There’s been much discussion about circular deals in AI; I myself am skeptical that these deals are cause for alarm. But there’s another circularity that’s flying under the radar: Robinhood (the stock) is one of the top holdings of the customers of Robinhood (the company).[1] As of December 2025, Robinhood is the 26th most popular holding of Robinhood customers. Another example is Moomoo, a retail broker similar to Robinhood. On the list of U.S. stocks most popular with Moomoo customers, you’ll usually find the ADR of Moomoo’s parent company.
You might call this pattern “circularity” “reflexivity,” “self-referential,” “perpetual motion,” “fiscal incest,” or “pyramidal.” Whatever you call it, we’ve seen it before. Broker stocks such as E*Trade and Ameritrade soared in the late 1990s, as their customers frantically traded shares. And among the shares they frantically traded were E*Trade and Ameritrade shares. In 1999, E*Trade was among the top ten stocks traded by E*Trade customers,[2] while Ameritrade was one of the top three most frequently purchased stocks among Ameritrade customers.[3]
How can we explain the phenomenon that customers of broker XYZ buy XYZ shares? One explanation is familiarity bias. A large academic literature shows that investors tend to buy what they know, preferring to buy shares in XYZ when the investors are:
- Employees of XYZ.
- Customers of XYZ.
- Geographically close to XYZ.
For example, Keloharju, Knüpfer, and Linnainmaa (2012) find that both for brokers and automobiles, customers of XYZ have a higher probability of buying XYZ shares.
“Buy what you know” is generally bad advice, at least for individual investors. Better advice is “buy what you don’t know,” that is, buy an index fund holding thousands of stocks that you’ve never heard of. Familiarity bias leads to underdiversification, often resulting in severe wealth destruction for unsophisticated individuals. You might think that familiarity with company XYZ would give individuals an advantage in trading XYZ, but on average it doesn’t work that way. Keloharju et al., like most studies, find that individuals do not exhibit skill when trading stocks that are familiar to them.
A second explanation is that sometimes broker XYZ gives incentives for its customers to buy XYZ shares. When Robinhood went public in 2021, it allocated one-third of its new shares to its customers. Democratizing finance in this way has downsides, as shown by the events of 1929. National City (the precursor to Citibank) sold its own shares to its own customers in what was called “the most flamboyant high-pressure bank stock selling campaign in all history.”[4] This practice led to disastrous outcomes and was subsequently vilified in the Pecora hearings of 1933.
A third explanation is that widespread trading of broker stocks is a sign of speculative froth. As I’ve previously discussed, retail broker stocks have soared recently, part of a general increase in speculative activity. Robinhood appears on lists of retail favorites, such as previously-discussed lists of stocks purchased by U.S. and Korean retail investors. Robinhood also regularly appears on the Moomoo list of popular stocks. Another sign of frenzy is that shares in many brokers (including Robinhood, eToro, and Moomoo’s parent company) now have 2X ETFs, allowing retail investors to take levered bets.
The seemingly circular trading of broker shares reminds me of crypto, and that’s not a good thing. On a conceptual level, crypto apologists often use the circular argument that crypto is good because it empowers people to trade crypto. Many crypto projects are circular by design. For example, Sam Bankman-Fried’s scheme involved multiple levels of circular logic. I will not attempt to explain the tangled web involving Alameda Research, the FTX exchange, and the FTT token, but instead will just quote Robert Armstrong of The Financial Times:[5]
A hedge fund’s solvency should not depend on a carnival ticket that was invented to facilitate, and which derives all its value from, trade on an exchange that is owned by the same person as the hedge fund. This is circular and creepy and probably not sustainable.
What does it all mean? Whenever you observe a phenomenon from 1999 recurring again, you’ve got to be concerned about a speculative mania. For me, XYZ customers buying XYZ shares is just another mile marker on the road to bubble city.
Endnotes
[1] References to this and other companies should not be interpreted as recommendations to buy or sell specific securities. Acadian and/or the author of this post may hold positions in one or more securities associated with these companies.
[2] “E-Trade Shares Rise 19% After Announcing 2-for-1 Stock Split,” L.A Times, January 5, 1999.
[3] “AOL, Ameritrade and Dell are online investor's favorite buys,” Deseret News, January 12, 2000.
[4] “Damnation of Mitchell,” Time, March 6, 1933.
[5] “Bank man, fried,” The Financial Times, November 9, 2022.
References
Keloharju, Matti, Samuli Knüpfer, and Juhani Linnainmaa. "Do investors buy what they know? Product market choices and investment decisions." The Review of Financial Studies 25, no. 10 (2012): 2921-2958.
