Content uploaded by Maria Milosavljevic
Author content
All content in this area was uploaded by Maria Milosavljevic on Dec 16, 2014
Content may be subject to copyright.
Available via license: CC BY 4.0
Content may be subject to copyright.
The Impact of Manipulation in Internet Stock Message
Boards
Jean-Yves Delorta,b, Bavani Arunasalama, Maria Milosavljevica, Henry
Leung∗,c
aCapital Markets CRC Limited, Sydney, NSW 2001, Australia
bDepartment of Computing, Macquarie University, Sydney, NSW 2109, Australia
cDiscipline of Finance, Faculty of Economics and Business
Room 402, H69 Economics and Business Building
The University of Sydney, NSW 2006, Australia
P: +612 9114 0554 F:+612 9351 6461
Abstract
This work analyses the impact of manipulation in internet stock message
boards on financial markets. Past research has demonstrated that messages
posted in such boards could be used to predict price return and activity. It
is therefore not surprising that people try to use such forums to influence
other users to act in a particular way. We focus on posts moderated as
manipulative due to ramping. We find that ramping occurs irrespective of
moderation. We show that ramping is positively related to market returns,
volatility and volume. We also demonstrate that firms with higher volume,
lower price level, lower market capitalization and higher volatility receive a
higher proportion of ramping posts. Furthermore, particular sectors (such as
Health Care and Energy) are more common targets of ramping.
Key words: Ramping, Manipulation, Internet Stock Message Boards,
Event Study, Market Reaction
∗Corresponding author
Email addresses: jydelort@cmcrc.com (Jean-Yves Delort), bavani@cmcrc.com
(Bavani Arunasalam), maria@cmcrc.com (Maria Milosavljevic),
h.leung@econ.usyd.edu.au (Henry Leung)
November 1, 2009
1. Introduction
Internet stock message boards provide an effective medium for investors
to communicate, to disseminate and discover new information. Researchers
have been attempting to measure the impact of these message boards on
stock markets, particularly as the number of users and the volume of posts
have increased. For example, Antweiler and Frank (2004) established the
impact of message boards on market volatility. However, the value and in-
tegrity of internet stock message boards is often criticised and investors are
warned not to trade on the information provided (Fraser, 2007). Further,
Australian regulation dictates that message board providers conform to spe-
cific guidelines1. Nevertheless, people continue to use forums for their own
purposes, and financial surveillance analysts at regulators and exchanges are
known to use message boards for identifying cases of market manipulation
and explaining unusual trading activity.
HotCopper2is the most popular internet stock message board in Aus-
tralia and forms the focus of our study. This site operates under a strict
code of conduct which prohibits unethical or illegal use of the forum. Ad-
ministrators moderate posts which do not conform to the guidelines and
can revoke access for users who consistently disregard the rules. Moderated
posts are labeled with the reason for moderation, for example ramping,flam-
ing or profanity. These posts serve to proxy manipulative behaviour. Using
an event study methodology, we investigate whether ramping is related to
stock price volatility, trade volume and returns. Further, we study the firm
characteristics targeted by rampers.
We find that on average, there are X posts moderated due to ramping ev-
ery day which indicates that ramping is a common phenomenon. In addition,
we find that the lead time to moderation is 7.95 hours, allowing sufficient
time for these posts to impact the market. Our results demonstrate that
ramping is positively related to market returns, volatility and volume. Firms
with higher volume, lower price level, lower market capitalization and higher
volatility also receive a higher proportion of ramping posts. Finally, rampers
target particular sectors such as Health Care and Energy.
The remainder of this study is organised as follows. In Section 2 we
1See Australian Securities & Investments Commission (ASIC) Regulatory Guide 162,
http://www.asic.gov.au
2http://www.hotcopper.com.au
2
present an overview of previous work. This serves as motivation for the
development of our hypotheses in Section 3. In Section 4 we describe the data
and empirical methodology used in our analysis. The results are reported
and discussed in Section 5, and Section 6 concludes and provides some areas
for further work.
2. Literature review
Stock market manipulation is generally defined as the creation and ex-
ploitation of arbitrage opportunities (Zigrand, 2006). Manipulation of stock
prices may occur through direct trading strategies or indirectly via the dis-
semination of distorted price sensitive information. The former may be ob-
served when trades are intentionally placed in the wrong direction or short
term losses are being undertaken to move prices in the desired direction
(Chakraborty and Ylmaz, 2004). The latter may occur through mediums
such as investment blogs, spam emails and online stock message forums
(Antweiler and Frank, 2004; Das and Sisk, 2005; Aggarwal and Wu, 2006).
However, the detection of both types of manipulative behaviour remains dif-
ficult due to the lack of proxy for the occurrence of manipulation.
The impact of trade strategy manipulation can be observed in the market
response to such strategies. For example, Aggarwal and Wu (2006) demon-
strates that stock market manipulation cases pursued by the U.S. Securities
and Exchange Commission (SEC) from January 1990 through October 2001
were associated with greater stock volatility, greater liquidity and higher
returns. Other evidence of stock market reaction to manipulation is doc-
umented by Hillion and Suominen (2004), who highlight a general rise in
volatility, trading volume and one-minute returns in the closing price of Paris
Bourse stocks between January and April, 1995. Further, the proportion of
partially hidden orders increased during this period.
Evidence of manipulation has also been found in certain types of deriva-
tives markets such as the futures market. It is shown that uninformed in-
vestors earn positive profits by creating a futures position and simultaneously
trading in the spot market. As a result, arbitrage free profit opportunities are
exploited to derive profitable cash settlement at the time of delivery (Kumar
and Seppi, 1992; Merrick et al., 2005).
Additional evidence of trade strategy manipulation include those found
by Guo et al. (2008), who uses low earnings quality firms to proxy for firms’
intention to manipulate. They find that these firms tend to use stock splits
3
to manipulate equity values before acquisition announcements. On the other
hand, Eom et al. (2009) find that investors strategically place spoofing in-
traday orders in the Korean Exchange. Spoofing orders have little chance of
being executed and are used to mislead other traders of an order book im-
balance and thereby influence the direction of stock price movement. They
conclude that stocks targeted for manipulation had higher return volatility,
lower market capitalization, lower price level, and lower managerial trans-
parency. Finally, large traders have also been found to carry out manipula-
tion due to its scale of operation and its ability to better sequence and time
trades (Gastineau and Jarrow, 1991; Allen et al., 2006).
Manipulative behaviour involving the intentional mis-interpretation and
dissemination of price sensitive information is more difficult to document.
In their attempt to elucidate market manipulation from the 1600s to the
internet technology frauds of today, Leinweber and Madhavan (2001) show
that through technology, information can be quickly disseminated to influ-
ence investor perception on particular stocks. Stock spam emails and posting
on internet message boards are two major avenues manipulators can affect
investor sentiment because of their global reach and high degree of accessi-
bility to investors. In particular, significant reactions of trading volume and
returns have been shown to respond to spam campaigns (Bohme and Holz,
2006). Hanke and Hauser (2008) then extends the works of Leinweber and
Madhavan (2001) and Bohme and Holz (2006) to provide corroborating ev-
idence that stock spam e-mails significantly and positively impact volatility
and intraday spread. Similar evidence is noted by Hu et al. (2009), who
found that pump and dump email campaigns show a statistically significant
decline in stock price from the peak spam day to the following day.
Online message forum is another type of medium sensitive to market
manipulation. Many cases of forum-based market manipulation have been
reported in the literature and in the media. Evidence of online message
board posts’ impact on stock activity is well documented in prior literature
(Antweiler and Frank (2004); Das and Sisk (2005); Aggarwal and Wu (2006)).
Antweiler and Frank (2004) use computational linguistics methods to ana-
lyze the effect of message sentiment (buy, sell or hold) in forums to determine
whether the information content of internet stock messages boards influence
the stock market. They find that stock messages help predict volatility and
disagreement between messages to be associated with increased trading vol-
ume. However, they find no correlation of message sentiments with the di-
rection of returns.
4
Das and Sisk (2005) utilize the medium of stock message board to ex-
amine the social structure of information flow driving stock prices. They
define a conceptual unit named the financial community, in which partici-
pants access the same information source and trades on clusters of similar
stocks. Investor opinions are linked in this way across stocks. Over 23 million
posted messages from January 2000 through April 2001 are extracted from
online stock message boards such as Yahoo Finance, Motley Fool, Raging
Bull and Silicon Investor and used to develop connected financial commu-
nities. They find that strongly connected communities stocks earn higher
mean returns than those not as well connected.
Chemmanur and Yan (2009) provide further evidence of the relation be-
tween advertising and stock returns. They conjecture via their investor at-
tention hypothesis that positive advertising is expected to be conducive to
larger stock returns. By using trading volume and the number of financial
analysts covering to proxy for investors’ attention on a stock, they find that
a greater amount of advertising is associated with a larger stock return. This
advertising effect is found to be stronger for small firms and growth firms.
This is consistent with Brown and Cliff (2004), who show that investor senti-
ment is strongly positively correlated with contemporaneous market returns.
Heavily advertised stocks are also related to trading volume. Barber and
Odean (2008) test and confirm the hypothesis that individual investors are
net buyers of attention grabbing stocks, for example, stocks in the news,
stocks experiencing high abnormal trading volume, and stocks with extreme
one-day returns. This is based on the reasoning that investors need to over-
come the difficulty in selecting stocks to buy from the thousands available.
Thus they choose the ones made known to them. Extreme news content may
also be related to trading volume. Tetlock (2007) uses the Wall Street Jour-
nal to show that high media pessimism predicts downward pressure on stock
prices and extremely high or low pessimism predicts high market trading
volume.
3. Hypotheses development
Unlike the online message forum data sets used in prior literature, this
paper employs moderated posts unique to the Australian HotCopper online
message forum. Improving upon prior literature such as Das and Sisk (2005),
Chemmanur and Yan (2009) and Brown and Cliff (2004), HotCopper mod-
erated posts allow us to proxy for manipulative behaviour directly.
5
Online message forums are exposed to manipulative behavior such as
ramping, as shown by the number of moderated posts in forums such as
HotCopper. Stock ramping through forums occurs when a ramper recom-
mends a stock and highlights its huge potential to rise in share price in the
very near future. The ramper buys a large quantity of a stock, and posts
to influence the masses in to buying the stock with the aim to drive up the
share price. By implying that readers of these posts are in possession of priv-
ileged information - such as insider knowledge of an impending takeover offer
- rampers seek to persuade the gullible into purchasing a particular stock.
If a significant enough number of easily-led individuals invest in the touted
stock, a ramper can “ramp up” the share price so that he or she could sell
their shares at a quick profit. Ramping can be both up or down.
First, the quality of available information on message boards remains
an open question. Even though regulators may require exchanges to moni-
tor illegal activity on stock message boards, difficulty remains in detecting
whether investors are using online forums to express subjective opinions or to
manipulate. Stock message board administrators may function as regulators
through post moderation. Although moderation serves to reduce the impact
of ramping posts on stock prices, the wide accessibility of internet technology
and the high speed at which investors absorb news may diminish the effect
of moderation. Therefore we want to test the following hypotheses:
Hypothesis 1. Moderation does not prevent ramping from taking place in
internet stock message boards.
Additionally, the direct measure of ramping through HotCopper moder-
ated posts allow us form a well specified model that tests whether ramping
is related to stock market activity. This analysis is followed by an investi-
gation into the types of firms targeted by rampers in the forum. Literature
also documents stock message board induced manipulation. Well advertised
(ramped) stocks, in particular, are related to higher stock returns (Das and
Sisk, 2005; Chemmanur and Yan, 2009; Brown and Cliff, 2004). Based on the
results consistent with the investor attention hypothesis proposed in Chem-
manur and Yan (2009), high levels of stock advertising is associated with a
larger stock return.
Stocks heavily promoted to investors via ramping posts may also gener-
ate high trade buy volume. One possible explanation proposed by Barber
and Odean (2008) may be investors’ need to select stocks out of the many
6
available and ones that attract their attention will be the ones most likely to
be included in their investment portfolio.
Furthermore, extreme news announcements have been shown to produce
high trade volume (Tetlock, 2007). Similarly, ramping posts, as a type of
extreme news announcement, may potentially create the same effect.
Both stock spam e-mails and online message forum messages rely on the
internet technology to increase its reach to investors (Leinweber and Madha-
van, 2001). Drawing upon the works of (Bohme and Holz, 2006) and Hanke
and Hauser (2008) on the impact of stock spam e-mails on trade volatility
and intraday spread, we hypothesize that manipulators may employ online
message forum message ramping to influence investor sentiment in similar
fashion.
If ramping posts does indeed have an impact on investor sentiment prior
moderation, we may form the following hypothesis to test the relation be-
tween moderated ramping posts and stock market response:
Hypothesis 2. Moderated online message board posts containing ramping
content are positively related to stock price volatility, volume and returns.
Finally, we attempt to describe the profile of rampers using HotCopper
moderated posts. Since manipulation is more effective when there is greater
uncertainty about the value of a firm, we expect that firms with a higher
return volatility would attract more ramping attempts (Eom et al., 2009).
Firm size has also been shown to be significantly positively associated
with the level of disclosure (Alsaeed, 2006). This can explained by the fact
that larger publicly listed firms receive higher analyst coverage. On the
other hand, smaller firms receive less coverage and thus is more susceptible
to price movements driven by private information release. To that end, we
expect smaller firms to be more prone to manipulation through ramping.
Hanke and Hauser (2008) reports that liquidity is a significant character-
istic of companies targeted by spammers. The lower the liquidity, the larger
the price impact. In addition, the impact of spamming on trade volume
is markedly higher for low-turnover-stocks. The reaction of high-turnover
stocks to spamming is generally lower because it takes a greater number of
participants to move prices. To similar effect, we expect rampers to target
low-turnover stocks (low liquidity).
The above in conjunction produces the following hypothesis:
7
Hypothesis 3. In forums rampers focus on stocks that are not followed by
too many users, that do not have a high volume and that are volatile.
4. Data and methodology
4.1. Data
The data for this study comes from HotCopper, Australia’s largest inter-
net stock message board. The time period for our study runs from January
2008 to December 2008 inclusive. Messages were downloaded from the Hot-
Copper website using software written by the authors, and we restricted our
collection of data to only include messages that have a ticker symbol field
representing an ASX listed company. Each message contains the following
fields: date, time, author, ticker symbol and content. In total, the data set
contains 1,146,223 messages for 1,825 firms listed on the Australian Securi-
ties Exchange (ASX).
An interesting feature of the HotCopper forum is that the site contains
moderated posts which were deemed as manipulative by the forum adminis-
trators. In accordance with Australian law regarding all information available
in the securities market, HotCopper users are expected to comply with a set
of strict usage guidelines3. Messages that do not comply may be moderated
by a moderation volunteer or by an administrator4. Such messages are not
removed from the forum, but their content is replaced by a message which
contains the following information: moderation time, moderation type, and
a comment specifying the reasons for moderation. The full set of moderation
types with their respective percentage composition are listed in Table 1; each
type is a simple phrase which describes the primary reason for moderation.
Some examples of textual comments are provided below:
•All you seem to be doing is plaguing threads with your downramping
on non held stocks. Find something else to do with your time instead
of wasting others. Snide remarks on stocks are not considered helpful
or useful posts.
•Rumours aren’t required thanks. They usually lead to misinformation
on a public forum. No more like this thanks.
3User Posting Guidelines, http://www.HotCopper.com.au/postingguidelines/
4List of moderators, http://www.HotCopper.com.au/forum modList.asp
8
Table 1: Moderation types found on HotCopper in 2008
Moderation cause Total Total %
Other 6,379 45.1%
Flaming 3,583 25.3%
Ramping 1,519 10.7%
Profanity 1,207 8.5%
Spamming 389 2.8%
Defamatory 351 2.5%
Unknown Reason 159 1.1%
Advertising 152 1.1%
Unlicensed Advice 72 0.5%
Duplicate 73 0.5%
Insider Trading 73 0.5%
Copyright 63 0.4%
Sexist 43 0.3%
Racist 43 0.3%
Blasphemous 33 0.2%
Total 14,139
•This post is being moderated because of the unsubstantiated informa-
tion particularly “Every deal done falls through.” This just isn’t true.
Please refrain from flaming other posters too.
In addition to the HotCopper data, we obtained the daily closing prices,
high and low intraday prices and volume data for all companies listed on
the ASX in 2008 from Reuters. We excluded all firms that had fewer than
50 posts during 2008 from our data set. Our final sample size consists of
1,083,913 posts for 938 firms.
9
4.2. Methodology
4.2.1. Stock Market Impact of Ramping
Figure 1: Regression equation for the impact study
Yi,t =β0+β1V oli,t +β2Markett+β3Commenti,t +β4Inf oi,t +X
m∈L
βmLabeli,m,t +i,t (1)
where:
Yi,t - the dependent variable that takes the values of RE Ti,t,S IGi,t ,SP Ri,t or V OLi,t
V oli,t - traded volume of stock ion day t
Markett- price of S&P/ASX200 on day t- used as a control variable for the market effect
Commenti,t - number of moderated posts regarding stock ion day tthat mention ‘ramp’ or ‘unlicensed
advice’ in the moderator’s comment
Inf oi,t - a dummy variable that indicates the release of a company announcement for stock ion day t
L- the set of moderation types given by
{Advertising, Blasphemous, Copyright, Defamator y, Duplicate, F laming,
Insider Trading, Other, P rof anity, Racist, Ramping, Sexist, S pamming,
Unknown Reason, Unlicensed Advice}
Labeli,m,t - number of posts regarding stock ithat were moderated with the label m∈Lon day t
and β0and i,t are the intercept and a random error term respectively.
We apply an event study methodology to analyze the market impact of
ramping in internet stock message boards. We define a trade event for a
certain stock as a day on which the market is open for trading in that stock.
We further define a posting event for a certain stock as a day on which at
least one message has been posted in the message board discussing this stock.
Similarly, we define a moderation event for a certain stock as a day on which
at least one message discussing this stock has been moderated. Finally, we
define a ramping event for a certain stock as a day on which at least one
message has been moderated because of an attempt to ramp this stock.
The performance of a stock is measured in terms of the dependent vari-
ables defined by Hanke and Hauser (2008) and also used by Hu et al. (2009),
that is, stock return, volatility, intraday volatility and turnover. Their defi-
nitions of these variables are shown in Figure 5.
10
As discussed in Section 4.1, HotCopper posts may be moderated due
to several reasons, but each moderated post can only be labeled with one
of the moderation types given in Table 1. This restriction and the close
similarity between some of the moderation types may result in the ramping
posts being labeled differently. For example, a post which could be labeled as
both ‘Unlicensed Advice’ and ’Ramping’ may only be labeled as ‘Unlicensed
Advice’. Generally, posts that are moderated due to multiple reasons are
labeled as ‘Other’ which explains the high percentage of moderation under
‘Other’ (45.1%) as given in Table 1.
We found a significant number of ambiguous moderated posts in which
the moderators assign one label but mention several causes for moderation in
the comment field. Since our focus is on investigating the impact of ramping,
it is important that we incorporate all posts that could potentially be seen as
ramping. As a result, we include posts that mention the keyword ‘ramp’ or
‘unlicensed advice’ in the moderator’s comment as an additional explanatory
variable in our regression analysis.
Figure 2: Regression equation for determinants of target firms
Yi=β0+β1V oli+β2MarketCapi+β3P ricei+β4SI Gi+i(2)
where the explanatory variables are:
V oli- traded volume of stock iduring the sample period.
P ricei- closing price of stock iduring the sample period.
MarketCapi- market capitalization of stock iduring the sample period.
SI Gi- volatility of stock iduring the sample period calculated as described in Section 4.2.1
and β0and iare the intercept and a random error term respectively.
The dependent variable is the proportion of ramping posts, and is calculated as:
Yi=Rampi
Modi
(3)
where
Rampi- the number of ramping posts of stock iduring the sample period.
Modi- the number of moderated posts of stock iduring the sample period.
11
The regression equation for this impact study is given in Figure 1. The
market and volume are used as added control variables. We also included in
our model a dummy variable that takes the value 1 if a company announce-
ment for stock iwas released on day tand 0 otherwise. We include this in
our model in order to control for the impact of company announcements on
the performance of stocks.
4.2.2. Determinants of target firms
As outlined in Section 2, particular types of firms are more common
targets for manipulation than others. We use the regression model shown in
Figure 2 in order to investigate the characteristics of the firms targeted by
rampers.
Figure 3: Regression equation for sectors of targeted firms
Yi=X
m∈S
βmSectori,m +i(4)
where
Yi- the proportion of ramping posts as calculated in Figure 2
S- the set of GICS industry sectors given by {Energy, Materials, Industrials, Consumer Discretionary,
Consumer Staples, Health Care, Financials, Information Technology, Telecommunication Services,
Utilities}
Sectori,m - a dummy variable for sector m∈Sthat takes the value 1 if stock ibelongs to sector mand
0 otherwise.
and iis a random error term.
We also investigate whether rampers target companies in certain industry
sectors. We use the sector codes given by the Global Industry Classification
Standard (GICS). The regression model used for this analysis is given in
Figure 3.
5. Results
5.1. Descriptive statistics
5.1.1. Moderation activity
We define moderation delay as the difference between publication time
and moderation time for a particular post on an internet message board. In
12
Figure 4: Ramping vs moderation activities
5 10 15 20
0.2 0.4 0.6 0.8 1.0 1.2 1.4
Hour of the day
Proportion
Moderated posts
Ramping posts
Hour of the day
Proportion
This figure plots the proportions of moderated posts and ramping posts divided by the proportion of posts
with respect to the hour of the day.
2008, for the 938 stocks included in the sample, the mean moderation delay
was 7.95 hours (σ= 15.93). In addition, 40.78% of moderated posts where
moderated after two hours. This is sufficient time for ramping on forums to
be effective, thus providing an environment which supports Hypothesis 1.
We found that in general, inappropriate posts tend to be published outside
of normal trading hours. This is demonstrated by the graph shown in Figure
4 which represents the proportion of moderated posts and with respect to
the hour of day. What is most interesting here is the daily pattern of posts
13
which are moderated because of ramping. The graph identifies quite clearly
that ramping is more active during trading hours which is quite different
to the general pattern for moderated posts. We believe that the reason for
this is that ramping messages created outside of normal trading hours are
less likely to be effective since they are more likely to be moderated before
market opening.
5.1.2. Dependent variables
Table 2: Mean and stddev for all dependent variables
Nb. events Price RET SIG SPR VOL
Panel A: Trade event 238252
Mean 1.9820 -0.0052 0.0785 0.3747 1
Median 0.24 0 0.3747 1 1.4717
St.dev. 7.7096 0.0785 0.7776 0.7116 0.518
Panel B: Posting events 100695
Mean 2.3754 -0.0022 0.0878 0.4817 1.1994
Median 0.265 0 0.4817 1.1994 1.6403
St.dev. 9.7972 0.0878 0.8489 0.8886 0.6476
Panel C: Moderation events 4870
Mean 2.7410 -0.0053 0.1151 0.5474 1.3636
Median 0.245 0 0.5474 1.3636 2.6274
St.dev. 9.1831 0.1151 1.019 0.9844 0.7723
Panel D: Ramping events 786
Mean 1.6275 0.0027 0.1395 0.6446 1.8063
Median 0.19 0 0.6446 1.8063 5.5706
St.dev. 5.7589 0.1395 1.2379 1.1167 0.9636
This table presents the mean, median and standard deviation for the dependent variables namely, Return
(RET), Volatility (SIG), Intraday volatility (SPR) and Turnover (VOL). Panel A presents the statistics
for all the firms across all the days for January 2008 through December 2008. Panels B, C and D present
the same statistics only for days with at least one post, days with at least one moderated post and days
with at least one ramping post respectively.
Table 2 reports descriptive statistics for our data sample. In 2008, for
the 938 stocks included in the sample, we have 768 ramping events occurring
with 100,695 posting events. As shown in Table 1, ramping events represent
10.7% of the total number of moderation events; from Table 2 we can see
that moderation events contain at least one ramping post 15.77% of the
time. Because of the increasingly high volume of messages that are posted
on internet stock forums, it seems likely that inappropriate posts will not be
discovered in the manual moderation process. Our data is clearly restricted
14
to posts which are discovered however it is reasonable to assume that the
real number of ramping events is higher than our figures indicate. This
would clearly not be limited only to ramping, but is relevant to all forms
of moderation events which co-occur with 4.8% of posting events. Note also
that messages are not posted every day for all stocks, and that posting events
represent only 42.26% of the total number of events for the entire sample.
The year 2008 represents an unusual case study because of the unprece-
dented global financial crisis which hit the market in the last quarter. On
an average day during January 2008 to December 2008, the mean return was
negative for trade events, posting events and moderation events. In con-
trast, the mean return for ramping events was positive. This indicates that
ramping co-occurs with price increase.
Ramping events have the lowest mean price of the four types of events. It
indicates that the “pump and dump” interest is the highest for small-priced
firms. Table 2 also shows that ramping events have the highest mean for
return, volatility, intra-day volatility and volume which supports Hypothesis
2. In section 5.3 we further investigate the impact of ramping on the four
dependent variables taking into account the firm characteristics.
5.2. Impact of manipulation in IMS
Table 3 shows the results of the regression analysis for the model given in
1. The results show that Volume, Market, the release of company announce-
ments and Ramping are significantly correlated to return, volatility, intra-day
volatility and volume. The correlation coefficients are positive for the four
variables, but the confidence is stronger for volatility, intra-day volatility
and volume. Insider Trading is also is also significantly correlated to volume,
intra-day volatility and return. We previously introduced the dummy vari-
able ‘Ramping in comment’ which corresponds to moderated events where
at least one moderation comment contain the word “ramp”. The variable is
also correlated to the two volatility measures, and volume to a lower extent.
For volume we find a positive correlation with Ramping, Insider Trading,
Other, Advertising, Flaming and Profanity. Other is a general category that
is used for posts moderated because of miscellaneous or mixed reasons (and
so can include Ramping, Flaming and Profanity).
Our results are limited by the fact that we only have closing prices and
do not currently have access to intra-day prices. This results in an inability
to distinguish between different moderated posts, that is, we cannot identify
the impact of a specific ramping post on the market. Indeed, a ramping
15
Table 3: Impact study results
RET SIG SPR VOL
Variable Est. t-stat Est. t-stat Est. t-stat Est. t-stat
Intercept 0.0166 15.2390 *** 6.9634 86.5380 *** 4.6550 29.8520 *** -3.3960 -64.8660 ***
Volume 0.0009 17.6740 *** 0.0550 170.8620 *** 0.0992 157.6600 ***
Market 0.0044 22.1930 *** -0.8165 -85.6260 *** -0.5583 -30.2240 *** 0.4563 74.1110 ***
Ramping 0.0131 2.5530 * 0.1428 4.2970 *** 0.2816 4.2890 *** 0.2560 11.5920 ***
Spamming -0.0163 -2.1080 * -0.0460 -0.8840 -0.1540 -1.4860 0.0004 0.0120
Insider Trading 0.0129 0.8020 0.1129 0.9590 1.2281 4.8220 *** 0.3740 4.7730 ***
Other 0.0047 2.0800 * -0.0233 -1.6500 . -0.0968 -3.4970 *** 0.0979 10.4490 ***
Unknown Reason 0.0033 0.2660 0.0306 0.3930 0.0187 0.1250 0.0745 1.4370
Unlicensed Advice 0.0047 0.2450 -0.0913 -0.7510 -0.2059 -0.8770 0.0652 0.8050
Flaming -0.0061 -2.1050 * -0.0039 -0.2130 0.0065 0.1810 0.0699 5.7930 ***
Defamatory -0.0175 -2.1110 * 0.0360 0.6720 -0.1362 -1.3150 0.0016 0.0460
Duplicate 0.0211 0.9990 -0.0148 -0.1240 -0.1753 -0.7450 0.0413 0.5200
Advertising 0.0102 0.8230 -0.0202 -0.2410 0.0408 0.2520 0.2176 3.9640 ***
Profanity -0.0061 -1.3240 0.0389 1.3000 0.1073 1.8130 . 0.2227 11.2000 ***
Copyright -0.0118 -0.6230 0.0070 0.0540 -0.1801 -0.7090 0.0197 0.2300
Ramping in comment -0.0004 -0.0630 0.0813 2.0880 * 0.2430 3.1460 ** 0.0472 1.8220 .
Company info 0.0080 13.2160 *** 0.0736 19.1700 *** 0.1368 18.4040 *** 0.1641 65.0160 ***
R20.010 0.1266 0.1044 0.0441
n104375 233585 228075 238236
This table reports the results of the regression analysis with four dependent variables: Return(RET), Volatility (SIG), Intraday volatility (SPR)
and Turnover(VOL) in turn. RE Ti,t =ln Si,t −ln Si,t−1where Si,t is the closing price of stock ion day t.SI Gi,t =r(RETi,t −RETi)2
σ2
i
where
RET iis the average RETi,t over the sample period and σ2
iis the average variance of stock iover the sample period. S P Ri,t =ςi,t
ςiwhere ςi,t is the
intraday price range of stock ion day tgiven by ςi,t =ln(intraday high)−ln(intraday low) and ςiis the average intraday price range stock iover
the sample period. V OLi,t =ln h1 + toi,t
toiiwhere toi,t is the turnover of stock ion day tand toiis the average turnover of stock iover the sample
period. The explanatory variables are as follows : V olumei,t is the traded volume of stock ion day t.Markettis the price of S&P/ASX200 on day
t. Ramping, Spamming, Insider Trading, Other, Unknown Reason, Unlicensed Advice, Flaming, Defamatory, Duplicate, Advertising, Profanity and
Copyright indicate the number of posts moderated under the respective category for the respective stock on a given day. Ramping in commenti,t
is the number of moderated posts regarding stock ion day tthat mention ‘ramping’ in the moderator’s comment. Company I nfoi,t is a dummy
variable that indicates the release of a company announcement for stock ion day t. All data are for January 2008 through December 2008. All the
explanatory variables are log transformed except for Inf oi,t.
16
Table 4: Determinants of target firms
(1) (2) (3) (4) (5)
Intercept -0.0992 * 0.1239 ** 0.0704 *** 0.0058 *** -0.0906
(-2.489) (2.980) (11.450) (0.453) (-1.341)
Volume 0.0123 *** 0.0186 ***
(4.025) (3.545)
Market cap -0.0036 -0.0059
(-1.506) (-1.243)
Price -0.0188 ** -0.0157
(-2.856) (-1.261)
Volatility 1.0479 *** 0.4135
(-2.306) (1.251)
R20.017 0.0029 0.0086 0.02203 0.0501
n936 758 935 896 750
This table presents the results of the regression analysis described in Section 4.2.2. The dependent variable
is the proportion of ramping posts given by Rampi
Modiwhere Rampiis the number of ramping posts of stock
iduring the sampling period and M odiis the number of moderated posts of stock iduring the sampling
period. V olumeiis the average turnover for stock iduring the sampling period, M arket capiis the market
capitalization of stock imeasured at the end of 2008, P riceiis the average closing price for stock iduring
the sampling period and V olatilityiis the average volatility of stock iduring the sampling period.
Table 5: Targeted sectors
Code Sector Estimate T-value
10 Energy 0.0789 *** 6.919
15 Materials 0.0627 *** 8.141
20 Industrials 0.0310 . 1.761
25 Consumer Discretionary 0.0475 * 2.111
30 Consumer Staples 0.1173 ** 2.875
35 Health Care 0.0706 *** 3.756
40 Financials 0.0306 . 1.831
45 Information Technology 0.0338 1.292
50 Telecommunication Services 0.0776 * 2.156
55 Utilities 0.0764 . 1.804
R20.147
n911
This table presents the results of the second regression analysis described in Section 4.2.2. The dependent
variable is the proportion of ramping posts given by Rampi
Modiwhere Rampiis the number of ramping posts
of stock iduring the sampling period and M odiis the number of moderated posts of stock iduring the
sampling period.
post cannot be measured independently of other moderated posts (such as
profanity) which occur on the same day. Hence, we cannot accurately identify
whether a specific ramping post has been effective in its cause.
17
5.3. Determinants of target firms
In this section, we present our findings in investigating the characteristics
of firms which are the most common targets for ramping in forums. Since
some of the independent variables are significantly correlated, we implement
both simple and multiple regressions in order to minimize the multicollinear-
ity problem. In the simple regressions (regressions (1) through (4)), firms
with higher volume, lower price level and higher volatility tend to receive
a higher proportion of ramping posts. Market capitalization is also nega-
tively correlated to the proportion of ramping posts but with a low signifi-
cant value. Regression (5), which controls for multicollinearity, confirms that
firms with higher volume, lower price level, lower market capitalization and
higher volatility receive a higher proportion of ramping posts. However, the
correlation is highly significant only for volume.
We now analyze if rampers target companies in specific sectors. Table 5
shows that the proportion of ramping posts is positively correlated to Energy,
Materials and Health Care. It is also correlated to a lesser extent to Consumer
Staples, Consumer Discretionary and Telecommunication Services.
6. Conclusion
By using a unique Australian HotCopper online stock message forum
database, we extract the posts moderated as ramping and examined its re-
lationship with market response. First, we find that ramping occurs despite
attempts by moderators to alert users of ramping. This is consistent with
our expectation that rampers in forums are difficult to identify even for mod-
erators. The possible reasons could be due to the fact that ramping can be
a long progressive process, by which a ramper inconspicuously disseminates
his ideas but attempts to hide the motivation to ramp. Ramping is positively
and significantly related to returns, volatility and volume. Further rampers
target firms with higher trade volume, lower price levels, lower market cap-
italization and higher volatility. Particular sectors such as Health Care and
Energy are common targets of ramping.
Our proxy for ramping is not perfect. Future works may include the devel-
opment of techniques to detect well-concealed smart manipulators amongst
posts not yet reviewed by moderators. For example, text mining techniques
could be employed to detect ramping in the following unmoderated posts
which by human interpretation obviously suggests the poster’s intention to
ramp.
18
•i heard a rumor that the plant is still being repaired and may be oper-
ating again later today. Does anyone know how to contact the fellow
from Atmos that is mentioned in this article?
•I’m focusing on oil stocks now, not effected by all this, and also I heard
a rumor even General Motors now admit our oil has run out on what
we can produce in one day and now in decline which is good news to
the oil stocks.
References
Aggarwal, R. K., Wu, G., 2006. Stock market manipulations. Journal of
Business 79 (4), 1915–1953.
Allen, F., Litov, L., Mei, J., 2006. Large investors, price manipulation, and
limits to arbitrage: An anatomy of market corners. Review of Finance 10,
645–693.
Alsaeed, K., 2006. The association between firm-specific characteristics and
disclosure: The case of saudi arabia. Managerial Auditing Journal 21 (5),
476 – 496.
Antweiler, W., Frank, M. Z., 2004. Is all that talk just noise? the information
content of internet stock message boards. The Journal of Finance 59 (3),
1259–1294.
Barber, B. M., Odean, T., 2008. All that glitters: The effect of attention
and news on the buying behavior of individual and institutional investors.
Review of Financial Studies 21 (2), 785–818.
Bohme, R., Holz, T., 2006. The effect of stock spam on financial markets.
Brown, G. W., Cliff, M. T., 2004. Investor sentiment and the near-term stock
market. Journal of Empirical Finance 11, 1–27.
Chakraborty, A., Ylmaz, B., 2004. Manipulation in market order models.
Journal of Financial Markets 7, 187–206.
Chemmanur, T., Yan, A., February 2009 2009. Advertising, attention, and
stock returns.
19
Das, S., Sisk, J., 2005. Financial communities. Journal of Portfolio Manage-
ment 31 (4), 112–123.
Eom, K. S., Lee, E. J., Park, K. S., 2009. Microstructure-Based Manipu-
lation: Strategic Behavior and Performance of Spoofing Traders. SSRN
eLibrary.
Fraser, J., 2007. The mysterious world of stock forums. Compare-
shares.com.au.
Gastineau, G. L., Jarrow, R. A., 1991. Large-trader impact and market reg-
ulation. Financial Analysts Journal 47, 40–72.
Guo, S., Liu, M. H., Song, W., 2008. Stock splits as a manipulation tool:
Evidence from mergers and acquisitions. Journal of Financial Manage-
ment (Winter), 695 – 712.
Hanke, M., Hauser, F., February 2008. On the effects of stock spam e-mails.
Journal of Financial Markets 11 (1), 57–83.
Hillion, P., Suominen, M., 2004. The manipulation of closing prices. Journal
of Financial Markets 7, 351–375.
Hu, B., McInish, T., Zeng, L., 2009. The can-spam act of 2003 and stock
spam emails. Financial Services Review 18 (1), 87–104.
Kumar, P., Seppi, D. J., 1992. Futures manipulation with cash settlement.
The Journal of Finance 47 (4), 1485–1502.
Leinweber, D. J., Madhavan, A. N., 2001. Three hundred years of stock
market manipulations. The Journal of Investing 10 (2), 7–16.
Merrick, J. J., Naik, N. Y., Yadav, P. K., 2005. Strategic trading behavior and
price distortion in a manipulated market: anatomy of a squeeze. Journal
of Financial Economics 77, 171–218.
Tetlock, P. C., 2007. Giving content to investor sentiment: The role of media
in the stock market. The Journal of Finance 62 (3), 1139–1168.
Zigrand, J.-P., 2006. Endogenous market integration, manipulation and lim-
its to arbitrage. Journal of Mathematical Economics 42, 301–314.
20
Figure 5: Dependent variables defined by Hanke and Hauser (2008)
Stock Return:
RETi,t =ln Si,t −ln Si,t−1(5)
where Si,t is the closing price of stock ion day t.
Volatility:
SI Gi,t =s(RETi,t −RE Ti)2
σ2
i
(6)
where RET iis the average RETi,t over the sample period; and σ2
iis the average variance of stock
iover the sample period.
Intraday volatility:
SP Ri,t =ςi,t
ςi
(7)
where ςiis the average intraday price range stock iover the sample period; and ςi,t is the intraday
price range of stock ion day tgiven by:
ςi,t =ln(intraday high)−ln(intr aday low)
Turnover:
V OLi,t =ln 1 + toi,t
toi(8)
where toi,t is the turnover of stock ion day tand toiis the average turnover of stock iover the
sample period.
21
