effectiveness and market reaction to the stock exchange's inquiry in australia

Journal of Business Finance & Accounting, 34(7) & (8), 1141–1168, September/October 2007, 0306-686Xdoi: 10.1111/j.1468-5957.2007.02020.x

Effectiveness and Market Reaction to theStock Exchange’s Inquiry in Australia

Ning Gong∗

Abstract: This paper examines a unique stock market monitoring program used by theAustralian Stock Exchange (ASX). When the ASX observes unusual share price or trading volumechanges of a listed company, it sends a letter demanding an explanation. Companies need torespond publicly to several stylized questions. Such public communications between the stockexchange and listed companies contain information. This paper documents how companiesrespond to the ASX inquiry and how the market reacts to the replies. It is found that somecompanies do release new information to the market when asked. After the firm’s reply is posted,the average trading volume and the bid-ask spread are reduced, and in most cases, the share priceis also stabilized with the following two exceptions: (1) The price will continue to rally on averageif the company releases only partial information when questioned after a significant price jump;(2) The downward price trend will be reversed if the company states that no new informationcould explain the decline. Furthermore, there are statistically significant, positive abnormalreturns for the first five trading days, which are not conditional upon the replies firms give tothe ASX inquiries.

Keywords: market monitoring, information disclosure, stock exchanges

1. INTRODUCTION

In competitive markets, share prices reflect all publicly available information. When afirm’s share price falls or rises significantly over a short period of time, investors rightlydeserve an answer as to what has happened to that firm.

The Australian Stock Exchange (ASX) has a well-established market monitoringprogram which sends inquiry letters to firms experiencing a significant price or

∗The author is from the University of Melbourne. He is grateful to Valdimir Atanasov, Bruce Grundy, BobOfficer, Chander Shekhar, Martin Walker (editor), and an anonymous referee for helpful comments. Anearlier version of the paper, titled ‘Stock Exchange as a Monitor: The Australian Experience’ was presentedat the Australian Graduate School of Management, Edith Cowan University, University of Western Australia,Melbourne Business School, 2001 FMA European Meetings, and the 13th Australasian Finance and BankingConference in Sydney. The author also thanks Elizabeth Melville-Jones of the ASX for providing theinstitutional details. Finally, thanks to Laurel Chen for her outstanding research assistance and Sylvia Jonesfor editing. (Paper received November 2006, revised version accepted January 2007. Online publication May2007)

Address for correspondence: Ning Gong, Melbourne Business School, University of Melbourne, 200Leicester Street, Carlton, VIC 3053, Australia.e-mail: [email protected]

C© 2007 The AuthorJournal compilation C© 2007 Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UKand 350 Main Street, Malden, MA 02148, USA. 1141

1142 GONG

volume change. The motive is to promote market efficiency through timely informationdisclosure and reduce the opportunity for corporate insider trading and self-dealings.1

How firms respond to the inquiry letters and how the market reacts to the publiclyposted replies are the two main questions I have investigated.

In this study, it is found that 30% of the companies in the sample do release newinformation to the market when queried. After the firm’s reply is posted, the averagetrading volume and the bid-ask spread are reduced, and in most cases, the share priceis also stabilized with the following two exceptions: (1) The price will continue torally on average if the company releases only partial information when questionedabout a significant price jump; (2) The downward price trend will be reversed if thecompany states that there is no new information which could explain the price decline.Furthermore, there are statistically significant, positive abnormal returns for the firstfive trading days, which are not conditional upon the replies firms give to the ASXinquiries about price declines.

This study has a number of policy implications. First, it affirms the important roleplayed by the ASX in promoting market efficiency. With the establishment of such amarket monitoring program, firms are more aware of the listing standard and disclosurerequirements, and some of them do release additional information when requested.This process may speed up the information flow in the market place and reduceinformation asymmetry among traders. Consequently, it improves liquidity and lowersthe cost of capital. Second, this is an opportunity to study how the market responds tosuch public communications between the stock exchange and listed firms. For example,it is found that the share price on average levels off when companies state that theyhave no new information to explain why the price ran up prior to the inquiry. It mustbe asked why the price does not fall after such a statement. Third, the ASX inquirysystem is an alternative mechanism which can be compared with practices in othercountries, in particular the US. For example, the New York Stock Exchange (NYSE)has Stock Watch, a computerized system, which automatically flags unusual volume orprice changes in listed stocks, helping the Exchange to guard against manipulation andinsider trading. The NYSE sometimes halts trading in a firm to investigate the possiblecauses of order imbalance. However, there is no public communications system in theNYSE similar to the one discussed here; hence, its main monitoring role is to guardagainst brokers/dealers’ abuse of their power.

Currently in many countries there is a debate on how to reduce the costs of marketfraud effectively and increase the level of information disclosure, whether throughreforming the shareholders’ class action litigation or enhancing stock exchanges’ mar-ket monitoring role. In the US, class action lawsuits are often used as a weapon againstmanagers’ misstatements under the Rule 10b-5 of the Securities and Exchange Act of1934. Such lawsuits are considered too expensive and lack deterrence. Most of them aresettled by out-of-court payments, even if some of the cases have little merit. Settlementallows managers to avoid personal liability by paying the claims with the corporation’smoney. In addition, the class action lawsuits focus too much on compensation paid to

1 For example, on the Australian Stock Exchange’s (ASX’s) website, the following paragraph summarizesthe ASX’s reasons for conducting share market monitoring: ‘Market integrity of the share market is vitalto everyone: private and institutional investors, listed public companies, stockbrokers, government and thecommunity. Without integrity, market participants could be discouraged through lack of confidence, fromraising capital in the primary market or investing in existing shares. Capital and markets could drift overseas.Growth in the Australian share market could slow and market efficiency could be adversely affected.’

C© 2007 The AuthorJournal compilation C© Blackwell Publishing Ltd. 2007

THE STOCK EXCHANGE’S INQUIRY IN AUSTRALIA 1143

misled investors, but not enough on curbing managers’ misbehavior. Pritchard (1999)argues that the low-cost alternative is antifraud monitoring and enforcement by thesecurities exchanges where the trading affected by the fraud takes place. The reasonsare that stock exchanges have the infrastructure to monitor share prices and incentivesto reduce fraud in order to promote market liquidity and increase their revenues. InAustralia, the ASX has an established market monitoring program and so class actionshareholder lawsuits against managerial fraud are not prevalent.2 Therefore, this studyoffers some support for Professor Pritchard’s low cost proposal in limited ways.

The paper is organized as follows. Section 2 offers a brief literature review on theneed for information disclosure and the role of stock exchanges. Section 3 outlinesthe mechanics of the market surveillance process by the ASX. Section 4 describes thedata base and examines the issues of who gets the inquiry letter and what the marketresponses are. Section 5 considers the after-inquiry market performances and analyses.The concluding section contains a brief summary.

2. INFORMATION DISCLOSURE AND THE MONITORING ROLE OF THE STOCKEXCHANGE: A BRIEF LITERATURE REVIEW

It is well known from the literature that an increased level of information disclosure bylisted companies and the resultant more accurate stock prices will benefit shareholdersand reduce the cost of equity capital. The benefits of information disclosure byfirms include: (1) reduced private information collection costs and improved risksharing among traders (Diamond, 1985); (2) reduced information asymmetry amongshareholders and improved stock market liquidity (Amihud and Mendelson, 1986;and Diamond and Verrecchia, 1991); and (3) more accurate tools for shareholdersto evaluate and scrutinize management, and enhanced corporate governance andmanagerial accountability (Diamond and Verrecchia, 1982; Fox, 1997; and Lowenstein,1996). Botosan (1997) provides empirical evidence of the link between disclosure leveland the cost of equity capital.

If the benefits of information disclosure by firms are well understood, why aremanagers sometimes reluctant to release information promptly and why do theyeven misrepresent material facts when they know true information will eventuallybe revealed? Several motives may be at work. (1) From the agency point of view,managers may conceal a decline in earnings or other bad news for fear of losing theirjobs if performance continues to suffer. They hope to buy time to turn the prospectsof the company around (Arlen and Carney, 1992). (2) Managers are motivated byopportunities for insider trading. It is often alleged that corporate insiders makedisclosures strategically so they can realize the maximum value of their own shares oroptions (Carter, Mansi and Reeb, 2003). (3) Managers protect shareholders’ interestsbecause full disclosure of commercially sensitive information may damage the firm’scompetitive edge (Gertner, Gibbons and Scharfstein, 1988).

In general, information disclosure is considered beneficial in a collective sense, butindividual firms and their managers may not find it desirable in certain circumstances,especially when they have bad news. Thus there is room for either government imposedregulation or industry-wide self regulation.

2 However, Donnan (2000) argues for the US style class actions in securities fraud cases in Australia inpromoting market fairness and investor protection. Clearly, the debate is going on.


1144 GONG

The Australian Stock Exchange monitors share prices and trading volumes to checkwhether listed firms follow the disclosure regulation. A sudden change in share priceor trading volume may not necessarily be caused by new information about the firm’sfuture cash flows or risk exposure. In the existing literature there are perhaps threeschools of thought with regard to asset prices in the market. In the more establishedefficient-market’s view, asset prices are rational assessments of expected future payoffsand discount rates, and changes in prices reflect the arrival of new information aboutfuture payoffs or interest rates. A sudden dramatic change in price without properinformation disclosure indicates that the managers of a firm may be withholdinginformation or misleading the market and thus violating the listing rule. The secondis a behavioral view in which there is a component of prices, termed as ‘fads’ or‘noises’, which is not tied to fundamentals, and in which changes in the asset pricesmay therefore reflect changes in the non-fundamental component as well as changesin fundamentals. In the third ‘middle-of-the-road’ view, rational price movements inasset prices arise from the trading process because the market’s initial reaction tonews does not fully reflect investors’ assessments of its implications for fundamentals.Further trading reveals additional information about those assessments.3 Therefore,in the latter two views, asset prices may move without news.

The ASX’s market surveillance approach is consistent with the efficient market view.The monitoring program tries to identify whether the market price moved by news andif so, companies need to conform to its listing standard. Challenging the company inpublic also sends a signal to those traders who might possess inside information thatthe stock exchange is watching. It reduces the exchange’s potential legal liability toinvestors if the firm later is found to be violating the listing rule.

Although many articles have been written on the benefits and costs of informationdisclosure for firms, and on the stock exchange’s motivations on market monitoringand surveillance,4 none has empirically documented the market responses of thesurveillance activity by the stock exchange. This is exactly the purpose of this study.

3. AN OVERVIEW OF THE ASX MARKET SURVEILLANCE PROCESS

The ASX itself is a publicly listed company. It is the main stock exchange in Australia,having offices in all capital cities. The ASX is an order driven market for listed securities.Orders are matched through its Stock Exchange Automated Trading System (SEATS).All transactions, unless they meet the criteria for Special Crossings, are handledautomatically and recorded on the ASX’s computer system. Special Crossings, whichinclude block trades and large portfolios, can be executed off-market but must bereported to the market through SEATS.5

Firms listed on the ASX have to comply with its regulations, in particular, the listingrule of continuous disclosure of material information. Continuous disclosure is the

3 Any standard textbook in finance has a chapter on the efficient market hypothesis. For the relationshipbetween investor psychology and market prices, see De Bondt and Thaler (1995). Grundy and McNichols(1989) and Romer (1993) offer several explanations on rational asset price movements without news.4 For example, Fischel and Grossman (1984), Mahoney (1997) and Huddart et al. (1999) argue securitiesexchanges have appropriate incentives to regulate the market because these rules are what the investorswant. However, Pirrong (1999 and 2000) argues that the investor protection offered by exchanges may besuboptimal because of interest group rent seeking behavior within the exchange membership.5 For detailed descriptions of the operating of the Australian equities market, see the ASX publication, AnOverview of the Australian Equities Market (March, 2000).



timely advising of information to keep the market participants informed of events anddevelopments which may affect market prices as they occur. Listing rule 3.1 of the ASXsays that:6

Once an entity is or becomes aware of any information concerning it that a reasonableperson would expect to have a material effect on the price or value of the entity’ssecurities, the entity must immediately tell the ASX that information.

However, there are a few exceptions to the rule, which is generally termed a ‘carve-out’rule of disclosure. Firms do not have to disclose an incomplete proposal or negotiation,nor information that is a trade secret or confidential. This ‘carve-out’ rule causesambiguities.

The Surveillance of Market Activity (SOMA) system of the ASX detects unusualtrading patterns in price and/or trading volume of a stock, which then trigger aninvestigation executed by a surveillance analyst, who may make a further inquiry to seewhether the price or volume change is related to any prior information disclosure oris an industry-wide or market-wide phenomenon. The analyst then refers the case tothe listing officer if preliminary investigations cannot explain the large movement ofthe share price or volume. What scale of change is considered unusually big so that itwarrants an inquiry? The ASX does not specify an exact threshold. In this study, thedistribution of change in share prices which triggered the inquiry is documented inFigure 1. It seems that the majority of the firms which were investigated had a share pricechange of 20% or greater. Some firms may be challenged with a smaller percentagechange for other reasons.

If the listing officer decides to make a formal inquiry, a short phone call will bemade to the firm regarding the matter and then a formal letter will be sent to thecompany, demanding answers within a specific time period (usually one trading dayor less). Companies’ responses are classified as ‘market sensitive’ and will promptly bemade public by the ASX.

There are three types of query letters which can be sent to companies by the ASX.

(i) Price Inquiry

A typical letter in this category asks the following four questions: (1) Is the companyaware of any new information that has not yet been announced? (2) If the answer to thefirst question is yes, then can an announcement be made immediately? The companyalso has an option to apply for a trading halt for its securities if it is aware of somenew market sensitive information, which cannot be announced immediately. (3) Canthe company offer some explanation for the price change of its securities? (4) Is it incompliance with the ASX listing rule of continuous disclosure of material information?A sample query letter is contained in Appendix A.

In some cases, when the letter is sent close to the half-yearly financial reporting date,two additional questions are asked. The first is whether the company believes that therewill be a change in the operating profit or loss before abnormal items and income taxsuch that the figure for the current financial year would vary from the previous yearby more than 15 percent. The second question is whether the company has reason to

6 The detailed rule can be found at the website: www.asx.com.au/ListingRules/chapters/ch03.pdf


1146 GONG

believe that it will record any material abnormal or extraordinary profits or losses forthe current year (or half year).

Obviously, the objective for such inquiries is to ensure that companies adhere to theASX listing requirements and disclose information which may be already used by someinvestors whose trades have caused abnormal price movements in the first place.

(ii) Volume Inquiry

If the trading volume of a company’s security increases dramatically, then the ASXasks the company to provide an up-to-date list of the top twenty shareholders, togetherwith the number of securities and the percentage of capital each holds. However, suchinformation is classified as ‘not market sensitive,’ in contrast to the response of shareprice changes even though it may be materially relevant. Large investors may haverecently changed their holdings because they have better information about the firmthan small investors; or they may be suspected of trying to get a ‘toehold’ in a potentialtakeover target.

The analysis of volume inquiry is handicapped because the period in which thevolume changes dramatically is not specifically pointed out in the ASX query letters.Therefore, in this paper, I will concentrate on the analysis of query letters related toprice changes although for completeness, a sample query letter on volume change isalso provided in Appendix A.

(iii) Inquiry Made on Other Issues

From time to time, the ASX also asks companies to clarify other issues which may ormay not be considered market sensitive. Such inquiries require firms to confirm orrefute speculations made by journalists or clarify the status of negotiations previouslyannounced. The ASX has expanded its queries since the April 2000 technologymeltdown and increasingly it asks questions on issues other than price or volumechanges. These may be about cash reserves for technology and mining companies,takeover rumors, or specific questions on quarterly reports, etc.

Since the SOMA system allows the analysts to identify every bid and offer, it is possibleto scrutinize the activities of a particular person trading in the stock. This function isimportant because such surveillance may prevent, or lead to prosecution for, insidertrading. However, the focus of this paper is on the monitoring of the share priceand trading volumes on the company level, not on the trading activities of individualtraders.

4. DATA AND SAMPLE

I begin the analysis by examining the written replies to ASX inquiries. An initial sampleof 1,233 responses to queries from the ASX about share price changes, trading volumes,and other matters in the period from July 1, 1998 to June 30, 2000 was collected by handmainly through the website of the Trading Room (www.tradingroom.com.au), whichposts company announcements to the ASX and offers price and volume informationas well. Price and volume data for each company and the All Ordinaries AccumulationIndex were obtained, and sometimes cross-checked from the same website, the CoreResearch Data of the Securities Industry Research Centre of Asia-Pacific (SIRCA) at



Table 1Summary of the Data Sample

Number of Cases Percentage

Total sample 1,233 100%

Price query 957 77.6

Sample used 934 75.8Further query 13 1.0Further response 3 0.2Sample excluded 7 0.6

Volume query 102 8.3

Other query 174 14.1

Market sensitive 15 1.2Non-market sensitive 159 12.9

Notes:This table summarizes the cases of the ASX inquiries from July 1, 1998 to June 30, 2000. The totalsample size is 1,233, including price query, volume query and other market sensitive or non-market sensitivequeries.This paper focuses on the data from the price query, which is the only subset that indicates the range andperiods of the price fluctuations. The first query letter is used even though the ASX may have made furtherenquiries because the company failed to give a satisfactory response or the company said it was ‘unaware ofany new information’ but released market sensitive announcement shortly afterwards. The same applieswhen the company made further corrections or provided new information after the first announcement,mostly on the next trading day. The data for five price inquiries are excluded because the query letter,periods of price change and price up to t = 30 are not available. Also excluded are two inquiries from anoverseas stock exchange.ASX requests information about the top 20 security holders in the volume query when there is a substantialincrease in the volume of the shares traded. Other queries request clarification of a company’s previousannouncements, newspaper articles, accounting reports, etc.

the University of Sydney, and the Datastream International. The bid-ask spread data atthe closing for relevant trading days were obtained from Research Application ServiceProvider (RASP) of the SIRCA.

As Table 1 indicates, over three-quarters of the ASX inquiries are about share pricechanges. Of 957 replies I exclude the following from further study: (a) In 13 cases, wherethe ASX issued a further query letter because the company had failed to answer some ofthe questions, or had indicated there was ‘no new information’ yet promptly releasedmarket sensitive announcements, only the first announcement is used for the study.However, where companies got multiple inquiries about different matters in the sampleperiod, all of these are included as effective data for analysis. (b) Also excluded are 3cases where the company made a correction or provided new information voluntarily,usually on the next trading day. (c) In addition, another 7 cases are excluded for otherreasons such as the fact that the letters were not posted, or did not indicate the periodof price change, or the query came from the New Zealand Stock Exchange, or the pricedata were not available. Therefore, a sample of 934 is used for further study.

The queries on volume and other matters are relatively less frequent. The sampleincludes 102 cases which account for 8.3% of the total queries. In the period underconsideration, the internet, telecommunications and related sectors went through


1148 GONG

Figure 1Distribution of Market Adjusted Price Changes Prior to ASX Query

4

1210

27

19

7

21

78

128

145

120

96

73

43

37

21

56

23

14

0

20

40

60

80

100

120

140

160

<=–5

0%

[–40

%,–

50%

)

[–30

%,–

40%

)

[–20

%,–

30%

)

[–10

%,–

20%

)

[0,–

10%

)

(0,1

0%)

[10%

,20%

)

[20%

,30%

)

[30%

,40%

)

[40%

,50%

)

[50%

,60%

)

[60%

,70%

)

[70%

,80%

)

[80%

,90%

)

[90%

,100

%)

[100

%,1

50%

)

[150

%,2

00%

)

>=20

0%

Range

Nu

mb

er

dramatic changes around the world. In addition, lots of junior Australian mining andresource companies were in financial distress due to low commodity prices and weakdemand, and many of them offered themselves as ‘shell’ for the so-called ‘backdoor list-ing’ for internet and other technology companies. This fact is documented in Table 2.It is clear that most of the inquiries (74.3%) were sent to media, technology andtelecommunications (MTT) or resources companies during the sample period. Thegrouping of industries follows the code system published by the ASX. The detail of thisis in Appendix B.

Overall, it is found that more queries are sent for cases where share prices go upthan down (91.5% vs. 8.5%), and the percentage of price changes prior to an inquiryis higher for cases where prices increase (mean 56.8%) than for cases where pricesdecline (mean −27.5%). These facts may reflect the generally bullish market situationduring the sample period. In addition, it takes a longer observation period for caseswhere share prices decline (mean 9.7 days) than for the price cases where prices rise(mean 4.3 days).

For the frequency distribution of market adjusted price changes prior to ASX inquiry,see Figure 1. It is found that although the majority of inquiries are for firms having 20%or bigger changes in their share prices, some are made despite relatively small changes.For example, 28 inquiries involve a change of less than a 10%. Further reading ofthe cases indicates that this usually happens when the firm’s price had a prior volatileperiod.

Finally, the bid-ask spread data are collected for 934 inquiries about price changesin the sample period. However, there is no quoted relative spread for eight cases inthe SIRCA data base. In another fourteen cases, the relative spread is quoted as zero.These twenty-two cases are deleted in the calculation. Therefore, the results are basedon a sample size of 912 observations.



Tab

le2

Sum

mar

yof

Mar

ketA

djus

ted

Ret

urn

and

Tra

ding

Day

sPr

ior

toIn

quir

y

Num

ber

ofC

ases

Tra

ding

Day

sPr

ior

toIn

quir

yM

arke

tAdj

uste

dPr

ice

Cha

nge

Num

ber

Perc

enta

geM

ean

Med

ian

Mea

nM

edia

nSt

d.D

ev.

Tot

al93

410

0%

Pan

elA

:Pri

ceIn

crea

se85

591

.54.

33.

00.

570.

440.

52R

esou

rces

436

46.7

3.9

3.0

0.64

0.51

0.54

Med

ia,T

elec

omm

unic

atio

nsan

dTe

chno

logy

213

22.8

4.2

3.0

0.50

0.38

0.54

Hea

lthca

re,B

iote

chno

logy

525.

64.

83.

00.

470.

340.

44Fi

nanc

ial

555.

95.

44.

00.

590.

410.

56In

dust

rial

s44

4.7

4.6

3.0

0.55

0.35

0.50

Oth

ers

555.

95.

73.

00.

380.

300.

26

Pan

elB

:Pri

ceD

ecre

ase

798.

59.

75.

0−

0.28

−0.

240.

15R

esou

rces

313.

36.

55.

0−0

.31

−0.2

80.

11M

edia

,Tel

ecom

mun

icat

ions

and

Tech

nolo

gy13

1.4

6.7

4.0

−0.2

6−0

.19

0.23

Hea

lthca

re,B

iote

chno

logy

40.

413

.59.

5−0

.27

−0.2

60.

17Fi

nanc

ial

80.

913

.512

.5−0

.23

−0.2

20.

08In

dust

rial

s6

0.6

20.5

10.5

−0.3

7−0

.36

0.18

Oth

ers

171.

810

.95.

0−0

.22

−0.1

70.

13


1150 GONG

5. ANALYSIS OF COMPANIES’ RESPONSES TO ASX QUERY LETTERS

In this section, I first study the information content in companies’ replies to the ASX.Then, I focus on the market price changes in response to the release of the companies’reply. In particular I try to assess how far the ASX market monitoring system helps toinform the market and improve liquidity.

(i) Do Companies Release Information in Their Replies?

While all companies state that they follow the ASX listing rules,7 some offer explana-tions or disclose new information in their responses. Some of these answers are quitelengthy and opaque. After reading all the replies, I classify them into three broadcategories for discussion.

Group 1: No New Information Release (Response = 0)

In this category, I have included three types of replies from the companies: (1) Thosethat clearly say no new information without any further elaboration; (2) Those thatprovide some explanation but no new information about the change of the firm’s cashflows or risk characteristics, attributing price changes to previous announcement(s),general market trend such as commodity price fluctuations and market interest inthe industry sectors, media coverage, broker/analyst reports, internet chat-room, ormarket speculations; (3) Those that repeat a previously published forecast or make ageneral statement that changes to the forecast, if any, will be contained in forthcomingreports for the ASX. Based on the Efficient Market Hypothesis, all previously disclosedinformation should be already incorporated into the market price. Therefore, thosestatements are classified as not containing any new information.

Group 2: Partial Information Release (Response = 1)

As mentioned in Section 3, while the ASX requires listed companies to disclosematerial information continuously, it does make some exceptions when the informationconcerns an incomplete proposal or negotiation, or information is a trade secret,or protected by law. Group 2 responses include companies which admit that theyare engaging in some activities which may affect the firm’s investment opportunitiesand then apply the ‘carve-out’ rule in the ASX listing rule 3.1, claiming incompleteand confidential negotiations and discussions, etc., and refusing to disclose anyconcrete information. This classification does not apply in the following circumstanceswhere:

� Such an intention has been previously disclosed to the market. In this case theresponse is classified as containing no new information.

� The response is a general statement of the company’s practice to investigatebusiness opportunities regularly without alluding to any concrete proposal. Sucha response is classified as containing no news.

7 See Question 4 in Query Letter 1 – Share Price in Appendix A.



Table 3Company Responses to ASX Inquiries

Response = 0 Response = 1 Response = 2

Total 659 72 203

Panel A: Price IncreaseNo of cases 601 70 184Percentage 70.3 8.2 21.5

Panel B: Price DecreaseNo of cases 58 2 19Percentage 73.4 2.5 24.1

Notes:We classify the entire sample into six categories, depending on whether the query is issued after anunusual share price increase or decline and how the firms respond. If a firm replies with new information,then it is classified as ‘Response = 2.’ If a firm does not disclose any new information, then it is classified as‘Response = 0.’ Partial information disclosure by a firm is classified as ‘Response = 1.’

� The company discloses the identity of discussions/negotiations/proposals (otherthan previously announced) and then stops giving the full details, citing the‘carve-out’ rule. In this case, the response is classified as providing new infor-mation to the market.8

Group 3: New Information Release (Response = 2)

This category includes responses which: (1) clearly state that a new announcementis to be released, or attached in response to the ASX query, or refer to the fact thata new announcement was made on the same date or immediately after receiving theASX query; (2) confirm variations on profits and abnormal items (either figures orranges) for ASX questions on earnings change (other than clearly stating that such aforecast has been previously released to the market); (3) clarify market speculations andmisinformation; (4) give other new information not previously released to the marketsuch as a detailed project progress report, change of status of contracts (e.g., fromconditional acceptance to unconditional acceptance), or the identity of the partner ina deal, etc.

Table 3 summarizes the responses to the ASX queries from companies. Althoughover 70% of the firms sampled responded to the ASX inquiry with no new information,close to 30% firms, when pushed by the ASX, did release new information to the marketto a certain degree. This simple fact indicates that ASX inquiry does indeed make adifference in prompting information disclosure among listed companies.

(ii) How Does the Market React to Companies’ Responses?

Table 4 summarizes the price performance of companies’ shares at t = 1, 2, 5 tradingdays after the event date. The definition of the event date, which is denoted as ‘t = 0,’needs clarification.

8 Whether the response belongs to partial or full disclosure in this situation is distinguished by whether theidentity of the potential partner/target is disclosed. I believe that, once the identity of the partner/target isidentified, the market can have a much clearer picture about its future cash flows, and investors and analystscould figure out the rest of the details relatively easily.


1152 GONG

Table 4Market Response to a Company’s Announcement of Different Information

Content

Market Adjusted Return

Number of Cases Mean (%) t-statistic Sig. (2-tailed)

Panel A: Price IncreaseResponse = 0 601

t = 0 0.574 1.027 0.305t = 1 0.940 1.118 0.264t = 2 0.670 0.723 0.470t = 5 0.879 0.639 0.523

Response = 1 70t = 0 4.557 2.192 0.032t = 1 6.321 2.508 0.014t = 2 8.636 2.886 0.005t = 5 8.545 2.513 0.014

Response = 2 184t = 0 −0.673 −0.598 0.551t = 1 −0.470 −0.329 0.742t = 2 0.342 0.170 0.865t = 5 −0.428 −0.244 0.807

Panel B: Price DecreaseResponse = 0 58

t = 0 5.182 3.543 0.001t = 1 8.188 3.372 0.001t = 2 8.466 3.341 0.001t = 5 9.194 3.348 0.001

Response = 1 2t = 0 −2.42 −0.518 0.696t = 1 −3.47 −2.339 0.257t = 2 0.366 0.605 0.654t = 5 −1.20 −0.666 0.626

Response = 2 19t = 0 4.324 1.929 0.070t = 1 2.910 1.217 0.239t = 2 2.711 1.231 0.234t = 5 6.447 1.968 0.065

Notes:This table includes the total sample. If the announcements were made before 15:45, t = 0 is thecalendar day of the firm’s announcement. If the announcements were made after 15:45, t = 0 is the nexttrading day.

In Australia, the stock market closes at 4:00pm. Between 4:00 to 4:05pm, limitedorders can be submitted, and the closing auction will be conducted from 4:05 to 4:10pm.The daily closing price reported is the price set in the closing auction. Some companiesrelease information after the market closes. Therefore, it is more reasonable to set theevent date as the next trading day. In addition, it is found that in Australia, there isevidence that news may be incorporated into the market price within 15 to 20 minutes.9

9 For example, Aitken et al. (1998) documents that the market digests the adverse news of short sales within15 minutes or 20 trades in the Australian share market. Also, a popular website: www.tradingroom.com.aucharges users for news within 20 minutes of company announcements.



To this end, I choose 3:45pm as the benchmark to decide which date is the event datewhen I set t = 0. For example, on Feb.16, 2000, firm A’s reply to the ASX query wasposted at 10:35am, while firm B’s reply was posted at 3:59pm. For firm A, Feb.16 is theevent day and ‘t = 0’ refers to this date. However, for firm B, the event day is on Feb.17, 2000, which is denoted as ‘t = 0’. This distinction fits the trading environment.10

The market adjusted net return is defined as:

Pi,t − Pi,−1

Pi,−1− Pm,t − Pm,−1

Pm,−1,

where Pi , t is the price of asset i at date t on the fully-diluted basis, and Pm , t is the levelof All Ordinaries Share Accumulation Index at date t. Again, keep in mind that theevent date is denoted as ‘t = 0.’ This method of calculating the market adjusted netreturn avoids the complication of having a pre-event estimation period for parameterssuch as betas and alphas. The reason for adopting such a restricted model is thatthe parameters estimated from the pre-event period do not reflect the firm’s businessand financial risk characters after the event in our study. As a matter of fact, lots offirms completely changed their business focus shortly after an event. For example,many junior mining companies were renamed and changed to internet companies.Introducing information from the pre-event period may just create noise. Such arestricted model is also used to study the under-pricing of initial public offerings, seeRitter (1990) for an example.

From Panel B in Table 4, it is found that when the ASX makes inquiries in caseswhere the price declines, the market prices of the firms reverse their downward trendand the market-adjusted accumulated abnormal returns are about 5.2 to 9.2% withsignificance at about the 0.1% level in the first five trading days after the firm’s indicatethat they have no new information (response = 0). This essentially means the firmsdeny knowing any bad news. Therefore, it is not difficult to see why the market reactswith optimism.

However, if the companies provide some new information or explain the pricedecline (response = 2), the positive accumulated abnormal returns for the sharesare between 2.7% to 6.4% during the first five trading days after the event. They areless significant as evidenced by the lower t-statistics, when compared with the case of nonew information. As a matter of fact, only the abnormal returns at t = 0 and t = 5 are atmeaningfully significant levels of about 7%. This observation seems to suggest that thenew information provided by the firms’ replies had already been partially incorporatedinto the price before the inquiry. Since there are only two replies which apply the ‘carve-out’ rule and release partial information (response = 1), no conclusions can be drawn.

When the ASX makes inquiries about price increases, the market treats it as if it is a‘non-event’ when the firm either provides no information (response = 0) or releasessome new information (response = 2) in the first five trading days after the reply. But thepositive abnormal returns are significant when companies release partial information(response = 1). On average, the abnormal return goes up by 4.6% at the closing ofthe event day, a total of 6.3% for the first trading day after the companies respond

10 The ASX allows after-hour trading among institutions, and the trades have to be reported to the ASX.However, the reported daily closing price is the one set in the closing auction, so the after hour trading doesnot affect the closing price.


1154 GONG

Table 5Market Response to Company Announcements, Irrespective of the Information

Content

Market Adjusted Return

Number of Cases Mean (%) t-statistic Sig. Level (2-tailed)

Panel A: Price Increase 855t = 0 0.632 1.281 0.201t = 1 1.077 1.541 0.124t = 2 1.251 1.523 0.128t = 5 1.225 1.138 0.256

Panel B: Price Decrease 79t = 0 4.783 3.969 0.000t = 1 6.623 3.501 0.001t = 2 6.877 3.525 0.001t = 5 8.270 3.819 0.000

and a further 2.3% for the second trading day. The positive market responses to thepartial information release are significantly different from zero for the first five tradingdays at less than 5% confidence levels. How do we interpret this finding? It seems thatthe market agrees with the firm’s reply when there is no new news but it discountsnew information released as it is too late. In either case, the market-adjusted abnormalreturns are not significantly different from zero. However, investors collectively treatpartial information disclosure as good news and expect more good news to come.

Unconditional on the contents of the replies, investors can get positive pricemomentum for the first five trading days after the replies to the ASX inquiries are postedfor the cases where prices decline. However, a significantly positive price momentumdoes not appear to be present for the first five trading days of the price increase, asshown in Table 5.

(iii) What Factors May Affect Firms’ Share Price Responses to the ASX Price Inquiries?

From Table 4, it is clear that there are only two cases where the market responses to theASX inquiries are significant: (1) when firms do not disclose any news about the recentprice declines; or (2) when firms disclose information partially about the recent risein their prices. We need to understand what factors may affect the share price changesin response to firms’ replies. In other cases, the abnormal returns are not significantlydifferent from zero. Therefore, it is not interesting to do further econometric analysisto see which factors may affect the share returns.

For this purpose, standard cross-sectional regression analyses are conducted for theabove two cases which have significant abnormal returns after the firms’ reply lettersare posted. The percentage change in the share price on the fully diluted basis is chosenas the dependent variable. Three independent variables are chosen for the regressionanalysis. The first independent variable is the prior price change, which is the rootcause of the ASX inquiries. This variable measures the short term price momentum.For example, does the degree of price decline prior to the reply during the observationperiod specified by the ASX positively or negatively affect the market response at



t = 1? The second factor chosen is the market movement, which is represented bythe movement in the All Ordinaries Accumulation Index. This factor measures theoverall Australian share market return at the corresponding time interval. The thirdfactor is the type of industry. Because in the sample period, the more speculative tradesare concentrated in the resources, technology and biotech industries, it is natural toask the question whether stocks in these industries react to the firms’ replies differently.Dummy variables are used to distinguish different sectors. The regression results aredocumented in Table 6.11

The regression results suggest that the degree of price decline prior to the inquiryis a significant factor (at the 1% level) for all four regressions in the subgroup. Whenfirms deny any bad things happened which could cause the stock price collapses, themarket price rebounds. The larger the degree of price decline prior to the inquiry,the bigger reversal of the price movement is after firms deny the existence of anynegative events. Clearly, the investors are relieved that there is no bad news after all.However, it is puzzling to note that such a price momentum variable does not appearto have a significant impact in the case when firms partially disclose information aboutshare price increases. As shown in Table 6, the coefficients are not significantly differentfrom zero at all conventional confidence levels. Industry factors, which are representedby the dummy variables, are also somewhat significant (two out of six regressions)in determining the degree of the market response to the inquiries. However, it isdifficult to generalize about the significant impact of industry factors on the marketprice response in these situations. The market factor does not significantly affect thepercentage changes in share prices in all regressions.

(iv) Does the Trading Volume Change After Companies Respond to the ASX PriceInquiry?

Trading volume is generally considered an indicator of information asymmetry in themarket place. High trading volume stems, presumably, from either different privateinformation or different interpretations of the same common information amongtraders, see Harris and Raviv (1993). Intuitively, when the market receives a company’sreply, the trading volume should be reduced because the information asymmetryamong traders about the company’s cash flows and its future growth potential isreduced. Trading advantages by informed traders are reduced once the news is madepublic. However, there should still be trading after the public announcement becausethe interpretations of the announcement may very well vary. He and Wang (1995)provide a theoretical model of how trading volume is related to the information flow inthe market. They explain that, in general, informed investors increase their speculativepositions just prior to a public announcement and reduce their positions immediatelyafterwards. Thus, high volume should be observed around such announcements.

11 The adjusted R2 coefficients of these regressions in Table 6 seem to be low, but they are not out of linewith many published studies in the capital market area. For example, Table VI of Bradley, Jordan and Ritter(2003) reports a range of adjusted R2 coefficients between 2.3% to 4.6% in the regression of the market-adjusted return with respect to analyst ratings. Table VI of Aggarwal (2000) reported a range of the adjustedR2 coefficients of between 4.4% and 9.2% in the analysis of after market returns for IPOs. In fact, Roll (1988)found that with hindsight the average adjusted R2 is only about 20% with daily data, which means that only20% of the variation in the daily stock prices can be explained ex post.


1156 GONG

Table 6Regression Results

Regressor Coefficients

Subgroup: price decrease, response = 0Constant −0.0494 −0.0278 −0.0273 −0.0245

(−1.5726) (−1.0187) (−1.1402) (−0.9733)[0.1215] [0.3129] [0.2592] [0.3347]

�Pi −0.3376 −0.3515 −0.3471 −0.3443(−2.7785) (−3.0801) (−2.7982) (−2.8255)

[0.0075] [0.0032] [0.0071] [0.0066]Mi 2.3912 2.0123 2.4849 2.3933

(1.0539) (0.9197) (1.0793) (1.0579)[0.2965] [0.3670] [0.2853] [0.2948]

D1 −0.0553(−2.1571)

[0.0355]

D2 −0.0418(−1.1007)

[0.2759]D3 −0.0411

(−1.0267)[0.3091]

Adjusted R2 0.0976 0.1123 0.0986 0.0971

Subgroup: price increase, response = 1Constant −0.0014 −0.0292 −0.0558 −0.04553

(−0.0398) (−0.8097) (−1.6646) (−1.3256)[0.9684] [0.4210] [0.1007] [0.1896]

�Pi 0.0861 0.0581 0.0738 0.0801(1.1933) (0.6797) (0.9447) (1.0701)[0.2370] [0.4991] [0.3483] [0.2882]

Mi 1.2963 1.4568 1.3765 1.3559(1.4274) (1.4689) (1.3023) (1.3292)[0.1581] [0.1466] [0.1973] [0.1884]

D1 0.0704(1.4607)[0.1489]

D2 0.0702(2.7120)[0.0085]

D3 0.0524(2.1120)[0.0385]

Adjusted R2 0.0266 0.0455 0.0278 0.0189

Notes:The percentage price change (on the fully diluted basis) for a firm’s stock, Ri , between t = −1 andt = 1 in response to the posted reply is regressed against the firm’s prior percentage price change �Piand market return Mi , and a dummy variable indicating the industry sector. Regression results are shownseparately for where there is no dummy; where the dummy includes the resources sector only (D1); wherethe dummy includes the resource sector and the technology sector (D2), and where the dummy includesresources, technology and biotech sectors (D3). The t-statistics, based on Newey-West HAC Standard Errors& Covariance, are given in parentheses (top). The p-values are given in brackets [bottom].



Table 7Trading Volumes Around the Company Responses to the ASX Query

on Share Price

Date Total Average

t = −3 1,416,082,081 1,516,148t = −2 2,309,887,503 2,473,113t = −1 5,037,419,694 5,393,383t = 0 3,990,689,124 4,272,686t = 1 2,475,134,972 2,650,037t = 2 2,346,249,085 2,512,044t = 3 3,114,660,768 3,334,755

This hypothesis is verified in our sample. As we can see from Table 7, the tradingvolumes of the companies queried on share prices in t = −1 and t = 0 are very high,but they are reduced from t = 1 to 3 when compared with that in t = −1 and 0 by aboutone third on the daily basis. Incidentally, the magnitude of this reduction in tradingvolume is similar to an earlier study by Beaver (1968), who found that volume in theweek of an earnings announcement is, on average, about 34% higher than the averagevolume over the corresponding non-report period.

(v) What is the Impact on the Bid-Ask Spread? 12

Liquidity refers to the degree to which a stock can be bought or sold quickly in themarket without affecting the stock’s prevailing price. An investor willing to trade in thestock market may either wait to transact at a favorable price by submitting a limit order,or insist on immediate execution at the currently quoted bid or ask price by enteringa market order. The quoted ask price on the limit order book is for immediate buying,and the bid price is available for immediate sale. Therefore, liquidity can be measuredby the bid-ask spread.

The existing literature on market microstructure suggests that bid-ask spreads aremainly driven by inventory holding costs, order-processing costs, and asymmetricinformation costs. Information asymmetry between uninformed market makers andinformed traders could increase the bid-ask spread because uninformed market makersmay be reluctant to trade in such an adverse environment thereby increasing the bid-askspread to protect themselves.13

With the implementation of the inquiry system, we expect that the degree ofinformation asymmetry between uninformed market makers and informed traderswill be reduced once the firms’ replies are posted. To test this hypothesis, the relativebid-ask spread data at the closing on relevant trading days were collected from SIRCAto match all the replies in response to the ASX price inquiries.

The relative spread is the relation between the spread (best ask - best bid) and themid-point of the spread. The result is a percentage. For example, if the best ask is $1.35

12 I thank Professor Martin Walker, the editor, for the suggestion of this analysis.13 For the literature survey on the determination of the bid-ask spread see O’Hara (1995) among others.Glosten (1994) discusses the equilibrium bids and offers in an open limit-order driven market which fits theASX trading environment well. The bid-ask spread in an order driven market is set by ‘patient’ limit-ordertraders, who trade against the market orders submitted by possibly informed investors.


1158 GONG

and best bid is $1.25, then the bid-ask spread is $0.10. Because the mid-point is $1.30,the relative spread will be 0.10/1.30 = 0.07692 = 7.69%.

Table 8 lists the results. t = 0 denotes the event date. For the entire sample of theinquiries related to price changes, the mean relative bid-ask spread increases first fromt = −3 to t = −1, and then reduces in the subsequent trading days. Perhaps the moreinteresting number to check is the relative spread one day before the event and afew days after the event. In Panel A, we see that the mean relative spread has beenreduced from 6.54% at t = −1 to 5.23% at t = 3, a 20% reduction. Also, there is abigger reduction in the mean relative spread for cases where the inquiries directed toshare price increases than for cases of share price decreases from t = −1 to t = +3. InPanels B and C, more detailed calculations are presented, corresponding to whetherfirms respond with new, partial, or no information disclosure. The reduction in themean bid-ask spread is the greatest at 21.8%, from 6.18% at t = −1 to 4.83% at t =3, in cases where the firms reply with no new information when queried about shareprice increase. However, when firms provide new information when queried upon pricedecline, the mean relative spread will decline from t = −1 to t = 1, and then rise slightlyin subsequent two trading days.

To check whether the means of the bid-ask spread on two different trading days arestatistically different from each other, we follow the standard procedure to conduct thet-test. Let m be the means, s the standard deviations, and n be the sample size. Thet-statistic is defined as:

t = mi − m j√(s2

i + s2j

)/n

,

where the subscript i represents the trading date.The results for the paired t-tests are contained in Table 9, which documents the

mean relative spread at t = −1 and the mean relative spread at t = 0, 1, 2, and 3,respectively. All four pairs are significantly different from each other at the significancelevel of 0.03 or less for the entire sample. In Panels B and C of Table 9, more t-tests areconducted for six sub-groups, depending on whether share price prior to the inquiryincreases or decreases and whether the reply is classified as containing new, partial orno information. In general, we see that the differences in the mean relative spreadsare significant for inquiries of share price increase than for share price decrease. Incases where the inquiries issued after a share price decline, the reduction in the meanspread is only significant at 1.7% for the pair t = −1 vs. t = 2, and 7% for the pair t =−1 and t = +3.

In summary, querying companies when there is an unusual movement of the shareprice and posting their replies on average can reduce the relative bid-ask spread byabout 20%. This of course will increase the liquidity of the shares under the study.Furthermore, such a reduction is more prominent for cases where the queries areissued after share price increases and when companies respond with either no or newinformation disclosure.

(vi) Does the ASX Follow Up With Further Inquiry?

The ASX rarely follows up inquiries after it has received the firms’ replies. Onlyin five cases, did companies get a further letter from the ASX within the next few



Table 8Relative Bid-ask Spreads Around the Company Responses to the ASX Query on

Share Price

Number of Cases Relative Spread

Max. Min. Mean St.Dev

Panel A: SummaryPrice Increase 835

t = −3 66.67 0.23 6.0538 7.5162t = −2 131.64 0.16 6.2156 8.5403t = −1 122.19 0.20 6.2666 8.4438t = 0 94.18 0.22 5.6214 6.6282t = 1 95.97 0.28 5.1339 6.3661t = 2 85.64 0.20 5.1227 6.5260t = 3 76.74 0.21 4.9739 5.7662

Price Decrease 77t = −3 66.67 0.36 7.5342 11.5664t = −2 79.38 0.53 8.3987 13.8140t = −1 86.32 0.59 9.4894 15.0131t = 0 78.26 0.41 9.9232 14.7465t = 1 78.26 0.41 8.7918 14.3535t = 2 69.73 0.52 7.9665 12.7477t = 3 66.67 0.58 8.0463 11.5683

Total 912t = −3 66.67 0.23 6.1788 7.9403t = −2 131.64 0.16 6.3999 9.1137t = −1 122.19 0.20 6.5387 9.2130t = 0 94.18 0.22 5.9850 7.7337t = 1 95.97 0.28 5.4427 7.4380t = 2 85.64 0.20 5.3628 7.2919t = 3 76.74 0.21 5.2333 6.5064

Panel B: Price IncreaseResponse = 0 592

t = −3 66.67 0.23 6.0618 7.8848t = −2 131.64 0.16 6.2417 9.1202t = −1 122.19 0.20 6.1750 8.3800t = 0 36.45 0.22 5.3044 5.0110t = 1 53.51 0.28 4.8890 4.8061t = 2 61.70 0.20 4.9681 5.6150t = 3 52.21 0.21 4.8271 4.8807

Response = 1 68t = −3 32.56 0.62 5.7861 5.6039t = −2 39.11 0.88 5.8140 6.6469t = −1 42.23 1.03 5.7828 6.9819t = 0 32.38 0.88 5.6430 5.6729t = 1 27.85 0.56 5.1587 5.4026t = 2 50.05 0.56 5.2247 6.6972t = 3 31.68 0.55 5.0211 5.7684


1160 GONG

Table 8 (Continued)


Max. Min. Mean St.Dev


Panel C: Price DecreaseResponse = 0 56

t = −3 55.45 0.42 7.5910 10.2536t = −2 79.38 0.53 8.9146 13.8051t = −1 86.32 0.59 10.1960 15.4792t = 0 78.26 0.58 10.8720 15.3808t = 1 78.26 0.76 9.5827 14.4901t = 2 69.73 0.52 8.1141 12.1588t = 3 54.89 1.05 8.2328 10.5117

Response = 1 2t = −3 1.59 1.55 1.5686 .0238t = −2 2.03 2.03 2.0301 .0039t = −1 5.77 3.49 4.6330 1.6117t = 0 11.80 3.02 7.4098 6.2061t = 1 5.34 2.89 4.1168 1.7335t = 2 3.16 2.50 2.8324 .4680t = 3 3.39 1.95 2.6677 1.0205


Notes:There are 934 inquiries about price changes in the sample period. However, there is no quotedrelative spread for eight cases in the SIRCA data base. In another fourteen cases, the relative spread isquoted as zero. These twenty-two cases are not used in the calculation. The following results are based on asample size of 912 observations. t = 0 denotes the event date. The numbers are in percentages. For example,the average relative bid-ask spread for the entire sample at = −3 is 6.1788%, at t = −2 is 6.3999%, etc. PanelA lists the summary statistics for the whole sample, as well as two sub-categories distinguishing whether theinquiries are about price increase or price decrease. Panels B and C give further detailed information aboutthe change in the relative spread when firms respond with either new, partial, or no information disclosure.

trading days because they failed to answer some of the specific questions. In anothereight cases, the ASX contacted the companies again when the share price of the firmsinvolved continued to fluctuate out of normal trading range, or the firms disclosedno information in response to the first query letter, but made a market sensitiveannouncement shortly after.



Table 9Paired t-tests Between the Mean Relative Spread at t = −1 and the Mean Spread

at t = 0, 1, 2 and 3, Respectively


Mean t-statistic Sig. Level(Differences) (2-tailed)

Panel A: SummaryPrice Increase 835

t = −1 vs t = 0 0.6525 3.207 0.001t = −1 vs t = 1 1.1327 4.687 0.000t = −1 vs t = 2 1.1439 4.622 0.000t = −1 vs t = 3 1.2927 5.307 0.000

Price Decrease 77t = −1 vs t = 0 −0.4338 −0.928 0.356t = −1 vs t = 1 0.6977 1.349 0.181t = −1 vs t = 2 1.5229 2.264 0.026t = −1 vs t = 3 1.4431 1.768 0.081

Total 912t = −1 vs t = 0 0.5607 2.941 0.003t = −1 vs t = 1 1.0960 4.860 0.000t = −1 vs t = 2 1.1759 5.035 0.000t = −1 vs t = 3 1.3054 5.595 0.000

Panel B: Price IncreaseResponse = 0 592

t = −1 vs t = 0 0.8806 3.397 0.001t = −1 vs t = 1 1.2860 4.284 0.000t = −1 vs t = 2 1.2069 3.924 0.000t = −1 vs t = 3 1.3479 4.507 0.000

Response = 1 68t = −1 vs t = 0 0.1398 0.462 0.646t = −1 vs t = 1 0.6242 0.877 0.384t = −1 vs t = 2 0.5581 0.630 0.531t = −1 vs t = 3 0.7618 0.900 0.371

Response = 2 175t = −1 vs t = 0 0.0811 0.204 0.838t = −1 vs t = 1 0.8118 1.720 0.087t = −1 vs t = 2 1.1584 2.620 0.010t = −1 vs t = 3 1.3121 2.787 0.006

Panel C: Price DecreaseResponse = 0 56

t = −1 vs t = 0 −0.6760 −1.100 0.276t = −1 vs t = 1 0.6133 0.895 0.375t = −1 vs t = 2 2.0819 2.460 0.017t = −1 vs t = 3 1.9631 1.846 0.070

Response = 1 2t = −1 vs t = 0 −2.7768 −0.855 0.550t = −1 vs t = 1 0.5163 5.996 0.105t = −1 vs t = 2 1.8006 2.227 0.269t = −1 vs t = 3 1.9653 4.702 0.133


1162 GONG

Table 9 (Continued)

Number of cases Relative Spread

(Differences) Mean t-statistic Sig. Level (2-tailed)

Response = 2 19t = −1 vs t = 0 0.5268 1.217 0.239t = −1 vs t = 1 0.9655 1.653 0.116t = −1 vs t = 2 −0.1538 −0.148 0.884t = −1 vs t = 3 −0.1446 −0.143 0.888

Notes:Panel A lists the summary statistics for the paired t-stat for the whole sample, as well as two sub-categories distinguishing whether the inquiries are about price increase or price decrease. Panels B andC give further detailed information about the t-stats when firms respond with either new, partial, or noinformation disclosure.

However, I do find that in the sample period, some companies received multiplequery letters. For example, one small technology firm got as many as 11 within a periodof 16 months, all but the first one were inquiries for price jumps.

(vii) How Often Do Companies Voluntarily Halt Their Trading?

In the ASX query letters, the firm is reminded of the option of requesting a tradinghalt if it is aware of some new information but an announcement cannot be madeimmediately; see Appendix A Query Letter 1.14 Firms seldom used this option. Thereare only eight cases in the sample where companies requested trading halts. All ofthese occurred after the firms experienced a price jump before the inquiry. The priceperformances of these eight companies after the market was reopened are listed inTable 10. It is clear that the distribution of the market adjusted absolute returns is quitedispersed and the t-test shows that the post-announcement market adjusted abnormalreturns are not significantly different from zero. It is difficult to draw any meaningfulconclusion from this small sample.

6. CONCLUSIONS

In this paper, I have examined the market monitoring program by the Australian StockExchange (ASX) from July 1998 to June 2000. During this time the economy, par-ticularly the internet and telecommunications sector, experienced dramatic changes.Prices and trading volumes were volatile. Under these circumstances, the effectivenessand market responses to the ASX inquiries, which were triggered by unusual tradingpatterns in prices, are documented. About 30% of the companies queried did releasesome new information to the market. After the replies were made public, tradingvolumes and bid-ask spreads were reduced, share prices stabilized in most cases withthe following two exceptions: (1) The price continued to rally on average if the companyreleased only partial information when questioned after a significant price jump; (2)The downward price trend reversed if the company stated that no new information

14 For a description of trading halts in the Australian Stock Exchange, see Aitken et. al, (1999).



Tab

le10

Pric

ePe

rfor

man

ceof

Com

pani

esC

allin

gfo

rT

radi

ngH

alts

inR

espo

nse

toth

eA

SXIn

quir

y

12

34

56

78

Ave

rage

t-sta

tistic

Sig.

(2-ta

iled)

Eff

ectiv

epr

ice

chan

ge45

.3%

15.2

%48

.6%

50.6

%25

.0%

71.0

%10

0.4%

69.4

%53

.2%

prio

rto

inqu

iry

Post

anno

unce

men

tmar

ket

adju

sted

retu

rnt=

1−4

.72

−0.1

610

.73

0.12

6.21

16.4

23.

0040

.60

9.03

1.84

50.

108

t=

2−1

1.42

−0.5

021

.91

−12.

32−0

.96

5.17

22.5

730

.11

6.82

1.21

80.

263

t=

5−5

.63

−0.3

116

.46

−18.

1628

.93

5.93

51.5

99.

3811

.02

1.40

10.

204


1164 GONG

could explain the decline. Furthermore, there existed statistically significant, positiveabnormal returns for the first five trading days, which did not seem to be related to thecontent of the replies about price declines.

There is a number of remaining issues. If one of the purposes of the marketsurveillance is to prevent insider trading, then monitoring the derivative prices is alsonecessary. For example, a large buy order for a deep ‘out-of-money’ put option maysuggest that someone expects the underlying share will collapse. Monitoring derivativeprices may help the exchange to deter insider trading. The ASX perhaps should alsotake this into consideration. Another potentially interesting issue would be to matchthe insider trading record with the price/volume movements. Although this is oneof the goals of the ASX’s market surveillance program, it is not clear whether this isactually done in practice. These issues deserve further research.

APPENDIX A

Sample ASX Query Letters

Two actual copies of ASX query letters which were used during the sample period areprinted here. The first one is a letter inquiring about unexpected share price movement,while the second one is an inquiry about volume changes. Phone number digits andthe listing officer’s name are changed. The company’s name is also omitted.

A. Query Letter 1 – Share Price

We have noted a change in the price of the Company’s securities from 64 centson 11 February 2000 to a high of 85 cents today.

In light of the price change, please respond to each of the following questions:

1. Is the Company aware of any information concerning it that has not beenannounced which, if known, could be an explanation for recent trading in thesecurities of the Company?

2. If the answer to question 1 is yes, can an announcement be made immediately?If not, why not and when is it expected that an announcement will be made?

Please note, if the answer to question 1 is yes and an announcement cannot bemade immediately, you need to contact us to discuss this and you need to considera trading halt (see below).

3. Is there any other explanation that the Company may have for the price changein the securities of the Company?

4. Please confirm that the Company is in compliance with the listing rules and,in particular, listing rule 3.1.

Your response should be sent to me on facsimile number (XX) XXXX-4114. Itshould not be sent to the Company Announcements Office.

Unless the information is required immediately under listing rule 3.1, a responseis requested as soon as possible and, in any event, not later than half an hour beforethe start of trading (i.e. before 9.30 a.m. E.D.S.T.) on Tuesday, 15 February 2000.



The response must be in a form suitable for release to the market. If you haveany concern about release of a response, please contact me immediately.

LISTING RULE 3.1

Listing rule 3.1 requires an entity to give ASX immediately any informationconcerning it that a reasonable person would expect to have a material effect onthe price or value of the entity’s securities. The exceptions to this requirementare set out in the rule.

In responding to this letter you should consult listing rule 3.1 and the guidancenote titled ‘Continuous disclosure: listing rule 3.1’.

If the information requested by this letter is information required to be givento ASX under Listing rule 3.1 your obligation is to disclose the informationimmediately.

Your responsibility under listing rule 3.1 is not confined to, or necessarily satisfiedby, answering the questions set out in this letter.

TRADING HALT

If you are unable to respond by the time requested, or if the answer to question 1is yes and an announcement cannot be made immediately, you should considera request for a trading halt in the Company’s securities. As set out in listing rule17.1 and the guidance note titled ‘Trading halts’ we may grant a trading halt atyour request. We may require the request to be in writing. We are not requiredto act on your request. You must tell us each of the following:∗ The reasons for the trading halt.∗ How long you want the trading halt to last.∗ The event you expect to happen that will end the trading halt.∗ That you are not aware of any reason why the trading halt should not be granted.∗ Any other information necessary to inform the market about the trading halt,

or that we ask for.

The trading halt cannot extend past the commencement of normal trading on thesecond day after the day on which it is granted. If a trading halt is requested andgranted and you are still unable to reply to this letter before the commencementof trading, suspension from quotation would normally be imposed by us from thecommencement of trading if not previously requested by you. The same applies ifyou have requested a trading halt because you are unable to release informationto the market, and are still unable to do so before the commencement of trading.

If you have any queries regarding any of the above, please do not hesitate tocontact the undersigned on XXXX 0011.

XYZLISTINGS OFFICER

B. Query Letter 2 – Volume

Australian Stock Exchange Limited (‘ASX’) has noticed the recent increase inthe volume of trading in the Company’s shares.


1166 GONG

In the interests of ensuring a fully informed market, ASX requests that theCompany provide an updated list of the names of the twenty largest shareholders,the number of securities each holds and the percentage of capital each holds.

This information should be up to date as at 5pm WST Wednesday 14 June 2000.Please direct your response, in a format suitable for release to the market, tofacsimile (XX) XXXX 2020 by 5pm WST Friday 16 June 2000.If you have any questions, please contact the undersigned on XXXX 0011.

XYZLISTINGS OFFICER

APPENDIX BGrouping of Industry Sector

The following lists the detailed industry grouping used in Table 2. The number in thebracket indicates the industry code assigned by the ASX. For example, coal (45) meansthat the ASX assigns code 45 for coal producers.

Group Industry Sector Number

1 All Resources 467

Base Metals (22)Coal (45)Diamonds (25)Gold Explorer (12)Gold Producer (11)Gold, Copper (15)Gold, Other Mining (13)Mineral Sands (23)Mining Explorer (27)Mining Investment (28)Mining Producer (26)Mining Services (223)Oil or Gas Explorer (42)

2 Media, Telecommunications and Technology 226

Advertising, Marketing (155)Computer and Office Services (226)Diversified Media (151)Equipment, Services (183)High Technology (228)Network Operator (181)Other Telecommunications (184)Television (153)

3 Healthcare, Biotechnology 56

Biotechnology (212)Health and Related Products (215)Health/Medical Services (214)Hospital Management (213)Pharmaceutical (211)



4 Financial 63

Developer, Finance (65)Entrepreneurial Investors (227)Equity Investor (192)Insurance Company (171)Miscellaneous Financial Services (195)Property Investor (193)Property Trust (201)

5 Industrials 50

Diversified Industrial (231)Electricity, Gas (52)Engineering Contractor (114)Fertilisers (102)Light Engineering (115)Miscellaneous Industrials (221)Other Infrastructure (54)Timber and Board (74)

6 Others 72

Brewer (81)Casinos or Gaming (241)Food (91)Leisure Activities (243)Manufacturer, Retailer (133)Miscellaneous Services (222)Retail (131)Soft Drink or Confectionery (94)Vintner (82)Wholesaler, Retail (132)

Total (Market Sensitive) 934

REFERENCES

Aggarwal, R. (2000), ‘Stabilization Activities by Underwriters after New Offerings,’ Journal ofFinance, Vol. 55, pp. 1075–103.

Aitken, M., A. Frino and R. Winn (1999), ‘The Effect of Information Motivated Trading Haltson Price Discovery: Evidence from the Australian Stock Exchange’, Unpublished WorkingPaper (University of Sydney).

——— ———, M. McCorry and P. Swan (1998), ‘Short Sales are Almost InstantaneouslyBad News: Evidence from the Australian Stock Exchange’, Journal of Finance, Vol. 53, pp.2205–23.

Amihud, Y. and H. Mendelson (1986), ‘Asset Pricing and the Bid-Ask Spread’, Journal ofFinancial Economics, Vol. 17, pp. 223–49.

Arlen, J. and W. Carney (1992), ‘Vicarious Liability for Fraud on Securities Markets: Theoryand Evidence’, University of Illinois Law Review, Vol. 1992, pp. 691–740.

The Australian Stock Exchange (2000), An Overview of the Australian Equities Market (mimeo).Beaver, W. (1968), ‘The Information Content of Annual Earnings Announcements’, Journal

of Accounting Research, Vol. 6 (supplement), pp. 67–92.Bhattacharya, U. and H. Daouk (2002), ‘The World Price of Insider Trading’, Journal of

Finance, Vol. 57, pp. 75–108.


1168 GONG

Botoson, C. (1997), ‘Disclosure Level and the Cost of Equity Capital’, The Accounting Review,Vol. 72, pp. 323–49.

Bradley, D., B. Jordan and J. Ritter (2003), ‘The Quiet Period Goes Out with a Bang’, Journalof Finance, Vol. 43, pp. 1–36.

Carter, M., S. Mansi and D. Reeb (2003), ‘Evidence on Quasi-Private Information and InsiderTrading’, Financial Analysts Journal, Vol. 59 (May/June), pp. 60–68.

DeBont, W. and R. Thaler (1995), ‘Financial Decision-Making in Markets and Firms: ABehavioral Perspective’, in R. Jarrow, V. Maksimovic and W. Ziemba (eds.), Handbooksin Operations Research and Management Science, Vol 9: Finance (Elsevier Press), pp. 385–410.

Diamond, D. W. (1985), ‘Optimal Release of Information by Firms’, Journal of Finance, Vol.40, pp. 1071–94.

——— and R. E. Verrecchia (1982), ‘Optimal Managerial Contracts and Equilibrium SecurityPrices’, Journal of Finance, Vol. 37, 275–87.

——— ——— (1991), ‘Disclosure, Liquidity, and the Cost of Capital’, Journal of Finance, Vol.46, pp. 1325–59.

Donnan, J. (2000), ‘Class Action in Securities Fraud in Australia’, Company and Securities LawJournal, Vol. 18, pp. 82–96.

Fischel, F. and S. Grossman (1984), ‘Customer Protection in Futures and Securities Markets’,Journal of Futures Markets, Vol. 4, pp. 273–95.

Fox, M. B. (1997), ‘Rethinking Disclosure Liability in the Modern Era’, Washington UniversityLaw Quarterly, Vol. 75, pp. 903–18.

Gertner, R., R. Gibbons and D. Scharfstein (1988), ‘Simultaneous Signaling to the Capital andProduct Markets’, RAND Journal of Economics, Vol. 19, pp. 173–90.

Glosten, L. R. (1994), ‘Is the Electronic Open Limit Order Book Inevitable?’ Journal of Finance,Vol. 49, pp. 1127–61.

Grundy, B. and M. McNicolls (1989), ‘Trade and Revelation of Information through Pricesand Direct Disclosure’, Review of Financial Studies, Vol. 2, pp. 495–526.

Harris, M. and A. Raviv (1993), ‘Differences of Opinion Make a Horse Race’, Review of FinancialStudies, Vol. 6, pp. 297–355.

He, H. and J. Wang (1995), ‘Differential Information and Dynamic Behavior of Stock TradingVolume’, Review of Financial Studies, Vol. 8, pp. 919–72.

Huddart, S., J. Hughes and M. Brunnermeirer (1999), ‘Disclosure Requirements and StockExchange in an International Context’, Journal of Accounting and Economics, Vol. 26,pp. 237–69.

Lowenstein, L. (1996), ‘Financial Transparency and Corporate Governance: You Manage WhatYou Measure’, Columbia Law Review, Vol. 96, pp. 1335–62.

Mahoney, P. (1997), ‘The Exchange as Monitor’, Virginia Law Review, Vol. 83, pp. 1453–71.O’Hara, M. (1995), Market Microstructure Theory (Blackwell Publishers Inc).Pirrong, C. (1999), ‘The Organization of Financial Exchange Markets: Theory and Evidence’,

Journal of Financial Markets, Vol. 4, pp. 329–57.——— (2000), ‘A Theory of Financial Exchange Organization’, Journal of Law and Economics,

Vol. 43, pp. 437–72.Pritchard, A. C. (1999), ‘Markets as Monitors: A Proposal to Replace Class Actions with

Exchanges as Securities Fraud Enforcers’, Virginia Law Review, Vol. 85, pp. 925–1020.Roll, R. (1988), ‘R2’, Journal of Finance, Vol. 43, pp. 541–66.Romer, D. (1993), ‘Rational Asset-Price Movements Without News’, American Economic Review,

Vol. 83, pp. 1112–30.


effectiveness and market reaction to the stock exchange's inquiry in australia

Documents