BEQUESTS AND INHERITANCES

Approximately 2.4 million adults died in the United States in 1999, leaving estates valued at over $196 billion to their families, charities, federal and state governments (via estate taxes), and others. Easily the largest share of estates goes to spouses. Children and charities also receive sizable amounts. Wealth transfers of this magnitude can have significant effects, yet researchers are still working to understand why people leave bequests. Several theories exist, but none fully explains the patterns of bequest distribution observed. The existence of estate taxes likely influences financial choices later in life, and perhaps the distribution of bequests as well. In 1999 the U.S. government collected $22.9 billion in estate and gift taxes—1.25 percent of all revenues collected. State governments collected over $6.1 billion in additional taxes.

Distribution of estates

One problem facing researchers is a lack of detailed information on why people accumulate wealth and how they distribute it at death. This is hardly surprising, given that most families consider death and the ensuing distribution of the decedent's possessions to be a fairly private matter. Since 1916, the federal government has taxed estates. The documentation of these estates that accompanies estate tax returns provides some insights into how people distribute their wealth at death. In 1999, only estates in excess of $650,000—just over 4 percent of all decedents— were required to file tax returns. Since these data focus on higher-income individuals they may not be representative of all decedents, but they do include the vast majority of wealth held by older individuals.

First, how do elderly individuals keep their wealth? One way to answer this is to look at wealth holdings at death. For estate tax returns filed in 1999, $79.4 billion of the wealth was in stocks—by far the largest category at 40 percent of wealth holdings. The next largest category was bonds, which include state and local bonds ($20.2 billion), federal savings bonds ($0.9 billion), other federal bonds ($6.0 billion), corporate and foreign bonds ($2.5 billion), and bond funds ($0.8 billion). Over 11 percent ($22.5 billion) of decedent wealth was held in cash or in cash management accounts. Another 10.7 percent ($21.0 billion) was held in real estate other than personal residences.

Table 1 shows how people distribute their wealth at death. The largest share, almost $55.7 billion, went to the surviving spouse. This is even more significant in light of the fact that only 42 percent of these decedents were survived by a spouse. Nearly 12 percent of decedent wealth went to the federal government for estate taxes, and another $6.1 billion was paid to state governments. Over $14.5 billion went to charities, $6.5 billion went to pay off existing debts and mortgages, and $4.5 billion went to pay funeral and other estate expenses. Bequests to children and others were estimated from 1989 data. The "Other" category includes many trusts that could not be connected to individuals. The data also suggests that daughters tend to receive slightly more in bequests than do sons.

Table 2 shows the average bequest amounts given by 1989 decedents, with the recipients divided by prebequest adjusted gross income (AGI). Clearly higher-income individuals receive significantly larger bequests on average than do lower-income individuals. Comparing the left-hand side of Table 2 with the righthand side indicates that unmarried decedents give more, and larger, bequests to relatives than do decedents survived by a spouse. This makes sense because spousal bequests are untaxed. Most likely the first member of a couple to die leaves most of his or her wealth to the spouse. This ensures that the surviving spouse has sufficient funds to live and allows the survivor to collect additional information about potential heirs before the estate is finally disbursed.

Reasons for leaving bequests

For years researchers have sought to understand why people leave bequests. The sheer quantity of wealth involved makes this an important question. Several broad categories of possible reasons for bequests exist, but none successfully explains all observed patterns. The most likely explanations of intergenerational bequests are intergenerational altruism and some form of exchange. Altruism or some type of "joy of giving" probably motivates bequests to charitable organizations. Most likely different people experience different bequest motives. This section describes them.

First one must wonder whether bequests to descendents are intentional or accidental; that is, simply the result of life ending before spending all the money. In a 1991 study, B. Douglas Bernheim examined the types of wealth held by retirement-age individuals. He concluded that the evidence strongly suggests bequests are indeed intentional. If people were interested only in ensuring that they had enough money for their own expenses, one would expect more people to purchase annuities. Annuities provide income for the rest of a person's lifetime, regardless of how long he or she may live. Their purchase relieves the concerns many may have about running out of money, but generally they cannot be passed to others at death. (Couples can purchase annuities that provide income until both of them die, and some annuities can be passed to nonspouses at death, but generally the benefits are substantially reduced.) Instead, elderly people hold their wealth in stocks, bonds, and other bequeathable forms. This suggests that people intend to leave their wealth to others after they die, and are not simply making sure their money lasts as long as they do.

Intergenerational altruism. Children, grandchildren, and other different-generation relatives receive more total bequests than do charitable organizations. A leading theory of intergenerational bequests suggests that people give such bequests because they are altruistic; that is, they care about the well-being of their children and grandchildren. An altruistic parent experiences a benefit from his or her children's happiness. One way a parent can influence a child's happiness is to give some of his or her wealth to the child. In a 1988 survey of older individuals, John Laitner and F. Thomas Juster (1996) found that 45 percent of respondents with children considered leaving an estate or inheritance to be very or quite important. During life, a parent may be reluctant to part with much of his or her wealth because of uncertainty about how long he or she will live and future financial needs. At death, the parent no longer has these concerns, and his or her wealth can be freely distributed to descendents.

To test this theory, researchers first assume that parents feel equally altruistic about all of their children. If this is true, then it would follow that parents distribute bequests in a manner that equalizes their children's after-bequest income and wealth. This theory is tested by examining how bequests are distributed among children to see if poorer children receive larger bequests and wealthier children receive smaller bequests. The data show that roughly two-thirds of all decedents distribute bequests almost equally to all children. Those who do not distribute bequests equally do indeed show some tendency to give larger bequests to poorer children, but the tendency is not as strong as expected.

Exchange-motivated bequests. The failure to confirm intergenerational altruism as a major bequest motive leads researchers to suspect that many bequests are motivated by some form of intergenerational exchange. B. Douglas Bernheim, Andrei Shleifer, and Lawrence Summers (1985) studied a large number of older people and the types and frequency of contact they had with their children. They also examined the types of wealth held by survey participants. The types of contact studied were phone calls and visits by the children to their parents. The researchers found that the more bequeathable wealth parents had, the more children contacted their parents. The hypothesis here is that the parents' wealth induces the children to make more frequent contact. Children may even compete for larger shares of the estate. In families with only one child, the child tended to contact the parents less frequently than did children in larger families. With no one to compete against, only children had less financial incentive to contact their parents frequently.

Other researchers have sought evidence that bequests are part of an intergenerational exchange. Some have examined bequest data for signs that parents give larger bequests to richer children, the theory being that richer children are more likely to have the financial resources required for frequent long-distance calls and visits. Unfortunately, these tests also fail to conclusively explain why people leave bequests to particular beneficiaries.

Developing accurate tests of the reasons people leave bequests is extremely difficult. One problem facing researchers is a lack of quality data. Estate tax data provide some insights into the bequests of wealthy individuals, but lack information on 96 percent of the population. They contain no data on child-to-parent contact and lack direct information on children's wealth. Other data sets contain information about wealth holdings of a broader set of parents and children, but since they survey living individuals, they have no information on bequests. Also, the initial assumption that parents care equally about all of their children is unlikely to be true.

The bottom line here is that one may never know the exact cause(s) of intergenerational bequests. Most likely each of these motives operates to different degrees for different individuals. Some are altruistic, and others are not; some reward children who call and visit, and others do not. The future challenge may be to further understand how much of the population acts from each of these motives.

Charitable bequests. Charitable bequests comprise roughly 14 percent of nonspousal bequests. These bequests are not divided evenly among different types of charities. Of the $10 billion left to charitable organizations in 1995, $3.2 billion (31.6 percent) went to educational, medical, or scientific organizations, and $3.1 billion went to private foundations. Only $970 million went to religious organizations and $273 million went to the arts and humanities. The remainder went to charities in a number of smaller categories.

Charitable bequests are most often the result of a sincere desire to help others. Some people are motivated to replace reductions in government-funded social services. Others may feel they have worked hard for their money and would rather put it to good use than have their children spend it. Occasionally, decedents seek to perpetuate the family name with a building or establishment of a scholarship, but these instances are clearly in the minority. Also, the structure of the estate tax (see below) clearly encourages charitable bequests.

Estate taxes

Enacted in 1916, the federal estate tax has gone through many changes over the years. Marginal tax rates have ranged from 1 percent to as high as 77 percent. There have been limits on how much can be transferred to spouses and others, and there have been changes in the ways estates are valued. This section briefly describes the major components of the federal estate tax as it existed through most of the twentieth century. Note that the Economic Growth and Tax Relief Reconciliation Act of 2001 includes gradual reductions in the estate tax, with complete elimination in 2010.

Table 3 shows the amount of tax collected from estates of different sizes. The progressive nature of the estate tax is quite evident here. Nearly half of the estates had a value of less than $1 million, and these estates paid 3.5 percent of the tax due that year. At the other extreme, a small portion (approximately 0.5 percent) of estates had a value above $20 million, and paid 23.8 percent of the total tax.

The federal estate tax has several objectives, the foremost being to raise revenue for the federal government. While the portion of federal revenues derived from the estate tax has generally been fairly small (typically less than 2 percent), it did exceed 5 percent in several Depression era years and approached 10 percent in 1936. Opponents of the estate tax have argued that this small contribution to federal revenues is not sufficient to prevent its elimination.

A second objective of the estate tax is to complement the income tax system to ensure all income is taxed. Much wealth is held in capital gains, which are not considered income until realized (i.e., when the underlying asset is sold). When a person holding an asset that has appreciated dies, no income taxes are paid on the appreciation. When the asset is transferred to another individual, the basis on which capital gains taxes will have to be paid becomes the then current value of the asset—usually much higher than the price paid by the original owner. The new owner will have to pay capital gains taxes only on the increase in value that occurs after he or she receives the asset. By collecting estate taxes on the value of the transferred asset, the estate tax complements the income tax and ensures that all capital gains are taxed. Without the estate tax, owners of capital assets would have to keep track of the original purchase price, regardless of who made the purchase and how long ago the purchase occurred.

A third objective is to reduce concentrations of wealth. Some view large inheritances as contrary to a major value of American society: equal opportunity for all. They see estate taxation as a way of "leveling the playing field." The values in Table 2 suggest that estate taxes do little to reduce familial wealth concentrations. However, it is possible that the difference between bequests received by lower- and higher-income individuals would be even greater in the absence of the estate tax.

A fourth objective is to reduce competition between states for wealthy individuals. Some states could seek to encourage wealthy people to move there by advertising their low (or nonexistent) estate tax rates. The federal estate tax ensures that decedents' estates will be taxed similarly regardless of their state of residence. While some differences between state estate tax laws exist, almost all of them are smoothed out by the structure of the federal estate tax. (See "Tax Credits," below.)

Determining estate taxes. To figure the tax due on an estate, one must first calculate the value of the estate. This includes the value of real estate holdings, stocks, bonds, pensions, businesses, cash, and proceeds from life insurance policies owned by the decedent. Often decedents have transferred assets to others prior to death. Some assets must be included in the calculation of a decedent's estate if they were transferred to others within three years preceding death.

Assets can be valued at the date of death or six months after death. This allowance recognizes that it can take a considerable amount of time to settle an estate and the value of the estate may decrease in that time. There is also a method of valuing property used in a closely held farm or business. Such property sometimes would have a greater value under a different use than under its current use. For example, a five-hundred-acre farm may have a certain value as farmland, but a much higher market value if sold for residential property. Rather than forcing the estate to consider the value of the land as residential, it can be valued as farmland. Certain restrictions apply to using this "special use" valuation, and there is a maximum as to how much of the market value can be excluded from the estate. In 1995, 456 estates used the "special use" valuation, for a reduction of $171 million in gross estate value.

Once the gross estate value has been determined, a number of deductions are allowed. The largest of these is bequests to the spouse. Initially this deduction was limited to a fraction of the gross estate, but this limit was removed in 1982. Recognition that estate taxes would be paid at the death of the surviving spouse made limiting the spousal deduction unnecessary. Other allowable deductions include charitable bequests, debts of the decedent (e.g., mortgages and medical debts), and expenses incurred by the estate (e.g., funeral expenses and attorney's fees). Table 1 gives 1999 values for these deductions.

Tax credits. Initially, the first $50,000 of an estate was exempt from taxation—similar to the standard deduction allowed for personal income taxes. The exemption amount was increased periodically (and decreased in 1932) until 1977, when it was repealed and replaced with a credit. This credit, called the "unified credit," is by far the largest credit provided for in the estate tax, reducing estate tax liability by roughly $18.5 billion in 1999. Since this credit eliminates all estate tax liability for most estates, it is often more convenient to speak in terms of the amount of an estate it effectively exempts from taxation. For example, in 1999 the unified credit amount was $211,300. An estate valued at exactly $650,000 (after deductions) would have had an initial liability tax of exactly $211,300. Applying the unified credit (i.e., reducing tax liability by the amount of the credit) leaves a final liability of $0. The initial tax liability on estates smaller than $650,000 is less than $211,000, so the unified credit eliminates all tax liability for these estates. For larger estates the unified credit effectively eliminates the tax that would be due on the first $650,000 of the estate. Under the Economic Growth and Tax Relief Reconciliation Act of 2001, the effective exemption rises to $3.5 million in 2009. This credit gets its name from the fact that it unifies the exemptions for the estate, gift, and generation-skipping transfers taxes.

The state death tax credit is the other major credit allowed under the estate tax. This credit provides a one dollar reduction in federal estate taxes due for each dollar of state estate taxes paid. The credit is capped at an amount that varies with the size of the taxable estate. Most states simply charge estate taxes equal to the maximum allowable federal credit amount. In fact, there is little incentive for states to do otherwise. States that charge a smaller amount do not save their constituents any money; they only allow them a smaller credit against their federal estate taxes— in effect letting the federal government collect tax revenues the state could have collected. States that charge a larger amount are providing a disincentive for wealthy people to live in their state. In 1999, the state death tax credit reduced federal estate taxes by approximately $6.0 billion.

Behaviorial effects

The estate tax can influence the behavior of wealthy older individuals, even if it simply causes them to consult a professional estate planner in order to minimize estate tax liability. Other behavioral effects may include increased work effort and savings, increased charitable bequests, and decreased capital gains realizations.

First, since estate taxes directly reduce the amount of a parent's wealth actually received by an heir, they can lead to increased work effort and savings. A parent intending an heir to receive a specific dollar amount would have to save more (and perhaps work more) than if there were no estate taxes. Other parents could decide estate taxes are so high that it is impractical to leave any sizable bequests. This could lead to decreased savings and work effort among older individuals. It is extremely difficult to determine how significant either of these behavioral effects may be in practice.

Second, by exempting charitable bequests from taxation, the estate tax directly encourages such bequests. A simple example will illustrate why this is so. Suppose Susan is deciding how to allocate her estate, knowing it will face a marginal tax rate of 55 percent. She has allocated all of her estate except for $10,000, which could be given either to charity or to a cousin. If it is given to the cousin, the government will collect its share, leaving the cousin with only $4,500. If it is given to a charity, the charity will receive the entire $10,000. Many people choose the charity when confronted with this actual choice of giving $4,500 to an heir or $10,000 to charity. Several studies of this issue produced similar results. For example, Gerald Auten and David Joulfaian (1996) concluded that charitable gifts could be as much as 12 percent lower in a world without estate taxes.

Third, the fact that the appreciated value of capital assets (e.g., stocks and bonds) in an estate is not subject to capital gains or income taxes encourages older individuals not to sell such assets. The wealth associated with these assets will be subjected to the estate tax whether or not the asset is sold before death. But if an individual sells an asset before death, he or she will be liable for capital gains taxes on the appreciated value. The individual will then have less to bequeath. In contrast, an individual who does not sell an asset before death can bequeath the entire value of the asset. Economists continue to debate whether or not estate taxes actually encourage older individuals to hold capital assets longer than they should.

Finally, the behavioral effects of bequests themselves should be considered. Studies show that the larger the inheritance received by a person, the more likely he or she is to reduce the work effort, and perhaps even to retire. The effect, however, is relatively small. Joulfaian (1998) reported that among 1995 beneficiaries, only 9 percent of unmarried bequest recipients had quit work three years later. Of couples with both spouses working when they received an inheritance, 98.5 percent still had at least one spouse working three years later. This result likely understates the effect of bequests because it cannot indicate whether retirement was accelerated but did not occur within the three-year study period.

Robert P. Rebelein

See also Assets and Wealth; Estate Planning; Financial Planning; Intergenerational Exchanges; Pension; Taxation.

BIBLIOGRAPHY

Auten, G., and Joulfaian, D. "Charitable Contributions and Intergenerational Transfers." Journal of Public Economics 59 (1996): 55–68.

Bernheim, B. D. "How Strong Are Bequest Motives? Evidence Based on Estimates of the Demand for Life Insurance and Annuities." Journal of Political Economy 99 (1991): 899–927.

Bernheim, B. D.; Shleifer, A.; and Summers, L. H. "The Strategic Bequest Motive." Journal of Political Economy 93 (1985): 1045–1076.

Johnson, B. W., and Eller, M. B. "Federal Taxation of Inheritance and Wealth Transfers" In Inheritance and Wealth in America. Edited by R. K. Miller and S. J. McNamee. New York: Plenum Press, 1998. Pages 61–90.

Johnson, B. W., and Mikow, J. M. "Federal Estate Tax Returns, 1995–1997." In Statistics of Income Bulletin. Washington D.C.: Statistics of Income Division, Internal Revenue Service, 1999. Pages 69–130.

JOULFAIAN, D. The Federal Estate and Gift Tax: Description, Profile of Taxpayers, and Economic Consequences. Office of Tax Analysis Working Paper no. 80. Washington D.C.: U.S. Department of the Treasury, 1998.

Joulfaian, D., and Wilhelm, M. O. "Inheritance and Labor Supply." Journal of Human Resources 43 (1994): 1205–1234.

Laitner, J., and Juster, T. F. "New Evidence on Altruism: A Study of TIAA-CREF Retirees." American Economic Review 86 (1996): 893–908.

SEATER, J. J. "Ricardian Equivalence." Journal of Economic Literature 31 (1993): 142–190.