What, if anything, can we learn from banks’ family office reports?
UBS is one of the many banks which put together reports providing insight into the family office sector, but what can we learn from these reports and do they all back each other up? (Reuters/Denis Balibouse)
Good morning. Family offices are becoming an increasingly common way for the very wealthy to manage their money.
But of course many of them are very private, certainly much more so than pension funds or other asset owners.
Many of the largest and most professional ones do publish data on their investment activity but this leads to the question of how we know what on earth family offices are up to.
Into this gap in the market have stepped several large banks, which publish annual family office reports claiming to offer insight into what this sector looks like.
Let’s take a look at how these banks claim family offices allocate their assets:
The data here all relates to 2025 because not every bank has published its 2026 report yet.
The other caveat is that RBC’s data relates only to North American family offices.
Now the first thing which jumps out is that the findings these banks provide have almost nothing in common.
The level of exposure to public equities falls in a range between 41 per cent and 15 per cent.
What’s going on here?
Someone who works in the family office sector had this to say: “The family office universe is private and heterogeneous - there’s no central reporting and definitions vary.
“Coverage often differs by banking institution - each bank sees only the subset of clients it services or interacts with depending on core markets/geographies, which can skew the sample.
“Methodologies and objectives will also vary - some reports are based on client data, others on surveys or market estimates, often produced with different commercial lenses.
“As a result, you tend to have dispersion in both size estimates and behavioural insights across published reports.”
Indeed there is little consistency within these reports year to year, when we compare 2024 to 2025 and pick two asset classes entirely at random (I couldn’t find JPMorgan’s historic reports so they’re not included).
What have we learnt from this? Possibly nothing, other than the fact family offices are each so sui generis that trying to generalise about them is essentially a waste of time. But maybe that makes them more interesting?
