We formed Epoch Investment Partners, Inc. in 2004 to take advantage of the changing dynamics of the global marketplace. Specifically, we perceived a paradigm shift in the order of the sources of returns: a phenomenon, which, in our opinion, no investor could afford to ignore. Over the years, our investment philosophy has revolved around this specific understanding of how the markets generate return. Now, in the first decade of the 21st century, this framework continues to provide new and valuable perspectives to the informed investor.
The heart and soul of our investment philosophy is captured in Figures 1 and 2.
Figure 1: Components of Compound Annual Total Returns for Trailing 10-year Periods (S&P 500 Index 1926-2006)
Source: Epoch Investment Partners, Inc. and Standard & Poor’s
Figure 2: Components of Total Returns by Decade (S&P 500 Index 1927-2006)
Source: Epoch Investment Partners, Inc. and Standard & Poor’s
Figure 1 shows an 80-year history of equity returns over rolling ten-year periods. Figure 2 shows the same information, but displayed decade by decade. In both figures, the black line for “Total Return” shows a wide range of outcomes: during some decades, returns approached 20% per year. At other times, however, returns were in the low single-digit territory.
To understand how these fluctuations occur – and to devise an investment model that will capture the resulting opportunities – it is necessary to disaggregate “Total Return” into its component parts: the sources of return. There are only three sources of return – earnings, dividends, and P/E ratios. In Figures 1 and 2, each of these sources is shown as an individual contributor to overall market performance. The important thing to keep in mind is that the sources of return never change, but their order of importance does. That, is, during any given decade, one of these variables exerts a more dominant influence on total equity return than the other two. In the 1980s and 1990s, for instance, the P/E ratio had the largest positive effect on the era’s spectacular market performance.
Let’s take a closer look, however, at the behavior of the P/E ratio. It is clear that P/E expansion can have a dramatic positive impact on total return. But it is crucial to note that the contribution of this variable is highly erratic: sometimes positive, sometimes negative. Specifically, when the P/E ratio expands, the stock market generally produces double-digit returns, and when the ratio contracts, returns fall into the single-digits.
The 20-year period between 1980 and 2000 was one of the eras in which P/E expansion significantly enhanced market performance. As noted earlier, these decades generated record-breaking equity returns, in terms of both size and duration. During this time, the S&P 500 (our proxy for equity returns in general) compounded at 17.6 % per year. As shown in Table 1, this growth was predominantly driven by P/E expansion.
Table 1: S&P 500 Total Return (1980-2000)
Source: Epoch Investment Partners, Inc. and Standard & Poor’s
What caused this historic expansion in P/E multiples? The answer is falling interest rates. Figure 3 shows how interest rates and P/E ratios are highly correlated in an inverse manner.1 When one falls, the other rises – and vice versa. For the two decades between 1980 and 2000, plummeting interest rates led to skyrocketing P/Es, which resulted in huge growth in total equity returns.
Figure 3: P/E’s and Interest Rates: Long-term History S&P Index (1954-2006)
Source: The Federal Reserve and Standard & Poor’s
But, as any investor knows, the current investment landscape has started to bear less and less resemblance to the bygone days of rampant P/E expansion. In our opinion, this paradigm shift is based around the belief that future interest rates are more likely to rise than fall and therefore, P/E ratios are more likely to fall than rise. This assertion has already proven true for the first six years of this decade, as shown in Figures 1 and 2. It is important to note that, in the early part of the decade, this P/E decline reflected the bursting of the equity and dot.com bubbles, in addition to the collapse of the NASDAQ. More recently however, the modest decline in P/E multiples reflects the fact that interest rates bottomed in June 2003 at 3.11%.2 Since then, the 150 basis point rise in interest rates has kept downward pressure on P/E ratios, which has contributed negatively to total equity returns in the 2000-2006 period.
The negative outlook for P/E ratios brings us back to our previous discussion of the sources of equity return. Again, there are only three sources: P/Es, earnings and dividends. So far, we have only discussed one of these sources; now, let’s consider the impact of the remaining two.
By revisiting Figures 1 and 2, it is clear that earnings and dividends have had a consistently positive effect on total return throughout the history of the markets. This distinguishes them from the highly volatile P/E ratio. Over the past six years in particular, earnings and dividends have continued to boost total return, while P/Es have hindered market performance. Among these three sources of return, earnings have been especially strong during this period, compounding at over 8% per annum. Like P/Es in the 1980s and 1990s, earnings have recently reached record highs, but may be poised for an upcoming decline from their recent double-digit growth rates.
The recent increase and potential decrease in the return-generating importance of the earnings variable can be explained by three factors. The most relevant factor behind the increase is globalization. The addition of three billion people into our capitalistic society has effectively doubled the world’s workforce. However, while these people represent a huge addition to available labor, they have brought no capital with them into the global marketplace. As a result, the world has seen a seismic shift in the relationship of economic returns between labor and capital. That is, return to capital has soared while return to labor has grown much more slowly.3 This fact is reflected in the strong earnings growth of corporations around the world.
In fact, return to capital (which is a function of earnings, or profit as a percentage of GDP) is now at an all-time high, as shown in Figure 4. Many forecasters anticipate an impending reversal to this growth pattern, but we do not believe such a decline will be the result of globalization. While corporate earnings (i.e. the profits to GDP ratio in Figure 4) may soon flatten or perhaps decline, the overall effect of globalization has been a permanent shift upwards to return on capital. This has occurred simply because we have re-wired the world’s factory floor and must continue the integration of hundreds of millions of new laborers into the global workforce. The application of Ricardo’s Law of Comparative Advantage has facilitated this remarkable process, while at the same time allowing for an acceleration of worldwide real GDP growth, higher global productivity and lower inflation.4 Figure 5 illustrates how globalization has resulted in the rapid rise of global trade as a percent of world GDP. In short, globalization has been a positive force, lifting both corporate earnings and human aspirations around the world. Although not every person or local industry may have benefited from our new borderless economy, it has been a collective win for all nations. We have seen this reflected in the recent prominence of earnings as a dominant source of total equity return.
Figure 4: G7 Corporate Profits as a Percentage of GDP (1980-2005)
Figure 5: Global Trade as a % of World GDP
Source: IMF, Morgan Stanley Research
The second most important explanatory factor for the recent gains in corporate earnings – and, perhaps, their potential decline – is the worldwide housing bubble. In the U.S., this has been the biggest housing bubble in 100 years – and it is over. While it remains difficult to predict the speed and impact with which this bubble will pop, its scale and perpetuation has already exerted a dramatic influence on company earnings.
In a previous white paper5, we illustrated the sources of corporate earnings over the past four years. Corporate profit gains over this period were driven by several variables, many of which were directly related to the housing bubble: consumption exceeded disposable personal income by $140 billion; investment spending rose by $440 billion, with residential investment companies generating over 65% of that number; the fiscal deficit grew by $380 billion; and the trade deficit rose by $410 billion (a subtraction to the profits total). The nearly $300 billion increase in residential spending also spawned the gap in consumer spending and disposable personal income through the aid of the home equity loan, “no doc” mortgages, and other sub-prime lending schemes. As a result of these combined factors, earnings continued to grow, boosting overall market returns.
However, as discussed in a more recent white paper from January 2007 – “Goldilocks is Dead” – residential housing is about to go into a recession, which will in turn negatively affect consumer spending. These two events will significantly slow corporate earnings gains.
In other words, earnings were turbo-charged by the residential investment spending and home equity loan expansion that comprised the recent housing bubble. Now, as the bubble pops, direct residential investment spending will actually contract this year from 2006 levels, which will lead to an earnings decline.
The third damper on continued earnings growth is the rise in unit labor costs. In many cases, unit labor costs can be used as a proxy for profit margins. Figure 6, shows how the combination of rising productivity and low-wage increase resulted in several years of falling unit labor, which had a beneficial impact on profit margins and, therefore, on earnings. This year, however, we are starting to see a reversal in this positive trend.
Figure 6: Non Farm Unit Labor Cost: YoY% Change, 1996-2006
Source: Bureau of Labor Statistic
Our discussion of earnings has led to the following conclusion: the double digit earnings gains of 2002 to 2006 are likely behind us. While earnings should continue to rise in 2007 and 2008, this growth will occur at a more modest, low-single-digit rate.
Now that we have discussed the first two sources of equity return – P/Es and earnings – we are left with the third and final source: dividends. Dividends, like earnings, are always additive to equity returns. And, in our view, they represent today’s most compelling investment return source.
In our recent book, Free Cash Flow and Shareholder Yield: New Priorities for the Global Investor6, we develop the argument for why free cash flow will emerge as the dominant allocator of capital in the public markets (as it already has in private equity and capital budgeting decisions within corporate Treasury departments). This leads to the conclusion that dividends – because they are often a preferred method of free cash flow deployment – will become a dominant driver of total equity returns.
Our definition of a dividend goes beyond the traditional cash dividend; it also encompasses share buybacks and debt paydowns. Behind this definition lies a fundamental investment truth: if a firm can invest an unencumbered (free) dollar of cash flow at its cost of capital or higher, it should do so, as it is accretive to shareholder value. Such an investment can take the form of an internal investment or an acquisition. On the other hand, if a firm cannot reinvest that dollar of free cash flow at its cost of capital, it should return that dollar to its investors through one of three means: cash dividends, share repurchases, or debt paydowns. In an economic sense, these three cash deployment options are all functionally dividends. We refer to these dividends collectively as shareholder yield.7
Dividends, free cash flow, and shareholder yield have never been more crucial to successful investing than they are today. Corporations, particularly in the U.S., have never been more liquid. Dividends have not kept pace, however, with this growth in free cash flow. Figure 7 shows that dividends as a percent of cash flow is almost at a 30-year low. The 2006 dividend payout ratio for the S&P 500 was 29%, the lowest in 12 years and much lower than its historical average level of 45% to 50%.
Figure 7: Dividends as a Share of Free Cash Flow 1974 through January 2007
Source: Corporate Reports, Empirical Research Partners Analysis. Largest 1,500 stocks; data smoothed on a trailing one-year basis. Excluding Microsoft’s special dividend in 12/2004.
Taken together, this data points to the market’s capacity for increased dividend payouts, whether in the form of cash or stock buybacks. In fact, the trend toward higher stock buybacks has already made itself apparent within the equity markets; according to Standard & Poor’s, S&P 500 companies repurchased $464 billion with of shares in 2006, nearly $100 billion more than 2005 which was $100 billion more than 2004. It is only a matter of time before case dividends and debt paydowns – the other two components of dividend-derived Shareholder Yield – follow suit. When this occurs, total equity return will certainly be impacted. We can already see evidence of this trend today: Morgan Stanley has pointed out, for example, that if one combines the current totals for cash dividends and share buybacks (net of share increases and option exercised) the effective yield on the U.S. market is over 4%.
The burgeoning popularity of dividend-based cash deployment strategies can also be seen in emergence of restricted stock units (RSUs). Corporate pay policies are abandoning options for RSUs, largely because options are now required to be expensed by the FASB. Plus, with interest rates more likely to rise than fall, windfalls from options are increasingly unlikely. RSU plans, on the other hand, allow for management and employees to receive the dividends paid on vested and awarded (not yet vested) shares. To us, this seems like an excellent indication that dividends – in the forms of cash dividends, share repurchase, and debt paydowns – are on the rise as an increasingly important source of total equity return.
In summary, the investment philosophy at Epoch is based around the changing order of the sources of equity return. We believe that, for the foreseeable future, P/Es are unlikely to expand, and earnings growth will be more moderate than in the recent past. Hence, dividends, in all three forms, may well emerge as the dominant component of positive equity returns. Because of this focus on dividends, investors should learn everything they can about a company’s free cash flow: its source, its transparency, its growth rate, its volatility and its optimum applications given the firm’s cost of capital. As we continue to face the realities of the new millennium, this strategy offers the best chance for successfully navigating the global investment landscape.
1The correlation coefficient over this 20-year period was -0.81. Such a correlation is statistically concluded to be highly significant.
2See our white paper, “3-11 – A Rate, Not a Date.”
3This can explain several recent phenomena: the move to populism and protectionism in the new Congress and the general proliferation of anti-Chinese sentiment.
4See our white paper, “Why the World Works in the Long Run.”
5See our white paper, “What the Market Has Not Discounted.”
6Free Cash Flow and Shareholder Yield: New Priorities for the Global Investor, by William W. Priest and Lindsay H. McClelland, published by John Wiley & Sons, December 2006.
7We have created a mutual fund exclusively for this purpose. We also manage similar funds for CI Funds in Canada and John Hancock.