by Editorial Team, AdvisorAnalyst
The finfluencer problem is bigger than anyone in the wealth management industry wants to admit. Twenty-six percent of U.S. investors now make investment decisions based on recommendations from social media personalities. Among investors aged 18 to 34, that number climbs to 61 percent. And according to a new working paper by Toshiaki Mitsudome of NYU's School of Professional Studies1, the behavioral engine driving all of it isn't ignorance — it's overconfidence.
Using the 2024 National Financial Capability Study Investor Survey — a nationally representative sample of 2,831 U.S. investors — Mitsudome isolates the specific cognitive bias that makes retail investors click, follow, and trade on the word of strangers on the internet. The finding is precise: a one standard deviation increase in the overconfidence gap raises the probability of making investment decisions based on finfluencer recommendations by 6.4 percentage points — a 23 percent relative increase against the population mean.
That number matters. It isn't a soft correlation. It survives demographic controls, risk tolerance adjustments, investment experience, and social media platform exposure. Overconfidence operates, in Mitsudome's words, "through an independent cognitive channel that persists after controlling for platform exposure."
What Overconfidence Actually Means Here
The paper defines overconfidence precisely — not as arrogance, but as miscalibration. It is the gap between what an investor believes they know and what they can actually demonstrate they know, measured by comparing self-rated financial knowledge scores to objective financial knowledge test results.
The descriptive statistics make the gap vivid. Finfluencer followers score an average of 4.9 correct answers out of 12 on the objective knowledge battery — versus 5.7 for non-followers. Yet followers rate their own knowledge at 5.2 on a 7-point scale, compared to 4.1 for non-followers. They know less and think they know more. That combination — not ignorance alone, and not confidence alone — is what Mitsudome identifies as the structural predictor.
His robustness checks confirm it cleanly. When subjective knowledge is separated from objective knowledge in the regression, the subjective score is positively associated with finfluencer susceptibility, while objective knowledge is negatively associated. As Mitsudome concludes: "the bias — not the ignorance — drives susceptibility."
The Advisor Amplification Problem
Here is where the paper delivers its most uncomfortable finding for the wealth management industry.
The conventional assumption is that working with a financial advisor provides a buffer against bad behavioral decisions. Professional relationships, accountability, fiduciary duty — the story goes that advised investors are more protected. Mitsudome's data say otherwise.
Overconfident investors who use financial advisors are more likely to follow finfluencer recommendations than overconfident investors who do not. The interaction coefficient is +5.5 percentage points, significant at the 0.001 level.
Mitsudome's explanation is structural. Finfluencer activity is client-initiated and platform-accessible. Unlike structured products or private placements — where an advisor's fingerprint is on the transaction — finfluencer-driven decisions happen before the advisor is in the room. "When a client acts on a finfluencer's advice, the advisor's reputational exposure is limited. This limited attribution weakens the incentive for advisors to actively monitor or correct overconfident clients' finfluencer engagement."
There is also a selection effect at work. Overconfident investors may actively choose advisors who validate their views rather than challenge them — a pattern consistent with prior research showing that investors gravitate toward advisors who share their risk appetite and beliefs. "Overconfident investors may have selected into advisory relationships that validate rather than challenge their beliefs," Mitsudome notes, citing Stolper and Walter (2019).
The implication for advisors is direct: having a client relationship does not automatically mean having influence over that client's social media-driven behavior.
Mobile Apps: The Lowest-Friction Path to the Worst Decision
The paper's third finding addresses trading channels. Mobile app traders — investors who place orders through apps independently — show the strongest amplification of overconfidence-driven finfluencer susceptibility in the entire study. The app trading coefficient is 0.222 (p < 0.001), and the overconfidence–app interaction adds another 0.064 (p < 0.001).
The mechanism is friction — or the absence of it. "Overconfident app traders face virtually no friction between receiving a finfluencer recommendation and executing it." No call to an advisor. No review period. No second opinion. A notification, a tap, an order.
For investors who are already prone to overweighting attention-grabbing signals — which is precisely what finfluencer content is designed to be — removing the human intermediary removes the last potential pause in the decision chain.
Key Takeaways for Advisors
- You are not a corrective buffer by default. The data show advisors amplify finfluencer susceptibility among overconfident clients, not reduce it. Addressing finfluencer behavior requires explicit, proactive conversation — not the assumption that your relationship provides protection.
- Identify miscalibration, not just knowledge gaps. Clients who score high on self-assessed knowledge but struggle with basic financial concepts are the highest-risk group. Ask questions that expose the gap between confidence and accuracy.
- The mobile trading channel is a direct line from finfluencer to portfolio. Clients who trade independently through apps are operating outside your visibility entirely.
Key Takeaways for Investors
- Knowing that you don't know something is more protective than thinking you know it. The research is unambiguous: the investors most susceptible to finfluencer advice are those who overestimate their own knowledge.
- Having a financial advisor does not make you immune. Your advisor cannot correct behavior they cannot see — and finfluencer-driven decisions often happen before any professional conversation occurs.
- The platform is designed to move you quickly. Slowing down the decision is the intervention.
Mitsudome is direct about what the regulatory response needs to look like: "The regulatory debate about finfluencers should extend beyond disclosure requirements to address the underlying behavioral biases that make investors susceptible." Disclosure rules assume investors will read, process, and apply warnings rationally. Overconfident investors, almost by definition, will not.
The finfluencer era is not a technology problem. It is a calibration problem. And the wealth management industry — advisors, regulators, and educators alike — is only beginning to understand what that means.
Footnote:
1 Mitsudome, T. (2026). "Overconfidence, Financial advisors, and Social Media Investment Influence: Evidence from U.S. Investors." NYU School of Professional Studies. Data: 2024 National Financial Capability Study Investor Survey (FINRA Investor Education Foundation).
