But should one be as confident as Glassman and Hassett that the process will continue until the risk premium shrinks to zero and the Dow reaches 36,000? I don't think so. It is easy to imagine that some short-term event might shake investor confidence in the long-term stability of the market and push the equity risk premium back up. In fact, a stock market crash might be precisely such a self-fulfilling event. My own guess is that while investors are now satisfied with a smaller equity risk premium than they have had in the past, the risk premium will never fall all the way to zero.That was his only problem with the argument? The model they used posited value as the ratio of dividends relative to the difference between the cost of capital and the long-term growth rate but Glassman-Hassett pretended profits equal dividends ignoring the fact that growth requires investment in new capital. Brad DeLong is happy to mock the past mistakes of our new CEA chair:
Glassman and Hassett get the math of the Gordon equation for valuing the stock market simply wrong. It's not the earnings yield that shows up in the numerator, it's the dividend yield. The book should have been called Dow 22000. Glassman and Hassett get the math of the equity premium wrong. The weighted average of the returns on bonds and stocks is the return on capital. The equity premium is a wedge between the rate of return on stocks and the rate of return on bonds. If the equity premium falls, the rate of return on stocks falls and the rate of return on bonds rises. Hassett calculated the effect of a fall in the equity premium by fixing the rate of return on bonds. The book should have been called Dow 15000. Glassman and Hassett could make the argument that the equity premium ought to go away, and someday might go away, but that is not the argument they make. The argument they make--back in the late 1990s--is that the equity premium will go away in the next three to five yearsYes – Glassman-Hassett got the math of the Gordon equation terribly wrong. Let’s pose a really simple model based on an all equity financed company so the return on assets equals the return on assets. Of course the expected return on assets is generally lower than the return on equity. As a lot of folks were saying the expected return to equity exceeded the risk-free by near 6% back then, let’s delever that risk premium to say 4% currently. Let’s also assume a risk-free rate equal to 3% and a growth rate equal to 2%. Now a pure IP company or a pure franchiser might not require tangible assets so profits equal cash flows, most companies do require tangible assets. Imagine for simplicity a company with $100 million in tangible assets that generates $10 million in expected profits – that is a return to tangible assets equal to 10 percent. What would its price/earnings (P/E) ratio be? If the risk premium is 4% so the cost of capital is 7%, P/E would be 16. Now if Glassman-Hassett were right that the risk premium suddenly fell to zero, P/E would be 80 not 100 as they argued. Even under Mankiw’s suggestion that the risk premium might fall to say 2%, our model suggests P/E would be less than 27. Daniel Gros has more on Hassett:
Hassett’s celebrity rests on being the less flamboyant half of the duo that penned Dow 36,000 one of the most wrong books ever published on investing and the markets.It was wrong not just because it assumed a zero risk premium. It was wrong as it fundamentally did not under the basic finance of the Gordon equation it purported to use.