Mimicking Repurchases

Mimicking Repurchases

Massimo Massa, Zahid Rehman, Theo Vermaelen

Journal of Financial Economics 84 (2007) 624-666

Presenter: Fangzhou "Ark" Shi

[Previous Research]

Two groups of previous research

  • The effect of repurchases in the prices of firms that announce a stock repurchases program.

This paper differs from the existing literature as it shows that an open market share repurchases announcement provides information not only about the repurchasing firm but also about its competitors.

  • Previous research has shown the ability of the market to infer something about the firm’s rivals based on various firm actions.

This paper focuses on the action of share repurchases. And it shows a negative impact of a share repurchases on the price of other firms in the same industry and g a step further by showing how firms react to repurchases of their industry counterparts thereby relating to the literature on strategic reaction.

[Two Theory Hypothesis]

Mimicking Hypothesis (MH)

As a firm repurchases, it generates expectations that the other firms within the same industry will also repurchases. It they do not, the market interprets it negatively and attributes the lack of the repurchase announcement to worse economic prospects and a weaker competitive position with respect to the repurchasing firm. This induces the other firms in the industry to repurchases, not to take advantage of a significantly undervalued stock price, but simply to correct the negative market perception by mimicking the behavior of their competitors.
Important factor: degree of strategic interaction within the industry.
And we use concentration as a proxy for the degree of strategic interaction

Similarity Hypothesis (SH)

Repurchases are made by mature firms for which the stock market has already partially anticipated a decline in investment opportunities.

  • Confirm the declining growth prospects of these firms.
  • Signal low agency problems for the repurchasing firm.

For repurchasing firm:
Given that the decline in growth had already been partially priced, the positive signal regarding low agency problems prevails and determines an appreciation in the stock price of the repurchasing firms.
For similar mature non-repurchasing firms located in the same industry:
Only the confirmatory signal about the decline in investment opportunities occurs, which leads to a depreciation in their stock price.
The underlying intuition of this alternative hypothesis is that mature firms within the same industry tend to act in similar ways because they share similar economic prospects.
Important factor: same industry, maturity.

[Data]

The data for this study comes from three sources: the Center for Research in Securities Prices (CRSP)-Compustat Merged Industrial Database for the accounting variables and the total payout dollar amounts by year for each firm, the Securities Data Corporation (SDC) Database for open-market repurchase program announcements and CRSP Monthly Stocks for dividend announcements.
Time Period: 1984~2002

[Research Method]

The authors give several hypothesis following both MH and SH. Use both univariate test and multivariate test to reject the insignificant hypothesis.

  • Maturity
  • Market Reaction to the Repurchase Announcement
  • The effect of a repurchases on other non-repurchasing firms within the same industry
  • The Decision to repurchase
  • Repurchases and long-run abnormal returns

[Maturity]

If concentrated industries contain firms that are more mature, then both the MH and SH would share some empirical predictions but the underlying rationale would still be very different.
SH:
1.Mature firms within the same industry share similar economic prospects.
2. Similar firms in the same industry repurchases at the same time because their similar economic conditions induce them to act in this way.
3. The repurchasing firms could be completely ignorant of the decisions of their rivals but could still end up making the same decisions.

MH:
1. The existence of a limited number of firms in concentrated industries makes strategic interaction more likely because the decision to repurchase acquires a strategic dimension.
2. Firms repurchase as a strategic reaction to the prior repurchases of other firms in the same industry.
3. The decision to repurchases is influenced or driven by the decision of a firm’s product market rivals.

The SH and MH yield similar predictions when concentrated industries are characterized by repurchasing firms that are more mature than the ones in unconcentrated industries. To distinguish the two theories:
Test whether repurchasing firms located in concentrated industries are more mature than their counterparts in unconcentrated industries.

[Maturity] – Test

1. Operating performance
Measure: ROCAA, ROS, ROE, CFROA.
Table 2: Univariate tests for analyzing the maturity of repurchasing stocks in high and low concentration industries using changes in operating performance measures.
Before the announcement, the changes in operating performance are significantly more positive for repurchasing firms in concentrated industries. However, after the announcement, we observe a similar decline in operating performance in both groups, which supports the Grullon and Michaely findings but does not support SH.
(Grullon and Michaely: Repurchases tend to be made by maturing firms.)

2. Risk
Risk of repurchasing stocks around the repurchase announcement.
According to Grullon and Michaely (2004), a contraction in the investment opportunity set must be accompanied by a decline in the firm’s risk. Therefore, if repurchasing firms in concentrated industries are more mature in comparison with their counterparts in unconcentrated industries, then the decline in their risk should be more significant in the post-announcement period.
Four methods:
1. Change in the standard deviation of stock returns around the repurchase announcement.
2. Changes in the systematic risk of equity using the Fama and French three-factor model.
3. Permanence of the cash flow shock measure of Guay and Harford (2000)
4. Dispersion in analyst forecasts

The univariate tests do not reveal any difference between the maturity of repurchasing firms located in concentrated and unconcentrated industries.

[Market Reaction to the Repurchase Announcement]

SH:
The SH implies that a repurchase announcement conveys two opposing signals to the market. For the repurchasing firm, the positive signal outweighs the negative one resulting in a net positive effect outweighs the negative one resulting in a net positive effect on the share price because of a repurchase announcement. However, giving the similarity of the firms in the industry, the market impounds the negative signal of the decline in profitability into the prices of other non-repurchasing firms in the same industry. Hence, these firms experience a more positive share price reaction when they repurchase later because a part of the negative signal reflecting the decline in profitability has already been anticipated and priced by the market. In the unconcentrated industries, since firms are less similar, this anticipation by the market does not occur and, therefore, the strength of the negative signal is stronger for repurchasing firms. We would expect the share price reaction to repurchase announcements to be more positive, on average, for firms located in concentrated industries.

MH:
There is no difference in reaction between repurchasing firms in concentrated and unconcentrated industries since the repurchase announcement never conveys a signal of permanent decline in future growth opportunities for the repurchasing firms.

(SH1a): The market’s reaction to a share repurchase announcement should be more positive for a firm which is located in a concentrated industry than for a firm located in an unconcentrated industry.
(MH1a): No difference is expected in the strength of the share price reaction to the repurchase announcement depending on the degree of concentration of the industry in which the firm is located.

SH:
According to Grullon and Michaely (2004), even though a repurchase announcement sends a negative signal about the firm’s profitability, it also signals, at the same time, a reduction in agency costs. Given that the impact of the signal depends on the market perception of the quality of governance of the firm the SH posits the strength of the positive signal to be greater if the repurchasing firms have poor governance.
This would signal a larger decline in agency problems in comparison to good governance firms.

MH:
MH does not require any difference in the abnormal return at the time of the repurchase either on the basis of better governance or on the basis of concentration.

(SH1b): The strength of the share price reaction to the repurchase announcement should be more positive for firms with poor governance than for firms with better governance.
(MH1b): The strength of the share price reaction to the repurchase announcement should not depend on the quality of governance of the firm.

[Market Reaction to the Repurchase Announcement] - Test

Table 3
Panel A: Univariate tests, CAR, both matched and unmatched quintiles
Panel B: Multivariate tests with control for other firm characteristics
Concentration Dummy: 0 for low concentration industries and 1 otherwise.

There is no statistical difference in market’s reaction across the two groups.
Concentration Dummy is insignificant in all the four specifications and there exists no difference between the two groups of stocks in terms of the market reaction to the repurchase announcement.
Announcement reaction does not vary either with the quality of governance or with the degree of concentration of the industry in any of the four specifications.

Reject SH1a & SH1b and accept MH1a & MH1b

[The effect of a repurchases on other non-repurchasing firms within the same industry]

SH:
Maturing firms experience a change in investment opportunities and the repurchase provides a confirmatory signal about this perception. This signal should have a negative impact on the value of mature non-repurchasing firms in the same industry at the time of the repurchase announcement.

MH:
The ability of the market to use the repurchase as a useful signal to draw inferences about other non-repurchasing firms depends on the degree of strategic interaction (use degree of concentration in the product market as proxy) within the industry. The higher the degree of strategic interaction within the industry, the more the market will interpret the lack of a repurchase as a sign of weakness of the non-repurchasing firm, leading to a drop in its stock price.

The two hypotheses make similar predictions only if concentrated industries are made up of firms that are more mature than the ones located in unconcentrated industries. By identifying a few firms in each type of industry that are most like the repurchasing firm, we can determine whether the reaction on their price is driven by declining profitability or whether it is a phenomenon related to industry concentration.

(SH2): The negative reaction on share prices of non-repurchasing rival firms should be significant only for mature firms and should be insignificant for non-mature firms, regardless of the concentration of the industries in which these firms are located.
(MH2): The negative reaction on share prices of non-repurchasing rival firms should be significant only for firms located in concentrated industries, regardless of their maturity.

[The effect of a repurchases on other non-repurchasing firms within the same industry] – Test

Table 4
Panel A: the average cumulative abnormal returns (CAR) for rival firms following the announcement for the time period specified.
Panel B: cross-sectional regression
Dependent variable: CAR of non-repurchasing rival firms
Average cumulative abnormal returns
Cross-sectional regressions
Rivals of matched repurchasing firms: Firms that belong to the same industry as the repurchasing firms and that have not made a repurchase announcement in the three years prior to the announcement or in the month following the announcement.
PS*Concentration Dummy: the interaction of Payout Size with Concentration Dummy. If the share price reaction for rival firms is stronger in concentrated industries, we would expect this term to be negative.
Maturity Dummy: 1 for mature firms and 0 for non-mature firms. Maturity is measured using six different proxies.
PS*Maturity Dummy: the interaction of Payout Size with Maturity Dummy.

There is a significant negative abnormal return for the non-repurchasing firms in concentrated industries. In the unconcentrated industries, the abnormal returns for the rival firms are not significant.

PS*Concentration Dummy turns out to be negative and significant in all the 12 specifications with the control for rival maturity. PS*Maturity Dummy is either insignificant or is significant with the wrong sign. The reaction on a firm’s share price as a result of a rival’s repurchase is not driven by the firm’s maturity but by the concentration of the industry.

Reject SH2 and accept MH2

[The Decision to repurchase]

SH:
Repurchase activity should be correlated among mature firms in the same industry since they share similar economic prospects. It is not the concentration of the industry that drives the correlation but the maturity of the firms located within these industries.

MH:
We should only observe correlated repurchasing activity in concentrated industries since it is precisely in these industries that the degree of strategic interaction between the firms is the highest and firms have a greater incentive to mimic each other. It is only the concentration of the industry that matters.

(SH3a): Repurchase activity should be correlated only among the mature firms in the industry, regardless of the industry’s concentration.
(MH3a): Repurchase activity should be correlated only in concentrated industries, regardless of the maturity of firms located in these industries.

SH:
Firms with declining profitability and availability of free cash flows repurchase to give these cash flows back to the shareholders. The incentive in doing this will be a function of the quality of governance of the firm. Well governed firms will distribute more, while less well governed firms will distribute less, if anything at all. Therefore, the quality of governance should make a difference in the incentive to repurchase.
Firms cluster their repurchases because they react to a common factor (use repurchase wave of other firms in the same industry as proxy). The sensitivity to the repurchases wave should be stronger in the case in which firms have good governance.

MH:
The decision to repurchase is affected by the repurchases of other firms in the industry per se and the repurchase wave is not a proxy for a common factor related to profitability changes. Hence, governance quality should not make any difference.

(SH3b): Repurchase activity should be more correlated among well-governed firms in the industry than among poorly governed firms.
(MH3b): The correlation of repurchase activity within the industry should not depend on the quality of governance of the firms.

[The Decision to repurchase] – Test

Table 5
Panel A: Tobit regression for the complete sample
Panel B: the breakdown according to maturity
Panel C: controls for the quality of corporate governance
Dependent Variable: ratio of repurchases to total payout, defined as the sum of dividends plus repurchases (between 0 and 1)
Based on both entire sample and payout increasing firms
Payout increasing firm (Jagannathan, Stephens, and Weisbach 2000): firms paid a dividend in at least one of the last two years and during the current year are either increasing their dividends, are initiating a repurchases, or are engaged in both simultaneously.
Conc: Concentration, Sum of the squared market share of each firm in the same industry during a year.
RW: Repurchase Wave. It is the total number of repurchase announcements, excluding those of the firm in question, which take place in the same industry over the past six-months.
Conc / RW interaction: the product of Concentration and Repurchases Wave. It measures the effect on the probability of a repurchase announcement resulting from an increase in both the concentration of the industry and the intensity of repurchase activity in the industry.
Conc / Gov: the product of Concentration and a governance dummy that takes a value of one if Governance Index (GI) exceeds 9 and is zero otherwise.

The Conc / RW interaction term is positive and highly significant across all the specifications. The strategic interaction with the competitors becomes stronger as the concentration of the industry increases and this is consistent with the MH.

Across all specifications, repurchase activity among mature and non-mature firms is equally correlated if the firms belong to concentrated industries.

Conc / RW interaction is positive and significant for both Good Governance and Bad Governance firms in all specifications. The Conc / RW / Gov Interaction is insignificant in all cases. It shows that firms with good and bad governance react alike to repurchase activity of other firms within their industry and the clustering of this activity cannot be explained by differences in the quality of corporate governance.

Reject SH3 and accept MH3

[Repurchases and long-run abnormal returns]

SH:
The information regarding the declining cost of capital is impounded into the share price of similar firms in the industry as more firms in the industry repurchase. Therefore, when later firms announce repurchases there might be no drift. This implies that, on average, repurchasing firms located in concentrated industries, may not show any share price drift. This is less so, however, for repurchasing firms in unconcentrated industries as these firms are less similar. Hence, these firms exhibit significant post-announcement share price drift.
Also, there should be no share price drift when the repurchasing firms in unconcentrated industries. Because there is no decline in the cost of capital.

MH:
Repurchases by firms in concentrated industries need not be followed by a share price drift since the repurchase decision is not motivated by market timing but is a response to the repurchase decisions of other firms in the industry. In contrast, firms in unconcentrated industries will only repurchase if their stocks are genuinely undervalued and are, therefore, more likely to experience a price increase in the post-announcement period.

In case of the SH, we expect to observe, within the set of repurchasing firms located in concentrated industries, a more significant share price drift for firms located in industries in which the flow of information is slower. This is because in these cases the market will not be very quick in pricing the decline in cost of capital for other firms in the industry. The speed of information flow should not matter for the MH.

Figure 1
Proxy: Information disseminates more slowly for industries that have lower levels of analyst coverage and trading volume and higher levels of analyst dispersion.

(SH4): The size of the post-announcement drift for repurchasing firms in concentrated industries should be negatively related to the speed with which information flows within the industry.
(MH4): No share price drift should occur for repurchasing firms located in concentrated industries.

[Repurchases and long-run abnormal returns] - Test

Figure 2: the average CARs for matched repurchasing firms in high and low concentration industries for the 36 months following the open-market repurchase announcement.
Abnormal returns are calculated using four different methods.

Table 6
Panel A: RATS (Ibbotson, 1975) regression for both matching and non-matching firms
Panel B: Calendar time portfolio regression (Ikenberry, Lakonishok, and Vermaelen 2000).
We construct portfolios made up of firms that have just announced an open market repurchase going long in the quintile of repurchasing firms in unconcentrated industries and short in the quintile of repurchasing firms in concentrated industries.
Panel C: Using “marching firms” methodology, comparing long-run performance of repurchasing and non-repurchasing control firms.

Table 7
Panel A: the impact of information flow on long-run performance
Panel B: the impact of maturity on long-run performance

In the case of low-concentration industries, the long-run (36 months) over-performance is never below 20%, regardless of the methodologies used to aggregate returns, while for high concentration industries, it is never above 10%.

In RATS regression, repurchasing firms in low concentration industries outperform their counterparts in high concentration industries (in both groups) by highly significant amounts over 12, 24, and 36 months.

CTPR: Portfolios of repurchasing firms in low concentration industries deliver significantly higher post-announcement returns than their counterparts in high concentration industries.

Repurchasing firms in unconcentrated industries outperform their controls by a margin greater than the one by which repurchasing firms in concentrated industries outperform their controls.

Within concentrated industries, the long-run returns for the repurchasing firms are insignificantly different from zero regardless of the speed with which the information flows within the industry. For the repurchasing stocks located in unconcentrated industries, we observe a positive significant share price drift regardless of the information flow characteristics of the industries.

No matter which proxy for firm maturity we use, non-mature firms in unconcentrated industries exhibit significant share price drift in the long-run. Moreover, both mature and non-mature repurchasing firms in low concentration industries out-perform their respective counterparts in high concentration industries.

Reject SH4 and accept MH4

[Conclusion]

In the case of strategic interaction between firms, repurchasing shares acquires a mimicking dimension. Repurchase in concentrated industries are chosen mostly as a reaction to other firms’ repurchase decisions and are not driven by the desire to take advantage of significantly undervalued stock prices.

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License