The extreme volatility of stock market values has been the subject of a large body of literature. Previous research focused on the short run because of a widespread belief that in the long run the market reverts to well-established fundamentals. The authors' research suggests this belief should be questioned. First, they show actual dividends cannot account for the secular trends of stock market values. They then consider a more comprehensive measure of capital income, which displays large secular fluctuations that roughly coincide with changes in stock market trends. Under perfect foresight, however, this measure fails to properly account for stock market movements. The authors thus abandon the perfect foresight assumption and instead assume that forecasts of future capital income are performed using a distributed lag equation and information available up to the forecasting period only. They find that standard asset-pricing theory can be reconciled with the secular trends in the stock market. This study, nevertheless, leaves open an important puzzle for asset-pricing theory: The market value of U.S. corporations was much lower than the replacement cost of corporate tangible assets from the mid-1970s to the mid-1980s.