Daniele Grassi, chief executive of Axyon AI, looks at market performance data and considers what the future holds for asset managers, which he expects will include the usage of more AI-based systems to manage through market anomalies.
Covid-19 continues to take its toll on financial markets, presenting major challenges for asset managers, from active funds to passive investments. The implications have been significant and central banks have now injected close to $100 billion to prop up investment funds hit by the market turmoil, raising questions about the systemic risks posed by the sector.
With the effects of the crisis likely to be felt for several years ahead, how can asset managers adapt to this ‘new normal’ and begin to prepare for future challenges?
Given the scale of Covid-19 and its impact on the global economy, asset managers were always likely to face high levels of volatility and challenging market conditions. However, the immediate reaction to the crisis produced varying results for different investment strategies and approaches.
Passive funds tracking equity and bond indexes, which have proved highly popular with investors in recent years, have been exposed to the full extent of market volatility from the very beginning of the crisis. These funds suffered significant losses in value throughout February and March, but they weren’t the only immediate losers.
Funds managed with traditional quantitative methods have experienced a similar struggle. These funds typically work on the assumption that patterns can be found in historical data and then used to inform investment choices. However, the current crisis has left most of these funds badly equipped to make decisions able to mitigate risk and provide positive performance. Covid-19 has been referred to by some as a ‘black swan’ event, pushing a large share of quantitative models to ingest chaotic market data, thus providing inaccurate predictive views of market changes.
Active equity funds initially performed well in late February and early March, with 89% of active UK funds outperforming, according to research from FE Fund Info. However, as the Covid-19 crisis has worsened, managers and funds from all sides of the investment spectrum have felt the full force of very challenging market conditions.
Our models at Axyon have picked up some interesting insights which have shown how profound the impact of the virus has been for the markets. In the commodities sector, for example, we have seen extremely high market anomaly scores since mid-February. This suggested a complicated environment coming up for traditional and quantitative portfolio managers to generate alpha.
In the European equity space during recent weeks, we have seen our predicting models focusing more and more on risk and value factors while reducing the importance of momentum from 50% to around 30%. Additionally, there has been a significant sector rotation in the models’ picks. While during March we saw the models showing a preference for information technology stocks, this weighting has been reduced following recent market gains. Instead, we have seen a preference being adopted on a few selected financial stocks, where a homogenous bearish view had previously been taken on the sector.
What could the future hold?
Just like other industries, asset managers are already facing a new normal as a result of the Covid-19 crisis. Fund managers are taking a retrospective look at their strategies to re-evaluate their portfolios and ensure they are prepared for the months ahead, with high levels of volatility going forward.
In general, we expect to see more firms looking beyond traditional risk management models and developing AI-based systems capable to better detect and handle market anomalies. This approach will ideally integrate deep learning - the leading-edge technology in artificial intelligence - to capture more complex, non-linear patterns in asset behaviour, unlock expanded usage of alternative data, and allow algorithms to adapt to changing market conditions in real-time. These systems will produce early warnings and enable more advanced portfolio construction, going beyond a mere look back at historical data.
Active managers could benefit greatly from this innovation also in alpha generation, making their strategies more resilient during future market crises. Deep learning will help these managers with several steps of the investment process, including asset allocation and selection by providing insights on the direction of the market and on the relative future performance of different asset classes or securities.
Across the sector, we are seeing firms battling challenging conditions, but changes are already starting to take place as asset managers look to the future. The Covid-19 crisis has caused a level of disruption not seen in recent memory, but it will lead to welcome shifts in the industry as managers innovate their processes and rethink strategies in light of this new normal.
© 2020 funds europe