## p.p1 0.0px 0.0px; font: 11.0px Helvetica} I proxy

p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 12.0px Helvetica}

p.p2 {margin: 0.0px 0.0px 0.0px 0.0px; font: 11.0px Helvetica}

I proxy price efficiency and information content by employing two groups of

measures for both price delay and idiosyncratic risk. The price delay measures are based on

the relative explanatory power of a simple market model compared to an extended market

model including 5 lags for market returns. The idea is that one observes greater price delays

in the incorporation of new information, if the extended market model has greater

explanatory power than the simple market model (Hou & Moskowitz, 2005). This price delay

decreases when share repurchases improve the speed with which information is assimilated

into share prices. The idiosyncratic risk measures quantify whether share price moves

together with the market. I use the R-squared between the simple and extended market model

as well as the absolute market correlation between stock and market returns to proxy for

idiosyncratic risk. When share prices increasingly co-move with the market the relative

amount of idiosyncratic risk declines. This implies that if share repurchase introduce noise

into share prices the idiosyncratic risk of a stock increases and the information content

declines (Busch & Obernberger, 2016). I use the manually collected weekly repurchase transaction data to construct two

measures of repurchase activity originally theorized by Busch & Obernberger (2016). The

first measure envelopes the monthly number of shares repurchased scaled by the total number

of shares outstanding in the previous month. The second measure approximates the remaining

number of shares that can be bought within the program scaled by the number of shares

outstanding at the beginning of the program. I furthermore incorporate firm fixed effects and

time effects to guarantee that my results are no affected by heterogeneity or macro-economic

factors (Busch & Obernberger, 2016).