Abstract. Cross-sectional momentum has been one of the most persistent return anomalies to provide high levels of abnormal returns in most markets and asset classes over long time periods. While scientific literature is still inconclusive on the core cause of the anomaly, a significant amount of research has been published confirming the existence of abnormal returns related to the phenomenon.

Momentum also has its downsides, or its moments, as previous researchers have expressed it. The strategy occasionally experiences large streaks of negative returns, which can wipe out a significant part of the value of momentum portfolios within only a few months. These momentum crashes can take decades to recover from for the strategy and are an important consideration for both researchers and investors seeking to profit from the abnormal returns or diversification benefits that the strategy has provided.

As momentum crashes have been found to happen during rebounding markets after market crashes, this thesis studies the momentum crash following the recent market downturn caused by the COVID-19 pandemic and takes a modern look at both momentum and momentum crashes in the US stock market by studying three different momentum strategies formed in previous research with data from January 1990 to March 2022. It also introduces a risk-managed momentum strategy that scales the weights of a traditional 1st decile momentum strategy based on the lagged value of the VIX index compared to its ten-year simple rolling average, up to the previous month.

The results show that momentum portfolios had large negative returns in the year following the market downturn caused by the COVID-19 crisis at the start of the year 2020. The negative returns for all studied momentum portfolios were caused by the highly positive returns of the shorted portfolio in the strategy during a market recovery period, similar to prior research results on momentum crashes. The Vix-based risk-managed momentum strategy successfully lowered the effects of momentum crashes compared to its base strategy and provided statistically significant abnormal returns and higher Sharpe ratios compared to the three traditional momentum portfolios throughout the studied time period. Successfully using a lagged value of a market-based index to predict the volatility of momentum has both practical implications, as well as possibly interesting implications for future research on momentum.

The traditional 1st decile momentum strategy saw significantly larger losses during momentum crashes compared to 3rd decile momentum strategies; however, the 1st decile portfolio still has higher mean returns than 3rd decile momentum portfolios over a long time period. This suggests that managing the downside risk of aggressive momentum strategies has been extremely important during the 21st century to maximize the benefits of the return anomaly