Do senders time information to take advantage of periods of high or low recipient attention? We explore this question with official US corporate information filings after-hours in our paper ‘Night Fever: Investor Inattention and the Timing of Corporate Filings’.
Corporate news is a great setting to study the interaction between senders and recipients of information for several reasons. First, regulation requires public firms to disclose relevant news in a standardized way that leaves little freedom in terms of content and format, except for the timing of the news release. The timing of news is an important object of interest because it is a strategic decision of the sender. Moreover, it is well defined empirically and it affects the recipient’s response. Second, institutional characteristics shape the distribution of attention and the ability to act upon corporate information at different times of the day, resulting in distinctly identifiable regimes.
Far from conventional wisdom, evenings are not a time of limited attention and hiding of bad news. Instead, there is a rich structure of attention, opportunities to trade, price formation, and opacity. We document three distinct information regimes within the hours after market closure:
- an early period (4.00-4.45pm) in which both investor attention (grey line in Figure I) and over the counter (OTC) liquidity is still high (green line in Figure II);
- a late period (4.45-5.30pm) in which investor attention is high (grey line in Figure I) but liquidity is low (green line in Figure II);
- an overnight period (5.30pm-6am) after the closing of the SEC filing system. Filings submitted during that time are not posted immediately. Instead, they are put in a buffer and become available simultaneously the next day at 6 am. This creates a decrease of information flow throughout the night as well as an information overload on the morning that decreases investors’ attention per filing (green dotted line in Figure I).
Figure I. Number of downloads by filing time.
Figure II. Number of filings and dollar volume across the day.
With a sample of 581,954 filings of current events (8Ks) from 1994 to 2012 we show that firms time news according to the three regimes to take advantage of these attention patterns and trading opportunities. In terms of aggregate numbers, the number of filings abruptly increase just after the official trading closes (black line). As the evening progresses, this high influx of filings changes character in terms of relevance, sign, and complexity of news (Figure III). Patterns also arise in terms of the type of news filed during each regime:
(1) big news is released right after market closure allowing for trading opportunities without price feedback;
(2) complex news later in the evening when investor attention is maximized; and
(3) bad news is more frequent in the overnight period, where the informational blackout generates minimal investor attention.
Big vs small news: trading volume in the first 24 hours after release
Complex vs simple news: length of a filing in words
Good vs bad nes: ex-ante good vs ex-ante bad news category
Figure III. The character of disclosure in the after-hours.
The effects are sharply discontinuous at the relevant thresholds (Figure III). News that seem negative on first inspection but turn out to be not so negative are abnormally represented in the late after-hours period, when attention is high, but prices are not posted simultaneously. These results jointly suggest that firms strategically allocate the timing of their news based on their relevance, complexity and obfuscation opportunities.
Besides describing the rich institutional structure of the after-hours period and establishing the motives for strategic timing of news within this period, our results imply that the timing of news is not exogenous. Firms strategically choose when to release news, even within the time periods of one day. Our finding that the timing of disclosure is strategic has implications for this literature, as well as the wider literature on news and asset pricing which often takes the arrival of news as exogenous. We hope that our findings will help future research in interpreting the content or reactions to news.
Vicente Cuñat is an Associate Professor of Finance at the London School of Economics and Political Science.
Moqi Groen-Xu is an Assistant Professor of Finance at London School of Economics and Political Science.