The Myth of 1926: How Much Do We Know About Long-Term Returns on U.S. Stocks?

Abstract

Investors are often told that the long-term return on U.S. stocks is a known quantity. Few realize that virtually all such statements are based on a data set that begins in 1926, and that because stock trading in the United States is much older, this dataset excludes more than half the total trading record. Even post-1926, certain types of stocks were systematically excluded, up until very recently. This article chronicles the biases and limitations of the post-1926 CRSP-Ibbotson dataset, and explains how these biases limit what is presently known about long-term returns on U.S. stocks.