It is one of the most important questions an investor should ask: What is the right price to buy (or sell) a security? Unfortunately, it is almost impossible to answer with any degree of certainty.
That does not mean, of course, that nobody has answers. Arguably, there are as many responses to this question as there are investors—and maybe even humans—in the world. Indeed, it is likely that anyone who cares has an approach or algorithm oriented towards this issue. These can include very simple methodologies based on one variable, such as whether one “likes” an investment. Most of us probably know someone—a spouse, perhaps?—who has perfected the use of this one-factor model. Some might even argue that there is much to be said for keeping things simple.
But for many investors, this is not enough. Take the equity market, for instance. Most would probably agree that there are a variety of factors that not only influence the price of a stock, but perceptions about its value. Moreover, the way in which these factors interact with one another can create additional crosscurrents, including feedback loops, that affect attitudes and behavior. The number of variables, then, is truly infinite.
The sheer vastness of the data set and the challenges of finding a workable approach to solving this problem led some very smart people to develop the Efficient Market Hypothesis. EMH postulates that it is impossible to devise an algorithm that enables investors to earn returns exceeding those of the market as a whole. That is because the interactions of participants are, in effect, an algorithm that incorporates all information required to establish the “efficient” or “correct” price of a security.
In other words, it is pointless for anyone to collect data and deploy algorithms in an attempt to predict the future path of share prices, because the latter will quickly and efficiently adjust to whatever new information might come along. That said, it is worth keeping in mind that EMH simply tells us that a security is fairly priced; it does not rule out the prospect that one could expect a positive return for a stock.
Given the stakes involved—not to mention the natural human desire to solve the seemingly unsolvable—it is not surprising that there is a whole industry, collectively referred to as active fund management, devoted to the task of trying to disprove EMH.
As one might have guessed, we are in that camp; we do not accept or agree with the theory’s seemingly logical conclusions. Allowing for some differences in our team’s individual perspectives, we believe it is possible to make predictions about expected long-term equity returns that can serve as guides to whether or not securities are attractively priced. We also believe it is possible to assess the risk to these forecasts. Generally speaking, there are three questions to consider:
- What is an attractive return to judge the expected return of a stock against?
- What is the expected return of a stock?
- What is an acceptable risk to that expected return?
We think the answer to the first, as we noted in “GDP: A Better Benchmark?” can be found in the growth rate of nominal gross domestic income (GDI) and the related economic measure, gross domestic product (GDP), which we believe are the most rational gauges for assessing investment performance.
First and foremost, if investors achieve a return on capital, net of expenses and taxes, in excess of these benchmarks, they will be increasing their capital at a rate that outpaces national output and income.
In fact, such a phenomenon has arguably played a key role in how many of today’s richest individuals and institutions achieved that status. It is, in part, a function of the fact that many capital assets earn returns in excess of the rate of growth in GDP, owing to the lack of competition stemming from some combination of risk aversion, regulatory constraints and barriers to entry. Amplified by ultra-low interest rates and seemingly boundless liquidity, the result has been a breathtaking concentration of wealth in relatively few hands.
To put it simply, those who had sufficient capital to begin with, because of luck, inheritance, criminal acumen, entrepreneurial genius or savings, have been afforded a rare opportunity to accumulate vast fortunes that are out of reach to even the most hard-working and thrifty of their fellow citizens.
The accumulation of wealth relative to the value of a nation’s output of goods and services, therefore, is not necessarily driven by skill, but rather by the interesting historical phenomenon: return on capital is often higher than the future growth of GDP. Eventually, of course, something has to give, and this is where revolutions and wars, natural disasters, pandemics and other mechanisms of self-correction regularly come into play.
Another reason why the well known gauges of economic activity can and should serve as the relevant investment benchmarks stems from the fact that a nation’s ability to produce goods and services likely represents the natural limit to the aggregate value of societal assets. Ultimately, the latter is determined by the output they can be converted into.
Certainly, there have been fluctuations in the relationship between output and the various types of incomes, as we noted previously. Indeed, variations in the distribution of income between rents, interest rates, wages and profits have made it possible for certain asset prices to outpace economic growth, at least for a time. Over the past decade, for instance, equity values increased dramatically as corporate profits assumed a historically large share of the national income mix.
Even so, other evidence points to a strong link between incomes and output. The empirical work of French economist and philosopher, Thomas Piketty, suggests that, over the long run, there is a relatively stable relationship between the value of GDP and society’s assets. While there may be periods when one is high relative to the other, the relationship has been remarkably stable, on average. Few of us have the patience, however, to stand pat until such variations occur, or the psyche to capitalize on opportunities when they arise. Moreover, even with the cyclical swings, the returns from most assets have over time been largely positive and the opportunity costs of waiting for just the right moment can be significant.
So, assuming that the minimum objective for investment activities is to earn returns that exceed the pace of economic growth, the critical question is: How do we calculate the expected return for any particular asset? The answer can be seen as simple or complicated, depending on one’s focus and perspective. In the case of a stock, for example, the expected return is typically calculated as the sum of the dividends that will be paid plus the change in the value of the security over time, expressed as an annual percentage rate.
Quantifying the expected return from dividends is relatively straightforward. Historic dividend payments are known; future payments, of course, are not. Nevertheless, history suggests that established businesses seek to maintain or increase their dividends, subject to limits imposed by earnings, free cash flow and balance sheet requirements. If one is uncertain about how to forecast future dividends, then current distributions can be used as a proxy. Dividing the annual payout by the current share price equals the dividend yield.
The other component of expected return is derived from the growth rate of the stock’s “terminal value”—that is, its market price. To estimate this variable, one approach is to view it from the vantage point of expected long-term economic growth. Generally speaking, the growth rate of earnings for all businesses in an economy will match that of nominal GDP. While there can be periods, as noted above, where that might not be the case, economic growth has tended to be somewhat—and, perhaps, amazingly—stable over time, which suggests this measure is a good starting point for defining the baseline standard.
Looked at another way, if a company’s profit is expanding at a faster pace than the overall economy, it means the business is either taking share from competitors, supplying an expanding market, paying less for labor and other production factors than competitors, or extracting a premium price for its products and services. History suggests that such a company will realize an above-average increase in its terminal value, and vice versa.
The challenge, of course, is to determine how fast earnings are and will be growing. If one assumes a dividend yield of, say, 2%, a company would need a 6% earnings growth rate for its stock to have an expected return of 8%, matching both the long-term annual return from equities and the objective of garnering a return above the pace of economic growth, net of expenses and taxes (Note that only the dividend received is taxable while any tax due on the cumulative increase in a stock’s terminal value is deferred until a sale is made).
The critical issue, then, is to evaluate whether it is reasonable to assume that a company whose stock is being considered as a possible investment will see 6% annual growth in earnings—typically, earnings per share, or EPS. One approach is to break down and evaluate expectations for a company’s earnings in the same way as one might assess the outlook for GDP growth. Among the questions that could be asked are:
- Will the company be able to increase prices at a rate above, below or near the rate of inflation?
- Will it be able to expand its sales at a rate above, below or near the pace of population growth?
- Will it be able to reduce its costs at a rate above, below or equal to productivity growth?
There are other considerations, including the amount of capital a company needs to grow it existing business. If the total equals the earnings it retains after paying dividends, the calculation is fairly straightforward: expected return equals the dividend yield plus anticipated nominal growth in earnings.
However, if the business generates more capital than necessary, the excess should be factored in. For example, with an expected dividend yield of 2%, an expected earnings growth rate of 6%, and excess capital of 2%, the expected return is 10%. The assumption here is that the excess capital is invested at a rate that is at least equal to the existing rate of return on capital. In contrast, if the firm is expected to require additional capital to achieve a presumed level of growth, the expected return would need to be adjusted downward.
There is little doubt that the generation and profitable deployment of excess capital can have a potentially significant impact on returns. If, for example, a company has an earnings yield of 8%, pays out 2% in dividends, reinvests 4% in existing businesses to keep growth on an even keel, and then reinvests the remaining 2% in new businesses that can generate returns and growth commensurate with established operations, that would likely signify an attractive investment opportunity.
In reality, many companies have not been careful stewards of capital. They have squandered vast sums on unsuccessful acquisitions and misguided investment projects. For this reason, it might be necessary to adjust expectations downward if there is a decent risk that those in charge might lead a firm on a similarly destructive course.
Whether a company distributes excess capital in the form of dividends or share buybacks, or by way of investments that presumably generate returns at least in line with current levels, all have different risks and tax implications in regard to forecasting expected returns. Under the circumstances, it is also necessary to anticipate whether the company’s management and board of directors will (or will continue to) act in shareholders’ best interests.
It is worth bearing in mind that any predictions about expected returns are predicated to a great extent on the supposition that the value the equity market assigns to future earnings will remain constant. While anything is possible, history suggests that this outcome is extremely unlikely. This then becomes yet another critical risk to the assessment.
Regardless, such analyses invariably leave numerous questions unanswered or lead to assessments with wide margins of error. There are simply too many variables, including the “unknown unknowns.” But this sort of approach can provide a structured way of thinking about the factors that influence prices and investor perceptions, including dividend yields, earnings growth and terminal values.
Of course, making predictions about expected earnings does not necessarily address the many risks that are inherent in such forecasts, nor does it take account of idiosyncratic factors such as investors’ willingness and ability to assume risk. Moreover, with the methodology detailed above, only company-specific issues have been considered, while the analysis itself rests on certain assumptions about the overall pace of economic activity.
In reality, there are a variety of other, more generalized risks that can impact outcomes. One such example is valuation risk. Essentially, the higher the valuation of a company is relative to its current or future earnings, the greater the risk that expected return forecasts won’t be realized.
In the approach outlined above, for instance, only the current dividend yield and expected earnings growth rate are featured as variables, and valuation is ignored. However, this is somewhat utopian. Practically speaking, the behavior of others can alter the dynamic. For a company where the expected return is estimated to be 8%, the prospect that such performance might not be realized is higher if its stock is trading at 20 times its earnings, rather than 10 times. That is because a larger multiple implies that investor expectations are high and, therefore, are subject to greater risk of disappointment.
Research undertaken by Yale University economics Professor Robert Shiller regarding the relationship between the cyclically adjusted price-earnings (P/E) ratio (CAPE) of the S&P 500 index and its future performance lends further support to the notion that the lower the current valuation of a company’s stock, the higher one’s confidence should be about expected return estimates.
There is also something of a fundamental twist to this argument. In the case of two stocks with the same 2% dividend yield, for example, logic would suggest that the company whose shares are trading at 10 times earnings is in a better position to increase its payout than the more highly valued alternative. After all, with a yield of 2% and a price-earnings ratio of 10, the former would only be distributing 20% of its earnings in the form of dividends; the latter, with a P/E ratio of 20, would be paying out 40%, or twice as much.
Finally, another category of risks that can put pad to expected return forecasts are those that fall under the broad heading of “macro.” These can range from higher taxes to natural disasters to the confiscation of private property to a collapse in economic activity stemming from geopolitical events, pandemics and large-scale policy mistakes. Because these can be difficult or even impossible to anticipate in terms of their timing and impact on any one company or sector, many investors simply choose to mitigate the risk through global and asset class diversification, the deployment of multiple investment strategies, and hedging.
In the end, the process of trying to figure out what a security is worth can raise more questions than it does answers. But without some sort of defined process and a measure of understanding of the factors involved, the balance may prove to be heavily—and unfavorably—skewed in favor of the former.
Please see important disclosures and disclaimers: http://www.odincapitalmanagement.com/legal.