### Table 1. Distribution Statistics for the Degree of Insider Trading, z

"... In PAGE 3: ...Using our iterative method, values of z were calculated separately for each race as described in the appendix, and some distribution statistics for z are reported in Table1 . The mean value is very similar to the average value reported by Shin for his sample of horse races, but note that there is much variation between races within the sample, the derived degree of insider trading ranging from less than 1% to 17%.... ..."

### TABLE 6 INSIDER TRADING: THE RELATIONSHIP OF THE SHIN MEASURE TO THOSE BASED ON PLUNGERS. (971 HORSE RACES)

### TABLE 7 INSIDER TRADING: THE RELATIONSHIP OF THE SHIN MEASURE TO THOSE BASED ON PLUNGERS. (1000 GREYHOUND RACES)

### Table 6 Introduction and Enforcement of the Insider Trading Law The first column reports the year legislation to curb insider trading was introduced, the second one the year the first case of prosecution took place. No means there has been no case of prosecution. Source: Bhattacharya and Daouk (2002).

"... In PAGE 10: ...ress.4 This is not simple cheap talk. Disclosure standards have improved throughout the Continent as have laws to protect minority shareholders. For example, in Table6 , we report the year of introduction of a law to prosecute insider trading and the year this law was applied for the first time. Before 1980 no EU country, except France and Sweden, had an anti insider trading law.... ..."

### Table 2.1: Cross table for white-collar criminal context. C = Conspiracy, F = Fraud, GL = Grand Larceny, IT = Insider Trading, OJ = Obstruction of Justice, and P = Perjury.

2005

### Table III Effect of Insider Trading Laws on Liquidity The pooled regressions are based on monthly data from 1969:12-1998:12. The dependent variable is quot;liq quot;, and it is the natural logarithm of the ratio of volume to market capitalization. The independent variables are the insider trading variables. They are coded as follows. The indicator variable quot;IT laws quot; changes from 0 to 1 in the year after the insider trading laws are instituted. The indicator variable quot;IT enforcement quot; changes from 0 to 1 in the year after the first prosecution was recorded. The equity data are from Morgan Stanley Capital International. p-values in brackets were computed using the procedures suggested by Newey and West (1987).

### Table IV Effect of Insider Trading Laws on the Cost of Equity (Unadjusted) The pooled regressions are based on monthly data from 1969:12-1998:12. The dependent variable is quot;rawret quot;. It is defined as follows. quot;Rawret quot; is raw returns, and is computed as continuously compounded returns. The independent variables are the insider trading variables. They are coded as follows. The indicator variable quot;IT laws quot; changes from 0 to 1 in the year after the insider trading laws are instituted. The indicator variable quot;IT enforcement quot; changes from 0 to 1 in the year after the first prosecution was recorded. The equity data are from Morgan Stanley Capital International. p-values in brackets were computed using the procedures suggested by Newey and West (1987).

### Table V Effect of Insider Trading Laws on the Cost of Equity (Implicitly Adjusted) The pooled regressions are based on monthly data from 1969:12-1998:12. The dependent variable is quot;div quot;. It is defined as follows. It is computed as the sum of the dividend yield forecast and the growth rate of the dividend yield (dividend based measure). The independent variables are the insider trading variables. They are coded as follows. The indicator variable quot;IT laws quot; changes from 0 to 1 in the year after the insider trading laws are instituted. The indicator variable quot;IT enforcement quot; changes from 0 to 1 in the year after the first prosecution was recorded. The equity data are from Morgan Stanley Capital International. p-values in brackets were compute using the procedures suggested by Newey and West (1987).

### Table 4: Proportion of observed to expected trades of assets that correspond to the correct solution

"... In PAGE 14: ... Comparison of insiders and outsiders: The impact of insiders on market behavior be- comes most evident if one analyzes the nature and frequency of the transactions (and other market activities) involving the informed insiders and the remaining outsiders . Table4 summarizes the frequency of trades involving assets that are part of the solution (I and IV), as a function of the information status of the seller and of the buyer. The ta-... In PAGE 15: ... Third, the observed/expected ratio of the number of trades diminishes with the number of insiders on the market, except for the case when insiders sell assets to outsiders. [ Table4 about here.] Finally, we investigate whether insiders managed to exploit their informational advan- tage by earning higher dividends than the uninformed subjects.... ..."

### Table VIII Market Timing and Overconfidence The table presents results about market timing. The dependent variable in columns (1) to (3) is equity market timing by firms. The dependent variable in columns (4) to (6) is market timing by insiders. Timing variables are computed for horizons of 6 months or 12 months (indicated in regression headers). All the regressions are OLS regressions. Accounting data are provided by from Compustat. Returns data is provided by CRSP. Insider trading data are provided by Thomson TFN. Variable definitions are provided in the Appendix. *, **, *** denote two-tailed significance at the 10%, 5%, and 1% level respectively. All regressions have intercepts that are not presented. Standard errors are clustered at the 2-digit SIC level.

2006