Printer Friendly

Managerial discretion costs and the acquisition of capital: evidence from forced warrant exercise.

Forced warrant exercise should elicit a stock price reaction only in response to inventory adjustments or unanticipated increases in agency costs. We find evidence of both effects. Firms calling warrants for redemption experience negative abnormal returns on the announcement date. Negative returns are concentrated among firms whose characteristics suggest that the call should have been deferred. We find lower announcement returns among inefficient firms with low leverage--firms for which the agency costs of managerial discretion may be high. All else equal, announcement returns are lower among inefficient firms that invested heavily in the year after forced exercise.

The agency costs of managerial discretion play an important role in the valuation process. Although managerial discretion costs cannot be explicitly measured, we can infer these costs from the market reaction to announcements that signal a change in their magnitude.

Studies of seasoned equity offerings (SEOs) and asset sales have documented a relation between managerial discretion costs and stock returns. In this paper, we examine the market reaction to another capital acquisition event in which managerial discretion costs may play a significant role: forced warrant exercise. Warrant exercise is forced when in-the-money warrants are called for redemption prior to expiration. Similar to asset sales and SEOs, forced warrant exercises bring new capital into firms with varying levels of both project quality and debt capacity. Thus, managerial discretion costs become a potential (though not exclusive) influence on the announcement returns.

In perfect markets, managers should call warrants as soon as they are in the money, in order to expropriate the time premium on the warrants. In the presence of managerial discretion costs, firms with poor investment opportunities or unused debt capacity should defer warrant calls. Forcing exercise and injecting equity into the company is bad news because shareholders fear that the proceeds will be misapplied. The announcement of a warrant call should lead to lower residual returns where non-value-maximizing behavior is more likely. These abnormal returns reflect the costs of managerial discretion that are associated with the unanticipated equity issue.

Our results are consistent with the agency explanation for the market's reaction to forced warrant exercise. We find that the market reacts adversely to the warrant call when the announcement reveals that the firm is either issuing equity when debt capacity exists or planning to invest when it has poor investment opportunities. The negative share price reaction to the announcement is concentrated among inefficient firms with low leverage. High-leverage firms experience lower abnormal returns the more they reduce their total borrowing in the year following the warrant call. Holding the impact of leverage constant, inefficient firms experience lower announcement returns the more they invest. Efficient firms fail to evince a similar relation.

An alternative (though not mutually exclusive) hypothesis to explain stock price reactions to a warrant call is that price movements are due to inventory adjustments by security dealers. We explore this hypothesis, but find that it cannot entirely explain market reactions to warrant call announcements. This result, coupled with the strong relation between announcement residuals and our standard proxies for managerial discretion costs, lends credence to the joint hypothesis that the warrant call conveys new information and that agency costs explain some of the negative stock price reaction.

This paper adds to prior research that documents the impact of agency costs on stock returns. Jung, Kim, and Stulz (1996) examine a sample of companies that announce SEOs and find that the announcement residuals are significantly negative only among firms that chose to issue equity despite having high free cash flow and inferior investment opportunities. Bayless and Chaplinsky (1991) report a similar finding. The studies suggest that the adverse market reaction to equity issues reflects the increased costs of managerial discretion when firms that should have issued debt sell equity instead, thereby incurring higher future agency costs.

Allen and McConnell (1998) show that average announcement returns are lower when firms retain the proceeds from an equity carve-out, instead of using the funds to reduce borrowing. Lang, Poulsen, and Stulz (1995) report a similar finding in a study of asset sales. Because firms with poor investment opportunities dominate both samples, the lower announcement residuals appear to reflect the greater managerial discretion costs that are expected to follow the acquisition of equity capital. McConnell and Muscarella (1985) find a negative share price reaction when oil companies, which had poor investment opportunities during their sample period, announce increases in planned capital expenditures.

I. Why Does the Market React Negatively to Forced Warrant Exercise?

In this section, we examine two possible causes for the negative reaction to the warrant call: 1) first, that the warrant call reveals new information about managerial discretion costs, and 2) second, that the warrant call causes inventory adjustments by shareholders and dealers, which results in a temporary price pressure effect.

A. Managerial Discretion Cost Hypothesis

Value-maximizing managers have an incentive to call their warrants as soon as they are in the money because forcing exercise transfers the value of the remaining time premium from the warrant holders to the stockholders (Ingersoll, 1977). In a perfect market, managers will call in-the-money warrants immediately to realize the gain from wealth expropriation. Managers will invest the proceeds from the forced exercise if the firm has good projects and otherwise return the proceeds to the shareholders.

Market imperfections such as agency costs and taxes create circumstances under which the managers might find it optimal to delay calling the warrants. Deferring a forced exercise is optimal for a firm with poor projects or unused debt capacity if the costs of disposing of surplus cash exceed the value of the expropriated time premium. Thus, in an imperfect market, both the firm's investment set and the costs of disposing of surplus cash determine the optimal call policy.

If the firm has good investment opportunities and no debt capacity, value-maximizing managers will call the warrants immediately and invest. No share price reaction is expected, because the quality of the investment opportunity set is known and investors anticipate the call.

If the firm has poor investment opportunities or has debt capacity, deferring the warrant call could represent the best course of action. Value-maximizing managers will delay calling the warrants when the costs of disposing of the exercise proceeds exceed the value of the expropriated time premium. These costs might include personal taxes triggered by special dividends or share repurchases, a less than optimal concentration of voting power in managers' hands generated by a stock buyback, the loss in value caused by investing the cash in bad projects, and increased agency costs generated by suboptimal debt reduction. The frequency of delayed warrant calls is significant: Schultz (1993a) examines 116 callable warrants and finds that 35 (30% of his sample) were not called even though they met the necessary conditions.

When the best policy is to delay forcing exercise, calling the warrants increases managerial discretion costs, because the infusion of capital makes it more likely that managers will overinvest either now or in the future. The announcement of a warrant call will generate a negative share price reaction because it signals a deviation from value-maximizing behavior. The abnormal return reflects the value of unanticipated managerial discretion costs.

If reducing financial leverage below optimal levels reduces monitoring and makes future overinvestment more likely, even using surplus cash to pay down debt can increase future managerial discretion costs. Forced exercise is a leverage-decreasing event, and firms that use the exercise proceeds to pay down debt experience an even greater reduction in financial leverage, which increases the likelihood of future overinvestment. This expectation follows from Jensen (1986) and Stulz (1990), who show that debt limits managerial discretion, since debt service limits free cash flow, and lenders have an incentive to monitor management to make sure they will be repaid. Thus, the optimal call policy for firms with poor investment opportunities or excess debt capacity is to delay forcing exercise.

We hypothesize that the costs of managerial discretion associated with forced exercise will be higher among three types of firms: 1) inefficient firms, because it is more likely that managers will use additional capital for value-destroying uses; 2) firms with low leverage prior to the call, because the firms are adding equity capital when there may be spare debt capacity; and 3) firms that reduce their industry-adjusted leverage with the call proceeds, thus increasing the likelihood of future overinvestment. We expect these increased agency costs to result in lower announcement returns in each case. (1)

B. Inventory Adjustment Hypothesis

A price pressure or inventory adjustment hypothesis offers an alternative (though not mutually exclusive) explanation for a share price reaction to forced warrant exercise. Under this hypothesis, the negative announcement-day return results from portfolio adjustments by warrant holders. Schultz (1993a) notes that warrant holders might sell their warrants rather than exercise them when they are called. Market makers or specialists who expect to buy these warrants might then lower their bid and ask prices to eliminate a long position in the stock, or perhaps short the stock, to hedge their positions in the warrants. At the completion of the warrant call (the last day on which the warrants can be exercised), the market maker raises prices and covers the short position by exercising the warrants.

The inventory control hypothesis fits well with the expectation that both the warrant call and the likely use of exercise proceeds would be fully anticipated. It implies that the total abnormal return on the stock between the announcement and the completion of the warrant call should be zero. Furthermore, there should be a strong negative correlation between the announcement and completion day returns, since the firms with the lowest announcement day returns should experience the greatest relief from downward price pressure on the day the warrant call is completed.

II. Data and Sample Characteristics

We began sample selection with a search of the Commerce Clearing House Capital Changes Reporter for companies that referenced the terms "warrant" and "called." The search returned a total of 992 companies, of which 233 reported completed warrant calls and 153 had returns available on the CRSP file. We were able to identify announcement dates for 123 of the remaining companies. We removed six firms because the warrant call announcement was contaminated by a concurrent earnings announcement, and dropped two companies because their warrants were called in anticipation of a merger agreement, leaving a final sample with 115 companies.

We obtain the estimated proceeds from the warrant exercise for 103 companies in the sample and compare those amounts to the market value of common equity at the end of the second day prior to the announcement of the warrant call. The average ratio of exercise proceeds to equity value is 17.6%, with a median value of 14.2%. The maximum ratio of exercise proceeds to equity value was 79.0%, and the minimum was 0.3%. We believe that it is reasonable to characterize the exercise proceeds from the warrant call as economically significant for the median firm.

Fifty-two of the 115 news stories that we use to identify warrant call announcements also list intended uses for the proceeds of the warrant call. Of these 52 stories, 43 (83%) note that the proceeds were to be used for investment purposes (working capital, capital expenditures, research and development, or acquisitions). Five list both investment and debt reduction, three note a desire to strengthen the balance sheet by raising equity, and one lists debt reduction as the sole reason for calling the warrants. Thus, a large majority of firms that report use of proceeds intend to invest the cash raised in the warrant call. (2)

Table I summarizes the distribution of warrant call announcements across the sample years. The majority of companies (64%) called their warrants after 1990. We draw approximately 56% of the sample (64 firms) from four industry groups, as defined by two-digit SIC codes (chemicals, instruments/electronic equipment, wholesale goods and business services). The remaining companies are distributed evenly across 34 industries.

The sample firms are, on average, small and young. The mean (median) market value of equity at year-end prior to the warrant call is $76.6 million ($37.1 million). The mean (median) length of time from listing on an exchange until the warrant call announcement is 4.3 years (2.0 years).

We find that the stock price performance of the sample firms prior to the warrant call announcement is very good, displaying a run-up that exceeds the abnormal investment performance of companies that conduct SEOs. Over the year prior to the warrant call announcement, the common stock of the sample firms produced an average return of 97.2%, compared to a return of 17.7% on the CRSP value-weighted index.

We compute buy-and-hold abnormal returns as the one-year buy-and-hold return on the sample firm's stock, minus the buy-and-hold return to an equal weighted portfolio matched on size, book-to-market equity ratio, and prior one-year return (Lyon, Barber, and Tsai, 1999). The median buy-and-hold abnormal return in the year (252 trading days) prior to the warrant call was 49.3%.

Recent research by Loughran and Ritter (1995, 2000), Spiess and Affleck-Graves (1995), Lee (1997), and McLaughlin, Safieddine, and Vasudevan (2000) shows that companies that conduct SEOs underperform non-issuing firms with similar size and book-to-market characteristics over the three years following the sale of equity. Since warrant calls, like SEOs, provide funds for investment and reduce leverage, we examine the long-run abnormal performance of the sample firms following the call date.

To compute long-horizon excess stock returns, we select control firms using procedures recommended by Lyon et al. (1999). We create a control firm portfolio that we match on the basis of firm size, book-to-market equity, and prior one-year stock return. Lyon et al. (1999) report results based on size and book-to-market portfolios, but recommend controlling for other relevant characteristics of the sample. Since our sample firms experienced a run-up in prices in the year prior to the announcement, we choose reference portfolios matched on past return as well. We compute 756-trading day (three-year) buy-and-hold returns for both the sample firm and the control portfolio, beginning two days after the announcement of the warrant call. The abnormal return in event time is the difference between the buy-and-hold returns to the sample firm and the control portfolio. The significance of the long-run excess returns is evaluated through the use of empirical p-values, as in Lee (1997). (3)

Table II contains the three-year buy-and-hold abnormal returns (BHARs) for the sample. Similar to companies that conduct SEOs, firms that call warrants for redemption also underperform firms with similar size and book-to-market characteristics during the three years following the warrant call.

A total of 103 firms have sufficient financial data for the computation of long-mn abnormal returns. The mean three-year buy-and-hold return to these 103 companies is - 14.5%. The mean three-year buy-and-hold return to the control portfolio was 24%, producing a buy-and-hold abnormal return (BHAR) of -38.5%. The empirical p-value is zero, reflecting the fact that in our 10,000 independent trials, all of the "pseudo-warrant call" portfolios outperform the portfolio of firms that call their warrants. The long-run buy-and-hold abnormal performance of our sample firms is similar to the performance of companies that conduct SEOs. (4)

We also examine the calendar-time performance of the sample using the Fama-French (1993) three-factor model and calendar-time control portfolios, as described in Lyon et al. (1999). Table III, Panel A, shows that the intercept in the Fama-French model using equally weighted returns is negative and significantly different from zero. The null hypothesis that :he intercept using value-weighted monthly returns is zero cannot be rejected.

Table III, Panel B, reports monthly average calendar-time abnormal returns (CTARs). We compute these returns using control portfolios that we match to the sample firms on the basis of both size and book-to-market ratios. We observe significant underperformance in both the equal-weighted portfolio (monthly excess return of -1.23%, implying an annual excess return of-14.8%) and the value-weighted portfolio (monthly excess return of -0.62%, implying an annual return of -7.7%). Again, as with the case of BHARs, we find that the long-run calendar-time abnormal performance of our sample firms is similar to the performance of companies that conduct SEOs.

Although the abnormal performances of companies following warrant calls and SEOs is similar, interpreting the economic meaning of these results is difficult. Recent research by Brav, Gezcy, and Gompers (2000), Eckbo, Masulis, and Norli (2000), Jefadeesh (2000), and Mitchell and Stafford (2000) argues that the magnitude and significance of the long-run-buy-and hold abnormal performance of companies following an SEO is driven by inadequacies in the empirical methods. Therefore, we are careful to avoid claiming that the firms in our sample underperform the market following the warrant call announcement. what is clear is that that the empirical characteristics of firms that force warrant exercise are similar in several respects to thoe of companies that conduct SEOs. Since other studies link the market reaction to the SEO announcement of a warrant call.

III. Announcement Returns.

As we have noted, a warrant call is a leverage-decreasing event that results in the acquisition of unrestricted cash. If the issuing firm lacks good investment opportunities or has spare debt capacity, a warrant call increases the agency costs of managerial discretion. The announcement residuals should reflect the portion of those costs that are unanticipated. Price-pressure effects resulting from dealer inventory adjustment could also influence announcement returns.

A. Announcement Returns and Completion Date Returns

We compute abnormal returns over the (-1,0) announcement window using the market model with a Scholes-Williams (1977) adjustment. As reported in Table IV, the average two day abnormal return is -3.12%, with a median value of-l.91%. Sixty-nine percent of the two-day announcement returns are negative. The null hypotheses that the mean and median are zero can be rejected at the 1% level of significance. Schultz (1993a) and Fields and Moore (1995) also report significant negative announcement residuals when firms call warrants. Their results, along with our own, find a market reaction to warrant call announcements that is similar to the adverse reaction to seasoned equity offerings (Jung et al., l996). (5)

Table IV also reports results for the 109 firms for which we could identify the completion date of the warrant call. The completion date is the last date that warrants can be exercised. Table IV shows that the mean two-day abnormal return on days (C-1, C), where day C is the completion date, is 2.22%, which is significantly greater than zero at the 1% level. The median completion day excess return of 1.19% is not different from zero at conventional significance levels. Schultz (1993a) reports a similar finding. As Schultz observes, this positive return cannot be attributed to new information, because the completion date is known in advance. This result leaves inventory adjustment as a likely explanation for the completion day return, and perhaps also at least a portion of the announcement date return.

As noted earlier, if inventory adjustments were entirely responsible for the announcement day return, we would expect to see a strong negative correlation between the announcement day and completion day returns. However, the correlation between these two excess returns is 0.093, so we see that there is a slight tendency for firms with lower announcement returns to also have lower completion date returns. This finding is not consistent with Schultz's (1993a) notion that inventory adjustment is entirely responsible for the negative announcement day returns.

Table IV also reports the cumulative abnormal returns for the period beginning one day prior to the warrant call announcement and extending to the completion day. The mean (median) length of time from announcement to completion is 33 (25) trading days. As argued earlier, if inventory adjustments are the sole cause of price movements around the warrant call, the total abnormal return over the warrant call period should be zero. As Table IV reports, the mean (median) abnormal return over the entire period is -7.7% (-9.0%). Both measures are significantly different from zero at the 1% level. Again, this is not consistent with the notion that inventory adjustment is entirely responsible for the negative announcement day returns.

B. Announcement Returns and Operating Efficiency

We first examine announcement returns for efficient versus inefficient firms. Since inefficient firms are more likely to add equity capital to poorly performing businesses, we hypothesize that the market reaction to forced warrant exercise will be lower for these firms. We define efficient (inefficient) firms as those with positive (negative) industry-adjusted asset turnover in the year prior to the warrant call. Industry-adjusted asset turnover is the firm's asset turnover ratio (sales/assets), minus the median asset turnover ratio in the firm's industry, defined at the two-digit SIC code level. (6)

The results presented in Table V are consistent with our hypothesis. Efficient firms have a mean announcement return of -1.81%, compared to -3.74% for inefficient firms, and the difference is significant at the 10% level. On a median basis, returns are -0.65% for efficient firms and -2.98% for inefficient firms, and the difference is significant at the 5% level.

C. Announcement Returns and Financial Leverage

Jensen (1986) and Stulz (1990) show that debt limits managerial discretion, since debt service limits free cash flow, and lenders have an incentive to monitor management to make sure they will be repaid. Any impact of forced warrant calls on agency costs should be greatest among companies with low financial leverage, because these firms are adding equity capital when spare debt capacity likely exists. Only six of our 115 sample firms have any public debt outstanding, so most of their long-term debt is privately held. Private debt holders (e.g., banks) are assumed to monitor firms more closely than public debt holders, which reinforces the notion that the high-debt firms in our sample are limited in their degree of managerial discretion.

Therefore, we examine firms with low leverage prior to the warrant call. We define low leverage as debt-to-assets (DTA) less than 10%. These firms have a mean DTA ratio of 2.04% in the year prior to the warrant call. The mean industry-adjusted DTA ratio is -17.35%, indicating that the average low-leverage firm has a DTA ratio 17 percentage points lower than the median firm in the same two-digit SIC code. Thus, the firms in the low leverage stratum have low amounts of debt on both an absolute and relative basis. If we assume that the industry median level of borrowing represents the appropriate amount of leverage, then the firms in this group used below normal levels of borrowing at the time of the forced warrant exercise.

Because a forced warrant call is a leverage-decreasing event, the acquisition of equity by the low-leverage firms cannot reduce financial leverage significantly, because it is already low. However, the extent of leverage reduction among high-leverage firms can vary, because they have differing amounts of exercise proceeds and because some firms will use the proceeds to pay down borrowing. Agency theory suggests that leverage reduction in connection with forced warrant exercise will lead to increased managerial discretion costs among firms that reduce their leverage below optimal levels. Therefore, we also examine the characteristics of two additional groups during the year of the forced warrant exercise: 1) high-leverage firms that experience significant leverage reductions and 2) high-leverage firms that continue to use significant amounts of debt. Table VI shows the characteristics of these two additional subsamples.

We first study high-leverage firms that experience a large leverage reduction in the year of the warrant call. We define a large leverage reduction as a reduction in the DTA ratio by at least a half from year -1 to year 0, where year 0 contains the warrant call announcement. These firms have an average DTA ratio of 32.48% in the year prior to the warrant call, which is reduced to 7.14% in the year of the warrant call. On an industry-adjusted basis, the mean DTA falls from 9.39% to -14.02%. Mean year 0 DTA is significantly lower than mean year -1 DTA at the 1% level, for both the unadjusted and industry-adjusted figures. If we assume that the industry median level of borrowing represents the appropriate amount of leverage, we see that the firms in this group decrease their borrowing to below-optimal levels in the year of the warrant call.

The second group contains the remaining sample firms, the high-leverage firms that do not reduce their leverage in the year of the warrant call. These firms have a mean DTA of 37.22% in year -1, and a mean DTA of 35.25% at the end of year 0. On an industry-adjusted basis, the mean DTA falls from 10.52% to 9.93%. The companies in this group have debt ratios that are high in comparison to the other sample firms and high in relation to the companies in their respective industries. Again, if we assume that the industry median level of borrowing represents the appropriate amount of leverage, we see that the firms in this group use above-normal levels of borrowing around the time of the forced warrant exercise. (7)

Table VI also reports announcement residuals for the three subsamples. The low-leverage sample experiences a mean announcement residual of -4.00%, while the leverage-reduction subsample has a mean announcement residual of -4.20%. Both of these figures are significantly different from zero at the 1% level. In contrast, the remainder of the sample, the high leverage firms that do not reduce their leverage, have a mean announcement residual of -1.28%, which is not different from zero at conventional levels of significance. Furthermore, the mean announcement residual of -1.28% for the rest of the sample is significantly different from the mean announcement residual of the combined low-leverage and leverage-reduction sub-samples, with a p-value of 0.021. This analysis suggests that the receipt of equity capital was "bad news" for companies either currently or prospectively operating at below normal levels of debt.

D. Interactions Between Financial Leverage and Operating Efficiency

In an agency framework, firms with a combination of low/reduced debt and below-normal operating efficiency should defer their warrant call, because they are susceptible to the greatest increase in managerial discretion costs.

Table VII reports summary statistics for the announcement returns to the subsample with low or reduced leverage and negative industry-adjusted asset turnover. The mean two-day announcement return for that group is -5.14%, with a median value of -3.37%. The null hypothesis of a zero mean (median) can be rejected at the 1% level. The remainder of the sample, comprised of firms with other combinations of leverage and efficiency, experienced a mean (median) announcement return of -1.36% (-0.92%). The differences between the two sample means (medians) are significant at the 1% (5%) level. Therefore, we conclude that the firms that would have been better off deferring the call experience the strongest reaction to the announcement that warrant exercise will be forced.

E. Announcement Returns and Future Operating Performance

We examine the operating performance of the sample firms beginning in the year prior to the warrant call and continuing until three years after. Table VIII contains data on industry-adjusted EBITDA/assets. We define EBITDA as earnings before interest, taxes, depreciation, and amortization (Compustat data item 13), divided by total assets. We adjust this ratio by subtracting the industry median ratio, where we define industries at the two-digit SIC code level.

As in Table VII, we divide the sample into two groups. The first subsample has low leverage and negative industry-adjusted asset turnover, and the second subsample contains the rest of the firms. As Table VII shows, only the first subsample has a significantly negative reaction to the announcement of forced warrant exercise. We examine subsequent operating performance to see whether the bad news apparently contained in the forced exercise announcement is in fact realized in subsequent operating results.

Table VIII shows that the inefficient/low leverage subsample has very poor operating performance in each of the five years we examine. In the year prior to the warrant call, the median inefficient/low leverage firm has an EBITDA/assets ratio that is 34.4 percentage points lower than the median firm in their industry. (8) The remainder of the sample firms outperform their industry by a median 0.95 percentage points. In the year of the warrant call, and in the three years after that, the performance of the median inefficient/low-leverage firm is consistently well below industry median levels. In contrast, the remainder of the sample has operating performance that closely matches industry median levels. In each of the five years we examine, we find that the median EBITDA/assets of the inefficient/low leverage subsample is significantly worse than median performance in the remainder of the sample. Thus, the poor performance of the inefficient/low leverage firms, which seems to be anticipated at the announcement o f the warrant call, is in fact realized in the years subsequent to the forced exercise of the warrants.

IV. Regression Analysis

The pattern of announcement returns suggests that the warrant call is bad news for companies with low levels of financial leverage and poor investment opportunities. We examine the announcement residuals in a cross-sectional regression framework, in order to control for other aspects of the sample that may influence announcement residuals.

A. Regression Variables

We define the following variables:

The low leverage dummy variable is equal to one if the firm belongs to the low-leverage subsample, and zero otherwise. A low leverage firm is defined as a firm with a debt-to-assets ratio of 10% or less. We use this dummy variable on the basis of the significant announcement residuals (-4%) generated by the low leverage sub sample.

The high leverage x change in industry-adjusted leverage variable controls for leverage changes within the remainder of the sample. We define high leverage as one if the firm has a debt-to-assets ratio in excess of 10%, and zero otherwise. Change in industry-adjusted leverage is equal to industry-adjusted leverage at the end of year 0, minus industry-adjusted leverage at the end of year -1, where year 0 is the year of the warrant call. The use of this variable is motivated by the results shown in Table V.

We measure our capital expenditures/total assets variable at the end of year 0, where year o is the year of the warrant call. Given the poor stock price performance of the sample firms after the warrant call, we hypothesize that firms that continue to invest heavily after the warrant call may experience poorer long run performance.

Our poor asset turnover dummy is measured at the end of year -1. Poor asset turnover is defined as one if the firm has negative industry-adjusted asset turnover and zero otherwise. Industry-adjusted asset turnover is equal to the difference between asset turnover at the end of year -1 and the asset turnover ratio of the median firm in the industry, defined by the two-digit SIC code. This variable identifies inefficient firms.

Our good asset turnover dummy is measured at the end of year -1. We define it as one if the firm has positive industry-adjusted asset turnover and zero otherwise. This variable identifies efficient firms.

We then define two variables, poor asset turnover dummy x capital expenditures and good asset turnover dummy x capital expenditures, which allows for different slope coefficients on capital expenditures/total assets for inefficient and efficient firms.

Our completion date abnormal return variable is the cumulative abnormal return experienced by the stock in window (C-1, C), where day C is the completion date of the warrant call. As noted earlier, if the only explanation for the negative announcement date returns is inventory control by dealers, then the announcement and completion day returns should be strongly and negatively correlated, since the firms with the lowest announcement-day returns should experience the greatest relief from downward price pressure on the day the warrant call is completed. Although we showed earlier that the simple correlation between and completion day returns is very small, here we control for completion day returns in a partial correlation framework.

We include a variable for redemption proceeds/MV equity to test the idea that small warrant redemption proceeds may have less of an impact on the stock price.

Our post-call three-year excess returns is the three-year buy-and-hold abnormal return, which we measure relative to a size, book-to-market, and past-return matched control portfolio. This variable controls for changes in the investment opportunity set following the warrant call.

B. Results

Table IX contains the results of our regression analysis. The coefficient on the low leverage dummy ranges between -3.0 and -4.0 and is significant, which is consistent with the univariate analysis presented in Table V. Leverage changes also matter within the high-leverage subsample. All else equal, low-leverage firms experience an adverse market reaction to the forced warrant exercise announcement. The positive coefficient on the change in industry leverage variable indicates that for firms with high leverage, announcement residuals are lower among the firms that reduce their industry-adjusted leverage. These results suggest that firms experience a more negative share price reaction to the forced warrant exercise if it leads to greater managerial discretion costs.

The coefficients on capital expenditures are also consistent with the agency hypothesis. Additional capital expenditures by inefficient firms lead to lower announcement returns. However, efficient firms that invest heavily after the warrant call avoid an incremental adverse reaction.

We find no evidence that inventory control is solely responsible for the announcement day results. The coefficient on the warrant call completion date return is never significant in any regression. We also regress the completion date abnormal returns against the independent variables in Table IX (excepting, of course, the completion date returns themselves). Our idea is that, if inventory adjustment is solely responsible for the announcement date returns, then in Table IX we have somehow identified

a set of instruments that are correlated with the tendency of dealers to lower prices on stocks of firms that call their warrants. If this is true, then the same set of instruments should be correlated with the completion day returns, when dealers raise their prices. In unreported regressions, none of the explanatory variables from Table IX had significant coefficients when regressed against completion date returns, and all regression models had insignificant F statistics. This finding lends further support to the argument that the explanatory variables in Table IX do in fact serve as proxies for the presence of agency costs.

We also find that the post-call abnormal return and size of the redemption proceeds are not significantly related to the residual.

V. Conclusion

Firms can force the exercise of outstanding warrants by calling them for redemption. Firms that force warrant exercise are typically small and on average see their share prices almost double in the year prior to the warrant call, despite concurrently poor operating performance. The announcement of the warrant call elicits a negative average abnormal return of 3%. The average residual return on the completion date is just over 2%. Given that the market knows the completion date in advance, the reversal of returns suggests that price-pressure effects explain at least a portion of these abnormal returns. However, the fact that the returns at these two dates are uncorrelated suggests that price pressure cannot be the sole explanation for the negative announcement returns.

In perfect markets, the forced redemption of warrants should not elicit a stock price reaction, because the warrant call can be fully anticipated and managers will apply the proceeds of the warrant call to value-maximizing uses. However, when there are agency costs of managerial discretion, it might be optimal to defer the warrant call. In this case, investors could react negatively to the acquisition of equity capital through a warrant call. This can occur if the benefits of expropriating the time premium of the warrants are outweighed by the loss in firm value as managers either apply the proceeds to value-destroying investments, incur costs in disgorging unneeded cash, or reduce leverage to suboptimal levels.

We examine the returns to 115 firms that force redemption of outstanding warrants. We find that announcement returns are related to variables that act as proxies for the presence of managerial discretion costs. The negative share price reaction to the announcement is concentrated among inefficient firms with low leverage. High-leverage firms experience lower abnormal returns the more they reduce their total borrowing in the year following the warrant call. Holding the impact of leverage constant, inefficient firms experience lower announcement returns the more they invest. Efficient firms fail to evince a similar relation.

Since low-leverage firms and inefficient firms are more likely to experience the agency costs of managerial discretion, our results suggest that these costs explain at least a portion of the share price reaction to a warrant call. Thus, we add to the evidence from studies on other capital acquisition events that show the valuation impact of agency costs. Our findings are consistent with previous research that shows an adverse reaction to the acquisition of equity capital.

We also call into question the use of unit IPOs to control agency costs. Schultz (1993b) argues that unit IPOs are structured to minimize issue costs and limit the amount of capital available at the offering date, which forces firms to prove themselves to investors in order to finish the capital raising process. The theory and evidence in this paper suggests that some companies may incur substantial agency costs when the warrants are called for redemption, which suggests that the unit IPO is an imperfect instrument for controlling agency costs.
Table 1

Frequency Distribution for Year of Warrant Call Announcement

This table reports the frequency of warrant calls for a sample of firms
following the announcement of the redemption of outstanding warrants.
The sample contains 115 Warrant calls announced over the period
1982-1997.

 Number of
Year Warrant Calls

1982 1
1983 4
1984 3
1985 4
1986 9
1987 7
1988 7
1989 6
1990 5
1991 11
1992 9
1993 14
1994 9
1995 17
1996 6
1997 14

Table II

Long-Run Buy-and-Hold Abnormal Returns to Firms that Calf Warrants for
Redemption

This table describeds the three-year buy-and-hold abnormal returns for a
sample of firms following the announcement of the redemption of
outstanding warrants. The sample contains 115 warrant calls announced
over the period 1982-1997. The control porfolio consists of firms
matched on the basis of size, book-to-market equity, and prior returns.
Excess return is the sample firm's long-run return minus a control
portfolio's long-run return.


Mean -38.50%
Empirical p-value 0.0000
Median -43.17%
Percent Negative 75.7%
N 103

Table III. Calendar-Time Analysis of Long-Run Excess Returns

The sample contains 115 warrant calls that were announced over the
period 1982-1997. We estimate the calendar-time return under the
Fama-French model with the following regression:

[R.sub.pt] - [R.sub.ft] = [[alpha].sub.i] + [[beta].sub.i]([R.sub.mt] -
[R.sub.ft]) + [s.sub.i][SMB.sub.t] + [h.sub.i][HML.sub.t] +
[[epsilon].sub.it]

where [R.sub.pt] is the simple average monthly return on the
calendar-time portfolio, [R.sub.ft] is the monthly return on
three-month Treasury bills, [R.sub.mt] is the return on a value-
weighted market index, [SMB.sub.t] is the difference in the returns
of a value-weighted portfolio of small stocks and big stocks, and
[HML.sub.t] is the difference in the returns to a value-weighted
portfolio of high book-to-market stocks and low book-to-market stocks.
Under the mean calendar-time approach, we calculate abnormal returns
as the difference between the monthly return to the sample firm and the
monthly return to a five-company control portfolio composed of firms
matched on the basis of prior-year return, size, and book-to-market
ratio. For n given month, we average the abnormal returns of the
companies that announced a warrant call within the prior three years.
We then average these monthly mean excess returns across the entire
sample period. The t-statistic is computed using monthly mean excess
returns that have been standardized by the within-month variance.

Panel A. Fama-French 3-factor Model Regressions

Variables Equal Weighted Value-Weighted
 Portfolio Portfolio

Intercept -1.07 ** -0.63
 (-2.30) (-1.27)
[R.sub.m] - [r.sub.f] -0.02 0.11
 (-0.16) (0.84)
SMB 0.45 ** 0.16
 (2.50) (0.82)
HML -0.15 -0.18
 (-0.67) (-0.79)
N 156 156
F -2.71 ** 1.40
[R.sup.2] 0.32 0.01

Panel B. Mean Monthly Calendar-Time Abnormal Returns

Variables Equal Weighted Value-Weighted
 Portfolio Portfolio

Mean -1.23% *** -0.62% ***
t-statistic (-5.14) (-3.00)

*** Significant at the 0.01 level.

** Significant at the 0.05 level.

Table IV

Cumulative Abnormal Returns Around Announcement (Day 0) and Completion
(Day C) of a Warrant Call

The table describes the abnormal returns for several windows, where day
0 is the announcement of a warrant call and day C is the completion of
the warrant call. The mean (median) number of trading days between day 0
and day C is 33 (25). We compute abnormal returns using the market model
with the Scholes-Williams (1977) adjustment. The sample contains 115
warrant calls that were announced over the period 1982-1997.

 AR(-1,0) AR(C-1,C) AR(-1,C)

Mean -3.12% 2.22% -7.74%
p-value 0.000 0.001 0.000
Median -1.92% 1.19% -9.00%
p-value 0.000 0.216 0.004
Percent Negative 69.6% 44.0% 64.2%
N 115 109 109

Table V

Analysis of Announcement Returns by industry-Adjusted Asset Turnover

The original sample contains 115 warrant calls that were announced over
the period 1982-1997 AR(-1,0) is the two-day announcement residual
computed using the market model with the Scholes-Williams (1977)
adjustment. Efficient firms have positive industry-adjusted asset
turnover, and inefficient firms have negative industry-adjusted asset
turnover is the firm's asset turnover ratio (sales/assets), minus the
median asset turnover ratio in the firm's industry, defined at the
2-digit SIC code level.

 Efficient Firms Inefficient Firms
 (Positive Industry- (Negative Industry-
 Adjusted Asset Adjusted Asset
 Turnover) Turnover)

Mean AR(-1,0) -1.81% -3.74%
p-value 0.027 0.000
Median AR(-1,0) -0.65% -2.98%
p-value 0.349 0.000
Percent Negative 57.5% 75.7%
N 40 70

 p-value for Test of
 Significance of
 Difference in Means
 (Medians)

Mean AR(-1,0) 0.080
p-value
Median AR(-1,0) 0.049
p-value
Percent Negative
N

Table VI

Analysis of Announcement Returns by Debt-to-Assets (DTA) Ratio

The original sample contains 115 warrant calls that were announced over
the period 1982-1997 There are 107 companies with complete event-year
data on financial leverage. AR(-1,0) is the two-day announcement
residual computed using the market model with the Scholes-Williams
(1977) adjustment. Firms with low financial leverage have DTA ratios of
10% or less at the beginning of the year in which forced exercise
occurs. Firms with a large leverage reduction are companies outside of
the low leverage group that cut their DTA by half or more during the
year in which forced exercise occurs. The remaining 43 companies do not
have low leverage and do not experience a significant reduction in
leverage over the event year. Industry-adjusted DTA is the firm's DTA,
minus the median DTA in the firm's industry, defined at the 2-digit SIC
code level.

 All Firms Firms with Firms with
 with Low Large
 Leverage Financial Leverage
 Data Leveraged Reduction

Mean AR (-1,0) -2.95% -4.00% -4.20%
p-value 0.000 0.000 0.000
Median AR (-1,0) -1.89% -2.37% -2.50%
Percent Negative 69.2 73.7 69.2

Year -1 Mean DTA 23.17% 2.04% 32.48%
Year 0 Mean DTA 17.88% 5.25% 7.14% ***

Year -1 Median DTA 20.18% 0.19% 27.25%
Year 0 Median DTA 9.58% 1.94% 4.42% ***

Year -1 Mean industry-adj. DTA 0.35% -17.35% 9.39%
Year 0 Mean industry-adj. DTA -3.84% -14.22% -14.02%

Year -1 Median industry-adj. DTA -1.97% -16.33% 12.01%
Year 0 Median industry-adj. DTA -10.28% -12.46% -12.95% ***

Number of Firms 107 38 26


 Remainder
 of the
 Sample (a)

Mean AR (-1,0) -1.28%
p-value 0.136
Median AR (-1,0) -1.19%
Percent Negative 65.1

Year -1 Mean DTA 36.22%
Year 0 Mean DTA 35.25%

Year -1 Median DTA 35.65%
Year 0 Median DTA 31.55%

Year -1 Mean industry-adj. DTA 10.52%
Year 0 Mean industry-adj. DTA 9.93%

Year -1 Median industry-adj. DTA 6.73%
Year 0 Median industry-adj. DTA 7.22%

Number of Firms 43

(a)The mean AR(-1,0) for the "remainder of the sample" is significantly
different from the mean AR(-1,0) for the combined "low-leverage" and
"leverage-reduction" subsamples. A t-test for the difference in means
has a p-value of 0.021.

*** Significantly different from year -1 mean (median) DTA at the 0.01
level.

Table VII

Analysis of Announcement Returns by Debt-to-Assets (DTA) Ratio and
Industry-Adjusted Asset Turnover

The original sample contains 115 warrant calls that were announced over
the period 1982-1997 There are 107 companies with complete event-year
data on financial leverage. AR(-1,0) is the two-day announcement
residual computed using the market model with the Scholes-Williams
(1977) adjustment. Firms with low financial leverage have DTA of 10% or
less at the beginning of the year in which forced exercise occurs. Firms
with a large leverage reduction are companies outside of the low
leverage group that cut DTA by half or more during the year in which
forced exercise occurs. Industry-adjusted asset turnover is the firm's
asset turnover ratio (sales/assets), minus the median asset turnover
ratio in the firm's industry, defined at the 2-digit SIC code level.

 Firms with Low or
 Reduced Leverage and
 Negative Industry-
 Adjusted Asset Turnover All Other Firms

Mean AR (-1,0) -5.14% -1.36%
p-value 0.000 0.061
Median AR (-1,0) -3.37% -0.92%
p-value 0.000 0.082
Percent Negative 80.0% 61.3%
N 45 62

 p-value for Test of
 Significance of
 Difference in Means
 (Medians)

Mean AR (-1,0) 0.008
p-value
Median AR (-1,0) 0.015
p-value
Percent Negative
N

Table VIII

Operating Performance in the Years Surrounding the Forced Warrant
Exercise

The original sample contains 115 warrant calls that were announced over
the period 1982-1997 There are 107 companies with complete event-year
data on financial leverage. EBITDA/Assets is earnings before interest,
taxes, depreciation and amortization divided by prior year assets. We
adjust this ratio by subtracting industry median EBITDA/assets, where we
define industries at the two-digit SIC code level. Firms with low
financial leverage have debt to asset ratios of 10% or less at the
beginning of the year in which forced exercise occurs. Firms with a
large leverage reduction are companies outside of the low leverage group
that cut the ratio of debt to assets by half or more during the year
that exercise is forced. Inefficient firms are firms with negative
industry-adjusted asset turnover, which is the firm's asset turnover
ratio (sales/assets), minus the median asset turnover ratio in the
firm's industry, defined at the 2-digit SIC code level.

 Industry-Adjusted EBITDA/Assets

 Year -1 Year 0 Year 1

Mean for Inefficient/Low and -49.63% -70.54% -45.59%
Reduced Leverage Firms
Mean for Remainder of Sample -5.23% -3.44% -11.40%

t for Difference in Means -4.97 *** -1.78 % -3.32 ***
Median for Inefficient/ Low and -34.40% -35.80% -30.95%
Reduced Leverage Firms
Median for Remainder of Sample 0.95% -0.60% -2.70%

z for Difference in Medians -6.01 *** -5.44 *** -4.07 ***

 Industry-Adjusted
 EBITDA/Assets

 Year 2 Year 3

Mean for Inefficient/Low and -41.79% -57.81%
Reduced Leverage Firms
Mean for Remainder of Sample -3.78% -5.69%

t for Difference in Means -3.37 *** -1.50
Median for Inefficient/ Low and -17.80% -15.20%
Reduced Leverage Firms
Median for Remainder of Sample -0.40% -0.60%

z for Difference in Medians -4.15 *** -2.65 ***

*** Significant at the 0.01 level.

* Significant at the 0.10 level.

Table IX

Ordinary Least Squares Regressions in which the Dependent Variable is
the Cumulative Abnormal Return on Days (-1,0) Surrounding the
Announcement of a Warrant Call The original sample contains 115 warrant
calls that were announced over the period 1982-1997. Low leverage dummy
equals one if a firm had low leverage (debt-to-assets less than 10%)
before the warrant call. High leverage dummy equals one if a firm had
high leverage before the warrant call. Poor (good) asset turnover dummy
equals one if industry adjusted asset turnover was negative (positive)
in the year before the warrant call.

Independent Models

Variables (1) (2) (3)

Intercept -0.015 -0.010 -0.025
 (-1.59) (-0.96) (-3.26) ***
Low Leverage Dummy -0.034 -0.032
 (-2.29) ** (-2.14) **
Low Leverage Dummy x Industry- 0.021
Adjusted Asset Turnover (2.14) **
High Leverage x Change in 0.002 0.002 0.001
Industry Adjusted Leverage (2.48) *** (2.63) *** (2.02) **
Poor Asset Turnover Dummy X -0.107
Capital Expenditures/Assets (-l.94) **
Good Asset Turnover Dummy X -0.028
Capital Expenditures/Assets (-0.26)
Completion Date Excess Return 0.093 0.100 0.100
 (0.93) (0.98) (0.99)
Redemption Proceeds/MV Equity

Post-call 3 Year Excess Returns

F 3.l0 ** 2.34 ** 2.88 **

Adjusted [R.sup.2] 0.059 0.063 0.053

N 100 100 100

Independent Models

Variables (4)

Intercept -0.010
 (-0.96)
Low Leverage Dummy -0.032
 (-2.14) **
Low Leverage Dummy x Industry-
Adjusted Asset Turnover
High Leverage x Change in 0.002
Industry Adjusted Leverage (2.15) **
Poor Asset Turnover Dummy X -0.140
Capital Expenditures/Assets (-1.65) *
Good Asset Turnover Dummy X -0.028
Capital Expenditures/Assets (-0.26)
Completion Date Excess Return 0.100
 (0.98)
Redemption Proceeds/MV Equity -0.022
 (-0.45)
Post-call 3 Year Excess Returns 0.005
 (0.89)
F 1.88 *

Adjusted [R.sup.2] 0.063

N 72

*** Significant at the 0.01 level.

** Significant at the 0.05 level.

* Significant at the 0.10 level.


(1.) There are two other information-based theories that explain the market's negative reaction to the announcement of a seasoned equity offering. The timing model holds that value-maximizing managers will issue new shares when they perceive them to be overvalued. However, the timing model cannot explain a negative share price reaction to a warrant call, because the firm has already agreed to sell the shares at a fixed price. Thus, regardless of whether managers perceive the shares to be fairly, over-, or undervalued, they will call the warrants to expropriate the time premium. Investors will not be able to infer anything about the stock price's relation to fair value by observing the warrant call. The other information-based theory is the pecking order theory of Myers and Majluf (1984). However, the pecking order theory explains the choice of securities (debt or equity) to be issued by the firm and has no implications for the optimal warrant call policy.

(2.) Other studies of capital acquisition and managerial discretion costs find that debt reduction is a more frequently expressed intention for use of proceeds. For example, Lang et al. (1995) and Allen and McConnell (1998) report that many firms that undertake asset sales or equity carveouts intend to return the proceeds to debtholders or shareholders. However, both studies also report that a typical firm in their sample is highly levered or has suffered poor operating performance. In contrast, firms that call their warrants have, by definition, recently experienced good stock price performance (to drive the stock price above the exercise price) and so are less likely to be distressed firms.

(3.)We select, for each sample firm, one randomly chosen firm from the same size and BM reference portfolio. We assume that the randomly selected company called warrants for redemption, and we compute its excess three-year return as described above. We compute excess returns for all of the companies in the sample, which yields an average excess return to the "pseudo-warrant call" sample. We repeat the process 10,000 times. The empirical p-value measures the percentage of times that the average excess return to the randomly drawn sample was less than the average excess return to the sample of companies that actually called their warrants.

(4.) Loughran and Ritter (1995), Spiess and Affleck-Graves (1995), and Lee (1997) report that firms that complete SEOs experience three-year excess buy and hold returns between -13.8% and -22.8%.

(5.) In contrast, Ederington and Gob (2001) do not find a permanent stock price reaction to a conversion-forcing call of a convertible bond. Although both a warrant call and the call of a convertible bond result in a decrease in leverage, only a warrant call results in an inflow of cash.

(6.) It would be interesting to examine the relation between managerial shareholdings and the announcement residuals. Unfortunately, the dominance of small firms within the sample and the subsequent poor performance of many companies within it have significantly reduced the availability of that information. The literature does suggest, however, that the impact of the missing data is minimal. Ang, Cole, and Lin (2000) use asset turnover ratios to measure agency costs and find a significant inverse relation with managerial shareholdings. In light of their finding that managerial ownership varies with operating efficiency in the cross-section, we are confident that our use of industry-adjusted asset turnover ratios captures much of the information that would be provided by managerial shareholdings.

(7.) One possibility for the difference in post-warrant call leverage ratios between the leverage reduction and high leverage subsamples is that leverage reduction subsample obtains a larger amount of equity from the warrant call compared to the high leverage subsample. The average value of warrant call proceeds/book value of equity for the leverage reduction sample is 137%, but for the high leverage subsample it was 71%. Measured as a percentage of the market value of equity, these figures are 18.8% and 19.4%, respectively.

(8.) This extremely poor operating performance is, surprisingly, not reflected in the pre-announcement stock price performance of the sample firms. The inefficient/low leverage subsample had an average market-adjusted buy-and-hold return of +72% in the year prior to the warrant call, compared to +85% for the remainder of the sample.

References

Allen, J.W. and J.J. McConnell, 1998, "Equity Carve-Outs and Managerial Discretion," Journal of Finance 53, 163-186.

Ang, J.S., R.A. Cole and J.W. Lin, 2000, "Agency Costs and Ownership Structure," Journal of Finance 55, 81-106.

Bayless, M. and S. Chaplinsky, 1991, "Expectations of Security Type and the Information Content of Debt and Equity Offers," Journal of Financial Intermediation 3, 195-214.

Brav, A., C. Geczy, and P. Gompers, 2000, "Is the Abnormal Return Following Equity Issuances Anomalous?" Journal of Financial Economics 56, 209-250.

Eckbo, B.E., R.W. Masulis, and [empty set]. Norli, 2000, "Seasoned Public Offerings: Resolution of the 'New Issues Puzzle,'" Journal of Financial Economics 56, 251-292.

Ederington, L.H. and J.C. Goh, 2001, "Is a Convertible Bond Call Really Bad News?" Journal of Business 74, 459-476.

Fama, E.F. and K.R. French, 1993, "Common Risk Factors in the Returns on Stocks and Bonds," Journal of Financial Economics 43, 3-56.

Fields, L.P. and W.T. Moore, 1995, "Equity Valuation Effects of Forced Warrant Exercise," Journal of Financial Research 18, 157-170.

Howe, J.S. and Su, T., 2001, "Discretionary Reductions in Warrant Exercise Prices," Journal of Financial Economics 61, 227-252.

Ingersoll, J., 1977, "An Examination of Corporate Call Policies on Convertible Securities," Journal of Finance 32, 463-478.

Jegadeesh, N., 2000, "Long-Term Performance of Seasoned Equity Offerings: Benchmark Errors and Biases in Expectations," Financial Management 29, 5-30.

Jensen, M.C., 1986, "Agency Costs of Free Cash Flow, Corporate Finance, and the Market for Takeovers," American Economic Review 76, 323-329.

Jung, K., Y. Kim, and R.M. Stulz, 1996, "Timing, Investment Opportunities, Managerial Discretion, and the Security Issue Decision," Journal of Financial Economics 42, 159-186.

Lang, L.H.P., A. Poulsen, and R.M. Stulz, 1995, "Asset Sales, Firm Performance and the Agency Costs of Managerial Discretion," Journal of Financial Economics 41, 3-38.

Lee, I., 1997, "Do Firms Knowingly Sell Overvalued Equity?" Journal of Finance 52, 1439-1466.

Loughran, T. and J.R. Ritter, 1995, "The New Issues Puzzle," Journal of Finance 50, 23-52.

Loughran, T. and J.R. Ritter, 2000, "Uniformly Least Powerful Tests of Market Efficiency," Journal of Financial Economics 55, 361-389.

Lyon, J.D., B.M. Barber, and C. Tsai, 1999, "Improved Methods for Tests of Long-Run Abnormal Stock Returns," Journal of Finance 54, 165-202.

McConnell, J.J. and C.J. Muscarella, 1985, "Corporate Capital Expenditure Decisions and the Market Value of the Firm," Journal of Financial Economics 14, 399-422.

McLaughlin, R., A. Safieddine, and G.K. Vasudevan, 2000, "Investment Banker Reputation and the Performance of Seasoned Equity Offerings," Financial Management 29, 96-110.

Mitchell, M.L. and E. Stafford, 2000, "Managerial Decisions and Long-Term Stock Price Performance," Journal of Business 73, 287-330.

Myers, S. and N. Majluf, 1984, "Corporate Financing and Investment Decisions When Firms Have Information That Investors Do Not Have," Journal of Financial Economics 13, 18-221.

Scholes, M. and J. Williams, 1977, "Estimating Betas from Nonsynchronous Data," Journal of Financial Economics 5, 309-328.

Schultz, P., 1993a, "Calls of Warrants: Timing and Market Reaction," Journal of Finance 48, 681-696.

Schultz, P., 1993b, "Unit Initial Public Offerings: A Form of Staged Financing," Journal of Financial Economics 34, 199-230.

Spiess, D.K. and J. Affleck-Graves, 1995, "Underperformance in Long-Run Returns Following Seasoned Equity Offerings," Journal of Financial Economics 38, 243-267.

Stulz, R.M., 1990, "Managerial Discretion and Optimal Financing Policies," Journal of Financial Economics 26, 3-27.

Helpful comments were received from an anonymous referee, John Howe, the Editors. and seminar participants at the University of Missouri-Columbia and the University of Missouri-St. Louis.

Michael J. Alderson and Brian L. Betker *

* Michael J. Alderson and Brian L. Betker are Professors of Finance at Saint Louis University.
COPYRIGHT 2003 Financial Management Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2003 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Alderson, Michael J.; Betker, Brian L.
Publication:Financial Management
Date:Mar 22, 2003
Words:9717
Previous Article:Survival and default of original issue high-yield bonds.
Next Article:Collusion and financial leverage: an analysis of the integrated mill steel industry.
Topics:


Related Articles
Shareholder dividend options.
Abatement of interest final regs.
Perceived managerial discretion: an analysis of individual ethical intentions.
Warrants issued to customers.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters