Do 70% of Families Lose Their Wealth by the Second Generation?

Updated August 7, 2026. Quick answer: the 70% is a real published number and it measures something else. The firm it is attributed to says that in about 70% of estate transfers families lose control of their assets and become divided — a statement about family unity and governance, not about wealth disappearing. And the “90% by the third generation” half appears nowhere on their site at all.

Where the number comes from

The claim is almost always attributed to research by the Williams Group, the firm behind the book Preparing Heirs. So we went and read what they actually say. Two sentences carry it, and we quote both in full because the wording is the entire issue.

From their About page:

“We have found that in approximately 70 percent of estate transfers, families lose control of their assets and become divided despite having excellent estate planning.”

From their Our Story page:

“Roy’s groundbreaking 20-year study of 2,500 families combined with an extensive research project including another 750 families with the Miami of Ohio School of Business set out to determine why only one-third of families succeeded in retaining their wealth into the next generation.”

Both read at thewilliamsgroup.org/about-us and /our-story on 7 August 2026.

There is no published study behind either sentence that we could find. No paper, no methodology, no description of how the families were selected, no definition of what counts as “retaining” wealth, and no stated period. Their research page returns a 404. That is not an accusation — a consulting firm is under no obligation to publish its methods. It does mean the number cannot be checked by anyone outside the firm, which matters a great deal once it is being repeated as a fact.

What that source actually measured

Read the first quote again slowly. What is being measured is that families “lose control of their assets and become divided” — governance, cohesion, who is in charge. That is not the same as the money being gone. A family can lose control of assets to trustees, to litigation, or to a fracture between siblings while the assets themselves remain entirely intact.

Now the part that settles it. We searched both pages for the rest of the circulating claim:

• “90 percent” — absent. It does not appear.
• “Second generation” — absent. It does not appear.

The second half of the claim is not in the source the claim is attributed to. The closest thing is “only one-third of families succeeded in retaining their wealth into the next generation” — one step, not two, and about a third of families succeeding rather than 70% failing by a specific generation.

The honest current figure

Where the numbers probably came from — offered as an observation, not a finding

There is a different, well-known pair of statistics about family businesses: roughly 30% survive into the second generation and about 12% into the third. The Williams Group’s own chief executive cites exactly those figures in a 2025 trade article — and attributes them to a different organisation, not to her own firm’s research.

Do the subtraction: 100 − 30 = 70 lost by the second generation. 100 − 12 = 88, which rounds to 90, by the third. The circulating claim is arithmetically identical to the family-business survival figures, inverted.

We are calling this an observation and not an attribution, deliberately. The arithmetic matching is suggestive and it is not proof. We did not trace the circulating wording back to that source, and we could not check the business-survival figures themselves — the organisation credited with them has a domain that no longer resolves. Noticing a pattern is not the same as establishing a chain, and this desk does not publish the first as though it were the second.

But the distinction underneath is solid and worth carrying away regardless of origin: whether a family business survives three generations and whether a family keeps its money are different questions. Most wealthy families do not own an operating business at all. Applying a business-continuity statistic to them is a category error even when the arithmetic is right.

What to cite instead

“The Williams Group states that in approximately 70% of estate transfers, families lose control of their assets and become divided – a finding about family governance and unity, not about wealth being dissipated. No published methodology accompanies it, and neither a 90% figure nor a second-generation framing appears in the firm’s own material.”
Source: Clear Money Guide, The Provenance File, 2026. Free to reuse with attribution under CC BY 4.0.

What this audit does not establish

This audit does not establish that families reliably keep their wealth. It establishes that a specific, very widely repeated sentence does not say what its cited source says. Those are different claims.

It also names no one as repeating the claim. Without a working search tool we could not reach a single third-party page carrying the exact wording, so we are not in a position to say who publishes it today. What is verified here is what the attributed source itself does and does not say, which we read directly.

And it is not a criticism of the underlying research. A twenty-year study of thousands of families may well be excellent work. The problem being described is what happens to a finding after it leaves the building — a claim about family unity becomes a claim about money, one generation becomes two, and a second figure attaches itself from somewhere else entirely.

Related

On what actually happens to money between generations: how many Americans have a will · what happens without one · what settling an estate actually costs.

This is an audit in the Provenance File. Our method, the six verdict classes and the desk laws are published at how we check money claims. If you think this is wrong, tell us — with a source. No advertising appears on this page and we earn nothing from it.