A Century of Global Stock Markets

Summary of working paper 5901

Investing in global stock markets has become the focus of a baby boom generation obsessed with saving for retirement. Behind the flood of money into domestic and international equities by savers is a widespread appreciation that stocks yield a high rate of return over the long term. However, estimates for the rate of return on equity capital typically are based on 20th century U.S. data, which show an equity return (in excess of inflation) of about 6 percent. NBER Faculty Research Fellow William Goetzmann and his coauthor Philippe Jorion point out that this high rate of return may be explained by the fact that the United States had the winning stock market in the 20th century.

In "A Century Of Global Stock Markets" (NBER Working Paper No. 5901) Goetzmann and Jorion construct a series on real capital appreciation for equity markets in 39 countries covering most of the 20th century. Their "financial archeology" shows that the United States is the exception in a global capital market frequently wracked by financial crisis, political upheaval, expropriation, and war. The United States has had by far the highest uninterrupted real rate of return at 4.73 percent a year. In sharp contrast, the median real appreciation rate for the other countries is only 1.5 percent annually.

Stock market volatility in the United States is about average for the world markets. Goetzmann and Jorion suggest that the rich U.S. equity premium over inflation is biased upward by the cumulative impact of survivorship for the world's most successful capitalist system. To be sure, countries such as Germany and Japan have enjoyed strong stock market runs in the post-World War ll period. Many emerging markets have sizzled in recent years. Yet in every case, stock market performance was abysmal in earlier eras.

Can differences related to survivorship make a difference? Goetzmann and Jorion demonstrate that small differences in return over long time horizons can compound into vastly divergent financial outcomes. Compounded at a 5 percent rate of interest, for example, the approximately $24 dollars Peter Minuit, the Governor of the West India Company, paid for Manhattan Island would have grown into about $1.6 billion in current dollars by 1995. But compounded at a 3 percent rate, the financial outcome is a mere $1.3 million.

The authors construct a unique global equity index which not only extends for 75 years but also includes (for the first time) defunct as well as surviving stock markets. Their "survived" index, which takes into account all markets since their last break, returned 5.35 percent between 1921 and 1995. The "continuous" market index, which skips all markets that suffer a permanent interruption but includes all markets with available observations and links any temporary closures, returned 4.86 percent. The "reconstructed" index, which attempts to create a complete series by assuming that markets that fail lose half of their value, returned 4.74 percent. The authors' volatility estimates offer additional support to the risk reduction benefits from investing internationally. Clearly, survivorship pays off big over time. Indeed, the authors conclude that "it should be clear that if we fail to account for the 'losers' as well as the 'winners' in global equity markets, we are providing a biased view of history which ignores important information about actual investment risk."

The authors' global stock market database is patched together from the Morgan Stanley Capital International Perspectives, the International Finance Corporation, the International Monetary Fund, the League of Nations, the United Nations, and the International Abstract of Economic Statistics. Although returns are calculated by using both local currencies and the dollar, the study focuses on real return figures that are derived largely by deflating nominal returns with wholesale price indexes. The monthly data offer a rich mine for gaining a better grasp of how stock markets and equity risk premiums respond to peace and crisis, and allow for comparing the behavior of winning markets to losing markets. For instance, markets that became extinct plunged by 27 percent the year before the break; markets that subsequently recovered dropped by 16 percent.