Foreign exchange reserves are held to finance imports and pay foreign currency debt obligations, and to provide a buffer against capital flight and sudden portfolio outflows – but they come with opportunity costs. Many countries hold an excessive amount, and strategic asset allocations are seldom interrogated to improve efficiency or cut the costs that can be incurred by unnecessary constraints or biases.
Now, however, the threat of structurally higher inflation, tightening developed market monetary policy and rising rates is forcing a re-think. There are relatively straightforward ways to improve efficiency.
The following example describes a hypothetical reserves investor that established a simple portfolio of 50% global government bonds and 50% euro and dollar cash 20 years ago, with a view to outperforming a blended inflation benchmark by one percentage point per annum.
This portfolio easily outperformed its benchmark until around 2012. Thereafter, it struggled to keep up as rates ran out of room to fall further and coupon and principal proceeds were steadily ploughed into low-yielding assets. By the end of 2021, its performance since inception had fallen behind the benchmark.
We propose that the investor should redirect 20% of the portfolio into a conservative investment tranche, split between investment-grade corporate bonds and US agency mortgage-backed securities. These assets have shorter duration than the typical portfolio of government bonds, as well as some credit risk exposure that would have helped to boost performance once the zero bound in rates was approached in late 2009. As a result, by the end of January 2022, the proposed allocation could have earned as much as 20 percentage points more cumulative return than the original allocation.
Volatility and maximum drawdown would have been lower too, but what about liquidity risk?
We applied haircuts to the value of the two portfolios using the high-quality liquid assets factors set out in the Basel III framework for liquidity risk measurement (Figure 1). These impose a valuation penalty that reflects the losses one could make if one attempted to sell assets during a period of market stress. Because the original allocation does not attract a penalising HQLA haircut but corporate bonds and MBS do, this is negative for the proposed allocation.
Figure 1. Diversifying would have enhanced performance without unduly compromising liquidity
Portfolio valuations after HQLA haircut, rebased to a value of 100 for the original allocation on 31 January 2001
Source: Neuberger Berman
Figure 2. HQLA haircuts assumed for each asset class
Source: Bloomberg, Neuberger Berman
On day one, those haircuts amounted to a 7.5% reduction in valuation at the whole-portfolio level. By June 2008, however, the superior overall performance of the proposed allocation started to close the gap. By the end of 2015, the proposed allocation would have had a higher value than the original allocation, even after taking account of the heavy penalties of the HQLA haircuts.
In other words, even with these stringent haircuts, we can see that this hypothetical reserves investor can afford to exchange at least some short-term liquidity for a likely enhancement to long-term asset growth – unless it may need to liquidate 90%–100% of its assets all at once.
We proposed a similar solution to one official institution that approached Neuberger Berman for help early in 2021. It ran a surprisingly large allocation to quasi-government bonds that were intended to gain a modest pick up in yield while preserving government bond-like liquidity. We questioned this approach because quasi-government bonds have become quite illiquid in times of risk aversion – less liquid than global investment-grade corporate bonds or investment-grade securitised debt tranches, which also tend to offer higher yields.
As with our hypothetical reserves investor, we were able to show this real-world investor how adding securitised credit would have raised the return with only a moderate increase in volatility, while diversifying with corporate bonds would have substantially reduced the volatility without giving up too much of the enhanced return.
These examples reflect the question we are increasingly asked by official institutions: how can we use a more flexible credit allocation to maintain returns, within tight constraints on liquidity, drawdown risk, credit quality, duration and other metrics specific to our mandate? We believe there is much that can be done to make asset allocations more efficient – and once that ‘beta is better’, the potential for added value from active management can be explored.
Jahangir Aka is Head of Official Institutions, Jon Jonsson is Senior Portfolio Manager, Multi-Sector Fixed Income, and Ziling Jiang is Head of EMEA Institutional Solutions, Neuberger Berman.
This article is an edited extract from Neuberger Berman’s white paper, ‘Could Your Beta be Better?’, published 9 May 2022.