Are we trading more than we should?

Written by: Nicola di Crescenzo, Anna Murru, Sebastiano Murru, Alessandro Orrù

1. Introduction

“The investors who inhabit the real world and those who populate academic models are distant cousins.” (Barber and Odean (2001))

In traditional financial models, market participants are described as rational decision-makers who tend to minimize transaction costs and hold diversified portfolios. However, this depiction fails to accurately describe real-world investor behavior. In reality, people are often driven by cognitive biases: they trade more than necessary, sell their winners and hold their losers (the disposition effect), while maintaining portfolios with high levels of idiosyncratic risk. Behavioral biases are systematic deviations from rational behavior arising from the interaction between reward/emotion systems (amygdala, ventral striatum) and cognitive/control system (prefrontal cortex), which eventually lead individuals to make decisions that differ from traditional rational models.

In the literature, overtrading is described as the practice of buying and selling investments more frequently than one’s financial situation requires. This behavior is particularly harmful because, on average, it results in poorer performance compared to that of a low-cost index fund- a gap that is partially, but not entirely, driven by transaction costs.

We begin by investigating whether the phenomenon of overtrading actually exists in the market. Subsequently, we explore the cognitive biases that drive it, and ultimately, we analyze its quantitative consequences on investors’ performance.

2. Do Investors Trade Too Much?

The paper Do Investors Trade Too Much, Odean (1999), attempts to answer the question of whether individual investors trade excessively by analyzing a sample of investors with discount brokerage accounts.

At the time of the study, the average annual turnover rate on the New York Stock Exchange was over 75% - a figure that remains similar today. While this appears exaggeratedly high for meeting investors’ needs, the lack of a definitive 1 model predicting what trading volume should be in real markets makes it difficult to affirm whether this is the case. Therefore, the paper tests for excessive trading by examining whether investors engage in trading even when their expected profits are insufficient to cover the costs, implying an overestimation of expected returns.

Furthermore, when computing the monthly returns on the “buy” portfolio (RBt) minus the “sell” portfolio (RSt) and regressing them against the Capital Asset Pricing Model (CAPM):

The results yield negative alphas for all time horizons, indicating negative performance after accounting for market risk. Similar results were obtained using the Fama-French 3-factor model, where the intercept term remained reliably negative.

3. Why Do Individuals Overtrade?

Having established that overtrading exists, we must examine the drivers behind it. While some trading activity is rational, driven by unexpected liquidity shocks, the need to rebalance portfolio weights (e.g., maintaining a 50/50 small-cap/large-cap split), or tax-loss harvesting, behavioral biases play a significant role in excessive frequency.

3.1 Overconfidence

Overconfidence is defined as the tendency to place excessive faith in one’s own knowledge and abilities. In the context of finance, this bias manifests in three primary ways: Overprecision: Overestimating the accuracy of one’s own knowledge. Overestimation of abilities: Believing one’s skills are superior to what they objectively are. Better-than-average effect: The belief that one is better than the median person at a given task, often disregarding average statistical outcomes. These biases lead investors to overestimate their understanding of market dynamics and their ability to anticipate price movements. This combination becomes particularly risky when investors disregard empirical evidence regarding the poor performance of active traders, believing such statistics do not apply to them. Studies support this link. A survey of 1,345 German investors found that individuals who considered themselves more knowledgeable than average churned their portfolios more frequently. Similarly, a study of 475 U.S. investors showed that those comfortable with their understanding of investment products had a higher propensity to trade.

Neuroscientific evidence helps explain the mechanisms that sustain overconfidence: functional magnetic resonance imaging (fMRI) studies have shown increased activation in reward-memory-related regions (bilateral striatum and hippocampus) associated with states of elevated confidence. This bias is also associated with altered activity in prefrontal regions such as the medial prefrontal cortex and dorsal anterior cingulate cortex, which are involved in self-evaluation, error monitoring, and uncertainty processing. Other studies highlight the’ tendency of people to feel more confident in the option they choose while underweighting contradictory information. This bias has been observed to be related to specific brain activity, especially in the anterior dorsal cingulate cortex.

3.2 Sensation Seeking

Beyond confidence, trading may be driven by the desire for stimulating experiences. Just as individuals may gamble to feel the adrenaline of uncertainty, trading can become an activity pursued for the sensations it produces rather than for financial utility. Research indicates that the sensations derived from trading are comparable to those of gambling. Surveys show that investors who agree with statements such as "I enjoy investing" or "Games are only fun when money is involved" trade twice as much as others. Further evidence of this substitution effect was observed in Taiwan, where trading volume on the stock exchange decreased by 25% when a legal lottery was introduced. This suggests that for many, trading is a form of entertainment.

Sensation seeking in trading, as well as in gambling, is triggered by the perception of a potential reward in the environment, and strongly associated with the mesolimbic dopaminergic reward system. Reward system lies along one of the major dopamine pathways in the brain: the meso-limbic pathway, which 3 projects from the ventral tegmental area through the nucleus accumbens in the limbic system, to the neocortex. Dopamine is a neurotransmitter known as the “pleasure” chemical of the brain, since, when dopamine centers are stimulated, the subject reports an intense feeling of excitement. Dopamine also plays a significant role in reward anticipation and motivation. High sensation-seeking individuals show heightened reactivity of the ventral striatum to high-risk stimuli, together with a weaker top-down control from lateral prefrontal cortical regions, resulting in increased impulsivity and risk-taking.

Gender Differences

The link between overconfidence and overtrading also reveals gender-based disparities. Men generally exhibit higher levels of overconfidence in fields perceived as male-dominated, such as finance. Consequently, research on German and U.S. investors has shown that men churn their portfolios at an annual rate of approximately 80%, compared to 50% for women. While gross returns are often similar between genders, the higher transaction costs incurred by men result in lower net returns for male investors compared to their female counterparts. However, these differences tend to diminish when adjusted for self-reported risk aversion levels.

The Impact of Online Trading

The transition to online trading has exacerbated these tendencies. In the study Online Investors: Do the Slow Die First?, Barber and Odean (2002) analyzed 1,607 investors who switched from phone-based to online trading. Prior to the switch, these investors beat the market by 2% annually. After the switch, their trading became more active and speculative, causing them to lag the market by more than 3%. In order to capture the effects of the switch, the online traders are size matched with other investors who have not traded online. The group of online investors exhibited a turnover of 70% before the switch, already higher than the 50% observed in the other group. Their level then spiked to 120% in the month immediately after the switch, before stabilizing at 90%. No similar trend is observed in the group which did not switch to the online. This confirms that the ease of execution provided by online platforms acts as a catalyst for overconfident investors to trade more frequently, often to their detriment.

4. Quantifying Overtrading Losses

Now that the behavioral causes have been widely discussed, it is crucial to quantify the economic impact of overtrading. The seminal study by Barber, Lee, Liu, and Odean (Just How Much Do Individual Investors Lose by Trading? Barber, Lee, et al. (2009)) breaks down investor losses into four distinct categories, based on a detailed analysis of the Taiwanese market (an excellent proxy for high-frequency retail trading).

4.1 Trading Losses (27%)

Often underestimated, trading losses come primarily from the bid-ask spread and adverse selection. Individual investors frequently place aggressive market orders driven by news or price movements. However, the counterparty is often an institutional investor with superior information or execution speed. Consequently, individuals tend to consume liquidity when it is most expensive or provide it at unfavorable prices.

4.2 Commissions (32%)

Despite the rise of "zero-commission" models, transaction costs remain a critical drain on performance, often disguised as wider spreads or Payment for Order Flow. In the study Trading Is Hazardous to Your Wealth (Barber and Odean (2000)), 66.465 US households were analyzed between 1991 and 1996, dividing them into five groups according to their monthly turnover. The top 20% of investors by turnover (averaging 258% annual turnover, meaning they completely reshuffled their securities portfolio twice a year) underperformed the lowest turnover quintile by 7 percentage points. Investors in the first quintile reported a return of 18.5% net of costs, while those in the last quintile reported only 11.4%. Much of this underperformance was attributable to direct transaction costs The hyperactive investor essentially works for the broker rather than for themselves.

4.3 Transaction Taxes (34%)

The largest share of losses, in specific tax contexts such as Taiwan (and applicable to varying degrees to other markets), is represented by transaction taxes. Every time an investor closes a position, a percentage of the capital is taken by the tax system. The higher the frequency of transactions (turnover), the more devastating the tax impact becomes. A buy-and-hold investor pays this tax once; an overtrader may pay up to several times a day.

4.4 Market-Timing Losses (7%)

Although smaller in percentage terms, market-timing losses are absolute destroyers of wealth. These occur when investors increase equity exposure after market rises (FOMO) and reduce it after falls (panic selling), driven by an inability to manage emotional responses.

Aggregate Economic Impact

The aggregate annual losses of individual investors in the Taiwan study amounted to approximately 3.8% of their total invested assets and 2.8% of the total per sonal income of the population. To put this in perspective, the authors noted that total trading losses were equivalent to 2.2% of Taiwan’s GDP comparable to the entire nation’s private spending on clothing and footwear.

5. Conclusion

 Overtrading is a pervasive phenomenon in financial markets, driven not by rational profit-seeking but by cognitive biases such as overconfidence and sensation seeking. It affects demographic groups differently, with distinct patterns observed between genders and regarding the medium of trading. Most importantly, the evidence is unequivocal: overtrading is detrimental to wealth. Whether through direct commissions, taxes, or poor market timing, hyperactive investors consistently underperform their more passive counterparts.

To conclude, overtrading reminds us that markets run on human cognition, and as we uncover how neural systems for reward, attention, and error monitoring shape choices, neuroscience is becoming an increasingly essential lens in modern finance - one that can guide better design, regulation, and investor behavior

References

Banca d’Italia (2025). Cognitive Biases in Finance. URL: https://economiapertutti.bancaditalia.it/aree-tematiche/trappole-comportamentali/eccessodi-fiducia/index.html?dotcache=refresh (visited on 11/23/2025).

Barber, Brad M., Yi-Tsung Lee, et al. (2009). “Just How Much Do Individ ual Investors Lose by Trading?” In: The Review of Financial Studies 22.2, pp. 609–632. URL: https://faculty.haas.berkeley.edu/odean/papers%20current%20versions/justhowmuchdoindividualinvestorslose_rfs_2009.pdf.

Barber, Brad M. and Terrance Odean (2000). “Trading Is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors”. In: The Journal of Finance 55.2, pp. 773–806. URL: https://faculty.haas.berkeley.edu/odean/papers%20current%20versions/individual_investor_performance_final.pdf.

Barber, Brad M. and Terrance Odean (2001). “The Behavior of Individual Investors”. In: Handbook of the Economics of Finance. URL: https://faculty.haas.berkeley.edu/odean/papers%20current%20versions/behavior%20of%20individual%20investors.pdf (visited on 11/23/2025).

Barber, Brad M. and Terrance Odean (2002). “Online Investors: Do the Slow Die First?” In: URL: https://faculty.haas.berkeley.edu/odean/papers%20current%20versions/online%20rfs.pdf (visited on 11/23/2025).

Chase Bank (2025). Overtrading: Definition, Examples, and How to Make Sure You Don’t Do It. URL: https://www.chase.com/personal/investments/learning-and-insights/article/overtrading-definition-examples-and-how-to-make-sure-you-dont-do-it (visited on 11/23/2025).

Molenberghs, Pascal et al. (2016). “Neural correlates of metacognitive ability and of feeling confident: a large-scale fMRI study”. In: Social Cognitive and Affective Neuroscience 11.12, pp. 1942–1951. URL: https://pubmed.ncbi.nlm.nih.gov/27445213/.

Odean, Terrance (1999). “Do Investors Trade Too Much?” In: American Economic Review 89.5, pp. 1279–1298. URL: https://faculty.haas.berkeley.edu/odean/papers%20current%20versions/doinvestors.pdf (visited on 11/23/2025).

Sacré, Pierre et al. (2019). “Risk-taking bias in human decision-making is en coded via a right-left brain push-pull system”. In: Proceedings of the National Academy of Sciences of the United States of America 116.4, pp. 1404–1413. URL: https://pubmed.ncbi.nlm.nih.gov/30617071/.

Sun, Fanru et al. (2025). “Confidence bias prescribes the neurocomputational mechanism of decision-making”. In: Cell Reports 44.5, p. 115563. URL: https://www.cell.com/cell-reports/fulltext/S2211-1247(25)00334-1