Mosaic ImmunoEngineering is a nanotechnology-based immunotherapy company developing therapeutics and vaccines to positively impact the lives of patients and their families.
Message: from a rb poster
You really want to know about reverse splits? or stay stupid?
When is a Reverse Split Good News?
Wealth Effects and Performance Implications of Reverse Splits
ABSTRACT
In this paper, we examine the motivations for and signaling implication of reverse
splits. Reverse splits are so named because unlike forward splits, which increase
the number of shares outstanding and decrease the price by some amount, they
reduce the number of shares outstanding and increase the share price by some
factor. Reverse splits are of interest because they provide a decidedly negative
signal to the market. They are a confirmation of poor performance particularly
when they are enacted for the purpose of maintaining a listing on an exchange
or in order to obtain a price at which institutional investors can buy the stock. Our
results indicate that the market views reverse splits unfavorably and that reverse
splits appear only to delay the inevitable delisting of many sample firms.
When is a Reverse Split Good News?
Wealth Effects and Performance Implications of Reverse Splits
1. Introduction
In the past couple of years, reverse splits have become very popular, with
well even such a well known firm as AT&T completing 1 for 5 split in November
2002. Many of the firms that went public in the 1990s during the tech/internet
boom have also engaged in reverse splits as their share prices plummeted
dramatically. Over the years, much has been written in finance literature about
forward splits (see Rozeff (2003) for a comprehensive long-term study of forward
splits). Much less has been written about reverse splits. In the case of forward
splits, the number of shares is increased and the share price is proportionately
decreased. With a reverse split, the number of shares is decreased and the
price per share is proportionately increased. While a split by itself should have
no effect on the value of the stock, it may convey signals about the future
prospects of the firm.
In this paper, we examine the motivations for and the signaling/valuation
implications of reverse splits. Thus, if a forward split is seen as providing positive
information about the future of a firm, a reverse split would imply just the
opposite. In fact, as we shall show reverse splits provide a decidedly negative
signal to the market in that they are a confirmation of poor performance. This is
particularly the case when they are enacted for the purpose of maintaining a
listing on an exchange because the firm doing the reverse split has not been able
to meet the listing requirements. Reverse splits may similarly reflect poor
performance when they are done or in order to obtain a price at which
institutional investors can buy the stock.
In this study, we use a sample of 1702 reverse split from 1992 through
2001, which is the largest sample of reverse splits ever studied. We find that as
shown in prior studies, the announcement of a reverse split results in significant
negative abnormal returns to shareholders, on the announcement dates,
approval dates, and the split ex-date. Cumulative abnormal returns upon
announcement are significantly negative. We find that firms enacting reverse
splits suffer from poor operational performance prior to the reverse split.. We also
find that announcement date abnormal returns are significantly positively related
to profitability, leverage, and the size of the split. Finally, we find that the long run
viability and accounting performance of reverse split firms is poor.
The rest of this paper is organized as follows. Section 2 reviews the
existing literature on forward and reverse splits. Section 3 describes our sample
and research approach. Section 4 provides sample descriptive data. Section 5
provides results, and section 6 concludes the paper.
2. Theoretical framework
2.1. Signaling implications of managerial actions
The literature on signaling, asserts that management’s actions, in addition
to having an objective in and of themselves, are used to transmit information to
shareholders regarding the future cash flow implications of the firm and/or the
firm’s systematic risk. The best signals of future market value improvement are
those which are unambiguous and can be relatively cheaply and effectively
transmitted to the market, but which cannot be mimicked by a poor performer.
Two such “credible signals” are share repurchases and dividend increases; firms
with managers who have superior information regarding poor prospects will be
loathe to buy overvalued stock or effectively commit to a higher dividend knowing
it cannot be permanently maintained. Accordingly, the literature indicates that the
positive abnormal returns observed upon announcement of a share repurchase
or an increase in dividends are consistent with the market’s perception of a
positive signal regarding future cash flows (see, eg., Dittmar, 2002).
Empirical evidence showing positive event study returns suggest that the
forward splits provide some positive signal to shareholders (Desai and Jain,
1995, Dittmar, 2002, Brennan and Copeland, 1988). Forward splits appear to
move the firm into a more marketable trading range (Baker and Gallagher, 1980).
Positive wealth gains are observed on the announcement date and split ex date
but not on the approval date; apparently, for the signal to be effective, it does not
need to be repeated (Kryzabowski and Zhang, 1991, 1993).
Despite the abundance of research on forward splits, very little has been
done regarding the motivation for and wealth gains or losses from reverse splits.
Descriptively, reverse splits are “the substitution of one new share for a certain
number of outstanding shares” (Han, 1995); they are nothing more than a
mechanism for reducing the number of shares outstanding and increasing the
price by some stated factor. Spudeck and Moyer (1985) argue that reverse splits
are a negative signal because they imply that the only mechanism by which the
firm can improve the stock price is through largely artificial means. Empirical
evidence on reverse splits implies the market views them as a negative signal
about future performance (Brennan and Copeland, 1988; Lamoureau and Poon,
1987).
Han (1995) raises the questions of why managers engage in reverse splits
and why, if they are wealth destructive, shareholders approve such actions,
especially when, as Peterson and Peterson (1992), indicate they are not required
by the exchanges on which the firms are traded. Several reasons have been
advanced: the requirement for marginability, the signal presented by a low stock
price and the demonstrable lack of interest in such securities on behalf of
institutions (Falkenstein, 1998), and irrationality. Under Federal Reserve Board
initial margin requirements rules, stocks selling for less than a dollar are not
marginable. Most institutional investors will not by stocks that are selling for less
than $5, and some avoid stocks selling for less than $10.
Woodridge and Chambers (1983) examine up to 54 reverse splits for
NYSE/AMEX stokcs occuring between 1962 and 1981. Mean-adjusted excess
returns are significantly negative following the announcement, approval and
effective dates. Lamoureux and Poon (1987) look at forward and reverse splits
on NYSE/AMEX. Using a sample of 49 reverse splits in the period July 1962
through December 1985, they find that volatiltiy decreases, liquidity increases
following this corporate action. Peterson and Peterson (1992) examine 196
reverse splits over the period July 1962 and December 1989 for all listed stocks.
They find that discretionary reverse splits have negative excess returns, while
that non-discretionary splits result in positive excess returns. Furthermore, they
find that .these firms’ total risk declines following the RS, while systematic risk
does not change. They also argue that marketability is enhanced resulting from
a higher price range. Han (1995) looks at the effect of reverse splits on the
liquidity of 136 NYSE/AMEX/NASDAQ firms from 1963 through 1990. He finds
there is a decrease in spreads, an increase in trading volume and a signifnicant
decline in the number of non-trading days following reverse splits, all indications
of increased liquidity. Using event study methodology, he also finds a negative
reaction to reverse splits. Finally, Desai and Jain (1997) long at the long term
effect of 5,596 forward splits and 76 reverse splits over the period 1976-1991.
The stocks are listed on all exchanges. For the latter, they find very large and
significant negative abnormal returns 1, 2 and 3 years following the RS.
Reverse splits are often associated with corporate failure. An example is
the Bank of San Francisco – described as a “black hole” for its fiscal profligacy
and poor performance – which used a reverse split combined with a name
change to “reinvent itself” (O’Hara, 1994). Companies which do reverse splits
sometimes fail to meet the listing requirements on float and later have to issue
more shares1. However, not all the motives for reverse splits appear at first
glance to be wealth destructive and capable of generating the kind of negative
price reaction observed in the literature. Reverse splits may be a signal to the
market that the firm is a candidate for a takeover, due to the fact that fewer
shareholders would have to be convinced to tender their shares2 Reverse splits
are also used: to eliminate large numbers of small odd lot holders, to deregister a
1 For instance, Greenbriar Corp announced after its reverse split, “The combination of a lower
number of shares outstanding and the unexpected demand have resulted in occasional shortages
in the number of shares available for sale. The Company has undertaken with the American
Stock Exchange that it will adjust its number of shares outstanding by means of a stock dividend
during the first quarter to provide at least 200,000 shares in the public float, which is the minimum
standard of the Exchange.”
2 An example of the use of a reverse split to signal takeover candidacy is San Francisco’s Pacific
Bank, where management informed shareholders that the split could likely result in a change of
company and to take it private. The latter is accomplished by making the reverse
split so severe that only a few shareholders are left with more than a single
share.
This is the first study to examine reverse splits with a sample following
significant changes in NASDAQ listing requirements starting August 22, 1997.
For firms which cannot improve earnings quickly enough to keep the listing, the
reverse split may be the only way to get the stock price up to the minimum
required level. In cases where the motivation is to maintain listed status, the
reverse split announcement reminds the public that the firm is in danger of
delisting and has no alternative mechanism to keep its listing. In addition to
looking at the price reaction following a reverse split, this study examines reverse
splits motivated by limits to qualify as a marginable security, and price restrictions
on institutional holdings to see if the stated objective of the reverse split is ever
achieved. This is an area that has not been studied before. Our study is the first
to examine the relationship between motive and performance outcome for
reverse splits, and sheds new light on the signal they provide to the market.
3. Research Design
3.1. Sample Selection
We obtain announcements of reverse splits from the Center for Research
in Security Prices (CRSP) file, by identifying all distributions for the period 1992
to 2001 which are coded as reverse splits. In order to obtain information
regarding the motive for the reverse split, and to identify any confounding
control of the bank (Calbreath,1993).
announcements occurring the same day as the announcement of the reverse
split, we obtain the corresponding news announcements from the Lexis-Nexis
database. We obtain accounting data for our sample firms from Standard and
Poor’s Research Insights for firms engaging in reverse takeovers. Stock return
data are also obtained from CRSP. Corporate events prior to the reverse split as
well as post-reverse split performance and delisting information are obtained
from searches in Lexis/Nexis, annual reports, proxy statements, First Call,
Market Guide, and Research Insights. This extensive search process resulted in
1,072 announcements of reverse splits. Our sample is the largest ever used in a
study of reverse splits. The largest previous sample used contained 483 reverse
splits and covered a time period previous to the one we are examining. In
addition, we are able to examine reverse splits motivated by changes in
NASDAQ listing requirements.
We also conduct empirical tests of our sample firms vs. a control sample
of firms that did not do reverse stock splits. The control sample is obtained by
selecting the closest match in terms of size (assets) and book/market within the
three digit SIC code of the sample firm.
Table 1, Panels A , provide sample descriptive characteristics and provide
evidence regarding the frequency of reverse splits over time. The results
indicate that reverse splits decreased dramatically in 1993 and increased
dramatically in 1998. That increase continued in subsequent years so that
41.7% of all the splits in our sample occurred in the four years 1998 through
2001. Panel B shows that most firms underwent only one reverse split.
However, 10 firms had three reverse splits, while 2 had 4. It is clear that for
some firms, reverse splits cannot save a stock price from continuing to fall when
the underlying business prospects are very poor.
In Panel C, we provide information on the number and percent of reverse
splits by industry. Two industries, SIC 3000 and SIC 7000 accounted for more
than 40% of all reverse splits. It is not clear why these two industries should
account for such a large proportion of reverse splits. Panel D show the number
and frequency of splits by exchange listing. By far the largest number and
proportion are stock that traded on NASDAQ, 871 and 81.125%, respectively.
This is not surprising given the fact that newer and smaller firms tend to list on
NASDAQ, as the listing requirements are not as stringent as those for the NYSE
or AMEX. Interestingly, in August 1997, NASDAQ increased the requirements
to remain listed. The net tangible asset requirement was doubled under one
method and the market value of the float was increased by $1 million to $11
million again depending on the method used to determine eligibility to remain
listed. It would appear that these change had a lot to do with the increase in
reverse splits starting in 1998.3
[Table 1 About Here]
3.2. Research methods
3.2.1 Event study
To test the market’s reaction to the announcements of reverse stock splits,
we use event study methodology with the significance tests based on the
3 We would like to thank NASDAQ for providing the listing requirements for the time period of this
study.
standardized cross-sectional method (Boehmer, Musumeci and Poulsen, 1991)
with Scholes-Williams betas. Returns are modeled using ordinary least squares.
A 100-day estimation period was chosen from t-110 to t-11 where t=0 is the
event date.4 Cumulative abnormal returns (CARs) are reported for the (-1,+1)
and (-1,0) event windows. We attempt to explain the source of value gains or
losses with cross-sectional regressions on CARs.
3.2.2 Buy and hold returns analysis
In order to examine the long run returns of firms engaging in reverse
splits, we utilize the Barber-Lyon (1997) methodology. We hypothesize that firms
employing reverse splits for discretionary reasons may experience improvements
in long horizon returns, given that they survive and are not delisted. We estimate
matching sample-adjusted long-horizon average holding-period abnormal returns
(AHAR) for our sample bidders. AHARs are calculated starting with the month
after the announcement date for 6, 12, and 18 month holding periods.
3.2.3 Changes in Risk
Like Petersen and Petersen (1992), we hypothesize that the reverse split
may have consequences for the risk profile of the firm. For each firm, we
estimate the changes in its risk characteristics surrounding the announcement of
the reverse split. The pre-announcement risk measure, VARPRE [postannouncement
risk measure, VARPOST ] is estimated over the period from t-110
4 We varied the parameter estimation period from 100 to 150 days. Results similar to the ones
reported were obtained. In some cases our total sample includes more than one announcing firm
for the same date. Accordingly, our significance tests adjust for cross-sectional dependence in
the regression residuals. We also adjust for serial dependence as well as for changes in event
induced variance (see the manual for Eventus, version 7.0). CARS are based on an equally
to t-11 [t+11 to t+110], where t=0 is the announcement date for both sample firms
and matched firms. We estimate (a) total risk, or Var (Ri ), where Ri is the security
return on the ith security, and (b) systematic risk, or βi.
4. Results
4.1. Characteristics of firms which engage in reverse splits
Based on the existing literature, we would anticipate finding that firms
engaging in reverse splits are poor performers which have adopted this particular
strategy as a result of actions taken over time. These have translated into the
firm being subject to being delisted. Descriptive statistics for the year prior to the
reverse split, provided in Table 2, indicate that this is in fact the case. Firms
engaging in reverse splits are, on average, very small with a mean (median)
sales of $175 ($14) million, asset size of $239 ($19) million, and market value of
equity of $119 (25) million. In addition, these firms are very unprofitable with a
mean (median) net profit margin of -30.09% (-8.91%), return on assets and
equity of –27.91 (-11.46%) and –32.29% (-12.27%), respectively. However, their
profitability ratios were not substantially different from their control counterparts,
indicating that perhaps the threat of delisting arose from adverse price
movements within the sector. Reverse split firms have negative cash flow as
well; they exhibit significantly lower free cash flow to total assets than their
control counterparts.
[Table 2 About Here]
weighted market portfolio. Similar results were obtained with the value weighted market portfolio.
Table 2 also indicates that being overlevered was not the likely reason that the
sample firms performed poorly; their debt/asset ratios do not differ significantly
from control firms. Nor was the liquidity of sample firms, measured by cash/total
assets, significantly different from control firms. From an accounting perspective,
sample firms were not nearer bankruptcy than control firms; their Z scores were
insignificantly different. Again, this supports the conjecture that reverse split firms
may have experienced share price declines as a result of adverse industry
conditions.
4.2. Event study results
Event study results shown in Table 3 indicate that the market does not respond
favorably to announcements of reverse splits; mean event date abnormal return
is -5.61 percent, which is significant at the 1% level. The results support the idea
the reverse split is in general a desperate measure taken by poorly performing
managers, and that it is not viewed positively by the market. Panel B breaks
down the sample by motives for doing the reverse split. The most significant
factors are threat of delisting, news release with no reason given, no news
release prior to the reverse split announcement, capital structure change, and
other reasons. The abnormal returns are significantly negative when the stated
reason for the split is delisting (a nondiscretionary reason), when the
management does not give a reason for the split, and when there is no news
release regarding the split at all.
[Table 3 About Here]
The results for reverse splits conducted for discretionary reasons – to attract
institutional investors by moving the price over the $5 limit on institutional
purchases, or as part of a reorganization, are very different. We observe that
when the motive is to attract institutional investors, the abnormal returns are
significantly positive and generate returns of 2.64% over the (-1,+1) event
window. The results for the reorganization motive are mixed; on the
announcement date itself, the abnormal returns are negative, but over the (-1,0)
window, they are significantly positive. It appears that corrective, rather than
survival, motives for firms doing reverse splits, generate wealth for shareholders,
and the differences in abnormal returns are significantly higher for corrective than
survival motives.
4.3. Multivariate Regression Analysis
We next examine the factors that affect reverse split announcement
returns. The results of these regressions are shown in Table 4. Two models are
shown for the purposes of demonstrating the robustness of results. All three
models are significant at the 1% level, as shown by their F-tests.
[Table 4 About Here]
In Model 1, we regress three day CARS on a liquidity, as measured by cash/total
assets, leverage, as measured by debt/total assets, the magnitude of the split,
profitability, as proxied by return on assets, and liquidity, as proxied by trading
volume. The coefficient of cash/assets is negative but insignificant, indicating that
the market does not reward firms with higher liquidity or penalize those with lower
liquidity. The coefficient of debt to total assets is positive and significant,
indicating the market perceives that firms with higher leverage benefit more from
the reverse split than those with lower leverage. Perhaps shareholders of firms
with high leverage feel that management is signaling a sale of additional shares,
which are easier to sell now with listing maintained and having a price high
enough to attract institutional investors. Return on assets is positive but
insignificant. Maybe shareholders see the firm as temporarily in distress, and if
it can just get the share price up and attract institutional investors, it can remain
listed. Trading volume is negative and insignificant, a surprising result since
liquidity and low volume were two commonly cited reasons managers give for
reverse splits.
The size of the split factor is also positive and significant. In CRSP,
reverse splits have a negative FACSHR value, so the larger the reverse split, the
more negative the FACSHR value will be. The positive significant coefficient on
the size of split variable indicates that more dramatic reverse split events appear
to be signs of increasing desperation, ie, a firm which has to do a 1:100 reverse
split is in worse shape than one which merely needs to double its share price to
meet the exchange listing requirements, and hence, is more likely to pull through
distress.
The results of models 2 and 3 are quantitatively similar. Size, as proxied
by market value, is insignificantly different from zero. Liquidity is significantly and
positively related to CARs; perhaps the market perceives that firms with higher
liquidity are more likely to pull through a momentary period of distress if they can
remain listed. However, trading volume is insignificant, indicating that abnormal
returns are not driven by concerns over the liquidity of the shares being traded.
Overall, the results suggest that while reverse splits are seen as evidence of
wealth destruction on average, the shareholders of firms doing reverse splits
view the performance outcomes more favorably for more profitable firms who use
greater relative amounts of debt capital.
4.4. Post-reverse split performance
The stated objectives of managers when they announce a reverse stock
split include more visibility to institutions, maintaining a listing, becoming a
marginable security, and signalling takeover target status. To answer the
questions of whether reverse split firms are successful in achieving these stated
intentions, and whether or not management is able to parlay the reverse split into
long term wealth maximization, we examine the long run survival of reverse split
firms as well as whether the firm achieved what the reverse split was supposed
to achieve. The results of this analysis are presented in Table 5
[Table 5 about here]
Managers of firms doing reverse splits often express tremendous
optimism regarding the firms’ future profitability, as one would expect in order to
prevent a large sell-off of shares. To some degree, this happens as shown in
Panel A. Return on assets and net profit margin increase significantly (1%
lelvel). However, the results in Panel B of Table 5 suggest that firms doing
reverse splits, are about as likely to be delisted within six months following the
completion of the reverse split as they are to survive. Only 37.3% are still listed
two years following the completion of the reverse split.
For those firms trading on NASDAQ, we also wanted to look at the six,
twelve and eighteen month equtiy returns following reverse splits, taking into
account two different listing regimes used by NASDAQ. As shown in Panel C,
returns are significantly negative (5% level) six months after the reverse split for
the total sample and listing regime 2. Returns are negative but not significant for
listing regimes 1 and 2 one year after the reverse split, but are not significant.
Finally, only for listing regime 1 are returns signicant eighteen months after the
reverse split. In this case, the shares experience a 24.05% decline vs. control
firms that do not do reverse splits.
4.5 Impact on Risk
We look at various risk measure before and after the reverse split for our
sample firms and in comparison to our matched sample. The results are
presented in Table 6. Total risk as measured by variance of returns increases
significantly (1% level) following the RS. However, when controlled for with the
matching sample, the change is negative and not significant. On the other hand,
systematic risk decreases significantly (at the 5% level) for the RS sample, but
increases control adjusted. This change in not significant. Our results differ from
Peterson and Peterson(1992), who find a significant decrease in variance after
reverse splits, but no change in systematic risk. It should be noted that Peterson
and Peterson did not test for changes using a matching sample. Finally,
unsystmeatic risk increases signficantly (at the 1% level) when there is no
matching, and decrease signficantly (at the 1% level) when the sample is match
adjusted.
[Table 6 About Here]
4.6 Probability of Survival
We run logistic regression to attempt to determine the probability that a
firm which does a reverse split will still be listed six months after the reverse split.
The results are shown in Table 7. We test three models. All have Chi-square
statistics that are significant at the 1% level. We find that listing regime 1 is
positive and highly significant, as is assets whiile financial leverage and the
motive to avoid delisting.is significantly negative. Looking at the size of the
coefficient, those firms with greater assets and less leverage following the splits
are more likely to then meet the relevant listing requirements.
Conclusions
In this paper, we examine the motivations for and signaling implication of
reverse splits. Reverse splits are so named because unlike forward splits, which
increase the number of shares outstanding and decrease the price by some
amount, they reduce the number of shares outstanding and increase the share
price by some factor. Essentially, reverse splits are of interest because they
provide a decidedly negative signal to the market – namely, that they are a
confirmation of poor performance, particularly when they are enacted for the
purpose of maintaining a listing on an exchange because the firm doing the
reverse split has not been able to meet the listing requirements or in order to
obtain a price at which institutional investors can buy the stock. Our results
indicate that the market does not view reverse splits favorably and that reverse
splits appear only to delay the inevitable delisting of many sample firms, in the
absence of other corrective governance events.
When is a Reverse Split Good News?
Wealth Effects and Performance Implications of Reverse Splits
ABSTRACT
In this paper, we examine the motivations for and signaling implication of reverse
splits. Reverse splits are so named because unlike forward splits, which increase
the number of shares outstanding and decrease the price by some amount, they
reduce the number of shares outstanding and increase the share price by some
factor. Reverse splits are of interest because they provide a decidedly negative
signal to the market. They are a confirmation of poor performance particularly
when they are enacted for the purpose of maintaining a listing on an exchange
or in order to obtain a price at which institutional investors can buy the stock. Our
results indicate that the market views reverse splits unfavorably and that reverse
splits appear only to delay the inevitable delisting of many sample firms.
When is a Reverse Split Good News?
Wealth Effects and Performance Implications of Reverse Splits
1. Introduction
In the past couple of years, reverse splits have become very popular, with
well even such a well known firm as AT&T completing 1 for 5 split in November
2002. Many of the firms that went public in the 1990s during the tech/internet
boom have also engaged in reverse splits as their share prices plummeted
dramatically. Over the years, much has been written in finance literature about
forward splits (see Rozeff (2003) for a comprehensive long-term study of forward
splits). Much less has been written about reverse splits. In the case of forward
splits, the number of shares is increased and the share price is proportionately
decreased. With a reverse split, the number of shares is decreased and the
price per share is proportionately increased. While a split by itself should have
no effect on the value of the stock, it may convey signals about the future
prospects of the firm.
In this paper, we examine the motivations for and the signaling/valuation
implications of reverse splits. Thus, if a forward split is seen as providing positive
information about the future of a firm, a reverse split would imply just the
opposite. In fact, as we shall show reverse splits provide a decidedly negative
signal to the market in that they are a confirmation of poor performance. This is
particularly the case when they are enacted for the purpose of maintaining a
listing on an exchange because the firm doing the reverse split has not been able
to meet the listing requirements. Reverse splits may similarly reflect poor
performance when they are done or in order to obtain a price at which
institutional investors can buy the stock.
In this study, we use a sample of 1702 reverse split from 1992 through
2001, which is the largest sample of reverse splits ever studied. We find that as
shown in prior studies, the announcement of a reverse split results in significant
negative abnormal returns to shareholders, on the announcement dates,
approval dates, and the split ex-date. Cumulative abnormal returns upon
announcement are significantly negative. We find that firms enacting reverse
splits suffer from poor operational performance prior to the reverse split.. We also
find that announcement date abnormal returns are significantly positively related
to profitability, leverage, and the size of the split. Finally, we find that the long run
viability and accounting performance of reverse split firms is poor.
The rest of this paper is organized as follows. Section 2 reviews the
existing literature on forward and reverse splits. Section 3 describes our sample
and research approach. Section 4 provides sample descriptive data. Section 5
provides results, and section 6 concludes the paper.
2. Theoretical framework
2.1. Signaling implications of managerial actions
The literature on signaling, asserts that management’s actions, in addition
to having an objective in and of themselves, are used to transmit information to
shareholders regarding the future cash flow implications of the firm and/or the
firm’s systematic risk. The best signals of future market value improvement are
those which are unambiguous and can be relatively cheaply and effectively
transmitted to the market, but which cannot be mimicked by a poor performer.
Two such “credible signals” are share repurchases and dividend increases; firms
with managers who have superior information regarding poor prospects will be
loathe to buy overvalued stock or effectively commit to a higher dividend knowing
it cannot be permanently maintained. Accordingly, the literature indicates that the
positive abnormal returns observed upon announcement of a share repurchase
or an increase in dividends are consistent with the market’s perception of a
positive signal regarding future cash flows (see, eg., Dittmar, 2002).
Empirical evidence showing positive event study returns suggest that the
forward splits provide some positive signal to shareholders (Desai and Jain,
1995, Dittmar, 2002, Brennan and Copeland, 1988). Forward splits appear to
move the firm into a more marketable trading range (Baker and Gallagher, 1980).
Positive wealth gains are observed on the announcement date and split ex date
but not on the approval date; apparently, for the signal to be effective, it does not
need to be repeated (Kryzabowski and Zhang, 1991, 1993).
Despite the abundance of research on forward splits, very little has been
done regarding the motivation for and wealth gains or losses from reverse splits.
Descriptively, reverse splits are “the substitution of one new share for a certain
number of outstanding shares” (Han, 1995); they are nothing more than a
mechanism for reducing the number of shares outstanding and increasing the
price by some stated factor. Spudeck and Moyer (1985) argue that reverse splits
are a negative signal because they imply that the only mechanism by which the
firm can improve the stock price is through largely artificial means. Empirical
evidence on reverse splits implies the market views them as a negative signal
about future performance (Brennan and Copeland, 1988; Lamoureau and Poon,
1987).
Han (1995) raises the questions of why managers engage in reverse splits
and why, if they are wealth destructive, shareholders approve such actions,
especially when, as Peterson and Peterson (1992), indicate they are not required
by the exchanges on which the firms are traded. Several reasons have been
advanced: the requirement for marginability, the signal presented by a low stock
price and the demonstrable lack of interest in such securities on behalf of
institutions (Falkenstein, 1998), and irrationality. Under Federal Reserve Board
initial margin requirements rules, stocks selling for less than a dollar are not
marginable. Most institutional investors will not by stocks that are selling for less
than $5, and some avoid stocks selling for less than $10.
Woodridge and Chambers (1983) examine up to 54 reverse splits for
NYSE/AMEX stokcs occuring between 1962 and 1981. Mean-adjusted excess
returns are significantly negative following the announcement, approval and
effective dates. Lamoureux and Poon (1987) look at forward and reverse splits
on NYSE/AMEX. Using a sample of 49 reverse splits in the period July 1962
through December 1985, they find that volatiltiy decreases, liquidity increases
following this corporate action. Peterson and Peterson (1992) examine 196
reverse splits over the period July 1962 and December 1989 for all listed stocks.
They find that discretionary reverse splits have negative excess returns, while
that non-discretionary splits result in positive excess returns. Furthermore, they
find that .these firms’ total risk declines following the RS, while systematic risk
does not change. They also argue that marketability is enhanced resulting from
a higher price range. Han (1995) looks at the effect of reverse splits on the
liquidity of 136 NYSE/AMEX/NASDAQ firms from 1963 through 1990. He finds
there is a decrease in spreads, an increase in trading volume and a signifnicant
decline in the number of non-trading days following reverse splits, all indications
of increased liquidity. Using event study methodology, he also finds a negative
reaction to reverse splits. Finally, Desai and Jain (1997) long at the long term
effect of 5,596 forward splits and 76 reverse splits over the period 1976-1991.
The stocks are listed on all exchanges. For the latter, they find very large and
significant negative abnormal returns 1, 2 and 3 years following the RS.
Reverse splits are often associated with corporate failure. An example is
the Bank of San Francisco – described as a “black hole” for its fiscal profligacy
and poor performance – which used a reverse split combined with a name
change to “reinvent itself” (O’Hara, 1994). Companies which do reverse splits
sometimes fail to meet the listing requirements on float and later have to issue
more shares1. However, not all the motives for reverse splits appear at first
glance to be wealth destructive and capable of generating the kind of negative
price reaction observed in the literature. Reverse splits may be a signal to the
market that the firm is a candidate for a takeover, due to the fact that fewer
shareholders would have to be convinced to tender their shares2 Reverse splits
are also used: to eliminate large numbers of small odd lot holders, to deregister a
1 For instance, Greenbriar Corp announced after its reverse split, “The combination of a lower
number of shares outstanding and the unexpected demand have resulted in occasional shortages
in the number of shares available for sale. The Company has undertaken with the American
Stock Exchange that it will adjust its number of shares outstanding by means of a stock dividend
during the first quarter to provide at least 200,000 shares in the public float, which is the minimum
standard of the Exchange.”
2 An example of the use of a reverse split to signal takeover candidacy is San Francisco’s Pacific
Bank, where management informed shareholders that the split could likely result in a change of
company and to take it private. The latter is accomplished by making the reverse
split so severe that only a few shareholders are left with more than a single
share.
This is the first study to examine reverse splits with a sample following
significant changes in NASDAQ listing requirements starting August 22, 1997.
For firms which cannot improve earnings quickly enough to keep the listing, the
reverse split may be the only way to get the stock price up to the minimum
required level. In cases where the motivation is to maintain listed status, the
reverse split announcement reminds the public that the firm is in danger of
delisting and has no alternative mechanism to keep its listing. In addition to
looking at the price reaction following a reverse split, this study examines reverse
splits motivated by limits to qualify as a marginable security, and price restrictions
on institutional holdings to see if the stated objective of the reverse split is ever
achieved. This is an area that has not been studied before. Our study is the first
to examine the relationship between motive and performance outcome for
reverse splits, and sheds new light on the signal they provide to the market.
3. Research Design
3.1. Sample Selection
We obtain announcements of reverse splits from the Center for Research
in Security Prices (CRSP) file, by identifying all distributions for the period 1992
to 2001 which are coded as reverse splits. In order to obtain information
regarding the motive for the reverse split, and to identify any confounding
control of the bank (Calbreath,1993).
announcements occurring the same day as the announcement of the reverse
split, we obtain the corresponding news announcements from the Lexis-Nexis
database. We obtain accounting data for our sample firms from Standard and
Poor’s Research Insights for firms engaging in reverse takeovers. Stock return
data are also obtained from CRSP. Corporate events prior to the reverse split as
well as post-reverse split performance and delisting information are obtained
from searches in Lexis/Nexis, annual reports, proxy statements, First Call,
Market Guide, and Research Insights. This extensive search process resulted in
1,072 announcements of reverse splits. Our sample is the largest ever used in a
study of reverse splits. The largest previous sample used contained 483 reverse
splits and covered a time period previous to the one we are examining. In
addition, we are able to examine reverse splits motivated by changes in
NASDAQ listing requirements.
We also conduct empirical tests of our sample firms vs. a control sample
of firms that did not do reverse stock splits. The control sample is obtained by
selecting the closest match in terms of size (assets) and book/market within the
three digit SIC code of the sample firm.
Table 1, Panels A , provide sample descriptive characteristics and provide
evidence regarding the frequency of reverse splits over time. The results
indicate that reverse splits decreased dramatically in 1993 and increased
dramatically in 1998. That increase continued in subsequent years so that
41.7% of all the splits in our sample occurred in the four years 1998 through
2001. Panel B shows that most firms underwent only one reverse split.
However, 10 firms had three reverse splits, while 2 had 4. It is clear that for
some firms, reverse splits cannot save a stock price from continuing to fall when
the underlying business prospects are very poor.
In Panel C, we provide information on the number and percent of reverse
splits by industry. Two industries, SIC 3000 and SIC 7000 accounted for more
than 40% of all reverse splits. It is not clear why these two industries should
account for such a large proportion of reverse splits. Panel D show the number
and frequency of splits by exchange listing. By far the largest number and
proportion are stock that traded on NASDAQ, 871 and 81.125%, respectively.
This is not surprising given the fact that newer and smaller firms tend to list on
NASDAQ, as the listing requirements are not as stringent as those for the NYSE
or AMEX. Interestingly, in August 1997, NASDAQ increased the requirements
to remain listed. The net tangible asset requirement was doubled under one
method and the market value of the float was increased by $1 million to $11
million again depending on the method used to determine eligibility to remain
listed. It would appear that these change had a lot to do with the increase in
reverse splits starting in 1998.3
[Table 1 About Here]
3.2. Research methods
3.2.1 Event study
To test the market’s reaction to the announcements of reverse stock splits,
we use event study methodology with the significance tests based on the
3 We would like to thank NASDAQ for providing the listing requirements for the time period of this
study.
standardized cross-sectional method (Boehmer, Musumeci and Poulsen, 1991)
with Scholes-Williams betas. Returns are modeled using ordinary least squares.
A 100-day estimation period was chosen from t-110 to t-11 where t=0 is the
event date.4 Cumulative abnormal returns (CARs) are reported for the (-1,+1)
and (-1,0) event windows. We attempt to explain the source of value gains or
losses with cross-sectional regressions on CARs.
3.2.2 Buy and hold returns analysis
In order to examine the long run returns of firms engaging in reverse
splits, we utilize the Barber-Lyon (1997) methodology. We hypothesize that firms
employing reverse splits for discretionary reasons may experience improvements
in long horizon returns, given that they survive and are not delisted. We estimate
matching sample-adjusted long-horizon average holding-period abnormal returns
(AHAR) for our sample bidders. AHARs are calculated starting with the month
after the announcement date for 6, 12, and 18 month holding periods.
3.2.3 Changes in Risk
Like Petersen and Petersen (1992), we hypothesize that the reverse split
may have consequences for the risk profile of the firm. For each firm, we
estimate the changes in its risk characteristics surrounding the announcement of
the reverse split. The pre-announcement risk measure, VARPRE [postannouncement
risk measure, VARPOST ] is estimated over the period from t-110
4 We varied the parameter estimation period from 100 to 150 days. Results similar to the ones
reported were obtained. In some cases our total sample includes more than one announcing firm
for the same date. Accordingly, our significance tests adjust for cross-sectional dependence in
the regression residuals. We also adjust for serial dependence as well as for changes in event
induced variance (see the manual for Eventus, version 7.0). CARS are based on an equally
to t-11 [t+11 to t+110], where t=0 is the announcement date for both sample firms
and matched firms. We estimate (a) total risk, or Var (Ri ), where Ri is the security
return on the ith security, and (b) systematic risk, or βi.
4. Results
4.1. Characteristics of firms which engage in reverse splits
Based on the existing literature, we would anticipate finding that firms
engaging in reverse splits are poor performers which have adopted this particular
strategy as a result of actions taken over time. These have translated into the
firm being subject to being delisted. Descriptive statistics for the year prior to the
reverse split, provided in Table 2, indicate that this is in fact the case. Firms
engaging in reverse splits are, on average, very small with a mean (median)
sales of $175 ($14) million, asset size of $239 ($19) million, and market value of
equity of $119 (25) million. In addition, these firms are very unprofitable with a
mean (median) net profit margin of -30.09% (-8.91%), return on assets and
equity of –27.91 (-11.46%) and –32.29% (-12.27%), respectively. However, their
profitability ratios were not substantially different from their control counterparts,
indicating that perhaps the threat of delisting arose from adverse price
movements within the sector. Reverse split firms have negative cash flow as
well; they exhibit significantly lower free cash flow to total assets than their
control counterparts.
[Table 2 About Here]
weighted market portfolio. Similar results were obtained with the value weighted market portfolio.
Table 2 also indicates that being overlevered was not the likely reason that the
sample firms performed poorly; their debt/asset ratios do not differ significantly
from control firms. Nor was the liquidity of sample firms, measured by cash/total
assets, significantly different from control firms. From an accounting perspective,
sample firms were not nearer bankruptcy than control firms; their Z scores were
insignificantly different. Again, this supports the conjecture that reverse split firms
may have experienced share price declines as a result of adverse industry
conditions.
4.2. Event study results
Event study results shown in Table 3 indicate that the market does not respond
favorably to announcements of reverse splits; mean event date abnormal return
is -5.61 percent, which is significant at the 1% level. The results support the idea
the reverse split is in general a desperate measure taken by poorly performing
managers, and that it is not viewed positively by the market. Panel B breaks
down the sample by motives for doing the reverse split. The most significant
factors are threat of delisting, news release with no reason given, no news
release prior to the reverse split announcement, capital structure change, and
other reasons. The abnormal returns are significantly negative when the stated
reason for the split is delisting (a nondiscretionary reason), when the
management does not give a reason for the split, and when there is no news
release regarding the split at all.
[Table 3 About Here]
The results for reverse splits conducted for discretionary reasons – to attract
institutional investors by moving the price over the $5 limit on institutional
purchases, or as part of a reorganization, are very different. We observe that
when the motive is to attract institutional investors, the abnormal returns are
significantly positive and generate returns of 2.64% over the (-1,+1) event
window. The results for the reorganization motive are mixed; on the
announcement date itself, the abnormal returns are negative, but over the (-1,0)
window, they are significantly positive. It appears that corrective, rather than
survival, motives for firms doing reverse splits, generate wealth for shareholders,
and the differences in abnormal returns are significantly higher for corrective than
survival motives.
4.3. Multivariate Regression Analysis
We next examine the factors that affect reverse split announcement
returns. The results of these regressions are shown in Table 4. Two models are
shown for the purposes of demonstrating the robustness of results. All three
models are significant at the 1% level, as shown by their F-tests.
[Table 4 About Here]
In Model 1, we regress three day CARS on a liquidity, as measured by cash/total
assets, leverage, as measured by debt/total assets, the magnitude of the split,
profitability, as proxied by return on assets, and liquidity, as proxied by trading
volume. The coefficient of cash/assets is negative but insignificant, indicating that
the market does not reward firms with higher liquidity or penalize those with lower
liquidity. The coefficient of debt to total assets is positive and significant,
indicating the market perceives that firms with higher leverage benefit more from
the reverse split than those with lower leverage. Perhaps shareholders of firms
with high leverage feel that management is signaling a sale of additional shares,
which are easier to sell now with listing maintained and having a price high
enough to attract institutional investors. Return on assets is positive but
insignificant. Maybe shareholders see the firm as temporarily in distress, and if
it can just get the share price up and attract institutional investors, it can remain
listed. Trading volume is negative and insignificant, a surprising result since
liquidity and low volume were two commonly cited reasons managers give for
reverse splits.
The size of the split factor is also positive and significant. In CRSP,
reverse splits have a negative FACSHR value, so the larger the reverse split, the
more negative the FACSHR value will be. The positive significant coefficient on
the size of split variable indicates that more dramatic reverse split events appear
to be signs of increasing desperation, ie, a firm which has to do a 1:100 reverse
split is in worse shape than one which merely needs to double its share price to
meet the exchange listing requirements, and hence, is more likely to pull through
distress.
The results of models 2 and 3 are quantitatively similar. Size, as proxied
by market value, is insignificantly different from zero. Liquidity is significantly and
positively related to CARs; perhaps the market perceives that firms with higher
liquidity are more likely to pull through a momentary period of distress if they can
remain listed. However, trading volume is insignificant, indicating that abnormal
returns are not driven by concerns over the liquidity of the shares being traded.
Overall, the results suggest that while reverse splits are seen as evidence of
wealth destruction on average, the shareholders of firms doing reverse splits
view the performance outcomes more favorably for more profitable firms who use
greater relative amounts of debt capital.
4.4. Post-reverse split performance
The stated objectives of managers when they announce a reverse stock
split include more visibility to institutions, maintaining a listing, becoming a
marginable security, and signalling takeover target status. To answer the
questions of whether reverse split firms are successful in achieving these stated
intentions, and whether or not management is able to parlay the reverse split into
long term wealth maximization, we examine the long run survival of reverse split
firms as well as whether the firm achieved what the reverse split was supposed
to achieve. The results of this analysis are presented in Table 5
[Table 5 about here]
Managers of firms doing reverse splits often express tremendous
optimism regarding the firms’ future profitability, as one would expect in order to
prevent a large sell-off of shares. To some degree, this happens as shown in
Panel A. Return on assets and net profit margin increase significantly (1%
lelvel). However, the results in Panel B of Table 5 suggest that firms doing
reverse splits, are about as likely to be delisted within six months following the
completion of the reverse split as they are to survive. Only 37.3% are still listed
two years following the completion of the reverse split.
For those firms trading on NASDAQ, we also wanted to look at the six,
twelve and eighteen month equtiy returns following reverse splits, taking into
account two different listing regimes used by NASDAQ. As shown in Panel C,
returns are significantly negative (5% level) six months after the reverse split for
the total sample and listing regime 2. Returns are negative but not significant for
listing regimes 1 and 2 one year after the reverse split, but are not significant.
Finally, only for listing regime 1 are returns signicant eighteen months after the
reverse split. In this case, the shares experience a 24.05% decline vs. control
firms that do not do reverse splits.
4.5 Impact on Risk
We look at various risk measure before and after the reverse split for our
sample firms and in comparison to our matched sample. The results are
presented in Table 6. Total risk as measured by variance of returns increases
significantly (1% level) following the RS. However, when controlled for with the
matching sample, the change is negative and not significant. On the other hand,
systematic risk decreases significantly (at the 5% level) for the RS sample, but
increases control adjusted. This change in not significant. Our results differ from
Peterson and Peterson(1992), who find a significant decrease in variance after
reverse splits, but no change in systematic risk. It should be noted that Peterson
and Peterson did not test for changes using a matching sample. Finally,
unsystmeatic risk increases signficantly (at the 1% level) when there is no
matching, and decrease signficantly (at the 1% level) when the sample is match
adjusted.
[Table 6 About Here]
4.6 Probability of Survival
We run logistic regression to attempt to determine the probability that a
firm which does a reverse split will still be listed six months after the reverse split.
The results are shown in Table 7. We test three models. All have Chi-square
statistics that are significant at the 1% level. We find that listing regime 1 is
positive and highly significant, as is assets whiile financial leverage and the
motive to avoid delisting.is significantly negative. Looking at the size of the
coefficient, those firms with greater assets and less leverage following the splits
are more likely to then meet the relevant listing requirements.
Conclusions
In this paper, we examine the motivations for and signaling implication of
reverse splits. Reverse splits are so named because unlike forward splits, which
increase the number of shares outstanding and decrease the price by some
amount, they reduce the number of shares outstanding and increase the share
price by some factor. Essentially, reverse splits are of interest because they
provide a decidedly negative signal to the market – namely, that they are a
confirmation of poor performance, particularly when they are enacted for the
purpose of maintaining a listing on an exchange because the firm doing the
reverse split has not been able to meet the listing requirements or in order to
obtain a price at which institutional investors can buy the stock. Our results
indicate that the market does not view reverse splits favorably and that reverse
splits appear only to delay the inevitable delisting of many sample firms, in the
absence of other corrective governance events.
4 Recommendations
Loading...
Loading...
New Message
Please
login
to post a reply