If a broker suspects you’re being scammed, experiencing cognitive decline, or simply can’t be reached during a market crisis, who should they call? Since 2018, financial firms in the United States have been required to ask their retail clients this very question under FINRA Rule 4512, which invites investors to voluntarily designate a “trusted contact.” The designation is free, takes minutes, and offers a layer of protection against financial exploitation. Yet most eligible investors have never done it.
A study published in the Journal of Behavioral and Experimental Finance examines why uptake remains so limited and what the pattern of adoption reveals about how light-touch financial regulation actually works in practice. The short answer: knowing more about finance helps, but only if you live somewhere people trust each other.
A voluntary safeguard with stubbornly low adoption
Ioannis Petrakis of Northumbria University drew on microdata from the 2021 National Financial Capability Study, zeroing in on 2,824 active retail investors with non-retirement accounts. These are the people directly exposed to broker-dealer compliance rules and presumably most likely to benefit from naming a trusted contact. Even so, only 37.9% had done it. Meanwhile, 21.9% had been named as a trusted contact by someone else.
That dual structure, naming someone versus being named, shapes the study. The first captures an individual’s own decision to adopt a protective measure. The second reveals something about social standing: who do peers perceive as reliable enough to oversee their financial affairs in an emergency?
The financial literacy puzzle
A natural guess is that more financially knowledgeable people would be quicker to adopt the safeguard. But when Petrakis ran a straightforward statistical comparison, financial literacy, measured by correct answers to four standard questions on interest, inflation, diversification, and mortgage pricing, barely budged the probability of naming a trusted contact.
That null result is itself suspicious. Financial literacy isn’t randomly distributed. People accumulate it through market participation, advisor relationships, and selective exposure to information, all of which plausibly correlate with unobserved traits that also shape whether someone engages with consumer protection tools. To separate cause from correlation, Petrakis used a technique called instrumental variables, which isolates a portion of literacy driven by something external to an individual’s choices.
The instrument here was whether a respondent attended high school in a state that mandated personal finance coursework. State legislatures adopted these mandates at different times, well before Rule 4512 existed, so the exposure affects adult financial knowledge without being tied to current investing decisions.
Once literacy was isolated this way, the picture changed sharply. Each additional correct answer on the four-question literacy test raised the probability of naming a trusted contact by roughly 6 to 11 percentage points, a large effect relative to the 37.9% baseline. The same instrument showed that each correct answer raised the likelihood of being named by a peer by about 7.7 percentage points, roughly a 35% increase over the baseline rate.
Petrakis interprets the gap between the naive estimate and the instrumented estimate as evidence that unobserved factors were masking the true relationship, an issue known as attenuation bias.
Portfolios, complexity, and the limits of attention
Portfolio structure also mattered, and in ways that suggest both salience and signaling are at work. Investors holding non-risky assets like bonds and mutual funds were about 16.6 percentage points more likely to name a trusted contact. Cryptocurrency owners were 4.4 percentage points more likely. Each additional asset category in a diversified portfolio raised the probability by 5 percentage points, and larger total investment values pushed it higher still.
These patterns held even after accounting for whether investors used professional financial advisors, which rules out the simple explanation that adoption just reflects delegation to professionals. Instead, the findings are consistent with the idea that bigger, more complex portfolios make the stakes feel real and the safeguard feel worthwhile.
There’s also a complementarity at play. Financial literacy and portfolio engagement reinforce each other: the effects of diversification and portfolio scale are substantially larger among more financially capable investors. Having complex holdings isn’t enough on its own; the cognitive tools to process what they mean appear to convert complexity into protective action.
The same portfolio features also predicted being named by others, often with larger effects. Diversified, experienced, and substantial investors seem to serve as focal points of trust within their financial networks.
Where trust flips the script
The most striking finding involves the geography of compliance. Adoption rates vary dramatically across the country, exceeding 50% in the South Atlantic region but falling below 30% in the Mountain states. Petrakis argues that this dispersion points to the role of social capital, meaning generalized trust, civic participation, and the density of informational networks.
When he examined how financial literacy interacts with state-level and regional social capital, a clear pattern emerged. In areas with low social capital, financial literacy had essentially no positive effect on trusted-contact designation, and in some specifications it was mildly negative. In high-social-capital areas, literacy became a strong positive predictor of both adoption and being named by peers.
Petrakis interprets this as a “trust-privacy trade-off.” In low-trust environments, financially sophisticated investors may rationally be more cautious about formalizing third-party access to their accounts. They understand the risks of privacy breaches, family conflict, or opportunistic misuse of financial information, and they may reasonably decline to name anyone. In high-trust settings, the same knowledge pushes them toward adoption because delegation feels credible and monitoring feels reliable.
At the 10th percentile of social capital, the marginal effect of literacy on designation was slightly negative. At the median it hovered near zero. Only at the 90th percentile did it become large and statistically positive. Regional variation in social capital more than doubled the effect of literacy at the upper end of the distribution while eliminating it at the lower end.
What this means for behavioral regulation
Rule 4512 belongs to a broader family of “light-touch” regulatory tools that rely on nudges rather than mandates. Firms must ask, but investors don’t have to answer, and nothing bad happens if they refuse. This design is intended to lower friction and respect autonomy. The study suggests it also produces uneven outcomes that depend heavily on cognitive and social resources people bring to the decision.
Petrakis argues that relying on firms’ “reasonable efforts” is unlikely to produce uniform protection. Where financial capability is high and trust is strong, even minimal prompts get internalized. Where either ingredient is missing, the same prompts may fail outright or, in low-trust settings, even deter the very investors who best understand the risks they’d be taking on.
The author suggests that bundling disclosure with financial education, community-based trust building, or advisor-mediated engagement may be needed to close structural gaps. He also notes several limits of the analysis. The data are cross-sectional, capturing a cumulative snapshot rather than the dynamics of adoption over time, so the study explains who has complied by 2021, not the timing or diffusion path. Longitudinal data would be needed to study learning and delayed uptake directly.
One alternative explanation, that the 2021 survey captured a pandemic-era surge of novice investors whose behavior might not generalize, was addressed with a separate analysis. Investors who started trading during 2019–2021 didn’t drive the literacy effect, and the results were stable when these recent entrants were isolated.
