Leavey Professor Gustavo Schwenkler Works to Unleash the Power of Data for Small Firms and Individual Investors
If we have learned anything from our increasingly digital world, it’s that data is invaluable. Whether it’s being used by marketers to target specific audiences online, computer engineers to teach human-like behaviors to algorithms, or investors to inform financial decision making, data is arguably the world’s most precious commodity.
Finance professor, Gustavo Schwenkler, understands the value of data in finance, having observed and studied investment decision making for years. However, during his research on the cyclicality of firm failures, he encountered two significant challenges as an investor and researcher when attempting to access data: cost and availability.
“I was at an institution that didn’t have access to the data I wanted and external data providers were extremely expensive. Unlike established entities like Goldman Sachs and Morgan Stanley who can easily purchase data from aggregators such as S&P, Bloomberg, and Morningstar, no matter the price tag, researchers and retail traders do not have this luxury,” explains Schwenkler. “On top of that, the information that providers were offering was sparse and not extensive enough to gain a broad understanding of the nature of firm defaults.”
Confronted with these challenges, Schwenkler set out to find a viable solution. Over the past eight years, his research has centered on leveraging sophisticated statistical tools to unlock information that would otherwise remain inaccessible or inaccurate. His innovative approach involves the development of natural language processing tools capable of extracting insights from financial news reports. These tools analyze mentions of competitors, their relation to the economy, chain reactions of shocks amongst firms, and linkages between assets, empowering traders to make more informed investment decisions.
Using financial news from outlets like CNBC, MarketWatch, and Yahoo Finance is deliberate and serves as another informative layer to Schwenkler’s algorithms. What appears in the news versus what doesn’t can help investors make decisions, and while it may be perceived that financial news reports facts, there are natural biases and business objectives at play.
“News outlets cater to paying agents, and paying agents want financial news to report on what investors care about most, which is typically risk,” Schwenkler describes. “Consequently, the majority of news coverage revolves around stocks deemed risky, with a propensity for higher future returns.”
Schwenkler tested this hypothesis in his new research on financial news reporting, confirming that stocks reported in the news did in fact generate higher returns in the future, driven in large part by the level of risk associated with these stocks.
The advantages of accessing high-quality information, both at the individual and aggregate levels, are immense. Individuals can construct more robust portfolios, trade with greater insight, and effectively manage risk. Regulatory bodies overseeing financial markets can utilize this information to identify emerging contagions and determine the need for interventions or bailouts, as demonstrated in the case of Silicon Valley Bank earlier this year.
The more information and data investors have access to and the faster information is processed can contribute to overall market efficiency, predictability, and volatility.
“Even though they aren’t moving around as much money as giant firms, it’s important for average investors to have access to high quality data too, " says Schwenkler. “Retail investing has risen a lot in the past few years and there is a big population of people investing and trading, I mean, think of GameStop in 2021!”
The rise of everyday traders, who not only rely on traditional news but also leverage social media, introduces a new dynamic of investing and financial literacy. Acknowledging the power of social media as a source of valuable information, Schwenkler aspires to extend his research to encompass this influential domain.
“It’s no secret that new models, machine learning, and algorithms are at the forefront in Silicon Valley. However, without access to high-quality data, their efficacy is compromised," notes Schwenkler. “Looking ahead, I aim to continue to provide access to otherwise elusive data, fostering further research and innovation.”