J Econ Finan (2015) 39:357–369

DOI 10.1007/s12197-013-9259-x

Multiple reverse stock splits (investors beware!)

Claire E. Crutchley & Steven Swidler

Published online: 23 May 2013

# Springer Science+Business Media New York 2013

Abstract This study compares firms that implement multiple reverse stock splits to
firms with only one reverse stock split. Reverse stock splits are usually implemented
by firms trying to increase their stock price to remain listed on stock exchanges or
widen stock ownership especially by institutional investors. Firms that declare mul-
tiple reverse splits tend to have lower returns following the reverse split and even less
liquidity than one reverse split firms. Sixty five percent of the firms with multiple
reverse splits end up being liquidated or delisted. If one reverse split is viewed as
desperation, then multiple reverse stock splits are a sign of extreme distress.

Keywords Reverse stock splits . Liquidity . Shares . Information

JEL Classification G14 . G34

1 Introduction

Reverse stock splits are frequently regarded as desperate attempts by the firm to raise its
share price enough to meet the minimum listing requirements of an exchange (Martell
and Webb 2008). Hwang (1995) finds news of a reverse split has an adverse effect, while
several others show under-performance after the reverse split (Woolridge and Chambers
1983; Lamoureux and Poon 1987; Peterson and Peterson 1992; Desai and Jain 1997) .1
If investors view one reverse split as a desperate act on the part of managers, then two or
more reverse splits by the same firm may be seen as a dire situation. The following
analysis chronicles the frequency of firms that have two or more reverse splits and
examines whether the outcome is better the second (or third or more) time around.
NASDAQ and NYSE require a common stock price above $1 in order for a
company’s stock to maintain its listing status. For some firms, the only short-run
1
While reverse splits lead to negative performance, on average, there may be ex-ante reasons for managers
to believe a reverse split will increase firm value. For example, if exchange listing maintains visibility,
announcement of a reverse split may increase value.
C. E. Crutchley : S. Swidler (*)
Department of Finance, Auburn University, 303 Lowder Business Building, Auburn, AL 36849, USA
e-mail: [email protected]
C. E. Crutchley
e-mail: [email protected]
358 J Econ Finan (2015) 39:357–369

solution to meeting the constraint is for the company to declare a reverse stock split.
Moreover, many institutional investors require that share prices be above $5, or in some
cases $10, before they can purchase them for their portfolio. Even if the stock maintains
the listing requirement, investors typically treat a reverse split as bad news. Radcliffe and
Gillespie (1979) show negative returns on the ex-date of the reverse split. In the long-
run, Martell and Webb (2008) find NASDAQ firms that declare reverse splits in down
markets tend to fare better than companies that reverse split in up markets. However,
they show in both cases the long run prospects are poor, and most firms that declare
reverse splits do not remain independent, publicly traded firms.
The demise of firms after a reverse stock split is not surprising in light of Kim et al.
(2008) finding that stock returns and earnings underperform for the four years following
the reverse split. While there are significant negative abnormal returns at the time of the
split, the market does not immediately impound the bad news. Kim et al. (2008) conclude
that stock prices do not quickly adjust as investors may not be able to short the stock.
Moreover, the stocks are often illiquid and transaction costs high so that market efficiency
is not violated. This is especially true of the stocks priced below $5 after the reverse split.
If the main motivation for reverse splits is to maintain exchange listing, it follows
that corporate boards are concerned about the firm’s public profile, trading of the
stock, liquidity and ultimately the survival of the firm.2 Han (1995) provides evidence
of increased liquidity and fewer zero volume days following a reverse split. Chung
and Yang (2010) find firms that boost share price above $5 increase institutional
ownership following the reverse split. While reverse splits may have the desired
increase in trading activity for some firms, for those companies needing another
reverse split, low volume clearly remains an issue.
To date, there has been little discussion of firms that declare more than one reverse
split over time. The following analysis documents how frequently multiple reverse
splits occur, whether this sample is similar to the set of firms with only one reverse
split and the long run performance of firms with multiple reverse splits. Finally, we
discuss implications for investors who might be interested in this group of firms.

2 Sample

The time period of analysis covers 1990 through 2010. Our sample consists of
all companies listed in the Center for Research in Security Prices (CRSP) with
a distribution code of 5523 indicating a stock split and a negative Factor to
Adjust Shares indicating a reverse split. This sample includes all reverse splits
for public companies trading in the United States regardless of stock exchange.
We classify companies as having “Only one” reverse stock split if there is one
reverse split between 1980 and 2010 and those with “Multiple” reverse stock
splits if the company implements more than one reverse stock split. Note that
we examine only the stock splits between 1990 and 2010 but we investigate

2
Martell and Webb (2008) suggest three additional reasons: 1) increasing institutional ownership, 2)
reducing the number of shareholders before taking the firm private and 3) reducing the number of small
shareholders to save servicing expenses. The first reason gets back to liquidity, while reasons 2 and 3 seem
to be of secondary importance, especially if there has been more than one reverse split.
J Econ Finan (2015) 39:357–369 359

Table 1 Frequency of reverse splits per company

Number of reverse splits Number of companies Percentage

One 1551 85.8 %

Two 216 12.0 %
Three 28 1.5 %
Four 7 0.4 %
Five 5 0.3 %
Companies 1807 100 %

This is the universe of reverse stock splits between 1990 and 2010 as reported on the CRSP stock file. One
reverse split indicates a company reverse split its stock once between 1980 and 2010. While only the
reverse splits between 1990 and 2010 are included in the sample, if a company implemented a reverse split
between 1980 and 1990, this will be counted in the number of reverse splits. For example, if a company
implemented two reverse splits between 1980 and 1990 and one reverse split between 1990 and 2010, this
company would be classified as having three reverse stock splits

reverse splits from 1980 to 1990 to correctly classify whether the firms in our
sample had an earlier reverse split.
Table 1 lists the number of firms that experienced reverse stock splits from 1990 to
2010. In total, there were 1,807 companies that declared at least one reverse split over
the 21 year period. Of those firms, 1,551 (85.8 %) only had one reverse split, while
256 (14.2 %) had two or more reverse splits. The latter category includes 5 firms with
as many as five reverse splits in our sample period.3
Reverse splits occur throughout the sample period and a distribution across
years appears in Table 2.4 The pattern for firms with only one reverse split is
similar to the multiple reverse split distribution. This information is displayed in
Fig. 1. Of note is that for both groups, the highest frequency happens in 1998
and may be the result of a significant decline in the market. From July to
October 1998, the NASDAQ fell by nearly 30 % in a reversal of what had
been a very strong bull market. Ironically, by December, the market had
climbed back to its earlier high and continued to increase in 1999 so that
reverse splits may not have been needed to increase share price.
In declaring a reverse split, the board’s goal is to increase the stock price.
Depending on the specific stock price desired, the number of old shares to new
must be a large enough factor to increase the stock price at least above the
listing threshold of $1 and perhaps the $5 limit for institutional investors.
Table 3 tabulates reverse split factors for only one incidence and for the first-
fifth multiple reverse split. The distributions are very similar. Specifically, in
both sub-samples approximately half of the firms have factors between five and

3
Of the five companies that initiated a reverse split five times, the most extreme case is Members Service
Corp who reverse split the stock five times within four years from 1990 to 1994. Members Service Corp’s
last two splits were 50 to 1 and 40 to 1. Shortly after the fifth reverse split the company was delisted, and
within a few years, the company was investigated by the SEC for fraud.
4
The first line of the table 1980–1989 shows there were 47 reverse splits for companies who also had at
least one reverse split between 1990 and 2010; all of these companies are classified in the multiple reverse
split category.
360 J Econ Finan (2015) 39:357–369

Table 2 Reverse stock splits by year and number of splits

Year All reverse splits Only one reverse split Multiple reverse splits

Number Percent Number Percent Number Percent

1980–1989 47 – 0 – 47 –
1990 106 5% 71 5% 35 7%
1991 98 5% 68 4% 30 6%
1992 156 8% 107 7% 49 9%
1993 115 6% 83 5% 32 6%
1994 92 4% 56 4% 36 7%
1995 101 5% 71 5% 30 6%
1996 98 5% 68 4% 30 6%
1997 105 5% 72 5% 33 6%
1998 184 9% 136 9% 48 9%
1999 114 5% 85 5% 29 6%
2000 74 4% 61 4% 13 2%
2001 125 6% 93 6% 32 6%
2002 124 6% 105 7% 19 4%
2003 90 4% 74 5% 16 3%
2004 43 2% 37 2% 6 1%
2005 60 3% 39 3% 21 4%
2006 66 3% 55 4% 11 2%
2007 55 3% 40 3% 15 3%
2008 75 4% 59 4% 16 3%
2009 68 3% 59 4% 9 2%
2010 124 6% 112 7% 12 2%
Total 1990–2010 2073 100 % 1551 100 % 522 100 %

We classify a company as having multiple reverse stock splits if it implemented more than one reverse stock
split between 1980 and 2010 although our sample begins in 1990. This table shows the number of reverse
splits each year between 1990 and 2010. The reverse splits shown from 1980 to 1989 are reverse splits for
sample firms who also had at least one reverse split between 1990 and 2010. Firms classified in Only one
reverse stock split implemented one reverse stock split between 1980 and 2010. Firms classified as Multiple
Reverse Splits implemented two or more reverse splits between 1980 and 2010. Total 1990–2010 includes
only reverse splits between 1990 and 2010

ten shares for one. The most popular reverse split factor is 10 (19 % for one
reverse split, 20 % for multiple splits).
While one reverse stock split may indicate a temporary problem in a low stock price,
multiple reverse splits indicate a recurring problem. As reverse stock splits tend to be
implemented by small companies, Table 4 reports the size and the stock prices of
companies with one and multiple reverse stock splits. The first two lines are for only
one reverse split companies; these companies are small with an average (median) market
value of $327 ($15) million and an average pre- and post-split stock prices equal to
$2.15 and $7.16 with a mean split factor of 9.2 shares to 1. Over half of the companies
have a stock price below $0.63 increasing to a post-split median price of $3.75.
J Econ Finan (2015) 39:357–369 361

200
180
Number of Reverse Splits 160
140
120
100
80
60
40
20
0
1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010
Year of Reverse Split

All Only One Multiple

Fig. 1 Number of Reverse Splits by Year based on number of reverse splits per Company. This figure
shows the yearly distribution of reverse splits in the 20 year period 1990–2010. The top line is all reverse
splits; the middle line is reverse splits for companies with only one reverse split over the time period and the
bottom line is reverse splits for companies with more than one reverse split over the time period

The second part of the table shows the market value, stock price, split factor and
years between splits for companies by multiple reverse splits. The median market
value of the companies is small and gets smaller with each split.5 Reverse stock splits
appear to be motivated by the $1 stock listing requirement as the median stock price
before the reverse split is $.30 to $.50 for all reverse splits. Even though successive
reverse splits generally exhibit declining post-split prices, the median post-split price
is always above the $1 threshold. The mean and median split factors are very similar
for all reverse splits, whether multiple or not. Finally, the median years between
reverse stock splits falls from approximately 3 years to 2 years although the numbers
of observations for the fourth and fifth split are very low.

3 Hypotheses and methodology

An underlying question in this paper is why a firm would try a second (or multiple)
reverse split(s). After all, the first reverse split apparently did not work, and reverse
splits are costly. In most cases, the firm again needs to increase share price to meet
listing or institutional ownership requirements. As mentioned before, the main goal of
listing is to keep a public profile among investors and encourage trading of the stock.
Quite simply, firms are concerned about liquidity to maintain demand for their stock
and ultimately ensure their survival.
It is possible that investors view a reverse stock split as either positive or negative
news. If investors see a reverse stock split as a signal that the only way management can
raise price is through a reverse split, a negative stock reaction would occur. If, on the

5
There are two outliers which skew the mean market value of those with a third reverse split; one has a
market value above $2 Billion and one has a market value of approximately $700 million. The next largest
company has a market value below $100 million.
362 J Econ Finan (2015) 39:357–369

Table 3 Reverse stock split factor

Only one reverse Multiple

Reverse split factor Number Percentage Number Percentage

Less than 4 342 22 % 116 22 %

4<=Size<=5 227 15 % 66 13 %
5<Size<10 468 30 % 154 30 %
Size=10 299 19 % 105 20 %
10<Size<= 20 131 8% 47 9%
20<Size<50 61 4% 30 6%
Size>=50 23 1% 4 1%
Total 1551 100 % 522 100 %

The reverse split factor is the number of old shares exchanged for one new share. All reverse splits are
included. Those splits in “only one reverse” are issued by companies who only implemented one reverse
stock split between 1980 and 2010. The Multiple reverse stock split is all reverse stock splits issued
between 1990 and 2010 by companies with more than one reverse stock split between 1980 and 2010

other hand, the reverse split signals management is enhancing stock value by keeping
the stock listed, this may be perceived as positive news by the market. Thus, the first set
of hypotheses test whether investors view a reverse stock split as either a good or bad
event. In testing the news content of reverse splits, the analysis considers abnormal
returns before, on and after the ex-split date and compares the performance of firms with
only one reverse split to those firms that experience multiple reverse splits.
Frequently firms do not announce a reverse split. For these companies the
first indication of the reverse split is on the ex-split date. Peterson and Peterson
(1992) use either the announcement date, if found, or the ex-split date. In our
sample, only 36 % have an announcement of the reverse split. For the few with
announcements, the median announcement date occurs five calendar days prior
to the split date, and 75 % are within 15 calendar days of the ex-split date.
Because of the lack of announcement dates, we follow the Martell and Webb
(2008) methodology and specify the ex-split day as our event date (day 0).
To calculate excess returns, we use the Fama French two step methodology on
WRDS Eventus and match stocks according to size, market value and market to
book. We collect data from day −255 to −45 from the daily CRSP files to estimate
returns in a pre-event (normal) period and then calculate abnormal returns from day
−30 to +30. We also calculate a simple holding period return as the percentage change
in price over the 61 days, unadjusted for the market.
If the main reason for a reverse stock split is to maintain liquidity by keeping the
stock listed or eligible for institutional investment, it is important to examine liquidity
around the time of the split. For many companies in our sample, there are days when
the stock does not trade at all. Thus, we measure liquidity in two different ways. The
first method examines non-trading days within a given period, and the second
liquidity measure considers volume in a given time frame.
The first approach calculates the number of non-trading days similar to Han
(1995). Using the CRSP daily files, we determine the number of days with zero
J Econ Finan (2015) 39:357–369 363

Table 4 Market value, stock price, split factor and years between reverse splits

Market Stock price Split Years between Number

value factor split

Reverse Split $1,000 s Before Reverse After Reverse

Only One
Mean 326,926 $2.15 $7.16 9.2 1551
Median 14,579 $0.63 $3.75 5
Multiple Reverse Splits
First
Mean 119,980 $1.35 $4.14 9.5 212
Median 8,965 $0.44 $2.69 5
Second
Mean 49,447 $1.52 $6.57 9.1 4.2 253
Median 6,933 $0.50 $2.95 6 3.2
Third
Mean 85,136 $2.56 $5.41 9.3 3.6 40
Median 5,797 $0.41 $2.34 5 2.8
Fourth
Mean 7,573 $0.44 $2.26 13.4 3.2 12
Median 1,635 $0.30 $1.87 6.0 1.9
Fifth
Mean 1,781 $0.83 $5.04 13.6 1.5 5
Median 1,210 $0.42 $1.67 8 1.8

This table reports information by reverse split for companies with reverse splits between 1990 and 2010. The
sample is split between those companies with only one reverse split between 1980 and 2010 and those with more
than one reverse split between 1980 and 2010. Number is the number of observations for each sub-sample. The
market value is the day prior to the reverse split in thousands of dollars and the stock price is the day before and the
day after the reverse split in dollars. Split factors are the mean and median number of shares exchanged for one
share at the reverse split. Years between splits is the number of years between the second and first (third and second,
fourth and third, fifth and fourth) for those companies with a second (third, fourth, fifth) reverse split. Note there are
522 multiple reverse stock splits between 1990 and 2010; the 47 reverse stock splits in 1980–1990 shown in line 1
Table 2 result in companies being classified as having multiple reverse stock splits but are not in the sample

volume in the 50 trading days prior to and following the reverse split. Non-trading
day comparisons are then made between firms with only one reverse split to those
firms that experience multiple reverse splits.
In method two, liquidity equals actual trading volume measured in dollars. We calculate
total volume over the twelve months prior to and following the reverse split. Once more,
the analysis examines differences between single and multiple reverse split firms.

4 Empirical results

To see whether exchange listing and institutional holding are of primary concern to the
vast majority of firms that declare reverse splits, Table 5 lists the frequency of reverse
stock splits by pre-split stock prices; prices are measured the last trading day prior to the
364 J Econ Finan (2015) 39:357–369

Table 5 Frequency of reverse stock splits by pre-split prices

Number of companies

Pre-split price First or only reverse split Second reverse split Third-fifth reverse split

Less than $1 1168 195 46

percentage 67 % 77 % 82 %
$1 to<$2 197 24 1
percentage 11 % 10 % 2%
$2 to<$5 234 19 6
percentage 14 % 8% 11 %
Over $5 139 14 3
percentage 8% 6% 5%
Total 1738 252 56

This table reports the frequency of pre-reverse split prices for companies with reverse splits between 1990 and
2010. First reverse split includes the reverse split of companies who only had one reverse split between 1990
and 2010 and the first reverse split of companies who had more than one reverse split between 1990 and 2010.
Second reverse split includes the second reverse split for those companies with more than reverse split between
1980 and 2010 and Third-Fifth includes the third fourth and fifth reverse splits between 1990 and 2010

reverse split. In the case of the first or only reverse split, 1,168 of 1,738 (67 %) pre-split
prices fall below the $1 listing threshold. Another 197 (11 %) are between $1 and $2 and
might thought to be in danger of violating price listing standards. In terms of institutional
holding, 1,599 (92 %) fail to clear the $5 benchmark.
For firms that have a second reverse split, exchange listing and institutional
holding appear to be even more problematic. There are 195 of 252 (77 %) firms with
pre-split prices below $1, twenty-four (10 %) between $1 and $2 and 238 (94 %)
below $5. For the third through fifth reverse split, even greater percentages miss the
exchange listing and institutional thresholds. There are 46 out of 56 firms with shares
falling short of $1 (82 %), one (2 %) between $1 and $2 and a total 53 (95 %) below
$5. Taken as a whole, Table 5 provides strong evidence that exchange listing and
institutional holding requirements provide compelling motivation for reverse splits
for a large majority of the firms in our sample.
While firms are motivated to retain exchange listing and boost institutional
holding, it remains an empirical issue whether investors view a reverse split as either
a positive or negative event. Table 6 reports returns around a reverse split and
considers separately the performance of firms with only a single split and those
companies with multiple splits. The latter sample is further divided into returns for
the first reverse split and returns for the second through fifth split.
In most cases, the mean and median return results are qualitatively similar. Daily
returns can be quite large (in absolute value) due to a low stock price which can lead to
extreme values for cumulative returns. Moreover, as we shall see below, a number of
stocks have at least one zero volume trading day in the event window. This results in
some daily returns being calculated over a longer time period. Thus, we focus on the
median return results to infer investors’ assessment of reverse stock splits.
In the run-up to the reverse split ex-date, the median risk-adjusted abnormal
returns are little different from 0 for all three sub-samples in Table 6, Panel A.
Table 6 Abnormal returns around reverse splits

Panel A Fama French adjusted returns

Single reverse First of multiple Second-fifth

Mean Median Mean Median

J Econ Finan (2015) 39:357–369

CAR −30 −2 3.53 %*** −0.77 % 7.94 %** 0.62 % 11.14 %*** 0.25 %
1520 207 308
Split return −1,0 −5.34 %*** −3.24 %*** −8.22 %*** −5.67 %*** −4.94 %*** −4.78 %***
1518 207 308
CAR +1, 30 −1.29 % −3.26 %*** −1.24 % −3.61 %* −4.68 %* −7.18 %**
1520 207 306
Car −30,+30 −3.10 %* −4.79 %*** −1.52 % −9.74 %* 1.54 % −5.58 %
Observations 1522 207 309
Panel B Unadjusted returns
HPR −30 +30 −11.19 %*** −13.57 %*** −10.99 %*** −20.00 %*** −12.96 %*** −18.83 %***
Percent negative 68.5 %*** 72.3 %*** 70.3 %***
Observations 1483 206 293

This table presents stock returns surrounding the ex-date of the reverse split, day 0. Day −1 is the trading day before the reverse split, −30 −2 is the 29 trading days prior to the
reverse split and +1-+30 is the 30 days following the reverse split. Abnormal returns are calculated using Fama French two-step model with a CRSP Value weighted market
portfolio. The estimation period is from days 255 to 46 trading days prior to the reverse split ex-date. Unadjusted returns are a simple holding period return: the percentage change
in price over the 61 days (adjusted for the reverse split) and the Percent negative is the percent of companies with positive holding period returns over the 61 day period. Single
Reverse Splits are the reverse splits for companies with only one reverse split, first of multiple is the first reverse split for companies with more than reverse split in the 1990–2010
period, and second-fifth includes the second, third, fourth and fifth reverse split (when relevant) for companies with more than one reverse split. Mean differences from zero and
differences in means across groups are calculated using a t-test. Median difference from zero and differences across groups is calculated using Wilcoxon signed rank tests
* ** ***
, , significantly different from zero (50 % for the percent negative) at the 10 %, 5 % and 1 % level
365
366 J Econ Finan (2015) 39:357–369

Table 7 Non-trading days surrounding reverse splits

Number of non trading days

Single reverse split Multiple reverse splits

First Second-fifth

Days surrounding reverse split Mean Median Mean Median Mean Median
Days −50 to −1 2.69*** 0 4.24***c 0 3.30***a 0
Days +1 to +50 3.48*** 0 4.35*** 1 3.91*** 0.51
*** b **
Difference in non-trading days 0.79 0 0.12 0 0.60 0
Observations 1551 212 310

Days −50 to −1 are the 50 calendar days prior to the reverse split and +1 to +50 are the 50 calendar days
following the reverse split
The number of non-trading days is the number of days the stock did not trade in that time period on a
trading day. Mean differences from zero and differences in means across groups is calculated using a t-test.
Median difference from zero and differences across groups is calculated using Wilcoxon signed rank tests
1
For the fifty days following second-fifth reverse splits, half the companies have zero non-trading days,
and half have at least one non-trading day resulting in a median of 0.5
* ** ***
, , significantly different from zero at the 10 %, 5 % and 1 % level
a b c
, , significantly different from single reverse split at the 10 %, 5 % and 1 % level

On days −1 and 0, median abnormal returns are −3.24 % for the single reverse
split firms, and even more negative for multiple split firms, −5.67 % for the
first split and −4.78 % for splits two through five. For days +1 to +30, single
reverse firms exhibit negative median returns equal to −3.26 %. Again, multiple
reverse split firms display relatively larger negative returns, −3.61 % for the
first split and −7.18 % for instances two through five. Taken over the entire
event window, −30 to+30 days, median abnormal returns for single reverse
split firms are −4.79 % compared to −9.74 % and −5.58 % for the first and
second through fifth reverse split sub-samples. However, pairwise differences
are not significantly different.
To further examine the news content of reverse splits, Table 6 reports
unadjusted returns over the sixty-one day event window. These returns demon-
strate a similar pattern as the Fama French risk adjusted results. Median
holding period returns for single reverse split firms are −13.57 % and are even
more negative for the first and second through fifth splits, −20 % and
−18.83 %, respectively. Furthermore, 68.5 % of single reverse split firms
exhibit negative returns for the event window compared to 72.3 % and
70.3 % for the multiple split cases. A simple binomial test rejects the null
hypothesis that fifty percent of 61 day returns are positive, and instead implies
that investors generally view reverse splits as a negative corporate event.
Overall, the results are consistent with previous findings of adverse effects of
reverse splits. Moreover, the results suggest that the effects may even be more
negative for multiple reverse split firms.
Turning next to issues of liquidity, Table 7 examines the number of non-
trading days before and after a reverse split. The mean number of non-trading
J Econ Finan (2015) 39:357–369 367

Table 8 Dollar volume twelve months prior to and after reverse splits

In $millions Mean only one Mean first of multiple Mean second through fifth

Volume before 658.5 278.6 124.0

split
Volume after 531.4 171.3 136.1
split
Increase in −140.6** −107.4 8.0
volume
Difference in Only one vs first One vs second through fifth First vs second through fifth
volume
Difference 380.0* 534.6*** 154.6
before split
Difference 360.1*** 395.3*** 35.2
after split

Dollar volume is the price times the number of shares each day. The volume is accumulated each calendar
month. Before split is the sum of the dollar volume each of the twelve calendar months prior to the reverse
split and After Split is the sum of the twelve calendar months following the split. The month of the reverse
split is not included
* ** ***
, , significantly different from zero at the 10 %, 5 % and 1 % level

days is between 2.5 and 4.5 in the fifty days prior to and following the reverse
split. If a split increases liquidity, we would expect the number of non-trading
days to decline. Instead, the mean number of zero volume days increases

70
Dollar Monthly Trading Volume (millions)

60

50

40

30

20

10

0
-12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12
Month Relative to Reverse Split

Only One First of Multiple Second through Fifth

Fig. 2 Monthly Trading Volume around the Reverse Split Ex-Date. The graph shows the mean of the
monthly dollar trading volume in the 25 months surrounding the reverse stock split by reverse split
category. The top line is dollar volume around the first and only reverse split, the middle line is the dollar
volume for the first of multiple splits and the bottom line (represented by triangles) is the dollar volume for
the second through fifth splits
368 J Econ Finan (2015) 39:357–369

subsequent to the split and is statistically significant for single reverse split
firms and for companies that have two or more splits. This evidence is contrary
to Han (1995) who finds fewer non-trading days following reverse splits prior
to 1990. There is also evidence that for days −50 to −1, multiple reverse split
firms exhibit a larger number of non-trading days compared to single reverse
split firms.
Consistent with these liquidity findings, Table 8 reveals that monthly vol-
ume, in dollars, also declines after the split for reverse split number one (both
only and first). There is no significant difference in volume for the second
through fifth reverse split, although there is a small dollar volume increase.
Additionally, firms with only one split have significantly greater trading volume
than multiple reverse split firms both before and after the split. A graphic
representation of these results appears in Fig. 2. The downward trend in volume
for reverse split one and the considerably lower trading volume for multiple
reverse splits provide further evidence that a reverse split strategy does not lead
to higher liquidity. Perhaps the strategy maintains exchange listing and even
qualifies the stock for institutional holding, but in the end it does not lead to
greater liquidity.
Table 9 documents the long term outcome of firms following reverse stock
splits. The sample includes all firms with reverse splits from 1990 to 2007,
although the outcomes are as of the end of 2010. Forty two percent of only one
reverse split firms have what might be considered a good outcome (19 % still
trading and 23 % mergers). This compares to the 34 % of multiple split firms
with good outcomes. Particularly noteworthy is that one split firms are twice as
likely to be part of a merger as multiple split firms. With respect to bad
outcomes, one split firms are less likely than multiple split companies to incur
liquidations and delistings, 57 % vs. 65 %. Generally speaking, one reverse
split firms do poorly, multiple split firms do worse.

Table 9 Long run outcome of firms with reverse stock splits from 1990 to 2007

One Multiple

Outcome Number Percentage Number Percentage

Still trading 248 19 % 53 22 %

Mergers 309 23 % 29 12 %
Exchanges 16 1% 3 1%
Liquidations 1 0% 2 1%
Delisted 747 57 % 156 64 %
Total 1321 100 % 243 100 %

This table shows the outcome of the companies who initiated a reverse stock split between 1990 and 2007.
The outcome is whether the company is still trading at the end of 2010 or whether it merged (Mergers),
exchanged for other stock (Exchanges), liquidated (Liquidations) or was delisted (Delisted). The sample is
split between companies with one reverse stock split and those with multiple splits. Unlike the rest of the
paper, this only includes reverse splits through 2007 to give 3 years following the reverse split to examine
long term outcome of the company
J Econ Finan (2015) 39:357–369 369

5 Conclusion

Companies that declare reverse stock splits often do so to maintain exchange listing
and public visibility among investors. Moreover, if the split factor is large enough to
get the share price over $5, institutional holders are able to purchase the stock. For
many firms with a low stock price, a reverse split is the only short run mechanism to
increase share price and potentially raise liquidity. However, for the majority of firms,
declaring a reverse split is delaying the inevitable and the majority end up being
liquidated or delisted from their exchange. Thus, a reverse split is often seen by
investors as a desperate tactic by management.
If one reverse split is viewed as desperation, then multiple reverse stock splits are a
sign of extreme distress. In later (second–fifth) reverse stock splits, the median stock
price has again fallen below $1. In addition, firms that declare multiple reverse splits
tend to have even less liquidity than one reverse split firms with an average of four
non-trading days following the reverse split and significantly lower trading volume
than even the single reverse split firms. Companies that declare multiple reverse stock
splits are delaying the inevitable; they are more likely to be liquidated or delisted.
Based on the evidence, investors should walk away from one reverse split firms, but
they should positively flee if a firm continues to declare reverse stock splits.

References

Chung KH, Yang S (2010) Reverse Stock Split and Institutional Investor Behavior, Working Paper
Desai H, Jain PC (1997) Long-run common stock returns following stock splits and reverse splits. J Bus
70:409–433
Han KC (1995) The effects of reverse splits on the liquidity of the stock. J Financ Quant Anal 30:159–169
Hwang CY (1995) Microstructure and reverse stock splits. Rev Quant Finance Account 5:169–177
Kim S, Klein A, Rosenfeld J (2008) Return performance surrounding reverse stock splits: can investors
profit? Financ Manag 37:173–192
Lamoureux CG, Poon P (1987) The market reaction to stock splits. J Finance 42:1347–1370
Martell TF, Webb GP (2008) The performance of stocks that are reverse split. Rev Quant Finance Account
30:253–279
Peterson DR, Peterson P (1992) A further understanding of stock distributions: the case of reverse stock
splits. J Financ Res 15:189–205
Radcliffe RC, Gillespie WB (1979) The Price Impact of Reverse Splits. Financial Analysts Journal, 63–67
Woolridge JR, Chambers DR (1983) Reverse Splits and Shareholder Wealth. Financ Manag, 5–15
Reproduced with permission of the copyright owner. Further reproduction prohibited without
permission.