By Victor Haghani and James White¹
ESTIMATED READING TIME: 9 min.

The Economist recently published an article with the playful title, “How to Invest Your Enormous Inheritance,” in which it discussed the recent dramatic growth in intergenerational bequests. The article reports that “In 2025, people across the rich world will inherit some $6trn, or around 10% of GDP – a figure that has climbed sharply in recent decades.”²

With The Economist’s gracious permission, you can read the full article here.

It closes with a quote from us:

“People often want to know how much they need to have to give each member of their family’s next few generations a modest income,” says Mr Haghani. “The answer is: a lot more than most anyone thinks.”

A few friends have asked us to explain what we meant by that. In the U.S., parents can leave about $30 million to their children – and grandchildren and beyond – with no Federal inheritance tax. We know this is a very large amount of wealth,³ but as we’ll see, with some reasonable assumptions, it doesn’t necessarily alter the financial position of those inheriting it as dramatically as one might casually imagine. As with so many other cases of financial decision-making that we have written about, taking account of uncertainty in investment returns, family fecundity and longevity paints a starkly different picture than a static analysis based on a central case.

Let’s take the case of parents in their early 60s, with three children of average age 28. The parents put $30 million into an intergenerational trust for the benefit of their three kids and future generations. Their intention is to provide $50,000 per year of after-tax income, in constant standard-of-living terms, for all of the trust’s beneficiaries.⁴ When we talk about dollar amounts below, we’ll be implicitly assuming they’re adjusted to a constant standard of living.

We’ve chosen $50,000 because while it’s not an inconsequential sum, it also won’t move an heir from the 50th percentile to the 99th – or even 90th percentile of income. It helps, but it doesn’t put the recipient on a lifetime gravy train, either. It likely even falls short of what Warren Buffett would call “enough so your children can do anything, but not enough that they can do nothing.”

Let’s try to answer this question: starting with $30 million, how long is it expected to take until the trust can no longer provide the $50,000 to at least one branch of descendants?

We need a few more assumptions:

• First, the family assets are passed down generation to generation per stirpes,⁵ which means that the assets are evenly split each generation equally between the children of that generation. We think that is a reasonable way to handle intergenerational transfers in that it doesn’t reward heirs who have more children relative to those who have fewer.
• Second, we need some demographic assumptions. For life expectancy, we’ll use current actuarial mortality tables for the top wealth-quartile, in which a woman born today has a life expectancy of about 88 years. We’ll assume about 30 years between generations, and that the probabilities of a descendant having zero to five children are approximately 20%, 20%, 30%, 15%, 10%, and 5% respectively.⁶ Even though our assumptions imply a slightly shrinking population at large,⁷ the number of beneficiaries of the trust are expected to be growing over time. In fact, at a long enough horizon, most everyone in the world would be a beneficiary of the trust, as long as descendants mate with non-descendants without geographic or other constraints.
• Finally, we need to make assumptions about financial returns and risk. We’ll use our Capital Market Assumptions to arrive at an assumption of a 4% expected real after-tax return for the investment portfolio, with 12% risk.⁸ This results in a 2.5% expected after-tax return adjusted for a further assumption that the cost of living will grow about 1.5% faster than CPI.⁹

With these assumptions in hand, we can run simulations of family tree growth and financial outcomes. One surprising finding is that with the 2.5% expected after-tax cost-of-living adjusted portfolio growth, there’s a greater than 50% chance that the trust would not be able to cover at least one branch of the descendant’s $50,000 per year during the lifetime of the first generation. Longer term, there’s an 80% chance that the $50,000 constant-standard-of-living annuity will run out on some branch while the second generation is still alive.

These results are surprising because they are so different from what you’d get from doing the analysis without incorporating uncertainty arising from the variability in the market value of trust’s assets, the natural dispersion in the size of family tree branches, and the uncertain longevity of individual descendants. If we remove all sources of uncertainty in our analysis by assuming the trust portfolio’s return is 4% risk-free, that each person in the tree will have exactly two children,¹⁰ and that they’ll have their children at age 34 and live exactly to age 88, we’d find that the $50,000 payout could be maintained for over 200 years. It’s the uncertainty in future outcomes that creates the high probability that the trust won’t be able to pay $50,000 to some branch of descendants.

In the charts below, we show the likelihood over time of at least one descendant not receiving the full $50,000, along with the median trust assets available to each branch, and the average number of living descendants.

Calculator

Of course, our base case assumptions are not appropriate for most families, and were meant just as an illustration. Please use our family tree financial calculator to explore other assumptions and family circumstances.

Concluding Thoughts

There are other ways of structuring payouts among descendants which will give different, and in many cases, more desirable, results. In particular, as we have discussed before, spending policies which adjust to evolving portfolio values are generally more optimal than tying fixed spending policies to risk-bearing portfolios. However, we thought the case we have explored in this note would give a sense for the limits on how far bequests can go under reasonable investment and family growth assumptions, and also would explain why we told The Economist that it takes a lot more savings than most anyone thinks to provide a modest income to the next few generations. We look forward to hearing your thoughts, and stay tuned for more research on this topic from us.

1. This is not an offer or solicitation to invest, nor are we tax experts and nothing herein should be construed as tax advice. Past returns are not indicative of future performance.
   Thanks to Larry Hilibrand, John Karubian, Arjun Krishnamachar, Aron Landy, Steve Mobbs and Jeff Rosenbluth for their helpful comments, and to our colleagues Jerry Bell and Steven Schneider for their contributions.
   
2. See also “A $105 Trillion Inheritance Windfall Is On the Way for US Heirs”, by Neligh, Cobo, and Tartar, Bloomberg, December 5, 2024.
   
3. For a discussion, albeit somewhat one-sided, of the societal costs and benefits of having such a large exclusion, you may enjoy hearing Milton Friedman discuss the topic in 1978 here.
   
4. Regardless of age, so payments would still be made to young children for their future benefit.
   
5. “Through branches” in Latin.
   
6. About one-third of women have no children, but we have adjusted that probability down for adoptions and also the fact that about 40% of families in the US are “step-couples” which adds to the probabilities of family size derived from data on number of children per mother, and proportionately more at the higher end of family size. Our assumptions are roughly consistent with current US fertility rates.
   
7. In most developed countries, the replacement fertility rate is about 2.1 children per woman. Although it might seem that each woman only needs to have two children (one to replace herself and one to replace her partner), the actual replacement rate is slightly higher than 2, primarily because not all children survive to reach reproductive age.
   
8. This return assumption is a about 0.75% below the median response from our survey where we asked people what is the lowest after-tax, risk-free return above CPI they would accept to forego all other investing for the rest of their lives. We felt the answers to that survey were just a bit too aspirational for use here.
   
9. See Chapter 15, page 236 in our book, The Missing Billionaires: A Guide to Better Financial Decisions where we explain:
   “Over long periods, it becomes clear that if your spending just kept up with CPI, you would probably feel a dramatic decrease in your welfare. For example, if an adult spending the average amount in the United States in 1900 kept spending the same amount adjusted for CPI each year, then in 2022 that (very old) person would be spending just $5,000 per year in 2022 dollars. She would undoubtedly conclude that her standard of living had declined over this long period, given average US per‐­capita spending is more than 10 times higher, at $55,000.”
   
10. Including generation zero, which we assume has two children, not three.