by Lars Kroijer
The large decline in the Barclays US High Yield index in 2008 was no surprise. Companies with high-yield bonds outstanding were dependent on a benign economic environment to repay their debts. With a collapsing market and grim forecasts as a result of the crash, those future repayments were put in doubt and investors sold high-yield bonds as a result.
Table 7.2 Various bond indices performance 2002–12 (%)
Generally, I would caution investors about reading too much from this short data period. There is no saying that future crises or correlations will be like those of the past. In fact, as a writer based in Europe I find it entirely conceivable that a future crisis could easily involve government credit issues with their bonds declining in value instead of being a safe asset. If so, you could easily see both equities and government and corporate bonds collapsing at the same time.
So while there is no certainty as to what might happen in a future crisis, adding other government and corporate bonds to the portfolio makes good sense. We are adding bonds from a very broad range of countries, maturities, currencies and risk levels, and with both government and corporate issuers. This kind of asset class diversification to the world equity portfolio as a return generator is likely to serve the portfolio well, with lower-risk and more-diversified returns.
Getting practical
As we look to add bonds to the portfolio, here are a few pointers:
Buy a broad-based bond exposure, both in terms of geography and type of bond.
Buy index-tracking products where available. You generally don’t want to pay the higher fees of active management, but in the case of adding bonds cheaply actively managed funds may be a good choice when index trackers are unavailable.
Look for new product development. Particularly in the bond space this will be important. My prediction is that the broadly available bond offering will be much expanded in the years ahead and you will probably be able to benefit from this.
Later, when discussing implementation (see Chapter 14) I’ll describe a couple of good alternatives to gain broad and cheap bond exposure as part of your rational portfolio.
Typical criticisms of adding bonds to a portfolio
Compared to equities there are fewer good bond indices and they are not as well known
True. Many investors don’t even know they exist. But what we care about is the availability of products that represent a broad range of bonds, both geographically and by type. Although this sector is still not up to the level of equities it is improving.
You will tend to overweight the debt of indebted countries and companies (they have more debt outstanding as a fraction of company value or GDP)
True. But the prices should reflect the higher indebtedness. In the future there could be government bond indices driven by the GDP of a country, but those that exist are still not that widespread.
In certain countries the trading of bonds is very expensive, rendering them a bad risk/return prospect
The key thing to figure out is if you are at a cost disadvantage compared to other market participants. If you are, it may make sense to stay away. If costs are just high for everyone, the higher costs should be reflected in the price and not affect the bond’s risk/return profile.
The income from bond coupon payments renders them tax inefficient
Important point. If you are unable to find tax-efficient products (ETFs, etc.) you may consider tax-advantageous bonds in your geographic area, such as municipal bonds in the US. A tax disadvantage can easily eliminate any investment advantage.
Bond products only represent a portion of the total bonds outstanding
It would be impractical and/or impossible to buy a small portion of all outstanding bonds in the world. Indices try to ensure that they can be practically implemented and thus avoid small and illiquid bonds; as with equity markets (to a lesser degree) this is a simplification we have to live with.
Bonds are mainly dollar denominated. This is an issue for non-US investors
Yes, a large portion is in dollars. This is because of the US’s dominance of government and corporate bonds, but also because many that issue bonds in currencies other than their home currency do so in dollars. You can partly alleviate this problem if your minimal risk asset is not US government bonds and thus exclude them from your other government bond allocation. Also, don’t ignore the large and increasing number of international corporate bonds.
It is expensive to trade bonds
Unless you are a big institution you should buy products like ETFs, bond index funds or even cheap managed bond funds that acquire the bonds for you. With the exception of buying government bonds directly from the treasury, you can typically only buy bonds in larger ticket sizes and by buying aggregating products like ETFs or index funds scale and cost advantages are gained that are hard for the individual investor to match. (This is also discussed later in Chapter 14.)
Corporate bond returns also depend on credit quality
Earlier we discussed how the equity risk premium to the minimal risk bonds is about 4–5% a year. What about risky governments and corporate bonds?
Corporate bonds rank above equities in the capital structure of firms and therefore have superior rights to cash flows or capital. It therefore seems reasonable that corporate bonds should have a lower return expectation than equities. How much lower obviously depends on the mix of corporate bonds in the index. At the time of writing, the yield to redemption for the Finra/Bloomberg US investment grade and high-yield indices were as follows (www.bloomberg.com/markets/rates-bonds/corporate-bonds):
Current yield
US investment grade 3.13%
US high yield 5.65%
For government bonds, we saw earlier what the yield on a 10-year bond was for various ‘risky’ countries. But as was the case with those government bonds we can’t simply deduce from the data above that high-yield bonds always will do better than investment grade ones. The high-yield bonds are likely to have a much higher default rate (just like higher-yielding government bonds will default more often), and the return net of those defaults will be lower than in the unlikely case where all the high-yielding bonds are repaid in full.
Return expectations of the rational portfolio
Below are some estimates for returns of the various asset classes we have discussed so far in this book. Based on a mix of academic research, historical returns, a study of the financial markets and my own judgement I have outlined what I would consider reasonable expected returns for each asset class, above the rate of inflation as follows:
Annual real return expectations
Minimal risk asset 0.50% (UK, US, German government debt or similar)
‘Risky’ government bonds 2.00% (sub-AA-rated countries)
Corporate bonds 3.00% (mix of maturities, countries and credit quality)
World equities 5.00% (4.5% equity risk premium)
In later sections, we look at individual attitudes towards risk. The minimal risk asset is obviously deemed to have little risk (thus the name), and we have discussed previously how a reasonable estimate of risk for equity markets is a 20% annual standard deviation. While we can reasonably estimate the risk of the government and corporate portfolio relative to the minimal risk asset and equities (with government bonds closer to minimal risk, and corporate bonds closer to equities), how the various allocations act relative to each other is harder to predict.
Although you have added diversification to your portfolio by adding risky government and corporate bonds, adding bonds is not always guaranteed beneficial. The correlation between those assets and the rest of your portfolio is likely to be higher during duress than in a steady state, even as some higher-rated bonds may increase in value as a safe haven during a storm.
Adjusting the rational portfolio
Before we introduced risky bonds into the portfolio things were easy. If you wanted no risk, you could pick the minimal risk asset; if you wanted a lot of risk, you could pick a broad-based equity portfolio.
If you wanted a risk profile in between the two, you allocated between the two. And do this in a cheap and tax-efficient way. Simple.
Adding other bonds to the portfolio gives additional diversification benefits to the portfolio, but does so at the expense of more complexity. When deciding on what portion of your portfolio you should allocate to bonds, start by going back to your premise as a rational investor. We assumed each dollar invested in the world markets is equally well informed, and as a result we should try to replicate the exposures of all markets.
The split of the approximately $100 trillion market that is world equities, and world government and corporate bonds, is roughly as shown in Figure 7.11. But after adjusting the portfolio to exclude the minimal risk bonds and other highly rated and low-yielding bonds, the split is quite different. Figure 7.12 is the same pie chart but excluding AAA/AA-rated government bonds.
Figure 7.11 World debt and equity split (in $ trillion)
Figure 7.12 Adjusted world debt and equity split (in $ trillion)
Using this mix of assets as a rough guide to our portfolio allocation we could allocate our risky investments as follows:
World equity markets 75%
Sub-AA government debt 10%
Corporate debt 15%
If you combine your investment in the minimal risk asset with investments in equities, risky government bonds and corporate bonds in those proportions you are doing well. You will have allocated your investments roughly in the same proportions as the aggregate market participants who have been choosing between a wide array of investable assets in search of the best risk/return. If you now do so in a cost- and tax-efficient manner, while thinking about your risk levels, you will have created a very strong portfolio. Graphically this updated portfolio is illustrated as point T in Figure 7.13.
If your risk preference is lower than point T, you combine the ‘T’ portfolio (as in Figure 7.13) with the minimal risk asset to get the desired portfolio risk.
Figure 7.13 The updated portfolio
Using equity risk insights in the context of a full rational portfolio
My reason for focusing on equities is that we have good data to gain some meaningful insights about the risk of investing in equity markets, whereas it’s harder to be exact about the overall portfolio risk. This is because while we can try to quantify the risk of government and corporate bonds, it is harder to predict how those move relative to each other and to equities (correlation) with any accuracy, and therefore what the aggregate portfolio risk is.
Extrapolating an understanding of equity market risk to the overall portfolio you need to bear the following in mind:
Try to understand the risks you are taking with an investment in the world equity markets.
Realise that your minimal risk asset is not entirely without risk, but has a far lower risk than the equity markets. Longer-maturity bonds will fluctuate more in value than shorter-term ones.
Other government and corporate bonds will typically have a risk level between equities and the minimal risk bonds. The riskier the bonds, the more they will be like equities and perhaps move in price more like equities if markets drop, but it is hard to be exact about that. As a very rough rule of thumb, assume that the other government and corporate bonds have slightly less than half the risk of equities, but will move more like equities in terrible markets than when markets are stable.
The rational portfolio allocations
So, now we have a rational portfolio as shown in Figure 7.14. Incorporating the points above about the relative proportions of risky government and corporate bonds relative to equities, the rational portfolio could look like that in Table 7.3.
The allocations shown in the table are the best because they are based on the proportions of values that the market already ascribes to them, with the caveat that I have excluded non-return generating, highly rated government bonds. So apart from those highly rated bonds, the ratio of equities, government and corporate bonds is in line with the market value proportions in the world today. And if we allocate along the same lines as the efficient markets we will achieve maximum diversification and the best risk/return profile.
Figure 7.14 The rational portfolio
Table 7.3 The rational portfolio at different risk preferences (percentages)
We need to take a combination of equities and other government and corporate bonds and combine that with our ‘safety asset’, the minimal risk asset. How much risk we want is then determined by how much of the minimal risk asset, and how much of the combination of the other asset classes, we want. Construct your portfolio in this way and you will have an outstanding portfolio for the long run.
In implementing the portfolios outlined above look for products that, as closely as possible, represent the various asset classes:
Assetclass Description
Minimal risk asset UK, US, German, etc. or equivalent credit quality of maturity matching investor’s time horizon.
Equities World equity index or as broad as possible.
Other government bonds Diversified, real return-generating government bonds of varying maturities, countries and currencies; we have used those rated sub-AA as a good indicator.
Corporate bonds Broad range of corporate bonds of varying maturities, credit risk, currency, issuer and geographic area.
If the strategy for investing above seems simple, it is because it is. As a rational investor who is willing to add a bit of complexity to the all minimal risk/equity mix from earlier, we are simply adding real return-generating bonds in the proportions that they exist in the world. We think that the normal functioning of the market has caused the prices of the many debt and equity securities to be such that they reflect the risk/return characteristic of that security.
Of course not every investment in the world is perfectly efficiently priced. If this was the case, and everybody believed it, then there would be no trading. Everyone would accept that the prices reflected all information and the security’s risk/return contribution. But that is not the point. The point is that we do not think that we are in a position to know better than the prices set by the market and as a result should not try to reallocate our portfolio to get better returns.
Special case: if you want a lot of risk
Theory and practice collide in the case where an investor’s risk preference is higher than point T in Figure 7.13. Theory suggests that the investor should borrow money and use that borrowed money to buy more of the ‘T’ portfolio.10 In reality many investors either don’t ever want to borrow to invest or simply can’t find the money to do so. Particularly post-2008, when geared investors got burned badly as loans got called at the worst possible time, investing with borrowed money is often not a real or desired alternative.
Figure 7.15 Higher risk-preference investors
For investors with a higher risk preference than point T, I would suggest buying more of the world equity portfolio and fewer bonds (instead of borrowing money to buy more of the mix of equities and bonds) – see Figure 7.15.
If you want even more risk than being 100% invested in world equities there are leveraged ETFs. Simplistically, the way they work is that the provider takes your £50 and uses that as collateral to borrow another £50, and then invests the £100 (in the case of a 50% leverage product). You will then have the exposure to the market of £100 despite only having invested £50. This obviously works well if markets are going up, but will quickly hurt badly in declining markets.
1 As discussed elsewhere these graphs are mainly for illustration, particularly as the risk and correlations change continuously.
2 Note that a lot of debt (about $28 billion) is issued outside the country of the issuer. This may be a French company issuing debt in the US in dollars. This is important in that you may buy a bond in the US, but have the underlying exposure as that of a French company.
3 At the time of writing, shorting Japanese government debt is a popular hedge fund trade as the managers deem the debt levels unsustainable,
but according to a Japanese hedge fund manager friend of mine the arguments used are hardly new and in his view this is not a ‘slam dunk’ trade. Time will tell.
4 As a first gut feel of why this ‘buy the world’ strategy may not work for everyone, consider that over half your government bond portfolio would be Japanese and US government bonds. This not only adds quite a bit of concentration risk to two issuers, but also adds currency risk and minimal real expected returns due to the low/negative real yields on those bonds.
5 One minor caveat to this argument; if the mix of maturities you added to get to your minimal risk asset is very different from the overall mix of maturities issued by that government then it makes sense to amend the minimal risk allocations. So if you only had very short-term bonds in your minimal risk allocation, yet are willing to add more risk in the form of equities and other bonds, it makes sense for this allocation to include some longer-term bond exposure from that same government.
6 There is of course the possibility that the other currencies appreciate against sterling. So if you held US bonds and the dollar went up in value relative to the pound, those bonds would be worth more in sterling terms. That being the case, although this risk can also lead to you making money it is not a risk you get compensated for taking in the form of higher expected returns.
