Disclosure regulation and the profitability of insider trading: Evidence from New Zealand

Pacific-Basin Finance Journal 12 (2004) 479 – 502 www.elsevier.com/locate/econbase

Disclosure regulation and the profitability of insider trading: Evidence from New Zealand Ahmad Etebari a,*, Alireza Tourani-Rad b, Aaron Gilbert b b

a Whittemore School, University of New Hampshire, Durham, NH 03824, USA Auckland University of Technology, Private Bag 92006, 1020 Auckland, New Zealand

Received 20 May 2003; accepted 6 September 2003 Available online 6 February 2004

Abstract This paper provides evidence on insider trading in New Zealand by examining transactions disclosed by corporate insiders for a sample of 93 listed companies over the 1995 – 2001 period. These transactions include two types of disclosures: immediate disclosures, as represented by substantial shareholder (SSH) notices, and delayed disclosures, as reported in annual reports. The results (2453 transactions) show that insiders earn significantly large abnormal returns on their transactions, with the gains coming largely from transactions involving delayed disclosure. In contrast, transactions involving immediate disclosure earn insignificant returns. The results also show that the size of the company, membership in a major stock index, the position of the insider and the percentage of the insider’s holdings being traded all affect the size of the abnormal returns. These findings lend strong support to amendments to securities laws that require continuous disclosure for all transactions. D 2003 Elsevier B.V. All rights reserved. JEL classification: G18; G14; K22; N27 Keywords: Insider trading; Disclosure; Securities regulation

1. Introduction Insider trading can be defined as trading by people with information superior to that possessed by the market. Typically, insiders are company directors, executives and large shareholders. In most countries, insiders are required to disclose their trading to the appropriate government agencies, such as the Securities and Exchange Commission in * Corresponding author. Tel.: +1-603-862-3359; fax: +1-603-862-3383. E-mail addresses: [email protected] (A. Etebari), [email protected] (A. Tourani-Rad), [email protected] (A. Gilbert). 0927-538X/$ - see front matter D 2003 Elsevier B.V. All rights reserved. doi:10.1016/j.pacfin.2003.09.005

480

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

the U.S., within a reasonably short period of time following the date of the transaction. The details of insiders’ transactions are then made available to the public through various media. This information receives a great deal of attention by investors and the press due to the perception that insider trades are driven by information asymmetry. Other parties privy to information include relatives of corporate insiders and constructive insiders such as lawyers and accountants of the company who acquire their information due to their contractual relationships with the company. Transactions by these parties are less widely followed by the market because of greater difficulty in obtaining information about the trades. The profitability of insider trading has been the subject of numerous studies. In general, these studies show that insiders are better informed about their companies’ prospects and that they trade profitably based on this information. For example, Lorie and Niederhoffer (1968), Jaffe (1974), Finnerty (1976) and Seyhun (1992), among others, show that insiders in the U.S. make abnormally large returns from their trades. Pope et al. (1990) and Friederich et al. (2002) examine insider trading in the U.K. market and report similarly large abnormal results. These studies report consistently significant results for purchases and conclude that purchases are more informative than sales. While there are a number of reasons to sell a stock, including liquidity needs and portfolio rebalancing, the main reason for buying stocks is profit (Jeng et al., 1999; Carpenter and Remmers, 2001; Friederich et al., 2002). There is also evidence showing that insiders can successfully time their transactions. That is, insiders buy shares after a decline in the price that is followed by price increases and sell shares after a price run-up, which then reverses and results in negative returns (Lakonishok and Lee, 2001). However, there is mixed evidence whether or not outsiders can profitably mimic insider transactions. Seyhun (1986) and Rozeff and Zaman (1988) found that outsiders following insider trades could not earn abnormal returns. In contrast, examining a sample of trades with a relatively shorter delay between trade and disclosure, Bettis and Vickey (1997) found that both insiders and outsiders could make significant abnormal profits based on purchases and sales over both short and long holding periods. Bettis and Vickey (1997) showed that, even after adjusting for transaction costs, outsiders could still make abnormal gains by simulating insider transactions. Despite the weight of evidence that insiders can earn abnormal returns, a small number of studies have found evidence to the contrary. Rozeff and Zaman (1988) found that, once transaction costs of 2% were taken into account, insiders in the U.S. market could not earn significant returns. For the Oslo Stock Exchange, Eckbo and Smith (1998) failed to find abnormal returns due to insider trading. Surprisingly, this study showed that insiders as a group underperformed professionally managed funds. Insider trading has been the subject of very little research in New Zealand. Studies of insider trading are restricted to Etebari and Duncan (1997) and Duncan and Etebari (1990), who studied insider trading in the years 1986 and 1993. Both studies analyzed price runups prior to corporate announcements and reported evidence of potential insider trading in New Zealand. Casey and Tourani-Rad (2001) examined data on directors’ trades disclosed in annual reports for the period 1993 – 1999. They showed that insiders could obtain 15.63% abnormal returns when they purchased but suffered a loss of 11.75% when they sold shares. This finding partly supports the evidence from other markets, with positive returns to purchases. However, the treatment of sales and the size of the purchaser’s profits are vastly different to those recorded in these markets.

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

481

Insider trading is now regulated in 85% of countries with stock markets due to its perceived harmful effects, most notably increased bid –ask spreads (Kyle, 1985; Chung and Charoenwong, 1998), reduced market liquidity (Friederich et al., 2002; Ausbel, 1990; Repullo, 1999) and reduced investor confidence in the market (Ausbel, 1990; Bhattacharya and Daouk, 2002; Leland, 1992). To mitigate these costs and curb the unfair advantage of those with access to private information, the laws typically call for insiders to disclose their trades in a timely fashion. (For example, the U.S. laws currently require the reporting of all insider trades within two days of transaction.) Public disclosure of insider trading enables the market to accelerate the price discovery process and reduce abnormal profits to such trades; it can also enhance market efficiency (Huddart et al., 2001). As stated by Carlton and Fischel (1983), ‘‘the greater the ability of market participants to identify insider trading, the more information such trading conveys.’’ Until recently, disclosure laws in New Zealand regarding insider trading were vastly different from those of other developed markets. Under the Securities Market Act 1988, only substantial shareholders (SSH), those shareholders with over 5% of the company’s shares, were required to disclose their trading in a timely fashion and even then only when their holdings changed by a cumulative 1% since the previous disclosure. As for directors and executives, only company directors were required to disclose their trading in their annual report, while executives who were non-board members were not required to disclose their trading at all. This meant that only large shareholders disclosed trading in a timely fashion, within 5 days of the trade, while ordinary directors disclosed, on average, 9 –10 months after the transaction. However, this changed on 1 December 2002, when the Securities Market Amendment Act 2002 came into effect. This act now requires all corporate insiders—i.e., large shareholders, directors and executives—to disclose the details of their trading to the market within 5 working days of the transaction. Given the weak enforcement of regulations in New Zealand, it would be interesting to examine the abnormal gains resulting from insider trading. Furthermore, the two-tiered disclosure regime in New Zealand provides a unique opportunity to examine the effect of timely disclosure on insider trading profits. In this paper, we analyze the effect of information asymmetry on insider trading profits in New Zealand. We use size, index membership and the insider’s position within the company as proxies for information asymmetry, as well as the size of the trade relative to the insider’s holdings, to explain any abnormal returns. We would expect that by delaying disclosure of their trades insiders would be able to earn significantly greater profits. Our sample consisted of 2453 trades including 1514 trades by directors and 939 trades by large block holders for 93 listed firms over the January 1995 to December 2001 period. The results show that insiders were active and that, over the study period, there was at least some activity in more than 71% of the firms. Over the 250 trading days following the trades, insider purchases (sales) were on the average associated with an abnormal gain of + 6.5% (
5.99%). For delayed disclosures, the average abnormal gain for purchases (sales) was + 8.92% (
7.70%); both of these results were statistically significant. For immediate disclosures by major shareholders, purchases and sales had market responses of + 4.12% and
4.49% respectively, but neither of these was statistically significant. The results of testing of transaction characteristics for the delayed sample show that small companies earned significantly higher returns than large companies, less-researched

482

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

companies outperformed those listed in NZSE 10—an index of the top 10 listed companies, and insiders with more information, such as Managing Directors and Chairmen, outperformed other classes of insiders. The results also show that small transactions in terms of percentage of the insider’s holdings generated higher returns than large transactions, with the exception of full sales which outperformed partial sales largely due to the information such a sale provides. The next section describes the data and the methodology. Results are presented in Section 3, and Section 4 gives a summary and concludes the paper.

2. Data and methodology 2.1. Data The study period ran from January 1995 to December 2001. We collected information on all companies that were listed on the New Zealand Stock Exchange (NZSE) during this period. Companies had to have share price data available for 250 trading days prior to and after each event (i.e., insider transaction) and had to have at least one substantial shareholder and annual report disclosure during the period. Any individual transactions lacking the required 250 days price information were excluded. The overall sample consisted of 1450 buy and 1003 sell transactions covering 93 companies across two sets (subsamples) of disclosures: immediate disclosures and delayed disclosures. The delayed disclosures subsample included the directors’ share dealings as disclosed in their respective company annual reports. Disclosure of this information is required by statute and is therefore available for all companies. The immediate disclosures subsample included trades for which SSH disclosure notices were reported to the public for the period 1995– 2001. The SSH notices are required to be presented both to the public issuer of the shares they relate to and to the stock exchange that publishes them. These announcements were obtained directly from the NZSE. The subsample in this case was limited to existing substantial shareholders who had changed their holdings sufficiently to require disclosure. Transactions that were disclosed in both the annual reports and in a SSH notice were removed from the delayed disclosures subsample. They were retained in the immediate disclosures subsample on the grounds that these disclosures would provide timely information to the market. These criteria resulted in an immediate disclosures subsample of 939 transactions (597 buys and 342 sells) and a delayed disclosures subsample of 1514 transactions (853 buys and 661 sells). The two subsamples differed drastically in terms of the delay in disclosure of transactions. SSH notices must be filed within 5 days of trading and the majority of them are indeed disclosed within this window. However, there are substantial delays in disclosures of directors’ trades in annual reports. Fig. 1 shows the distribution of the delay in directors’ trades, where the delay is measured by the difference between the date of each transaction, as listed in the annual report, and the financial year-end of the respective company. The disclosures have a mean

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

483

Fig. 1. Days between directors’ transactions and financial year end.

delay of 181 days for purchases and 175 days for sales, or roughly 6 months, in reporting directors’ transactions. Considering that it typically takes an additional 3 or 4 months before the annual report is published, the average delay in disclosure of directors’ trades is closer to nine months. Fig. 1 also shows that insider trading activity is lowest around the release dates of financial year-end results and half-yearly results. (In Fig. 1, the release of half-yearly results corresponds with days 150– 210.) In New Zealand, the Insider Trading (Approved Procedure for Company Officers) Notice 1996 and its 1993 predecessor of the same name require insiders to avoid trading in the month prior to a half-yearly report and the 2 months prior to a full-year report. In Fig. 1, there is evidence of decreased trading prior to the reporting dates, but there is also evidence that trading does not cease altogether, suggesting that the regulations intended to restrict such trading might not have been fully enforced. The decrease in trading following the reporting of the financial results is likely to be due to the fact that information asymmetry following the reporting dates would be at its lowest level, offering insiders fewer opportunities to make profitable trades. Share price data for the study was obtained from the Datastream and Datex databases. 2.2. Measuring insider trading Studies of insider trading often examine net insider sales, defined as open market or private sales minus purchases. This definition is used on the grounds that insiders could, for instance, act on positive information by not only purchasing shares but also by delaying sales until the information is public. Net sales reflect the effects of either or both actions. However, in New Zealand, we expect that a study of sales and purchases separately would be more informative. Given that there are already doubts about the Securities Commission’s effectiveness as an active enforcement agency, we would find it almost impossible for the

484

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

Commission to have been able to make a case against an insider for delaying a planned sale of securities until after a substantial rise in the price. The measure of insider trading used in this paper is the basic purchase or sale transaction. This measure assigns the same weight to each buy or sell transaction irrespective of the trade size involved. Barclay and Warner (1993) report that the optimal size of an information-driven trade tends to be medium to small because larger trades tend to increase the probability of detection and can result in legal and market penalties. In addition, Seyhun (1986) suggests that smaller transactions are often more informative than larger transactions. Furthermore, we do not adjust the purchase or sale transactions for normal trading that goes on in a security. Therefore, we assume that each transaction was an unusual trade. To make a meaningful adjustment for normal trading, we would need data for a longer period than is currently available in New Zealand. 2.3. Methodology We used event study methodology to estimate abnormal security returns due to insider trading. Given the difference in the signal provided by insider purchases versus sales, we examined purchases and sales separately for each subsample. Previous research has shown that insider purchases are more informative than sales, as sales might be driven by such factors as liquidity needs and portfolio rebalancing, as opposed to private information. The estimation period used in this study was 190 trading days starting 250 trading days prior to each event, i.e., the date of each transaction. For each event, we estimated the market model over days t =
250 to t =
61, as follows: Rit ¼ ai þ bi Rmt þ eit

ð1Þ

where Rit and Rmt are, respectively, the returns of stock i and that of the market portfolio on day t of the estimation period, defined by log of daily price relatives and calculated from prices adjusted for capital changes. We used the NZSE All Ordinaries Index to measure the market portfolio. Using the estimates from Eq. (1) above, we then forecast daily abnormal returns of each security over a test period run from 60 trading days prior to the event date to 250 trading days after that date. We chose a long post-event window to show the long-term effect of insider trading, although we also report the abnormal returns for shorter subperiods surrounding the event date. Daily abnormal return for security i on day t of the test period (t =
60 to t = + 250) is given by the following equation: ARit ¼ Rit
ðai þ bi Rmt Þ:

ð2Þ

Further, as in Etebari and Duncan (1997), we also measured the abnormal returns using market-adjusted returns. This measurement did not change our results; hence, in this paper, we report and discuss the results from risk-adjusted procedures only. Daily abnormal returns were averaged using the formula AARt ¼

n 1X ARit n i¼1

ð3Þ

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

485

and tested for their significance according to the following t-statistic: AARt ð4Þ rE where rE = standard deviation of the error terms over the estimation period. Daily average abnormal returns were then accumulated over the entire test period, as well as selected subperiods, to give cumulative abnormal return, CARt. The CARs were standardized and tested for their significance according to the following formula: t-stat ¼

CARt t-stat ¼ pffiffiffi ð5Þ n  rE Further, the differences in results between various subsample CARs were tested according to the following t-statistic: ðCAR1
CAR2 Þ ffi qffiffiffiffiffiffiffiffiffiffiffiffiffi ð6Þ S12 S22 n þ n pffiffiffi pffiffiffi where S1 ¼ n*rE1 ; S2 ¼ n*rE2 , and rE1 and rE2 are standard errors of the two subsamples over the estimation period. Because the purpose of the SSH notices is to improve the speed with which the market reacts to insider trading and to prevent illegal transactions by insiders, it should follow that immediate disclosure shortens the period that insiders can earn abnormal profits. A test of the difference in the abnormal returns between the two subsamples would help to establish if the timing of disclosures has any effect on the insiders’ ability to profit from private information. t-stat ¼

3. Results 3.1. Summary statistics Table 1 reports summary statistics for the overall sample and the two subsamples, delayed disclosures and immediate disclosures. As reported in the table, the overall sample included a total of 1450 buy and 1003 sell transactions over the 1995 –2001 period involving 93 companies. In both subsamples, there were more buys than sales, and overall there were more delayed disclosure transactions than immediate disclosure transactions. This goes against the evidence from the U.S. market, which shows more sales than purchases. One explanation for this could be the differences in the usage of stock options as a part of executive compensation. Whereas in the US the use of stock options is very prevalent, in New Zealand their use is limited to very large companies. While directors in the US frequently sell exercised shares, this does not occur as much in New Zealand, with the result that the New Zealand market sees fewer share sale transactions than purchases. With the exception of years 1995 and 2001, the number of transactions was distributed fairly uniformly across the years covered in the study. The number of transactions in 2001 was smaller because at the time many companies had not yet filed annual reports for the fiscal year 2002. Also transactions were spread fairly equally across weekdays, with no

486

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

Table 1 Summary statistics on insider transactions and firms Panel A: Summary statistics on individual transactions Delayed disclosures (annual reports)

No. of transactions By year 1995 1996 1997 1998 1999 2000 2001 By day of the week Monday Tuesday Wednesday Thursday Friday Saturday Sunday By index membership NZSE 10 NZSE 40 (excl. NZSE 10) Non-NZSE 40

Immediate disclosures (SSH notices)

Overall

Purchases

Sales

Purchases

Sales

Purchases

Sales

853

661

597

342

1450

1003

166 109 128 155 122 127 30

117 143 87 102 99 108 21

41 62 111 128 150 67 38

10 18 78 76 88 45 27

207 171 239 283 272 194 68

127 161 165 178 187 153 48

140 165 149 157 196 35 11

118 118 149 98 159 9 10

110 111 128 107 141 0 0

68 76 43 79 76 0 0

250 276 277 264 337 35 11

186 194 192 177 235 9 10

153 170

132 131

49 146

28 84

202 316

160 215

530

398

402

230

932

628

Panel B: Summary statistics on individual companies

Market to book ratios Market value Firms with more than 1 trade

Average

Median

Maximum

Minimum

Standard deviation

1.32 NZ$362 85 (92% of firms)

0.93 NZ$50 65 (71% of firms)

0.73 NZ$429 86 (93% of firms)

1.01 NZ$282 74 (80% of firms)

0.12 NZ$47

apparent relationship to the day of the week effect documented in Etebari and Lont (2001). The table also shows 65 weekend trades. These trades are limited to the delayed disclosures subsample and most likely reflect transactions in overseas markets, such as ADRs in the U.S., or private, off-market transactions. Table 1 also shows that the majority of the trades involved firms not listed in a major stock index, such as NZSE 10 or NZSE 40. The NZSE 10 and NZSE 40 companies had similar numbers of transactions for the delayed disclosures subsample, but the NZSE 10 had significantly fewer transactions in the immediate disclosures subsample. The pattern evident for the immediate disclosures subsample also holds for the overall sample, with companies not in a major index making up the majority of the transactions.

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

487

Panel B of Table 1 presents information on the market-to-book ratios and the market values of the sample companies. The average market-to-book ratio for the sample companies was 1.32, with a median figure of 0.93. The average market value of the companies in the overall sample was NZ$362 million; however, the median number was much smaller, a mere NZ$49.5 million. Furthermore, 47% of the companies had a market value of less than NZ$50 million, while 15% had a market value exceeding NZ$500 million. Thus, the sample covered a wide range of market values, from the very small to the largest in New Zealand. Panel B also shows that over the study period at least 71% of the firms had one or more transactions in the buy or sell category. 3.2. Insider trading and abnormal security returns 3.2.1. Overall sample In this section, we examine the market’s response to insider transactions for the overall sample and the two subsamples of delayed and immediate disclosures. While we report returns for various days or intervals surrounding the dates of insiders’ purchases and sales, the abnormal returns at t = 0, day of transaction and the cumulative returns over days t =
1 to t = + 250 are the focus of our discussions. These returns respectively reflect the immediate impact and the long-term effect of insider transactions. The results for the overall sample are reported in Table 2. Panel A shows the average abnormal return (ARt) and the percentage of ARt values that are positive. Panel B reports the CAR over selected event windows. As shown in the table, for overall purchases the AR at t = 0 is a positive but statistically insignificant return of + 0.21%. The results are also insignificant throughout the window of days
5 to + 10. Further, these results do not reveal any noticeable difference between the proportions of positive and negative abnormal returns for each day. For t =
1 to t = + 250, roughly one calendar year following each transaction, the CAR for insider purchases is a statistically significant return of + 6.5%. As reported in Panel A of the table, the average abnormal return results for overall sales are also mostly insignificant and, for each day of the test period, there are roughly equal numbers of positive and negative abnormal returns with the exception of day 3, which has 39% of positive ARs. The AR at t = 0 is an insignificant return of
0.2%, reflecting no strong immediate reaction by the market to insider sales. However, the CAR over t =
1 to t = + 250 is a statistically significant
5.99%. The results for the overall sample suggest that insiders can time the market to get longterm gains only. In contrast, the evidence from overseas shows that insiders could time their transactions to make short-term as well as long-term gains. Past evidence has also shown that insiders sell after a price run-up and buy after a price decrease, which reverses shortly after the insider’s trade. Our results show little evidence of this pattern. For purchases, there is only one day in the five days prior to the transaction date that has a negative abnormal return. For sales, the evidence shows three days with negative returns in the five days prior to the transaction date. Fig. 2 gives a graphic illustration of the results for the overall sample. It presents some interesting evidence on the inability of insiders to time their trades. For instance, insider sales typically occur during a period of increasing prices. Insiders also sell near the peak of

488

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

Table 2 The overall sample results Panel A: ARs for overall purchases and sales transactions Event date


5
4
3
2
1 0 1 2 3 4 5 6 7 8 9 10

Purchases

Sales

Abnormal returns

% of ARs positive

Abnormal returns

% of ARs positive

0.0004
0.0002 0.0005 0.0008 0.0017 0.0021 0.0013 0.0018 0.0016
0.0016 0.0019
0.0008 0.0046*
0.0004 0.0006
0.0012

44% 46% 49% 51% 60% 56% 54% 52% 48% 50% 49% 53% 53% 53% 52% 51%


0.0003 0.0011 0.0014
0.0010
0.0004
0.0020
0.0001
0.0014
0.0011 0.0023 0.0020 0.0017 0.0011*
0.0001
0.0024 0.0000

53% 59% 53% 48% 53% 47% 48% 46% 39% 58% 58% 60% 50% 47% 44% 53%

Panel B: Event window CARs for overall purchases and sales Windows


60, 250
1, 250
1, 30
1, 5
1, 1
1, 0
5, 10

Purchases

Sales

Cumulative abnormal returns

Cumulative abnormal returns

0.0725* 0.0650* 0.0230* 0.0127 0.0038 0.0017 0.0132


0.0409
0.0599** 0.0090
0.0018
0.0024
0.0004 0.0008

* Significant at 10%. ** Significant at 5%.

the increase, with prices flattening out before they decline. As for their purchases, insiders buy when the price is increasing. On average, insiders buy about 10 days into a significant price run-up that typically lasts 3 months. These results differ from those documented in previous research based on the North American markets, where insiders have apparently been able to successfully time their trades in the short run, selling at the peak of an increase immediately prior to a reversal and purchasing just before an upturn in a company’s share price. There is little evidence in either the data in Table 2 or Fig. 2, however, to suggest that insiders in New Zealand can successfully time the market in the short-run. 3.2.2. Delayed disclosure versus immediate disclosures The insider purchases and sales were then separately analyzed for delayed disclosures and immediate disclosures, with the results for purchases presented in Table 3. The most

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

489

Fig. 2. CARs for overall insider purchases and sales.

noticeable feature of these results is that insiders seem able to earn significantly greater returns by delaying the disclosure of their share purchases. Over the period t =
1 to t = + 250, the cumulative return for the delayed disclosures subsample due to insider purchases is a statistically significant + 8.92%, compared with an insignificant + 4.12% return for the immediate disclosures subsample. There is still, however, little evidence of a short-term effect and the insiders’ ability to time their purchases. The results in Table 3 are graphically illustrated in Fig. 3, which depicts the CARs for delayed disclosure and immediate disclosure purchases. From this figure, it can be seen that the two subsamples perform virtually identically for the first 20 days, after which CARs for the delayed subsample tend to out perform the CARs for the immediate subsample. This trend continues up until about 6 months after the transaction date. CARs for the immediate subsample stay at the same level, while CARs for the delayed subsample continue to increase over the remainder of the test period, reflecting the presence of a significant, prolonged effect. The results for insider sales are presented in Table 4. As can be seen, CARs due to insider sales for the two subsamples are noticeably different from those of the purchases. Despite a significantly negative AR at t = 0 for immediate disclosures, the majority of days returned positive abnormal returns for the first 10 days following the transaction. Over t =
1 to t = + 250, delayed disclosures post a significant CAR of
7.70%, whereas immediate disclosures earn an insignificant CAR of
4.49%. Again, CARs for the delayed subsamples are nearly double that of the immediately disclosed subsample. The results for insider sales are also graphically portrayed in Fig. 4. This figure shows that delayed disclosure sellers seem to time the market better, selling at the peak of the

490

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

Table 3 Insider purchases: delayed vs. immediate disclosure Panel A: ARs for purchase transactions Event date


5
4
3
2
1 0 1 2 3 4 5 6 7 8 9 10

Delayed purchases

Immediate purchases

Abnormal returns

% of ARs positive

Abnormal returns

% of ARs positive

0.0009 0.0010 0.0018
0.0024 0.0001 0.0060* 0.0014
0.0001 0.0017
0.0031
0.0004
0.0023 0.0073*** 0.0011 0.0002
0.0006

46% 45% 49% 51% 55% 61% 55% 52% 54% 49% 49% 52% 51% 56% 55% 52%


0.0001
0.0014
0.0008 0.0039 0.0033
0.0017 0.0013 0.0037 0.0014
0.0001 0.0042* 0.0007 0.0020
0.0018 0.0009
0.0019

41% 47% 48% 51% 65% 52% 54% 52% 42% 52% 49% 53% 54% 51% 49% 51%

Panel B: Various event window CARs for purchase transactions Windows


60, 250
1, 250
1, 30
1, 5
1, 1
1, 0
5, 10

Delayed purchases

Immediate purchases

Cumulative abnormal returns

Cumulative abnormal returns

0.0960** 0.0892** 0.0294** 0.0057 0.0061*** 0.0001 0.0128

0.0494 0.0412 0.0166 0.0120** 0.0016 0.0033 0.0136*

* Significant at 10%. ** Significant at 5%. *** Significant at 1%.

price run-up which then slowly decreases over the next 30 days. The CAR for immediate sales continues to trend upward for quite some time, nearly doubling in the 30 days after the transaction. Afterwards, it trends downward, reaching a similar level to the CAR for the delayed subsample. The similarities at the end can be seen from the figure and also from Panel B of Table 4, where the cumulative results for the
60 to + 250 window show less than a 2% difference in the end results. These results suggest that insiders delaying disclosure are able to time the market better in the short-term and consequently earn a higher return. We tested the difference in CARs due to purchases and sales between the delayed and immediate disclosures subsamples for various intervals, with the results presented in Table 5. As shown in the table, the difference in CARs due to purchases starts out insignificant, but by day + 5 the immediate purchases post a statistically larger CAR. From that point

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

491

Fig. 3. CARs for insider purchases: delayed vs. immediate disclosure.

on, the difference in CARs takes on a different sign at every event window, indicating that CARs are larger and more significant for the delayed disclosures. At the end of the period, the difference in CARs between the two subsamples is nearly 5%. The results for sales are similar to those of purchases. The immediate subsample results are initially larger, but by day + 30 the delayed subsample posts significantly larger returns. Over days
1 to + 250, the difference in CARs for sales between delayed and immediate disclosures is a statistically significant return of 3.2%. These results show a far greater long-term effect for both insider purchases and sales when disclosure is delayed. The results show that, overall, insiders earn abnormally large returns from either purchasing or selling shares in their companies, with the returns being slightly larger than those reported for the foreign markets. The results also support the finding that sales are less informative than purchases. When the sample is split into delayed and immediate disclosures the results reveal that the profits in the overall sample are the result of the superior abnormal returns earned by delayed disclosure transactions. The immediate disclosures subsample produced results in the expected direction; however, those results were statistically insignificant, while the abnormal returns to the delayed disclosures subsample both had the expected sign and were statistically significant. The differences in the CARs for the purchases and sales between the two subsamples support the initial hypothesis that delays in disclosure result in significantly higher abnormal returns. This also supports the findings in Huddart et al., (2001) that timely disclosure results in a reduction in insider trading profits. As can be seen in both the purchase and sales categories, the abnormal returns for the immediate disclosures subsample are roughly half that of the delayed disclosures subsample. Our specific subsample results give clues as to the best way to reduce the profits of insiders: timely disclosure. The extent of the abnormal gain in New Zealand is

492

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

Table 4 Insider sales: delayed vs. immediate disclosures Panel A: ARs for sale transactions Event date


5
4
3
2
1 0 1 2 3 4 5 6 7 8 9 10

Delayed sales

Immediate sales

Abnormal returns

% of ARs positive

Abnormal returns

% of ARs positive


0.0005
0.0004 0.0014
0.0007
0.0002 0.0019
0.0019
0.0038* 0.0008 0.0042*
0.0020 0.0009 0.0015 0.0000
0.0035 0.0016

56% 61% 50% 52% 52% 53% 47% 41% 39% 59% 53% 59% 47% 47% 45% 55%


0.0002 0.0023 0.0014
0.0014
0.0006
0.0054* 0.0016 0.0007
0.0026 0.0007 0.0055* 0.0023 0.0008
0.0002
0.0013
0.0014

49% 58% 56% 45% 55% 41% 49% 51% 40% 58% 63% 60% 53% 48% 42% 52%

Panel B: Various event window CARs for sales transactions Windows


60, 250
1, 250
1, 30
1, 5
1, 1
1, 0
5, 10

Delayed sales

Immediate sales

Cumulative abnormal returns

Cumulative abnormal returns


0.0509
0.0770***
0.0012
0.0011 0.0017
0.0002
0.0008


0.0322
0.0449 0.0179
0.0002
0.0060***
0.0006 0.0021

* Significant at 10%. *** Significant at 1%.

significantly larger than it is in other countries with more transparent disclosure requirements that encourage more timely disclosure of insider trading. A likely reason for such large returns in New Zealand is the insiders’ ability to make multiple trades before the market is informed. Without these signals, the market will be slow in readjusting and will continue to provide insiders with advantageous trading opportunities. Timely disclosure will reduce the ability of insiders to exploit these opportunities over the long-term. 3.2.3. Information asymmetry and abnormal returns The literature on insider trading has established that the level of information asymmetry between insiders and the market is a major determinant of the profitability of insider trading. The extent of information asymmetry depends, in turn, on the degree of

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

493

Fig. 4. CARs for insider sales: delayed vs. immediate disclosure.

mispricing of the company that the insider trades in and the amount of knowledge that the insider has about the mispricing. Frankel and Li (2001) and Lakonishok and Lee (2001), among others, show that small companies and less-researched firms provide insiders with greater opportunities to make abnormal returns. Seyhun (1986), Friederich et al. (2002) and Lakonishok and Lee (2001) conclude that insiders have a more accurate perception of the value of their company’s shares and therefore know when the company is undervalued. In small companies, which tend to be less followed by fund managers and security analysts, information asymmetry is greatest. The fewer analysts following these companies results in a relatively less efficient market for their stocks. Given that trading by insiders is

Table 5 Difference in CARs between delayed and immediate disclosures Windows


60, 250
1, 250
1, 30
1, 5
1, 1
1, 0
5, 10 ** Significant at 5%. *** Significant at 1%.

Purchases

Sales

CAR difference: delayed less immediate disclosure

CAR difference: delayed less immediate disclosure

0.0466*** 0.0480*** 0.0129***
0.0063** 0.0046
0.0032
0.0008


0.0187***
0.0321***
0.0191***
0.0009 0.0077*** 0.0004
0.0030

494

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

a valuable source of information about these companies, when such trades are not disclosed in a timely fashion, there is little chance of the insider’s information being incorporated into the share price. This results in ongoing mispricing and greater opportunities to earn abnormal returns. This informational advantage should therefore be apparent when the sample is separated by the size of the company and membership in a major stock index. Past research has also shown that the knowledge of the individual insiders about the correct price of their companies’ shares depends upon their position and rank within their companies. Carpenter and Remmers (2001) and Seyhun (1998) both examined the effect of the position of the insider within the company and found that the more information an insider had, the higher the return that they earned. Arguably, top-ranked managers who know the most about their companies and hence are better able to price their companies’ shares could potentially earn the greatest returns from insider trading. The impact of information asymmetry on insider trading profits was examined by separating each subsample on the basis of company size, membership in a stock index and the position of the insider in the company. Only the results for the delayed disclosures subsample were significant. The results for the immediate sample were insignificant; hence, they are not reported in the paper. 3.2.3.1. Effect of size and index membership. The company size was measured by market valuation of the company at the time the transaction was made. Transactions that occurred when the company had a market capitalization of less than NZ$50 million were classified as small, while those associated with companies that had a market capitalization of greater then NZ$500 million were classified as large. Those falling within this range were analyzed but not reported in the paper in order to highlight the contrast between the two extremes. Index membership was examined by separating the affected companies into three categories: those falling within the NZSE 10, those included in the NZSE 40 after removing the firms in the NZSE 10 and those not included in the NZSE 40. Firms in the NZSE 10 are the largest firms and are more heavily researched by analysts. Those outside the NZSE 40 are typically less researched; hence they offer a significant informational advantage to insiders. Once again, to highlight the contrast in the results, we report and discuss the results for the firms in the NZSE 10 and those outside the NZSE 40. The results are reported in Panels A and C of Table 6. Focusing on purchases, it can be seen that when the sample is split on size or index membership, small and less-researched firms outperform their counterparts. Over the 250 days following the transactions, small firms and non-NZSE 40 firms post large abnormal returns of 21.53% and 11.57%, respectively. Surprisingly, large firms and NZSE 10 firms both lose (1.59% and 1.29%, respectively) over days
1 to + 250, although these returns are statistically insignificant. From these results, it can be concluded that smaller and less-researched companies account for a disproportionately large share of the abnormal returns we reported earlier for delayed disclosure purchases. For sales, size and index membership also produce similar results. Over the period
1 to + 250, small firms outperform large firms and less-researched firms marginally outperform NZSE 10 companies. However, in this case, all the firms, regardless of our classifications, respond negatively to insider sales. This is a significantly different result

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

495

Table 6 CARs by firm size and index membership Panel A: Sample separated by firm size Windows


60, 250
1, 250
1, 30
1, 5
1, 1
1, 0
5, 10

Purchases

Sales

Large companies

Small companies

Difference

Large companies

Small companies

Difference


0.0397
0.0159 0.0188** 0.0092** 0.0036**
0.0010
0.0042

0.2800*** 0.2153*** 0.0365
0.0016 0.0070 0.0020 0.0239


0.3198***
0.2312***
0.0177*** 0.0108**
0.0034
0.0030
0.0280*


0.1044***
0.1245***
0.0113
0.0072 0.0015 0.0002
0.0088


0.1582**
0.1569***
0.0096
0.0079
0.0048
0.0018
0.0126

0.0539*** 0.0324***
0.0017 0.0007 0.0063 0.0020 0.0039

194

203

Panel B: Transaction details separated by firm size Number of transactions

250

192


NZ$89,510.48
NZ$305,588.22 1,143,154.85 547,030.93

Average profit NZ$70,404.27 NZ$153,008.19 Average shares 857,569.65 314,283.24 traded Average profit 0.08 0.49 per share

(0.08)

(0.56)

Panel C: Sample separated by index membership Windows

Purchases NZSE 10


60, 250
1, 250
1, 30
1, 5
1, 1
1, 0
5, 10


0.0681
0.0129 0.0128 0.0114* 0.0032
0.0014 0.0054

Sales Non-NZSE 40 Difference 0.1403*** 0.1157** 0.0321* 0.0019 0.0064** 0.0005 0.0146


0.2084***
0.1286***
0.0193*** 0.0096**
0.0032
0.0018
0.0092

NZSE 10

Non-NZSE 40

Difference


0.0851**
0.1014***
0.0047
0.0080
0.0013
0.0014
0.0062


0.0949**
0.1090***
0.0026
0.0036 0.0021
0.0012
0.0070

0.0098*** 0.0076**
0.0021
0.0044
0.0034
0.0002 0.0009

130

398

Panel D: Transaction details separated by index membership Number of transactions

140

523

Average profit NZ$42,778.86 NZ$71,696.23 Average shares 278,436.04 280,698.48 traded Average profit 0.15 0.26 per share


NZ$570,936.97
NZ$127,234.11 703,713.45 800,105.08 (0.81)

Large companies=mkt. cap>500 million, small companies=mkt. cap<50 million. * Significant at 10%. ** Significant at 5%. *** Significant at 1%.

(0.16)

496

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

than we reported for purchases, where the abnormal returns were driven largely by small and less researched firms. Lakonishok and Lee (2001) document similar results for the U.S. stock markets, except that they do not find any market reaction to insider sales in larger companies. This may be due to the differences in disclosure requirements and executive compensation schemes between the two countries. Panels B and D give details about the individual transactions and the average profitability of transactions for both purchases and sales. The results for purchases are generally supportive of the presence of a greater informational advantage to small and lessresearched firms. Small firms, for instance, make twice the average return with less than one-half the average number of shares, resulting in a very high per share average profit of NZ$0.49 as opposed to just NZ$0.08 for large firms. The results are similar for non-NZSE 40 companies except that the average number of shares traded in NZSE 10 and non-NZSE 40 firms is very similar. The average profit is, however, still nearly double, as non-NZSE 40 firms earn NZ$0.26 per share as opposed to NZ$0.15 per share for NZSE 10 firms. The positive results for the large and NZSE 10 companies shown for short-term event windows compared to the negative CARs for the period
1 to + 250 suggest that these insiders are able to pick good times to buy to some degree. They buy more when the transaction earns a larger 250-day CAR. The results for sales, however, significantly depend upon index membership, but not upon the company size. With regard to size, insiders in small companies make triple the average return with half the shares, earning a per share return of NZ$0.56. As for index membership, NZSE 10 and non-NZSE 40 firms trade similar numbers of shares on average, but the NZSE 10 insiders earn significantly higher average profit. They earn NZ$0.81 per share as opposed to their non-NZSE 40 counterparts who earn just NZ$0.16 per share. Again, this result suggests that NZSE 10 insiders are able to time when to sell shares. It is likely that NZSE 10 insiders sell heavily when their trade is driven by information but lightly for other reasons. The results, overall, support the notion that information asymmetry between the company and the market affects the return earned by insiders. The CAR results indicate that both small companies and less-researched non-NZSE 40 companies earn significantly higher returns for both purchases and sales, although the difference is significantly larger for purchases than it is for sales. Furthermore, in terms of the average profit per share, purchases of both small and non-NZSE 40 companies generate significantly higher profits. The latter results also hold for small firm sales, but not for index membership, where NZSE 10 firms earn more than four times the average profit per share of non-NZSE 40 companies. 3.2.3.2. Effect of insider position. As shown by Carpenter and Remmers (2001) and Seyhun (1998), the more information that an insider has, the higher the abnormal returns they earn. In this section, we examine the extent to which the position of the insider within the organization explains the abnormal returns of delayed purchases. Managing Directors, for instance, have more information than do independent directors due to their close involvement with the company. As with the testing for company size and index membership, the position of the insider was determined at the time of the transaction. Although a number of directors moved up in rank from director to chairman or from executive director to managing director, very few went from a position of more information to one of less. In the latter circumstances, the majority of the changes were

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

497

associated with retirements from the board. Whenever a director moved up in rank, we used the new position in the study. However, given the small number of such occurrences, we do not expect these cases to bias the results. Table 7 shows that purchases and sales had similar percentages of transactions undertaken by each class of insider. In both samples, about half of all transactions were undertaken by non-executive directors. This is not surprising, given that non-executive directors make up more than 50% of most boards. Managing directors were responsible for 18% of both purchases and sales transactions and ranked as the second most active purchasers of shares. Executive directors made more sales, which may be related to the use of option and share incentive schemes by a few companies. As shown by Ofek and Yermack (2000), insiders getting shares via executive option schemes tend to sell more shares as they seek to diversify their holdings. In both samples, chairmen of the board made the least number of transactions at 13% and 15% for purchases and sales, respectively. The abnormal return results by the insider’s position within the company are shown in Panel A of Table 8. These results reveal that the insider’s position influences the return they can earn. Over the period t =
1 to + 250, on their purchases chairmen earn the most, almost 1% above managing directors, who earn more than executive directors. Nonexecutive directors, those with the least access to information, do not earn significant returns. The same pattern is observed for sales, except over the same period that managing directors/CEOs outperform chairmen. The other difference is the inability of executive directors to make returns off sales. Whereas non-executive directors make a statistically significant return of 9.44%, executive directors make a loss of 3.27%. The sales results might have, of course, been driven by diversification or liquidity motives. The purchases results are, however, unambiguous in showing that insiders with greater access to information, such as, chairmen and those who hold executive posts, earn higher returns on their share purchases than those in non-executive posts. These results are consistent with those reported by Carpenter and Remmers (2001) and Seyhun (1998). Panel B of Table 8 reports total profits and profits per share for each group of insiders. These results are generally supportive of the CAR results reported in Panel A. The only exceptions are that in their purchases non-executive directors earn higher profits per share than either executive directors or chairmen. Also, managing directors making either purchases or sales earn significantly higher profits per share than the other classes of insiders. Chairmen earn the highest average profits; they also trade twice as many shares on average. The results for the executive directors’ sales show that, on average, they earn NZ$0.01 per share as opposed to a loss of 3.27% from the CAR tests. Again, this may be related to the sale of shares from executive share purchase plans by managing directors for Table 7 Delayed disclosure transactions by insider’s position Purchases

Managing director/CEO Chairman Executive director Non-executive director

Sales

# of transactions

% of sample

# of transactions

% of sample

153 115 114 471

18% 13% 13% 55%

122 96 127 316

18% 15% 19% 48%

498

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

Table 8 CARs and transaction details by insider position Panel A: CARs by insider’s position Windows

MD/CEO

Chairman

Purchases
60, 250
1, 250
1, 30
1, 5
1, 1
5, 10

0.1112** 0.1136*** 0.0089
0.0094 0.0019
0.0097

0.1603* 0.1219* 0.0172 0.0009
0.0024 0.0096

Sales
60, 250
1, 250
1, 30
1, 5
1, 1
5, 10


0.0949
0.1257***
0.0183
0.0081 0.0049
0.0102


0.1129*
0.1041*
0.0273
0.0099
0.0017
0.0216*

Executive director Non-executive director 0.0700 0.1067* 0.0166 0.0219 0.0147 0.0124

0.0648 0.0327
0.0077
0.0068
0.0021
0.0019

0.0586 0.0478 0.0348** 0.0103 0.0027 0.0208*


0.0742
0.0944** 0.0081 0.0026
0.0009 0.0063

Panel B: Transaction details separated by insider’s position

Purchases Number of transactions Average profit Average shares traded Average profit per share

MD/CEO

Chairman

Executive director Non-executive director

148 NZ$44,491.97 173,742 NZ$0.26

114 NZ$115,622.63 1,018,044 NZ$0.11

114 NZ$32,450.91 347,862 NZ$0.09

Sales Number of transactions 122 95 127 Average profit
NZ$194,602.39
NZ$287,749.20
NZ$2,970.01 Average shares traded 559,559 1,527,470 222,122 Average profit per share (0.35) (0.19) (0.01)

465 NZ$88,380.40 536,035 NZ$0.16

315
NZ$118,681.49 722,158 (0.16)

* Significant at 10%. ** Significant at 5%. *** Significant at 1%.

liquidity or diversification reasons. Overall, the results support the finding that higher ranked insiders perform better. In particular, managing directors and chairmen outperform lower ranked insiders such as executive directors and non-executive directors. The greater the difference in knowledge between the individual and the market, the greater the returns that insider can earn. 3.2.3.3. Transaction size and stealth trading. A number of studies refer to a stealth trading hypothesis (Kraakman, 1991) arguing that insiders trade in such a way as to avoid alerting the market to their trading (Friederich et al., 2002). One way to engage in stealth trading is to make numerous small trades as opposed to a single large trade. We tested this hypothesis by examining delayed disclosure purchases of insiders to determine if transaction size relative to the insider’s holdings in the company is related to the abnormal

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

499

Table 9 CARs and the size of insider’s holdings Purchases

Average % holding Average CAR

Sales

Full sales

Small

Large

Small

Large

Full sales

Partial sales

0.59% 27%

5055.57%
6%

0.73%
14%

83%
3%

100%
15.34%

14.60%
4.70%

profits. The regulatory environment in New Zealand presents the perfect conditions to observe clustered small trades as the lack of timely disclosure gives insiders an incentive to undertake such trades. Provided that the market is not alerted, an insider could trade frequently on superior information by making a number of small trades. Table 9 presents the results for large and small transactions, where large and small categories represent the top and bottom 20% of transactions based on the percentage of the insider’s holdings being traded. The table also shows the results for partial sales and full sales of holdings. As can be seen, there are substantial differences in the results between the smallest and largest transactions. For purchases, the smallest transactions were, on average, 0.59% of the insider’s holdings but earned a significantly large return of 27%. In contrast, the largest transactions, which involved insiders greatly increasing their holdings in the company, lost 6.03%. Unlike purchases, large insider sales did still earn a positive abnormal return, i.e., avoiding a loss of 2.95%; however, small sales earned 14.09%, a significantly higher return than large sales. These results support the findings in previous research that insiders hide their transactions by trading only small amounts to avoid detection. Small trades of an insider’s holdings are unlikely to provoke much market interest, but as revealed by the results these small transactions are extremely profitable to the insider. One situation where there can be more information in a large transaction than in a small one is the full sale of holdings. In the case of an insider fully selling out, the transaction contains a considerably large amount of information about the company’s future prospects, often signaling bad news. Given the significance of the information, these transactions should generate larger abnormal profits. As is shown by the results, full sales earn over 10% higher abnormal returns than partial sales, indicating that full sales are indeed more informative than partial sales. The results clearly show that both small to medium sized trades and trades where the insider completely sells out of the company contain material information that can be used to earn significant abnormal returns. Full sales signal that the company will perform extremely poorly in the near future. Small transactions tend to indicate that an insider has information that they wish to keep undisclosed in order to preserve the profits that they can earn from the information. Either way, insiders are informed traders and informed trades make profitable trades.

4. Conclusions In this paper we examined the relationship between the timing of mandatory disclosures of share transactions by insiders and their ability to profit. The trades subjected to analysis

500

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

were taken from New Zealand, where until recently a two-tiered disclosure regime was in effect for reporting insider trades. Depending on the size of the transaction in question and its relationship to the individual’s holdings, transactions were either disclosed immediately or reported in the affected company’s annual report. Disclosures in company annual reports represented a delay of at least several months while the reports were being printed and distributed. This provided ample opportunity for directors to act on privileged information and undertake multiple transactions without the market knowing. The delayed disclosures were the focus of our study. Using a large sample of 2453 insider trades for 93 listed firms in New Zealand, we find that over the January 1995 to December 2001 period insiders actively traded in their companies’ shares and earned significant long-term returns. The cumulative abnormal returns for purchases and sales over the calendar year following transactions were + 6.5% and
5.99%, respectively. Further tests show that much of the results were due to our delayed disclosures subsample, i.e., trades reported in company annual reports long after they occurred. For this subsample, we find significantly large returns of + 8.92% for purchases and
7.70% for sales. The results for our immediate disclosures subsample were insignificant for both purchases and sales. Further analysis of the delayed disclosures subsample revealed that company size, analyst following, and insider position were significant contributors to the results. That is, small companies earned significantly higher returns than large companies; companies not included in a major stock index, i.e., those less followed by analysts, outperformed those listed in the NZSE 10; and insiders privy to more information, such as managing directors and chairmen, earned higher abnormal returns than other classes of insiders. The results also showed that small transactions in terms of the percent of the insiders’ holdings being traded resulted in more significant returns than large transactions, with the exception of full sales, which outperformed partial sales largely due to the information these transactions convey. A key finding of our study is that by delaying disclosure of their trades insiders can earn superior profits. Delays in disclosure add to information asymmetry in the market, thereby increasing transaction costs of trading, reducing market liquidity and increasing the cost of capital. In contrast, timely disclosure reduces trading opportunities for the insiders, thus enhancing investor confidence and the integrity of the market. In a major extension of the previous legislation, the Securities Markets Amendment Act 2001 requires continuous disclosure of trades by all insiders, including the previously covered substantial shareholders, as well as directors and company executives. Based on our results, we suspect that such timely disclosure required by this act has the potential to reduce profitability of insider trading while also improving market efficiency. To produce tangible results, however, the newly mandated disclosure requirements need to be more effectively policed and enforced. Previously, in New Zealand there were major concerns regarding the monitoring and enforcement of the existing laws and regulations. There was a perception that the laws were less effectively policed and enforced than those of most other developed countries, like the U.K. or the U.S., that have had laws in effect governing timely disclosure of trades for some time. A major reason for this weakness was the perceived lack of enforcement power of the New Zealand Securities Commission, the principal regulatory watchdog. The new act gives the Securities Commission extra authority to transition the agency into an effective investigatory and enforcement unit.

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

501

Whether or not the new disclosure requirements, in combination with effective enforcement, produce the desired outcome is an empirical question. Future research will be needed to study the effectiveness of the new act. Acknowledgements An earlier version of this paper was presented at the 7th meeting of the New Zealand Finance Colloquium, Palmerston North, 2002, the 15th Australasian Banking and Finance Conference, Sydney, 2002, and seminars at University of Waikato, Otago University and Massey University, Albany. We wish to thank participants for their helpful comments. We would also wish to thank an anonymous referee and the Editor of the Journal for their valuable suggestions. Any remaining errors are ours. References Ausbel, L., 1990. Insider trading in a rational expectations economy. American Economic Review 80, 1022. Barclay, M., Warner, J., 1993. Stealth trading and volatility: Which trades move prices? Journal of Finance 42, 1245 – 1260. Bettis, C., Vickey, D., 1997. Mimickers of corporate insiders who make large-volume trades. Financial Analysts Journal 53, 57. Bhattacharya, U., Daouk, H., 2002. The world price of insider trading. Journal of Finance 57, 75. Carlton, D., Fischel, D., 1983. The regulation of insider trading. Stanford Law Review 35, 357 – 895. Carpenter, J., Remmers, B., 2001. Executive stock option exercises and inside information. Journal of Business 74, 513. Casey, Q., Tourani-Rad, A., 2001. Are Directors Deals Dodgy? Some New Zealand Evidence. Unpublished Paper, University of Waikato. Chung, K., Charoenwong, C., 1998. Insider trading and bid – ask spread. Financial Review 33, 1. Duncan, K., Etebari, A., 1990. Insider trading, information leakage and pre-announcement price movements: evidence from the New Zealand market. In: Rhee, G., Chang, R. (Eds.), Pacific Basin Capital Markets Research. North-Holland, Amsterdam, pp. 227 – 241. Eckbo, E., Smith, D., 1998. The conditional performance of insider trades. Journal of Finance 53, 467 – 498. Etebari, A., Duncan, K., 1997. The efficacy of insider trading laws: New Zealand evidence. New Zealand Investment Analyst 18, 17 – 25. Etebari, A., Lont, D., 2001. Monthly and daily patterns in equity returns: NZ evidence. New Zealand Investment Analyst 22, 13 – 17. Finnerty, J., 1976. Insiders and market efficiency. Journal of Finance 31, 1141 – 1148. Frankel, R., Li, X., 2001. The characteristics of a firm’s information environment and the predictive ability of insider trades. Working Paper, Massachusetts Institute of Technology. Friederich, S., Gregory, A., Matatko, J., 2002. Short-term returns around the trades of corporate insiders on the London stock exchange. European Financial Management 8, 7 – 30. Huddart, S., Hughes, J., Levine, C., 2001. Public disclosure and dissimulation of insider trades. Economica 69, 665 – 681. Jaffe, J., 1974. Special information and insider trading. Journal of Business 47, 410 – 428. Jeng, L., Metrick, A., Zeckhauser, R., 1999. The profits to insider trading: A performance-evaluation perspective. Working Paper, Harvard University. Kraakman, R., 1991. The legal theory of insider trading regulation in the United States. In: Hopt, K. (Ed.), European Insider Dealing. Butterworths, London, pp. 39 – 55. Kyle, A., 1985. Continuous auctions and insider trading. Econometrica 53, 1315 – 1335. Lakonishok, J., Lee, L., 2001. Are insider trades informative? Review of Financial Studies 14, 79 – 111.

502

A. Etebari et al. / Pacific-Basin Finance Journal 12 (2004) 479–502

Leland, H., 1992. Insider trading: Should it be prohibited? Journal of Political Economy 100, 859 – 887. Lorie, J., Niederhoffer, V., 1968. Predictive and statistical properties of insider trading. Journal of Law and Economics 11, 35 – 53. Ofek, E., Yermack, D., 2000. Taking stock: Equity-based compensation and the evolution of managerial ownership. Journal of Finance 55, 1367 – 1384. Pope, P., Morris, R., Peel, D., 1990. Insider trading: some evidence on market efficiency and directors share dealings in Great Britain. Journal of Business Finance and Accounting 17, 359 – 380. Repullo, R., 1999. Some remarks on Leland’s model of insider trading. Economica 66, 359. Rozeff, M., Zaman, M., 1988. Market efficiency and insider trading: New evidence. Journal of Business 61, 25 – 44. Seyhun, N., 1986. Insider profits, costs of trading and market efficiency. Journal of Financial Economics 16, 189 – 212. Seyhun, N., 1992. Why does aggregate insider trading predict future stock returns? Quarterly Journal of Economics 107, 1303 – 1331. Seyhun, N., 1998. Investment Intelligence from Insider Trading. MIT Press, Cambridge, MA.