The non-psychological literature concerning problematic trading
We included 11 papers in our review from the non-psychological area (see
Table 1
), which cover the years between 1995 and 2024. Most of these papers (seven) are from US scholars – with one involving both US and Spanish scholars – one from Spain, one from Belgium, and two from Germany. While seven research groups analyzed problematic trading from an individual perspective, two papers adopted a group-level analysis, and two offered a mixed perspective. In line with the topic of our review, which concerns a (problematic) financial behavior, the area of study – except for the sociologist Weidner (11) – is economy and law.
Remarkably, none of these studies conceptualized problematic trading as a phenomenon in itself by analyzing the specific elements of trading as a problematic/excessive behavior. In line with this, only three papers analyzed this problematic behavior through the lens of gambling (10, 11, 31), hence using gambling criteria to define trading as pathological. Klontz et al. (31) suggested ten gambling-based questions to be addressed to online traders, specifying that the person was moving from investing to gambling in case of positive answers. Similarly, Tierney (10) suggested that trading might be a rational substitute for gambling. Finally, Weidner (11) did not provide a specific criterion to define trading as problematic; however, he argued that some traders (both private and institutional) use the financial market to gamble.
Enlarging the results about the criteria used to define trading as problematic, these are not available for five (including 11) papers; two papers – previously described – refer to gambling (10, 31), while the remaining four used behavioral-based indicators. Bhattacharyya and Nanda (25) pointed out impaired long-term performance and preference for short-term performance. Corzo et al. (26) reported instead an increase in trade volume. Khlön (30) listed trading too often after periods of high market returns, taking excessive risks, not learning or learning too little, and failing to exit the market at the right time. Finally, Inghelbrecht and Tedde (28) referred to a trading behavior characterized by higher frequency and higher quantity of money, which is associated with lower performance. Their theoretical paper also includes research, and they found that their hypothesized lower performance was not present in their sample.
The no-psychological literature, however, gave rise to the field of study concerning the analysis of excessive/problematic trading as an irrational behavior, with most papers focusing on the role of overconfidence. Mahoney (32) noted two theoretical explanations for excessive trading in the literature. The noise trade model – which starts from excessive volatility – suggests that a subset of irrational traders confuse noise with information in the market. The heterogeneous expectation model by Stout (42) – which starts instead from excessive trading – posits that all investors are rational: all act based on the best information available (even if there might be a mistaken trader). Mahoney (32) suggests that even Stout (42) model should account for irrationality. He proposes that irrational trading exists and that excessive trading might result from the incentive structures facing investors and financial intermediaries (i.e., excessive trading could result from a conflict of interest between investors and intermediates whose compensation is based on transaction volume). Similarly, Tierney (10) also suggested that brokers with control over customer accounts might have incentives to increase their compensation (and trade excessively). However, Stout (33) reaffirmed that excessive trading is not due to irrationality. Possibly, the investor trades because he/she believes others in the market may be irrational.
Overconfidence is indicated as an explication variable of problematic trading by five studies (26, 28–30, 43), with Inghelbrecht and Tedde (28) particularly referring to overestimation, while De Bondt et al. (43) also reported the inclination to choose among those issues that have already caught attention. Additionally, Tierney (10), even if not explicitly referring to overconfidence, stated that (over)trading might be a “rational” substitute for gambling, but that for some people, it might be unintentional and due to irrational belief.
A few scholars have also addressed the role of financial markets structures in influencing problematic trading. Tierney (10) highlighted that gamified investing might incentivize excessive trading. Similarly, Weidner (11) pointed out that the financial industry has addictive properties like the gambling industry, attracting risk-seeking individuals. More generally, Klhön (30) suggested that the constitution of financial markets influences investors’ overconfidence, which he, along with other scholars, suggested contributes to excessive trading.
Finally, Klhön (30) pointed out that male investors trade more aggressively, while Bhattacharyya and Nanda (25) used portfolio pumping (that they defined as trading in the direction of the existing holdings of the risky asset to bolster the short-run measured value of the fund) as explication variable. For Klontz et al. (31), no explication variable arose from the reading of the paper.
The psychological literature concerning problematic trading
The 12 papers included in this review from the psychological area (see
Table 2
) cover the years between 2014 and 2023. They were conducted in Western and Eastern countries: the UK, Australia, Canada, Italy, Slovakia, Sweden, Spain, Brazil, Turkey, and China. All of them adopted an individual-level study of problematic trading.
Interestingly, excluding the papers by Wang (41) and Grežo (35) – who did not describe problematic trading as a distinct phenomenon in itself nor associated it with gambling disorder – the other ten papers consistently adopted the behavioral addiction framework, usually referencing gambling disorder. Concerning Wang (41) and Grežo (35), in line with the lack of definition of problematic trading as a distinct diagnosis, there is not a transparent reference to the criteria used to define it as pathological (35), or there is just a reference to having an income that is hardly enough to compensate for transaction cost (41).
Problematic trading as a form of gambling
Among the ten papers that utilized the behavioral addiction framework, six did not present problematic trading as a separate entity. Instead, they suggested that it could be integrated into the concept of problematic gambling.
Three papers referred to gambling criteria to define trading as problematic (6, 37, 39). Håkansson et al. (37) advised including trading among the aspects to be analyzed with gambling tools. Sonkurt (6), in his case report of a cryptocurrency trader, used the DSM-5 (17) gambling criteria to evaluate the patient. Finally, Roza et al. (39) suggested evaluating patients with gambling-like behavior in trading by adapting the gambling criteria of the last edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5 TR; 44). It is important to note that Roza et al. (39) wrote that there is not enough evidence currently supporting a distinctive diagnosis of problematic trading. Therefore, they considered that, in the future, there might be scientific evidence supporting the recognition of problematic trading as a disorder on its own.
The other three studies did not explicitly refer to gambling criteria to define trading as problematic but referred to behavioral features previously observed in gamblers. Andrade and Newall (34) cited engaging in frequent, short-term trading and continuing despite experiencing losses and the tendency to chase losses. Newall and Weiss-Cohen (9) similarly highlighted in problematic traders signs of behavioral dependency like gambling: spending more time engaging in the activity and thinking about the activity more than wanted. However, they also referenced previous studies that created scales specifically addressing problematic trading by resembling gambling instruments (45, 46). Finally, Johnson et al. (38), while suggesting cryptocurrency trading as a form of gambling, pointed out that the investment characteristics of cryptocurrency holders (i.e., excessive trading, chasing losses, borrowing money) suggest a structure similarity with problematic gambling.
Problematic trading as a disorder in its own right
Four research teams suggested conceptualizing problematic trading as a distinct diagnosis, hence analyzing the specific elements of trading (instead of merely proposing an overlap with gambling). More specifically, Arthur et al. (8) pointed out that besides considering problematic trading as a potential contributor to gambling, it is essential to also analyze it as a distinct phenomenon (and, more specifically, as a new behavioral addiction), especially concerning speculation. Roza et al. (39) previously suggested that financial investment and gambling are different activities that show considerable overlap, especially for financial speculation (e.g., day trading and cryptocurrency investing), as it is characterized by higher risk and shorter-term investments. Arthur et al. (8), instead, suggested that it is just speculation to be considered as a phenomenon on its own when compared to gambling as they posited that speculation is conceptually intermediate between investing and gambling (with whom hence share some features), whose specific qualities are related to time frame, expected returns, asset purchase, and economic utility. More specifically, speculation: i) has a variable time-frame (which is usually short for gambling and long for investment); ii) its expected returns are mixed and highly variable (while for gambling is usually negative and with low variability and for investment is usually positive and somewhat variable); iii) sometimes there is an asset purchase (which is absent in gambling and present in investment); iv) it is characterized by a mixed economic utility (while the economic utility is low for gambling and high for investing). The activities and instruments used in speculation are pretty distinct from gambling but less distinctive from investment. Moreover, the level of risk of speculation is high both in speculation and gambling (it is low for investment), while the role of chance is high across the three activities. Stakes are present in both speculation and gambling (not in investment). Finally, in gambling, there is a definitive event and outcome; in speculation, it is usually present (in investment, it is usually not present). However, Arthur et al. (8) did not present this last variable as a feature of speculation. In sum, Arthur et al. (8) conceptualized trading (in the specific form of speculation) as problematic by pointing out the need to recognize it as a contributor to problematic gambling (hence, to be assessed when evaluating gambling behaviors) but also as a new behavioral addiction in its own right (but they did not provide specific elements to definite it).
Similarly to Arthur et al. (2016), Delfabbro et al. (27) assimilated problematic trading – in the form of cryptocurrency trading – to online gambling and excessive social media use as sharing elements of risk involved in these two excessive behaviors. However, they also recognized its specific features: its 24-hour availability and long-form nature, the extreme volatility of outcomes, and the strong influence of sentiment and social influence. Therefore, regarding the criteria used to conceptualize trading as problematic, they adopted the behavioral addiction framework, with a specific reference to overspending and compulsive checking.
Guglielmo et al. (36) specifically stated to consider pathological trading, or “trading addiction,” as an (overlooked) form of behavioral addiction, pointing out the scant literature investigating addictive-like behavior among investors and criticizing the use of pathological gambling criteria or unspecified addiction criteria. They defined trading addiction as a persistent or recurrent maladaptive trading behavior that disrupts family, personal, and/or professional activities with associated progressive loss of control over trading, tolerance, and withdrawal symptoms (similar to substance use disorders). Moreover, they suggested (addiction-based) DSM-like pathological criteria specifically designed for trading, also considering that trading addiction might occur both in professional and non-professional traders. So, they proposed two criteria. The first criterion (A) posits persistent and recurrent problematic trading behavior (associated with impairment or distress) as indicated by at least five (up to 13) symptoms in the last 12 months. The list of symptoms includes, for example, tolerance, being restless or irritable when trying to cut down or stop trading, often trading when feeling distressed, and chasing losses. For some symptoms, Guglielmo et al. (36) pinpoint that the behavior should occur outside working hours when evaluating a professional trader. The second criterion (B) specifies that other psychological diagnoses should not better explain trading behavior. Finally, they also stressed that in their view, pathological trading reflects Goodman (47) concept of addiction, including both dependence and compulsion.
Finally, more recently, Vismara et al. (40) listed “compulsive online trading” as a type of digital form of Obsessive-Compulsive Related Disorders (OCRDs) while concurrently highlighting its overlap with gambling disorder and encouraging further studies on the topic to understand if it might be defined as an autonomous disorder, also considering its differences with gambling: stock exchange operations lie on individual knowledge and skills, and a longer-term perspective usually characterizes investment. Moreover, about the criteria used to define trading as problematic, they wrote that (online) trading is characterized by urge-driven, compulsive checking and investing in stock exchange transactions, with the consequent expansion of time dedicated to trading and consequent impairing of different functional areas. It is interesting to note that, even if Vismara et al. (40) included problematic trading among ORCDs, besides highlighting its overlap with gambling (i.e., a behavioral addiction), they also referenced the neuroimaging study by Knutson and Bossaerts (48) that supported the conceptualization of compulsive online trading as principally an addiction disorder, with the associated compulsive component as a second (additional) component.
Explicative variables of problematic trading
The psychological literature concerning problematic trading – regardless of its conceptualization as a disorder in its own right or as a form of gambling – highlighted some variables that might explain excessive/problematic trading. Among the 12 psychological papers included in our review, only three of them did not analyze these variables (6, 36, 40), while the other papers mainly focused on overconfidence (already studied in the non-psychological literature) and other risk factors previously observed in gamblers.
Cognitive risk factors similar to the ones previously observed in gamblers include being overconfident in own investment skills (8), having a confirmation bias – or only attending to information that confirms the person’s opinion (8), having an illusion of knowledge and control (8, 27, 39), preference for lottery-type stocks – that is, low-price stock with a slight chance of highly increasing their value (8), and high loss aversion (8), which might be conceptually linked to chasing ones’ losses, suggested by Roza et al. (39) as a risk factor. Moreover, Delfabbro et al. (27) considered preoccupation/salience, corresponding to continuously thinking about the activity. There are other scholars citing overconfidence (generally) as a risk factor for problematic trading (9, 41). Moreover, Grežo (35) provided a valuable contribution concerning this variable by distinguishing in his meta-analysis between indirect measures and three direct measures of overconfidence: overplacement (of the person’s abilities or performance relative to others), overestimation (of the person’s knowledge, abilities or performance compared to the actual performance), and overprecision (of the person’s ability to make accurate probability judgment). Regarding indirect measures, which showed a higher impact on financial decision-making, Grežo (35) underlined that some studies did not measure overconfidence but used excessive trading as a proxy. Concerning the three direct measures of overconfidence, overplacement showed a high effect, while overestimation only had a trivial effect, and overprecision was not statistically significant.
Regarding personality features, studies highlighted risk-taking behavior (8), impulsiveness (9, 34, 38), sensation seeking (8, 9, 34), novelty seeking and low cooperativeness (34, 38), and Fear of Missing Out (FOMO) – which, in the context of trading, consists in regretting not having made an investment or having sold in a specific period (27, 38, 39). Regarding FOMO, however, it should be noted that the study (49) included in Johnson et al (38) review showed that it is associated with an increased likelihood of cryptocurrency trading activity. However, it was not a predictor of problematic cryptocurrency trading.
Concerning motivational features, Arthur et al. (8) listed leisure (i.e., trading for fun and excitement) and “gambling” motives (i.e., aspiration of high payoffs and sensation seeking as the drivers). In this area, it is possible also to include anticipated regret (27), which cognitive psychology has recognized as a factor influencing many decisions (50, 51).
As additional (less analyzed) risk factors, Delfabbro et al. (27) and Roza et al. (39) highlighted the role of social learning and reinforcement, while Johnson et al. (38) suggested that young males are more at risk for problematic trading. Also, Andrade and Newall (34) pointed out market manipulation and scams as risk factors previously observed in gambling, while Håkansson et al. (37) suggested that COVID-19 (and similar crises) might increase trading frequency.