The next generation of robo-advisors may move into the world of short-term trading and stock selection.
A recent study by researchers at the University of Chicago shows that AI, especially large-scale language models trained on vast amounts of text and capable of generating natural responses, can analyze financial statements as well as or better than human analysts. But before you ask ChatGPT to suggest your next deal, let's go a little further.
The researchers provided GPT-4 with financial statements, with company names and personally identifiable details removed, and asked it to predict the direction of future revenue. They found that GPT-4 predicted revenue changes more accurately, outperforming human analysts. Moreover, trading strategies based on GPT-4's predictions delivered higher returns than strategies based on other models.
So what does this mean for the future of financial analytics? In the short term, AI's ability to process vast amounts of data and generate insights quickly and accurately will undoubtedly impact the financial services industry. Ultimately, for investors, AI will strengthen the case in favor of a passive investment approach.
AI can analyze trends faster than humans, calculate key financial ratios, and provide narrative insights into a company's future performance. Just as software developers already do, experts will add an AI co-pilot to their toolkit.
But the long-term impact on retail investors is more subtle. As more firms adopt AI techniques, the initial advantage they provide will fade. Just as high-frequency trading firms, which rely on speed to gain a trading advantage, have competed with each other to reduce the time it takes for orders to reach exchanges (one high-frequency trading firm once paid $14 million for land next to the Chicago Board of Trade to install antennas that would route orders one microsecond faster), AI-enhanced financial analysis will likely become an arms race.
In other words, if an advantage exists, the market will eat it up until there is no longer a sustainable advantage.
This is why the work of American economist and Nobel Prize winner William Sharpe is timeless. In his 1991 paper, “The Arithmetic of Active Management,” Professor Sharpe argued that in a world where all investors are actively trying to outperform the market, their collective efforts will cancel each other out. For every investor who outperforms the market, there will always be others who underperform.
Thus, when costs are taken into account, an average dollar of active management will underperform an average dollar of passive management (because the costs of active management are higher than those of passive management).Professor Sharpe's insight is based on the idea that markets are generally efficient, and that all available information is already reflected in stock prices.
Combined with the rise of AI-powered financial analytics, Professor Sharpe’s logic leads to a compelling conclusion: the best way for most investors to benefit from these technological advances is through a low-cost passive indexing approach.
By investing in a broad market index, investors can gain exposure to the entire market and benefit from the collective insights of all market participants, including those using advanced AI tools. Importantly, they can do this while minimizing costs, which can be a significant drag on long-term investment returns.
For individual investors, this means focusing on low-cost index funds that offer broad market exposure. It also means resisting the temptation to constantly tweak your portfolio in an attempt to outperform the market.
The current generation of robo-advisors typically use low-cost index funds, but some actively manage exposure to a variety of passive products in their portfolios. Five-year annualized returns on growth portfolios from Canadian robo-advisors, reported by the Globe last fall, ranged from 3.67% to 5.98% for the period ending Sept. 30, 2023.
Research consistently shows that the more frequently investors trade, the lower their returns tend to be, often due to trading costs, taxes, and the psychological pitfalls of trying to time the market.
The secret to successful passive investing is staying on course rather than reacting to short-term market fluctuations. This approach is consistent with passive investing principles that emphasize buy-and-hold strategies and minimal trading activity. Doing so allows investors to make the most of the market's long-term upward trajectory without being distracted by short-term volatility.
But it's important to realize that staying passive is easier said than done. One of the hardest things about passive investing is staying disciplined during market downturns. Investors often believe they can keep their emotions at bay and stick to their plan, but it is precisely at times like these that they are most tested.
The behavioral challenge of staying on track cannot be underestimated; it requires a solid understanding of investor psychology.
Artificial intelligence can be used to provide insights and warnings that help investors avoid common pitfalls, such as panic selling during market downturns or over-trading in an attempt to outperform the market. For example, AI can be used to better match investors with the most appropriate risk profile.
Pure DIY investors are prone to taking on high risks due to overconfidence. Predictive analytics can help you better understand investor psychology and guide you to a portfolio that is more likely to stick from the start.
AI has the potential to revolutionize financial statement analysis, but its greatest contribution to investors may not be in stock selection, but in helping them maintain discipline.
Preet Banerjee He is a consultant to the asset management industry with a focus on the commercial application of behavioral finance research.
