Over the last 130 years, U.S. equities have generated a total return of 8.9% p.a. including dividends. While there has been significant variation in year over year returns, the long-term investment returns paint a picture of consistent compounding of wealth as seen in Figure 1 below.
The total investment return from equities can be decomposed in four components, namely dividends, inflation, real growth in business value of all listed businesses, and change in Mr. Market’s “perception” of the value of those businesses. Equation 1 below formulates the decomposition of total investment returns among the four components mentioned above.
Total Return from Equities = Dividend Yield + Inflation + Real Gr in Business Value + ∆ Valuation ascribed to Businesses …………………………… 1
While dividend yield and inflation are straightforward numbers, we need some proxies for growth in business value and changes in valuations ascribed by market participants to businesses. For the purpose of our analysis, we have used growth in cyclically adjusted earnings as the proxy for business value growth and changes in CAPE multiple as the proxy for changes in valuation ascribed to businesses by market participants. Table 1 below shows the contribution to total returns from U.S. equities from these four factors over the last 130 years. Clearly, dividends have been the largest component of the total investment returns.
Table 1: Decomposition of Total Return from U.S. Equities
Table 2 below shows decomposition of total returns for each decade. While Figure 1 earlier showed a rather consistent compounding of wealth by U.S. equities, the table below shows that while there were some decades that yielded extraordinarily high total returns (1921-1930, 1941-1950, 1951-1960, 1981-1990, and 1991-2000), there were decades that yielded very low total returns (1881-1890, 1911-1920, 1931-1940, and 2001-2010). In fact, decades of 1911-1920 and 2001-2010 were associated with negative real total returns of -4.9% and -1.2% respectively with the decade of 1971-1980 yielding a real total return of 0%. Interestingly, there isn’t much to distinguish between decades with high total return and decades with low total return in terms of inflation or EPS growth. Further, dividends have made a much smaller contribution to total returns since the nineties.
Modified Equation for Approximating Total Returns
While changes in valuations ascribed to businesses is an important driver of returns over short and intermediate terms, its impact on total investment return over very long periods should be insignificant. The reason why this proposition holds true over very long periods is that the two primary drivers of the multiples assigned by market participants to cash flows or assets of businesses, namely interest rates and growth expectations, tend to revert to mean over long-term.
Our contention that over very long periods changes in business valuations should be a tiny component of total returns is supported by Table 1 which shows that changes in valuation ascribed to businesses as proxied by changes in CAPE multiple was a very small contributor to total investment returns at 0.3%. With this insight in mind, the equation to approximate total return from equities can be rewritten as follows:
Long-Term Total Return from Equities ≈ Dividend Yield + Inflation + Real Gr in Business Value …………………………… 2
Further, inflation plus real growth in business value approximate the nominal growth in the business value and so the equation above can be further modified as below:
Long-Term Total Return from Equities ≈ Dividend Yield + Nominal Gr (Business Value) …. 3
Stock Price Appreciation and Business Value Growth
Given that the total return from equities are equal to the sum of dividend and stock price appreciation, equation 3 makes it clear that over very long periods, capital gains component of return from investing in equities is largely attributable to underlying business value growth. Using the data from Table 1 above, capital appreciation contributed about 4.0% to total return from equities (Inflation + CAE Growth: Real). From the discussion above, we deduce that the underlying business value growth of companies included in the S&P 500 has approximated to an annual average rate of 4.0% over this period.
This assertion that over long periods of time, price returns earned from investing in equities approximately equal the underlying business value growth is also supported from the Flow of Fund Accounts of the United States published by the Federal Reserve. Based on the data published in the Flow of Funds report, the net worth of nonfinancial corporate businesses in the US grew at a CAGR of 7.1% between 1945 and 2014. As growth in net worth of all businesses is a good proxy for underlying business value growth, this tells us that the business value of nonfinancial corporate businesses in the US grew at about 7.1% over that time. Over this same time frame, S&P 500’s price increased at a CAGR of 7.2%!
We conclude with the assertion that price appreciation component of the total return from equities is largely composed of growth in underlying business value.
 For the purposes of this paper, U.S. equities refer to S&P 500 and its components.  Using data provided on Robert Shiller’s website http://www.econ.yale.edu/~shiller/data.htm  In “The Triumph of Optimists: 101 Years of Global Investment Returns”, Elroy Dimson Et al. showed that dividends account for a significant part of the total investment returns over long periods. From 1900 to 2000, a US portfolio that included reinvested dividends would have generated 86 times the wealth generated by the portfolio solely relying on capital gains.  Mr. Market is an allegory created by Benjamin Graham. While explaining the fluctuations in stock prices, Graham says that “Think of you as owning a share in a business in partnership with others. One of your partners, say Mr. Market, is somewhat of a neurotic who on any given day will offer to buy your share or sell you his at a specific price. His moods can fluctuate anywhere between incredible optimism and overwhelming depression. One day he will nominate a higher price to buy or sell, the next day he might increase it, lower it, or even appear uninterested in whether he buys or sells.”  There is also a residual component which is largely a factor of the geometric interactions of the various factors when they are compounded over the long-term. However, for the purposes of our analysis this has been ignored as it tends to be a very small component compared to the other four components.  Cyclically adjusted earnings (CAE) refer to the smoothed earnings measure popularized by Robert Shiller. It is calculated by taking a 10-year average of inflation adjusted trailing earnings. Cyclically adjusted PE (CAPE) is calculated by taking the inflation adjusted prices of S&P 500 and dividing it by CAE.  Note that we have excluded the impact of CAE multiple expansion which contributed 0.3% to the total returns.  Table B.102: Balance Sheet of Nonfinancial Corporate Business, Net worth (historical cost), row number 46.  Over this time frame, CAPE multiple expansion accounted for 0.8% of total returns. Using equation 1, we can calculate that the change in SPX price that was related to nominal business value growth was 6.4%. We contend that the difference between the growth of net worth of all businesses of 7.1% and growth in business value of S&P 500 of 6.4% is primarily related to the fact that smaller sized businesses have enjoyed higher returns / business value growth over this time.