Does Population Aging Affect Financial Markets?

10/18/2011
Featured in print Bulletin on Aging & Health

With the baby boom generation nearing retirement age and life expectancies continuing to increase, the U.S. population is aging rapidly. By 2030, the share of the U.S. population that is over age 65 is projected to be higher than it is in Florida today.

Population aging may affect financial markets if individuals tend to amass assets during their working years and spend them during retirement. When there is a large cohort such as the baby boom, there may be more demand than usual for corporate stock and other assets while the cohort saves for retirement. This demand may abate after the cohort retires. Some analysts believe that the rise in stock prices in the 1990s can be partially attributed to this, and have forecast sharp declines in asset prices in coming decades as boomers sell their assets to the smaller baby bust generation that follows them.

In The Impact of Population Aging on Financial Market, (NBER Working Paper 10851), James Poterba examines the potential impact of population aging on asset returns, the valuation of assets, and the demand for various financial assets and products.

The author begins by noting that several factors may mitigate the expected effect of the baby boom and baby bust cohorts on asset prices and returns. If boomers anticipate low returns in the future, they may save less today. If the size of the capital stock can adjust when the demand for capital increases, the change in asset prices will be smaller than with a fixed supply of assets. If global capital markets are well integrated, then asset prices will depend on global demographic forces rather than trends within a single country. If the age structure in the population affects the rate of productivity growth, this could swamp any asset price effects arising from demography-induced changes in asset demand. Finally, if investors anticipate that the retirement of boomers will lead to lower asset prices, this future price decline should be incorporated into the current price of assets. Consequently, asset prices should not experience a sharp decline in the future.

Next, the author turns to evidence on household asset holdings by age to see whether households accumulate assets during their working years and decumulate them during retirement. He cautions that it may be impossible to completely separate the effect of an individual getting older (age effect) from the effect of being part of a particular cohort whose preferences may differ from those of other cohorts (cohort effect) and the effect of having experienced particular events such as a period of high asset returns at a given point in time (time effect).

He shows that mean and median asset holdings both rise with age until about age 60. After this, there is no noticeable decrease in asset values with age, although correcting for the fact that wealthier households are more likely to survive to older ages results in some decumulation during retirement. Poterba notes that there will be additional decumulation of assets in retirement outside of household portfolios, as balances in defined benefit pension plans are sold to finance benefits.

The author then estimates the relationship between the age structure of the population and asset prices and returns using data from the 1926-2003 period. He first examines real annual returns for three assets - Treasury bills, long-term government bonds, and stocks in large corporations. He finds that the effect of the age structure on asset returns is greatest for Treasury Bills, while there is no effect for stocks. However, the relationship for Treasury Bills exists only prior to World War II. When he examines the effect of the age structure of the population on the price of corporate stocks, he finds that an older population is associated with an increase in stock prices, although the results are very sensitive to the particular choice of specification.

Finally, the author examines how population aging may affect the demand for different types of assets by tabulating age-specific holdings of various assets. This exercise is subject to the same difficulty in separating age, cohort, and time effects. Poterba finds that households age 65 and above currently hold about one-third of all corporate stock held by the household sector, and roughly the same fraction of bonds. The over-65 group is projected to hold nearly one-half of these assets in 2040. These older households are also projected to hold nearly two-thirds of annuity contracts in 2040, up from half today.

In concluding, Poterba notes that economic theory clearly predicts that a baby boom should drive asset prices up and asset returns down for its cohort. However, he writes, "none of the empirical findings provide a strong and convincing measure of the amount by which asset prices will change as the population of the United States and other developed nations ages." None-theless, given the inherent difficulties in estimating this relationship, he believes that "the theoretical models should be accorded substantial weight in evaluating the potential impact of demographic shifts."


The author gratefully acknowledges financial support for this research from the Hoover Institution, the National Institute on Aging, and the National Science Foundation.