COMMENTS OF THE AMERICAN BAR ASSOCIATION SECTION…

             COMMENTS OF THE AMERICAN BAR ASSOCIATION
       SECTION OF ANTITRUST LAW, SECTION OF INTERNATIONAL LAW,
       SECTION OF BUSINESS LAW, AND SECTION OF CRIMINAL JUSTICE
          ON THE DRAFT INFORMATION BULLETIN ON SENTENCING
              AND LENIENCY IN CARTEL CASES ISSUED BY THE
                    COMPETITION BUREAU OF CANADA

                                          July 14, 2008

        The Section of Antitrust Law, the Section of International Law, the Section of Business
Law, and the Section of Criminal Justice of the American Bar Association (ABA)(collectively,
the Sections) appreciate the opportunity to present their views on the Draft Information Bulletin
on Sentencing and Leniency in Cartel Cases (Bulletin) published for public consultation by the
Competition Bureau of Canada. The views expressed herein are presented on behalf of the
Sections. They have not been approved by the House of Delegates or the Board of Governors of
the American Bar Association and, accordingly, should not be construed as representing the
policy of the ABA.

                                   EXECUTIVE SUMMARY

        The Sections fully support the Bureau's effort to establish a transparent and consistent
process for sentencing and leniency in cartel cases. The fundamental requirement of any
effective settlement program is that it provides sufficient incentives to encourage the parties to
settle. To that end, the Sections respectfully suggest four criteria that are essential for an
effective settlement program and the Sections encourage the Bureau to establish a program that
meets these four criteria. The sentencing program must have procedural transparency, generous
or significant settlement discounts, legal certainty and the protection of confidentiality and
privilege.

        The Bureau's role in sentencing and leniency is to make recommendations to the Director
of Public Prosecutions (DPP), and the DPP then makes recommendations to the Court, which has
the final sentencing power. Accordingly, the Bureau cannot guarantee specific sentencing
outcomes. To the extent that there is inconsistency between recommendations made by the
Bureau and by the DPP, it will reduce the incentives on cartel participants to come forward and
seek leniency. The Sections appreciate the organizational structures which give rise to the
division of responsibilities between the respective organizations, but, to the extent that there can
be explicit and transparent cooperation arrangements or understandings between the Bureau and
the DPP on sentencing and leniency issues in cartel cases, the Sections would encourage such
cooperation. Such cooperation should be achieved by the Bureau and DPP working closely and
seamlessly together to achieve transparency and legal certainty.

       The sentencing considerations that the Bureau describes in the Bulletin are similar to the
considerations for calculating a sentence in other jurisdictions, including the United States. The
Bureau's use of volume of commerce to demonstrate economic harm is appropriate. The
proposal to calculate the overcharge through use of a proxy of 20% of the volume of commerce,
however, is not based on credible empirical evidence and has been questioned by the U.S.


                                                -1-
Antitrust Modernization Commission. The Sections encourage the Bureau to expressly permit
defendants and leniency applicants, as well as the Bureau itself, to rebut any presumption based
on the proxy, by establishing through evidence the overcharge multiplier based on the actual gain
or loss from the cartel behavior, where such evidence is practically available. That would be
fair, proportional and transparent.

       The Bureau articulates aggravating and mitigating factors that will shape its sentencing
recommendations. The factors are general and subjective and the Sections encourage the Bureau
to provide some form of scoring that allows more transparency and legal certainty.

        The Sections approve of the Bureau's recognition of the continuing use of prohibition
orders as an available remedy in cartel cases, which are particularly appropriate in cases where
there has been no admission or finding of guilt. The Sections encourage the Bureau to include
language making it clear that a prohibition order is available to leniency applicants without a
guilty plea.

        The sentencing process requires a leniency applicant to enter into an agreement to plead
guilty to the offense and to make explicit (and public) admissions about the cartel. Consistent
with the objective of transparency, it would be helpful if detail on this process, the nature of the
agreement and the requirement for admissions were set out clearly in the Bulletin. Inclusion in
the Bulletin of the circumstances in which the Bureau will recommend prosecution of individuals
and a lenient sentence would be of great assistance to potential leniency applicants.

        Transparency, legal certainty, generous settlement discounts and confidentiality are
necessary criteria for making the leniency program attractive to potential defendants, corporate
and individual. In that context, the Sections believe that the weight given by the Bureau to two
of the primary leniency factors, namely, the timeliness of cooperation and the value of the
evidence, must be clearly articulated and should not result in the inconsistent application of the
Bureau's leniency program.

       Similarly, the Sections respectfully submit that the effect that multiple leniency
applications would have on the leniency process should be more clearly expressed to give
prospective leniency applicants more clarity and predictability of outcome. A transparent and
predictable leniency process is at least as important as clearly articulated leniency considerations.

         Finally, the Bureau has properly recognized the need for confidential treatment of the identities of
leniency applicants and the information they provide. The Sections believe, however, that the exceptions
to confidentiality, in both the Bulletin and an earlier informational bulletin, are too broad, and may
constitute a substantial disincentive to seeking leniency in numerous cases. The Sections recommend that
the Bureau expressly provide that the confidentiality provisions of the Bulletin supersede those of that
earlier informational bulletin, and otherwise narrowing and clarifying the scope of exceptions to
confidentiality.




                                                    -2-
                                                    COMMENTS

Introduction

        The Sections welcome the Bureau's proposal on sentencing and settlements as strong and
positive. The Sections, individually or jointly, have previously commented on matters relating to
the investigation and criminal prosecution of cartel conduct in the United States and Canada, and
issues in many jurisdictions relating to international cartel enforcement, including sentencing and
settlement issues. 1 In the sentencing and settlement area, the Sections have the benefit of
substantial experience in several jurisdictions where they have observed and practiced in a
variety of settlement and sentencing programs. The Sections commend the Competition Bureau
of Canada for seeking to bring clarity and transparency to the sentencing process in criminal
competition law cases, particularly as it complements the Bureau's Immunity Policy. The
Sections believe that, because of the increasingly serious penalties in cartel cases around the
globe, enforcement agencies must articulate their sentencing policies fully and fairly to the
public, the prosecutors and the courts. The ultimate aim is to provide sufficient information for
the parties to be able to assess whether settlement is advisable, as well as incentives to settle
where liability is clear, rather than attempting to litigate each and every potential matter. The
Sections encourage the Bureau to evaluate its proposal utilizing four criteria that will make
settlement attractive and compelling to defendants where settlement can produce a just result:
procedural transparency, generous settlement discounts, legal certainty and protection of
confidentiality. 2 Our comments will review the program by considering these four criteria.

I.      The Decision-Making Role of the DPP

        While the sentencing proposal described in the Bulletin is a program and policy of the
Bureau, which makes recommendations to the DPP of prosecutions and sentences, it is critical
for transparency and legal certainty purposes that the DPP and the Bureau agree to common
principles on which such recommendations and decisions will be based, understanding that the
Court will ultimately determine the actual sentence based on the DPP's agreement with the
defendant. The DPP's official approach to the plea process is set out in Chapter 20 of the
Federal Prosecution Service Desk Book (the Deskbook). That document establishes the general
practice of the DPP in relation to the wide range of criminal matters it considers but it contains
nothing specific in relation to cartels. The Sections note with approval, however, that the
Deskbook does explicitly recognize that "because of the benefits that flow to the administration


1
  With respect to Canada, see Submission of the American Bar Association Sections of Antitrust Law and
International Law in Response to the Canadian Competition Bureau's Request for Comments on its Upgraded Draft
Bulletin on Corporate Compliance Program (May 2008), available at http://www.abanet.org/antittust/at-
comments/2008/05-08/comments-canadian-draft.pdf; and Comments of the ABA Section of Antitrust Law in
Response to the Canadian Competition Bureau Request for Public Comments Regarding Immunity Program Review
(May 2006), available at http://www.abanet.org/antitrust/at-comments/2006/05-06/com-canadian-leniency.pdf.
2
  Joint Comments of the American Bar Association Section of Antitrust Law and Section of International Law in
Response to the Commission of the European Communications' Request for Public Comments on the Draft EU
Settlement Procedures (December 2007), available at http://www.abanet.org/antitrust/at-comments/2007/02-
07/comments-EU-Draft-Settlement.pdf.



                                                     -3-
of justice from early guilty pleas, the Crown should make its best offer to the accused as soon as
practicable," and that "in Competition Act matters, because of the frequent involvement of
counsel during the investigative stage, discussions may take place before charges are laid." 3

        Predictability about the benefits of cooperation is particularly importance in cartel cases
because cartel behavior is inherently covert. The role of immunity and leniency applicants is
crucial in a way which is rarely the case in other criminal matters. Generally, the only witnesses
to the cartel violations are the participants themselves and the witnesses often need to explain
their words or the deeds that followed them. Accordingly, it would enhance the objectives of the
Bureau if there were an understanding or agreed upon course of conduct between the DPP and
the Bureau as to the principles that the DPP will adopt in deciding whether to accept a
recommendation of the Bureau. An alternative may be for the DPP to itself develop a position
specific to cartel conduct which is consistent with the Bureau's approach to sentencing and
leniency. At the very least, the Sections recommend that Chapter 20 of the Deskbook be
amended to provide a statement of support for the Bureau's leniency program. Such a statement
would not be at odds with the principles that generally inform Crown counsel's approach to
sentence negotiations. In that regard, Chapter 35 of the Deskbook which addresses Immunity
Agreements devotes a section of the Chapter specifically to Provisional Guarantees of Immunity
for Competition Act Offences. 4

         The Bulletin appears to separate the sentencing and settlement process of the Bureau
from that of the DPP. In practice, the Bureau and the DPP have worked together regularly on the
development and resolution of cases, from the interrogation of witnesses to settlement. The idea
that the Bureau and the DPP would work together to a resolution of the case is consistent with
the criteria that will provide incentives for settlement. There is a procedural transparency that is
apparent if the defendant is dealing with what appears to be one organization that is able to
assure the defendant how the second organization will approach sentencing (or how both
organizations will do so together), rather than having to negotiate and make its deal twice with
no legal certainty that both organizations are like minded and will approve the same settlement.
The generosity of the settlement discounts also appears more compelling to the defendant if it
feels it is dealing with one party and does not fear that the second agency, the DPP, will not take
the deal made with the Bureau and whittle it down. Finally, and importantly, the DPP is actually
involved in the settlement negotiations to which privilege and confidentiality is accorded. A
separate negotiation with the Bureau, which is not the determinative negotiation, may not enjoy
the safety of the settlement privilege, adding to the concerns the defendant will have in the civil
damage process in Canada, the United States and other jurisdictions. Providing a more unified
negotiation that has the Bureau and the DPP involved simultaneously and seamlessly, as exists in
cartel enforcement actions today, is the best opportunity to provide incentives to settlement in
these criminal cartel prosecutions.



3
    See Chapter 20, The Federal Prosecution Service Deskbook, The Department of Justice Canada, at footnote 12.
4
  The Sections note that the Deskbook has not yet been updated to account for changes to the Bureau's Immunity
Program introduced in October 2007.



                                                       -4-
II.     Sentencing Principles in Cartel Offenses

        The Bureau's articulation of the factors it will take into consideration at sentencing are
important steps in creating transparency and legal clarity. Whether these factors create or
establish generous settlement discounts that provide incentives to settle is the critical question.
The Sections will comment on the sentencing principles, looking both at the process used in the
United States for sentencing antitrust violations and looking more generally at the tension
between flexibility and transparency ­ a tension at the heart of effective implementation.

        A.       Economic Harm

        The problem of how to calculate a criminal fine based on the economic impact of the
offense has been a challenging issue in all jurisdictions that have sought to address it. While the
Sections appreciate that difficulties can arise in calculating the economic harm, the Sections
submit that sometimes these difficulties can be overstated. The Sections urge the Bureau to
calculate a sentence based on the actual economic harm in the case and to rely on presumptions
only where that approach is unworkable. If the Bureau's proposed estimation formula is used,
the two relevant factors are (1) the volume of commerce in Canada, and (2) the applicable
multiplier, or "overcharge factor."

        Volume of commerce: The Bureau states that "[i]n most cases, it is necessary to include
only direct sales in Canada," but "it may also be appropriate to include indirect sales." In
furtherance of the criteria of transparency, generous discounts and legal certainty, the Sections
recommend that more detailed guidance be provided on what sales are included in the volume of
commerce. The Bureau needs to define, simply and directly, how the volume of commerce is
calculated and whether only direct Canadian sales are included. If there are circumstances in
which the Bureau would seek to include more than direct Canadian sales in its volume of
commerce calculation, it is important to understand what those circumstances are, as well as
what the Bureau might consider appropriately included as "indirect sales." There do not appear
to be any principles for application of this indirect sales concept contained in the Bureau's
Bulletin. The Sections are also concerned that the inclusion of indirect sales may increase the
potential level of fine in a manner which would not necessarily be adopted by the Courts if the
matter were contested. The uncertainty engendered by that possibility would create a serious
disincentive to participation in the leniency program, and would be highly undesirable from a
policy perspective.

       Overcharge factor: The Bulletin states that "Numerous studies have estimated that the
amount of the `overcharge' resulting from cartel activity is in the order of at least 10 percent,"
and proposes 20 percent of the volume of commerce affected as "a proxy of the economic harm."
The Bureau thus proposes a presumption concerning the overcharge or severity of the crime. 5

5
  The Guidelines employ similar presumptions. The ABA Section of Antitrust Law ("ABA Antitrust Section") has
questioned, and continues to question, whether the current presumption in determining criminal fine levels is
empirically sound or good public policy. See Comments of the ABA Section of Antitrust Law In Response to the
Antitrust Modernization Commission's Request for Public Comment on Criminal Remedies (November 14, 2005),
available at http://www.abanet.org/antitrust/at-comments/2005/11-05/criminal remedies-comm.html); ABA
Antitrust Section, Comments on the Proposed Amendments to the United States Sentencing Guidelines for Antitrust


                                                     -5-
Of course, the presumption that all antitrust conspiracies result in the same level of harm could
be inequitable and disproportionate ­ in either direction ­ in any given case. The Sections
recommend, in the interest of fairness and proportionality, that defendants and parties seeking
leniency or to settle cartel allegations be expressly entitled to an opportunity to rebut the
presumption by providing evidence of the actual overcharges.

        The experience in the United States is relevant to the Bureau's proposal. Since 1991,
with the adoption of the U.S. Sentencing Guidelines (U.S.S.G. or Guidelines) for antitrust
offenses, the Guidelines have included a provision that volume of commerce would be the means
of determining a base fine and that 20 percent would be an appropriate proxy for actual
overcharge because the determination of an overcharge or harm was very difficult to establish
through evidence. The use of any irrebutable proxy has been subject to serious criticism in the
United States, resulting in April 2007 with the Antitrust Modernization Commission (AMC)
presenting its report and recommendations to the President and the Congress of the United
States. Among the AMC's recommendations were two relating to the 20 percent proxy:

             52.    Congress should encourage the Sentencing Commission to reevaluate and explain
                    the rationale for using 20 percent of the volume of commerce affected as a proxy
                    for actual harm, including both the assumption of an average overcharge of 10
                    percent of the amount of commerce affected and the difficulty of proving actual
                    gain or loss. 6

             53.    The Sentencing Commission should amend the Sentencing Guidelines to make
                    explicit that the 20 percent harm proxy (or any revised proxy) ­ used to calculate
                    the pecuniary gain or loss resulting from a violation ­ may be rebutted by proof
                    by a preponderance of the evidence that the actual amount of overcharge was
                    higher or lower, where the difference would materially change the base fine. 7

        The Section of Antitrust Law has also questioned any inflexible presumption in the
legislative and policy context, including before the AMC. 8 Empirical studies have found
significant variation in the minimum, maximum, median, mean, and range of cartel




Sentencing (March 2005), available at http://www.abanet.org/antitrust/comments/2005/03-05/ussg-05.html; and
ABA Antitrust Section, Comments on HR 1086: Increased Criminal Penalties, Leniency Detrebeling and the
Tunney Act Amendment (January 2004), available at http://www.abanet.org/antitrust/comments/2004/reports.html.
6
  Antitrust Modernization Commission Report and Recommendation (April 2007) at 19, 300-01, available at
http://govinfo.library.unt.edu/amc/report_recommendation/toc.htm.
7
    Id. at 19, 302-03.
8
 See Comments of the ABA Section of Antitrust Law In Response to the Antitrust Modernization Commission's
Request for Public Comment on Criminal Remedies (November 14, 2005), available at
http://www.abanet.org/antitrust/comments/2005/03-05/ussg-05.html.



                                                    -6-
overcharges. 9 Accordingly, any proxy or set presumption is likely to be a poor estimate of the
actual economic harm in any given case.

         As with the U.S.S.G., the Bureau's stated reason for employing a proxy calculation is that
"[i]n most cases it may be difficult to quantify and establish (to the criminal standard) the actual
economic harm resulting from cartel behavior." 10 The Sections understand that calculating the
actual harm with any degree of precision is often difficult or impossible, especially at the stage
of an investigation when settlements and plea agreements tend to be negotiated. While a proxy
may be useful in many cases where proof of the actual harm cannot be developed as a practical
matter, the Sections recommend that the Bureau explicitly provide for the right of defendants and
parties seeking to settle cartel investigations to rebut the application of the presumptive proxy
overcharge percentage by proof of actual harm. This is consistent with the alternative fine
statute in the U.S., and with the practice of the U.S. Department of Justice Antitrust Division in
those cases where such proof is available. In 1996, for example, in United States v. Archer
Daniels Midland Co., 11 the Antitrust Division of the U.S. Department of Justice calculated and
proved to the court the alternative maximum fines under 18 U.S.C. § 3571(d), which requires the
determination of the actual gain or loss resulting from the violation to increase the fine above the
statutory maximum. While the Division uses proxies for actual harm because such proof has not
been developed or is difficult to adduce, basing fines on the actual harm rather than any proxy
generally leads to a more equitable result. The AMC's recommendation that the 20 percent
proxy should be a rebuttable presumption opens the process to fairness and proportionality in
cases where the presumption is faulty. The concept of a generous settlement discount as an
incentive for settlement is greatly enhanced by permitting, where feasible, the imposition of fines
more consistent with the actual harm inflicted by the defendant than by an inflexible, "one size
fits all" proxy.


9
 Id. See, e.g., Margaret C. Levenstein & Valerie Y. Suslow, "What Determines Cartel Success," 44 J. Econ. Lit. 43,
80 (2002)(looking at over twenty cartels that inflated prices 0 percent to 100 percent, with a median, and mean,
overcharge of nearly 45 percent); Jonathan B. Baker, "The Case for Antitrust Enforcement," 17 J. Econ. Persp. 27,
28-29 (2003)(citing studies showing cartel overcharges from 6.5 percent to 32.5 percent); Richard A. Posner,
ANTITRUST LAW, 303-04 (2d ed. 2001)(twelve cartels with a median overcharge of nearly 40 percent and a mean of
50 percent); John M. Connor & Robert H. Lande, "How High do Cartels Raise Prices: Implications for Optimal
Cartel Fines," 80 Tulane L. Rev. 513 (2005)(over two dozen cartels with a median overcharge of about 25 percent
and a mean overcharge of about 40 percent); John M. Connor, "Price-Fixing Overcharges: Legal and Economic
Evidence," Purdue U. Working Paper (Jan. 10, 2005)(very large sample set, finding median overcharges of 18 to 32
percent and a mean of 49 percent); John M. Connor, "Private International Cartels: Effectiveness, Welfare, and
Anticartel Enforcement," Purdue U., Dept. of Ag. Econ., Staff Paper 03-12 (2003)(seventy cartels raised prices 4 to
100 percent, with a mean of 28 percent); OECD, Report on the Nature and Impact of Hard-Core Cartels and
Sanctions Against Cartels under National Competition Laws, Annex A (2003)(median overcharges of 33 percent);
Gregory J. Werden, "The Effect of Antitrust Policy on Consumer Welfare: What Crandall and Whinston Overlook"
(U.S. Dep't of Justice Antitrust Division, Discussion Paper, No. EAG 03-2 (Jan. 2003)(finding overcharges with a
mean of 21 percent and a median of 18 percent in over a dozen cartels); and Mark A. Cohen & David T. Scheffman,
"The Antitrust Sentencing Guideline: Is the Punishment Worth the Costs?," 27 Am. Crim. L. Rev. 331 (1989)(an
older study finding a median overcharge of about 8 to 14 percent, and a mean of 8 to 10 percent).
10
     Bulletin, at ¶ 28.
11
     Crim. No. 96-CR-00690 (N.D. Ill. Oct. 15, 1996).



                                                        -7-
        Minimum recommendations: The Bulletin further provides that, in principle, "the
Commissioner does not submit a fine recommendation to the DPP that represents less than 10
percent of the affected volume of commerce for a business organization, unless it would exceed
the statutory $10 million maximum per count." The Bureau provides no explanation for its
position. The Sections recommend that, if the circumstances of a particular case suggest that the
actual economic harm resulting from the alleged cartel conduct is an amount less than 10 percent
of the affected volume of commerce, that the Bureau should recommend a fine below that level,
consistent with fairness and proportionality and the generous settlement discount concept.

       B.      Aggravating and Mitigating Factors

        The aggravating and mitigating factors set forth in the Bulletin are reasonable and are
primarily the same factors used in U.S. sentencing. They are used to differentiate the penalties
for defendants for a particular offense based on the defendant's activities. Unlike the sentencing
system in the United States, the Bulletin does not assign offense level "points" to categories in
aggravation or mitigation. Instead, it appears to allow the Bureau wider latitude in its
recommendations. While the U.S. system is more specific in assigning points for bid rigging,
volume of commerce, role in the offense, obstruction of justice and acceptance of responsibility,
it does create some parameters that allow a certain degree of transparency and proportionality.
The Bureau's more flexible and numerous factors allow much more latitude but tend to be far
more subjective, with the potential consequence that they will be less predictable, thereby
compromising both transparency and legal certainty. For that reason, the Bureau should
articulate these factors carefully, and provide guidelines to ensure that it applies apply these
factors consistently.

       The aggravating factor of duration of the conspiracy may also be duplicative of the
volume of commerce factor since length of the conspiracy is reflected directly in increased
volume of commerce. The Bureau may wish to consider whether to utilize an additional
aggravating factor for duration in light of this duplication.

        The ultimate question is how will these aggravating and mitigation factors be calculated
into an equitable sentence, consistent with transparency, legal certainty and generous settlement
discounts. The Sections submit that some concrete "scoring" similar to that used in the U.S.S.G.
will add to the transparency and proportionality to the leniency process. This is an area where
guidelines prepared by the Bureau and embraced, directly or indirectly, by the DPP, may be
effective in articulating the sentence and explaining clearly the benefit of early cooperation.

III    Prohibition Orders

        Section III of the Bulletin describes prohibition orders as a remedy available to the
Commissioner for cartel offenses. The Sections fully support the Bureau in its ongoing use of
prohibition orders, particularly pursuant to subsection 34(2) in cases where there has not been an
admission or finding of guilt. For greater transparency, the Sections recommend that the Bureau
provide further guidance on the circumstances in which a prohibition order on consent without a
guilty plea would be appropriate. In addition, because the Bulletin addresses both sentencing
and leniency in cartel cases, the Bureau may want to consider including language that explicitly



                                               -8-
states that a prohibition order without a guilty plea under subsection 34(2) would be available to
leniency applicants.

IV.         Sentencing Recommendations: The Process

            A.        General Considerations

        In the Bulletin, the Bureau briefly outlines its approach on recommending particular
terms of punishment for organizations and individuals. As noted above, the Bureau does not
have the final say on ultimate penalties imposed; the Bureau's views ultimately remain
recommendations. Because the Bureau's cases can be brought to a Canadian court only by the
DPP, the DPP has the ultimate discretion in this area (although it is bound to consider the
agency's views in taking the case forward). However, the DPP's position (even if agreed to by
the defendant) will not bind the trial court in view of the traditionally broad level of discretion
granted Canadian courts in the sentencing of offenders.

        These features, which are important components of the criminal antitrust process in
Canada, point to the need for the Bureau to outline and explain, in general terms, the nature of
the criminal guilty plea process. The Bulletin makes only passing reference to the fact that a
leniency applicant will be required to enter into "an agreement to plead guilty to the
offence(s)." 12 A more detailed description of the steps of the plea process will be of assistance
to familiarize the defendants (particularly those unfamiliar with the criminal process) with the
requirements and potential implications of seeking leniency in Canada. This description would
also add to the procedural transparency and the legal certainty criteria that provide strong
incentives to settlements.

        A few notable features of the Canadian process should be identified, including the fact
that defendants entering a plea of guilty in Canada will be expected to come to agreement on an
explicit (and public) statement of admissions which will outline the nature of the conspiracy, the
defendant's role in it, and such detail as the overall volume of commerce in the relevant product
by the defendant over the period of the conspiracy.

        Plea practices differ between Canada and the United States, the most notable distinction
being the degree of predictability a defendant can expect in the plea process. A Canadian court
has full discretion to accept or reject a sentencing recommendation, whereas defendants
resolving their cases in the U.S. can often be considerably more secure in knowing that they may
withdraw from the plea process should a sentencing court not agree to impose the jointly
recommended penalty. While outright rejection of jointly submitted sentencing
recommendations is rare in Canada, the appeal cases indicate that trial courts sometimes will not
align themselves with the positions of the parties, and that the only remedy from a non-
conforming sentence imposed by a trial court will be an appeal. 13


12
     Bulletin, at ¶ 87.
13
     See R v. Moser [2002], 163 C.C.C. (3d) 286 (Ont. C.A.).



                                                         -9-
            B.        Sentencing Recommendation for Individuals

        In Canada, individuals accused under the conspiracy and bid-rigging provisions of the
Act are liable to fines or a maximum of five years imprisonment. The Bulletin outlines
sentencing recommendations in Part B(V), stating that "there are a number of considerations the
Commissioner applies in developing sentencing recommendations for individuals ... [including]
the degree to which the individual personally profited from the offence (including salary,
bonuses and career enhancement), whether the individual has been sanctioned for participating in
other cartels or in the same cartel in another jurisdiction, and whether the individual has been
punished in other ways (for example, loss of employment)." 14 Consideration will be given as to
whether the individual has the ability to pay. 15

        As discussed in Section II.B., supra, the Bulletin sets out the aggravating and/or
mitigating factors that will play a role in any sentencing process. Based on past experience,
where mitigating factors have more persuasive influence, the Commissioner is likely to
recommend to the DPP a fine and prohibition order as an appropriate sentence for an
individual. 16 However, the Commissioner is more likely to recommend imprisonment for the
most serious violations where the presence of one or more aggravating factors suggests that a
more severe sentence for the potential accused is appropriate. 17

        While the sentencing guidance in the Bulletin provides a basic description of the factors,
it remains unclear under what circumstances the Bureau will actually prosecute individuals
separately from the company which has been involved in an offense and has obtained leniency.
Given the limited Canadian jurisprudence in this area 18 as well as the absence of formal
sentencing guidelines, there are no Canadian benchmarks to provide guidance for potential
leniency applicants. While many of the aggravating circumstances 19 in the Bulletin could form
the basis for a prosecution policy, the Bureau does not speak to when the Bureau and the DPP
will carve individuals out from corporate plea arrangements and prosecute them individually.
The current lack of clarity may unnecessarily complicate efforts by the parties to provide the


14
     Bulletin, at ¶ 56.
15
     Id. at ¶ 57.
16
     Id. at ¶ 59.
17
     Id. at ¶ 58.
18
   The only reported cartel case to date in which an individual was sentenced to incarceration was in the matter of
Russell Cosburn, a former Vice-President of Chinook Group Limited, who was sentenced in September of 1999 to a
term of 9 months' imprisonment (to be served in the community) for his role in a global price-fixing and market
allocation cartel involving chlorine chloride. See the Bureau's report of the case, available at
http://www.competitionbureau.gc.ca/epic/site/cb-bc.nsf/en/00608e.html.
19
     Bulletin, at ¶¶ 37-45.



                                                      - 10 -
necessary full and frank disclosure that is a primary goal of the leniency program. Development
of an accompanying prosecution policy for carve-outs of individuals in cartel cases will allow
both corporations and associated individuals to know where they stand, lead to greater
transparency and certainty in the process, and generally enhance the effectiveness of the leniency
process in Canada.

        The Bulletin states that in making sentencing recommendations for both organizations
and individuals, the Bureau and the DPP will rely on precedential case law involving similar
facts and offenses. 20 This suggests that these would not be a marked change from the Bureau's
approach to date, which has predominantly featured fines rather than incarceration as the
approach to punishing individuals in criminal cartel cases.

        Canada has proceeded in a distinctly different direction from the United States where
individual incarceration in antitrust maters is a cornerstone of the Division's policy. 21 Other
jurisdictions have taken steps towards punishment of individuals for antitrust violations. 22
Different approaches by different jurisdictions is, of course, appropriate. In its third report on the
implementation of the Council recommendation concerning hard core cartels, the OECD
observed that "... each country must determine its own `right' mix of sanctions that has the most
effective deterrent effect against cartels." 23

        In the 1995 amendments to Canada's Criminal Code, judges were directed to consider all
available sanctions that are reasonable in the circumstances before imposing imprisonment. 24
The Bulletin contains a mix of sentencing proposals that should not, in practice, violate this
principle in imposing appropriate penalties for criminal cartel cases.

V.          Leniency Considerations in Sentencing Recommendations

        The Sections believe that the Bulletin would benefit from greater clarity if it addressed, in
greater detail, the relationship between leniency considerations and sentencing
recommendations. For example, Part II of the Bulletin describes cooperation with authorities as

20
     Id. at ¶ 56.
21
  Indeed, defendants prosecuted by the Division in 2007 were sentenced to serve 31,391 jail days, representing the
highest number imposed in any given year, and more than doubling the previous record from fiscal 2005. Notably,
87% of individual defendants charged by the Division received jail time. "Recent Developments, Trends, and
Milestones in the Antitrust Division's Criminal Enforcement Program," Scott D. Hammond, Deputy Assistant
Attorney General for Criminal Enforcement Antitrust Division U.S. Department of Justice, March 26, 2008, page 1,
available at http://www.usdoj.gov/atr/public/speeches/232716.htm.
22
  Three executives were recently sentenced to up to three years of incarceration for their participation in an
international cartel in the global marine hose industry, under a plea arrangement with the U.K. Office of Fair
Trading (OFT). See OFT Press Release (June 11, 2008), available at http://www.oft.gov.uk/news/press/2008/72-08.
23
 OECD Policy Roundtable ­ Cartel Sanctions Against Individuals, Executive Summary, OECD Secretariat,
DAF/:COMP (2004) 39, at 7, available at http://www.oecd.org/dataoecd/61/46/34306028.pdf.
24
     See Criminal Code Section 718.2(e).



                                                      - 11 -
a mitigating factor to be considered by the Bureau in its sentencing assessment and
recommendations. Full cooperation with the authorities is also an important condition for
eligibility for lenient treatment. The Bulletin does not explain whether a party's cooperation will
be considered by the Bureau to reduce the otherwise applicable sentence and whether that
sentence will again be reduced by 50 percent for the "first-in" leniency applicant or by 30
percent for subsequent leniency applicants, or whether the 50 percent or 30 percent reduction for
applicants will be a reduction of a sentence that has not already been reduced for mitigating
factors.

         There are several potential situations that the Sections recommend the Bureau consider
exploring in greater detail, perhaps in the form of frequently asked questions about the leniency
program. For example, will the Bureau be prepared to recommend leniency to a corporation
whose director, officer or employee is the immunity applicant? Similarly, will the Bureau be
prepared to recommend leniency to a parent corporation whose Canadian subsidiary has
independently sought immunity for a section 46 offense? How will the percentage reductions be
calculated? Will they be explained and illustrated? Given that the ultimate aim of a leniency
program should be to provide sufficient incentives for the parties to settle rather than attempting
to litigate each and every potential matter, the Sections submit that leniency should be granted
liberally to provide incentives for parties to accept responsibility, cooperate and settle these
matters.

        A significant feature of the leniency system will be the extent to which the Bureau needs
additional evidence when the leniency candidate comes forward to settle. The "value added"
that the candidate brings forth will likely affect the size of the discount that it ultimately receives.
The Bureau needs to be very clear in articulating the value of the cooperation lest a candidate
believe that it is eligible for a bigger discount simply because it came in early but provided only
cumulative evidence (although the Bureau should bear in mind the usefulness of corroboration
from two or more sources). To make this process most successful, the Bureau will create a
"prisoners' dilemma" that should define the incentives for parties to cooperate fully. This must
be transparent and certain, thus the clarity of the Bulletin is a critical factor to creating the race to
the Bureau with "value added" cooperation.

VI.     The Leniency Process

       As recognized by the Bureau, "lenient treatment" is not a novel concept under Canadian
law. The Bureau has recommended lenient treatment in the past for cartel participants who have
cooperated with Bureau investigations. However, to date, the Bureau's approach to leniency has
been case specific and sometimes inconsistent. The Sections commend the Bureau for proposing
a more formal leniency policy. A transparent and predictable leniency program will complement
the Bureau's Immunity Program and support the effective and efficient enforcement of the Act.

         The Bureau's approach to the incentives to settlement is completely consistent with the
Commissioner's statement that parties are more likely to come forward and cooperate (rather
than litigate) where they are fully aware of sentencing and leniency considerations and are
confident that the Bureau will follow them in its recommendations to the DPP.




                                                 - 12 -
        Transparency, legal certainty, generous settlement discounts and confidentiality are
necessary criteria for making the leniency program attractive to potential defendants, corporate
and individual. In that context, the Sections respectfully submit that the weight given by the
Bureau to two of the primary leniency factors, namely, the timeliness of cooperation and the
value of the evidence, should be more clearly articulated to prevent the inconsistent application
of the Bureau's leniency program.

       Similarly, the Sections submit that the effect that multiple leniency applications would
have on the steps in the leniency process should be more clearly expressed in order to give
prospective leniency applicants more clarity and predictability of outcome. A transparent and
predictable leniency process is equally as important as clearly articulated leniency
considerations.

        The Bulletin sets out five steps to the leniency process. In general, these steps mirror the
steps in the Immunity Program:

               Step 1 ­ Contacting the Bureau

               Step 2 ­ The Proffer

               Step 3 ­ Conditional Recommendation of Leniency to the DPP

               Step 4 ­ Full and Frank Disclosure

               Step 5 ­ Sentencing Recommendation to the DPP

        The Bulletin states that the Bureau will make a preliminary assessment of the potential
contribution of the applicant to the investigation and the range of a possible leniency
recommendation as soon as the proffer is made. The Bulletin specifies that, at the proffer step in
the process, the Bureau will need to know with detail and certainty what evidence or testimony a
potential witness can give, the nature of any records the applicant can provide and how probative
the evidence is likely to be. The Bureau may request, as with immunity applications, an
interview with witnesses or an opportunity to view certain documents.

        As with immunity applications, the Bulletin states that the "proffer" should typically be
made within 30 days of initial contact. It is foreseeable, especially in situations where immunity
has been granted and a search or "dawn raid" has been conducted, that multiple parties will come
forward within a short period of time seeking lenient treatment. The Bureau needs to clarify
whether the Bureau's Leniency Program is a "race for the door" as it is with the Bureau's
Immunity Program or how the Bureau will weigh the value of the evidence in making its
leniency recommendation. The Bureau does not discuss what effect a second-in leniency
applicant will have on a first-in leniency applicant if the second-in proffers evidence of greater
probative value. That is an issue the Bureau will need to address to provide transparency and
legal certainty to all potential leniency applicants. The process of how the Bureau will work
with the DPP on steps 3 and 5 is also a fundamental issue that requires full explanation ­ and
certainty.



                                               - 13 -
         As the Section of Antitrust Law has previously recommended, "[t]he timeliness of a
party's cooperation should be the principal criterion considered by the Bureau, as it complements
the `first-in' nature of the Immunity Program." "The guiding principle in determining incentives
should be to encourage and reward parties for coming forward early," 25 thus, the Sections
recommend that the Bureau clarify that only the first in leniency applicant is entitled to a 50
percent reduction in its fine regardless of the probative value of the evidence of subsequent
leniency applicants.

         The Sections are of the view that the Bulletin would greatly benefit from further clarity as
to the timing of the proffer and the weight given to the value of the evidence at the proffer stage
in the process in the context of multiple leniency applications. In this regard, the Sections
recommend that the Bureau consider publishing responses to frequently asked questions about
the Leniency Program similar to those published for the Bureau's Immunity Program.

        Once the Bureau has assessed the information proffered by the leniency applicant, the
third step in the process is the making of a conditional leniency recommendation by the Bureau
to the DPP. According to the Bulletin, once the DPP has received the Bureau's recommendation
the DPP will decide whether to consider lenient treatment and, if so, to what extent. The Bulletin
does not however indicate what, if anything, is communicated by the DPP to the leniency
applicant at this stage of the process and before the leniency applicant is required to provide full,
frank and truthful disclosure. The Sections strongly recommend that the DPP provide the
leniency applicant with an agreement or letter or commitment that lenient treatment will be
provided conditioned on ongoing cooperation. Given the prosecutorial discretion that exists in
the Canadian process, such a commitment from the DPP prior to the full disclosure stage of the
immunity process will provide a certain degree of comfort to leniency applicants. The Sections
note that Chapter 20 of the Federal Prosecution Service Deskbook provides that "crown counsel
who conduct sentence negotiations shall have full authority to enter into binding agreements."

        With respect to the fourth step in the process, full and frank disclosure, the Sections
recommend that, for greater certainty, the Bureau address what happens to directors, officers or
employees of a leniency applicant who have cooperated with the Bureau's investigation if their
employer ceases to cooperate or fails to comply with its leniency obligations. The analogous
issue is addressed in the Bureau's Immunity Program Responses to Frequently Asked Questions
(FAQs). 26




25
  Comments of the ABA Section of Antitrust Law in Response to the Canadian Competition Bureau Request for
Public Comments Regarding Immunity Program Review (May 2006), available at
http://www.abanet.org/antitrust/at-comments/2006/05-06/com-canadian-leniency.pdf.
26
  The FAQs are available at http://www.competitionbureau.gc.ca/epic/site/cb-bc.nsf/en/02482e.html. In response to
question 32, the Bureau responds: "Revocation of immunity will only affect the party that does not co-operate or
that otherwise fails to comply with the Immunity Program requirements. A company's immunity can be revoked
while its co-operating directors, officers, employees, or agents who were covered retain their protection. Likewise,
an individual's immunity can be revoked while the individual's employer company remains covered."



                                                       - 14 -
        The Bulletin should also consider the general procedural due process implications of a
"failed" leniency application. While the Bureau indicates that the full and frank disclosure
process is to be "without prejudice," both direct and indirect prosecutorial use of evidence or
information derived from the unsuccessful leniency applicant in an indictment against it should
be absolutely barred. Such a direction would not only enhance procedural clarity for potential
applicants, but contribute to the overall integrity of the Bureau's cartel investigations and
regulation process.

VII.        Confidentiality

        The Sections commend the Bureau for its recommendation, in the Bulletin, for
confidential treatment of the identify of a party requesting leniency and the information provided
by the party, except where the identity or the information has already been made public, or where
disclosure is required by law, to a Canadian law enforcement agency, for the purposes of the
administration or enforcement of the Competition Act, or authorized by a party or necessary to
prevent the commission of a serious criminal offense. 27 The Sections further note with approval
that Section 29, subsection 10(3) of the Competition Act requires that all inquiries be conducted
in private, and that, in the Information Bulletin on the Communication of Confidential
Information Under the Competition Act (2007)(the Confidentiality Bulletin), the Commissioner
of Competition and the Bureau have committed, in normal circumstances, to commenting
publicly on the existence of an inquiry or examination only if it has become public through
another source. 28 The recognition under Canadian law of a settlement privilege ­ the so-called
"without prejudice" rule ­ that protects from disclosure settlement communications, 29 is another
laudatory protection of confidentiality that, in the context of leniency applications, will provide a
meaningful incentive to potential applicants.

        These protections seem possibly inconsistent, however, with the statement in the Bulletin
that "information and evidence gathered in the course of the Leniency Program will be used by
the Bureau to advance its investigation of the cartel," with no assurance of any effort to preserve
the confidentiality of the information or its source. 30 Moreover, the Confidentiality Bulletin
provides for communication of confidential information in numerous other circumstances,
including when: (1) "eliciting from market participants, such as customers, suppliers or
competitors, that may be used as evidence to determine whether the Bureau's or a third party's
assessment of a matter is accurate" (though, in such circumstances, the Bureau takes "care to
refrain from, or to minimize, the communication of confidential information" and "[s]uch
communication will only occur if it is not otherwise reasonably possible to obtain necessary
information from these third parties"); (2) "obtaining an opinion or analysis by an industry, legal,

27
     Bulletin, at ¶¶ 88-89.
28
  Confidentiality Bulletin, ¶ 3.3, available at http://www.competitionbureau.gc.ca/epic/site/cb-bc.nsf/vwapj/info-
bulletin-confidential-info-e.pdf/$FILE/info-bulletin-confidential-info-e.pdf.
29
     John Sopinka et al., THE LAW OF EVIDENCE IN CANADA, at 817 (2d ed. 1999).
30
     Bulletin, at ¶ 84.



                                                       - 15 -
economic or other expert on some or "all aspects of the matter in question"; (3) "coordinating
enforcement actions with foreign law enforcement authorities"; and (4) "when assessing the
accuracy or the evidentiary value of information"; as well as during proceedings before the
courts or the Competition Tribunal. 31

       While the Sections applaud the establishment of a general rule of preservation of
confidentiality, the Confidentiality Bulletin appears to provide fewer and narrower exceptions to
the general rule protecting confidentiality of information than does the Bulletin. The Sections
therefore recommend that the Bureau expressly provide that, where inconsistent, the
confidentiality provisions of the Bulletin supersede those of the Confidentiality Bulletin, and
otherwise consider strengthening the assurances of confidential treatment of information
provided by leniency applicants. 32

Conclusion

        The Bureau's leniency proposal provides a sound foundation for an effective program to
encourage settlements in cartel cases. The Sections encourage the Bureau to create a system
with procedural transparency and legal certainty ­ to the extent that is possible ­ and settlement
discounts that are sufficiently generous to encourage parties to cooperate fully and settle early.
The Bureau and the DPP also need to provide confidentiality and privilege assurance to protect
the parties in future damage proceedings. This type of system will enhance the Bureau's ability
to obtain early and complete cooperation and reduce the need for contested proceedings. These
benefits will help the Bureau achieve its important goals of deterring and punishing cartels.

                                              Respectfully submitted,



                                              American Bar Association
                                              Section of Antitrust Law,
                                              Section of International Law,
                                              Section of Business Law,
                                              and Section of Criminal Justice




31
     Id., at ¶ 4.2.1.2.
32
   See Joint Comments of the American Bar Association Section of Antitrust Law and Section of International Law
in Response to the Commission of the European Communications' Request for Public Comments on the Draft EU
Settlement Procedures (December 2007), at 15-23, available at http://www.abanet.org/antitrust/at-
comments/2007/02-07/comments-EU-Draft-Settlement.pdf (discussing: the need to avoid a requirement for written
submissions by leniency applicants, or if required, that the submissions be strictly limited and protected as
confidential; the use of hypothetical fact patterns and proposed statements in lieu of submissions, especially in light
of the risk of statements in U.S. civil litigation and other enforcement proceedings; and the need to adopt specific
safeguards to protect against the disclosure of settlement submissions).



                                                         - 16 -