Does Cash Flow News impact stock market prices?

Jan. 22, 2020

Congratulations Riccardo Sabbatucci whose paper “Cash Flow News and Stockprice Dynamics” written with co-autors Davide Pettenuzzo and Allan Timmermann, was recently accepted for publication in the Journal of Finance.

What is your paper about?

–The main research question of the paper is whether high frequency cash flow news has an impact on the stock market. Most existing studies focus on low frequency dynamics, but this disregards a lot of, potentially, valuable information. We are the first paper that tries to address this issue and present a new, quite technical, method to do it.

Why is this an important question?

– We know that stock prices should move either because of changes in expected returns (risk story) or expected cash flows (fundamental story). A lot of papers argued that expected returns are the only component that matter, and this is puzzling. We show that cash flow news, at high frequency, is indeed important.

What is high-frequency cash flow news?

– We define high-frequency cash flow news as the daily surprise in aggregate expected dividend growth. This aggregate news component is extracted from the daily announced dividends of individual firms.

We show that news about the expected dividend growth component helps predict stock returns, and this is comforting, meaning that news about what should be important for explaining movements in stock price is also to some extent driving stock price variation. 

How did you conduct the research?

– I have been working on the topic (dividend predictability) before, and was always puzzled by the fact that we study high frequency dynamics in asset returns but only low frequency dynamics in cash flow. Together with my co-authors, we found a way to deal with this aggregation problem.

What is your contribution to the literature?

– We are the first to provide the field with a measure of daily cash flow news by developing a method to extract an aggregate persistent dividend growth measure. To the best of our knowledge, no one has ever attempted this before because firms announced dividends only up to four times per year. Hopefully researchers will start using our cash flow measure in their empirical works.