a fixed-income portfolio. It also becomes more difficult to generate alpha, because pricing doesn’t always reflect the difference between good bonds and bad ones.”
Equity correlations, meanwhile, have fallen precipitously over the past 12 months. One function of this trend has been a ‘decoupling’ between equity and debt prices, most strikingly evident in the US energy sector. Fears over weak oil prices have spooked equity investors in oil and gas companies in 2017, even as bonds from the same issuers have proven resilient. This was particularly evident in the first five months of the year, when shares of energy-sector companies in the S&P 500 fell by 11 per cent, wiping more than $165 billion off their value, while energy-sector bonds rose by three per cent (see figure two).
Figure 2: US energy high yield spreads (inverted) vs S&P 500 energy index
This divergence is partly a function of the outsized impact of monetary stimulus on credit. By buying bonds and holding interest rates low, central banks spurred a flow of capital into riskier parts of the market, causing spreads to tighten. Companies in the oil and gas sector are now issuing bonds at comparable levels to those that prevailed in 2014, despite the fact that the price of crude has plunged 50 per cent in the intervening period.1
When the tide goes out
But Hurren believes the dynamics causing this disparity in correlations between the equity and debt markets are set to change, and that credit correlations will begin to fall over the next 12 months or so, especially if – as may be the case – rising inflation causes central banks to hike interest rates more quickly than the market expects.
“The impact of quantitative easing and easy monetary policy is twofold in relation to credit markets,” he says. “On the one hand, there is the direct impact of the central bank literally taking supply out of the market. On the other is the signalling effect, in that the central bank is demonstrating it wants to keep a lid on the cost of capital to support the economy, which boosts investor confidence. Markets may be underestimating how much spreads can move when those two factors are taken away.”
In fact, the turning point in monetary policy may already have arrived. The Federal Reserve has raised rates four times in the current cycle and is already unwinding its asset holdings. On November 2, the Bank of England raised the official bank rate from 0.25 per cent to 0.5 per cent and signalled its intention to implement two more hikes before 2020. And the ECB confirmed in October that it is winding down the scale of its bond-buying programme (although the bank’s president Mario Draghi said it would continue to buy “sizeable quantities” of corporate bonds into 20182).
During this period of strong correlation, the resilience of issuers’ balance sheets – as well as other factors such as their geographic location and the regulatory regime they fall under – have been all but irrelevant to bond pricing, favouring passive, beta-driven strategies based on simple market exposure.
But as central bank liquidity is withdrawn, fundamentals will come to bear on valuations once again, and may expose weaknesses in riskier parts of the market. As Warren Buffett famously put it, “it’s only when the tide goes out that you learn who’s been swimming naked”.
The decompression trade
But if the shift in the market brings risks, it may also present opportunities, says Hurren. Using credit derivatives to buy protection on sectors in which valuations are out of step with fundamentals – such as oil and gas, and autos – while selling protection on the wider index offers a way for investors to profit if these bonds reprice relative to the market. Given the low levels of market volatility, such trades are relatively cost effective to implement.
Fixed-income investors should also be aware that dispersion is likely to increase within sectors, favouring companies with stronger balance sheets, says McAlevey. By way of illustration, he points to a possible divergence between US and Australian financial companies, which is not currently reflected in bond valuations.
“Credit markets are currently pricing US and Australian bank risk at the same level, and this overlooks some big differences,” he says. “US banks currently look relatively strong, given that they have been compelled by regulators to bolster their balance sheets in recent years.
“By contrast, Australian banks hold a lot of residential real estate assets on their balance sheets in what looks to be a slowing economy. If there is a housing bubble forming, banks would be particularly vulnerable. There is much greater risk in the Australian market than in the US and this will start to become clear in valuations once correlations begin to fall,” McAlevey adds.
The return of fundamentals
There are signs that correlations are already beginning to break down in some sectors. Dispersion among US high-yield bonds is rising compared with investment-grade credits (see figure three), partly due to widening yield spreads among bonds from US high-street retailers. Investors are increasingly concerned about the disruptive impact of e-commerce on those companies, whose challenges were starkly illustrated by the bankruptcy of highly-indebted retailer Toys ‘R’ Us in September.
European bonds in the retail sector have yet to correct in the same fashion, which may reflect the effect of the ECB’s quantitative easing programme. The ECB does not have an internal process for credit analysis, and in buying bonds without paying heed to underlying fundamentals it continues to distort the market. The ‘yield to worst’ on the ICE Bank of America Merrill Lynch Euro High-yield Index sank to below two per cent in early November, an all-time low and reflective of a market driven by technical factors.
Investors should guard against complacency, however. When stimulus is withdrawn careful due diligence on individual issuers will become paramount. “After a period in which security selection was hamstrung in a high-correlation, low-volatility environment, fundamentals will make a comeback as the key driver of returns,” says Hurren.
Figure 3: Dispersion in US investment-grade credit versus high yield
1 ‘Energy debt sells like it’s 2014 even with crude oil at $50 a barrel,’ Bloomberg, October 2017
2 ‘ECB will keep buying substantial quantities of corporate bonds,’ Financial Times, October 2017