Over the past decade, the boom in ETFs and other passive investment products has significantly increased herding and overcrowding in financial markets. Many active managers looking to avoid
crowded trades and identify new sources of alpha have flocked to alternative data. A combination of
demand from the buy side and advances in artificial intelligence grew the number of alternative data providers more than threefold from 2010 to 2018. In turn, money managers looking to maintain an edge over peers have increased their spend on alternative data significantly in recent years.
Alternative data can provide predictive insights on publicly traded companies that traditional data sources (e.g. earnings statements, press releases, SEC filings) might miss. For instance, analysts often gather information on companies’ future plans and financial forecasts from public earnings calls.
However, executives can feel pressure from investors and other stakeholders to deliver forecasts in
which they may not be fully confident, rendering this information less useful for managers. Advances in
artificial intelligence have allowed for the development of “tonal analysis” technology. When applied to earnings calls, tonal analysis can get to the heart of not only what an executive says, but their level of confidence in these statements. If an executive were to deliver a strong forecast for the upcoming
quarter but tonal analysis revealed a level of deception, a manager using only information from the earnings call might receive a strong “buy” signal, whereas a manager who also incorporated tonal analysis data would likely end up with a weaker buy signal or a sell signal.
While the global datasphere is expanding rapidly, presenting meaningful opportunity and a plethora of choices for managers, incorporating alternative data in investment strategies presents certain challenges.