The objective of diversification is to dampen the effects of a bad outcome of one asset in a portfolio. This can manifest in bonds (or XTBs over them; we use the two terms interchangeably) as either the underperformance of a particular bond compared to the rest of the portfolio or in the worst case, albeit unlikely, the default of any of the issuing companies.
Non-default asset prices
Unlike equities, bonds have both a maturity date and a maturity value. This manifests itself in a known future price of the bond on a known date. We should note that for this section we ignore the possibility of a default – we discuss this later in the article.
The impact of having a known future price on a particular date on bond pricing is substantial when compared to equities, which can triple or quarter in value. Because a bond must finish at a certain price on a certain day, it makes it difficult for bond prices to deviate too much from that particular future value. Naturally, the further you are from that date the more the price can deviate. It is this feature which gives bonds inherently lower volatility than equities.
To illustrate this, we take market data from both the S&P/ASX 200 and the Bloomberg AusBond Credit 0+ Yr Index – both of which are used to describe generally their underlying markets. We then use this information to perform 250 simulations of the potential prices of an ‘average equity’ and 250 simulations of potential prices of the ‘average bond’. Both charts are done over a 4-year time horizon.
The 250 equity price simulations looks like the following:
Chart 1: 250 Equity Price Simulations
This chart is a graphical simulation of how equity prices may behave, based on model assumptions and calibrated historical data. It may not reflect actual price movements of any particular equity.
The 250 bond price simulations looks like the following:
Chart 2: 250 Bond Price Simulations
This chart is a graphical simulation of how bond prices may behave, based on model assumptions and calibrated historical data. It may not reflect actual price movements of any particular bond.
The important distinction here is that equities tend to be a ‘fan’ shape where the end prices cover a wide range in either direction. Bond prices, however, converge on a single point.
There are two important findings for diversification in this. Firstly, that a potential downside move for bonds is significantly lower than with equities. Secondly, if held to maturity, there is almost no need to diversify among bonds, since the future price is ‘set’. In other words, the need to protect against a downside move is annulled. This is because on the maturity date there is only one possible price.
Likelihood and Impact of a default
S&P provides the global probability of an investment grade bond defaulting within 1-year as 0.11%1. In Australia, there have been two recent Investment Grade Corporate Bond defaults – Pasminco in 2002 and Babcock and Brown in 2009. However, both were originally rated Investment Grade in a pre-GFC world where commentators today suggest higher ratings were more easily obtained.
In addition, because senior corporate bonds rank high in the company’s capital structure, the loss for investors in the event of a default is often significantly less for bond investors. By way of example, the North American average recovery rate upon default of all Senior Unsecured Corporate Bonds from 1987 to 2010 (including non-investment grade) is 49.2%¹. For simplicity, we use a number of 50% later in this article.
To crunch the numbers, if by small chance (0.11%) one of the bonds in an 8 XTB Investment Grade Portfolio does default, based on a recovery rate of 50%, an investor is likely to lose: 50% x (1/8) = 6.25% of their investment.
Therefore, to determine how many bonds you need to protect against this risk, you simply need to assess how much downside you are willing to accept in the 0.11% probability of an investment grade default. If you are willing to accept a 6.25% loss you only need 8 bonds. If you are only willing to lose 3.125% of your capital, then you will need 16 bonds.
Of course, there is always a pay-off. The yields achievable and lower brokerage on 8 XTBs might be more attractive than the reduced loss in the case of a 0.11% event.
¹ Corporate Default and Recovery Rates, 1920-2010, 28 February 2011 – Moody’s Investor Services