Number Affirmative/Percent of Responses MORE RESTRICTIVE
Activity/Reporting During the Past Eight Years
300/59.4% 188/37.2% 17/3.4% 505
Number Affirmative/Percent of Responses MORE INVOLVED Changes In Company Involvement Pertaining
To Insider Trading Activity/Reporting During the Past Eight Years
318/74.3% 106/24.8% 4/0.9% 428
TABLE 4
Comparison of trade frequency and dollar volume per firm-day for trades of all insiders of firms that do have blackout period versus those that don’t and for firms that do have policies/procedures restricting insider trading versus those that do not. Trading rates are reported in aggregate and for purchases and sales separately. Sample consists of 593 firms (members of the American Society of Corporate Secretaries) that provided relatively complete responses to our November 1996 survey, for which we are able to obtain insider trading data from SEC filings, and are present on CRSP and Compustat. The insider trading data cover the period January 1, 1992 to June 30, 1997. Trading rates and dollar volumes are calculated firm by firm and then averaged across firms.
TRADES PER FIRM-DAY DOLLAR VOLUME OF TRADES PER FIRM-DAY
FIRMS WITH
TRADES PER FIRM-DAY DOLLAR VOLUME OF TRADES PER FIRM-DAY
FIRMS WITH
TABLE 5
Maximum likelihood estimates for various logit models. In models 1 and 2, the classification variable equals one if the firm has a blackout period in place; zero otherwise. In models 3 and 4, the classification variable equals one if the firm has a trading policy in place that limits insider trading; zero otherwise. Sample consists of 626 firms (members of the American Society of Corporate Secretaries) that provided relatively complete responses to our November 1996 survey and appear on CRSP and Compustat. Firm assets and market-to-book are based on the five year average using data from Compustat for 1992-1996. The standard deviation of monthly stock returns is calculated over this same period. For each firm, the fractional ownership of all insiders is calculated as the total shares owned by insiders as reported in DISCLOSURE. There are only 589 observations for fractional inside ownership. Wald chi-squared in parentheses.
Dependent Variable/
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2
Trading Frequency
Per Firm-Day
1 2 3 4 5 6 7 8 9 10 All
Others Day Relative To Beginning Of Blackout Period And Trading Window
FIGURE 1. Trading frequency per firm-day, aggregating purchases and sales, for trading days and blackout days for the first 10 days of each type of period matched. Sample consists of 286
firms with easily-defined trading windows and blackout periods.
Window Day
Blackout Day
TABLE 6
Pooled time-series cross-section regressions of the number of trades per firm-day and dollar volume of trades per firm-day on indicator variables for various insider trading policies and control variables. Trading policy variables include an indicator equal to one if the firm has trading policies but no blackout periods, and indicator equal to one if the firm has trading policies and blackout periods, and indicator for blackout days, and indicator for the post 1994 period, and the interaction between the post 1994 and blackout day indicator. The control variables include the log of total assets, market-to-book, indicator variables for banks and utilities, and the standard deviation of monthly stock returns. The sample consists of all open market purchases and sales by insiders in 284 firms with blackout periods and 119 firms without blackout periods from the period January 1, 1992 to June 30, 1997. T-statistics in parentheses.
DependentVariable/
Log of Total Assets --- 0.0073***
(23.57)
TABLE 7
Pooled time-series cross-section regressions of the number of trades per firm-day and dollar volume of trades per firm-day disaggregated by purchases and sales on indicator variables for various insider trading policies and control variables. Trading policy variables include an indicator equal to one if the firm has trading policies but no blackout periods, and indicator equal to one if the firm has trading policies and blackout periods, and indicator for blackout days, and indicator for the post-1994 period, and the interaction between the post-1994 and blackout day indicator. The control variables include the log of total assets, market-to-book, indicator variables for banks and utilities, and the standard deviation of monthly stock returns. The sample consists of all open market purchases and sales by insiders in 284 firms with blackout periods and 119 firms without blackout periods from the period January 1, 1992 to June 30, 1997. T-statistics in parentheses.
DependentVariable/ Log of Total Assets -0.0001
(-0.679) Bank or Thrift 0.0093***
(11.89)
TABLE 8
Pooled time-series cross-section regressions of the number of trades per firm-day and dollar volume of trades per firm-day on indicator variables for various insider trading policies and interaction terms with the one month period following earnings announcements. The variables include an indicator equal to one for the one month period following an earnings announcement, an indicator if the firm has blackout period(s), an indicator for blackout days, and interaction terms between the post-earnings period indicator and the trading policy variables. The sample consists of all open market purchases and sales by insiders in 284 firms with blackout periods and 119 firms without blackout periods from the period January 1, 1992 to June 30, 1997. T-statistics in parentheses.
DependentVariable/
Independent Variable
Number of Trades Dollar Volume
Intercept 0.0432***
Firm Has Blackout Period 0.0053**
(2.11)
TABLE 9
Weighted least squares regressions of percentage spread and price impact measures on an indicator variable equal to one if the day is a blackout day and equal to zero if the day is an allowed trading day, the log of market capitalization from CRSP, the log of daily share volume from CRSP, the inverse of the average intraday trade price measured from bid-ask midpoints, the standard deviation of intraday price measured from bid-ask midpoints, an indicator variable equal to one for NASDAQ firms and zero otherwise, and an indicator variable equal to one for the day of and two days following earnings announcements and zero otherwise. The spread measures are calculated as daily averages based on trades for each firm day and the weights used in the regressions are the number of observations used to compute each firm-day average. The sample consists of 284 firms with blackout periods and 119 firms without blackout periods from January 1 to December 31 of 1996 comprising 82,872 firm-day observations. T-statistics in parentheses.
Dependent Variable/ Indicator For Trading Policy But No
Blackout Period(s) Indicator For Trading Policy With
Blackout Period(s) Indicator For Blackout Day -0.021***
(-18.13)
-0.026***
(-3.99)
0.007 (1.09) Log Of Market Capitalization -0.028***
(-59.63)
-0.060***
(-22.83)
0.030***
(11.43)
Log Of Daily Share Volume -0.007***
(-16.48)
0.032***
(13.42)
-0.041***
(-17.28) Inverse of Average Intraday Trade Price 5.392***
(315.04)
2.756***
(28.98)
3.880***
(40.35) Intraday Standard Deviation Of Price 0.114***
(73.91)
0.381***
(44.73)
-0.260***
(-30.27) Indicator For NASDAQ Firms 0.197***
(113.41)
0.118***
(12.29)
0.097***
(9.94) Indicator For Day Of And Two Days
Following Earnings Announcement
TABLE 10
Regressions of the dollar volume weighted cumulative abnormal returns to insiders for open market purchases and sales on indicator variables for various insider trading policies. The abnormal return is cumulated over the 120 days following the transaction and is measured relative to a benchmark portfolio based on five groups of book-to-market equity and five size groups. For sales the abnormal return is multiplied by negative one. The insider trading policy variables include an indicator equal to one if the firm has policies on insider trading, an indicator equal to one if the firm has policies that include blackout periods, and an indicator equal to one if the trade was made during a blackout day. The sample consists of all open market purchases and sales by insiders in 284 firms with blackout periods and 119 firms without blackout periods from the period January 1, 1992 to June 30, 1997 which have nonmissing values of book-to-market equity and size.
Model 1 Model 2 Model 3 Model 4 Model 5 Model 6
Variable Purchases
Indicator For Blackout Day -0.0119**
(-2.129)