Fixed income investments are cornerstones of almost every institutional investment programme. Most investors use allocations to fixed income to create a balance to the more riskier assets in their portfolio. Traditionally, the starting point for investors was to buy bonds issued by their government or locally-domiciled companies. However, as global capital markets became more integrated, investors increasingly invested in offshore instruments as well, either directly or through managed funds. Not surprisingly, offshore investments are particularly relevant for smaller economies, such as New Zealand. In fact, it is not uncommon to see New Zealand investment programmes without any dedicated domestic fixed income exposure. In contrast, especially in the retail market, some investors still rely exclusively on investing in New Zealand bonds.
So how should New Zealand investors decide on the split between domestic and offshore bonds? What trade-offs do they face in making this decision?
The beauty of hedging: why we can ignore differences in headline interest rates
A naïve investor just focusing on headline interest rates might conclude that it is not worth investing in offshore fixed income markets. Why invest offshore in markets like the UK, Europe or Japan - with long-term interest rates around 0-2% - when you can buy investments with double the yield in New Zealand? Furthermore, why invest offshore when you expose yourself to the fluctuations of the exchange rate by doing so?
This is where currency hedging provides an elegant solution. Investors enter into currency hedging arrangements to eliminate the risk of a large, adverse movement in the exchange rate. This negates the concern about the fluctuations of the exchange rate.
However, a somewhat less obvious consequence of currency hedging is the return component. A hedged investor essentially receives or pays the difference in interest rates as part of the hedging contract 1. If rates offshore are lower, the investor gains the difference in rates as part of the currency forward contract. If rates offshore are higher, an investor must pay away the difference.
From a theoretical point of view, this should not be surprising: in an integrated, competitive global capital market, the expected return from bonds of similar risk characteristics should be identical. Differences in interest rates that can be exploited without risk should be eliminated quickly by arbitrage 2.
Table 1 shows that it would have been costly for New Zealanders investing in domestic fixed income to have been blinded by the illusion of higher headline interest rates. While there were periods of underperformance, investors in hedged offshore government bonds largely outperformed those with a domestic focus over the last three decades 3. Furthermore, table 1 shows that hedged, global investors did not have a more volatile ride over this period.
Table 1 – Risk and return measures - January 1990 – March 2018
|Global Sovereign Bonds NZD Hedged||New Zealand Government Bonds|
|Annualised Standard deviation||3.2%||3.5%|
|Maximum monthly drawdown||-1.9%||-2.3%|
|Worst 12-month return||-1.8%||-3.9%|
Data source: estimated using S&P/NZX NZ Government Bond Index, FTSE Government Bond Index. Indices are unmanaged and cannot be invested in. Past performance is not necessarily a good indicator of future performance.
Rather than focusing on the headline yields, we should always compare hedged offshore returns with New Zealand returns. Historical data shows not only a higher return for offshore investments, but also lower volatility and smaller drawdowns.
However, it is arguably more important to consider the forward-looking risk and return expectations when deciding how to allocate between investments. Consequently, we turn our attention to the inherent risks and related premia investors can earn off the underlying fixed income instruments, here in New Zealand and offshore.
Risk premia in fixed income markets – and how they stack up around the globe
In our view, investors can access three key premia4 – over and above a cash return – in fixed income:
- Premium for default risk (credit risk premium): generally, investors are compensated with a higher yield for investing in securities with higher risk of issuer default.
- Term premium: investors can typically earn an additional return by investing in longer dated securities, which carry higher interest rates and inflation risks than shorter-term issues.
- Illiquidity premium: less liquid investments should have a higher return than more liquid investments.
These premia are not necessarily independent. For example, default risk and illiquidity are typically highly correlated. Nevertheless, by decomposing historical return premia we should be able to explain what led to the historical outperformance of global fixed income. We can then try and infer a forecast of return premia in fixed income and decide on whether New Zealand investors should access them onshore or offshore.
It is important to realise that government bonds also carry default risk. While yields on government bond are often used as 'risk-free' rates for valuation purposes – such as in the capital asset pricing model – it would be naïve to assume that they do not carry any risk.
One common approach to assess this risk is to use ratings issued by the major rating agencies (such as S&P, Moody's or Fitch). Other investors favour a more fundamental, purely quantitative analysis, for example using debt-to-GDP ratios as an indicator of risk.
However, rather than thinking about how to determine which rating agency is the most accurate or how to combine the different quantitative measures into one aggregate view, we believe it is more valuable to compare credit default swap (CDS) spreads.5
CDS data suggests that investing in offshore sovereign bonds has not led to a significant increase in risk for New Zealand investors. Even if the probability of a default event is higher, given the considerably better issuer diversification the average severity of a default event will likely be a lot smaller than in a New Zealand-only portfolio.
Having ruled out default risk as a major driver of the historical return difference, we now turn our attention to the term premium. The term premium should compensate investors for taking on additional interest rate and inflation risk by investing in longer dated fixed income instruments.6 A somewhat crude, but intuitive way to estimate the term premium is to deduct the return of a cash index from the return of a government index with longer duration.
Our analysis suggests that the term premium has, in the past, been a significant contributor to the return advantage of global markets. However, the data also illustrate that the difference is by no means constant over time. This should not be a surprise, as risk premia in financial markets are typically subject to fluctuations over time.
However, without knowing the direction of interest rates, the best forward-proxy for a potential difference in the term premium should be the difference in duration between the two markets. As we will see further below, this will lead us to assume that there could very well be a return advantage for offshore markets in the future as well.
The illiquidity premium is typically not directly observable. While for some of the biggest issuers – such as the US Department of the Treasury – a comparison of on-the-run and off-the-run securities allows some detailed measurement, investors typically have to make some approximation or rely on modelling to determine how much of the yield is due to a lack of liquidity.
In general, illiquidity is assumed to increase with credit spreads and term. While we have considered the default risk to be similar, the term is clearly longer for securities in the global market. In contrast, however, issuance size, another factor that has shown to be related to liquidity, is generally a lot smaller in New Zealand.7
Rather than doing detailed modelling of what is considered, outside of crisis events, a relatively small portion of bond returns, we inform our analysis based on conversations with market participants. This leads us to assume that the New Zealand fixed interest market is generally less liquid than global markets. As such, one would expect to access a return advantage when investing in New Zealand securities rather than in the global market.
Risk premium diversification
Our analysis above suggests that the main driver of the return advantage of global fixed income in the past was the term premium, and not the default premium or illiquidity premium. Having ascertained the impact of the different premia, we can now consider how much we can diversify these premia in New Zealand and offshore. To do this, we now include corporate issuance and government-related issues in our analysis.
Table 2 compares the Bloomberg Barclays Global Aggregate Bond Index with the S&P/NZX Government Bond Index and the Bloomberg NZ Credit 0+ Yr Index:8
Table 2 – Diversification potential in New Zealand and global fixed income
|S&P/NZX GOVERNMENT BOND INDEX||BLOOMBERG NZ BOND CREDIT 0+||TOTAL NZ (GOVERNMENT + CREDIT)||BLOOMBERG BARCLAYS GLOBAL AGGREGATE||NZ AS A PERCENTAGE OF GLOBAL|
|Amount outstanding (NZD)||63 billion||22 billion||85 billion||~50 trillion||~0.2%|
|Number of issues in the index||9||87||96||22,301||~0.4%|
|Average issuance size (NZD)||~7 billion||~250 million||~900 million||~2.2 billion||~40%|
|Estimated duration||4.8 years||2.9 years||~4.4 years||6.9 years||–|
Data source: Bloomberg, S&P
Table 2 clearly illustrates that it is significantly easier to diversify by issuer in a global portfolio. Looking further below the surface, we also find significantly more sector diversification in global credit markets. In contrast, there is a large concentration in the banking sector in New Zealand.
On the flipside, investors need to accept a somewhat higher duration (i.e., interest rate risk) when investing offshore. However, this should be, compensated by an additional term premium, as illustrated above. In addition, higher duration in global fixed income should amplify the diversification benefits in the event of an equity market downturn. This is often the main reason for holding bonds in a multi-asset portfolio.
Other important considerations
There are other important considerations that investors may want to consider in their fixed income decision making. Most notably, we believe that at least some investors should have regard to duration matching. For all investors, it is also important to consider the role of active management.
Our analysis shows that global fixed income has provided a significantly higher return than New Zealand fixed income in the past. In large part, this has been driven by a higher term premium in offshore markets. In line with this, we should consider the interest rate risk (duration) to be higher in offshore markets as well. However, for a multi-asset investor, increased exposure to duration is often desirable as an offset to more risky assets.
Furthermore, the significant improvement in diversification improves the attractiveness of offshore investing. Finally, investors may find it more useful to delegate the country allocation to an active manager rather than trying to determine the optimal split in a static allocation themselves. For most investors, having no dedicated exposure to domestic markets is therefore a defendable starting point.
However, investors with a need to match the duration of contractual liabilities, or investors who face significant regulatory or peer risk may choose to allocate more significant parts of the portfolio to the New Zealand market.
In any case, we recommend that all New Zealand investors read more than the one page in the book of fixed income, and consider a material allocation to offshore markets in their fixed income portfolios.
1 This is because the instruments used for hedging - forward currency contracts - are priced based on the difference in interest rates. For a more detailed discussion of currency hedging see Johnson, A., The mechanics of currency hedging using forward exchange contracts, Russell Communique, Q4 2013.
2 In practice, currency forward contracts typically have significantly shorter terms than the bonds they are protecting in a diversified portfolio. This creates a mismatch that would require a sophisticated and detailed analysis of the yield curve differences between different countries, which is beyond the scope of this paper. We believe that the discussion of the difference in term premiums in the next section is an adequate approximation.
3 Given the regime shift that occurred in monetary policy in New Zealand as a result of the Reserve Bank Act 1989, we believe it is meaningless to analyse periods prior to 1990.
4 For a discussion of return drivers see Fitzpatrick, G. and Ross, L., Credit, illiquidity, term: a discussion of three fixed income return drivers, Russell Investments, February 2015.
5 A credit default swap (CDS) is a contract that insures the buyer against the default of an underlying bond issue. The insurance premium is typically known as the CDS spread.
6 For an in-depth discussion of the literature on the term structure and term premium see for example, Gurkaynak, R.S. and Wright, J.H., Macroeconomics and the Term Structure, Journal of Economic Literature, June 2012, Volume L, No. 2.
7 However, there is a risk that average issuance size may lead one to confound the liquidity advantage with the default risk as well: the more debt issued, the more active and liquid a market may be but in that case the default risk should increase with additional (sizeable) issuance as well.
8 As at 31 July 2018.