Fixed income instruments, commonly referred to as bonds, are an important component of our multi-asset class unit trusts such as the Allan Gray Balanced Fund and Stable Fund, as well as the specialist Allan Gray Bond Fund. Mark Dunley-Owen outlines some of our fixed income principles and explains our current investment view.
Bonds are boring, or so the saying goes. A better interpretation is that bond returns are less volatile than equity returns, most of the time. This is positive for investors who value stability, such as retirees. It is negative for investors with long time horizons since higher risk investments, such as equities, typically outperform over long time periods. R1 in 2000 would be worth R10.65 today if you had invested it in the FTSE/JSE All Share Index (ALSI) versus R5.67 in the JSE All Bond Index (ALBI).
Boring or not, the diversification benefits of bonds make them relevant to many investors. They tend to perform well when equities perform poorly, and vice versa. It is prudent to maintain bond exposure within diversified investment portfolios, and vary this according to your objectives and your view on relative risk and return. This principle underlies our multi-asset class unit trusts, in which bonds typically have a material weighting and are a meaningful contributor to performance.
How we manage fixed income
In a recent podcast, Malcolm Gladwell discusses the free throw record of NBA Hall of Fame basketball player Rick Barry. Free throws are uncontested attempts to score points after a player is fouled in the restricted area. Historically they were particularly important in the last few minutes of a game – interested readers should google ‘Hack-a-Shaq’.
Barry’s career free throw percentage of 89.3% means he was successful in almost nine out of ten free throw attempts. This is the seventh highest career free throw percentage in the history of the NBA. For current context, Stephen Curry, sometimes referred to as the greatest shooter ever, has a 90.4% career free throw percentage.
What is unusual about Barry’s free throw record is that he threw the ball underhand, using so-called ‘granny style’. He is the only NBA player to do so consistently and was ridiculed for doing so, but persevered because it maximised his scoring ability. Malcolm Gladwell describes Barry as a low threshold personality who prioritises results over popularity.
We believe there are upside risks to South Africa’s long-term inflation and real yields.
Allan Gray is similar in that we will deviate from common wisdom if it improves our ability to generate long-term wealth for our clients. Some of our fixed-income principles are thus different from other managers:
- We believe investment skills are asset class independent. Our fixed-income investment team is part of our broader investment team, with the same analysts covering equities and bonds.
- We apply the same investment philosophy and process across asset classes. We do fundamental research to determine the fair value of an asset, using sustainable cash flow as a core input. We buy when the price is below this fair value, and sell when the opposite is true. This applies to both equities and bonds.
- We target long-term absolute returns.
This thinking has helped deliver attractive risk-adjusted returns for our clients. Graph 1 shows 1, 3, 5 and 10-year annualised returns for the Allan Gray Bond Fund in red and the ALBI in grey. The two lines in the graph look similar, except that the Bond Fund is further to the left. This indicates that the Fund has generated similar returns as the ALBI over various time periods, but at lower volatility or risk, meaning the Fund’s risk-adjusted returns are better. Adjusting returns for risk is important if one believes, like we do, that the last few decades have been unusually good to bond investors. The benefits of lower risk are often overlooked during bull markets only to become apparent when markets turn.
Our investment outlook for bonds
We caution against making broad assumptions about the future, such as macroeconomic forecasts, as these are seldom correct or indicative of investment performance. This is unusual in fixed-income investing where it is popular to have macroeconomic opinions, often over the short term. Instead, we spend our time thinking about key variables that we believe will impact long-term bond returns. Two of these are South Africa’s long-term inflation and the real return required to make South African bonds attractive to investors.
The three-year moving average of SA inflation has ranged between 4% and 8% over most of the last 20 years. This is remarkably stable considering the economic fluctuations over this period, testament to a supportive global interest rate environment and the competency of the South African Reserve Bank. Few investors remember that SA inflation was regularly higher than 15% in the 1980s, and even fewer think future inflation will return to these levels.
We likewise do not expect double-digit inflation anytime soon, but our view is that global inflation is likely to rise as some of the disinflationary trends of the last few decades moderate, such as global trade, fiscal prudence and excess savings. Since much of South Africa’s costs are imported, this will place upward pressure on South Africa’s inflation. Even if we are wrong, it is unlikely that SA inflation will fall sustainably below current levels while we remain an open economy with below average productivity and above average wage pressure.
The second variable, namely the required real return from bonds, is difficult to predict. A reasonable starting point is the historical real yield on the 10-year government bond, which is 4.1% over the last 20 years. The real 10-year government bond yield at the end of March was 2.6%, which suggests investors are more comfortable or complacent about South African risks today relative to history. This is surprising when one considers the economic consequences of recent events such as the sovereign ratings downgrade to junk status. It is even more surprising when one looks at South Africa’s underlying fundamentals:
- Real GDP growth was negative in December 2016, and the lowest it has been in post-apartheid South Africa outside of the Global Financial Crisis of 2009 and the Asian/Russian Crisis of 1998.
- Government gross loan debt is at 51.7% of GDP, the highest it has been since 1970. This ignores record high debt at the state-owned enterprises. Eskom and Transnet’s combined net debt was R444bn as of September 2016, double what it was in 2012 and equivalent to another 10.5% of GDP.
- Long-term growth drivers such as education, capital investment and policy certainty are deteriorating from already-low levels. Most readers are aware of South Africa’s education and policy challenges, but equally worrying is that government fixed capital spend is at the lowest percentage of GDP since 1970.
There appears to be a dichotomy between the real return investors are accepting from South African bonds, and the underlying risks these bonds are exposed to. A reasonable explanation is that investors are looking forward to improving fundamentals. While a cyclical recovery may have seemed likely earlier this year, recent events have changed this. Furthermore, investors like us who focus on long-term absolute returns, will find it difficult to forecast improvements to the fundamentals necessary to justify current real returns.
Approach with caution
We believe there are upside risks to South Africa’s long-term inflation and real yields. In some scenarios a correction in one may negatively impact the other. For example, the government has three broad ways to reduce its debt to GDP – higher growth, austerity or inflation. Higher growth and austerity would be positive for bonds but are difficult to implement within South Africa’s socio-economic realities. Inflation, as the path of least resistance, is likely to be at least part of the solution. This would be negative not only for bonds, but also for the currency and South Africa’s long-term competitiveness, which could in turn force investors to require a higher real return from South Africa’s bonds. If we are right, future bond returns are likely to be disappointing. Our caution is reflected in limited bond duration, a measure of risk, within our clients’ portfolios. The duration of the Allan Gray Bond Fund is 5.4 years, materially lower than its ALBI benchmark duration of 7.2 years. The duration of the Allan Gray Balanced Fund and the Allan Gray Stable Fund is lower still, at 2.3 years and 1.2 years respectively.