Introduction

Driving increases in shareholder value is one of the most important responsibilities of any business leader, but many are handicapped by not having a deep understanding of what actually drives value. Understanding valuation is central to being able to develop strategies and make day-to-day decisions that increase shareholder value.

Some people see the stock market as arbitrary and sometimes random in determining value. For the most part, I believe the market is very efficient in incorporating available information to value a company’s shares. Certainly problems with liquidity and other factors can knock valuations out of line for a period of time, but over a longer horizon, not only is the market is efficient, but valuations are easy to explain.

The factors that drive the overall market are actually simple and easy to understand. Through this paper, I will explain those factors. This includes what I believe to be a new and original approach (of my own invention) that combines generally accepted approaches to valuation with a simple but new approach to estimating the Equity or Market Risk Premium (MRP) that produces very good explanations of market P/E ratios and overall market levels. The chart below shows how the Predicted P/E when applied to S&P Operating Earnings explains levels of the S&P 500 over the past 48 years.

My approach to estimating the Market Risk Premium is the most original part of my overall hypothesis. (I have not found any similar approaches through my research, so please let me know if you have seen similar applications.) Many, if not most theorists, assumed the Market Risk Premium was a constant based on historical premiums above the risk free rate. Using the constant earnings growth model (see Constant Growth Model), I plotted P/Es where:

Predicted P/E = 1/ (Cost of capital – Real Growth – Inflation)

Where:

Cost of capital = Risk Free Rate + Market Risk Premium

There seemed to be no relation between predicted and actual PE, while using any range of risk premium constants.

I suggest that that the Market Risk Premium is not a static number but a variable that fluctuates in direct proportion to the long-term risk free rate. The central idea is that the risk premium is fixed as a percentage of the risk free rate, not a fixed premium. Regression analysis showed that when applied to the constant growth model, cost of capital could be determined by the following formula for 1982 - 2002:

Market Risk Premium = Risk Free Long-Term Rate x Risk Premium Factor (0.75)

Intuitively, this seems to make sense. Conventional theory would hold that if the Market Risk Premium constant were 6.0% and 10 year treasury’s were 4.0% then investors would expect equities to yield 10%, but if 10 year Treasury yields were 10% then investors would seem to require a proportionately smaller premium with a 14% return. I propose that using this MRP factor, investors would only expect equities to yield 7.0% when Treasury yields were at 4.0% and 17.5% when they were at 10%, the same proportionate compensation for risk.

This Risk Premium Factor (RPF) seems to hold steady for long periods of time, changing just twice from 1960 – present (April 2009). The RPF was 1.22 from 1960 -1980, 0.75 from 1981 – June 2001 and 1.46 from July 2002 – present. As shown above, the market does a very good job of predicting market levels. The model holds nicely through the present and, assuming that 10 Year Treasury yields are artificially low with the current flight to quality, even through the current financial crisis. Yields have been 2.2%-2.7% over the past few months (ending March 2008), compared to a range of 4.1% to 5.1% in 2006 and 2007 and rarely falling below 4% since 1960.

For more on underlying assumptions behind the calculations please see "The Risk Premium Factor: A New Model for Understanding the Volatile forces that Drive Stock Prices" (Wiley Finance) available on Amazon.  (http://amzn.to/jRHEAe )

Next