**3. Sample**

This study collects open market share repurchase announcement dates of non-financial and non-utility firms in the U.S. from January 1996 to September 2014 from Thomson Reuters SDC Platinum. Price and accounting variable data of these firms were obtained from Thomson Reuters Datastream. Market-adjusted announcement period abnormal returns were computed with market value-weighted returns obtained from Kenneth French's website8 as the benchmark. This study does not use the market model to estimate abnormal announcement returns as several firms in the sample repeat their announcements in less than a year period, and thus would bias the measurement of normal period return.9 Merging the data obtained from SDC Platinum and Thomson Reuters Datastream reduces the sample size to 3122 announcement-year observations. To mitigate the effects of outliers, these variables are winsorized at the 1% and 99% levels.

Table 1 shows the number of non-repeat and repeat announcements from 1996 to 2014, their corresponding announcement period returns, and the associated size programs during the sample period. In Panel A of Table 1, the total number of open market share repurchase announcements increases from 227 in 1996 to 405 in 2008 and then declines to 82 in 2014. The lowest (highest) total number of announcements is in 2012 (1998) with 36 (405) announcements. The percentage of announcements made by repeat repurchasers decreases from 59% in 1996 to less than half in 1999. Since then, it has increased to, on average, around three-fourths of all repurchase announcements

<sup>8</sup> http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data\_library.html#HistBenchmarks.

<sup>9</sup> This methodology is also used in studies on repeat acquisitions, such as (Karolyi et al. 2015; Billett and Qian 2008).

every year in the sample period. Panel B shows that the lowest (highest) 2-day cumulative abnormal returns or CARs (days 0, +1) is 0.4% (4.7%) in year 2006 (1999). The average 2-day CARs since year 2000 is 1.78% per year, which is much lower than those reported in earlier studies in the open market repurchase literature. The average cumulative announcement abnormal return for the whole sample period is 2.4%. We observe that the size of repurchase programs increases over time with the average smallest (largest) program launched in 2004 (2012). On average, repurchasing firms plan to buy back around 7.56% of their outstanding stocks. The (untabulated) correlation coefficient between the percentage of repeat announcements and announcement returns is negative. These statistics clearly show that open market share repurchase announcements in more recent periods are dominated by repeat repurchase announcements with significantly lower announcement period returns.


**Table 1.** Sample distribution and market adjusted returns.



**Table 1.** *Cont.*

ONLY ONE refers to firms that announce only one repurchase during the sample period. REPEAT is firms that announce to repurchase the first time and will repeat during the sample period. SIZEPROG is percent of shares authorized at initial authorization date.
