Calculating the Intrinsic Value of the S&P 500
Current insightful approaches for evaluating whether the S&P 500 is cheap or expensive include Professor Robert Shiller’s CAPE (cyclically adjusted price-earning) ratio and Warren Buffett’s ratio of market capitalization to GDP. I suggest an additional approach - estimate the intrinsic value of the S&P 500 directly and use that in a comparison to the aggregate cost of buying each company, based on current stock prices. It is almost axiomatic that the intrinsic value of the S&P 500 is the future shareholder distributions (dividends and stock buybacks) of all companies in the S&P 500, discounted to present value.
In valuing a business, I learned from the writings of Warren Buffett the importance of focusing on the business’s return on the capital that it has invested in its business and on the growth in this capital. In fact, if return on capital and growth in capital are assumed, the value of the business is a math calculation for any given discount rate.[1] This should apply also to the entire S&P 500.
I carried out this calculation in January 2023, with the assistance of the business data from my friends at FactSet, whose generosity made this work possible. The graph below shows the calculated return on capital over the last ten years for the companies in the S&P 500, with adjustments to reflect the addition and removal of companies over the time period.
Invested Capital and Return on Capital for S&P 500
[graph removed by application]
Data provided by FactSet. Invested capital (bars and left axis, in trillions); Return on capital (dots and right axis). Invested capital for a given year is the ending invested capital for the prior year. Return on capital is on a pre-tax basis.
This indicates that the S&P 500’s return on capital and growth in capital is remarkably consistent over the last ten years – quite different in this respect than almost all individual companies, in my experience. The graph above represents the summing of the individual data for each component company.
Based on the above, the average pre-tax return on capital is 12.4% over the last ten years and 12.2% over the last five years. I will use the latter. The growth in capital over the last ten years is 5.1%. Invested capital in total for the component companies is $18.3 trillion. This is based on company financial statements through 2021. Extrapolating a pre-tax return on capital of 12.2% and a growth in capital of 5.1% over 80 years, the intrinsic value of the S&P 500 is set forth in the chart below for various discount rates (more on discount rates later). [2]
The last step is to compare the intrinsic value of the S&P 500 to the aggregate cost of buying each company, based on current stock prices. The total cost of buying each company in the S&P 500 on January 20, 2023, based on closing stock prices and 2021 10-Ks, is $43.5 trillion. This includes $36.7 trillion for stock, the payoff of $7.9 trillion in loans and other liabilities and credit for $1.1 trillion of certain cash held by the companies assumed to be free cash. The S&P 500 closed on January 20 at 3972.61.
[chart removed by application]
The value ratio in the above chart compares the intrinsic value of the S&P 500 to the aggregate cost of buying each company in the S&P 500 by dividing the former into the latter. A value ratio of 1.0 indicates that the S&P 500 is fairly valued, a ratio substantially below 1.0 indicates that the S&P 500 is cheap, and a ratio substantially above 1.0 indicates that the S&P 500 is expensive.
These calculations indicate that the S&P 500 is fairly valued at a 6.8% discount rate, which suggests that going forward over the long run one might expect the S&P 500’s return to be substantially below its historic return but still perhaps better than what one might expect from many other diversified, widely available investments. If one uses Professor Aswath Damodaran’s equity risk premium, that would call for a discount rate of approximately 10% and indicate that the S&P 500 is substantially overvalued. In my opinion, the equity risk premium implicitly demands to a large extent that the S&P 500 perform as well in the future as it has performed in the past, and, as shown by the above value ratio of 2.5, under the assumptions outlined in this article, this expectation over the long run cannot be met.
My evaluation of the S&P 500’s invested capital used generally the approach illustrated in my book, One Model for Value Investment Calculations, with some deviations, relating mostly to certain component companies, particularly, for banks I used equity instead of invested capital, and for many REITS I reversed 40 – 65% of the depreciation expense. In the case of some of the companies that I am more familiar with, I reflected some adjustments in the evaluation, based of course on publicly available information. I am unfamiliar with many of the component companies.
1 This concept is widely illustrated, including in my book, One Model for Value Investment Calculations.
2 This assumes a corporate tax rate of 20% for the entire period. This also assumes that in year 80, the shares are sold at a PE of 15.
