The earnings announcement premium is the finding that stocks have, on average, earned higher returns in the periods when they are scheduled to announce earnings, regardless of whether the news turns out good or bad. It is a statement about the calendar, not about surprises. The effect was documented in small stocks in 1988, measured at roughly 7% a year in a large U.S. portfolio test in 2007, and has since been given two competing explanations: investor attention and compensation for risk. This article walks through the evidence, the debate, how the premium differs from post-earnings announcement drift, and why event risk makes it hard to capture with a single trade.
The Earnings Announcement Premium by the Numbers
| Figure | Value | Source |
|---|---|---|
| Long-short return, stocks expected to announce vs those not | 61 basis points a month, t-statistic over 5 | Frazzini & Lamont (2007) |
| Same result, annualized | About 7% a year | Frazzini & Lamont (2007); arithmetic: 61 bp × 12 ≈ 7.3% |
| Abnormal return of firms scheduled to report | 9.9% annualized | Savor & Wilson (2016) |
| How long the premium lasts | About 20 days | Frazzini & Lamont (2007) |
| Earliest samples showing the effect | Back to 1927 | Frazzini & Lamont (2007) |
| Countries with a significantly positive premium (among 20 with enough data) | Nine (June 2011 working-paper draft) | Barber et al. (2013), draft numbers |
What the premium is
Every public company reports earnings on a schedule that is largely known in advance. The earnings announcement premium says that a stock’s average return is higher in the window around that scheduled date than in other periods. It is an average over many stocks and many quarters. It does not say that a given company’s report will be well received, and it does not depend on the sign of the news.
The idea has a long lineage. Beaver (1968) studied 506 earnings announcements for 143 firms and compared trading volume and volatility in the announcement week with non-event periods. That was a study of how much the market reacts, not of a signed return premium, but it established that announcement dates are unusual in the data.
The early evidence: small firms
Chari, Jagannathan and Ofer (1988) documented abnormal returns around quarterly announcements. In their words, “small firms show large positive abnormal returns and a sizable increase in the variability of returns around these dates,” while “large firms show no abnormal returns around announcement dates.” Ball and Kothari (1991) then examined risk and abnormal returns in the days around quarterly announcements, using a design that lets risk vary daily in event time, and tested whether announcement-period risk and return vary with firm size.
At this stage the effect looked like a small-stock curiosity. That changed with the next study.
Frazzini and Lamont: a large, robust premium
Frazzini and Lamont (2007), in an NBER working paper, reported that “on average, stock prices rise around scheduled earnings announcement dates,” and that the premium is “large, robust, and strongly related to the fact that volume surges around announcement dates.” They built a long-short portfolio that held stocks expected to announce in a given month and shorted those that were not. It earned 61 basis points a month, or about 7% a year, with a t-statistic over 5. The CAPM could not explain it.
The authors also reported that the premium is quantitatively substantial especially among large-cap securities, lasts about 20 days, and is evident in samples going back to 1927. The large-cap finding sits uneasily with the earlier small-firm result, and we return to it below. Note also that the paper is an NBER working paper; we verified the working-paper version only.
Explanation one: attention and volume
Frazzini and Lamont linked the premium to trading volume. Stocks with high past announcement-period volume earned the highest premium, and the authors concluded that “the premium is driven by buying by small investors when the announcement catches their attention.” The mechanism is a demand story: an announcement draws attention, attracted investors buy, and the price is pushed up for a few weeks.
A working-paper draft of Barber, De George, Lehavy and Trueman (June 2011) also supports attention. The authors report evidence of “an increase in the attention paid to firms around the time of earnings releases, creating upward pressure on stock prices.”
Explanation two: compensation for risk
Savor and Wilson (2016) offer a different reading. They report that “firms scheduled to report earnings earn an annualized abnormal return of 9.9%.” Their account is risk-based: the covariance of firm-specific and market-wide cash-flow news spikes around announcements, so exposure to announcement risk is priced. In support, they find that announcer returns forecast aggregate earnings, that the premium is persistent across stocks, and that early announcers earn higher returns than late ones.
The two explanations are not settled in favor of either. Barber et al. report that, in their international data, there is “no evidence that higher levels of systematic or idiosyncratic risk around the time of earnings releases is a significant driver of the premium.” Savor and Wilson reach the opposite conclusion from U.S. data. That is a live disagreement, not a closed case.
Evidence outside the United States
Barber, De George, Lehavy and Trueman published “The earnings announcement premium around the globe” in the Journal of Financial Economics in 2013. We could not retrieve the published abstract, so the figures here come from their June 2011 working-paper draft and may differ from the final version. In that draft, using data from 46 countries, nine of the 20 countries with enough data for a within-country analysis showed a significantly positive premium, and the premium was significant in 16 of 20 sample years. The draft also reports that the premium for the smallest stocks exceeds that for the largest by roughly 6 percent a year, which fits the Chari et al. small-firm result better than the large-cap emphasis of Frazzini and Lamont. The draft describes the U.S. estimate of Frazzini and Lamont as over 7 percent a year, consistent with the figure above.
Two cautions follow. The premium is not found in every market, and the size pattern is not uniform across studies.
How this differs from post-earnings announcement drift
The two effects are often confused because both involve earnings. They are distinct:
- The earnings announcement premium is about timing. Returns are higher around the scheduled date for the average stock, whatever the surprise.
- Post-earnings announcement drift is about what happens afterwards. Bernard and Thomas (1989), “Post-Earnings-Announcement Drift: Delayed Price Response or Risk Premium?”, studied the drift that follows an announcement and depends on the sign of the earnings surprise. Their title poses the same question of delayed reaction versus risk that the premium literature asks.
In short, the premium rewards holding a stock during the announcement window; drift rewards trading in the direction of the surprise afterwards. For the drift effect, see our article on post-earnings announcement drift, and for how surprises are measured, standardized unexpected earnings and earnings surprise stocks.
Why it is hard to trade: event risk and implied-volatility crush
A premium that holds on average across a diversified portfolio is not the same as a dependable gain on one stock. A single report can move the price sharply in either direction, and the premium is small relative to that range. Dubinsky, Johannes, Kaeck and Seeger (2019) studied option-implied earnings risk and found that “the anticipated price uncertainty is quantitatively large, varies across time, and is informative about the future return volatility.”
Options add a second problem. Leading up to an announcement, option prices often rise because of the increased uncertainty, and afterwards implied volatility falls sharply, a pattern traders call implied-volatility crush. A buyer of options ahead of earnings therefore pays for that uncertainty up front, and can lose even when the stock moves in the hoped-for direction if the move is smaller than the price implied. We do not have a verified size for the crush, so we make no numerical claim. For more on the pricing, see implied volatility explained, options Greeks explained and how to trade earnings.
Limitations
The headline numbers come from specific designs. The 61 basis points a month is a long-short portfolio result from a working paper, not the return of buying one stock before its report. We did not capture the exact sample period behind it. The 9.9% figure uses a different method and is not directly comparable, and the two should not be added or averaged. None of these figures is net of trading costs or of the price pressure created by acting on a widely known effect.
The evidence is also incomplete in important ways. The explanations conflict, the size pattern differs between studies, and the international numbers we cite come from a draft. Several papers are decades old, and we have not tested whether the premium has persisted since publication. Published anomalies often weaken once they are widely known, so past averages are not a forecast. Nothing here is investment advice.
FAQ
What is the earnings announcement premium?
It is the tendency of stocks to earn higher average returns in the weeks they are scheduled to announce earnings. Frazzini and Lamont (2007) measured about 61 basis points a month, or roughly 7% a year, for a long-short portfolio of announcers versus non-announcers.
Is the premium the same as post-earnings announcement drift?
No. The premium concerns returns around the scheduled announcement date, whatever the news. Drift concerns the continued price movement after the announcement, in the direction of the earnings surprise.
Why does the premium exist?
There are two main explanations. Frazzini and Lamont tie it to attention and buying by small investors when volume surges, while Savor and Wilson (2016) argue it compensates for announcement-related risk. The evidence has not settled the question.
Can I capture it by buying stock or options before earnings?
The research describes an average across many stocks, not a reliable gain on any one report. Single-stock event risk is large, and options typically carry elevated implied volatility before the event that falls after it, so the premium does not translate into a simple trade.
References
- Ball, R. & Kothari, S. P. (1991). “Security Returns Around Earnings Announcements.” The Accounting Review, 66(4), 718–738. doi:10.2308/tar-9605070402
- Barber, B. M., De George, E. T., Lehavy, R. & Trueman, B. (2013). “The earnings announcement premium around the globe.” Journal of Financial Economics, 108(1), 118–138. doi:10.1016/j.jfineco.2012.10.006 (figures in this article are from the June 2011 working-paper draft).
- Beaver, W. H. (1968). “The Information Content of Annual Earnings Announcements.” Journal of Accounting Research, 6, 67–92. doi:10.2307/2490070
- Bernard, V. L. & Thomas, J. K. (1989). “Post-Earnings-Announcement Drift: Delayed Price Response or Risk Premium?” Journal of Accounting Research, 27, 1–36. doi:10.2307/2491062
- Chari, V. V., Jagannathan, R. & Ofer, A. R. (1988). “Seasonalities in security returns: The case of earnings announcements.” Journal of Financial Economics, 21, 101–121. doi:10.1016/0304-405x(88)90033-5
- Dubinsky, A., Johannes, M., Kaeck, A. & Seeger, N. J. (2019). “Option Pricing of Earnings Announcement Risks.” Review of Financial Studies, 32, 646–687. doi:10.1093/rfs/hhy060
- Frazzini, A. & Lamont, O. A. (2007). “The Earnings Announcement Premium and Trading Volume.” NBER Working Paper 13090. doi:10.3386/w13090
- Savor, P. G. & Wilson, M. (2016). “Earnings Announcements and Systematic Risk.” The Journal of Finance, 71(1), 83–138. doi:10.1111/jofi.12361
Track Earnings Events With Evidence, Not Hype
Alpha Suite builds its earnings-related signals on published research and reports what the evidence does and does not support.
Get Started with Alpha Suite