LIQUIDITY CHANGES ASSOCIATED WITH OPEN MARKET REPURCHASES

Before the tax-law changes of 1986, common stock repurchases received favorable tax treatment relative to cash dividends, yet more than 80% of the New York Stock Exchange-listed firms did not use repurchases fo r distributing value to their stockholders. Prior research suggests a possible resolution of the puzzle by examining the effect of open mark et repurchases on the liquidity of the firms' stocks. The liquidity of the stock as measured by bid-ask spread may be affected by stock repu rchases in any one or all of the following ways. First, when managemen t undertakes to reacquire the firm's shares in the open market they ar e, in effect, competing with the market makers of the stock. This open market repurchase activity, in the absence of information asymmetry, should result in greater liquidity or lower bid-ask spread. Second, a direct consequence of open market repurchase announcements may be incr eased trading in the secondary market. Increased trading volume makes it easier for the market maker to reverse his position in the stock. T herefore, the inventory holding cost component of bid-ask spread shoul d decline upon announcements of open market repurchases. A decline in the bid-ask spread would be consistent with this explanation. Third, p rior research suggests that open market repurchase announcements are a ssociated with increased trading by informed traders in the secondary market for the firm's stock. Informed traders trade with market makers only at favorable prices. Hence, the adverse selection component of t he bid-ask spread should increase in the post-announcement period. Thi s information-asymmetry- based explanation predicts increased bid-ask spreads following announcements of open market repurchases. However, i t should be noted that an asymmetric-information-based explanation doe s not necessarily imply that other market participants face the inform ed traders in all trades at all times. Specifically, the likelihood of trading with an informed trader is greater when stock prices are lowe r rather than higher. Also, repurchases only involve buying and not se lling by informed traders. Hence, the risk of trading against the info rmed trader is much less. An earlier study reports that bid-ask spread increases when firms announce their intention to reacquire common sto ck. It is argued that the increased bid-ask spread leads to an increas ed cost of capital. This hidden cost associated with open market repur chases may explain the preference for cash dividends. However, due to data constraints, the earlier study uses annual bid-ask spreads to doc ument the changes in liquidity. The use of annual data imposes certain limitations. Using annual data, it is not possible to examine when th e percentage spread increases with respect to the announcement date, t hat is, before or after the event. For instance, if the percentage spr ead is measured at year-ends and the event occurs in between, then an increase in post-event percentage spread may suggest that the event in duces the change in the spread. However, the spread could have increas ed prior to the event and remained stable thereafter. The availability of daily bid-ask spread data for our sample of National Market System (NMS)-listed stocks facilitates a close examination of liquidity chan ges surrounding the announcement of open market repurchase programs. W e find a significant increase in the percentage spread in the 60-day p eriod immediately preceding the event. Further, this increase in perce ntage spread is more pronounced in the 30-day period immediately prior to the announcement day. In subsequent periods, we find no further ch ange in percentage spreads. Our results indicate that percentage sprea ds increase prior to open-market repurchase announcements. These findi ngs are not consistent with the proposition that open-market repurchas e program announcements cause an increase in percentage spreads. We te st die robustness of our result by using a control sample of firms tha t are matched with our sample firms by market value of equity. Cross-s ectional regressions indicate that change in common stock price is the most important determinant of the change in percentage bid-ask spread . The change in spread does not differ between the firms announcing th e repurchase program and the control sample. The matched-sample result s support our univariate test results that the change in prices explai ns the change in percentage bid-ask spreads. Recent studies suggest th at firms buy back shares when their common stock is substantially unde rvalued. We also find that open-market repurchase programs are initiat ed following a period of steep decline in stock prices. The announceme nts are associated with significantly positive average stock price rea ctions. It appears that managers initiate repurchase programs to shore up declining investor confidence. This may be a partial explanation o f why open market repurchase programs are observed less frequently tha n cash dividends.