The wealth owned by individuals in the United States currently stands at about $33 trillion.  It may come as a surprise that most of this wealth is not, as one might suspect, held in portfolios of stocks, bonds,
mutual funds, options, and other financial instruments.  

Paul Schervish          

John J. Havens and Paul G. Schervish
October 19, 1999

The following overview presents the first new wealth transfer estimates to be generated since 1990, the assumptions used to generate these estimates, and a glimpse at the implications of this wealth transfer for charitable giving.  In a forthcoming full report with the same title, we will present a more detailed portrait of the pending wealth transfer, the methodology we used to obtain this estimate, and the implications for the level of charitable giving, especially by wealth holders, over the first half of the next century.  We will also suggest strategies by which fundraisers and financial advisors can guide wealth holders to shift even greater portions of their net worth to charity, in the form of both inter-vivos gifts and bequests.  These new forecasts were developed by the authors at the Boston College Social Welfare Research Institute with support from the T. B. Murphy Foundation Charitable Trust (Michigan) and the Lilly Endowment, Inc. 

On the basis of a recently developed Wealth Transfer Microsimulation Model (WTMM), we estimate that the forthcoming transfer of wealth will be many times higher than the almost universally cited 55-year figure of $10 trillion.  Our low-range best estimate is that over the 55-year period from 1998 to 2052 the wealth transfer will be $41 trillion, and may well reach double or triple that amount.  Depending upon the assumptions we introduce into the model (for instance, in regard to the current level of wealth, real growth in wealth, and savings rates) we estimate the wealth transfer will range from a lower level figure of $41 trillion to an upper level figure of $136 trillion.  These estimates are not back-of-the-envelope projections.  They emerge from what to our knowledge is a first-of-its-kind microsimulation model of wealth accumulation and transfer.  The new estimates update the figure published in 1990 by Robert Avery and Michael Rendall[1]--and regularly cited ever since then--indicating that over the 55-year period from 1990 to 2044, the value of estates in the United States passing from adults with children 50 years and older would be $10.4 trillion. 

Until our estimates circulate and others have a chance to review and criticize our work, we suggest that the focus be placed on the low-end estimate of $41 trillion.  It is not because we believe our middle level estimate of $73 trillion to be unreasonably high, or our upper level estimate of $136 to be implausible.  For instance, the $73 trillion estimate assumes a maximum real growth rate of 3% for the next 55 years, and assumes that the value of assets held by individuals in 1998 was $32 trillion, which is 18% lower than the $39 trillion figure recently cited in Worth magazine (September, 1999, p. 97) and lower than a recent Federal Reserve estimate of private wealth of $37.4 trillion.  However, because no simulation estimates are any better than the assumptions on which they are based, and because we anticipate refining our estimates in the light of suggestions and criticism by others, we suggest that for the time being most credence be given to the $41 trillion estimate.  Emphasizing the $41 trillion lower-level estimate with its 2% secular growth rate helps protect against potential charges of “irrational exuberance” arising from our not yet having modelled periodic recessions, a world economic crisis, or depression. 

“The $10 Trillion Transfer”

Avery and Rendall’s original figure has become known in academic, professional, and media circles as the “forthcoming $10 trillion intergenerational transfer of wealth” or, more simply, as “the $10 trillion transfer.”  Although Avery and Rendall did not calculate this figure primarily for the purpose of estimating the amount of wealth transfer, over the past decade their estimate has become the most frequently cited statistic in discussions about charitable giving.  Avery and Rendall were originally interested in saving behavior and sought to learn whether the size of an expected inheritance led to a tendency by heirs to save less.  As such, Avery and Rendall calculated the inter-generational wealth only for families whose head was age 50 or older and had living children.  From time to time, commentators have adjusted the $10.4 trillion figure to reflect changes in the cost of living, producing wealth transfer estimates in the range of $12-$13 trillion.  Sometimes, when writers or reporters informally calculate the amount of the pending wealth transfer that will come from the very wealthy, they come up with a transfer estimate somewhere around $7 trillion.

We want to be clear that we are not criticizing any aspect of Avery and Rendall’s estimate or research, other than the fact that they have been misunderstood and that their estimate has ended up being used for purposes for which it was not designed.  Many who make reference to the $10 trillion inter-generational transfer not only fail to realize that the original estimate was a by-product of other research on prospective savings rates by heirs, and limited to a specific demographic group.  They also regularly overlook the fact that the $10 trillion wealth transfer is an amount projected to occur over 55 years, and not over a ten- to twenty-year period.  Equally important, most of those citing the $10 trillion figure mistakenly assume that the $10 trillion is fully discretionary, thereby encouraging the erroneous expectation that that the full $10 trillion is to be divided only between heirs and philanthropy.  Overlooked is the fact that a large part of the transfer will go to estate taxes and, to a much lesser extent, fees.  It is important to be aware of these misconceptions when, for instance, we indicate that our lower level 55-year estimate of $41 trillion will entail a $6 trillion transfer to charity.  The figure of $6 trillion, which is approximately 15% of the total $41 trillion, should not be contrasted to the current figure of $10 trillion but to a figure of about $1.5 trillion that would be bound for bequests if in fact the forthcoming transfer were $10 trillion instead of $41 trillion.  Similarly, when we speak of a twenty-year lower level wealth estimate of a $12 trillion total wealth transfer and a $1.7 trillion transfer via charitable bequests, these figures should be compared to what the 55-year $10 trillion dollar transfer would look like over 20 years, namely, a $2.8 trillion transfer of wealth and about $0.4 trillion in charitable bequests.

The $41 Trillion Transfer of Wealth

Our microsimulation projections are detailed in Tables 1 - 3.  All dollar figures in these tables represent millions of 1998 dollars.  The tables follow the same format.  The first panel in each table presents the results of the simulation for the 20-year period from 1998 through 2017.  The second panel in each table presents the results of the simulation for the 55-year period from 1998 through 2052.  The columns in each table categorize estates by the value of the final estate in 1998 dollars at the time of death of the final decedent.  The rows of each panel present the estimated number of final estates, the aggregate value of these estates, the aggregate burial and estate fees for these estates, the aggregate federal and state estate taxes levied against these estates, the aggregate estimate of bequests to charities from these estates, and the aggregate bequests to heirs (other than spouse) from these estates.

As we detail in a subsequent section of this overview, the low, middle, and high level scenarios differ mostly in the combined rates of change used to simulate modifications in household wealth.  The lower level scenario assumes a sustained 2% rate of real growth in wealth in combination with low lifecycle savings and high lifecycle dissavings rates.  The middle level scenario sets up a baseline case of a sustained 3% rate of real growth in wealth in combination with lifecycle variations in the rates of change in wealth.  The upper level simulation model assumes a sustained 4% rate of real growth in wealth in combination with  high lifecycle savings and low lifecycle dissavings rates.

All three scenarios reproduce the underlying distributional dynamics currently occurring in estates.  The larger the estate, the greater the proportion that goes to charity and the smaller the proportion that goes to heirs.  For final estates valued at $20 million or more, approximately 39% in each scenario will go to charity, 23% to heirs, 34% to taxes, and 3% for fees and burial costs.  For final estates valued at $1 to $4.9 million the current and simulated figures are approximately 8% to charity, 66% to heirs, 22% to taxes, and 4% for fees and burial costs.  In all scenarios, the simulated growth rates operate both directly on the value of wealth accumulated and indirectly through influencing the value of simulated inheritances received during the period.)

Download "Millionaires and the Millennium: New Estimates of the Forthcoming Wealth Transfer and the Prospects for a Golden Age of Philanthropy" in its entirety.