Right up until Donald Trump decided Thursday night was just as good a time as any to launch 59 Tomahawk missiles at an airbase in Syria, investors were focused squarely on 10Y yields and, by extension, on speculation about what it will mean for Treasurys (and for the dollar) when the Fed begins to shrink its still bloated balance sheet.
Wednesday’s release of Minutes from the FOMC’s March meeting only served to make the chorus of balance sheet commentary grow louder.
And while we and plenty of others have spilled gallons of digital ink discussing the details and possible implications for yields and the dollar of a Fed that decides to effectively substitute balance sheet rolloff for a FF hike (or two) when it comes to tightening, the more general question (i.e. skipping happily over the nuance and mechanics) is this: what does a post-central-bank-“put” world look like in terms of cross-asset correlations? And further: how are the market dynamics we’ve grown so accustomed to going to change?
Well, no one has the answer to that. But what we do know is that eight-ish years of near constant liquidity injections have changed the way markets function in important ways.
So as you wind down your Friday and get set for a well-deserved two-day break from the incessant market and geopolitical headline hockey to which we’ve all been subjected over the last five days, do consider the following brief bit from Citi’s credit team who has a thing or three to say about the above.
What do credit traders look at when they mark their books? Well, these days it is fair to say that they have more than one eye on the equity market.
Reversing the central bank distortions. Much of the correlation not just between these two, but also with many other asset classes seems closely associated with the ongoing central bank balance sheet expansion. As that slows, or even begins to reverse over the coming quarters, we expect the negative impact on credit will be more than proportionate.
Just look at the intra-day correlation between iTraxx spreads and the pan-European equity index, STOXX Europe 600 (in levels). This high-frequency data taken on 10- minute intervals has very consistently shown that spreads tighten when equities rally, i.e. negative correlation (Figure 1). In contrast, although the relationship with bond yields is often quite strong on a given day, over time it is highly irregular, switching signs all the time (Figure 2).
Evidently the causation can run the other way also. In our experience, there is no shortage of equity people seeking informational value in credit spreads. And perhaps credit does have ‘an edge’ on specific issues like funding conditions and/or strength of the capital structure.
Statistically, over the last couple of years both markets have been influencing (“Granger causing”) each other. But considering the relative size, depth and liquidity of (not to mention the resources dedicated to) the equity market, we’d argue that more often than not, the asset class taking the passenger seat is credit.
Yet the relationship was not always so cosy.
Over the long run, the correlation in recent years is actually unusual. In the two decades before the Great Financial Crisis, three-month correlations between US credit returns and the S&P 500 returns tended to oscillate sharply and only barely managed to stay positive over the long run (Figure 3).
While things might not quite revert to the historical norm, to our minds, there are at least three reasons to suspect that the relationship between debt and equity won’t stay this cozy for much longer:
- Risk/reward is skewed heavily in favour of the equity market at these valuations.
- The cycle is maturing.
- And central bank distortions are diminishing.
Why is the strong correlation between debt and equity markets still holding? To our minds, the obvious explanation is central banks. Without turning this into yet another essay on QE and negative rates, we think their policies have overridden normal market behaviour in response to the evolving cycle. As illustrated in Figure 7, a wide spectrum of asset classes has been exceptionally correlated since the end of the GFC when viewed on a normalised basis. The common factor in all of these (approximated by a simple average) in turn correlates remarkably well with the rate of expansion in central bank holdings of securities (Figure 8)
This suggests there may be distortions in all. But to our minds credit is clearly more distorted than most others, like equities, are – especially in Europe. Policies intended to flatten the curve and bring long-dated real interest rates down have left total return buyers with precious little return potential from taking rates risk. Taking credit risk instead has been an obvious choice, encouraging inflows and, in turn, spread tightening.
It’s no accident that this rise in return correlations between equity and credit (Figure 3) occurred almost exactly at the time when the central banks effectively ‘took over’ markets, in our view.
However, with the Fed now tightening faster than the market anticipated not long ago, and our economists expecting that it will cease to reinvest maturing securities in its portfolio from December this year, the unwind is underway. To us, this adds to the asymmetry in risk/reward between credit and equities here.