A changing fixed-income world: asset allocation challenges

06 Nov 2018

With US rates likely to remain influential, the equity/bond correlation could shift and bonds may offer less hedging protection to cross-asset portfolios. Future returns could also be weaker, and volatility higher, than at the peak of QE.

As markets face central bank quantitative tightening, there are signs that we are on the brink of a regime change across major asset classes. We believe US bond yields are moving structurally higher, driven by: strong US GDP growth, expansionary fiscal policy and future bond supply, less foreign investor demand for US Treasuries, the US Federal Reserve (Fed) moving into restrictive territory and depressed real rates and term premia reversing.

Perhaps the most important implication for cross-asset investors and asset allocators is a possible shift in the equity/bond correlation as inflation and term premia pick up in the move away from quantitative easing (QE). The reversal of the currently high correlations between asset classes suggests a return to an environment favouring ‘asset-pickers’ where portfolio managers can add more alpha again.

We believe the future likely comprises weaker returns than during the height of QE, and importantly, more market volatility. For us as asset allocators, this means being ever more tactical in managing portfolios.

In this note, we explore some of the shifting sands in the macro/market backdrop and the implications for cross-asset investors and asset allocators.

 

Fixed-income markets: brace for impact?

US rates

US Treasuries (USTs) are currently affecting many asset classes. 10-year UST yields have broken out of their multi-decade downward sloping yield channel established since the mid-1980s.

Figure 1: 10-yr UST yields breaking out of 30-yr fixed-income bull-trend & above taper tantrum highs

10-year-ust-yields

Source: Bloomberg and BNPP AM, as of 26/10/2018

 

Interestingly, we find that the push higher in nominal yields this year is almost exclusively driven by higher real rates, while breakeven inflation rates moved sideways. Even during the recent equity market correction, yields did not reverse much, with real yields in particular remaining near their highs.

The break higher in yields on the back of higher real rates could hint at a structural shift in the macro/market backdrop. It is occurring at an interesting juncture:

  • Higher yields are consistent with still-strong US GDP growth
  • Expansionary fiscal policy means a higher deficit, leading investors to seek higher yields
  • Currency hedging costs have risen vastly and are eating into otherwise attractive-looking yield pick-ups: a 10-year UST currency-hedged into JPY yields only around 10bp; currency-hedged 30-year USTs yield only 28bp. This leaves Japanese investors better off sticking with their domestic market
  • Higher Fed policy rates may eventually start to hurt activity
  • The anchoring of yields on the back of the monetary policy actions of recent years should reverse and abnormally depressed term premia could finally revert back to normal.

All of this leads us to expect structurally higher rates for the medium term, even though for short-term trading, we note that short positioning is starting to look stretched and positioning squeezes should not be ruled out.

 

Other rates markets

While UST yields are at new highs above their 2013 taper tantrum levels, Bunds are still below those levels, but that said, the correlation between USTs and other fixed-income markets has remained high. So, with UST yields likely pushing higher, the other markets should follow, especially given their high valuations.

 

Higher yields and equities: shifting correlations?

Many market participants will likely recall a negative correlation between stocks and bond returns in the last two decades. But this has not always been the case. Before the mid-1990s, correlations were positive (Figure 2).

Figure 2: Equity/bond correlation mostly negative since mid-1990s (S&P 500 vs. UST weekly return correlation)

equity-bond-correlationSource: Bloomberg and BNPP AM, as of 22/10/2018

 

We find that low and anchored inflation may be one of the causes for this negative correlation (see the average level of inflation before and after the mid-1990s in Figure 2). But the question of what ultimately drives yields higher in the first place, and how sharply the fixed-income sell-off is, matters too. If yields rise in response to a better economy, and this comes with better company earnings, yields can rise and equities can rally in tandem. When yields rise aggressively, say in response to a shock (e.g. higher inflation or fiscal expansion), the sudden increase in the discount rate frightens the equity market and stocks sell off, while yields push higher.

That said, the short-term correlation between equities and bonds can flip around quite quickly, especially in risk-off periods. What is perhaps striking is that in the recent equity correction (preceded by higher yields), UST yields receded only briefly and rebounded quickly thereafter. This begs the question whether bonds are still a good equity/portfolio hedge.

Unfortunately, our analysis suggests the answer is ‘no’. In the current cycle, and especially in recent price action, bonds have not offset equity market losses. In fact, in 2018, yields have tended to rise as equities have suffered. This is a sign that the equity/bond correlation may be shifting into positive territory.

This is worrisome, especially in the face of likely larger equity corrections when we eventually hit end-cycle. If bonds cannot hedge smaller setbacks in the bull run, how can they be portfolio hedges when stocks enter a bear market? While in such a scenario, a safe-haven bid may propel bonds, the low starting point of yields means that bonds will be worse hedges. To illustrate this, 10-year yields would need to fall to around 1% from the current 3.1% to give a return of 20% that would match the average historical equity market sell-off. In fact, equity sell-offs have been clearly bigger in some recessionary periods.

 

Other implications for asset allocators

The shifting macro/market backdrop will, in our view, also set other challenges for asset allocators.

  • Volatility should generally rise, particularly given that QE has been a big contributor in depressing market volatility in recent years.
  • With valuations looking more stretched and no central bank put underpinning markets, return expectations should also be much lower for the foreseeable future.

One of the effects of post-crisis monetary policy action was a rise in correlations across asset classes. This should also reverse, with individual assets likely to revert to being driven by their own fundamentals/news and not purely by QE flows. This unwinding may already be under way.

 

Conclusions

US rates will likely remain in the ‘driving seat’ and affect many asset classes. As the equity/bond correlation shifts, bonds may now offer less protection to cross-asset portfolios. Future returns look likely to be weaker than at the height of QE, and volatility higher. For us, this means being ever more tactical in managing portfolios.

The unwinding of QE should also help reverse high correlations between asset classes. This suggests a return to an environment favouring ‘asset pickers’, where portfolio managers can again add more alpha.

 


This is a condensed version of A changing fixed-income world: asset allocation challenges dated 30/10/2018. For the full paper, click here

 

Acknowledgements

The author would like to thank Matteo Digrandi (Intern, MAQS) for his contribution to the content of this publication. His contribution was made in accordance with the firm’s systems and procedures on financial promotions. Views expressed are those of the authors, as of November 2018. Individual portfolio management teams may hold different views and may make different investment decisions for different clients.