by Lars Kroijer
So what is the difference between the minimal risk bonds and the other government bonds we are now looking to add? Think of your minimal risk bonds as the core of your portfolio. This is not where you are looking to make money, but where you are looking to take a minimal risk. I am then adding all other government bonds, other than the minimal risk bonds that are already in your portfolio. So if your minimal risk bonds are UK government bonds, then the other government bonds you should consider adding are all but UK government bonds. It only makes sense to add government bonds that add expected returns to the portfolio (see later). If you added German government bonds to the UK ones you had as the minimal risk asset then the low yield of the German bonds would not add returns to your portfolio. And since you had the capital-preserving minimal risk asset already in the form of UK government bonds the German ones would not add much to your portfolio (unless you wanted a couple of different bonds as the minimal risk asset and included the German bonds for this purpose).
It may come as a surprise to some that the bond markets in the world actually exceed the equity markets by a healthy margin of about $30–40 trillion, or more than double the US annual GDP (US government debt/GDP is approximately 100% at present).2
Adding government bonds
There are good reasons to add government bonds to your rational portfolio in addition to those you already hold as your minimal risk asset. Despite the world’s increasing international interdependence government bond portfolios are geographically diversified. Figure 7.2 shows government debt by geographical sector.
It is probably not surprising to many that the US debt is right at the top, considering the size of its economy, but that Japan is there alongside it may surprise some. With its very large debt/GDP ratio of over 200% Japan has managed to remain very indebted, but without incurring high, real interest rates as a result.3
As with equities, when adding other government bonds we would normally try to invest as broadly and cheaply as possible, and allocate in accordance to the relative values that the market has already ascribed to the various securities. However, you should amend this and not buy other government bonds in the proportions of the bonds outstanding that are shown in Figure 7.2.4
Earlier, we discussed how a highly rated government bond in the right currency provided the best investment for risk-averse investors, albeit at very low interest rates (at the time of writing). As an example, if I’m a UK-based investor looking for the lowest-risk investment for the next five years in sterling, I should buy five-year UK government bonds, perhaps in the form of inflation-protected bonds.
Figure 7.2 Government debt in $ billions
Based on data from Bank for International Settlements, end quarter 2 2012, www.bis.org
As we are now considering adding other government and corporate bonds to the rational portfolio of minimal risk bonds and equities, we should not just blindly add all the world’s government bonds (see Figure 7.3). We already have exposure to UK government bonds and therefore do not need to include more of these in the portfolio,5 we would just be doubling up on an exposure already taken to be the minimal risk investment.
How much the exclusion of the minimal risk asset from the world government bond portfolio changes the profile of the remaining portfolio depends on your base currency. If your base currency is $ or yen, then you would have reduced the universe of government bonds by a quarter. On the other hand, if your base currency is my native Danish kroner (with AAA-rated government bonds), then the impact on the world government bond portfolio would be negligible.
Figure 7.3 Don’t blindly add in more minimal risk government bonds
Only add government bonds if they increase expected returns
We discussed earlier how when adjusting for inflation investors in several highly rated government bonds should actually expect to earn a negative return, at least for short-term bonds. At the time of writing, the major countries with AAA or AA rating that offer a safe haven in their domestic currency but with little or no real return include Australia, Switzerland, Japan, Germany, the UK and the US, among others.
Consider the example of a sterling-based investor with UK government bonds as her minimal risk portfolio, contemplating adding other government bonds to her rational portfolio. From the UK bonds she gets almost no real returns, but also takes almost no risk. If she were to add bonds from the AAA- or AA-listed countries above she would also get no return, but would be taking currency risk. So she would get no greater returns from the foreign bonds, but take more risk.6
The rational investor thus has little to gain from adding AAA or AA government bonds to her portfolio other than as a minimal risk asset. If she was after a lower-risk portfolio she could add more of the minimal risk bonds (UK government bonds in this example). If she was willing to accept more risk in the portfolio she could get additional expected returns by either adding equities or government bonds that had a higher real return expectation than that offered by her minimal risk asset.
One caveat to excluding the other AAA/AA government bonds from the rational portfolio (see Figure 7.4): if you think there is credit risk in your minimal risk asset, then it may make sense to spread your investments among other AAA/AA credits. For example, if you are a UK investor and don’t consider the UK government’s credit entirely safe, then you could split your minimal risk investment into a couple of different AAA/AA bonds to diversify the credit. By diversifying you decrease the concentration risk of having your minimal risk asset from just one issuer (the UK government in this case), although it means taking currency risk with the other government bond holdings.7
The above is a departure from portfolio theory. According to portfolio theory you should add all the world’s investable assets in proportion to their values, and combine those investments with the risk-free asset to get to your desired risk level. Here I’m suggesting that you should only add assets that have a positive real expected return higher than your minimal risk government bonds, as you are otherwise adding currency risk without adding real expected returns.
Figure 7.4 Typically, do not add other AAA/AA credits
The government bonds we should add to the rational portfolio
If the above seems like excluding a lot of government bonds, remember that you have only removed those already in the portfolio (the minimal risk asset) and other government bonds without meaningful additional expected real return (but with currency risk).
After omitting the minimal risk asset and other AAA/AA government bonds because of their low yield, the bonds you should consider adding to your portfolio are real return-generating government bonds (the rest) and corporate bonds (see Figure 7.5). And here we are still talking many trillions of dollars of potential investments.
Which government bonds you are left to invest in depends on how highly rated are the bonds that you want to eliminate. If you were only to eliminate bonds that were AAA rated and wanted to invest in the remaining then those would be distributed as shown in Figure 7.6.
What you will notice is that this list is dominated by the US and Japan, ‘only’ AA rated by S&P at the time of writing, and thus not deemed entirely without risk. But if you, like most investors, take the view that these AA-rated bonds still did not offer enough expected real return to be worth adding to the portfolio (besides being the minimal risk for investors in those currencies), and that you only wanted to add government bonds rated below AA (to get additional yield), then the remaining world government bonds would be distributed as shown in Figure 7.7.
Figure 7.5 Add real return-generating government bonds to the portfolio
Figure 7.6 Below AAA government debt in $ billions
Based on data from Bank for International Settlements, end quarter 2 2012, www.bis.org
The way to read Figure 7.7 would be as follows:
I am an investor who, in addition to my minimal risk asset and equity portfolio, wants to add a diversified group of other government bonds. Since I already have exposure to my minim
al risk government bonds and don’t think government bonds from countries rated AAA/AA offer enough yield to be interesting, which government bonds should I be adding?
While in principle you should buy the bonds rated below AA according to their market values, in reality it is not practical for some investors to find investment products that represent so many different countries. Instead you might buy an emerging market government bond exchange traded fund (ETF) and combine that with a product covering sub-AA eurozone government bonds. The combination you end up with would not be exactly in the proportions of the sub-AA-rated bonds in Figure 7.7, but get you a long way towards adding a diversified group of real return generating government bonds to your rational portfolio.
Bond yields move a lot. Even at the height of the 2008 financial crisis the yield on 10-year Greek government debt was 5–6% compared to the current unsustainable levels. This was considered a safe investment although events since suggest that it wasn’t and provide a good example of how the credit quality of an individual government can decline at an alarming pace if the markets lose confidence in repayment. Despite the fact that the government bonds in the sub-AA chart (see Figure 7.7) are geographically diversified investments, you should expect some correlation between them. Not only do some of the European countries operate in the same currency and open market (the EU), but all countries are subject to changes in the world economy, besides their unique domestic changes, and are similarly vulnerable to changes in market sentiment.
Figure 7.7 Below AA government debt in $ billions
Based on data from Bank for International Settlements, end quarter 2 2012, www.bis.org
Table 7.1 shows the yields on various selected 10-year government bonds that fit into the real return-generating government category at the time of writing. While the table below is for 10-year bonds only (you should try to get a mix of maturities) it gives you an idea of the interest you can expect from these lower-rated bonds.
Table 7.1 Ten-year yields: selected sub-AA governments
Based on data from www.tradingeconomics.com
These yields are in local currency-denominated bonds. Since the expected inflation on the Brazilian real is greater than that of the euro the inflation-adjusted return is not as high as suggested above. In other words, Brazil is not as poor a credit risk as suggested by the table.
In addition, it is unfair to say, for example, that you would expect to make a nominal 11.4% return from Greek bonds. The high return implies an increased probability that Greece will default or that you will somehow not be paid in full.
There are some further points when considering adding other government bonds to the portfolio:
As you add additional government bonds from these lower-rated countries do so in a range of maturities. On top of diversifying geographically this will avoid concentration of the interest rate risk. Practically, adding these bonds is best done via buying a range of ETFs or low-cost investment funds that buy the underlying bonds for a low fee. For example, you could buy an emerging markets bond ETF that will give you an underlying exposure to the wide range of lower-rated government bonds in different maturities that you are looking for. Likewise, with some of the sub-AA-rated developed country government bonds. By holding the bonds via ETFs or investment funds you don’t have to worry about buying new bonds as old ones mature. The provider will do this for you and ensure that your maturity profile remains fairly stable, which is what you want.
Be careful that you don’t add concentration risk when you are meant to be diversifying. In Table 7.1 (that excludes the AAA bonds only) if your minimal risk asset had been US bonds and thus excluded, Japanese bonds would have been over 50% of the remaining amount. These charts and tables should serve to remind you to broadly diversify your government bond holdings to those that add real returns – not increase concentration risk to one issuer (e.g. Japan).
Buying the government bonds listed above in proportion to their shares of all sub-AA-rated government bonds would be an expensive administrative headache, and there are no access products like ETFs or index funds that do exactly that for you. But if you don’t get the proportions exactly right that is fine too. Perhaps buy some low-cost emerging market government bond funds and add to those some exposure to below-AA developed market government bond funds. If you do this roughly in proportion so that each country’s or region’s share is roughly similar to that of its share in Table 7.1 then that is a good approximation.
Keep an eye out for changes in the make-up of real return-generating government bonds. Some of the bonds rated lower than AA may have increased in credit quality to the point where they don’t really add real returns in excess of your minimal risk asset, or perhaps more likely some of the governments rated AA or above may have declined in credit quality to the point where they are worth adding as a real return generator. I look forward to re-reading this book in 10 years’ time and with the benefit of hindsight seeing which governments moved up or down in credit quality. Look to make these changes as you rebalance your portfolio occasionally. Since your real-return generating government bond portfolio is well diversified, changes to it will hopefully not be too dramatic, but likewise keep in mind that unlike the minimal risk asset, world equities and corporate bonds, this is the one segment of the rational portfolio where a broad-index-type access product like an ETF or index tracker will be unlikely to suit your needs. You probably have to put a few products together yourself to create a portfolio of sub-AA-rated government bonds, the make-up of which will change over time.
Adding corporate bonds
In addition to adding sub-AA government bonds you should consider adding a board portfolio of corporate bonds to your portfolio (see Figure 7.8).
Traditionally, whenever investment books like this one have proposed investment in corporate bonds (as most do) they were referring to US-based bonds. This was because their audience was often made up of US or dollar-based investors, and besides, foreign bonds were expensive and impractical to buy. While the ease of investing in non-US bonds is rapidly improving, the US dominance is still prevalent, at least relative to the US share of world GDP, at around 20%. We saw earlier that the world equity portfolio’s largest constituent by a wide margin is also the US, and if you only add US corporate bonds you will not get the diversification benefits of international exposure. But this is not just true of US investors. Any investor that adds corporate bonds only in their home geography may have diversified asset classes, but at the same time have increased geographic concentration. Adding a broad portfolio of international corporate bonds can rectify this concentration issue.
Figure 7.8 Add diversified corporate bonds to the portfolio
Figure 7.9 World corporate debt in $ billions
Based on data from Bank for International Settlements, end 2011, www.bis.org
Looking to the future, the non-US portion of world corporate debt is likely to increase further and thus augment the importance of getting both the asset class diversification of adding bonds and also the geographic diversification of adding international ones to your rational portfolio.8
When you add corporate bonds to your rational portfolio, consider Figure 7.9 and make sure you diversify internationally. At this time, around 55% of the world’s corporate bonds are non-US, and like the US ones represent thousands of individual bonds of different maturities, industries, geographic areas and credit qualities. Ignoring the great diversification benefits from adding index-tracking ETFs or funds made up of these many thousand foreign bonds to your rational portfolio would be an omission.
Figure 7.10 Placement in the capital structure
It makes sense that while there are very high-yielding bonds, in general return expectations from bonds are lower than equities. As a bond holder you are a lender – to either a corporation or government – whereas as an equity holder you are an owner. The seniority of the capital structure reflects this. In receiving the distribution of the cash flows of a company the debt holders are entit
led to their interest payments before dividends are paid to equity holders. Likewise, in default, debt holders have the first claim on the assets of a company. A lower expected return is the price of this superior place in the capital structure (see Figure 7.10).
Just as all the various layers of seniority in the capital structure are represented in the world corporate bond portfolio, the bonds vary significantly in maturity. And as with the case of government bonds the longer-maturity corporate bonds of similar credit quality typically yield more than their shorter-term peers.9
If history is any guide …
Compared to equity returns, figuring out historical returns for broad bond indices is not straightforward. Until fairly recently there was a dearth of investable index products available to investors who wanted bond exposure, other than perhaps that of the major country government bonds or US corporate bonds. Things are slowly getting better and the next decade will see further expansion in the amount of fixed-income products available for the retail investor.
The historical indices that do go back some time have a heavy US bias and until recently broad-based indices were hard to come by, much less ones you could actually create as a product. Table 7.2 shows the performance data for some broad bond indices.
Although the time period shown in Table 7.2 is far too short to make meaningful conclusions, 2008 stands out as an interesting data point. Both the US aggregate and global government bond indices had positive returns in a terrible equity market.
The outperformance of highly rated bonds in a tough market environment points to the potential advantage of adding fixed income to the rational portfolio. As equity markets collapsed, investors sought security in highly rated bonds. There was a belief that whatever happened, the bonds would be repaid at maturity, while nobody knew what would happen to equities.