The COVID-19 pandemic shook financial markets across the globe, leaving investors grappling with uncertainty. One intriguing area of study during this period has been the behavior of investors in stock markets, particularly the concept of herding. Herding occurs when investors follow the crowd instead of relying on their information and strategies, often leading to irrational market trends.
In the Romanian stock market, herding behavior had been a notable characteristic before the pandemic. But how did the crisis affect this behavior? Did the pandemic intensify the tendency to follow the crowd, or did it push investors to make more independent decisions? This article summarizes a study that delves into these questions, exploring investor behavior on the Bucharest Stock Exchange (BVB) before and during the COVID-19 pandemic.
What is Herding and Why Does it Matter?
Herding is a key concept in behavioral finance, where investors mimic the actions of others, often ignoring their research or the market’s fundamentals. This behavior is particularly common during periods of uncertainty or market volatility, as investors seek comfort in following the majority. While herding can provide short-term stability, it often leads to market inefficiencies, price bubbles, and eventually, crashes.
The Romanian stock market, being a relatively small and less liquid market compared to its Western European counterparts, is especially vulnerable to such behavioral biases. Low transparency and high levels of state ownership among listed companies often make it challenging for investors to make well-informed, independent decisions. As a result, the market has historically been prone to herding behavior.
Investigating Herding Behavior During COVID-19
To understand how the pandemic impacted investor behavior in Romania, the study analyzed stock data from 30 of the largest companies listed on the BVB. The data covered the period from January 2, 2018, to January 28, 2022. This period was split into two phases: pre-COVID (January 2018 to February 2020) and during-COVID (February 2020 to January 2022). The BET Index, the primary stock index of the Bucharest Stock Exchange, was used as the market benchmark.
Using the Cross-Sectional Absolute Deviation (CSAD) method—a statistical model introduced by Chang et al. (2000) to detect herding—the study measured the dispersion of individual stock returns from the overall market return. Under normal market conditions, the dispersion (CSAD) should increase as market returns rise, because stocks react differently to market information based on their fundamentals. However, if herding is present, this dispersion will decrease, indicating that investors are moving together, ignoring their information, and following the market trend.
The CSAD value is calculated using the following formula:
Where: Ri,t is the return of stock i on day t, Rm,t is the aggregate market return on day t, N is the number of stocks in the sample.
Once the CSAD values were calculated, they were used in a linear regression model to explore the relationship between market returns and stock return dispersion. The goal is to identify whether this relationship is non-linear, which would indicate herding behavior.
The regression equation used in this analysis was:
Where: CSADt is the cross-sectional absolute deviation of returns at time t and the dependent variable, |Rm,t| represents the absolute market return term, R2m,t is the square of the market return term which detects herding, ϵi,t is the error term.
The key variable of interest here is β2. If β2 is negative and statistically significant, it suggests a non-linear relationship between the market return and the dispersion of individual stock returns. This is evidence of herding behavior, where investors are moving in the same direction and minimizing the spread between individual stock returns.
What the Data Revealed
The first step in the analysis was to compare market behavior before and during the pandemic. Descriptive statistics showed that market volatility increased significantly during the COVID-19 crisis. The table below summarizes key statistics from the study:
Table 1. Descriptive statistics:
The average market returns were higher during the pandemic, but so was the volatility, indicating more unpredictable price movements. Interestingly, the CSAD values, which measure the dispersion of stock returns, also increased during the pandemic. This suggested that instead of moving together with the market, stocks were more likely to behave independently.
Evidence of Herding Before COVID-19
Before the pandemic, the study found clear evidence of herding behavior in the Romanian stock market. Investors were following market trends rather than acting on independent information. This is reflected in a statistically significant negative coefficient β2 in the CSAD model (see Table 2).
In simpler terms, before COVID-19, Romanian investors were more likely to mimic each other’s behavior, leading to less variation in individual stock performance. This behavior is often driven by fear of missing out on profits or the desire to conform to the majority.
A Shift During the Pandemic
Surprisingly, the pandemic marked a significant shift in investor behavior. The study found no significant evidence of herding during the COVID-19 period. While market volatility increased, investors seemed to move away from following the crowd and began making more independent, informed decisions. This is somewhat counterintuitive, as one might expect the uncertainty of a global crisis to push investors towards herding for safety.
Table 2. Estimated regression coefficients for the subsamples:
Several factors could explain why herding diminished during the pandemic. The global crisis caused market volatility to soar, making it more challenging for investors to rely on general market trends. Additionally, the sheer level of uncertainty and the constant influx of new information may have rendered herd behavior ineffective.
Another explanation could be the rise in informed trading during the pandemic. As information asymmetry grew, some investors likely acted on private information rather than following the crowd, thus reducing the overall level of herding behavior in the market.
Conclusion: A More Independent Market?
While it is often assumed that crises like the COVID-19 pandemic would heighten herding behavior, the Romanian stock market demonstrated the opposite. Investors appeared to pivot toward more independent strategies during the pandemic, signaling a behavioral shift that contrasts with the pre-pandemic period.
The Romanian stock market’s response to the pandemic provides a compelling case study in investor behavior during crises. As markets continue to evolve in response to global challenges, understanding the role of psychological biases like herding will remain crucial for investors and policymakers alike.
Comments