1 Introduction
The financial markets landscape is a complex tapestry woven from the intricate interplay of countless factors. Central to this complexity is the phenomenon of asset pricing anomalies, which consistently attract the attention of investors, traders, and academic researchers. These anomalies refer to patterns that contradict the Efficient Market Hypothesis (EMH), a theory positing that security prices fully reflect all available information (Fama, 1998). They present profit opportunities resulting from market inefficiencies (Bartram & Grinblatt, 2021), offering a captivating challenge to traditional financial models and theories.
One of the most prominent categories of these anomalies is the calendar or time-series anomalies, revealing systematic variations in stock returns over specific periods. Day-of-the-week effects, documented as early as 1984, have unveiled patterns where stock returns differ significantly based on the day of the week (Rogalski, 1984). This directly contradicts the EMH, which asserts that returns should follow a random walk, unaffected by the day of the week.
Following these, other temporal anomalies such as the January effect, which posits that stock returns in January are consistently higher than other months, have further strained traditional market theories (Bhardwaj & Brooks, 1992). Investigations into these anomalies have taken varied and fascinating paths, exploring even the influence of weather patterns on stock market dynamics. For example, research into the Halloween effect suggests higher returns in November compared to the rest of the year, with suggestions that this anomaly may be linked to seasonal affective disorder (Jacobsen & Visaltanachoti, 2009). Other researchers have extended this hypothesis, exploring whether weather patterns and the resulting human mood swings can significantly impact stock returns (Andrikopoulos et al., 2019; Symeonidis et al., 2010).
In the same vein, studies have examined quirky and unexpected market behaviors tied to specific events, leading to the identification of anomalies such as the Super Bowl effect (Krueger & Kennedy, 1990). This effect suggests a correlation between the outcomes of the Super Bowl and the subsequent year’s stock market performance, challenging rational explanations.
However, it’s important to underscore that calendar anomalies represent only one facet of the broad landscape of asset pricing anomalies. Research has also extensively probed into other anomalies related to firm-specific factors, market conditions, investor sentiment, and more.
For instance, anomalies like post-earnings announcement drift (PEAD) have been identified, where stock prices continue to drift in the direction of earnings surprise for several weeks post-announcement (Teoh et al., 1998). This finding contradicts the EMH’s prediction that prices should adjust immediately and fully to new information. Similarly, the post-merger performance of acquiring firms presents another anomaly, where the stocks of acquiring firms tend to underperform after mergers, challenging traditional asset pricing models (Agrawal et al., 1992).
Furthermore, certain market conditions and firm-specific characteristics have been associated with anomalous returns. For example, low share prices and high transaction costs have been linked to the January effect (Bhardwaj & Brooks, 1992). Firm characteristics such as size and book-to-market ratios have been shown to have systematic effects on returns, thus giving rise to the size and value anomalies (Agarwal & Poshakwale, 2010; Fama & French, 2008).
Cross-sectional anomalies represent another critical subset of asset pricing anomalies that have attracted significant academic scrutiny. These anomalies emphasize the role of firm-specific attributes in influencing stock returns. An example of this is the size effect, where smaller firms are observed to generate higher average returns than larger firms (Agarwal & Poshakwale, 2010). A related anomaly is the book-to-market effect, where firms with high book-to-market ratios are found to yield superior returns (Chou et al., 2010; Fama & French, 2008).
However, some scholars have criticized these anomalies as manifestations of omitted variable bias. For instance, Chou et al., 2010) suggest that the size and book-to-market anomalies are artifacts of neglected leverage risk. The authors propose that once leverage risk is appropriately accounted for, the purported anomalies dissolve. This perspective underscores the complexity of financial markets, where overlapping factors and interactions can create illusionary patterns, fostering misinterpretations and misguided investment strategies.
In the same vein, financial markets often exhibit anomalies linked to information asymmetry and investor sentiment. The PIN (Probability of Informed Trading) anomaly is one such illustration, where mergers and acquisitions announcements lead to higher informed trading, altering stock returns (Aktas et al., 2007). The underlying theory is that in periods leading up to significant corporate events, insiders and informed investors might take positions based on private information, leading to detectable price movements.
Sentiment-based anomalies, driven by behavioral biases, have also been documented extensively. For instance, the SAD (Seasonal Affective Disorder) anomaly revisited by Kelly (Kelly & Meschke, 2010) suggests that changes in mood due to seasonal variations can impact financial markets, leading to higher returns in the autumn. Taffler (Taffler, 2018) further elaborates on this aspect of emotional finance, highlighting how unconscious biases can influence investment decisions and drive market anomalies.
Moreover, anomalies like the ‘turn-of-the-year’ effect, where stocks tend to exhibit strong returns at the beginning of the year, often coincide with tax-motivated trading (Griffiths & White, 1993). This points to the influence of macro-level institutional factors in shaping stock market dynamics. Interestingly, a similar monthly pattern is observed in dividend payments, giving rise to the dividend month premium anomaly (Hartzmark & Solomon, 2013).
Alongside these, trading-related anomalies have been detected, linked to trading volumes, liquidity, and transaction costs. For instance, James (James & Edmister, 1983) reports a positive association between common stock returns, trading activity, and market value, while Daves et al., (Daves & Ehrhardt, 1993) explore the impact of liquidity and reconstitution on the value of U.S. Treasury Strips. On the other hand, Bhardwaj and Brooks,1992) illustrate how low share prices and high transaction costs could be contributing to the January effect.
The pervasive nature of anomalies across different time horizons and markets suggests potential opportunities for generating superior returns. The evolution of trading strategies like momentum, contrarian, and calendar-based trading is fundamentally rooted in these anomalies. Despite this, researchers continue to examine the persistence and exploitable nature of these anomalies.
The insights derived from these explorations not only offer potential strategies for market participants but also have significant implications for regulatory authorities, who must remain vigilant of market inefficiencies that could destabilize the financial system. It is worth noting that a subset of anomalies like the Super Bowl effect (Krueger & Kennedy, 1990) and the weather effect (Gerlach, 2010; Jacobsen & Marquering, 2008; Symeonidis et al., 2010) may appear irrational or even whimsical, they represent the far-reaching influence of human behavior on market dynamics.
In conclusion, asset pricing anomalies present a fascinating testament to the intricacies of financial markets. Their existence and persistence challenge traditional financial theories, prompting the exploration of alternative models and theories. While some anomalies may offer exploitable opportunities, market participants must exercise caution in interpreting these patterns, considering the complex interplay of multiple influencing factors. Continuous advancements in computational power and data analysis techniques promise to further deepen our understanding of these anomalies and their implications for financial markets.
In the following sections of this paper, we will delve deeper into the aforementioned topics and their relation to financial anomalies. In Section 2, we embark on an exhaustive literature review that encapsulates several dimensions of financial anomalies and the attendant puzzles.
We start by exploring the anomalies and the notion of market efficiency in Section 2.1. We then transition into the realm of behavioral finance, focusing on investor sentiment and its role in shaping market outcomes in Section 2.2.
Subsequently, in Section 2.3, we turn our attention to the calendar effects and seasonal anomalies, phenomena that have perennially intrigued researchers due to their recurrent patterns that seem to defy traditional market theories.
In Section 2.4, we delve into the relationship between risk factors and asset pricing, scrutinizing the impact of these factors on market anomalies.
We then move on to examine how information flow influences trading strategies and its potential role in the formation of anomalies in Section 2.5.
Finally, in Section 3, we consolidate our findings and offer concluding remarks, providing a recap of the significant observations made throughout the paper, outlining the research gaps, and charting potential directions for future investigations in this fascinating area of financial economics. Our overarching aim is to illuminate the complexities and subtleties of financial anomalies, thereby contributing to the ongoing discourse in this vibrant field of research.
2 Literature review
2.1 Anomalies and efficiency of financial markets
The notion of market efficiency stands as a cornerstone in the financial economics literature, offering an intellectual paradigm that has fueled much of the scholarly inquiry and theoretical developments in this field. According to the Efficient Market Hypothesis (EMH) proposed by Fama (1998), financial markets should inherently reflect all publicly available information, and therefore, it is impossible to consistently achieve returns higher than the market average. Yet, this core premise of financial economics has been challenged and complicated by the empirical identification of so-called ‘anomalies’.
Anomalies, as defined by Fama and French (2008), represent empirical results that appear to be inconsistent with maintained theories of asset pricing behavior. They indicate patterns in returns that seem to contradict the EMH, suggesting possibilities of earning excess returns by capitalizing on these detected inconsistencies. These anomalies take various forms, encompassing calendar effects, such as the day-of-the-week effect (Rogalski, 1984), the January effect (Bhardwaj & Brooks, 1992), the Halloween effect (Jacobsen & Visaltanachoti, 2009), and even more novel phenomena, like the weather effect (Andrikopoulos et al., 2019; Gerlach, 2010; Jacobsen & Marquering, 2008; Symeonidis et al., 2010).
The persistence and breadth of these anomalies have been substantial, with more than 100 documented in the literature (Jacobs, 2015). Yet, a key question arises: if these anomalies allow for consistent earning of excess returns, why aren’t they arbitraged away by rational investors, thereby restoring market efficiency? Several explanations have been posited, focusing on issues like transaction costs (Bhardwaj & Brooks, 1992; Fortin, 1990) and investor psychology (Loewenstein & Willard, 2006); Taffler, 2018).
A closer look at the anomalies can be divided into broad categories: firm-specific anomalies, event-based anomalies, calendar and time-based anomalies, and investor sentiment anomalies.
Firm-specific anomalies like the size and book-to-market ratio anomalies suggest that smaller firms and those with high book-to-market ratios tend to earn higher returns than predicted by the Capital Asset Pricing Model (CAPM) (Agrawal et al., 1992; Fama, 1998). Yet, these anomalies might be linked to omitted risk factors like leverage risk, rather than being true departures from market efficiency (Agarwal & Poshakwale, 2010; Chou et al.,, 2010).
Event-based anomalies highlight patterns surrounding corporate events, such as mergers and acquisitions (Agrawal et al., 1992; Aktas et al., 2007), seasoned equity offerings (Teoh et al., 1998), and earnings announcements (Battalio & Mendenhall, 2011). Some of these patterns could be attributed to information asymmetries and investor sentiment (Gençay et al., 2010; Lam et al., 2012; Shu & Chang, 2015).
Calendar and time-based anomalies cover patterns linked to specific periods, such as the day of the week (Rogalski, 1984), month of the year (Bhardwaj & Brooks, 1992; Hartzmark & Solomon, 2013), holiday effects (Jacobsen & Visaltanachoti, 2009), and even the daylight-saving anomaly (Gerlach, 2010); Gregory-allen et al., 2010). These effects could be driven by changes in investor risk preferences, institutional practices, or information dissemination over time (Kolb & Gay, 1985; Ma et al., 1988; Ülkü & Andonov, 2016).
Investor sentiment anomalies are associated with changes in investor moods and behaviors. Studies have demonstrated the impact of sentiment on stock returns and volatility (Bird & Casavecchia, 2007; Ciccone, 2011; Hirshleifer, 2020; Kelly & Meschke, 2010). Furthermore, emerging research has pointed out that these anomalies are not confined to traditional markets, extending to the realm of digital assets like cryptocurrencies (Qadan et al., 2022).
Recent studies highlight that these anomalies might not persist when considering trading costs, market frictions, and data snooping biases (Hsu et al., 2016; Tobek & Hronec, 2021; Zaremba et al., 2020). A few anomalies have also been explained by time-varying risk premiums, indicating that they may not necessarily contradict the EMH (Alti & Tetlock, 2014; Zhang, 2006). Furthermore, it has been observed that some anomalies diminish as markets mature, indicating a potential self-correcting mechanism in financial markets (Jacobs, 2016).
In conclusion, the existence of financial market anomalies presents a complex interplay between theoretical conceptions of market efficiency and empirical observations of apparent inefficiencies. The anomalies and the continual debate about their persistence and exploitations underscore the dynamic nature of financial markets and the evolving understanding of market efficiency. Future research directions should continue to investigate these anomalies, with a keen focus on separating truly exploitable inefficiencies from phenomena resulting from omitted variables, data biases, or changing risk premiums.
2.2 Behavioral finance and investor sentiment
Behavioral finance seeks to explain why investors sometimes act in ways that are inconsistent with classical economic theory. It draws on insights from psychology to explore how cognitive biases can impact financial decisions and market outcomes. Fama’s work on market efficiency, long-term returns, and behavioral finance provides a foundation for our understanding of the relationship between investor sentiment and market behavior (Fama, 1998).
One of the key ideas in behavioral finance is that investors may not always act rationally or in their best interests. This is evident in phenomena such as earnings management, where companies manipulate their earnings reports to meet investor expectations, resulting in the underperformance of seasoned equity offerings (Teoh et al., 1998). The principle of market efficiency holds that markets should fully incorporate all available information into prices, but Fama has dissected various anomalies that challenge this notion (Fama & French, 2008). One of these is information uncertainty, where stock returns can be influenced by the level of uncertainty about a company’s future prospects (Zhang, 2006).
Behavioral finance also extends to phenomena observed post-merger, where the performance of acquiring firms often falls below expectations, despite market efficiency dictating that any potential gains should already be priced in (Agrawal et al., 1992). This phenomenon also appears in the realm of mutual funds, where Hendricks (1993) noted a persistence in the short-run performance of top-performing funds, suggesting that investors may be influenced by recent performance when choosing funds (Hendricks et al., 1993).
Market anomalies provide further examples of departures from market efficiency that may be influenced by investor sentiment. For example, the day-of-the-week effect, where returns on certain days of the week are systematically higher or lower than on others, may reflect investor psychology rather than changes in underlying value (Rogalski, 1984). The January effect is another such anomaly, where low share price, transaction costs, and bid-ask bias may combine to generate abnormal returns in January (Bhardwaj & Brooks, 1992).
Jacobs (2015) attempted to explain the dynamics of 100 anomalies and found that they were influenced by various factors including time, market conditions, and investor sentiment (Jacobs, 2015). Jacobsen and Marquering (2008) found that weather influences investor sentiment and in turn, stock returns, illustrating how psychology can affect financial markets (Jacobsen & Marquering, 2008).
The influences of investor sentiment are not limited to traditional equity markets. For example, in the world of cryptocurrencies, Qadan (2022) identified several seasonal and calendar effects that affect the price efficiency of cryptocurrencies, suggesting the influence of investor sentiment and behavior in these markets as well (Qadan et al., 2022).
Investor sentiment is also a crucial factor in explaining asset pricing (Colacito & Croce, 2013). Asset prices often deviate from their fundamental values due to biased beliefs and investor sentiment, as explored by Alti (2014) (Alti & Tetlock, 2014). A model proposed by Lam (2012) even highlighted the role of pseudo-Bayesian behavior in understanding financial anomalies and investor behavior.
One common manifestation of investor sentiment is overreaction, where investors respond too strongly to new information, leading to price swings that are larger than warranted by the fundamentals (Davidson & Dutia, 1989). Contrarian investing, a strategy of buying assets that have recently performed poorly and selling those that have performed well, takes advantage of such overreactions (Chin et al., 2002).
Furthermore, certain calendar and seasonal effects, like the ‘Halloween effect’ or the ‘turn of the year effect’, which have been observed in U.S sectors and intraday studies respectively, can be attributed to behavioral aspects of investors (Griffiths & White, 1993; Jacobsen & Visaltanachoti, 2009). Even macroeconomic news and events like the Super Bowl have been found to have impacts on stock market anomalies, further reinforcing the importance of sentiment in understanding market behavior (Gerlach, 2007; Krueger & Kennedy, 1990).
Emotional factors also play a significant role in investment decisions. Taffler (2018) argued that unconscious emotions often guide financial decisions, which leads to certain investment outcomes (Taffler, 2018). Emotional finance suggests that financial health indicators can significantly vary for value and growth stocks based on investor sentiment (Bird & Casavecchia, 2007).
The complexity of the relationship between investor sentiment and market outcomes is further highlighted by the role of external factors like weather (Symeonidis et al., 2010). Weather effects can cause significant variations in stock market volatility, suggesting that investors’ mood states influenced by weather conditions can lead to observable effects in financial markets (Gerlach, 2010).
Moreover, research has shown that the sentiment of retail investors can play a key role in inflating stock price bubbles, revealing the potential downside of sentiment-driven trading (Henker & Henker, 2010). Likewise, the way investors perceive risk and react to it can significantly influence the profitability of certain trading strategies. As an example, investors’ biases may lead to poor timing in investing, causing underperformance even when following successful strategies (Hsu et al., 2016).
In summary, the field of behavioral finance provides a rich perspective on how investor sentiment can drive financial market outcomes. Whether it is earnings management, post-merger performance, market anomalies, or asset pricing, investor sentiment appears to play a critical role. This sentiment, shaped by a range of factors from psychological biases to external events and even the weather, leads to a variety of behaviors that can cause markets to depart from the predictions of classical economic theory. Understanding these behaviors and their implications is key to navigating the complexity of financial markets.
2.3 Calendar effects and seasonal anomalies
The efficient market hypothesis (EMH), initially posited by (Fama, 1998), suggests that financial markets fully reflect all available information, thus implying that generating abnormal returns on a consistent basis would be an improbable task for investors. However, various observed market irregularities, or "anomalies", often related to calendar effects and seasonality, challenge this fundamental premise.
Perhaps one of the most extensively documented of these phenomena is the "January effect". This effect refers to the tendency of stocks, especially small-cap stocks, to outperform the broader market in January (Bhardwaj & Brooks, 1992). Explanations for the January effect have revolved around tax-loss selling, institutional investors’ "window dressing", and liquidity effects (Griffiths & White, 1993). Tax-loss selling refers to the practice of selling securities at a loss to offset a capital gains tax liability, which typically occurs at the end of the year and may contribute to lower prices in December and a rebound in January. Window dressing is the strategy used by fund managers near the year-end to improve the appearance of the fund’s performance before presenting it to clients or shareholders. Liquidity effects, meanwhile, refer to the impact of changes in market liquidity on asset prices and returns.
Another significant calendar effect is the ‘weekend effect’, or the tendency for returns to be lower on Mondays than on other days of the week (Rogalski, 1984). While a multitude of theories have been put forth to explain this phenomenon, the most prominent revolve around settlement procedures, the timing of corporate announcements, and the psychology of traders (Tong, 2000).
Contrasting with the more universal January and weekend effects, certain calendar anomalies are specific to particular geographic regions. For instance, the ‘Halloween effect’ or ‘Sell in May and go away’ strategy is observed primarily in Western countries. It refers to the observed tendency of stocks to perform better between November and April compared to the period from May to October (Jacobsen & Visaltanachoti, 2009). Despite the persistent existence of this anomaly across several markets, conventional risk-based theories have yet to provide a convincing explanation (Jacobsen & Marquering, 2008).
Beyond the traditional asset markets, there is evidence to suggest that calendar effects also pervade the world of cryptocurrencies. (Qadan et al., 2022) found substantial seasonality in cryptocurrency prices, which they attributed to cycles in investor attention. However, the exact mechanisms behind such phenomena in this burgeoning domain remain a topic of ongoing research.
As the field of behavioral finance has grown, researchers have attempted to explain calendar effects through the lens of investor sentiment and behavioral biases. (Kelly & Meschke, 2010) provided evidence that seasonal affective disorder (SAD), a psychological condition that fluctuates with seasonal changes, significantly influences stock returns. Similarly, research has found that weather conditions and changes in daylight saving time can affect stock market volatility and investor sentiment, thereby contributing to the presence of seasonal anomalies (Gerlach, 2010; Symeonidis et al., 2010).
However, it is worth noting that the existence of calendar effects is not equivalent to the existence of profitable trading strategies. Practical barriers such as transaction costs can significantly reduce, if not entirely eliminate, the profitability of trading strategies built on these anomalies (Teoh et al., 1998; Duran & Bommarito, 2011). Moreover, anomalies often weaken or even disappear after they have been publicly identified, potentially due to market participants’ adaptive behavior (Fama & French, 2008; Jacobs, 2016).
While the existence of calendar effects appears to contradict the EMH, these phenomena might be better understood as symptoms of the larger complexities and potential inefficiencies present in financial markets. (Zhang, 2006) suggests that information uncertainty is a key factor in explaining stock market anomalies. This perspective postulates that anomalies are a result of investors’ responses to uncertainty and their preferences for stocks with less ambiguous information.
It is also important to note that despite the wealth of literature supporting the existence of calendar effects, there is also a body of work questioning their validity. Some researchers argue that these effects are statistical artifacts that emerge as a result of data snooping bias or methodological flaws in the employed statistical tests. These criticisms call for a more rigorous analysis of calendar effects using robust statistical techniques.
In summary, the study of calendar effects and seasonal anomalies uncovers intriguing patterns in financial markets and provides important insights into market dynamics, investor behavior, and the limits of market efficiency. Although these anomalies seem to contradict the EMH, they might better be viewed as evidence of the intricacies and potential inefficiencies of financial markets. Understanding the mechanisms behind these anomalies, whether they are rooted in investor psychology, institutional practices, or information dynamics, can potentially offer valuable insights into the functioning of financial markets and the development of effective investment strategies. Nevertheless, the practical implications of these anomalies for investment strategies should be carefully considered in light of transaction costs, adaptive market behavior, and the risk of data snooping bias.
2.4 Risk factors and asset pricing
Understanding risk factors and their impact on asset pricing is fundamental in financial economics. Asset pricing models have evolved over time from the single-factor model, like the Capital Asset Pricing Model (CAPM), to multifactor models that incorporate various dimensions of risk. The implication of these factors for market anomalies is an area of considerable research.
Fama and French (2008) made significant contributions in this regard, dissecting numerous anomalies in terms of risk factors. Fama challenges the notion that anomalies provide exploitable returns, suggesting instead that they can be understood and explained through risk factors. By extension, this implies that a thorough understanding of risk factors can potentially mitigate some of the uncertainty around asset pricing and inform better investment decisions.
Research by Agrawal et al. (1992) further illustrates the importance of risk factors in explaining anomalies. In a re-examination of the post-merger performance of acquiring firms, Agrawal found that traditional risk factors alone could not fully account for the observed underperformance. This insight underscores the complexity of asset pricing and the importance of considering a broad range of risk factors.
Building upon this research, Gultekin and Gultekin (1987) scrutinized the relationship between stock return anomalies and the arbitrage pricing theory (APT). His work emphasized the possible existence of unobserved risk factors that may explain such anomalies. These findings lend support to the notion that asset pricing is influenced by a wide array of factors, some of which may not be captured in conventional models.
Investigating risk factors further, Zhang (2006) proposed information uncertainty as a significant determinant of stock returns. He argued that an increase in uncertainty could lead to higher required returns and subsequently lower prices, manifesting as an anomaly.
In the context of globalized financial markets, research by Colacito and Croce (2013) highlighted the importance of incorporating global risk factors in asset pricing models. They introduced a recursive preference model for international asset pricing, thereby expanding our understanding of risk factors beyond domestic or regional contexts. Similarly, Bartram and Grinblatt (2021) provided a comprehensive examination of global market inefficiencies, emphasizing the relevance of global risk factors in understanding asset pricing and anomalies.
Aktas et al. (2007) examined the ‘PIN’ anomaly around merger and acquisition announcements, revealing a different perspective on risk factors. The "PIN" anomaly, which pertains to private information-based trading, illustrates how information asymmetry can influence asset pricing. This highlights the importance of considering market microstructure and information dynamics when examining risk factors.
Moreover, Alti and Tetlock (2014) presented a structural approach that incorporated biased beliefs as a potential risk factor. Alti argued that such beliefs could lead to asset mispricing, ultimately manifesting as market anomalies. His work indicates that understanding investor psychology and behavioural biases is crucial in assessing risk factors and their role in asset pricing.
While Daves and Ehrhardt (1993) focused on liquidity as a critical risk factor, particularly for U.S. Treasury strips, Ammann and Verhofen (2012) proposed an alternative three-factor model for international markets. These studies demonstrate the variety and evolution of risk factors considered in asset pricing research.
Adding another dimension to risk factors, Chou et al. (2010) and Agarwal and Poshakwale (2010) argued that traditional risk factors might not fully explain the size and book-to-market anomalies. They proposed that relative leverage and relative distress are also significant factors.
Research by Henker and Henker (2010) and Hirshleifer (2020) emphasizes the role of investor behavior and social transmission bias, respectively, in understanding stock market anomalies. These contributions highlight the need to consider non-traditional risk factors, such as investor sentiment and social influences, when analyzing asset pricing.
Consequently, the literature provides a rich array of perspectives on risk factors and asset pricing. Traditional measures like size and value, as well as less observable factors such as information uncertainty, biased beliefs, and global market risks, are all relevant. Incorporating these diverse risk factors can improve our understanding of financial markets, asset pricing, and the anomalies observed therein.
2.5 Information flow and trading strategy
The efficient market hypothesis argues that market prices should reflect all available information at any given moment, making it impossible to consistently outperform the market (Fama, 1998). In spite of this, numerous studies have documented patterns in stock returns that seem to violate market efficiency, known as anomalies.
The existence of such anomalies implies that information flow is not perfect and trading strategies may exploit these patterns. For instance, a well-documented anomaly is the underperformance of firms following seasoned equity offerings, which might suggest that insiders have access to private information about their firms (Teoh et al., 1998). Other anomalies include patterns in stock returns that are related to firm characteristics such as size and book-to-market ratio (Fama & French, 2008). Moreover, evidence suggests that stocks with higher information uncertainty earn higher returns (Zhang, 2006).
There are many other types of anomalies which have been observed and analyzed over the years. Mergers and acquisitions often lead to underperformance for acquiring firms, pointing towards possible issues in deal pricing (Agrawal et al., 1992). Additionally, there is a persistence of short-run relative performance in mutual funds (Hendricks et al., 1993).
Market returns also exhibit certain calendar effects. For instance, the day-of-the-week effect shows consistent differences in returns depending on the trading day, with Monday returns being lower than returns for the other days (Rogalski, 1984). Similarly, the January effect indicates higher returns in January, especially for low-priced stocks (Bhardwaj & Brooks, 1992).
In terms of explaining these anomalies, research has proposed various factors. Jacobs (2015) argues that market frictions, sentiment-driven mispricing, and varying risk premiums could be behind the dynamics of many anomalies. Other researchers have pointed to the influence of factors such as weather (Jacobsen & Marquering, 2008; Symeonidis et al., 2010), international asset pricing (Colacito & Croce, 2013), the timing of mergers and acquisitions (Aktas et al., 2007), and asymmetry of information flow between volatilities (Gençay et al., 2010).
The market’s maturity may also influence mispricing, as suggested by Jacobs (2016) who observed that young firms are more prone to mispricing. Certain seasonal effects, such as the Halloween effect, have been shown to influence returns in various sectors of the U.S. economy (Jacobsen & Visaltanachoti, 2009).
The role of investor sentiment and behavioral factors in the creation of anomalies has also been explored. For instance, Hirshleifer (2020) demonstrates the existence of social transmission bias in economics and finance, where people’s judgements and behaviors influence each other and impact the market. Kelly and Meschke (2010) re-examines the Seasonal Affective Disorder (SAD) anomaly, noting that sentiment does indeed influence stock returns.
Other anomalies related to calendar effects include the tax-induced trading and the turn-of-the-year anomaly (Griffiths & White, 1993), the dividend month premium (Hartzmark & Solomon, 2013), and the relation between common stock returns, trading activity, and market value (James & Edmister, 1983).
Beliefs and biases of investors have been shown to impact asset prices and investment decisions (Alti & Tetlock, 2014), and the liquidity and reconstitution of U.S. Treasury Strips can significantly affect their value (and Ehrhardt (1993). Moreover, investor behavior can be modeled to help explain financial anomalies, as shown by Lam (2012).
These anomalies persist on a global scale as well. There is international evidence of weekend anomalies (Tong, 2000), and the computation of returns in tests of the stock market overreaction hypothesis has been examined (Dissanaike, 1994).
Sentiment and financial health indicators are also important for value and growth stocks, as demonstrated in the European experience (Bird & Casavecchia, 2007). Furthermore, emotional finance explores the role of unconscious factors in investment decision-making (Taffler, 2018).
One peculiar anomaly is the so-called ‘Super Bowl Predictor’, which suggests that the stock market’s performance can be predicted by the outcome of the Super Bowl (Krueger & Kennedy, 1990).
There are also findings related to the effects of macroeconomic news and stock market calendar and weather anomalies (Gerlach, 2007), the impact of investor sentiment on financial market volatility (Shu & Chang, 2015), the relationship between investor optimism, false hopes, and the January effect (Ciccone, 2011), and the limits of investor behavior (Loewenstein & Willard, 2006).
Finally, despite the vast number of identified anomalies, recent research argues for global market inefficiencies and discusses their persistence and potential explanations (Bartram & Grinblatt, 2021).
In summary, the flow of information in the market is complex and does not always lead to perfect efficiency. Trading strategies that take into account various anomalies could potentially achieve above-average returns. However, it is important to note that these strategies often involve additional risks and complexities. Moreover, as our understanding of these anomalies improves, market participants may adjust their behavior, which can lead to the disappearance of these anomalies over time (Zaremba et al., 2020). Therefore, careful research and analysis are essential when attempting to exploit these patterns.
3 Limitations and implications for future research
Despite the wide-ranging discussion and examination of various market anomalies in the present review, certain limitations should be acknowledged. These limitations, in turn, open up avenues for future research, further contributing to our understanding of these complex phenomena.
3.1 Limitations
The following are some of the key limitations of the current review:
Scope of Literature: Although we have attempted to provide a comprehensive overview of market anomalies, due to the vastness of the subject, not all anomalies have been included in the review. Therefore, our discussion does not represent the entire spectrum of financial anomalies.
Cultural and Geographic Differences: Most of the studies reviewed in this paper are predominantly based on western markets. Anomalies in emerging and non-western markets have not been extensively discussed. The cultural and institutional differences in these markets can give rise to unique anomalies that are not seen in western markets.
Dynamics of Cryptocurrency Markets: While we touched on the efficiency of cryptocurrency markets, a detailed examination of anomalies in these relatively new and rapidly evolving markets was beyond the scope of this review.
Methodological Limitations: The majority of the studies discussed in this review utilized traditional statistical methods. The use of machine learning and advanced data analytics techniques is limited in the current literature on financial anomalies.
Exogenous Shocks and Anomalies: The impact of exogenous shocks (like global pandemics, political events, and natural disasters) on market anomalies was not explored extensively in this review. These events can significantly influence market dynamics, giving rise to unique anomalies.
3.2 Future research
Building on the limitations mentioned above, the following are some promising directions for future research:
Expanding the Scope of Literature: Future studies could expand the scope of market anomalies by including less explored anomalies. This will enrich the current body of literature and provide new perspectives on financial anomalies.
Non-western and Emerging Markets: Given the potential for unique anomalies in non-western and emerging markets, future research could explore these markets in depth. This will allow for a more comprehensive understanding of market anomalies that includes cultural and institutional diversity.
Cryptocurrency Market Anomalies: As cryptocurrency markets continue to grow and evolve, future studies could focus on identifying and understanding the anomalies unique to these markets. This could pave the way for new trading strategies and regulatory interventions in these markets.
Advanced Data Analytics Techniques: The use of machine learning and advanced data analytics techniques can provide deeper insights into financial anomalies. Future research could focus on developing and applying these advanced techniques in the study of market anomalies.
Exogenous Shocks and Anomalies: Further research is needed to understand the impact of exogenous shocks on market anomalies. This could help in predicting the emergence of new anomalies and improving our ability to manage risks associated with these shocks.
4 Conclusion
In conclusion, the extensive body of research on market anomalies and puzzles provides a multifaceted perspective on market efficiency and the influences affecting stock market returns. Notably, the studies examined in this literature review cover a broad spectrum of market anomalies and provide evidence of their persistence across various domains of the financial market.
These anomalies range from long-term return anomalies (Fama (1998), earnings management effects on seasoned equity offerings (Teoh et al., 1998), anomalies associated with net stock issues and accruals (Fama & French, 2008), information uncertainty and price continuation anomalies (Zhang, 2006), to post-merger performance (Agrawal et al., 1992), performance persistence in mutual funds (Hendricks et al., 1993), day-of-the-week returns (Rogalski, 1984), the January effect (Bhardwaj & Brooks, 1992), and even weather-induced mood shifts on stock returns (Jacobsen & Marquering, 2008).
Despite the Efficient Market Hypothesis’ argument that these anomalies may be chance results, several studies provide evidence of their persistence and potential influences on market dynamics. The variation in findings across these studies suggests that these anomalies may be a regular feature of financial markets rather than aberrations. Furthermore, it indicates the potential impact of factors like information uncertainty, behavioral biases, and corporate strategies on these anomalies.
Looking forward, it’s evident that more research is needed to understand the true nature and causes of these anomalies. Future work may involve developing new methodologies and theoretical models to better comprehend these phenomena, as well as exploring other potential influencing factors.
More specifically, while existing research has identified and analyzed many anomalies, new data sources and computational methods could enable the identification of previously undetected anomalies. At the same time, the development of more robust and comprehensive theoretical models could help improve our understanding of these anomalies’ origins and the mechanisms driving them.
Finally, the influence of other factors, such as technological innovations, market regulations, and macroeconomic variables, should be examined in greater depth. The complex and interconnected nature of financial markets suggests that a broad range of factors could potentially contribute to these anomalies and understanding these influences could provide valuable insights into market dynamics and efficiency.
