The extraordinary returns earned by common stock investors during the 1990s, and the associated growth in the value of stock market wealth relative to traditional measures of economic activity, has made the stock market a subject of popular and media attention. At the end of 1999, the market value of the shares in domestic corporations was 1.8 times the gross domestic product of the United States. This ratio has been higher in the last few years than at any other time in the period since World War II.
The rising equity market, along with other structural shifts in the U.S. economy such as the growing importance of self-directed retirement plans and the growing importance of stock options in employee compensation, have stimulated very broad investor interest in the stock market. This has been reflected in rapid growth in the number of individuals who own corporate stock.
The Survey of Consumer Finances provides a consistent set of statistics for various years on the number of shareowners in the United States. The Survey of Consumer Finances (SCF) is an ongoing survey conducted by the Survey Research Center at the University of Michigan on behalf of the Federal Reserve Board. It is widely used in academic and government studies of household asset and liability profiles, and more generally in investigations of household financial behavior. The Survey has been carried out every three years since 1983. The most recent survey was conducted during 1998, and information from this survey was released in 2000. Kennickell (2000) documents the generally high quality of the responses to this survey, and Kennickell, Starr-McCluer, and Surette (2000) report summary findings on household net worth.
Most of the core questions in the survey remain the same from one survey to the next. This makes it possible to track stock ownership and other aspects of household financial behavior in a comparable fashion over time. Unlike stand-alone surveys that provide only a snapshot of the number of stockholders at a single point in time, the SCF makes it possible to measure changes as well as levels of stock ownership.
The SCF is particularly valuable for studying shareownership because it over-samples high net worth and high-income households. Since these households account for a substantial share of total stock ownership, the SCF should yield more reliable estimates on ownership patterns than surveys that contact random samples of the population without regard to their income or wealth.
The survey focuses on household assets and liabilities. This is because for most purposes, the net worth of a household is the key determinant of the well being of individuals within the household. The survey asks if anyone in the household owns corporate stock. For comparability with the long line of previous NYSE Shareownership reports, this study nevertheless focuses on the number of individuals who own corporate stock. For single-person households, the individual coincides with the household. For married couples, if the couple owns stock, this is assumed to translate into two shareowners. In the vast majority of households, this is likely to be the appropriate treatment of stock holdings. For some households, however, there are other "unaffiliated adults" who will not be counted by this procedure. A married couple living with an adult child would be one example of a household with such an adult; a couple living with an elderly parent would be another. In cases such as these, the head of household is assumed to be the stockholder, and the unaffiliated adults are assumed not to own stock. This approach excludes unaffiliated individuals from the population of potential stockholders in calculating the fraction of the population that owns stock.
The Number of Shareowners
Table 1 reports the number of individual shareholders in the U.S. population at four points during the last eleven years. It shows estimates of the total number of stockholders from the 1989, 1992, 1995, and 1998 Surveys of Consumer Finances. The table employs several different definitions of shareownership. The first row uses the narrowest definition. It shows direct stockholders, those who own shares of stock in individual companies. In 1998, there were 33.8 million direct stockholders.
The second definition of stockownership, reported in the second row of Table 1, adds stock ownership through a taxable mutual fund to direct ownership. The third definition adds ownership through retirement plans such as Individual Retirement Accounts and 401(k) plans. Many individuals who own stock through retirement plans hold mutual funds in their IRAs or 401(k)s. These shareowners are not included with those who own taxable mutual funds in the second definition of shareownership.
The last and most expansive definition of shareownership defines a shareowner as someone who owns stock directly, through a mutual fund, in a self-directed retirement saving account, or through a defined contribution pension fund. Defined contribution pension plans differ from self-directed retirement accounts in that the pensioner typically does not control the allocation of assets in these plans.
The bottom row of Table 1 presents the latest data for the broadest measure of shareholding. It shows that there were an estimated 84.0 million shareowners in 1998. This represents an increase of 21 percent from 69.3 million in 1995, and a 61 percent increase from 52.3 million in 1989. The 1998 SCF suggests that there were 162.2 million household heads and spouses of household heads at the time of the survey. There were another 30.9 million unaffiliated adults, defined as persons over the age of 18 but who were neither heads of household nor spouses of heads of household. The results in Table 1 imply that 51.8 percent of all individuals who are either household heads, or spouses of household heads, own stock. This is the first time, since the modern version of the Survey of Consumer Finances began in 1983, that this probability of shareownership exceeds fifty percent.
The estimate of the total number of shareholders can also be compared with the total number of heads of household, spouses of heads of household, and unaffiliated adults over the age of 18 in the Survey of Consumer Finances. In this case, the probability of shareownership is 43.5 percent. This comparison is similar in spirit to calculations reported in earlier Shareownership reports, which scaled the number of shareholders by the total adult population in the United States. However, this probability under-estimates the probability that an individual in the population owns stock, since it does not allow for the possibility that unaffiliated adults own equity.
The estimate that 51.8 percent of household heads and their spouses own corporate stock is slightly higher than Kennickell, McCluer, and Surette's (2000) estimate, also based on the 1998 SCF, that 49.7 percent of households own corporate stock. Because married couples are more likely to own stock than single persons are, however, the fraction of the individuals owning stock is greater than the fraction of households owning stock. These results are therefore quite consistent.
The number of shareholders increased by nearly 15 million between 1995 and 1998. This is a larger increase than in either of the two preceding three-year periods. This corresponds to nearly an eight-percentage point increase in the percentage of household heads, and spouses of household heads, who own stock either directly or indirectly.
The four definitions of shareownership in Table 1 highlight the difference between direct and indirect stock ownership. Roughly 34 million individuals, or forty percent of all shareholders, reside in households that own stock directly. While the number of direct stockholders declined in both absolute numbers and as a percentage of the population in the 1992-1995 period, this trend reversed during the 1995-1998 period. The number of direct stockholders rose by 6.4 million between 1995 and 1998.
The rapid growth of direct stockownership between 1995 and 1998 was matched by a sharp increase in indirect ownership. The data may somewhat overstate the total number of indirect owners, since some stock holders who own stock through IRAs or 401(k) plans own individual shares, not mutual funds, in their retirement accounts. The SCF does not distinguish such stock ownership from equity ownership through financial intermediaries, so since the majority of equity investments through retirement plans are made through financial intermediaries, the calculations treat all equity investments through retirement accounts as indirect investments.
The data in Table 1 show the total number of individuals who own stock, according to increasingly broad definitions of stock ownership. With the exception of direct holding of stock, the table does not show the total number of individuals who own stock through each distinct channel of ownership. These statistics may nevertheless be of independent interest. The 1998 Survey of Consumer Finances suggests that 33.8 million heads of household, or spouses of heads of household, own stock directly. The survey also shows that a total of 26.8 million household heads, or spouses of heads, owned stock through an equity mutual fund, while 33.9 million owned stock through a self-directed retirement account and 48.3 million owned stock through a defined contribution pension plan.
The "grand total" number of individuals who own stock is less than the sum of the number who own stock in each way, because many individuals own stock in multiple ways. For example, while 33.8 million individuals own stock directly and 26.8 million own stock through equity mutual funds held outside retirement plans, the total number of stockholders who owned stock either directly or through equity mutual funds was 48.5 million. (This is the entry in the second row of Table 1.) This total reflects the fact that of the 26.8 million equity mutual fund owners, 12.1 million also owned stock directly. The total number of shareholders owning stock in either way was therefore 33.8 + 26.8 - 12.1 = 48.5 million.
The incremental number of shareowners who own stock through self-directed retirement plans (the third row in Table 1, less the second row) is 27.3 million, even though 33.9 million owned equity through some form of a self-directed retirement plan. This reflects the overlap of 6.6 million individuals who owned stock through a self-directed retirement plan and also held stock directly, or through an equity mutual fund, or through both of these alternative channels.
Finally, the incremental number of individuals who own stock through a defined contribution retirement plan, the last row of Table 1 minus the third row, is 8.2 million. This incremental value is low, even though 48.3 million individuals held some equity through defined contribution plans, because of the substantial overlap between those who hold equity through this channel and through the other channels considered in the earlier rows of the table.
Individuals who participate in defined contribution plans that hold corporate stock are considered stockholders, while those who participate in defined benefit pension plans are not. Defined contribution pension plan participants receive retirement benefits that are linked to the performance of the plan assets. This implies that such individuals benefit and suffer when stock prices vary. In contrast, individuals who are covered by defined benefit pension plans, those that promise retirement benefits based on the individual's wage history and years of service, are not considered shareowners even when their pension plans hold corporate stock. These workers have a pension plan that promises benefits that are not a function of asset returns. In this case, pension plan assets effectively belong to the owners of the corporation that promises pension benefits to its workers. The data in Table 1 suggest that the number of individuals who own stock only through a defined contribution pension plan was relatively stable between 1995 and 1998.
The growth of shareownership between 1995 and 1998 is not due to the expansion of any single means of owning stock. There was growth in direct stock ownership, in ownership of stock through equity mutual funds, and in ownership of stock or stock mutual funds through self-directed retirement plans. The rise in direct stock ownership marks a departure from recent years, when growth of mutual fund ownership and stock ownership through retirement plans accounted for most of the expansion in the number of stockholders.
Characteristics of the Owners of Corporate Stock
It is difficult to characterize the individuals who own stock, because they are drawn from a broad cross-section of the United States population. Individuals in all income and demographic categories are represented in the ranks of stockholders, and they are dispersed across the various regions of the United States.
Table 2 provides an overview of shareowners in the United States. It reports information on the entire shareowner population, as well as on two subgroups. These subgroups are "Baby Boomers," who were born between 1947 and 1962, and "Senior Citizens," those born before 1933. The latter group would have been at least 65 years of age at the time of the 1998 Survey of Consumer Finances.
In 1998, the median shareholder was 44 years old, employed, and had a family income of $57,000. (The median is the value that half of the shareowners are above, and half are below. Thus, half of all shareowners had a family income of less than $57,000.) The value of the median shareowner's stock portfolio, including assets held in retirement accounts, was $28,000. This represents a substantial increase from 1995, when the median value was $15,500. This increase is particularly striking in light of the sharp expansion in the number of shareholders, since new shareowners typically have smaller equity holdings than those who have held shares for a longer period of time. The strong returns on corporate stocks over the 1995-1998 period are a key factor explaining the growth of portfolio values.
Some shareowners have stock portfolios that are much larger than the median holding. The mean value of equity holdings was $148,500, and for individuals over the age of 65, the mean value was $267,700. For "Baby Boomers," the median equity portfolio had a value of $31,000, while the mean value of all equity portfolios was $115,800.
Table 2 presents some information on the structure of portfolios for typical shareowners. Just over thirteen percent of all shareowners have direct stock investments in only one stock. Since 40.2 percent of all shareowners have direct stock holdings, this implies that roughly one third of the investors who have direct stock investments hold only a single stock. Only three percent of the heads of households in the SCF, or spouses of these heads, own only stock in a single company. Among those who are over the age of 65, however, six percent own a one-stock portfolio.
There has been a gradual trend toward growing diversification among the shareholding population. In the 1989 Survey of Consumer Finances, for example, 22 percent of direct stockholders reported that they held shares in only one firm. In 1995, the comparable number was 15 percent.
The mean number of stocks held by shareowners who invest directly is 3.4. This value is lower (2.1) for "Baby Boomers," and it is higher (5.2) for those over the age of 65. Because the number of direct stockholders shown in Table 1 is less than half of the total number of shareowners, the median shareholder does not hold any stocks directly. This pattern has persisted for many years, underscoring the importance of indirect shareownership.
More than half of all shareowners, and more than two thirds of those in their retirement years, have a self-directed retirement account such as an Individual Retirement Account or a Keogh plan. Roughly one third of all shareholders, and slightly less than half of those over the age of 65, own at least one mutual fund. Thirty percent of all shareowners, but 46 percent of those over the age of 65, have brokerage accounts.
Table 2 also presents some data on the demographic profile of shareowners. Eighty-two percent of shareowners are married, 38 percent have completed college and one in six has done some post-graduate work. Seventy-seven percent of all shareholders, and 88 percent of Baby Boomer shareholders, are currently employed, while only 21 percent of those over the age of 65 are employed.
Despite the rapid growth of the number of individuals who own stock, the median demographic characteristics of shareholders have remained stable over time. In the 1992 Survey of Consumer Finances, for example, the median age of all shareowners was 45, median family income was $52,000 ($1992), and 79 percent of shareowners were married. There have been significant changes in several shareowner characteristics. These include the percentage of shareowners with IRAs or Keogh Accounts (30 percent in 1992, more than half in 1998), the percentage owning mutual funds outside of retirement accounts (24 percent in 1992 vs. 32 percent in 1998), and the value of the median shareowner's equity portfolio ($14,000 in 1992 vs. $28,000 in 1998).
To place the change in median portfolio value over the period 1992-1998 in context, it is helpful to consider the change in share values during this period. The average value of the NYSE Composite Index was 229.01 in 1992, compared with 550.26 in 1998 (see U.S. Council of Economic Advisers (2000)). Thus, an investor who held $14,000 in a broadly diversified equity portfolio in 1992, and who did not reinvest dividend payouts or sell his shares, would have had a portfolio worth $33,640 in 1998. The actual median portfolio value in 1998 was $28,00, which is probably below the appreciated value of the 1992 portfolio because of many new shareholders who have smaller holdings.
The summary statistics in Table 2 do not capture the rich heterogeneity in the composition of U.S. shareowners. This can be illustrated with a number of more specific tables that address a broad range of issues about shareholder characteristics.
Age Distribution of Shareholders
Table 3 presents summary information on the age distribution of shareowners. The upper panel in the table considers the broadest measure of shareownership, while the lower panel focuses on direct stock ownership only. For the broadest shareownership measure, the data show that roughly two thirds of shareholders are between the ages of 35 and 64. The percentage in the lower panel of Table 3, which focuses on direct stockownership alone, is nearly identical. Thirteen percent of shareholders are over the age of 65, while 22 percent are under the age of 35, when the broadest measure of shareownership is used. When shareownership is defined as direct ownership of corporate stock, a smaller fraction of the shareowners are under the age of 35, and a higher fraction is over the age of 65. (In evaluating these data on the age distribution of shareowners, it is important to recall that the SCF does not capture shareholders who are not household heads or spouses of these heads. In particular, minor children who may own shares in their own right are excluded from this analysis.)
The average value of stock portfolios rises with the age of the shareowner, which means that the age distribution of stock owned is different from the age distribution of shareowners. The last column of Table 3 shows that while individuals under the age of 35 account for 22 percent of all shareowners, on the broadest definition of shareholding, they account for only 6 percent of all shares owned. The analogous statistics for direct stock ownership are 18 percent of the shareowners under the age of 35, but only 4 percent of the stock owned by these individuals. Those over the age of 65 own roughly twice as much stock as their relative frequency among shareowners would suggest.
Table 4 presents data on the probability that individuals in different age groups owned stock in 1998. To streamline the presentation, the tabulations in this table, and in all subsequent tables, are restricted to the broadest measure of direct and indirect shareownership. The data in the first column of Table 4 show that individuals between the ages of 35 and 64 have higher probabilities of owning stock than either older or younger individuals. For those between the ages of 35 and 64, the probability of owning stock is slightly greater than 60 percent. Individuals over the age of 65 have the lowest probability of owning stock, 35 percent, while those under 35 have a 45 percent chance of ownership.
One striking development that Table 4 illustrates is a shift in the relative probabilities of stock ownership for young and old individuals. In 1989, the probability that an individual under the age of 35 owned stock was roughly equal to the probability that an individual over the age of 65 owned stock. In 1998, the ownership probability for younger individuals was substantially higher than that of older individuals.
Table 4 also presents information on the evolution of shareownership probabilities for individuals in given birth cohorts. The SCF samples different households in different years, so it is not possible to track the shareholding behavior of the same individuals over time. Since the SCF represents a random sample of the population in each year, however, it is possible to compare the stock ownership probabilities of individuals who were born in the same set of years across different surveys. This provides some insight on the ages at which shareholding probabilities are changing most rapidly.
Table 4 shows that for those who were under the age of 35 in 1989, and who had a 28.9 percent probability of shareownership in that year, the probability of shareownership in 1998 was 51.8 percent. The probability of shareownership increased between 1989 and 1998 for each age group except those over the age of 65, for whom ownership rates stayed relatively constant.
These data tracking age cohorts are important for understanding the source of rising shareownership. They demonstrate that the growth of the number of stockholders is not simply the result of rising shareownership among young individuals who are entering the adult population, but rather the result of rising share ownership amongst individuals in most age groups. For individuals who were between the age of 45 and 64 in 1989, for example, and thus were between the ages of 54 and 73 in 1998, the probability of shareownership rose by 13 percentage points, from 38.9 to 51.7 percent, over this nine-year period. For those aged 35 to 44 in 1989, the ownership probability increased by 17.5 percentage points.
Where Do Shareholders Live?
To explore the geographical pattern of shareownership in the United States, Table 5 presents data on the number and the percent of shareholders who reside in different Census regions and the probability of owning shares for individuals in these regions. The South accounts for the largest number of shareholders, but this is only because it is the region with the largest population. While one third of all stockholders reside in the South, 35.3 percent of the population is in this region.
The probability of shareownership is lower in the South (48.8 percent) than in other regions. It is very similar in other regions, ranging from 52.6 percent in the Northeast to 53.9 percent in both the North Central and West regions. There has been some convergence over time in the shareownership probabilities in different regions, while rates of shareholding have increased in all regions. As recently as 1995, the probability of shareownership was substantially higher in the Northeast (51 percent) than in the South (39.4 percent) or the West (41.7 percent). The twelve percentage point increase in the probability of shareownership in the West, and the nine percent increase in the South, are very significant changes for a three year period. The growth in shareholding in these regions may reflect the diffusion of employment-linked stock ownership plans. The growth in stockholding in these regions is evident in both direct and indirect means of ownership, so it cannot be attributed only to a single factor.
Education, Occupation, and Income of Shareholders
Table 6 presents detailed information on the educational attainment of shareowners. It shows that stockholders have varied educational backgrounds. Thirty-four percent of shareowners have no more than a high school education, while slightly more than one quarter attended college but did not complete a four-year degree. One shareholder in six completed college and did some post-graduate work.
Table 6 shows both the fraction of shareholders in different educational categories, as well as the percentage of shares owned by different investor classes. The amount of shares that each individual owns is the sum of the value of directly held shares and an estimate of the value of equity held through mutual funds or other indirect channels. Imputations are needed when the SCF does not report a precise value for the amount of shares held. For example, an SCF respondent may indicate that he holds $50,000 in an Individual Retirement Account that is invested in a mutual fund that holds stocks and bonds. In this case the account value would be divided equally between the two assets, so the individual would be assigned $25,000 of stock holdings.
Table 6 shows that on average, the amount of stock owned (conditional on being a shareholder) is greater for those with more education. This implies that the share of stock held by those in higher education categories exceeds the fraction of shareowners in these categories. While individuals with college degrees or post-graduate training account for 38 percent of all stockholders, they hold 63.8 percent of the outstanding shares.
Table 7 presents information on the occupation rather than the educational attainment of shareowners. The data show that individuals who work in executive or professional occupations account for roughly one third of shareowners. While this is the largest single category of shareholders by occupation, there is great diversity in the occupations of shareholders. Roughly one fifth of all stockholders are in clerical, technical, or sales occupations. The average stockholdings of this group are modest, however, since this group accounts for a smaller percentage of shares held (12.5 percent) than of shareholders (20.3 percent). Retirees comprise eleven percent of shareholders, but they hold roughly twenty-two percent of the outstanding stock. This reflects the rising size of average equity portfolios as households age.
Table 8 shows the family income distribution for shareowners in 1998. Shareowners are widely dispersed across the income distribution, while the amount of shares owned is somewhat more concentrated. The table indicates that nearly sixty percent of shareowners lived in families with 1998 annual family income between $25,000 and $75,000. Eleven percent of shareholders had family incomes below $25,000, while nearly 18 percent of shareholders had family income of more than $100,000. Less than four percent of shareowners had family incomes in excess of $250,000.
The last column in Table 8 shows the percentage of outstanding shares owned by shareowners in various family income categories. Because average portfolio size is an increasing function of family income, ownership of shares is more concentrated across income groups than the number of shareholders. Those with family incomes of more than $250,000 held 35 percent of the outstanding equity, while the 9.6 million shareholders with family incomes of less than $25,000 held only 3.5 percent. Shareowners with family income of at least $100,000 owned almost two thirds of outstanding shares. The data on the concentration of shareownership for 1998 are very similar to the data for 1995 and 1992.
Methods of Stockownership and Portfolio Size
Table 1 provided some information on the diverse pathways by which individuals hold stock. Table 9 presents additional information on the method of stock ownership for shareholders in different age groups. The first column shows the percentage of shareowners of different ages that own stock through self-directed retirement accounts such as IRAs or Keoghs. Thirty-two percent of shareholders of all ages own stock through this route. The probability that a shareowner owns shares through this channel peaks, at 37.7 percent, between the ages of 45 and 64. It is lower for currently-older shareholders than for those who are currently of working age in part because the former group was less likely to contribute to these accounts while working, and in part because older individuals may have already "cashed out" their retirement account balances.
The second column in Table 9 describes investors who own only a single stock. Just over thirteen percent of shareowners fall into this category. The probability that an individual shareowner owns only one stock are higher for young adults, and for the elderly, than for those in middle age.
The third column in Table 9 considers stock ownership through mutual funds. Just fewer than 32 percent of all shareholders (26.8 million individuals) own stock through mutual funds, and 14.7 million of these individuals own stock only through these funds. The probability of owning a mutual fund outside of a retirement plan is greatest (45.7 percent) for shareholders over the age of 65. There is no clear pattern in the probabilities of mutual fund ownership for shareholders at younger ages. The probability of owning shares through a mutual fund has increased since 1995, when the overall probability was 27.9 percent.
Finally, the last column of Table 9 shows the percent of shareowners who report that they have a brokerage account. Thirty-one percent of the population falls into this category. This is a substantial increase from 1995, when the comparable statistic was 26.5 percent. Older shareholders are substantially more likely than young shareholders to have such an account. The probability of having a brokerage account is 46 percent for those over the age of 65, but only 23 percent for those under the age of 35. Both values are higher than the comparable values, 38 percent and 19 percent respectively, in the 1995 Survey of Consumer Finances. Thus it appears that the growth of direct equity ownership during the 1995-1998 period coincided with an increase in the number of shareholders with brokerage accounts.
Table 10 presents information on the value of individual stock portfolios. (Recall that in measuring the value of an individual's portfolio, the value of total equity holdings for married couples is divided in half, with half assigned to each spouse.) The table shows the distribution of shareowners by the total amount of corporate stock that they hold. It reports the percentage of shareowners in each category, along with the percentage of shares held by stockholders in each category.
The findings in Table 10 suggest that slightly more than one stockowner in five has an equity portfolio worth less than $5,000. This fraction has declined since 1995, when the comparable statistic was 26.6 percent. One quarter of shareowners have equity portfolios worth at least $100,000, and 10.1 percent have more than $250,000 in equities.
The data in Table 10 illustrate the concentration of shareownership. Those with at least $250,000 in their equity portfolios hold 72 percent of the stock that the Survey of Consumer Finances attributes to individual investors. In a similar vein, the just under one third of shareowners who own less than $10,000 in stock account for less than one percent of outstanding stock holdings.
Mutual Fund Ownership
Equity mutual funds provide a mechanism for investors to achieve diversified portfolio holdings without managing investments in many individual securities. The data in Table 11 suggest that the growing investment in equity mutual funds has been an important factor in the expansion of shareholding in the last decade. The SCF provides enough information to shed some light on potential differences between the individuals who own equity mutual funds, and those who own stocks in other forms.
Table 11 presents information on the educational background of mutual fund investors, and Table 12 presents parallel information on the family income of these investors. Both tables contrast the fraction of adults in various education or income categories who own stock directly with the fraction that owns stock through an equity mutual fund. The tables also distinguish two forms of mutual fund ownership: direct ownership and indirect ownership through a self-directed retirement account.
Table 11 shows that the probability of owning a mutual fund directly is lower than the probability of owning stock directly for all education groups. It also shows a narrowing proportional difference between these probabilities as an individual's educational attainment increases. While less than ten percent of those with a high school diploma or less own equity mutual funds, more than a third of those with post-graduate degrees own at least one fund. When indirect ownership of mutual funds through retirement plans is combined with direct ownership, the sharp gradient in ownership probabilities by educational level remains. The probability of mutual fund ownership with this expanded definition exceeds the probability of direct ownership at all educational levels. For those with at least a college degree, there is roughly a two in three chance of either direct or indirect equity fund ownership.
Table 12 presents similar information with individuals stratified by family income level. The findings are broadly similar. The chance of direct equity ownership exceeds the chance of direct mutual fund ownership at all family income levels. The two probabilities are closest at middle income levels. For individuals who live in very high income families, the probability of owning stocks directly is between ten and twenty percentage points greater than the probability of directly owning mutual funds. For individuals from families with incomes of at least $50,0000, there is more than a two-thirds chance of either direct or indirect ownership of equity mutual funds.
Table 13 explores another dimension of mutual fund ownership: the number of funds that individuals who own mutual funds invest in. The table presents comparable data from 1989, 1992, 1995, and 1998. It shows an ongoing trend toward multiple fund ownership on the part of mutual fund investors. In 1998, twenty-three percent of those who owned mutual funds directly held only one fund, while 25 percent held two and 19 percent held more than six. Just over nine percent of those who owned mutual funds reported holding ten or more funds. To place the 1998 data into perspective, in 1992, 37 percent of those who held mutual funds held only one fund and less than ten percent held more than six funds. As recently as 1989, nearly half of fund investors held only one fund and less than five percent held more than six funds.
The importance of multiple mutual fund ownership, illustrated in Table 13, underscores the difficulty of using data on number of fund accounts to draw inferences about the time profile of the number of shareowners. The Investment Company Institute (2000) reports that in 1989, there were 20.8 million separate accounts at equity mutual funds in the United States. This number rose to 32.7 million in 1992, 69.4 million in 1995, and 119.8 million in 1998. The data in Table 1 show that total mutual fund ownership increased over this nine-year period, and the total number of shareholders increased as well. The growth in aggregate shareownership does not keep pace, however, with the six-fold increase in the number of equity mutual fund accounts over this period.
Risk Taking
The Survey of Consumer Finances asks several questions about how individuals think about the risks and rewards associated with portfolio investment. Since equities are risky assets, there is some interest in understanding what investors believe about the desirability of risky portfolio investments. Each household in the SCF is asked if they are willing to take substantial, above-average, or average levels of risk, or if they are unwilling to take risk. While it is difficult to precisely calibrate the responses to this question, and it is particularly difficult to understand the no risk response among those who hold stocks, it can provide some information on risk-taking.
Table 14 summarizes respondents' views on risk aversion. Almost half of shareholders report that they are willing to bear average levels of risk. Only seven percent report that they are willing to take substantial risk; this percentage is higher at younger ages. Twenty-nine percent respond that they are prepared to bear above average levels of risk. Eighteen percent of all shareholders, and 30 percent of shareholders over the age of 65, report that they are not willing to bear any risk. The responses regarding risk aversion are very similar to the responses in previous Surveys of Consumer Finances.
The Survey of Consumer Finances and Other Surveys of Stock Ownership
The estimates of the number of shareholders in the United States and the sketch of shareholder characteristics are broadly consistent with other surveys that provide data on asset ownership. Because the SCF targets higher income and higher net worth households that are more likely to own equities, it sometimes yields higher estimates of the value of assets. It also tends to suggest a greater number of asset holders than other surveys. In 1999, for example, the Panel Survey of Income Dynamics, a survey administered by the University of Michigan which oversamples low-income families, included a "wealth module" as part of the survey instrument. The results suggest a somewhat smaller number of shareholders than the SCF data (21.4 percent of families with direct stock holdings or direct mutual fund holdings) but this is probably due to sample differences.
The survey that most closely resembles the SCF from the standpoint of information collected on shareownership is a survey conducted by the Investment Company Institute and the Securities Industry Association in 1999. The findings from this survey were reported in the 1999 report Equity Ownership in America. Like the Survey of Consumer Finances, the ICI-SIA survey over-sampled high-income and high net worth households. The survey included a roughly similar number of respondents (5,235 for the ICI-SIA survey, including a group of 393 high net worth respondents, versus 4,309 respondents for the 1998 SCF, with 2,496 of these respondents from a high net worth sample). Many of the questions in the ICI-SIA survey were identical to questions in the Survey of Consumer Finances, so it is possible to make direct comparisons along some dimensions.
The ICI-SIA survey was conducted several months after the 1998 Survey of Consumer Finances. While the responses do not correspond to precisely the same period, they should nevertheless be broadly comparable. The characteristics of stockholders are unlikely to have changed substantially in the interim between the two surveys.
The results are quite consistent for the two sets of surveys. With respect to the total number of shareholders, the SCF-based estimate of 84.0 million is slightly higher than the ICI-SIA estimate of 78.7 million. This difference is too large to explain simply on the basis of sampling error in the two surveys. When a survey is used to estimate the total number of shareholders, there is always some imprecision in the resulting estimate. To gauge this imprecision, consider a survey that contacts 5,000 respondents, each of whom truly has a 50 percent chance of owning stock. Assume that the resulting estimate of the probability of owning stock is applied to an aggregate population of 160 million persons. In this case, there is a 95 percent chance that the true number of shareholders lies in a band centered on the resulting estimate, but including values up to 2.2 million shareholders higher, and 2.2 million shareholders lower, than the actual estimate. In the case of the 1998 SCF estimate, this suggests that a range between 81.8 million shareholders, and 86.2 million shareholders, has a 95 percent chance of including the true, but unobserved, number of actual shareholders.
The "95 percent confidence interval" for the SCF-based estimate does not include values as low as 78.7 million shareholders, which is the outcome of the ICI-SIA study. A similar calculation for the ICI-SIA estimate would show that its range of uncertainty does not include the SCF value. The difference between these estimates therefore probably reflects some differences in the set of respondents who answered the SCF and the ICI-SIA surveys.
Comparisons of the attributes of shareholders in the two surveys suggest that the two questionnaires may have reached slightly different populations. It is not clear what accounts for this difference. The median age of shareholders in the ICI-SIA survey, for example, was 47, contrasted with 44 in the 1998 SCF. Median family income for stockholders in the ICI-SIA survey was also slightly higher than in the SCF ($60,000 vs. $57,000), and median equity portfolio size was substantially greater ($50,000 compared with $28,000 in the SCF). The larger average portfolio size in the ICI-SIA report may reflect a different rule for dividing family equity holdings amongst family members in that survey and in the current SCF-based analysis. These summary statistics suggest that for some reason, the ICI-SIA survey reached fewer young households that own corporate stock than the Survey of Consumer Finances did.
While it is not clear how to evaluate differences between surveys, there is some presumption to favor the findings from the SCF. A number of studies by Federal Reserve Board researchers describe the range of internal consistency checks for data quality that are incorporated in the SCF. Moreover, these checks have been developed and refined over a period of many years. The SCF also provides respondents with advance warning, before the survey is administered, that they will be asked about their portfolio holdings. This provides respondents with an opportunity to prepare for the survey, and probably results in more precise estimates of asset values than a survey based only on telephone contact.
Despite the modest differences between the SCF findings and those from the ICI-SIA study, the broad consistency between the two studies is an important confirmation of the validity of both. Taken together the two studies suggest that the total number of shareholders in the U.S. economy is currently around 80 million. The cross-year comparisons with the SCF leave little doubt that the number of shareholders grew rapidly during the 1990s, and that a broad range of individuals are represented in the ranks of shareholders.
Aggregate Data on Individual and Institutional Ownership of Corporate Stock
Household survey data such as that from the Survey of Consumer Finances provides important information on the set of individuals who own stock, but it does not illuminate overall importance of individuals and institutions in equity markets. To investigate this issue, it is necessary to turn to aggregate information on the ownership of corporate stock, such as that published by the Federal Reserve Board in the Flow of Funds accounts. These accounts track the net purchases and sales of various financial assets by domestic financial institutions, by households, and by foreign investors. They include detailed information both on the level of asset holdings by different sectors, and on changes in these asset holdings over time.
Table 15 presents data from the Flow of Funds on the aggregate amount of corporate stock held by investors in different sectors at the beginning of each decade since 1950, and in 1998. The entries are end-of-year balances, so the 1998 entry is roughly coincident with the timing of the 1998 Survey of Consumer Finances. Table 16 presents similar information in percentage form. It shows the percentage of outstanding corporate equity that is held by households and other financial sectors.
Several features of the Flow of Funds data warrant special comment. First, the aggregate measure of corporate stock that these accounts track includes some closely held corporate stock as well as the equities traded on public stock exchanges. Second, the accounts are designed to measure the outstanding liabilities of U.S. corporations, so they exclude equity issued by foreign corporations, but traded on U.S. stock exchanges, from the aggregate total. Such equities would be counted in the equity portfolios of individuals responding to the Survey of Consumer Finances. Third, the "household" sector of the Flow of Funds accounts includes individuals as well as nonprofit institutions. The corporate stock holdings of these institutions, which are tabulated by the Federal Reserve Board for a subset of the years shown in Table 15, are roughly five percent of aggregate household sector holdings. Finally, the Flow of Funds accounts disaggregate stock ownership by several groups of financial institutions, such as mutual funds and private pension funds, for which individuals are the ultimate owners. To develop a clear measure of aggregate stock ownership by individuals, it is therefore necessary to combine the holdings in various sectors.
The data in Tables 15 and 16 show that there has been a significant and ongoing decline in the share of corporate stock that is directly owned by individuals in the United States. Between 1950 and 1980, this decline coincided with an increase in the share of institutional equity ownership, with a pronounced increase in ownership by defined benefit pension plans. Part of the apparent decline over this interval is a figment of the data collection process, however. The data for 1970 are the first in Tables 15 and 16 that divide separate household equity ownership through bank personal trusts and estates from other equity holding. This category accounts for 10.9 percent of outstanding corporate stock in 1970, which leads to a seeming decline in household direct equity ownership between 1960 and 1970. Without this artifact of the data, the decline in direct household equity ownership during the 1960s would be significantly less dramatic.
In the period since 1980,the fraction of shares held directly by households has declined, but it has been largely offset by an increase in indirect equity ownership. The growth of indirect ownership has included shareholding through equity mutual funds, and through defined benefit pension plans. In 1980, indirect household equity holdings accounted for 14.6 percent of outstanding corporate stock. At the end of 1998, they accounted for 27.5 percent of all outstanding shares, and for 41.1 percent of the corporate stock owned by the household sector.
The aggregate share of corporate stock in U.S. companies held by U.S. households has declined from 93.5 percent in 1950, to 84.5 percent in 1970, to 72 percent in 1990, and to 68.6 percent in 1998. The growth of stock ownership by investors from other nations, and in holdings by the defined benefit pension plans managed by state and local governments, have been central contributors in the recent decline.
The decline in direct stock ownership documented in Tables 15 and 16 may have implications for the operation of equity markets, even if the distinction between direct and indirect ownership is of no consequence for household balance sheets. There may be differences in trading patterns or other aspects of investor behavior that vary between institutional and individual investors. The shift toward a greater fraction of equities held by institutional investors may therefore affect the way equity markets react to new information or to other external shocks.
Recent and Future Trends in Shareownership
The foregoing data document a substantial expansion of the shareholder population in the last decade, but they do not provide any insight on the prospects for continued shareholder growth. Four factors, however, suggest that recent trends are likely to continue at least for the foreseeable future.
First, a key driver in the recent growth of shareholding has been the growth of self-directed retirement accounts. This trend appears to be continuing. Recent legislation has created a number of new saving vehicles, such as Roth IRAs, that permit investors to reduce the tax burden on their investment income while saving for retirement. There has been rapid growth in the utilization of employer-sponsored retirement saving plans, such as 401(k) plans. The U.S. Department of Labor (2000) reports that in 1996, the most recent year for which comprehensive data on 401(k) plans are available, there were 30.8 million active participants in these plans. Only five years earlier, the number was 19.2 million.
Industry analyses based on surveys of 401(k) plan providers suggest that there has been continued rapid growth in 401(k)-plan availability and participation in the years since 1996. There is also evidence that the fraction of plan participants who choose to invest at least some of their account balance in equities is increasing. This may be due in part to the very strong returns delivered by corporate stock portfolios during the 1990s. It is more difficult to predict the future fraction of 401(k) investors who will choose to hold equities.
One potential change in the structure of retirement saving arrangements that could affect the number of future stockholders is Social Security reform. If the current pay-as-you-go Social Security program is ever replaced or supplemented with a system of "individual accounts," and if these accounts permit or require individuals to hold some of their account balance in corporate stock, there could be a significant increase in stockownership.
Second, the growth of equity mutual funds during the 1990s has contributed to the expansion of the number of shareholders in the U.S. population. The most recent data from the Investment Company Institute suggest that the number of accounts at equity mutual funds is growing rapidly. Institutional changes within the mutual fund marketplace, such as the emergence of increasingly specialized funds that offer investors both underlying security portfolios and charge structures that are suited to particular preferences, suggest that there may be additional expansion in future years.
One influence that may reduce the growth of the mutual fund industry in future years is the rise of low-cost on-line brokerage firms that permit individual investors to trade stocks and to assemble equity portfolios at costs substantially below those of even a decade ago. Even if this set of market innovations slows the growth of the number of individual investors with mutual fund accounts, it may nevertheless result in continued growth in the total number of individuals who hold corporate stock.
Third, stock options have become increasingly common in compensation packages, not just for high-level executives but for rank-and-file workers as well. The National Center for Employee Ownership estimates that as many as ten million workers currently receive stock options as part of their pay, and these workers become stockholders when they exercise their options. The shifting industrial mix of the U.S. economy toward the technology-intensive sectors that have historically emphasized option-based worker pay packages is likely to further expand this source of individual share ownership.
A final factor that is likely to contribute to the growth in the number of shareholders is the aging of the U.S. population. Data from the Survey of Consumer Finances and other surveys show that the probability that an individual owns stock rises for most of his or her working life. As the "Baby Boom" generation moves closer to retirement, and as the members of this generation save in anticipation of retirement, there is likely to be continued interest in stock ownership as a means to financial security.
The historical record of the twentieth century suggests that investments in corporate stock have yielded higher returns, on average, than investments in fixed income securities or short-term notes. The very high stock market returns of the 1990s have undoubtedly made a significant group of individual investors more aware of this pattern, and this may have resulted in some shifts toward equity investment. One of the most difficult aspects of predicting future growth in shareownership is predicting how the population of individual investors will respond if equity returns do not continue at their recent pace.
. Non-U.S. Equity Holdings By U.S. Shareowners
Global financial markets have become increasingly integrated in the last two decades. The costs of moving financial capital across borders has declined, both as a result of changes in legal barriers to international investment and as a result of financial innovation that has reduced the transaction costs of global trading. For equity investors, investment specialists have long recognized the potential advantages to holding a diversified global portfolio. While there is no convincing evidence that the expected returns to equity investing are greater in one equity market than in another, there are well-documented diversification gains associated with investment in stocks from more than one nation. By holding a portfolio that consists of stocks from multiple nations, investors can reduce the variability of their returns without foregoing expected returns. Solnik (1996) summarizes the existing evidence on portfolio diversification in some detail. While some of the gains from holding shares in companies based outside the United States can be achieved by holding shares in diversified multinational firms based in the United States, this strategy does not seem to generate the full diversification effect of owning shares in non-U.S. firms.
This special report explores the prevalence of equity ownership in non-U.S. firms on the part of the U.S. shareowner population. It relies on data from the 1998 Survey of Consumer Finances, as well as information collected by the Investment Company Institute and the Securities Industry Association, to develop estimates of the number of individual investors who hold stock in non-U.S. firms. The central finding is that while relatively few investors hold non-U.S. stocks directly, a very substantial group of equity investors is exposed to the returns of non-U.S. firms through ownership of equity mutual funds that hold stock in these firms. Directly or indirectly, more than forty percent of U.S. shareholders have some equity investment in non-U.S. firms.
How Can Investors Hold Shares in Non-U.S. Firms?
There are three primary ways for U.S. shareowners to invest in the shares of firms that are headquartered in other nations. The first two, which are unfortunately impossible to distinguish in available data sets that track non-U.S. equity ownership by U.S. investors, involve direct ownership of shares, while the third involves indirect ownership.
The first, and probably the least common, is to purchase equity shares that are traded on stock exchanges outside the United States. U.S. investors can buy stock in companies that are traded in London, Tokyo, or Frankfurt, or on other exchanges. Most U.S. brokerage firms can carry out such transactions, although the commission costs would typically be higher than those associated with trading U.S. equities.
A second, and more straightforward, method of purchasing shares in non-U.S. firms is to purchase shares, either ordinaries or American Depositary Receipts (ADRs), in non-U.S. firms that are traded on the New York Stock Exchange or other U.S. markets. At the end of 1999, 406 firms from 49 nations had their stock listed on the NYSE. The global market capitalization of the non-U.S. listed firms was $5.6 trillion. While the non-U.S. listed firms on the NYSE represent only a limited fraction of the firms that investors could purchase with direct equity investments, they account for a nontrivial fraction of the total market capitalization of non-U.S. equity markets.
A third method of owning shares in non-U.S. firms involves indirect ownership through equity mutual funds that hold stock in firms based outside the United States. This is an increasingly popular means of international investment for U.S. shareholders. Shareowners can own shares in equity mutual funds that hold stock in non-U.S. firms either as taxable investments, or as tax-deferred investments in Individual Retirement Accounts, 401(k) plans, or other defined contribution accounts.
At the end of 1999, the Investment Company Institute (ICI) reports that there were 950 equity mutual funds that held stock in companies based outside the United States. This total includes global equity funds, which hold shares in both U.S. and non-U.S. companies, international equity funds, which hold shares exclusively in companies headquartered outside the United States, emerging markets funds, which hold shares based in some or all of the less developed nations, and regional funds, which hold shares in companies based in a particular part of the globe (such as the Pacific Rim or Europe). These 950 funds represented 24 percent of the number of equity mutual funds listed by ICI. Ten years earlier, in 1990, the ICI listed 155 funds that held stock in non-U.S. companies. This total equaled 14 percent of the total number of equity funds. The 1990s therefore witnessed a substantial expansion in the share of equity mutual funds that hold stock in companies based outside the United States.
Table 17 presents information on the total assets held in equity mutual funds that hold stock in non-U.S. companies. It distinguishes global, international, emerging markets, and regional funds. The table shows the assets in these funds, as well as the number of shareholder accounts and the net cash flows to these funds for the 1990-1999 period. The table shows that 14 percent of the assets in equity mutual funds at the end of 1999 were held in funds that held non-U.S. shares. In the early 1990s, the fraction of equity mutual fund assets invested in funds with some holdings of non-U.S. stocks nearly doubled. More recently, however, there has been a gradual recent decline in both the share of assets and the share of accounts at such funds. In 1996, the net cash inflow to equity funds that hold non-U.S. stocks was 22 percent of the net inflow to all equity funds. Two years earlier, it was 39 percent. In one year in the early 1990s, this flow accounted for as much as 53 percent of new money inflows to equity mutual funds. The increase in the assets of domestic equity funds relative to the assets of funds that hold stock in non-U.S. firms reflects a decline in new cash inflow to the latter group, as well as the strong performance of U.S. equity markets in the second half of the 1990s. These strong returns have increased the total assets in domestic equity funds.
Direct Shareholders of Non-U.S. Stock
The 1998 Survey of Consumer Finances, which was used in the earlier section of this study to describe patterns of stock ownership in the late 1990s, includes some information on direct ownership of stock in non-U.S. companies. It does not include any information on the extent of ownership of equity mutual funds that hold stock in non-U.S. firms. It is also uninformative on the issue of whether direct ownership of stock in non-U.S. firms takes the form of ordinaries or ADRs. Information from other sources suggests that ownership of equity mutual funds that hold stock in non-U.S. firms is widespread, so developing an accurate picture of ownership of non-U.S. equities by U.S. shareholders requires imputing ownership of funds that hold non-U.S. companies to those who hold mutual funds.
Table 18 presents information on direct ownership of non-U.S. stocks. The table shows the number of investors holding such stock, disaggregated by age groups. It also shows the percentage of all stockholders in each age category that own some non-U.S. stocks. ("Shareholders" are defined in this table as all individuals who own stock directly, through equity mutual funds held in a taxable account, or through equity holdings in tax-deferred accounts. This is analogous to the definition used in the last row of Table 1, above.)
Table 18 shows that in 1998, individuals of all ages had a 20.8 percent chance of owning any stock directly, and a 2.3 percent chance of owning a non-U.S. stock directly. This suggests that conditional on owning stock directly, an individual would have an 11 percent chance of owning stock in a company based outside the United States.
The probability that direct equity holders own stock in non-U.S. companies has trended upward over time. Table 18 shows that in the 1992 SCF, this probability was 8.3 percent, and that it rose to 8.6 percent in 1995 and to 10.9 percent in 1998. The probability that someone who owns stock directly will report some ownership of a non-U.S. stock is increasing in the shareholder's age, although the precise relationship between age and non-U.S. equity ownership seems unstable across the different survey years. In 1998, the chance that an over-65 direct shareowner would report ownership of a non-U.S. stock was roughly twice that of an under-35 direct shareholder. In 1995 and 1992, the probabilities that shareowners in these same cohorts (for example, those over the age of 62 in 1995, and under the age of 32) would report ownership of foreign stock were much closer.
Table 19 presents the information from the 1998 Survey of Consumer Finances in a somewhat different form. It tabulates the total number of shareowners who report direct ownership of non-U.S. stock: 3.7 million. This represents 4.4 percent of the 84 million individuals who are categorized as shareowners in the 1998 SCF. The table also shows the percentage breakdown of stockholders who own non-U.S. stock, by age. It shows that only 14 percent of those who own stock in non-U.S. firms are under the age of 35, while 62 percent are over the age of 45 and 26 percent are over the age of 65. The last column shows the probability of owning stock in non-U.S. firms for shareowners of different ages.
Table 18 and 19 show that most of the disparities across the younger-than-65 age groups in the total number of shareholders who own non-U.S. stock is due to differences in the likelihood of holding any stock directly. The chance of owning non-U.S. stock, conditional on owning any stock directly, is not very different across these age groups.
Ownership of Equity Mutual Funds That Hold Non-U.S. Shares
The total number of shareowners who own stock in non-U.S. firms is the sum of the number who own such stocks directly, and the number who own such stock indirectly through equity mutual funds. Unfortunately, the Survey of Consumer Finances does not ask detailed questions about the nature of the equity mutual funds that investors own. This means that to estimate the importance of indirect holdings of non-U.S. stocks, it is necessary to use data from other sources to supplement the Survey of Consumer Finances. Fortunately, there is another data source that provides roughly contemporaneous data on the structure of common stock and mutual fund holdings, and that includes detailed information on the type of mutual funds that investors own. This is a household survey conducted by the Investment Company Institute and the Securities Industry Association in conjunction with their 1999 report on Equity Ownership in America.
Table 20 presents summary information from the ICI-SIA survey on the probability that individuals who reported owning equity mutual funds report that they held at least one equity mutual fund that held stock in non-U.S. companies. The results suggest that slightly more than half of all mutual fund investors have at least some exposure to the equity of companies headquartered outside the United States through their mutual fund holdings. For shareowners who hold equity mutual funds in tax-deferred accounts, the conditional probability of holding an equity fund that holds non-U.S. shares, given ownership of an equity fund, is 59 percent. For those who hold equity mutual funds outside tax-deferred accounts, the analogous probability is 52 percent. This information can be combined with data from the 1998 SCF to estimate the overall likelihood of owning stock in companies based outside the United States.
Since the ICI-SIA data suggest that the conditional probability of non-U.S. stock ownership is much greater for individuals who own equity mutual funds than for those who directly own
O Clubeinvest.com informa que nenhuma da informação
aqui facultada deverá ser entendida como conselho ou recomendação
de qualquer tipo de transacção ou investimento.
