engleski

Downside risk measures during times of crisis: evidence from Croatian stock market

Concern of investors about losses and poorer portfolio performance has roots in behavioural finance. As investors are more sensitive to losses compared to gains in their utility functions, they will demand additional compensation for holding stocks with high sensitivities to downside market movements. Besides the financial crisis in 2008, which has been a significant event for almost all research, nowadays investors are facing challenging times given the COVID-19 pandemic. Instead of using a common measure of systematic risk, in order to test whether aversion to loss is rewarded, especially during times of financial crisis and/or pandemic, we use two downside risk measures: downside beta and tail beta. Tail beta is similar to downside beta, however, it focuses further on the left-tail of the return distribution. As a reference point, different percentiles can be used. In this research, the 10th percentile of the distribution of the market’s excess return is used as a reference point for downside deviations. The literature suggests a significantly positive relation between systematic downside risk and equity returns. In this research, we re-examine this evidence in a less liquid and undeveloped Croatian stock market, with special emphasis on the period during the financial crisis and the first wave of the COVID-19 pandemic. Weekly excess returns of 67 stocks that were included in the CROBEX index, albeit not continuously, in the period from January 2005 till September 2020 are used. Both downside risk measures are calculated for multiple samples based on a 52-week rolling window of returns. Stocks were included in the samples if they had at least 26 weeks of trading (in the 52-week period) and were traded in the last week of the sample. Univariate portfolio analysis using equally weighted returns is used where each portfolio was held for 1-week out-of-the sample resulting in time series of estimates of weekly returns calculated as an arithmetic average return. Obtained data is further used to calculate the long-short portfolios for each downside risk measure separately. Corresponding risk premium is calculated so that the zero-cost strategy could be tested. Research findings show that only tail beta seems to be able to capture risk premium associated with going long in the portfolio implying the highest loss aversion and going short in the one with the lowest loss aversion. However, we do not find the detected risk premium to be statistically significant. Results hold for both the financial crisis and pandemic period as opposed to the results for downside beta where risk premium is present only in the period concerning the first wave of the pandemic. In the case of downside beta, for almost the whole analysed period results reveal discount suggesting irrational negative relationship – an increase in loss aversion causes a decrease in return. The results for downside beta are not surprising entirely given the fact that earlier research for the Croatian stock market did not provide evidence of positive relation between downside risk and return. Nevertheless, capturing positive risk premium in less liquid and undeveloped Croatian stock market when tail beta is used seems to provide good grounds for further analysis.

downside risk ; financial crisis ; COVID pandemic ; illiquid and undeveloped equity market

nije evidentirano