Dear Immanuel,
Thanks for the question. Allow me to explore it from three angles: (1) Is it possible to make money from bonds when yields are near zero? (2) Are overall returns going to be lower? (3) Have bonds been able to play the stabilising role in a portfolio? Are there better alternatives?
The short answer to the first question is yes, bonds can have a positive return even when yields are zero or negative. In 2019, bonds had a tremendous year despite yields being so low, even negative in many developed countries – as you rightly point out, an unprecedented environment. There are three sources of return for bonds.
The first source of return is the yield.
The second is tied to the term structure and can be captured by a “roll down” strategy. So say you buy the 5-year bond and hold it to maturity and the return is 0.5% p.a., this is what you get. But, if a curve is upward sloping, i.e., the 5-year minus 3-year is a positive yield, you can employ a strategy such that you can sell an originally 5-year bond that had you had bought two years ago that has “rolled down” to 3 years, and sell it into the market which will price it at the lower yield (i.e. higher price) taking reference from the new 3-year issues. Because of this second source of return from “roll-down yield”, even negative-yielding bond markets like those in Europe saw positive return because the bond curves were steep: the negative yield was more than offset by the differences in yields between the terms. This is what Dimensional does for its bond funds in MoneyOwl’s portfolios.
There is a third source of potential return, which comes from changes in the yield or curve, i.e., forecast interest rate and bond yield movements. Unlike the first source of return (current yield) and the second source of return (the roll down yield/ steepness of curve), these are unknowns and arguably the equivalent of market timing in the equity markets based on reading economic data and taking fund manager risk. Dimensional has estimated that over time, the expected return from such forecasting activity is 0. In other words, it is hard to make such calls consistently. But a lot of the movement and return on bonds this year came from this third source of price change.
But I can appreciate that the niggling concern over low bond yields remain. How can it pan out any other way than bond yields normalising and thus bond returns being negative eventually, and overall returns lower? I think that it is not necessarily so.
Firstly, there is the all-important issue of time frame. Returns from bonds could be negative and overall market returns lower, but it might be for a time rather than permanently. A successive interest rate increase could very well in the short run cause pain in the bond market. But after this repricing, what we can expect then is higher yields overall and depending on the circumstances, the steepness of the curve could again offer “roll down” returns. In fact, insurance companies would be glad because while their bond portfolios might suffer mark to market losses, they end up having higher yielding assets. Obviously, I am venturing a bit into the territory of crystal ball gazing which I would normally prefer not to, but I’m just illustrating what is plausible, that it is not Armageddon.
Secondly, we should not view bonds in isolation, but the return of the whole portfolio. At MoneyOwl, we believe that the return from investment that you need to grow your wealth over the long term will come mainly from equities. The same macro factors that cause bond yields to “normalise” might actually be positive for equities. Granted that we have not seen it happen this way, from a low bond yield level. But we have seen periods of high inflation before – which should have been very bad for bonds – and despite those times an investment into a 60/40 portfolio or even an equities portfolio would still have provided a positive return as long as investors stayed invested for a sufficient time horizon.
When asked about negative bond yields, Warren Buffett said that they puzzled him but did not scare him. If we take a step back from the technical price levels and the unprecedented nature of bond yields, and remind ourselves what bonds are, perhaps we would agree that it is not scary. Ultimately, bonds are the result of production and economic activity. They are i.O.Us or debt issued by the government or companies when they borrow from investors to raise funds – because there is productive activity to be funded. Government use taxes and other income to pay back bondholders, while companies use business earnings to pay back bondholders. In that regard, the similar conviction we have about why global stock markets go up in the long term would apply also why bonds are probably here to stay: because population growth and a desire for increase in standards of living around the world give birth to aggregate demand, which have be met by companies or governments through the use of capital. Thus as long as the market economy is still valid and as long as we believe in capital markets, we should continue to have a belief in bonds having a role to play in our economy and also in our portfolios.
On whether bonds can continue to play a stabilizing role in portfolios, what we have seen during recent stress times is yes. When we monitor the portfolios, we see that the portfolios with progressively higher proportion in bonds have a less negative return in a severe equity market downturn than portfolios with less in bonds. This means that the portfolios are acting the way they should. Even when there was temporary extreme dislocation like in March 2020, credits suffered more than Treasuries – as the case should be.
On the possibility of gold being an alternative, we did see gold prices increase quite a lot in this year. Interestingly, Bloomberg had held a webinar saying that part of this was to do with a technical issue about the delivery of physical gold being affected by COVID. But the theory is that the opportunity cost of holding gold has fallen now that bonds are also not yielding anything, and perhaps gold can replace bonds.
My main discomfort with this is that gold has a very inconsistent relationship with equities. Unlike bonds, the negative correlation with equities movement is not strong. This is because gold is ultimately a speculative asset and unlike bonds and equities, which are productive assets. The return from gold you get is purely dependent on the change in gold price after you buy it. Prices are set by buyers and sellers, and while gold is bought and sold for jewellery, it is the speculative buying and selling of gold that determines almost all of the day-to-day changes in gold price. Being a speculative asset, therefore, gold cannot replace equities as the core of a portfolio that you invest in to secure your retirement nest egg, or for any return objective for that matter. There is no disagreement among investment professionals on this. I have never seen a portfolio with 40% in gold, for example and I don’t expect this to happen anytime soon. Even if there is a flight to safe havens, it is complex because gold is priced in USD and both gold and USD can be safe havens. This is an added complexity for the SGD-based investor. Because gold is a market of emotion, any allocation to the gold that is substantial enough to shift your return significantly when gold prices increase (maybe 10% or so), would also expose you to risk of loss if gold prices fall. This, then, adds to, rather than smoothens the volatility of your portfolio.
In portfolio construction, we would like to cross a high bar of evidence before we use an asset class for a particular purpose. We look for a long term, pervasive relationships or characteristics. We have evidence about long-term stock markets. With bonds, we also have long-term evidence of their characteristics and relationship to stock market price movements. With gold, there may have been recent movement demonstrating a hedging effect to equities, but the evidence is not pervasive enough across time to convince us to make a substantial strategic allocation.
Thanks again for the question and for allowing me to share these thoughts!
Dear Immanuel,
Thanks for the question. Allow me to explore it from three angles: (1) Is it possible to make money from bonds when yields are near zero? (2) Are overall returns going to be lower? (3) Have bonds been able to play the stabilising role in a portfolio? Are there better alternatives?
The short answer to the first question is yes, bonds can have a positive return even when yields are zero or negative. In 2019, bonds had a tremendous year despite yields being so low, even negative in many developed countries – as you rightly point out, an unprecedented environment. There are three sources of return for bonds.
The first source of return is the yield.
The second is tied to the term structure and can be captured by a “roll down” strategy. So say you buy the 5-year bond and hold it to maturity and the return is 0.5% p.a., this is what you get. But, if a curve is upward sloping, i.e., the 5-year minus 3-year is a positive yield, you can employ a strategy such that you can sell an originally 5-year bond that had you had bought two years ago that has “rolled down” to 3 years, and sell it into the market which will price it at the lower yield (i.e. higher price) taking reference from the new 3-year issues. Because of this second source of return from “roll-down yield”, even negative-yielding bond markets like those in Europe saw positive return because the bond curves were steep: the negative yield was more than offset by the differences in yields between the terms. This is what Dimensional does for its bond funds in MoneyOwl’s portfolios.
There is a third source of potential return, which comes from changes in the yield or curve, i.e., forecast interest rate and bond yield movements. Unlike the first source of return (current yield) and the second source of return (the roll down yield/ steepness of curve), these are unknowns and arguably the equivalent of market timing in the equity markets based on reading economic data and taking fund manager risk. Dimensional has estimated that over time, the expected return from such forecasting activity is 0. In other words, it is hard to make such calls consistently. But a lot of the movement and return on bonds this year came from this third source of price change.
But I can appreciate that the niggling concern over low bond yields remain. How can it pan out any other way than bond yields normalising and thus bond returns being negative eventually, and overall returns lower? I think that it is not necessarily so.
Firstly, there is the all-important issue of time frame. Returns from bonds could be negative and overall market returns lower, but it might be for a time rather than permanently. A successive interest rate increase could very well in the short run cause pain in the bond market. But after this repricing, what we can expect then is higher yields overall and depending on the circumstances, the steepness of the curve could again offer “roll down” returns. In fact, insurance companies would be glad because while their bond portfolios might suffer mark to market losses, they end up having higher yielding assets. Obviously, I am venturing a bit into the territory of crystal ball gazing which I would normally prefer not to, but I’m just illustrating what is plausible, that it is not Armageddon.
Secondly, we should not view bonds in isolation, but the return of the whole portfolio. At MoneyOwl, we believe that the return from investment that you need to grow your wealth over the long term will come mainly from equities. The same macro factors that cause bond yields to “normalise” might actually be positive for equities. Granted that we have not seen it happen this way, from a low bond yield level. But we have seen periods of high inflation before – which should have been very bad for bonds – and despite those times an investment into a 60/40 portfolio or even an equities portfolio would still have provided a positive return as long as investors stayed invested for a sufficient time horizon.
When asked about negative bond yields, Warren Buffett said that they puzzled him but did not scare him. If we take a step back from the technical price levels and the unprecedented nature of bond yields, and remind ourselves what bonds are, perhaps we would agree that it is not scary. Ultimately, bonds are the result of production and economic activity. They are i.O.Us or debt issued by the government or companies when they borrow from investors to raise funds – because there is productive activity to be funded. Government use taxes and other income to pay back bondholders, while companies use business earnings to pay back bondholders. In that regard, the similar conviction we have about why global stock markets go up in the long term would apply also why bonds are probably here to stay: because population growth and a desire for increase in standards of living around the world give birth to aggregate demand, which have be met by companies or governments through the use of capital. Thus as long as the market economy is still valid and as long as we believe in capital markets, we should continue to have a belief in bonds having a role to play in our economy and also in our portfolios.
On whether bonds can continue to play a stabilizing role in portfolios, what we have seen during recent stress times is yes. When we monitor the portfolios, we see that the portfolios with progressively higher proportion in bonds have a less negative return in a severe equity market downturn than portfolios with less in bonds. This means that the portfolios are acting the way they should. Even when there was temporary extreme dislocation like in March 2020, credits suffered more than Treasuries – as the case should be.
On the possibility of gold being an alternative, we did see gold prices increase quite a lot in this year. Interestingly, Bloomberg had held a webinar saying that part of this was to do with a technical issue about the delivery of physical gold being affected by COVID. But the theory is that the opportunity cost of holding gold has fallen now that bonds are also not yielding anything, and perhaps gold can replace bonds.
My main discomfort with this is that gold has a very inconsistent relationship with equities. Unlike bonds, the negative correlation with equities movement is not strong. This is because gold is ultimately a speculative asset and unlike bonds and equities, which are productive assets. The return from gold you get is purely dependent on the change in gold price after you buy it. Prices are set by buyers and sellers, and while gold is bought and sold for jewellery, it is the speculative buying and selling of gold that determines almost all of the day-to-day changes in gold price. Being a speculative asset, therefore, gold cannot replace equities as the core of a portfolio that you invest in to secure your retirement nest egg, or for any return objective for that matter. There is no disagreement among investment professionals on this. I have never seen a portfolio with 40% in gold, for example and I don’t expect this to happen anytime soon. Even if there is a flight to safe havens, it is complex because gold is priced in USD and both gold and USD can be safe havens. This is an added complexity for the SGD-based investor. Because gold is a market of emotion, any allocation to the gold that is substantial enough to shift your return significantly when gold prices increase (maybe 10% or so), would also expose you to risk of loss if gold prices fall. This, then, adds to, rather than smoothens the volatility of your portfolio.
In portfolio construction, we would like to cross a high bar of evidence before we use an asset class for a particular purpose. We look for a long term, pervasive relationships or characteristics. We have evidence about long-term stock markets. With bonds, we also have long-term evidence of their characteristics and relationship to stock market price movements. With gold, there may have been recent movement demonstrating a hedging effect to equities, but the evidence is not pervasive enough across time to convince us to make a substantial strategic allocation.
Thanks again for the question and for allowing me to share these thoughts!