Financial Advisers: What the Evidence Shows
Key takeaways · 12 min read
- About 7% of US advisers had a misconduct record, and past offenders were five times as likely to reoffend.
- Almost half of advisers who left after misconduct found another industry job within a year.
- Mystery shoppers with low-cost index portfolios were told to change strategy 85% of the time.
- Canadian clients paid more than 2.5% a year for advice that depended more on the adviser than on the client.
Most people who hand their savings to a financial adviser are buying two things: expertise they do not have, and someone to trust with decisions they would rather not make alone. Both can be worth paying for. But advice is also a business, and the person giving it is often paid in ways that depend on what the client buys.
For a long time, how that worked out for clients was a matter of opinion. Over the past fifteen years, economists have measured it. They have used regulators’ complete records of adviser misconduct, sent trained mystery shoppers to advisers’ offices, and followed hundreds of thousands of advised clients and the advisers’ own personal accounts.
The results are more nuanced than either the industry or its critics usually admit. Misconduct is concentrated among a minority who often keep working. Advice is expensive and surprisingly uniform. Yet for some people, advice still leaves them better off than going it alone. This article sets out what the evidence shows.
How common misconduct is
In the United States, every registered broker has a public record on FINRA’s BrokerCheck system. In 2019, Mark Egan, Gregor Matvos and Amit Seru published an analysis of all of those records from 2005 to 2015, about 1.2 million people, in the Journal of Political Economy. They counted six types of disclosure as misconduct, including settled customer complaints, arbitration awards against the adviser, regulatory and criminal actions, and being let go after allegations.
By that definition, 7.28% of advisers had at least one misconduct record, about one in thirteen. At some of the largest firms the share was more than 15%. In parts of Florida and California, about one in five advisers had a record. Misconduct rates were 19% higher in counties with below-average income and education, where clients may be least able to check.
The disputes were not trivial. The median settlement or award was $40,000, the mean about $550,000, and the total close to half a billion dollars a year. About one in four complaints alleged unsuitable investments, and about a third alleged misrepresentation or omission.
Repeat offenders keep working
The most striking findings concerned what happened next. Advisers with past misconduct were five times as likely as the average adviser to commit new misconduct, even compared with colleagues at the same firm in the same county and year. After a misconduct record, the chance of another one was 11% in the first year, about 4% by the fifth and 1.5% by the ninth, against an annual rate of about 0.6% for advisers overall.
The risk of a new misconduct record
Annual probability of a new misconduct disclosure for US advisers, by years since a previous one.
Egan, Matvos and Seru, Journal of Political Economy, 2019. FINRA BrokerCheck records, 2005–2015.
Firms did react. Within a year of misconduct, 48% of advisers had left their firm, against 19% of others. But 44% of those who left found a new job in the industry within a year, usually at firms that paid less and employed more advisers with records of their own. The authors called this matching on misconduct. In a 2024 follow-up, the same authors reported that publicly identifying the firms with the highest misconduct rates was associated with a drop of about 10% in misconduct.
Punishment was also unequal. In a 2022 paper titled ‘When Harry Fired Sally’, the same team found that after misconduct, female advisers were 20% more likely to lose their jobs and 30% less likely to be rehired than male advisers, a gap that shrank at firms with more women in management.
What happens after misconduct
US advisers with a misconduct record, 2005–2015.
Egan, Matvos and Seru, Journal of Political Economy, 2019.
What advisers recommend
Misconduct is the extreme. The more common question is whether ordinary advice is good. In 2008, Sendhil Mullainathan, Markus Noeth and Antoinette Schoar sent trained mystery shoppers on 284 visits to financial advisers. Each shopper described one of four existing portfolios: one chasing a sector that had just done well, one heavy in their employer’s shares, a diversified low-fee index portfolio, or all cash.
Advisers recommended actively managed funds in about half of the visits and index funds in only 7%. Shoppers who arrived with the low-cost index portfolio were told to change strategy 85% of the time, more often than those chasing past returns, who were told to change 59% of the time. Advisers supported the index portfolio in only 2.4% of visits. The authors noted a possible reason besides commissions: advisers may feel they must recommend something different to show that they add value.
Mystery shoppers and the advice they received
284 audit visits to US financial advisers, 2008.
Mullainathan, Noeth and Schoar, NBER Working Paper 17929, 2012. About 45% of visits produced no specific allocation advice.
Other studies point the same way. A 2009 paper by Daniel Bergstresser, John Chalmers and Peter Tufano found that US mutual funds sold through brokers from 1996 to 2004 delivered lower risk-adjusted returns than funds bought directly, even before distribution fees. The authors were careful: brokers might provide benefits the data could not see. In a 2019 study of structured bonds called reverse convertibles, Mark Egan found that cheaper and more expensive versions of essentially identical products were sold side by side, that customers often bought the worse ones, and that brokers typically earned about twice the fees on them.
The price of advice
The most detailed evidence on cost comes from Canada. Stephen Foerster, Juhani Linnainmaa, Brian Melzer and Alessandro Previtero studied 581,044 clients of 5,920 advisers at three dealers from 1999 to 2012, published in the Journal of Finance in 2017. Advised clients paid more than 2.5% a year in fees, roughly the entire equity premium that investors hope to earn for taking stock-market risk.
The advice was also strikingly uniform. A wide set of client characteristics, including risk tolerance, age and investment horizon, explained only 13% of the variation in how much clients held in risky assets. Which adviser they happened to have explained a further 20%. Moving from an adviser at the 25th percentile to one at the 75th meant about 20 percentage points more in risky assets, regardless of who the client was.
A 2021 follow-up by Linnainmaa, Melzer and Previtero asked whether this was a conflict of interest or something else. They looked at the advisers’ own portfolios. The advisers traded often, chased returns, preferred expensive active funds and were poorly diversified, just as their clients were. Their own net returns were about 3% a year below the benchmark, and they kept holding expensive portfolios after they left the industry. The authors concluded that many advisers sincerely believed in the strategies they sold.
What the studies of advice cost and quality found
Selected findings from large studies of retail advice.
| Study | Setting | Finding |
|---|---|---|
| Foerster and others, 2017 | 581,044 Canadian clients | Fees above 2.5% a year; adviser identity mattered more than client traits |
| Linnainmaa and others, 2021 | Advisers’ own accounts | About −3% a year net; same habits as clients |
| Bergstresser and others, 2009 | US mutual funds, 1996–2004 | Broker-sold funds underperformed direct-sold funds |
| Chalmers and Reuter, 2020 | Oregon university pension plan | Target-date funds beat broker-advised portfolios on risk-adjusted return |
Journal of Finance, Review of Financial Studies and Journal of Financial Economics. Canadian and US data.
Estimates of the total cost are contested. In 2015 the White House Council of Economic Advisers estimated that conflicted advice cost American retirement savers about 1 percentage point a year, or about $17 billion a year across $1.7 trillion of affected savings. Industry economists argued that the report leaned on small studies and ignored the benefits of broker services. The council itself noted that even half that effect would cost savers more than $8 billion a year.
The case for advice
None of this means advice is worthless. In a 2015 paper called ‘Money Doctors’, Nicola Gennaioli, Andrei Shleifer and Robert Vishny argued that people pay for trust as much as expertise. Trust lets anxious investors take risks they would otherwise avoid, so even an adviser who underperforms the market after fees can leave a client better off than keeping everything in cash. Foerster’s team found exactly that effect: taking on an adviser raised clients’ share of risky assets by about 30 percentage points.
The comparison that matters is with what the client would do otherwise. John Chalmers and Jonathan Reuter studied an Oregon university pension plan that removed its brokers and added target-date funds. Brokers had helped participants hold market risk, but steered them to higher-commission options. Afterwards, similar new members mostly chose target-date funds, with similar risk and better risk-adjusted returns. Conflicted advice beat no advice only when there was no good default.
Automated advice helps some investors. Francesco D’Acunto, Nagpurnanand Prabhala and Alberto Rossi studied a robo-adviser launched by an Indian brokerage in 2015. Clients who held fewer than five stocks became more diversified and performed better, and a common bias, selling winners too early while holding on to losers, fell by about 30%. Clients who were already diversified traded more and gained nothing. A 2024 paper by Jonathan Reuter and Antoinette Schoar argued that households with low financial literacy gain most from advice, even conflicted advice, and are also least able to spot misconduct.
What the records cannot show
Misconduct records are an imperfect measure in both directions. Egan and colleagues excluded pending, denied and withdrawn complaints, so their measure is conservative, and they found pending complaints themselves predicted misconduct. On the other side, a settlement is not proof of wrongdoing, and advisers can apply to have records removed. FINRA calls this expungement an extraordinary remedy, and tightened its rules in October 2023.
A formal fiduciary duty is also no guarantee. The misconduct patterns in the 2019 study appeared among registered investment advisers, who owe clients a fiduciary duty, as well as among brokers. The authors concluded that holding all advisers to a fiduciary standard might not be enough to deal with misconduct.
What regulators have done
The rules have moved back and forth. In the United States, the SEC’s Regulation Best Interest requires brokers to act in a retail client’s best interest when recommending investments, together with a short relationship summary called Form CRS, and has applied since 30 June 2020. The Department of Labor’s 2016 fiduciary rule for retirement advice was struck down by a federal appeals court in March 2018. Its 2024 replacement was blocked by Texas courts in July 2024 and formally vacated, with the Labor Department announcing in March 2026 that it had no current plans for new rules.
The United Kingdom went further in 2013, when its Retail Distribution Review banned commissions on investment advice. An early review for the regulator found fewer sales of products that had paid high commissions, but also that people with small sums found advice harder to get. Only 9% of UK adults received regulated advice in the year to May 2024, and a new, lighter form of targeted support began in April 2026.
Rules on advice since 2012
Main changes in the United States and the United Kingdom.
| Date | Where | What changed |
|---|---|---|
| Dec 2012 | UK | Commissions on investment advice banned |
| Mar 2018 | US | Labor Department’s 2016 fiduciary rule struck down on appeal |
| Jun 2020 | US | SEC Regulation Best Interest and Form CRS apply |
| Jul 2024 | US | Labor Department’s 2024 Retirement Security Rule blocked by courts |
| Mar 2026 | US | That rule formally vacated; no new rules planned |
| Apr 2026 | UK | New targeted support regime begins |
SEC, US Department of Labor, Financial Conduct Authority. UK advice-gap figures reported via an FCA-cited summary.
Recent cases
Enforcement continues at scale. In 2025, FINRA reported 625 new disciplinary actions, $99.6 million in fines and 187 individuals barred from the industry. In January 2025, the SEC settled charges against Wells Fargo and Merrill Lynch over cash sweep programmes in advisory accounts, which were often the only cash option offered and paid far less than alternatives as interest rates rose, with penalties totalling $60 million. In August 2025, Vanguard’s advisory arm paid $19.5 million to settle charges that it had failed to disclose properly how its advisers were paid for enrolling and keeping clients. The firms settled without admitting or denying the findings.
The research suggests questions rather than answers for anyone choosing an adviser: how is this person paid, and does it depend on what I buy; what does their public record show; what would I do without advice; and what does the total annual cost add up to? Those costs compound in the same way as the returns we described in our article on sequence-of-returns risk.
Questions people ask
How many financial advisers have misconduct records?
About 7% of US advisers had at least one misconduct record in 2005–2015, rising above 15% at some large firms and to about one in five in some counties.
How can I check an adviser’s record?
In the United States, FINRA’s BrokerCheck and the SEC’s adviser search show registration and disclosure histories. Other countries keep their own registers.
Is a fiduciary adviser safer?
A fiduciary duty sets a higher legal standard, but one large study found similar misconduct patterns among fiduciary advisers and brokers.
How much does financial advice cost?
In a large Canadian study, advised clients paid more than 2.5% a year in total fees. Costs vary widely by country, firm and payment model.
Is financial advice worth it?
It depends on the alternative. Studies find advice raises risk-taking and helps some undiversified investors, but low-cost defaults often did as well or better.
The short version
- About 7% of US advisers had a misconduct record, and past offenders were five times as likely to reoffend.
- Almost half of advisers who left after misconduct found another industry job within a year.
- Mystery shoppers with low-cost index portfolios were told to change strategy 85% of the time.
- Canadian clients paid more than 2.5% a year for advice that depended more on the adviser than on the client.
- Advice can help investors who would otherwise stay in cash or hold a few stocks; whether it helps depends on the alternative.
This article summarises published research and official information on financial advice. It is not financial, investment, tax or legal advice, and it does not recommend or criticise any particular adviser, firm or product. Regulations differ between countries and change over time. Check an adviser’s registration with your national regulator before relying on them.
Further reading. Egan, Matvos and Seru, ‘The Market for Financial Adviser Misconduct’, Journal of Political Economy, 2019, and their 2024 overview in the Journal of Economic Perspectives. Chalmers and Reuter, Journal of Financial Economics, 2020, is the clearest test of advice against a realistic alternative.
- The Missing Billionaires, Victor Haghani and James White (2023). Two former fund managers on sizing risk and making better financial decisions. Mathematical in places, and aimed at investors who want to think rather than delegate.
- Noise, Daniel Kahneman, Olivier Sibony and Cass Sunstein (2021). Why professionals facing the same case reach different judgments, which is what the Canadian advice data show.
- The Psychology of Money, Morgan Housel (2020). Short essays on behaviour and money. Popular rather than academic, and a useful counterweight to purely technical advice.
Sources
- Egan M, Matvos G, Seru A. The market for financial adviser misconduct. Journal of Political Economy, 2019;127(1):233–295. doi:10.1086/700735.
- Egan M, Matvos G, Seru A. The problem of good conduct among financial advisers. Journal of Economic Perspectives, 2024;38(4):193–210. doi:10.1257/jep.38.4.193.
- Egan M, Matvos G, Seru A. When Harry fired Sally: the double standard in punishing misconduct. Journal of Political Economy, 2022;130(5):1184–1248. doi:10.1086/718964.
- Mullainathan S, Noeth M, Schoar A. The market for financial advice: an audit study. NBER Working Paper 17929, 2012. doi:10.3386/w17929.
- Foerster S, Linnainmaa JT, Melzer BT, Previtero A. Retail financial advice: does one size fit all? Journal of Finance, 2017;72(4):1441–1482. doi:10.1111/jofi.12514.
- Linnainmaa JT, Melzer BT, Previtero A. The misguided beliefs of financial advisors. Journal of Finance, 2021;76(2):587–621. doi:10.1111/jofi.12995.
- Bergstresser D, Chalmers JMR, Tufano P. Assessing the costs and benefits of brokers in the mutual fund industry. Review of Financial Studies, 2009;22(10):4129–4156. doi:10.1093/rfs/hhp022.
- Egan M. Brokers versus retail investors: conflicting interests and dominated products. Journal of Finance, 2019;74(3):1217–1260. doi:10.1111/jofi.12763.
- Chalmers J, Reuter J. Is conflicted investment advice better than no advice? Journal of Financial Economics, 2020;138(2):366–387. doi:10.1016/j.jfineco.2020.05.005.
- Gennaioli N, Shleifer A, Vishny R. Money doctors. Journal of Finance, 2015;70(1):91–114. doi:10.1111/jofi.12188.
- D’Acunto F, Prabhala N, Rossi AG. The promises and pitfalls of robo-advising. Review of Financial Studies, 2019;32(5):1983–2020. doi:10.1093/rfs/hhz014.
- Reuter J, Schoar A. Demand-side and supply-side constraints in the market for financial advice. NBER Working Paper 32452, 2024.
- Council of Economic Advisers. The Effects of Conflicted Investment Advice on Retirement Savings. Executive Office of the President, February 2015.
- US Securities and Exchange Commission. Regulation Best Interest and Form CRS; statement on the 30 June 2020 compliance date.
- US Department of Labor. Press release on the vacatur of the Retirement Security Rule, 18 March 2026.
- Financial Conduct Authority. Early indications that reforms to financial advice are working. Press release, December 2014.
- FINRA. Key statistics, 2025; Expungement of customer dispute information.
- US Securities and Exchange Commission. Press release 2025-16 (Wells Fargo and Merrill Lynch cash sweep programmes), January 2025; Administrative proceeding IA-6912 (Vanguard Advisers), August 2025.
