This is the third posting in a series on the overall portfolio structure implied by a focus on a dividend-growth portfolio. A previous posting, “A Dividend-Growth Portfolio Isn't a Strategy; It's a Class of Assets,” (A Dividend-Growth Portfolio Isn't A Strategy, It's A Class Of Assets) presented the plan for the entire series on the topic.
Many investors treat bonds as if they are separate asset classes. As is discussed below, it is equally legitimate to view them as simply substitutes for the key assets: equities and cash. As substitute equities and substitutes for cash, bonds can be used to hedge some of the risks inherent in the key assets. However, they only provide a hedge to the degree their performance is not positively correlated with the performance of the dividend- growth portfolio or the cash. So, while they may mitigate volatility, they do not constitute an effective hedge.
Cash is not the only asset that an investor can use to protect their dividend-growth stocks. Many investors consider a bond portfolio as a way of protecting and balancing their equity positions. When viewed that way, the bond is a substitute for cash. The relationship of a bond to cash is, in some ways, analogous to the relationship of an option to the underlying equity. The bonds provide the option to have cash in the pre-specified amounts at particular intervals of time. An option provides the opportunity to make a transaction in equity at a pre-specified level within a particular time. However, depending on how interest rates move, the price of the bond can be far less stable than cash; the price of the bond in cash may either go up or down. So, a bond is neither a perfect substitute nor a systematic hedge.
With bonds there is a liquidity issue that doesn't exist with cash. In a liquid market, it is possible to interchange the two fundamental asset classes: cash and equities. The same is not true of bonds and equities. There is an intermediate step which usually involves converting the bond to cash before converting it to a different asset (note that I have ignored convertible bonds and bankruptcy proceedings). That liquidity issue affects not just the ability to shift between bonds and equities; it also affects the ability to shift between bonds and cash, the asset for which bonds are a substitute. In an illiquid market, the price of bonds can fall more sharply than that of stocks. Basically, bonds can become illiquid at exactly the point where liquidity has its greatest value: during a financial crisis.
Consequently, bonds may be an acceptable substitute for cash if one is willing to accept the risks associated with potentially surrendering liquidity. However, one then has to evaluate the difference between the return on bonds versus cash. That difference has to be assessed against the risk of loss of the liquidity. In my opinion, in the current low-interest-rate environment with the monetary authorities announcing intentions to tighten policy, holding bonds implies an unrealistically low assessment of the risk of a significant drop in bond prices and/or liquidity.
That hasn't always been the case. There is a second approach to bond holdings. The second approach views bonds as a part of the income-producing portfolio. In the parlance of this discussion, bonds are being used as a substitute for dividend-growth stocks. Within this second approach, there are a number of different asset allocations that can seem appropriate. All of them in one form or another involve having a balanced portfolio. A balanced portfolio can be used to place oneself in a comfortable position of investing in a mix of securities with intermediate risk and intermediate returns. The bonds ensure that the portfolio can't totally disappear, while the stocks ensure some growth. At one point, the approach was strongly recommended for retirees because of the then higher yield of bonds versus stocks. That justification has been totally undercut by the Federal Reserve Board. At current interest rates, one can't justify bonds based upon higher cash flow.
Another traditional justification was the supposed stability of the principal in the bond portfolio. However, that stability is dependent upon the bond maturing at the appropriate time. Further, in terms of real value versus nominal value, the supposed stability requires stable rates of inflation. Between maturity and issuance, the bond’s price can fluctuate. Consequently, one can argue that the supposed stability of the principal is an allusion. Further, it can be argued that the cash flow from a balanced portfolio with bonds will be more stable. That, of course, depends upon picking bonds whose interest payments are more reliable than the dividend flows from stocks. The fluctuations in value of the underlying principle and cash flow can be particularly perverse if one gives up control by investing in a bond fund. In any case, that supposedly stable principal and cash flow is achieved by sacrificing the potentially higher cash flow (and total return) that can be generated by the dividend-growth portfolio.
Another justification for holding bonds is their supposed countercyclical behavior. They're supposed to be uncorrelated to stocks. If that were true all of the time, then bonds would have an important role as a component of a portfolio designed to minimize volatility while it is being liquidated. However, if no liquidation of the portfolio is planned, their countercyclical behavior is irrelevant. What matters is the relative performance of the bonds versus the dividend-growth stocks through the cycle. The longer one’s investment horizon, the less likely it is that bonds can perform as well as a dividend-growth portfolio.
It has also been argued that the countercyclical behavior of bonds will allow an investor to achieve returns equal to or above those that can be achieved with the portfolio that is not balanced. The arguments rest upon the assumption that rebalancing will allow the investor to overcome the lower return on bonds by timing the transition of some of the assets between bonds and equities. It is a market timing approach that can work. Since it can work, it can be made to look very attractive in a backcast. However, no matter how one sets up the rebalancing, it is inherently an effort to try to time the market. It may succeed, or it may fail.
A closely related argument is that a balanced portfolio with bonds will allow the investor to achieve an equal or higher return at a lower risk. That, however, is one of the weaker arguments. It's totally dependent upon choosing an arbitrary period of time over which to measure volatility. It is also worth noting that the entire argument is based upon the assumption that one needs to diversify into a different asset class in order to have assets with the desired countercyclical behavior. That is not the case. One can diversify within equities and achieve a reduction in volatility of the same order of magnitude as can be achieved with a balanced portfolio. There are stocks that move in the same direction as bond prices. They are often referred to as bond substitutes. Frequently, they not only move in the same direction, but they move more in price. Thus, exposure to those bond substitutes can achieve a similar result to a balanced portfolio with a smaller investment in the hedging effort.
Keep in mind that much of the research that supports the argument in favor of the mixed portfolio is based on data on the aggregate performance of asset classes without any reference to the variation in the performance of the individual assets within those classes. Bond substitutes among equities are only one illustration of diversification within the supposedly distinct asset classes. Many analysts approach junk bond investing in the same way and use the same analysis as is used in equity investment. The only distinction between bonds and equities is if one tries to manage duration. Bonds inherently have a termination date. Otherwise, they are analogous to equities, but, as in the case with cash, they are an imperfect substitute. The imperfection arises from the fact that they have a termination date.
Lest someone think that my attitude toward bonds reflects ignorance about modern portfolio theory, rest assured, I'm very familiar with it, and, more importantly, I understand the underlying assumptions and when it will or won't work. Modern portfolio theory and all the trappings are totally irrelevant if the correlation matrix between stocks and bonds is unstable. First-hand experience investing during the financial crisis left no doubt that many bond prices fell more, and more rapidly, than stock prices. In fact, in some cases, that was true of the stock and bonds of the same company. The theory is nice when it works, but it's hardly something to bet the ranch on or base one's retirement on. That's especially the case since there are other assets that are uncorrelated and inversely correlated to stocks.
If bonds are exposed to liquidity risk, then the logical hedge would be cash. Further, bond’s countercyclical behavior is limited. Any reduction in volatility would only occur when volatility is within a limited range. In a market crash, a liquidity crisis is a normal side effect, and, as was amply demonstrated in a recent financial crisis, bonds provide only minimal protection. While stocks declined significantly, some bonds literally became illiquid, or put differently, they became absolutely valueless. One might argue that some bonds, Treasuries, retained their value even during the financial crisis, but that argument only supports the point that there is more diversification to be achieved within the asset classes than is commonly acknowledged. It also should be noted that Treasuries retained their value because they retained their liquidity. Some stocks also went totally illiquid as the companies went bankrupt, while others held up quite well. So, a bond portfolio needs to be hedged against the market crash just as an equity portfolio must be hedged against the market crash. Part of that hedging can be done within the asset class, but some of the hedge has to be constructed using alternative assets.
It's worth noting that this discussion of bonds has proceeded without any reference to inflation-indexed bonds. Inflation indexed assets, primarily bonds, will be discussed subsequently as an alternative asset. That approach reflects the principle benefit that differentiates inflation-indexed bonds from other bonds. It is not a pure differentiation, but is sufficiently significant to justify treating inflation-indexed bonds as alternative assets. Inflation-indexed bonds hedge the risk of unanticipated inflation. That risk applies equally to any dividend-growth portfolio or a bond portfolio. In that respect, bonds are interchangeable with dividend-growth stocks, but being interchangeable in the sense of sharing the same risk is hardly a benefit.
Disclosure: I am/we are long ANY HOLDING MENTIONED.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: The purpose of this posting is to express my opinion regarding the risks associated with holding bonds. I doubt very much my opinion is shared by all investors and please do not take it as a recommendation. Each investor should do their own due diligence and make their own investment decisions. The fact that I have decided that inflation and interest rate risk undermine any justification for holding bonds should be viewed as just one opinion.