31
Bonds for Long-Term Investors?

Only six years out of business school, what would, over the next half-century, become a persistent questioning of investments in bonds was launched with this analysis back in 1970. The use of bonds beyond a liquidity reserve continues, in my view, to be dubious policy for long-term investors, particularly endowments and pension funds. Would that there were an easy way to show how much it costs long-term investors to invest in bonds—presumably to offset some of the notoriously uncomfortable short-term price fluctuations of common stocks. If the “opportunity cost” of accepting much lower returns over the long run were made clear as the true cost of reducing portfolio price fluctuations in the near term, would rational and objective long-term investors continue to accept large allocations to bonds? Not likely!

Since everyone knows that long-term bond yields are unusually attractive these days, at least in comparison of historical yields, this may be an interesting time to reconsider the merits of investing in bonds for the long term. After all, if bonds are satisfactory investments, why not pick up some good bargains now? And if they are not attractive now, will they ever be?

Should pension and endowment funds have long-term investment in bonds? This may seem at first to be a curious question to which the only answer is “Of course.” Certainly, most major funds now have large bond holdings. But what persuasive, logical argument can be made in favor of sustained long-term investment in bonds that would explain satisfactorily why the custodians of almost every pension fund, endowment fund and large personal trust in the nation has owned, now holds, and plans to continue investing in long-term corporate and government bonds with a major portion of their assets? It seems appropriate to question the wisdom of this policy for long-term investors.

Proponents of long-term, continuous investment bonds in large portfolios argue four main propositions:

1. Preservation of principal is assured because at maturity the obligation must be repaid in full.
2. The yield on bonds is typically higher than common stock yields (and the extra income is often needed now). Moreover, interest income is assured as to amount and time of payment.
3. If the national economy should suffer a severe and prolonged depression, bonds would once again prove invaluable.
4. Trustees of pension funds, trusts and endowments are bound by the Prudent Man Rule of fiduciary obligation to invest in bonds to have a balanced portfolio.

Let's analyze these propositions carefully, taking them in reverse order, beginning with the Fourth. The Prudent Man Rule governing the duty of a Trustee in the Commonwealth of Massachusetts and now recognized quite widely states: “He is to observe how men of prudence, discretion and intelligence manage their own affairs, not in regard to speculation but in regard to the permanent disposition of their funds, considering the probable income as well as the probable safety of the capital to be invested.”

Unfortunately, it seems that too many trustees of endowments and pensions have too fully adopted a position regarding bonds that may be well enough suited to personal estate problems, without carefully evaluating the important differences between the investment problems and responsibilities of estate planners versus those of managers of endowment funds, pension funds, and other long-term portfolios—including many personal fortunes that will remain invested for the long term.

Trusteed capital management is historically based on experience in managing the financial affairs of mortal women and men who unfortunately do not live very long. The investment manager of a personal trust that will not last forever and is subject to an uncertain date of termination may well emphasize conservation of capital for the individual's descendants while planning for an estate settlement and portfolio liquidation. In contrast, endowments, pension funds, mutual funds and insurance companies have one unique and distinguishing characteristic: they will continue for very long. virtually indefinite, periods. And there is no separation of interest between those who get current income versus those who get capital later. For most institutional portfolios, money is fungible.

An endowment fund has an unusually long term financial role for at least a considerable portion of the activities of a presumably very long-lived organization or institution. A pension fund also has a long-term obligation to provide for the well-being of many workers during their retirement. Even most personal investments will be looked to for such long-term needs as education, retirement, and family security.

These investors cannot be bound by the rules of estate planning because those rules give no explicit attention to the growth of capital and income which any growing institution would naturally seek from its endowment, which any corporation would expect of pension-fund investments, and which any individual expects of his or her capital. Rules, particularly those formed for different times or purposes, must not be accepted blindly.

Let us analyze the more substantial propositions of the bond partisans. With regard to their third argument (i.e., the risk of economic depression), a powerful case can be built to support the view that, excluding an externally caused calamity such as a major world war, this nation should suffer neither severe nor prolonged economic setbacks. Students of economic history can offer myriad statistics to show how different our economic position is today from the 1920s. High and widespread personal income provides stability, while research, technology and education provide growth. The high proportion of workers in services and white-collar jobs reduces cyclicality, while large investments in plant and equipment per worker support economic growth.

Meanwhile, political scientists will point to the important institutional changes that have so greatly changed our economic structure such as FHA and VA loans, FDIC deposit insurance, IMF reserves, unemployment compensation, Social Security, progressive taxation, enormous government spending at Federal, State, and local levels, the SEC, FPC, FCC, and other regulatory agencies, and the trend towards full funding of pensions. In particular, they will point to the Employment Act of 1946 which posits responsibility for economic growth, price stability, and low unemployment with the Federal Government, objectives which have been reinforced by the commitment of recent Presidents to fulfill this obligation. Ours is a greatly different economy from the economy of our fathers and grandfathers, and long-term investment policy should be commensurately different.

Perhaps the most favorable change over the past 40 years is that our economy is now a managed economy in which both Government and businessmen take part. Business managers have gained substantial control over the uncertainties which have in the past caused large fluctuations in inventories and capital spending which have in turn been the major progenitors of past business cycles. And the Federal Government has learned a great deal about the effective uses of fiscal and monetary policy to guide the economy away from inflation,¹ on the one hand, and away from accelerating declines, on the other hand. These managers of our economy are equipped with more voluminous, more accurate, and more timely data than could have been imagined in the 1920s. The development of computers and the advent of econometric model building have made forecasting, analysis, and evaluation increasingly rapid and reliable. We are steadily gaining understanding of the way a complex industrial service economy operates and how effective management can avoid serious imbalances and beneficially influence developments.

This is not intended to suggest that we live in a “new era” from which recessions are banned, but it does seem highly probable that we do not face the prospect of either prolonged or severe economic depression. Consequently, no major portion of long-term portfolios should sacrifice the opportunity to invest more positively in our dynamic economy merely to defend against the remote prospect of sustained economic adversity.

This positive outlook does not deny the possibility of unforeseen economic, business or investment adversities, and a contingency reserve may be desirable. On the other hand, it is hardly necessary to allocate 30, 40 or 60% of the fund to bonds to protect against a possible decline in investment income which may or may not develop and which is highly uncertain as to timing.

The degree of protection or insurance needed by most large and long-term portfolios can be accomplished in most cases with an expendable reserve of only 5 to 10% of the portfolio. On the other hand, a large bond portfolio incurs too great a long-term opportunity cost in investment profits foregone, as will be shown below, to warrant using large bond holdings as massive insurance against an unlikely and uncertain adversity.

The remaining pro-bond propositions—assurance of income at a high level and capital preservation—are the key investment considerations and can be tested by comparing bonds with a conservative portfolio of, say, utility common stocks as represented by Moody's Utility Average. In questioning the long-term financial validity of bond investments, we will use ten-year time periods as the basis for evaluating bonds and the equity alternative. This test period is only for analytical convenience, and the reader should keep in mind that ten years is actually a very short-term proxy for the long-term character of the funds with which we are concerned.

Turning now to the bond advocates' Proposition Two and comparing cash income from bond interest to cash income from utility common-stock dividends, the record presented in Table 31.1 shows that over each ten-year period since World War II, total cash income from utility commons purchased in the first year and held for ten years exceeded the income from long-term Aa utility bonds bought in that same first year. On average, over a ten-year span, Moody's utility dividends returned 6.4% on cost versus a peak yield of 4.6% for the bonds, or a minimum increase in income earned of 40% over the bond yield.

Over longer periods, the advantage of equities increases substantially. On a pure rate-of-return basis, a dividend yielding 4% currently and growing at 6% annually is equivalent in cash income over a 20-year period to a bond yielding 6.5%. And over even longer periods, the algebra is inexorable. What bond could compete for long with a utility portfolio that currently yields 4% and is growing at 6%? This means the dividend will double every 12 years producing 8% on cost in 12 years, 16% yield on cost in 24 years, 32% in 36 years and, to carry the proposition to a century time span which will test our capacity to think in truly long terms, the dividend would yield 1,024% in the 96th year!

Granted utility dividends have yielded and are expected to yield more cash income than bonds over long periods, are not bond interest receipts more predictable? It is quite clear that the amount of common-stock dividend received over a period of several years cannot be forecast exactly. But the very probable rate of growth in earnings can be translated into a highly probable pattern of dividends, particularly for a portfolio of common stocks.

And although the pattern of dividend income will be less certain than the pattern of interest income received on a known, present portfolio of bonds, just the reverse will be true for a large, continuing, and therefore always changing, portfolio of bonds. While we know precisely what interest will be paid and on what date for each individual bond now owned, most issues in any present portfolio will have matured or been called in 20 years and will be replaced with other bonds at currently unknown future interest rates.

Viewed in this long-term perspective, we would expect a bond portfolio's income to be not more, but less certain than the income from a portfolio of conservative utility stocks because we do not usually know whether bond yields will trend higher or lower, whereas the utility portfolio dividend income will surely trend higher, only the rate of increase uncertain.

The essential conclusion is that while the yield of the present bond portfolio is highly certain, the yield of a future bond portfolio cannot be accurately predicted and is less predictable than the future yield on cost of a utility common stock portfolio. The evidence substantiates this view. While interest rates have declined nine times on a year-to-year basis, utility dividends for Moody's Average never once declined in the post-war period. Thus the utility portfolio actually provides both a higher and a more predictable level of income than a bond portfolio.

While this discussion of the very long-term advantages of common stocks over bonds has been based on a portfolio of electric utility common stocks, portfolio managers can and should consider a far broader list of equities. A review of total corporate earnings and dividends indicates quite dramatically that while aggregate dividends rise with increasing earnings, dividends generally do not fall when earnings drop in recessions. Thus, in the 20 years since World War II, aggregate dividends have declined on a year-to-year basis only once and even then, by only by a mere 2.3% in 1952. Yet, during that same 20-year period, dividends rose by nearly 400% or at an average annual rate of 7.2% compounded. On the record, bonds are inferior as a source of reliable income when compared to equities.

Regarding Proposition Four, preservation of capital, it is curious that advocates of bond investment appear so convinced that the contractual nature of a bond is always an advantage to the bond buyer. Granted that the contract protects the investor from receiving less than stipulated. it also prohibits the investor from receiving any more. This situation can be viewed as a source of risk when we consider inflation which erodes the future purchasing power of both income and capital. In fact, if inflation continued at the long-term historical rate of 2%, the assurance that a bond buyer will only recover at maturity the nominal dollars he puts up, is the assurance of an effective capital loss in real purchasing power terms.

Table 31.2 displays an historical comparison of bond and utility portfolios. During each decade, the market value of the utility common stocks rose significantly. The amount of increase ranged from 44% to 155% with an average appreciation of 115%. Eliminating the effect of changes in P/E, which did rise during this postwar period, appreciation due solely to earnings increases would have ranged from 46% to 77%. Not surprisingly, the utility portfolio produces an important capital advantage over bonds. The magnitude of this advantage is impressive. Using the earlier expectation for utilities of 6% growth in earnings, and assuming no change in Price/Earnings ratio, capital would increase over a century—if we can contemplate such a long time period—to an amount 256 times its present size.

The remarkable result of this historical analysis is that a portfolio of conservative equities has been and is likely to continue to be greatly superior to a bond portfolio on all counts:

1. Equities produce much higher income.
2. Equities increase capital substantially.
3. Equity income is more predictable.
4. Capital is safer from inflation in equities than in bonds.

The evidence is impressively in favor of investment in conservative equities as the preferred means by which a conservative, long-term investment portfolio can achieve its goals. Yet, the question remains: Why do most pension funds, endowments, and other large funds continue to commit a large percentage of portfolio capital to long-term investments in bonds?

The explanation lies partly in the experience fiduciaries have had with terminal estate planning, and partly in the difficulty all investment managers face when asked to deal astutely with time periods in excess of five years (which for many investors is a working definition of infinity). For an investment manager faced with a heavy volume of daily business demanding immediate decisions, the really long term is a most awesome challenge to the imagination. Thus, the real problem is perhaps not whether bonds are a better source of income and capital values over the longer term, but rather how the investment manager and his fund trustees can shift to a strange and unfamiliar time dimension in which truly relevant long-term policy can be formulated.

If it is decided to change away from a policy of holding a large permanent portfolio of bonds toward a portfolio of conservative common stocks, how should the change in policy be implemented? At least two choices are available: (1) the change can be made in a single rapid program when conditions are deemed propitious; and (2) a program of dollar averaging can be used to make the transition from bonds to stocks over a period of years. The choice depends in part on the decision makers' confidence in their ability to time the transition from bonds to equities; in part on their confidence that the policy change is soundly conceived; and in part on the risk that if near-term market developments go against the long-term trend and expectation, a sound long-term policy decision may be interrupted or reversed for essentially short-term reasons.

In almost any situation, depending on the politics of policy formulation, a sound means can be chosen to achieve the end result of a policy of holding bonds only as needed for permanent defensive reserves. No other long-term bond investments should be held for the long-term investors.