Building a Comprehensive White-Collar Violations Data System

The author(s) shown below used Federal funds provided by the U.S.

Department of Justice and prepared the following final report:

Document Title: Building a Comprehensive White-Collar

Violations Data System, Final Technical Report

Author(s): Sally S. Simpson, Peter Cleary Yeager

Document No.: 248667

Date Received: March 2015

Award Number: 2012-R2-CX-K016

This report has not been published by the U.S. Department of Justice.

To provide better customer service, NCJRS has made this federally

funded grant report available electronically.

Opinions or points of view expressed are those

of the author(s) and do not necessarily reflect

the official position or policies of the U.S.

Department of Justice.

P a g e | 1

Final Technical Report

Building a Comprehensive White-Collar Violations Data System

Sally S. Simpson and Peter Cleary Yeager

Report submitted to the Bureau of Justice Statistics, Grant No. 2012-R2-CX-K016

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 2

Contents

Abstract…..…………………………………………………………………..….....3

I. Overview of the White-Collar Crime Data Series…………………………..5

II. Conceptualizing the Series………………………...……………………….10

III. Prior Efforts to Measure White-Collar Offenses……..………………….....19

IV. Design Elements for Anticipated Data Series.…………………………......45

V. Assessing Available Data..………………………………………………....70

VI. Criminal and Civil Case Data: Strengths and Weaknesses………………...99

VII. Lessons learned from the Agency Briefs………….……………………...109

VIII. Conclusions and Recommendations………………….…………………..112

Bibliography……………………………………………………………………..121

Statutes Cited………………………………………………………………….…125

Appendix: Agency Briefs……………………………………………………….127

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 3

ABSTRACT

Despite its voluminous collections of data on conventional crimes and the legal responses

to them, the Nation has long lacked systematic data on white-collar offenses and the sanctions

employed against them. Because this void hampers research on and policy development for such

offenses, scholars and political leaders have advocated the development of an ongoing data

system that would systematically capture, measure, and describe these violations and the legal

responses to them. The federal Bureau of Justice Statistics (BJS), as the country’s leading

repository for the collection, analysis and dissemination of data on crime and justice processing,

is uniquely situated to develop a data series on white collar offenses. This BJS-funded project

proposes a design for such a series at the federal level and assesses opportunities and challenges

in its implementation.

The design elements for the data system include a definition of white-collar offenses that

distinguishes them clearly from other forms of offending and that corresponds to both

professional and popular conceptions of such violations. To achieve these goals we use a role-

centered definition that locates the motivations and opportunities for violations in legitimate

occupational and organizational roles. The data system comprises all violations of federal

laws—those sanctioned by regulatory/administrative, civil or criminal procedures—to represent

accurately the array of offenses and sanctions employed against them. From the federal agencies

and departments that enforce laws against white-collar offenses, it would systematically and

regularly collect enforcement case data on key factors: sources of identification of cases of

offenses, characteristics of offenses and offenders, and case outcomes.

Project staff assessed enforcement data characteristics, quality and availability in two

ways: through meetings and interviews with enforcement and data management personnel from

selected departments and agencies, and through examination of criminal and civil data held by

BJS as well as data made publicly available on a sample of agency web sites. Among the

findings and conclusions from these approaches are that (1) currently available data held by the

federal government’s enforcement units could contribute valuably to the formation of a data

series, but the data vary in terms of completeness and ease of accessibility across agencies;

(2) some agency personnel see the effort to build the data system as valuable to their own efforts

to improve the quality of their enforcement data; (3) the two agencies from which the project

sought agreements to share data with BJS to initiate the series both demonstrated early reluctance

to share the data; (4) agency web sites with publicly available enforcement data are a promising

source of information for the data series; (5) with current data sources we are commonly not able

to identify individuals who offend in legitimate occupations because role often cannot be

determined unless persons are listed as co-defendants with organizations; even then

distinguishing white-collar civil and criminal cases (as defined herein) from conventional

offenses can be uncertain because the available data do not clearly differentiate between

legitimate and illegitimate organizations; (6) demonstration projects between Department of

Justice enforcement divisions and agencies that share jurisdiction with them have promise for

both assessing data quality and completeness and promoting ongoing data-sharing for the white-

collar offenses data system; (7) BJS should continue its efforts to develop the data system.

Among other steps it should seek to form an ongoing working group that includes relevant

personnel from key enforcement units and agencies to share information on enforcement data

management challenges, needs and goals, to discuss the purposes and goals of a white-collar

offenses data system, and to discover the synergies between the enforcement units’ efforts to

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 4

improve their own data systems and the ongoing development of the new BJS data system for

white-collar offenses.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 5

I. Overview of the White-Collar Crime Data Series

A. Framing the Issue

Since its inception the Bureau of Justice Statistics (BJS) has collected,

analyzed, published, and disseminated information on crime, criminal offenders,

victims of crime, and the operation of justice systems at all levels of government.

A notable exception to this comprehensive coverage is the lack of information

about white-collar offenses, offenders, and justice system responses, even though

the collection of such data was one of the original tasks outlined for the agency.

Although the United States has long had annual accounting systems for

conventional crimes (including data collected and held by BJS) that have

underwritten countless research investigations and informed public policy

deliberations, the data landscape for white-collar offending is significantly more

constrained. There is no systematic accounting of white-collar offenses and

available data are substantially limited in scope and content. BJS does hold some

case and defendant-level criminal and civil enforcement data

on “white-collar”

crime from which it has issued select reports on Federal Enforcement of

Environmental Laws (Scalia, 1999) and white-collar offending at the state- and

federal-level (BJS, 1986; Mason, 1987), but the information is not organized in

BJS was created under the Justice Systems Improvement Act of 1979 to “promote the collection and analysis of

statistical information concerning crime, juvenile delinquency, and civil disputes.” It was mandated, among other

tasks, to collect “information concerning criminal victimization, crimes against the elderly, and white-collar crime”

(emphasis added).

The civil data have not been explored or used to the same extent as the criminal case processing data.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 6

such a way that enforcement actions against individual or organizational

defendants can be followed over time or linked across criminal, civil, and

regulatory justice systems. Indeed, regulatory enforcement data on organizations

are for the most part absent, and details about offenders and victims are sparse.

Consequently, it is impossible to track the extent of known white-collar violations

in the United States over time, to describe the characteristics of cases and

defendants, or to capture the full array of sanctions levied in a particular case or

against perpetrators.

The inability of the federal government consistently and accurately to report

known instances of white-collar offending has far-reaching consequences for

general knowledge, scientific investigation, and evidence-based policy in this

important area. “Policies are formed and legislation is passed, often seemingly

rationalized by little more than carefully selected anecdotes that seem to support a

particular policy when taken in isolation—and that can easily be countered by

other anecdotes that are supportive of an opposing point of view” (Dunworth and

Rogers, 1996: 499-500). Such an empirically deficient approach toward white-

collar crime virtually ensures “combativeness and policy mistakes”.

This policy-related observation was made by Dunworth and Rogers after their assessment of data on big business

litigation in Federal courts, but it is equally apt for white-collar crime.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 7

Confounding the development of a comprehensive data collection and

management system is the lack of conceptual clarity surrounding the definition of

white-collar crime. Because white-collar crime is not a legal category, definitions

abound. Criminologists and legal scholars have defined and classified white-collar

crime in a variety of ways. Some definitions focus on offenses committed by

companies and their managers to achieve the goals of the business, while others

emphasize offenses committed by individuals that may or may not involve

organizational or business resources but tend to be tied more to self-interest and

guile (e.g., embezzlement or tax fraud). Consistent with Sutherland (1949), social

scientists often prefer offender-based definitions of white-collar crime which call

attention to the social position of the actor. For Sutherland, white-collar crime was

“committed by a person of respectability and high social status in the course of his

occupation” (1949: 9). The behaviors he had in mind were not necessarily (or

even typically) pursued, prosecuted, and punished in the criminal justice system.

Instead they were handled commonly through civil and administrative justice

processes. But they met the general criteria of criminal behavior, i.e., offenses

defined by the “legal definition of social injuries and (ultimately the) legal

provision of penal sanctions” (Sutherland, 1983: 52). Although Sutherland’s

definition emphasized individual-level characteristics, his empirical research

focused on the offenses of corporations—an inconsistency that caused definitional

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 8

confusion and the subsequent parsing of white-collar crime into criminal behavior

systems that helped to organize, classify, and make sense of the extensive range of

behaviors captured by Sutherland’s conceptual and empirical work (Clinard et al.,

1994).

In contrast, offense-based definitions of white-collar crime focus on the

means through which the offense is perpetrated and its characteristics, as in

Edelhertz’s (1970: 3) definition: “an illegal act or a series of illegal acts committed

by nonphysical means and by concealment or guile to obtain money or property, to

avoid the payment or loss of money or property, or to obtain personal or business

advantage”. Edelhertz recognized distinct types of white-collar offending within

this broad definition, including personal crimes (individuals who act by themselves

for personal gain in a non-business setting); abuses of trust (people operating

within legitimate businesses and other organizations or professions who violate

their duties to an employer or client); business crimes (crimes that further business

interests but are not the primary focus of the firm); and con games (illegal acts by

an illicit organization whose business is white-collar crime) (Edelhertz, 1970: 19-

20). This focus on the offense itself rather than the social position of the actor

emphasized the diverse and extensive nature of fraudulent behavior.

The Bureau of Justice Statistics originally incorporated both offender- and

offense-based approaches into a single definition of white-collar crime. The

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 9

second edition of the Dictionary of Criminal Justice Data Terminology (1981:

215), for instance, described white-collar crime as “nonviolent crime for financial

gain committed by means of deception by persons whose occupational status is

entrepreneurial, professional or semi-professional and utilizing their special

occupational skills and opportunities; also nonviolent crime for financial gain

utilizing deceptions and committed by anyone having special technical and

professional knowledge of business and government, irrespective of the person’s

occupation.” Practically, however, most criminal justice data management systems

do not collect information about the offender’s occupational status or special

knowledge utilized to commit the offense. This reality ultimately restricted the

operational definition of white-collar crime to “nonviolent crime for financial gain

committed by means of deception” (Mason, 1986: 2) and produced a classification

scheme in which forgery, counterfeiting, fraud, and embezzlement constituted

white-collar crime. Notably, the classification of white-collar crimes in this

manner was not driven not by conceptual considerations but rather by data

limitations.

B. Justification for the Series

Recognizing existing data deficiencies and the need for research and

development on the topic, BJS contracted with Professor Sally S. Simpson

(University of Maryland) and Professor Peter C. Yeager (Boston University) to

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 10

assess the feasibility of developing a statistical series that would integrate criminal

and civil data that BJS receives from the Executive Office for U.S. Attorneys

(EOUSA) and the Administrative Office of the U.S. Courts (AOUSC) with

enforcement data from federal regulatory agencies. The ultimate goal of this effort

is to describe comprehensively the federal response to white-collar violations and

offer recommendations for series design and content. This Technical Report

summarizes the conceptual and methodological approach adopted by Simpson and

Yeager to map the data landscape. It also utilizes and assesses data from specific

agencies to demonstrate the feasibility of this approach.

II. Conceptualizing the Series

A. Principles and Aims

The conceptual basis for the data series comprises a number of key principles

and aims. A data series on white collar violations of federal laws should, of

course, meet the standards of reliability and validity. In practice these translate to

the requirements that the measurement schema being employed are

comprehensive, clear and replicable, and that what is being measured comports

with a concise, analytically-based and clear definition of the phenomenon of

interest.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 11

Because white collar offenses comprise a large and diverse array of illegal

behaviors, a data series for them must, in pursuit of comprehensiveness, be able to

integrate diverse existing databases into a single system of accounting. This

requires a coding system with two important characteristics that exist in a certain

tension with each other: that it be broad enough to capture the key points in the law

enforcement handling of cases of white collar offending, and that its data collection

categories be adequately concise to capture reliably and consistently these data

points from among the widely varying federal data systems in which the violations

and enforcement data originate.

Finally such a series should be maintained in timely fashion, and it must be

an ongoing endeavor with regularly scheduled additions of data on new cases—and

new developments in existing cases—to the database. And as a routine component

of the maintenance and growth of the series, there should be ongoing assessments

of data quality, of the series’ utility for both policy and research purposes, and of

potential improvements in collection and measurement. Such efforts will not only

ensure the quality of the data series, but should also have the effect of working

synergistically with parallel efforts of participating federal agencies to improve

their own data management processes toward the more uniform, effective and

useful measurement of white collar offenses and enforcement in the federal

government.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 12

The principal justification for and value of a data series on white collar

offenses are the contributions it will make to research and public policy. An

ongoing series of the sort envisioned here would not only provide unprecedented

opportunities for research on patterns of white collar law-breaking and the federal

response to it. Its availability would also promote such research in an area of

investigation that has been sharply limited by the lack of available data. To the

extent that the data series on white collar offenses permits and promotes this

research—much as the well-established National Crime Victimization Series and

the FBI’s Uniform Crime Reports have long underwritten the voluminous research

on conventional crimes—to that extent will law enforcement, legislators and

judicial personnel have access to a body of knowledge that will contribute to

improved policies for addressing this form of offending.

Key to a successful data series is an efficient and uniform system for data

collection and coding of information. This database system needs also to capture

information on the key factors or ‘moments’ in the handling of cases of white

collar offending. Figure 1 offers a simplified schematic presentation of those key

factors.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 13

The three primary types of information for inclusion in the series are the

sources of identification of cases of violation, data on offenses and offenders that

are processed for enforcement, and information on case outcomes, including

sanctions and referrals for further legal action. The latter is especially relevant for

cases originally processed and/or sanctioned by regulatory agencies that later refer

them to U.S. Justice Department attorneys for criminal prosecution. For each of

the three categories of data it is necessary to construct a parsimonious coding

system that permits and guides the translation of contributing enforcement

agencies’ diversely structured and variably inclusive data into reliable and

homogeneous indicators of the key variables. The development of an effective

‘crosswalk’ system for the purposes of this translation is essential to the

construction of the data series on white collar offenses.

Figure 1

Case

Source

•Complaints

•Referrals

•Investigations

Case

Processing

•Case Types

•Charges

•Defendants

Case

Resolution

•Violations

•Penalties

•Referrals

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 14

B. Methodological Matters

A white-collar crime data series should classify conceptually similar acts

using a sensible, commonly understood and culturally shared definition. Among

other reasons this is because such a series requires both legal and cultural

legitimacy—in an arena that is highly contested and often misunderstood. Because

U.S. laws hold that legal entities such as corporations are generally to be treated

under the law as persons, the series would encompass both offending individuals

and organizations (inclusive of for-profit, nonprofit and governmental actors). The

scope of the series would include all federal agencies’ criminal, civil, and

regulatory enforcement cases that are consistent with the definition. Although

there is broad variation in the types and quality of data available from regulatory,

civil and criminal law sources, the breadth of extracted data would capture key

offense, offender, and sanctions variables that are commonly defined, counted, and

measured across sources. The series should link initiated and processed cases of

offending across the distinct legal fields, and track respondents and defendants

(individually and co-offenders) across stages in the justice process over time. In

sum, it should maximize data comprehensiveness and compatibility and set a

strong foundation for future agency collaborators in the series.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 15

Challenges in Regulatory Data. On the face of it, the collection and

integration of regulatory data into a Federal White-Collar Crime Statistical Series

should be straightforward. Each regulatory agency was created by Congress

through enabling legislation that defines its purposes, tasks, and powers. Agencies

monitor actor compliance with specific statutes and regulations using a wide array

of mechanisms (e.g., inspections, self-reports). These agencies conduct

investigations, obtain reports from firms, keep records of investigative findings,

and hold hearings to establish violations of regulations and laws. Investigations

may result in case referral for criminal, civil judicial, and/or administrative

proceedings.

The reality, however, is much more complicated. Because agencies vary in

their specific tasks and powers, the data they collect and maintain also varies.

First, the quantity of data varies greatly between agencies. Some agencies

investigate and pursue hundreds of violations yearly while the capacity of others is

much smaller. Second, the mix of data also diverges by agency. A number of

regulatory agencies have broad authority to pursue civil litigation and criminal

investigations using their own investigators and attorneys, others do not. The

Environmental Protection Agency (EPA), for instance, can pursue violations of

law as civil administrative cases, civil judicial cases, or criminal investigations

(EPA investigators have warrant and arrest authority, although on finding evidence

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 16

of criminal offenses it must refer the cases to Justice Department attorneys for

prosecution). Other agencies’ powers, however, are more restricted. Enforcement

activity by the Federal Trade Commission (FTC), for instance, is limited to

administrative proceedings or the pursuit of civil actions in Federal court. Cases

thought to merit criminal charges are referred to the Department of Justice

Antitrust Division for investigation and prosecution.

Similarly, the Securities and

Exchange Commission (SEC) has the authority to bring a civil case in federal court

or before an administrative law judge within the SEC. Criminal enforcement of

the Federal securities laws is pursued through the U.S. Department of Justice and

the individual U.S. Attorneys General throughout the country. Third, case data

held by specific agencies will not always be exclusive to a single agency because

cases stemming from the same incident for the same conduct may be brought

simultaneously in different venues. For example, a defendant in a criminal

securities fraud case brought by Justice Department prosecutors may also be

subject to civil justice processing by the SEC. Further, there is some evidence that

criminal actions by the government are more likely to be brought in cases

involving both environmental and employee safety laws.

In such situations the

Because the Department of Justice Antitrust Division and the FTC share statutory authority for certain sections of

the Clayton Act, and because the FTC can challenge conduct under the Sherman Act, the two agencies must

coordinate with one another to determine, as each case arises, which agency is most appropriate for handling the

matter.

See, e.g., http://www.oshalawupdate.com/2012/12/18/osha-criminal-referrals-on-the-rise/.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 17

two agencies’ case data will overlap to the extent that they report on the

investigations that led to referrals to the Justice Department for criminal

prosecution.

This “duplication” of cases is complicated by whether individuals and

organizations within a case share similar or distinct sanction processes. So, for

example, a case of securities fraud may name five defendants in a civil case—two

organizations and three individuals—but parallel criminal charges may be brought

only against the organizations and not the individuals. Given that one of the key

goals of the white-collar crime statistical series is to track case-specific sanctions

against defendants across justice processes, this nettlesome problem must be

resolved.

Challenges in Criminal and Civil Data. Federal criminal and civil data

sources, available from the Executive Office of the U.S. Attorneys (EOUSA) and

Administrative Office of the U.S. Courts (AOUSC), must be merged with the

regulatory agency data in order to describe comprehensively the federal response

to white-collar violations.

The EOUSA data track criminal and civil cases and

defendants from matters presented to and/or pursued by the U.S. Attorneys, while

the AOUSC data contain civil and criminal case and defendant court filings and

Both EOUSA and AOUSA data are extracts from the case management systems used by federal prosecutors and

federal courts, respectively, for their specific administrative purposes. Therefore, the type of information we might

require for the data series on white-collar offenses may not always be available or accessible.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 18

outcomes in the U.S. Courts. Both data sources contain fairly detailed case

descriptions. However, there are notable insufficiencies in these data. Although

we will provide more details later in this report, several key differences in what is

recorded (and required to be reported) and how cases are counted between these

sources impact data comparability, completeness, and reliability.

On the criminal side of things, the EOUSA data include a variable for

referring agency and a flag (drawn from the variable participant type) to highlight

whether the case-defendant is an organization (which includes collective groups of

all types). The AOUSC data lack this flag and do not report referring agency.

Neither source offers much detail about organizational defendants, nor is the flag a

required data element. AOUSC data provide more detail on the violations charged

in a case (up to five unique charges) but the EOUSA reports only the lead charge

(and do not contain the statute violated for Matters Referred for Prosecution).

For civil court cases both data sources are less useful for our purposes. For

instance, the AOUSC civil data lack referring agency, a participant type flag, and

plaintiff/defendant information; it also is not possible to disaggregate defendants in

the same case. In addition, the data make it impossible to distinguish regulatory

agency action from DOJ action, from third party action, and so on. Because the

unit of analysis in the EOUSA data is case-defendant and the data currently lack

case or docket numbers, it is not possible to link defendants within the same case, a

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 19

connection that the criminal case data allow. Both data sources report one statute

associated with the case, but it is unclear if the case type variable

is the same

across the two datasets.

From this brief review, it is clear that the design and structure of a

comprehensive white-collar crime data series will be challenged by the strengths

and weaknesses of the kinds of data available to build the series. To better

understand what this means for our current efforts, in the next section we

summarize and assess how others have approached this problem.

III. Prior Efforts to Measure White Collar Offenses

This project is far from the first attempt to conceptualize and measure white

collar violations in a systematic manner. Although we have drawn and built upon

these earlier efforts, we have not adopted full-scale any approach as each has its

strengths and limitations. Most notably, there are differences in the focus and

scope of white-collar crime across studies that, in turn, affect the source and kind

of data utilized (Johnson and Leo, 1993). It is also important to keep in mind that

the empirical studies were designed to answer specific research questions and not

to build a data series. Below, we briefly summarize these efforts and highlight the

The AOUSC and EOUSA data include offense categorizations (such as “environmental offenses” and “antitrust”)

but it is unclear if these categories encompass the same statutes across the two databases. If it were possible to

verify which statutes are included in these categories, and confirm that they are consistent across databases, they

would be useful for efficiently identifying particular offenses of interest, without having to search for specific titles

and sections of statutes.)

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 20

key elements that have informed our project, beginning with the seminal work of

Edwin Sutherland.

A. Research-Based Efforts: Goals and Purposes

Edwin Sutherland. Although his definition of white-collar crime focused

on individuals, giving rise to offender-based approaches (“crime committed by a

person of respectability and high social status in the course of his occupation,”

1983: 11), Sutherland’s empirical research focused on companies. Sutherland

provides no rationale for this disconnect but it is possible that it was more

expedient at the time to sample, identify, and track firms as compared to

individuals. Moreover, the fact that crimes by business attack the fundamental

principles of American institutions—a key element of white-collar crime for

Sutherland—is also a strong justification for the company focus. To quote

Sutherland (1983:13), “white collar crimes violate trust and therefore create

distrust; this lowers social morale and produces social disorganization. …Ordinary

crimes, on the other hand, produce little effect on social institutions or social

organization.”

Sutherland studied the life history of 70 of the largest publicly- and

privately-owned corporations in the United States with regard to their violations of

law. Specifically, he used case and offense data reported by federal agencies and

the New York Times to examine restraint of trade, misrepresentation in

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 21

advertising, unfair labor practices, financial fraud and violation of trust, violations

of war regulations, and a small number of miscellaneous offenses.

To measure

legal violations, Sutherland employed formal decisions and orders of the court and

administrative agencies against a firm, including stipulations accepted by the court

or agency, settlements ordered or approved by the court, confiscation of food (in

violation of the Pure Food Act), and a few other ex post facto cases that had been

dismissed earlier. Sutherland thus relied on cases prosecuted, litigated, and

brought in regulatory/administrative, criminal, and civil justice venues. This

approach allowed him to then examine offending patterns over time, within firms,

across industries, and by legal venue.

Sutherland’s unit of count is a decision within a case. His rules for counting

decisions were as follows: (1) when three companies are defendants in a law suit in

which a decision is made, the decision is counted three times—once for each firm.

(2) If parallel cases are brought in different legal venues and found against one

company for essentially the same overt behavior, two decisions are counted. (3)

One decision may summarize multiple charges and behaviors that have taken place

over many years (e.g., price-fixing). In effect, all the charges and years are rolled

Sutherland did not specifically sample firms on any basis other than size (68 of the companies were listed on two

lists of the 200 largest non-financial corporations in the United States) and specialization—excluding corporations in

one industry and public utility corporations—although he did examine 15 of the largest power and light corporations

for comparison purposes.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 22

into a decision with a corresponding count of one (1983: 19). Depending on the

rule, Sutherland’s methods may increase the potential for over-counting or

undercounting offenses.

The key elements of Sutherland’s approach are as follows: (1) The decision

to track offending by legitimate businesses; (2) offending is captured across

criminal, regulatory, and civil venues (including private suits); (3) offense types

are constrained to five broad categories of crimes and some miscellaneous

violations; (4) an offense occurs only when a decision is determined against the

firm; (5) time is censored by the life history of the firm and the conclusion of the

study (1949).

Yale Studies. A very different approach was adopted by Stanton Wheeler

and his collaborators at Yale University (Wheeler et al., 1988; Weisburd et al.,

1991; Weisburd and Waring, 2001).

These researchers studied individual

offenders in seven federal courts between 1976 and 1978 who were prosecuted for

and found guilty of violating one (or more) of eight offenses in the federal criminal

code. This approach is consistent with a statute-based strategy for counting and

measuring white-collar crime and an offense-based definition of white-collar

crime. Wheeler and his colleagues (1982: 642) define white collar crimes as

A number of other researchers were involved in these efforts, but our concern here is how white collar offenders

and offenses were defined and measured in the Yale studies. Thus, we cite the works most relevant for our

purposes.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 23

“economic offenses committed through the use of some combination of fraud,

deception, or collusion.” The specific offenses selected for study included

securities violations, antitrust violations, bribery, bank embezzlement, mail and

wire fraud, tax fraud, false claims and statements, and credit- and lending-

institution fraud.

Once all offenders who met the inclusion criteria were identified, the

researchers then stratified and sampled among the offenders. Because there were

so few securities fraud and antitrust offenders,

all of these offenders were

selected for the study, whereas a random sample of the other offenders was drawn.

With sample in hand, the study then focused on the characteristics of the offenders

AND the offense committed. Offender and offense information in the study came

primarily from Pre-Sentence Investigative reports—documents prepared by the

probation officer, often with input from law enforcement and prosecutors—that

judges can use to inform sentencing decisions.

The Wheeler et al. (1988) approach to counting and measuring white-collar

crime has several key features, including the decision to focus on: (1) criminal

offenses; (2) individuals and not organizational offenders, although the researchers

track whether a corporate indictment was issued in the case; (3) a limited number

The fact that so few of these cases are criminally prosecuted says something about the limitations of the criminal

data source.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 24

of specific offenses (eight) and federal courts (seven) from which to select

offenders; (4) the sentencing stage of the criminal justice process (i.e., guilt had

been determined); and (4) a constrained set of years (1976-78)

. In addition, the

study depends completely on the availability of PSI reports to link relevant

offender and case information.

Wisconsin Study. The NIJ-funded research project on corporate crime

(Clinard et al., 1979; Clinard and Yeager, 1980; 2006) has more in common with

Sutherland’s approach to white-collar crime than it has with Wheeler’s. In the

Wisconsin study, the researchers focused on legitimate corporate actors and

tracked offending across administrative, civil, and criminal justice venues—actions

taken by a total of 25 federal agencies (Clinard and Yeager, 2006:110). Unlike

Sutherland, however, the Wisconsin study adopted a definition of crime that was

consistent with the subject of their research. Corporate crime was defined as “any

act committed by corporations that is punished by the state, regardless of whether

it is punished under administrative, civil, or criminal law” (Clinard and Yeager,

1980:16). Corporate crime is a subtype of white-collar crime that has distinct

features in that it is organizational in nature and occurs in the context of complex

corporate relationships.

Weisburd and Waring (2001) extended this time period by collecting official measures of criminality (arrests)

from FBI rap sheets for the original sample through 1990.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 25

The Wisconsin study’s sample of companies was substantially larger than

Sutherland’s (477 of the largest publically-owned U.S. manufacturing companies,

plus an additional 105 public companies in the wholesale, retail, and service

industries), but the increase in the sample size negatively affected the length of

time firms could reasonably be followed (1975-1976).

The Wisconsin project operationalized crime more broadly than did either

Sutherland or the Yale studies. Researchers gathered information at a point earlier

in the justice process and included all known initiated cases and enforcement

actions taken against a corporation. This technique is comparable to studies of

street crime that operationalize crime using police statistics (crimes known to

police). While the previous studies can be criticized for ignoring the winnowing or

funneling process whereby cases are diverted out of the legal system (if brought at

all), the Wisconsin study can be criticized for including actions that companies

may not, in fact, have committed or for which they are not legally responsible.

The range of violations covered in the Wisconsin study was far broader than

that covered by Sutherland, but there was overlap as well. The specific types of

violations included administrative, environmental, financial, labor, manufacturing,

and unfair trade practices (Clinard and Yeager, 2006: 113-116). An additional

special feature of the Wisconsin study was that researchers created a classification

scheme to rank violations as serious, moderate, or minor. The authors point out

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 26

that because most agencies did not have severity criteria (Clinard and Yeager,

2006: 118), their determinants of ranking were tied to things such as repetition of

the same offense by the company, intent, the spread of the crime within a

company, harm to victims (calculated in several different ways), firm refusal to

take pro-social actions (such as recall, reinstate or rehire employees, refusal to

honor agreements), threatening actions by the firm, and the length of time of the

violation.

A main goal of the study (1979) was to understand the etiology and patterns

of, as well as responses to, corporate crime. Therefore, in addition to capturing

initiated and enforcement actions against companies from a variety of different

sources, the study also utilized information about firm characteristics (such as

company size, financial performance, and market characteristics) to analyze the

relationship between these and firm offending records.

Yeager Study. Yeager’s (1987, 1991) research focused on the enforcement

of federal environmental laws against business polluters. It examined the social

and political factors that shaped both offending and enforcement decisions. While

its data base was a narrowly construed one—compliance and enforcement among

industrial violators of the federal Clean Water Act in New Jersey—its construction

illustrates some of the key matters that a data series on white collar offenses must

manage.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 27

For example, at the time of the research the Environmental Protection

Agency’s compliance and enforcement data systems varied significantly by

regulatory region in the U.S. Only Region II, which includes New York and New

Jersey, maintained longitudinal electronic records on polluting facilities.

Moreover, the two states varied in their enforcement authorities. In New Jersey,

the EPA itself monitored industrial compliance and enforced the law against

violators, while in New York State the Agency had delegated enforcement of the

Clean Water Act to the state’s environmental regulators, as allowed by the law.

Yeager’s analysis focused on the New Jersey data, which avoided the need to

ascertain any differences between the two states in data coding protocols and

enforcement priorities.

The unit of count for the study was the firm-violation, the goal being to

ascertain the number of violations of the law for each company over the period of

time during which its effluent had been regulated. Because the EPA data were

kept at the facility level, it was necessary to aggregate violation (and sanctions)

counts for facilities owned by the same company. This was possible because the

data included the name of the facility owner, but at times the aggregation required

careful matching of names as some facilities were owned by subsidiaries of major

corporations.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 28

Because of the way information was entered into the data base, it was also

necessary to construct analyses of offenses (counts) that normalized them across

facilities and firms. Many violations were reported on self-monitoring reports that

facilities were required to submit to EPA on schedules set for each facility, e.g.,

monthly or quarterly. Therefore, the number of offenses reported per year could in

part be an artifact of the number of reports required during the year, requiring

normalization of the counts.

Simpson Studies. Two separate white-collar (corporate) crime studies

conducted by Simpson (1985; 2007) also inform our strategic approach to the BJS

data series. The first study tracked the offending behavior of 52 “survivor”

companies over a 55-year time period (1927-1981) in the United States. The

companies were randomly selected from seven basic manufacturing industries.

The only criteria for selection were that they stayed in business for most if not all

of this time period (in some form), they continued to operate within the industry in

which they operated in 1927, and that the firms were US-based. The study focused

on only one type of illegal activity—alleged anti-competitive behavior—and used

two data sources to connect criminal, civil, and regulatory offending information

with companies. Cases were drawn from the Federal Trade Commission Case

Decisions and Trade Cases, which contain "texts of decisions rendered by federal

and state courts . . . involving antitrust, Federal Trade Commission, and other trade

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 29

regulation law problems" (Trade Cases, Introduction). Simpson also was

interested in corporate crime etiology and enforcement, so additional data were

drawn from sources such as Compustat, U.S. Census of Manufacturers, firm 10K

Reports, and the Statistical Abstract of the United States. These data were matched

with firm offending records to examine the economic and political context in

which offending occurred. To establish the proper temporal ordering between

economic characteristics and offending, cases were coded according to the year in

which the offense allegedly occurred (as per agency case documents), not when the

case was brought. Offense counts were created when a case was brought against a

company. Cases were tracked over time so that resolutions could also be

ascertained and classified by outcome (e.g., settlement, guilty finding, cease and

desist order). If a case was appealed, that was noted in the data. Simpson also

created a seriousness measure to rank the anti-competitive offenses. Offense type

was not utilized to ascertain seriousness because there were so many different

kinds of cases captured within the same offense type. Instead, two categories of

seriousness were created (serious and trivial) based on the degree of harm each act

engendered and victim class similarity (e.g., suppliers, customers, competitors).

Serious violations, such as all forms of price-fixing, predatory pricing, conspiracy

to monopolize and control territories, interlocks, and illegal mergers, were those in

which the cost of the act was high for another or potential competitor, resulting in a

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 30

substantially less competitive market. Most (but not all) cases of unfair advertising

and warranty violations were coded as trivial, i.e., those acts that, while widely

dispersed, have marginal economic effects on victims. A middle category of

seriousness (moderately serious) was created but not utilized due to the higher

potential for coding errors within this category.

The second Simpson study (Simpson, Garner, and Gibbs, 2007), funded by

the National Institute of Justice, used a triangulated research strategy that included

interviews with inspectors, secondary data analysis, and factorial surveys to assess

the deterrent effects of different kinds of state responses to firms that failed to

comply with environmental regulation, specifically the National Pollutant

Discharge Elimination System as authorized by the Clean Water Act. Therefore,

the study was designed to compare the effect of cooperative versus punitive

approaches on corporate recidivism. Only the construction of the secondary

analysis data set is relevant for our purposes here.

Simpson, Garner, and Gibbs (2007) randomly selected firms in four

manufacturing industries (pulp, paper, steel, and oil) in 1995. This strategy

produced a distribution of 30 pulp and paper companies (the two industries were

combined because of substantial overlap in the firms and facilities in the two

industries), 18 steel companies, and 19 oil companies (N=67). The companies

were followed through the end of 2000 by which time, due mostly to mergers, the

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 31

total number of firms was reduced to 55. The study relied on information collected

by the EPA in their Performance Compliance System (PCS) database and docket

files (administrative, civil, and criminal case files) to measure company violations.

Enforcement data in both data files (PCS and EPA dockets) were likely to overlap

to some degree, but there was no method for tracking the same violation by the

same offender across data sources.

The violation data (pollution and compliance schedule violations, as well as

EPA enforcement activity) are captured at the facility level, but because the focus

of the study was on the firm and not the facility, facility violations were counted,

aggregated and matched to the specific companies in the sample. If a company

owned several facilities, the firm offending count was a sum of the violations

counted at each of its facilities. Annual aggregation (normalization of count) was

also necessary as some compliance schedule requirements are monthly or

quarterly. A unique feature of this study is that it captured both inspection data and

“self-report” data—information that allowed researchers to create measures of

opportunity and to construct rate variables (number of violations/number of reports

required).

The EPA study also collected economic and structural information about the

firms in the sample that allowed the researchers to address two additional research

questions: (1) Are certain characteristics of companies associated with a greater

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 32

offending risk? (2) Do firm characteristics affect the type of intervention and

punishment a company receives when a violation occurs?

Schlegel SEC study. Another study of white-collar crime, funded by the

National Institute of Justice, was undertaken by Kip Schlegel and his research

associates at Indiana University (Schlegel, Eitle, and Gunkel, 1994). The topic of

this project was securities lawbreaking. To study securities fraud and the

enforcement response, Schlegel and his associates collected quantitative data on

enforcement actions taken by Self-Regulatory Organizations (National Association

of Securities Dealers, the NY Stock Exchange, and the American Stock Exchange),

the Securities and Exchange Commission, and the US Department of Justice.

Arguing that the study must consider fraud in its broadest context, the researchers

justify this approach by suggesting “it is a far more taxing yet potentially more

rewarding approach to try to examine what has not been caught, or what has been

caught by different means, and to study the net closely to determine the changes to

be made and the alternatives available” (Schlegel et al., 1994: 37).

The length of the available archival record varied, depending on the data

source. SEC enforcement actions (called “releases”) were bound and published in

a document called the SEC docket. These data spanned 1985 to 1991. DOJ data

on security actions were analyzed for the years 1984-1991. The self-regulatory

data extended from 1988 through 1992. All actions during these time periods were

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 33

extracted from the archival record for both individuals and companies. There were

relatively few of the latter, and they were found mainly in the administrative legal

realm—less so in criminal and civil actions.

Comparisons were made as to what

kinds of offenses were discovered and sanctioned, the nature of the offense, the

victim(s) involved, the distribution of offenses over time, perpetrator

characteristics (males, females, firm), and the distribution of sanction type over

time. The different data sources varied in their level of analysis. Civil cases were

case-based records while the criminal and administrative cases were organized by

individual defendant. Consequently, the study recorded the number of offenders

involved (including firms), but did not record each offender’s case disposition—in

part because most sanctions during the time of the study involved injunctions

(Schlegel et al., 1994: 39).

Analysis of the data focused on cases formally entered/actions initiated and

cases disposed and resolved. The researchers also tracked when administrative

actions involved a “parallel proceeding” in civil or criminal court. Like previously

discussed studies, Schlegel and his associates collected additional data from other

sources, including interviews with relevant enforcement staff and a written survey

administered to select offices of the FBI.

In the criminal area, only four cases involved firms or companies during the period of the study.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 34

Schlegel and his collaborators highlight some of the data difficulties they

encountered in their use of archival records. Specifically, they note that

information varied within and between the different archival sources. This fact,

coupled with the presentational and stylistic differences in the authors of the SEC

litigation releases, made it difficult to capture the same information from the

sources, create a consistent coding instrument, and reliably interpret the meaning

of the information (1994: 134). Undoubtedly, these problems are not unique to the

Schlegel study.

Karpoff Studies. Jonathan Karpoff and his colleagues have built a series

of detailed databases from publicly available data sources. These databases have

been created to investigate company (and in some cases manager) participation in a

variety of different corporate offenses (financial misrepresentation, bribery,

environmental violations), and to assess the legal and extra-legal consequences

associated with enforcement. To some degree, each research question generates its

own database because different crimes are of interest to the researchers. So, for

instance, if the researchers are interested in foreign bribery (Karpoff, Lee, and

Martin, 2014), a sample of publicly traded companies against whom enforcement

actions for foreign bribery have been initiated by the U.S. Department of Justice

(DOJ) and Securities and Exchange Commission (SEC) is generated. If financial

misrepresentation is of interest, the researchers generate a list of firms targeted by

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 35

SEC enforcement actions for financial misrepresentation (Karpoff, Lee, and

Martin, 2008). Environmental violations (Karpoff, Lott, and Wehrley, 2005)

follow the same process, but enforcement data are gathered from The Wall Street

Journal Index, under its "Environment" and "Environmental Crime" listings, and

NOT from the EPA (no doubt because EPA data are facility- and not firm-based).

Across all of the studies, there are common primary data sources from which

information is drawn:

[T]the SEC website (www.sec.gov), which contains SEC press and selected

enforcement releases related to enforcement actions since September 19,

1995; the Department of Justice, which provides information on

enforcement activity through a network of related agencies with particular

emphasis on high-profile enforcement actions available at www.usdoj.gov;

the Wolters Kluwer Law & Business Securities (Federal) electronic library,

which contains all SEC releases and other materials as reported in the SEC

Docket since 1973 and select Federal Securities Law Reporter releases from

1940 to 1972; Lexis-Nexis’ FEDSEC:SECREL and FEDSEC:CASES

library, which contains information on securities enforcement actions; the

PACER database, which contains lawsuit-related information from federal

appellate, district and bankruptcy courts; the SEC’s Electronic Data

Gathering, Analysis, and Retrieval (EDGAR) system; and Lexis-Nexis’ All

News and Dow Jones’ Factiva news source, which includes news releases

that reveal when firms are subject to private civil suits and regulatory

scrutiny (see Karpoff, Koester, Lee, and Martin, 2014: 10-11).

We call attention to the Karpoff data because it demonstrates that useful data can

be electronically scraped from publicly available sources, and that once collected

the data can be used for a multitude of different purposes. Over time, archival data

scraping will become even easier as more and different kinds of sources become

electronically available and increasingly sophisticated software becomes available.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 36

B. Applied Efforts

There are a number of archival data collection efforts in the white-collar

crime area that have been used by consulting firms to advise government, law

firms, and corporations about issues relevant to policy, regulation, and litigation.

National Economic Research Associates (economic consultants), for instance, used

data from FinCen enforcement actions and BankersOnline.com BSA/AML

penalties list to report on recent trends in Bank Secrecy Act and anti-Money

Laundering enforcement (2014). The data can be used to identify the types of

institutions targeted for enforcement actions, how patterns of enforcement have

changed over time for penalties with and without fines, counts of filings,

comparison of BSZ/AML with other kinds of Suspicious Activity Report (SAR)

activities (including check fraud, mortgage fraud, and identity theft), and the ratio

of SARs filings to enforcement actions, among other purposes. Similarly, NERA

recently published a report on securities class action litigation (Comolli and

Starykh, 2014) that utilized federal court filings to assess litigation trends. Case

filings were broken down by circuit courts, type of violation, foreign country

domicile and year, and by sector.

Consultants are not the only ones to utilize available data in the white-collar

crime area to generate reports and assist clients. The legal practice of Morvillo

Abramowitz Gran Iason & Anello PC has generated a document in which they

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 37

report on all of the SEC Enforcement Division’s new case filings for the entire

calendar year 2013.

There is little evidence that the data collected and utilized by consultants and

legal practitioners have been subjected to rigorous verification and validation.

Given what we know from our efforts at data cross-validation, there are apt to be

substantial source discrepancies as well as large amounts of missing cases and

information that affect overall comprehensiveness and quality of the applied

databases.

A somewhat different approach has been taken by the National White Collar

Crime Center (NWCCC). In association with the Bureau of Justice Assistance, the

NWCCC has administered three national victimization surveys, most recently in

2010 in which 2,503 adults reported (via telephone) household white-collar

victimization experiences within the past 12 months. The purpose of the survey is

to discern the prevalence and types of white-collar victimizations, whether victims

reported to law enforcement or other agencies that could assist victims, and

perceptions of crime seriousness. The 2010 survey, compared with earlier

versions, also offered a more comprehensive assessment of “corporate” crimes.

Unfortunately, and not unlike the previous versions, the telephone-based response

The different databases put together by Karpoff and his associates have been evaluated and cleaned for data

problems such as case redundancy and contradictory details (personal communication with Gerald Martin).

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 38

rate was very low (13% for the landline samples and 18% for cellular, Huff et al.

2010:38). Such low response rates seriously challenge comparisons between data

sources to determine convergent and nomological validity. As acknowledged by

the survey administrators, such deficiencies affect the ability “to assess the true

frequency of various types of white collar crime” (Huff et al. 2010: 13).

Even with these caveats, the authors of the survey compare victimization

rates from the 2010 survey to data from the 2008 National Crime Victimization

Survey to “show” the extensive nature of white-collar victimizations. The NCVS

“computed a victimization report rate of 135 households per thousand (13.5%) for

property crime and 19.3 individuals age 12 or over per thousand (1.93%) for

violent crimes. Even at an understated rate of 24.2% (for households), white collar

crime victimization is occurring much more frequently than property crime and

violent crime combined” (Huff et al. 2010: 22, emphasis added). Unfortunately,

such comparisons and conclusions are problematic. The NCVS has a response rate

of 95%, providing confidence that the property and violent crime rates are closer to

the true rates in the population. With a response rate of 13% and 18%, the

epistemic correlation for white-collar victimization will likely be low. And

although the NWCCC assumes that the numbers will understate the prevalence,

frankly it is unclear the direction in which the data might be biased. If the

respondents are interested to report their experiences with white-collar crime

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 39

because they have been victimized, the bias will move in the direction of over-

estimating—not underestimating—victimization levels.

Drawing from the survey results, what does “white collar crime” look like

from the perspective of the victim? On average, 24% of the respondents reported

some kind of white-collar victimization in their household within the past 12

months (respondents could report one or more victimizations per household).

Credit card fraud is the most commonly reported offense (38.7%), followed by

price misrepresentation (28.8%), unnecessary repair (22.8%), and monetary loss on

the internet (14.3%). More than half of the victimizations (54.7 %) were reported

to at least one entity but, because nearly 46% were not reported, there is likely a

great deal of bias in the cases known to authorities generally let alone law

enforcement specifically. Not surprisingly, because credit card fraud was the most

commonly reported victimization, credit card companies were notified most often

(30.9%), followed distantly by police (18.8%), banks (15.6%) and the perpetrating

business/person (14.8%).

For those interested in the hidden figure of corporate crime, the survey data

do not allow differentiating offenses committed by legitimate businesses

(organization or employee representing the company) from those perpetrated by

individuals or illicit organizations. Unfortunately, this is more than a simple

coding problem. It derives from how questions are constructed. For example, “In

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 40

the last 12 months, has anyone succeeded in getting someone in your household to

invest money or time in a business venture such as a work-at-home plan, a

franchise, or stock purchase that turned out to be fake or fraudulent?” Similarly,

“In the past 12 months, has someone in your household paid for repairs to a

vehicle, appliance, or a machine in your home that were later discovered

unperformed OR that were later discovered to be completely unnecessary?”

Crimes by businesses are bound with those committed by individuals or illicit

organizations. The problem is also compounded in the coded response categories

for reporting the victimization. To the question “To whom was this incident

reported,” a potential response category is: “Business/person involved in the

swindle.” In only one part of the survey (hypothetical scenarios) are legitimate

organizations and their employees differentiated from individual fraudsters, but the

differences between legitimate businesses and criminal enterprises remain

unexamined.

The differences between legitimate businesses/managers, individual

perpetrators of white-collar crimes, and illicit organizational schemes are important

to tease out. Conceptually, these are quite different and distinct offenders.

Importantly, the survey also reveals that white-collar offenses are viewed as

slightly more serious than traditional crime, and that organizational level offenses

are viewed more harshly than those committed by individuals, but we do not know

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 41

the mechanisms that drive the perceptions. Is it social organization, type of crime,

degree of harm, whether a business is licit or illicit? While more clarity is needed

to answer these key questions, the issues raised demonstrate the importance of not

mixing apples and oranges in this same way for a series on white-collar violations.

Overall, victimization surveys like the NWCCC effort appear better suited to

estimate the dark figure of non-corporate kinds of white-collar offending as these

acts are more likely to be recognized and reported by crime victims. Moreover, it

is extremely difficult to measure victimization at the corporate level. Although

corporations are often victims of price-fixing, industrial espionage, insider trading,

hacking, and employee theft, the firm (like individual victims) may not know it has

been victimized.

C. Broad Measurement Models

Several projects funded by BJS have addressed the problem of building a

federal database on white-collar crime or a white-collar crime reporting system. In

the section below, we highlight details from two of these projects.

Reiss and Biderman. In their examination of the social organization of

conceptualizing, classifying, and counting white-collar law violations, Reiss and

Biderman (1980) emphasize that a successful approach to classifying and counting

illegalities, their consequences, and dispositions depends on causal models. That

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 42

is, to build a meaningful data series requires sensitivity to substantive theories of

white-collar offending, the social organization of data collection and reporting, and

the kinds of information these systems provide. Yet, the authors concluded (at that

time) that the current state of agency information made it difficult to develop social

indicators on white-collar crime.

Defining white-collar violations as those violations of law to which penalties

are attached and that involve the use of a violator’s position of significant power,

influence, or trust in the legitimate economic or political institutional order for the

purpose of illegal gain, or to commit an illegal act for personal or organizational

gain (1980: 4), Reiss and Biderman focused exclusively on federal statistical

systems of data, acknowledging multiple and varied sources or information

routinely collected on white-collar crime. Yet the authors were optimistic that

some data can be consolidated into a coherent statistical series (“uniform statistical

reporting system”) under the following requirements. Paraphrasing from the

Executive Summary of their report (xxxii-xxxvi), the authors recommend:

1. The statistics must draw from administrative, civil, and criminal

jurisdictions.

2. The series should be monitored for ways in which changes in

overlapping and concurrent jurisdiction may affect the statistical series.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 43

3. A standard definition of white-collar crime and classification procedures

for events (detection, case processing, and outcome) must be adopted.

4. Multiple counts of the same events must be taken into account.

5. Decision rules as to how to count white-collar crime violations must be

clearly defined.

6. Agencies must report information in a consistent manner to allow

merging of data.

7. Central coordination of the processing and reporting of information to

insure uniformity and compliance must be established.

8. Relevant violations of law, regulations, or standards must be

systematically and regularly reported by each and every agency, whether

it is operating under a mandate for law enforcement, regulation, or

adjudication.

9. Agencies must have explicit criteria for defining referrals and their

sources so that their sources of variability can be investigated and so that

referrals can be reliably merged.

10. Standardization of data collection, analysis, and reporting must be

achieved.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 44

11. Within the data systems, it should be possible to identify successive

violations by the same violators.

Wellford and Ingraham. Charles Wellford and Bartram Ingraham (1990)

offered a tripartite division of white-collar crime into business and professional

crimes, occupational crimes, and individual frauds (see also Clinard and Quinney,

1967). In their BJS report, the authors suggested that a federal white-collar crime

reporting system could be created by using agency data from the Securities and

Exchange Commission, the Internal Revenue Service, the FBI, and the Office of

the Comptroller of the Currency (with five federal regulatory agencies under its

umbrella). Their approach documented how specific agencies collect and manage

their case data and provided an early version of a judicial system crosswalk in

which administrative, civil, and criminal justice systems are compared for banking

violations along 11 dimensions. Specifically, data comparisons should center on:

Who can be targeted (individual or entity), the cause of the action, investigation of

the case, forum, presiding officer, rules of procedure, publicity, burden of proof,

decision-maker, appeals, and sanctions.

Both approaches are sensitive to Sutherland’s point that although criminal

laws define certain behaviors as criminal acts, responsible officials most typically

respond civilly or administratively to essentially the same acts. (However, as we

have learned from interviews conducted with regulatory agency personnel,

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 45

decisions for how cases flow into the different justice systems may also be affected

by offender characteristics.)

The definitions also are inclusive of individuals and

firms as offenders, and differentiate self-interest from organizational gain. Finally,

the two approaches correspond with data already collected by BJS that detail

offense categories by representative crime (e.g., embezzlement, fraud, forgery,

counterfeiting, antitrust, food and drug, other regulatory offenses, tax law, bribery,

wildlife offenses, environmental, all others). However, as we have already noted,

these classifications do not neatly correspond across legal venues.

Although we see the merits of both approaches, a single definitional

approach (e.g., Reiss and Biderman’s) is more parsimonious than the typological

approach of Wellford and Ingraham. Like Reiss and Biderman, our definitional

approach is parsimonious yet consistent with key elements in both approaches.

IV. Design Elements for Anticipated Data Series

Our approach to the data series on white-collar offenses incorporates a

number of key design decisions. These decisions range from defining the

phenomenon of interest—white collar offenses—to constructing a sampling frame

from which to select initial data sources that represent the range of relevant

offenses for the series, to identifying the key variables of interest for inclusion. In

The type of offender (legitimate business versus perceived bad actor) can affect the case referral process.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 46

this section we describe and explain the design decisions that together constitute

the architecture for constructing the data series.

A. A Systemic Approach to Defining White-Collar Offenses

In order to be most useful, a definition of white collar law-breaking must

meet a number of criteria. First, it should be clear in its terms and concise in its

construction. Second, it should clearly distinguish the focal phenomena from

related matters. In this case it must identify traits that distinguish white-collar

offenses from other types of violations of law. It is their distinctive traits that lend

specific theoretical significance and unique policy relevance to types of offending.

Finally, the definition should meet the test of cultural legitimacy. That is, if it is to

have analytic power and policy relevance it must resonate with both popular and

professional understandings of the nature of such offenses. For example, the

definition should distinguish the offenses of interest from the much broader

category of fraud, which contains many sorts of offenses (e.g., con games, personal

income tax evasion) that do not fit with dominant cultural notions of either white-

collar offenders or offenses.

While this is not the place for a full explication of such dominant understandings, we simply note here that most

of the research literature adopts, explicitly or implicitly, definitions of white-collar offenses that distinguish them

from garden-variety frauds. One exception is Edelhertz’s (1970: 19-20) early definition: “an illegal act or a series of

illegal acts committed by nonphysical means and by concealment or guile to obtain money or property, to avoid the

payment or loss of money or property, or to obtain personal or business advantage.” Similarly, most journalistic

coverage of white collar crime concerns cases of offenses committed for financial gain by persons and organizations

in the context of legitimate economic or professional functions.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 47

As we have noted earlier, research and programmatic statements on white

collar offenses have developed a number of definitions of this type of law-

breaking. Prominently among them are two types: offender-based definitions that

focus on attributes of the violator (e.g., status, occupation) and offense-based

definitions that focus on characteristics of the act (e.g., fraud, deception, violations

of trust). Both of these types fail to meet one or more of the criteria noted above.

What is required is a definition of white collar violations that isolates a unique

context of motivation and opportunity, one that distinguishes it from other forms of

offending and that resonates with common public and policy understandings of the

phenomenon. For purposes of this work we assert the following definition:

White-collar offenses are any violations of (federal) law committed

by persons or organizations in the conduct of their legitimate

occupational roles or organizational functions.

We emphasize a number of features of this definition. First, it

parenthetically references federal law only because of the scope of this project,

although we have no reason to believe that the definition could not be applied

usefully at the state level. Second, it includes violations committed by both

individuals and organizations. By organizations we mean any formally organized

and recognized entities pursuing legitimate ends in any of society’s institutions,

including the market, government, religion, politics and the not-for-profit sector

(e.g., charities, private schools, civic organizations).

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 48

Third, the definition does not limit offenses to only those prosecuted under

criminal law. Instead, it includes all violations of federal laws whether they are

sanctioned by administrative (regulatory), civil or criminal law processes.

Researchers from Sutherland (1949) onward have long noted that many types of

white collar offenses are handled largely with noncriminal legal processes, and

criminal prosecutions of corporations are especially rare, a fact recently manifested

in the pattern of sanctions employed against responsible financial organizations in

the wake of the 2008 financial crisis. Notably, decisions as to whether to prosecute

cases criminally or to proceed with noncriminal procedures commonly reflect

preferences and practical considerations (e.g., relative burdens of proof, available

legal resources, collateral consequences, resistance to criminal prosecution by

powerful violators) rather than only assessments of criminal blameworthiness (see,

e.g., Brown, 2001; Yeager, 2015). It is noteworthy, too, that in comprising the full

range of the federal government’s sanctioning processes the definition asserted

here corresponds to those employed in the two earlier studies that took up the

matter of formulating a national data system for white collar offenses (Reiss and

Biderman, 1980; Wellford and Ingraham, 1990).

Fourth, rather than being other offense- or offender-centered, our definition

is role-centered. Its subject is offenses that abuse or misuse legitimate

occupational or professional roles to achieve a benefit, whether material or

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 49

psychological, and whether personal or organizational. Constraining the definition

to legitimate work roles is not a mere academic matter. It emphasizes instead that

the motivations, temptations and opportunities afforded by these roles have

distinctive implications for public policy approaches to punishing and limiting

such offenses, implications that differentiate them from other types of offenses that

manifest similar behavioral forms (e.g., individual check-kiting, organized scam

rings).

Relatedly, the definition is not statute-restricted. It does not restrict white-

collar offenses only to those infractions that violate (and are pursued under) laws

expressly aimed at the control of commercial or occupational activities. So, for

example, violations of federal Occupational Safety and Health Administration

(OSHA) rules that lead to the death of workers and to a prosecution of corporate

(and/or individual corporate managers) under the law of homicide are included as

white-collar offenses. So would be the theft of credit card numbers by a food

server, because the employee’s legitimate occupational role provided the unique

opportunity (and perhaps some of the motivation) for the offense. In contrast,

members of organized credit card theft rings would not be included among white-

collar offenders because they offend in the context of illegitimate roles.

In sum, the project definition of white-collar offenses locates intent in

occupationally- and organizationally-situated motives and opportunities,

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 50

distinguished by the dimension of legitimacy. The mode of behavior involves

abuses of relations of power, authority and trust related to legitimate occupations

and organizations. The actors are commonly violators who largely commit only

such offenses and no other types, such as conventional violent and property crimes.

In terms of numbers of lives affected and dollar losses these offenses are among

the most serious violations of law that societies confront. Notably, however,

unlike the case for conventional crimes the seriousness of white-collar offenses is

not well-indexed by the gravity of the sanctioning responses to them. Criminal

sanctions are rare, especially for the offenses of powerful perpetrators, and even

large civil fines are small relative to the size and wealth of many such offenders.

Definitional Extensions. Finally, the project definition can be situated

among other, related definitions of offenses in a typological manner. This exercise

not only helps to demonstrate axes of connection between our working—or core—

definition and broader ones, but it also provides avenues of extension that can be

employed in the face of constraints in existing federal data systems for white-collar

offenses. For example, some such systems may lack information regarding the

role in which the perpetrator was acting when the offense was committed. To the

extent that they are used, any such extensions should be as prudent and limited as

possible if the construct of white-collar violations is to meet adequately the key

definitional criteria noted above.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 51

One way to conceptualize the relationship between our core definition and

broader constructs is illustrated in Figure 2.

Figure 2

The core can be extended outward to draw in other offenses such as

illegitimate enterprises feigning legitimacy; individuals taking advantage of

D. Core Definition Plus B Plus C

Plus Offenses Committed by con

artists, credit card and identity

thieves, Nigerian scammers who

create criminal opportunities.

C. Core Definition Plus B Plus

Offenses Committed by citizens that

exploit existing opportunities (e.g.,

income tax fraud, welfare fraud, bad

checks)

B. Core Definition Plus Illegal

Enterprises that mimic legitimate

organizations (e.g., OC, diploma

mills, fraudulent medical clinics,

home improvement scammers)

A. Core Definition: Offenses

committed in course of legitimate

occupational /organizational roles

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 52

existing opportunities to deceive, or criminals creating new opportunities to run

scams and cons. The first stage extension (B) from the core definition is to

offenses that most resemble those in the core definition. These are the full-time

criminal activities organized to mimic legitimate businesses and professional

organizations and to use the masquerade to illegally enrich the perpetrators. Here

are diploma mills, fake medical clinics, home improvement scammers and even

organized crime offenders, to the extent that the latter control ostensibly legitimate

operations as either fronts for or means of illegal activities (e.g., unions, casinos,

waste removal companies). These offenders are often sustained and enhanced by

highly organized routines and relationships that are similarly illegitimate but often

quite sophisticated.

The next extension (C) is to offenses committed by people whose law

violations are neither organized in criminal enterprises nor committed in

occupational roles. Persons may commit these offenses more than once, but they

are neither orchestrated with others nor are the perpetrators career criminals in the

sense that they organize their lives importantly—even principally—around the

commission of crimes. Here are citizens committing income tax fraud, writing bad

checks, or lying to gain welfare benefits to which they are not legally entitled

(typically committed by poor persons on the margins of eligibility). These

offenses are similar to those comprised by the core definition on a number of

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 53

dimensions relevant to many WCC definitions in the literature: nonviolent

fraudulent acts enabled by (non-occupational) aspects of roles and circumstances

of trust.

The final extension (D) adds ‘career’ offenders, those persons whose income

generation and identities are organized principally—or in large measure—through

illegal activities. They work either alone or in small informal groups of

confederates. These offenders include such offenders as insurance fraudsters,

computer hackers, identity thieves, and others whose livelihoods and identities are

organized around criminal acts.

All of the extensions illustrated in the figure add types of offenses and

offenders that raise matters of public policy and social control that differ from

those associated with the project’s core definition. Hence the need to consider

carefully the implications of making any such extensions in the face of limitations

in existing federal data sources.

B. Offense Domains

Having defined the matter of interest, the next step in the construction of a

data series is identifying the range and types of white collar offenses, and the data

repositories for them. There is a broad range of such violations under U.S. law,

and dozens of organizations and web sites that hold and/or publish data on them.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 54

Mapping these offenses and data sources is necessary for both sampling and

analytic purposes. Given the nature and size of the task, building a federal data

series on white collar violations is a developmental process that necessarily will

proceed in stages over a period of years. It entails forging relationships with

agencies, arranging ongoing mechanisms for data sharing, formatting data from

diverse sources into standard categories for counts and analyses, and working with

cooperating agencies over time to improve progressively the quality of data. To

begin the task, therefore, we need a sampling frame of the universe of offenses

from which to choose initially selected offense types and the organizational

sources of data on them.

This step is also important for analytic reasons. Offense classifications by

type are useful to the search for explanations, which are often crime- or person-

centered, an idea long established in research on conventional offenses (see, e.g.,

Gibbons, 1965; Clinard and Quinney, 1967; Moffitt, 1993; Clinard et al., 1994) but

challenged of late by generalists (Gottfredson and Hirschi, 1990). White collar

offenses have also been sociologically typified, if not in as detailed a manner as

conventional crimes. In particular white collar violations in business can be

classified according to the institutional spheres in which they occur. This

classification distinguishes between economic and social regulation (see, e.g.,

Yeager, 1993: 9; Wiedenbaum, 1979). Economic regulation refers to rules and

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 55

agencies that sanction violations of the rules of market competition and exchange,

the rules essential to the efficient functioning of capitalist markets. Social

regulation, on the other hand, sanctions violations of rules that protect against the

harmful effects of market activities—market ‘externalities’ in the language of

economics—such as environmental pollution, the manufacture and sale of unsafe

products, and the exposure of employees to unsafe work conditions.

Each sphere comprises a number of types—or domains—of regulations and

their associated violations. Economic regulation includes rules that protect against

abuses in the financial system (e.g., misleading investors, insider trading, contract

fraud

), against unfair

competitive behaviors (e.g.,

monopolistic practices,

price-fixing), and against

unfair manipulation of consumers (e.g., false advertising, selling fraudulent

products and services). We illustrate this sphere and its associated domains of

enforcement in Fig. 3.

The most publicized types of contract fraud are those against government programs, such as overcharging in

military contractor fraud and Medicare and Medicaid fraud. Such frauds can also victimize private-sector

businesses as well. By extension one can also include here the various forms of consumer fraud and employee theft,

as these offenses violate the rules of fair and efficient market exchange.

Figure 3

Domain

Sphere

Economic Regulation

Financial

System

Market

Competition

Consumer

Protection

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 56

Social regulation includes rules that protect against harms to the

environment (e.g., air

and water pollution),

against the marketing of

unsafe products to

consumers (e.g., autos,

pharmaceuticals, toys), and against violations of employee rights (e.g., unsafe

workplaces, discrimination in hiring and promotion).

Fig. 4 illustrates this sphere

and its domains of enforcement. There is great variability within these specific

domains as to statute, offense and offender types, offense seriousness (scope and

harm), and enforcement responsibility. Yet, this kind of framing or classification

will be more likely to yield cases consistent with our working definition for the

series than an approach that organizes white-collar offenses by statute, broad

offense types (e.g., fraud), or enforcement response (criminal, civil, or regulatory).

The domain of employee rights includes the offenses of sexual harassment in the workplace and sexual assault of

co-workers. Just as does the former type of offense, sexual assault of co-workers is enabled by the context of

employment and the power relations within that context. This is an example of a crime that would be prosecuted

under conventional assault statutes but that constitutes a white-collar offense because of the specific structures of

motive and opportunity that condition its occurrence. Similarly, and for the same reason, the sexual assault of a

client or customer in the workplace would also constitute a white-collar offense.

Figure 4

Domain

Sphere

Social Regulation

Environmental Product Safety

Employee

Rights

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 57

C. Sampling Domains

Given the breadth of white collar offenses and the large number of federal

data systems that track them, it is necessary to launch the data series with a small

sample of enforcement data. We employed a number of criteria for selection of the

first sets of offenses for inclusion in the series. First, they should be representative

of the spheres of offenses as defined above. Second, each set should comprise

offense types that are sanctioned with the array of federal enforcement responses:

administrative/regulatory, civil and criminal sanctions. Third, the selection should

target agencies whose data on offenses are plentiful, accessible, and that generally

include largely types of violations that meet the definitional criteria for inclusion in

the series, as defined above. Underlying this sample selection plan—to begin with

the most accessible and relevant data (the ‘low hanging fruit’)—were the goals of

demonstrating both the potential for and some of the challenges facing the series,

and building a foundation of data for analysis that would demonstrate to other

agencies the value of their eventual participation in the series.

Numerous agencies and departments are charged with enforcing the laws

against the broad array of federal white collar offenses. Principal organizations are

listed in Table 1. These organizations vary substantially in the numbers of laws

and regulations they enforce, the volume of offenses they process annually, and in

the structures, operations, and accessibility of their enforcement data systems.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 58

Table 1: Federal Agencies Regulating

White Collar Offenses

Centers for Medicare and Medicaid Services

National Labor Relations Board

Commodity Futures Trading Commission

Nuclear Regulatory Commission

Consumer Financial Protection Bureau

Occupational Safety and Health Admin.

Consumer Product Safety Commission

Office of the Comptroller of the Currency

Environmental Protection Agency

Securities and Exchange Commission

Equal Employment Opportunity Commission

U.S. Coast Guard

Federal Aviation Administration

U.S. Department of Agriculture

Federal Deposit Insurance Corporation

U.S. Department of Commerce

Federal Energy Regulatory Commission

U.S. Department of Defense

Federal Housing Finance Agency

U.S. Department of Energy

Federal Reserve System

U.S. Department of Health & Human Services

Federal Trade Commission

U.S. Department of the Interior

Food and Drug Administration

U.S. Department of Justice

Internal Revenue Service

U.S. Department of Labor

National Highway Traffic Safety Admin.

U.S. Department of Transportation

U.S. Department of Treasury

Other federal units that keep data on white collar offenses are the Administrative

Office of the U.S. Courts, the Executive Office for U.S. Attorneys, and the U.S.

Sentencing Commission.

Based largely on our experiences over the years in working with federal

agency enforcement data,

early on in this project we settled upon the

Environmental Protection Agency (EPA) and the Securities and Exchange

Commission (SEC) as our primary regulatory agencies of focus, eventually

supplemented by examination of publicly accessible data from the Federal Trade

Collectively we have conducted research on data from the majority of these federal agencies, but with greatest

experience with those from the Environmental Protection Agency, the Occupational Safety and Health

Administration, and with anti-competitive/antitrust offense data. In addition, in recent years we have discussed

enforcement data with information technology specialists and enforcement officials at the Securities and Exchange

Commission, the Food and Drug Administration, the U.S. Department of Justice, the Occupational Safety and

Health Administration, and the Environmental Protection Agency.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 59

Commission (FTC), the new Consumer Financial Protection Bureau (CFPB), and

the Food and Drug Administration. Importantly, the EPA and SEC met the

selection criteria noted above, although accessibility proved to be more challenging

than we had anticipated (see Section V).

The EPA, established in 1970 by President Richard M. Nixon, is a major

federal agency in the sphere of social regulation. It undertakes or oversees the

majority of enforcement activities under the nation’s environmental laws, although

it shares this regulatory domain with several other units of government, including

the Departments of the Interior, Agriculture and Commerce, and the U.S. Coast

Guard. In particular the EPA is responsible for enforcing the nation’s principal

environmental laws, including the Clean Air Act (42 U.S.C. §7401 et seq.), the

Clean Water Act (33 U.S.C. §1251 et seq.), the Resource Conservation and

Recovery Act (42 U.S.C. §6901 et seq.), the Comprehensive Environmental

Response, Compensation, and Liability Act (also known as the ‘Superfund’ law)

(42 U.S.C. §9601 et seq.), the Safe Drinking Water Act (42 U.S.C. §300f et seq.),

and the Federal Insecticide, Fungicide, and Rodenticide Act (7 U.S.C. §136 et

seq.).

Violations of environmental laws are sanctioned with the range of

enforcement responses. The EPA employs non-judicial administrative civil actions

(notices of violation, administrative orders), civil judicial actions and criminal

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 60

actions. In the latter two types of sanctioning the agency refers cases to the U.S.

Department of Justice for filing in the federal courts. The agency also shares

enforcement with many of the states’ regulatory authorities to whom EPA has

delegated responsibility for enforcement of the laws. The vast majority of

violations of federal environmental laws fit the core definition of white collar

offenses (committed in the pursuit of occupational or organizational roles).

Finally, the EPA has long maintained enforcement and compliance data systems

that are among the most advanced among federal law enforcement agencies, if not

the most advanced. Indeed, the agency makes its extensive violations and

enforcement data available on-line at the EPA web site (www.EPA.gov), and

during the project period the agency was further refining its data systems in the

direction of greater integration of its enforcement data.

The Securities and Exchange Commission, established in 1934, is a principal

agency in the sphere of economic regulation. In the domain of rules protecting the

integrity and efficiency of the financial system in the U.S., it shares enforcement

responsibilities with the Federal Reserve, the Treasury Department, the Office of

the Comptroller of the Currency, and the Commodity Futures Trading

Commission. The SEC has primary responsibility for enforcing the nation’s

This point was underscored in a March 2013 meeting that project staff had with several EPA officials. An EPA

staffer asked how we were defining white collar offenses and, upon hearing the definition, replied that virtually

everything that EPA did fit the definition.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 61

securities laws, including the Securities Act of 1933 (15 U.S.C. §77a et seq.), the

Securities Exchange Act of 1934 (15 U.S.C. §78a et seq.), the Investment

Company Act of 1940 (15 U.S.C. §80-1 et seq.), the Investment Advisers Act of

1940 (15 U.S.C. §80b-1 et seq.), the Sarbanes-Oxley Act of 2002 (Pub.L. 107-204,

116 Stat 745; codified at various sections of the U.S.C.), and the Dodd-Frank Wall

Street Reform and Consumer Protection Act of 2010 (Pub.L. 111-203, 124 Stat.

1376; codified at various sections of the U.S.C.).

The SEC shares enforcement responsibilities under some of these laws. For

example, under Dodd-Frank it shares enforcement responsibility with the new

Consumer Financial Protection Bureau that was created by the statute, and under

Sarbanes-Oxley it shares responsibility with the Public Company Accounting

Oversight Board (again, created by the statute). It also shares securities

enforcement responsibility with a number of self-regulatory organizations, such as

the Financial Industry Regulatory Authority (FINRA) and the Municipal Securities

Rulemaking Board, and with 20 registered stock and futures exchanges.

The SEC also utilizes the full range of sanctioning activities. It employs

administrative sanctions such as cease-and-desist orders and securities trading

suspensions, and files civil suits in the federal district courts seeking such sanctions

as injunctions, monetary penalties and disgorgement of illegal profits. It refers

cases of suspected criminal activity to Department of Justice prosecutors. It is

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 62

likely that most enforcement cases fit the core definition of white collar offenses,

although some proportion may involve frauds and thefts committed outside of

legitimate occupational or organizational roles (Schlegel, Eitle, and Gunkel, 1994).

According to agency personnel the SEC maintains an effective enforcement case

tracking system,

and the agency’s publicly available data (www.SEC.gov)

provide quite detailed information on the nature of the offenses and offenders in

cases it sanctions.

D. Unit of Count

As with any data series, a key decision is determining what is to be counted.

In criminological data, researchers and policy makers may be interested in any of a

number of types of counts that are distinct from each other: offenses, offenders,

cases. For example, in prosecutions of corporate crime a single legal case may

charge a number of offenses against several individuals, as well as against the

organization itself. In addition, in such matters the federal government may bring

parallel civil and criminal cases against all or a subset of the defendants. In

creating the data series on white collar offenses, the goal is to specify a focal unit

of count that enables the analysis of the range of counts that are of policy and

conceptual interest.

Phone meeting with SEC enforcement personnel, May 21, 2013.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 63

Toward that end we have specified the case-defendant as the unit of count.

In the first instance this enables counts of the number of violators of particular

types and time periods. Case and other identification numbers for the legal actions

taken by the Government will also permit analyses of cases, for example by type or

time period.

It will be necessary to use care in identifying and linking defendants and

cases where case numbers for the same legal matter vary across regulatory and

criminal justice units, such as the regulatory agencies, the Executive Office for

U.S. Attorneys, and the Administrative Office of the U.S. Courts. The same may

be said for linking parallel civil and criminal proceedings against the same legal

matter.

E. Key Data Points for the Series

The series on white collar offenses should comprise data generated by the

key stages in case processing of violations. These elements include information on

how the suspected offenses were detected, number and nature of offenses or

charges, number and characteristics of offenders (e.g., individuals or

organizations), and sanctions imposed.

Figure 5 illustrates the general agency case management processes through

which the relevant enforcement data are generated.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 64

Figure 5

The processes begin with the initial detection of offenses, whether by regulatory

agencies or the Department of Justice, and whether by internal investigations,

outside complaints or self-reports. Depending on the nature of the violation, the

detecting organization may forward the matter to a more appropriate investigative

entity. There follows further investigation that determines facts, evidence and

responsible parties, and either the pursuit of charges against defendants or case

dismissal. Cases pursued by regulatory (administrative) agencies typically result in

the application of administrative or civil sanctions, or the referral of cases to

federal prosecutors for civil or criminal filings in the courts. The Department of

Agency Offense

Detection

External Complaints

Internal Investigations

Self-Reports

Agency Case

Processing

Investigations

Enforcement Actions

Dismissing Charges

Case Resolution

Administrative Sanctions

Civil Penalties

Referrals to DOJ

DOJ Case Processing

Cases Declined

Counts Dismissed

Civil Cases Filed

Criminal Cases Filed

Case Resolution

Civil Penalties/Order

Criminal Penalties

Cases Dismissed (not guilty/not

liable)

DOJ Offense Detection

External Complaints

Internal Investigations

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 65

Justice prosecutors may take any of several actions in response, including filing the

cases or dropping some or all of the charges. The Department may also generate

its own cases through complaints received and internal investigations, through such

units as the Criminal Division and the Environment and Natural Resources

Division.

In civil cases filed by U.S. attorneys, typical sanctions being sought include

monetary penalties/restitution and court orders (whether consent orders or imposed

by the court). In criminal cases the sanctions include fines, probation and terms of

incarceration. Notably in cases of corporate law-breaking, criminal prosecutors

may use non-prosecution or deferred prosecution agreements with offenders. Such

agreements operate much like pre-conviction probation in allowing offending

companies to cease their violations and to implement changes to prevent future

offenses, in exchange for prosecutors’ agreement not to pursue formal charges

against them. In a minority of filed cases defendants will be found not guilty or

not liable.

Coding of the variables of interest is to be done at the defendant level rather

than the case level, in keeping with the unit of count, the case-defendant. Figure 6

illustrates schematically the extraction of case-defendant data from agencies’ data,

which are commonly registered in terms of enforcement cases or matters.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 66

Figure 6

Agency

Case

Case Number

Date

Initiated

Defendants

Individuals

Organizations

Defendant A

Defendant B

Defendant C

Charges

Counts

Charges

Counts

Charges

Counts

Resolutions/Dates

Pending

Sanctions*

Referrals**

Resolutions/Dates

Pending

Sanctions*

Referrals**

Resolutions/Dates

Pending

Sanctions*

Referrals**

*Sanctions range from criminal penalties, through civil

monetary penalties and order, to administrative warnings

and orders.

**Referrals to other agencies for further processing,

especially Department of Justice.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 67

The variables of interest to the data series range from information on the

origins of cases (in investigations, complaints or referrals), through information on

offenses and offenders, to case resolution data. Key to the design are case and

defendant identifiers, which among other purposes are necessary for tracking of

cases through the data systems managed by the various enforcement agencies that

process them. The identifiers include case numbers, nature of defendants

(organizational or individual), and defendant characteristics (age, gender,

occupation, type of business or industry). Violations data of interest include the

type of legal action taken (administrative, civil or criminal), the number and types

of offenses charged, dates of violations and of initiation of legal actions, and

seriousness of violations. Case resolution data comprise number and types of

violations sanctioned, types of resolution (e.g., dismissal, fines, incarceration),

amounts of any fines or periods of incarceration, whether the case(s) was referred

to another agency, and whether there were appeals. An important aspect of the

data series is the ability to track what happens to defendants who are diverted out

of the legal system or from one system to another.

We anticipated that the various agency and DOJ data bases would not

routinely collect information on all of the factors of interest to the series. To

ascertain the degrees to which the data systems included these factors we

canvassed them using the form shown in Table 2.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 68

Table 2: Inventory of Data Elements

Data Elements

Available

Not

Available

Data

Source

Comments

Case Origin

Internal Investigation/Referral from

another agency/Complaint

Case Number(s)

Defendant(s) Identities:

For Organizational Defendants:

Distinction between legitimate

and criminal purpose organizations

For Individual Defendants:

Age of Defendants

Gender of Defendants

Occupation/Positions of Defs.

Address(es) of Defendant(s)

Type of Industry/Business

SIC, NAICS, Other Indicators

Type of Legal Action Taken

Administrative/Civil/Criminal

Data Action Initiated/Filed

Violations Alleged (each defendant)

Number of counts by statute

Number of counts by regulation

Other Description of Offense(s)

Date(s) of Alleged Violations

Seriousness of Violations

Date(s) of Case Resolution

For each defendant

Violations Penalized

For each defendant

Number of counts by statute

Number of counts by regulation

Resolution(s) (each defendant)

E.g., Case dropped, Order, Injunction,

Fine, Disgorgement, Incarceration

Fine/Disgorgement Amounts

Probation (months)

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 69

Incarceration (months)

Case(s) referred to other agency

For each defendant

Agency to which case(s) referred

Case numbers for referred cases

Resolution(s) of referred cases

For each defendant (e.g., referrals

declined; charges/counts penalized;

penalties imposed)

Case(s) Resolution(s) Appealed

Venue of appeal(s)

Outcome(s) of appeal(s)

F. Approaches to Data Gathering

We employed several methods to determine the nature, quality and

accessibility of agencies’ enforcement data for white collar offenses. These

include meetings with agency representatives, interviews with both present and

former agency officials, and assessments of the publicly available enforcement

data on agencies’ web sites, as with the use of Table 2. In Section V below we

describe the results of these efforts, including the focused data briefs the

researchers prepared for the Securities and Exchange Commission, the

Environmental Protection Agency, the Federal Trade Commission, the Food and

Drug Administration, and the Consumer Financial Protection Bureau.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 70

V. Assessing Available Federal Data

A. Introduction

In this section we describe the findings from our research into agencies’ data

systems. Our research efforts comprised: (1) a comprehensive web search of

federal departments, commissions, and regulatory agency sites to describe

available data that might be useful for a white-collar crime series; (2) meetings

with officials from both regulatory agencies and units of the U.S. Department of

Justice; and (3) close inspection of the publicly available data for several of these

agencies.

The research team compiled a list of federal agencies with enforcement

responsibilities and data that could fit (at least in part) with our definition of white-

collar violations. Forty-four separate entities were identified in this manner.

Websites were then searched to discern information availability and access as well

as the kind of data collected and held. Finally, to get a better sense of the scope of

the agency data, we calculated the percent of total criminal referrals the agency

contributed in 2010 using an extract from the LIONS database.

The meetings ranged from gatherings with members of several agencies to

those with officials from single agencies. A few of the meetings took place in

phone conversations. The officials included enforcement attorneys, information

technology personnel, and other specialists. The agencies represented in these

discussions include the fraud and environmental crime units of the Department of

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 71

Justice, the Environmental Protection Agency (EPA), the Federal Trade

Commission (FTC), the Securities and Exchange Commission (SEC), the

Administrative Office of the U.S. Courts, the Office for Victims of Crime, and the

Department of Health and Human Services. Publicly available violations and

enforcement data were examined as presented on the agency web sites of the EPA,

SEC, FTC, FDA and CFPB.

Project staff also examined the criminal and civil data contained in the

Federal Justice Statistics Program for their potential utility in a white-collar crime

series. PACER (Public Access to Court Electronic Records), a public electronic

database of federal court cases, was used to cross-check FJSP data for

completeness.

In this section, we first report what we discovered from agency website

searches. Next, we review our findings and conclusions from meetings and

interviews with agency enforcement and information technology personnel. We

then discuss the utility for a data series on white-collar offenses of the data

publicly available on selected agency web sites. Finally, we review the FJSP data

for their potential value for such a series. The aim in the section is to establish

both strengths and weaknesses of data sources in order to determine their current

levels of utility for the series as well as needed future developments to improve

their usefulness.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 72

B. Findings and Conclusions from Comprehensive Review of Agency

Websites

As noted above, the federal agency search identified 44 departments,

commissions, centers, agencies, and bureaus that could hold relevant data for the

series. The mission statement for 33 of these agencies fit with our definition of

white-collar crime for this project (the fit was closer for some than for others). Of

these agencies, 12 had online data and/or codebooks available; 14 had some

accessible but limited information; and seven had no online data or codebooks that

could be searched.

In terms of the kind of information gleaned from the sites, we assessed

whether data were publicly available, partially available or restricted; what

enforcement data were reported (e.g., final orders, stipulations, inspections),

including the form it took (quantitative data, PDF); the number of cases reported,

the years for which data are available, and the types of offenses reported (i.e., the

scope of the data); and the breadth of legal coverage (criminal, civil judicial and/or

administrative).

Overall, the search revealed that several available on-line data sources may

be useful for the white collar violation series in both the social (OSHA, EPA,

MSHA, and NLRB) and economic (SEC, OCC, FDIC) enforcement realms.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 73

However, the data may not be readily quantifiable (reported in PDF files, for

instance), in which case data scraping would be required to access and code

relevant information from the sites. Further, sites report data at different points in

the decision process. The Federal Reserve and the NLRB, for instance, report only

final orders, whereas the International Trade Administration within the Department

of Commerce, releases summary information (limited) that describes

investigations. Some sites report only civil or administrative outcomes while

others offer a mix of enforcement outcomes (such as the Environmental Protection

Agency and the Antitrust Division of the DOJ).

There is also great variety in the level of case detail provided by each

agency. In some sites (see, e.g., Federal Housing Finance Agency), cases are

described in great detail (from when the alleged illegal activity occurred, nature of

the action, who is named as co-defendants and their role in the violation, factual

allegations, to the specifics of the enforcement action). Other sites highlight yearly

accomplishments in which select (and therefore incomplete) cases and outcomes of

interest are described (see, e.g. http://www.ftc.gov/reports/annual-highlights-

2013/enforcement) or aggregate counts of enforcement activities such as total

inspections, enforcement actions, and violations are summarized.

This review confirmed the original suspicions of the research team regarding

the varied nature and quality of online data. However, the search did reveal that

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 74

some agency data are better (more comprehensive, broader in scope, and more

easily accessible) than others. For these reasons, we recommend that the white-

collar crime series incorporate the most complete and comprehensive data systems

from the available agencies. To examine and describe the extent to which agencies

meet these criteria, we have developed five agency briefs. The goal of these briefs

is to describe the full enforcement terrain that is captured within agency data and

show how that portrait comports with data collected from other sources.

C. Findings and Conclusions from Discussions with Agency Personnel

The purposes of project staff’s discussions with agency personnel were (1)

to communicate the nature and needs of the proposed data series on federal white-

collar offenses; (2) to ascertain the nature of agencies’ violations and enforcement

data systems and their potential usefulness for the series, and (3) to encourage

agencies’ involvement in future activities toward the development of the series,

including the regular sharing of data for inclusion in it.

In summary form the key conclusions from these discussions are several:

 Agencies vary substantially in the structures and the qualities of their

enforcement and compliance data systems. The data systems range

from those comprising largely legal documents (e.g., court filings,

administrative orders) kept in electronic form in computer files to

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 75

those comprising developed electronic records that track the

compliance and enforcement records of the regulated entities over

time.

 The agency data systems have been created first to serve the needs of

investigative and enforcement personnel in the conduct of their work

(e.g., tracking case developments and workload allocation) and,

secondarily (in some instances), for use by citizens seeking on-line

information on compliance and enforcement. In addition, agency

data on compliance and enforcement can be responsive to changing

enforcement priorities and budgets, complicating analyses of

enforcement trends. There has been no consistent or government-

wide template of key factors for agencies to track as investigations

unfold, as cases are filed and work their way through the legal

process, and as sanctions are issued for violators. Notably the data

systems were not created with an eye toward their utility for external

research or for combination with other agencies’ enforcement and

compliance data.

 Agency personnel were frank in discussing the limitations of their

data systems, whether for their own investigative/enforcement

purposes or for the purposes of research. Challenges include

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 76

variations across federal regions and districts in the completeness and

consistency of data input and the lack in some agencies of electronic

data files that can be easily searched for purposes of constructing

counts and rates of offenses, offenders and case outcomes (e.g.,

sanctions). Linking specific cases across agencies and even within

agencies presents additional challenges.

 Personnel from a number of agencies indicated either that their

agencies were pursuing improvements in their data management

systems or were interested to do so. Some suggested that the BJS

effort to construct a data series on white collar offenses might

provide the opportunity for agencies to share information on best

practices for the development of their data systems.

 The BJS effort to have agencies enter into memoranda of

understanding (MOUs) to share their enforcement and compliance

data with the Bureau for purposes of the data series stalled in both

efforts made during the present project. After successful initial

meetings with officials at both the Environmental Protection Agency

and the Securities and Exchange Commission, during which the

nature and characteristics of their data were discussed, project staff

efforts to engage in subsequent discussions regarding the terms of

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 77

data sharing with BJS failed when agency personnel became

unavailable to continue the discussions. Somewhat speculatively we

infer that among the possible reasons for the cessation of these

conversations are the perceived budgetary costs to the agency of

selecting and preparing data for transfer on an ongoing basis, and

uncertainty about the potential political risks to the agency of data

sharing, such as those that might arise from second- or third-party

analyses of enforcement and compliance data from single agencies or

from agencies in combination.

What follows are discussions of key points from program staff conversations with

agency officials that underlie the above conclusions.

Federal Criminal and Civil Data. In a separate section below we assess

the strengths and weaknesses of the criminal and civil data as presently included in

the Federal Justice Statistics Program. Here we note relevant points about the

criminal and civil data made by Department of Justice officials in our project

meetings.

The Justice Department to date has not organized its data processes with an

eye toward measuring the incidence and prevalence of white-collar offenses.

Instead, as with most agencies the focus has been on creating and managing data

systems that meet its programmatic needs for prosecuting and tracking cases. One

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 78

important limitation of the existing data systems is that offenses are not identified

with the level of detail needed to categorize them as white-collar offenses as

defined in this project. For example, the data kept by the Administrative Office of

the U.S. Courts register offenses by statute rather than by more precise descriptors

of the offenses. Many cases are prosecuted under various fraud statutes, but these

laws cover numerous crimes that would not qualify as white-collar crimes, such as

fraudulent individual claims against federal welfare programs or frauds perpetrated

by organized criminal groups.

In addition, this data system does not consistently

record whether organizations are defendants in cases, and does not make possible

the discovery of parallel civil and criminal cases against the same violators.

LIONS (Legal Information Office Network System) is the system that

tracks U.S. attorneys’ monthly case data. It records only the first offense listed in

an indictment, and not necessarily the most serious violation. In addition, the data

system is said to be missing considerable information. For example, it may not

include some major cases brought by litigating units (e.g., inspectors general)

without the involvement of U.S. attorneys’ offices. On the other hand, if criminal

Two possible remedies for this are consulting the fee-based data system PACER (Public Access to Court

Electronic Records) to access specific cases’ docket sheets, a time-intensive process, and to rely on information on

referring agency (where such information is available), since referring agencies such as the EPA and the SEC

generally pursue offenses that constitute white-collar violations as defined. The strengths and weaknesses of using

PACER to cross- validate case data for this project are described in the next section of this Report.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 79

cases are shared by both U.S. attorneys and other litigating units, and data are

available from both sources, it is important to avoid double-counting cases.

Another potential source of data for white-collar offenses are specialized

investigative and prosecutorial units, such as the Health Care Fraud Prevention and

Enforcement Action Team (HEAT) and the Financial Fraud Enforcement Task

Force. These are multi-agency groups formed in 2009 to combat federal offenses

such as fraud against the Medicare and Medicaid programs and financial frauds

such as mortgage-backed securities offenses, procurement frauds, identity theft and

Ponzi schemes. These units have their own data bases and may be a rich source of

information on cases of white-collar offenses. Department of Justice officials

raised a number of cautions about such data, however. The health care fraud data

contain information on a number of important factors: filing date, type of fraud,

unique offender identifier, name of company, etc. But if the cases are brought only

by assistant U.S. attorneys attached to local (federal) district strike forces, the cases

may not be included in the centralized data system (the Legal Information System

Network, or LIONS, which is part of the FJSP data under the EOUSA heading).

Similarly, case data associated with the Financial Fraud Enforcement Task Force

are decentralized among numerous working groups. It was noted that data

management was not among the official goals for the Task Force, and that data

collection was fragmented.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 80

A further implication for measuring white collar offenses of data compiled

by such task forces is that their prosecutorial work necessarily focuses on specific

priorities, such as “target rich” areas where rates of offending are expected to be

high and types of offending that pose the greatest public and policy-level concerns

(e.g., mortgage fraud and related securities frauds). Such priorities shape the

profiles of offenses and offenders in particular ways that may not be representative

of the universe of such violations and violators, a consequence that is only made

more salient by the fact that enforcement priorities often shift over time. Of course

the same can be said of virtually all forms of enforcement against white-collar

offenses, whether by criminal, civil or regulatory authorities.

Department of Health and Human Services: The Data Bank. Another

example of a focused source of white collar offense data are two data systems kept

by DHHS that track sanctions for misconduct by medical practitioners and

companies in the health care field. The National Practitioner Data Bank compiles

reports from such sources as malpractice insurers, hospitals, state licensing boards

and professional societies. These sources report sanctions they have imposed on

practitioners for violations of professional standards and related rules, such as

license suspensions and revocations and censures. The Healthcare Integrity and

Protection Data Bank compiles reports from federal and state government

agencies, and from health plans. The reports concern sanctions ranging from

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 81

reprimands and license revocations to civil judgments and criminal convictions.

For the latter data LIONS is the key source. Current planning calls for these two

databases to be merged. In general this is a quite comprehensive set of data and

one that would add an important component to a data series on white collar

offenses: those committed by professionals working in a domain vital to both

individual and public well-being.

DHHS officials noted some caveats regarding these data. Consistency in the

reporting and entering of data is uncertain. Among other reasons, DHHS cannot

compel hospitals and other medical entities to report sanctions. Officials were also

uncertain about the completeness of criminal case data. Because the purpose of the

database is to ascertain simply whether criminal convictions were related to

medical misconduct, information on sanctions is unnecessary for the purpose and

may not be included in the data. Nonetheless such missing data would appear to

be available, requiring only procedures for regularly extracting it from sources

such as LIONS.

Regulatory Agency Data. Project staff had detailed conversations with

enforcement and/or information technology personnel at three major federal

regulatory agencies: the Environmental Protection Agency (EPA), the Securities

and Exchange Commission (SEC), and the Federal Trade Commission (FTC).

These agencies were selected because they represented enforcement against white-

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 82

collar offenses in the spheres of social (EPA) and economic regulation (SEC and

FTC). The EPA and SEC were also selected because experience and project team

research has shown that these two agencies maintain substantial data bases on

offenses, offenders and enforcement. Further, each agency is responsible for most

enforcement cases while both offer assistance (including investigative and staff) to

DOJ and US Attorney’s Offices in the pursuit of criminal cases. Thus, EPA and

SEC also track some criminal case files which may be useful to assess and verify

case data from other sources. It is also important that the three agencies enforce

cases largely against white-collar violations as defined in this project. Their data

include little “noise” of the sort represented by cases against organized criminals or

persons not offending in the course of their legitimate occupations.

Environmental Protection Agency

The U.S. Environmental Protection Agency maintains some of the most

advanced data management systems in the federal regulatory system. In

combination the EPA’s systems track both compliance/noncompliance with the

federal environmental statutes and enforcement actions taken against

noncomplying facilities. They also track some data on criminal referrals that EPA

makes to the Department of Justice. To date these systems have primarily been

used for internal management purposes and for providing searchable compliance

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 83

and enforcement data to the public on the Agency’s elaborate web site. The EPA

maintains a commitment to the public utility of its data and has regularly engaged

in improving its data systems. During the time frame of this project the EPA was

further developing and integrating its compliance and enforcement data systems, a

process estimated to be completed by fall 2014. In all, the Agency’s data hold

substantial promise for inclusion in a data series on white-collar offenses.

As noted earlier, the EPA is responsible for enforcing numerous federal

statutes, including the Clean Air Act, the Clean Water Act, the Safe Drinking

Water Act, the Resource Conservation and Recovery Act, and the Comprehensive

Environmental Response, Compensation, and Liability Act (CERCLA), and the

Federal Insecticide, Fungicide, and Rodenticide Act. The Agency is responsible

for most of the enforcement activity under these statutes, although as with all

regulatory agencies it must refer suspected criminal cases to the U.S. Department

of Justice. In addition, it shares enforcement of CERCLA with the U.S. Coast

Guard, which is responsible for enforcement in coastal zones, the Great Lakes, and

ports and harbors.

Also worth notice is that primary enforcement of some environmental

statutes rests with other units of the federal government. These statutes are

focused on specific environment-related activities. For example, non-criminal

enforcement of the Endangered Species Act (16 U.S.C. § 1531 et seq.) is shared

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 84

among the National Oceanic and Atmospheric Administration’s Marine Fisheries

Service, the Department of the Interior’s Fish and Wildlife Service, the Treasury

Department, the U.S. Coast Guard, and the Department of Agriculture. The

Surface Mining Control and Reclamation Act (30 U.S.C. § 1201 et seq.) is

enforced by Interior’s Office of Surface Mining Reclamation and Enforcement,

while civil and criminal cases are initiated by the Department of Justice. In

addition, enforcement of a number of federal environmental statutes may be

undertaken by private parties under citizen suits’ provisions.

Thus, while the EPA is the agency most responsible for enforcing the

nation’s environmental laws, ultimately a comprehensive statistical series on

environmental offenses and enforcement will integrate data from these other

sources of monitoring and enforcement as well.

As noted above, the vast majority of EPA’s enforcement cases involve

offenses that constitute white-collar violations: violations of the nation’s

environmental statutes by companies, municipalities, farms, nonprofit

organizations (e.g., universities, hospitals), and government entities. According to

EPA compliance and enforcement officials, the infrequent exceptions would be

such cases as those involving the prohibited use of pesticides in homes or

violations by fraudulent businesses.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 85

The principal information system for EPA’s enforcement and compliance

data is the Integrated Compliance Information System (ICIS). It is a secure system

available only to EPA and state officials enforcing federal environmental laws.

underlies two data web tools: the Online Tracking Information System (OTIS), a

secure system accessible to employees of government environmental agencies, and

Enforcement and Compliance History Online (ECHO), available to the public via

EPA’s web site and which contains compliance and enforcement data for three

major EPA enforcement programs.

ICIS, the Agency’s most comprehensive enforcement and compliance data

system, includes all actions enforcing the federal environmental laws that have

been undertaken by the EPA (including convictions in referred criminal cases), and

all Clean Water Act enforcement actions taken by state environmental authorities

under delegation from the EPA. By the end of 2014 ICIS was to have been

enhanced by the incorporation of data on enforcement cases undertaken by state

enforcement agencies under EPA delegation to enforce the Clean Air Act.

Enforcement data are kept at the level of the facility that is being regulated

under federal environmental statutes, and include names and addresses of the

Upon delegation by the EPA, states can enforce various federal environmental laws. For example, under the

Clean Water Act’s National Pollution Discharge Elimination System (NPDES), EPA has delegated enforcement to

all but a handful of states.

These are the Clean Air Act’s stationary source program, the Clean Water Act’s direct discharge program, and the

Resource Conservation and Recovery Act’s hazardous waste generation program.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 86

regulated facilities. The data include violations as discovered by required periodic

self-reports (e.g., monthly, quarterly) or via government inspections, and sanctions

ranging from warning letters to criminal penalties. The EPA distinguishes between

informal and formal enforcement actions, with warning letters generally

considered an informal enforcement action and formal action comprising notices of

violation, administrative orders, civil orders and penalties, and criminal penalties.

For all of their relative comprehensiveness, the ICIS data also present a

number of limitations of varying degrees of importance. First, while federal

officials enter informal enforcement data for cases they enforce, state agencies to

whom EPA has delegated enforcement of federal laws do not enter all of their

informal enforcement cases. Second, the EPA distinguishes between major and

minor polluters—a distinction related to the pollution load being discharged—and

enforcement policy prioritizes the major dischargers. Moreover, some data are less

available for the latter. For example, nationally the self-report

compliance/noncompliance data under the Clean Water Act were 93 percent

complete for major dischargers but only 41 percent complete for minor dischargers

(fiscal year 2009 data).

Relatedly, data management officials noted that for

It was noted in our discussions that the distinction between formal and informal enforcement actions varies

somewhat by media being protected, and that enforcement policies are documented for each environmental medium.

Referring to the entry of facility discharge monitoring reports into the data. See U.S. Environmental Protection

Agency, NDPES Data Completeness Table (FY 2009), June 2010. Available at

http://www.epa.gov/Compliance/resources/reports/performance/cwa/2009-data-completeness-chart.pdf.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 87

smaller facilities Clean Water Act violations are included commonly only when

found in inspections, and that the quality of data for minor facilities is lower than

for major dischargers.

Third, as noted the data are registered by facility name, rather than by

corporate ownership names where the facility is named differently or owned as a

subsidiary of a larger corporation. This, combined with the fact that corporate

ownership may change over time, complicates research seeking to identify

violation rates, trends and sanctioning responses at the corporate level. This matter

also connects to a policy-level concern expressed by EPA officials: that the

Agency’s case data lack unique identifiers such as corporate IDs that would allow

it to connect easily to the enforcement data of other agencies on the same facilities.

Agency officials noted that it would be desirable to be able to link EPA data with

those at OSHA, the Department of Energy, the Department of Justice, the FBI, and

the U.S. Census. Differences in data structures among agencies complicate this

process presently.

Nonetheless, the EPA appears to be at the forefront of federal regulatory

agency efforts to address such challenges over time and to improve its data systems

by increasingly filling gaps and integrating the violations and enforcement data

across the environmental media the statutes are designed to protect. The prospect

that the BJS initiative to create a data series on white-collar offenses and the

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 88

federal enforcement response to them would both further those efforts and develop

synergy between them and those of other enforcement agencies would seem

promising.

In addition to regulatory enforcement the EPA also engages in civil and

criminal enforcement activities and maintains data on these. Civil administrative

actions are enforcement activities undertaken by the EPA or by states under EPA

delegation and do not involve judicial action. Sanctions include notices of

violation and administrative orders, sometimes including penalties. Civil judicial

actions are formal lawsuits filed in courts either by the U.S. Department of Justice

on behalf of the EPA or by States’ Attorneys’ General, and bring sanctions

including consent decrees, civil penalties, injunctions and supplemental

environmental projects as part of settlements. The EPA’s Criminal Investigation

Division (CID) conducts investigations and makes arrests in the more serious

cases, typically those allegedly involving willful or knowing violations of

environmental laws. The EPA refers such cases to the Department of Justice for

criminal prosecution. Sanctions include fines and imprisonment. Project staff also

met with representatives of the Agency’s criminal enforcement program.

The CID has approximately 200 law enforcement agents allocated among

47 field offices, 70 forensic scientists, and 45 attorneys specializing in

environmental crimes enforcement. In fiscal year 2010, CID opened 1824 leads, of

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 89

which 348 (19 percent) became formal investigations, which led to 208 cases (60

percent) prosecuted with indictments. The majority of open cases—51 percent—

involved alleged violations of the air and water pollution laws.

A substantial

minority of criminal cases are prosecuted as Title 18 offenses,

charging such

violations as false statements (18 U.S.C. § 1001), mail fraud (18 U.S.C. § 1341),

wire fraud (18 U.S.C. § 1343), conspiracy (18 U.S.C. § 371), and obstruction of

justice (18 U.S.C. § 1501 et seq.). In such cases, having offender identification

information will be key to correct classification as to white-collar versus non-

white-collar offenses. In addition, some environmental crime cases are not

handled by EPA, but instead are filed with DOJ by other agencies, such as

Interior’s Fish and Wildlife Service and the U.S. Coast Guard. Future work will

require assessments of the quality of data consistency and information availability

across these agencies in order to ensure reliable measurement of environmental

white-collar crimes.

According to EPA criminal enforcement officials, the Agency maintains the

Criminal Case Reporting System (CCRS) that tracks outcomes in cases referred to

DOJ and that includes information on defendants, fines and sentences in cases of

criminal convictions. They noted that identifying the specific statutes under which

See EPA’s Criminal Investigation Division by Tara Donn, Special Agent, Denver Area Office, and U.S. EPA

available at http://www.ncsl.org/documents/environ/TDonn62311.pdf.

Estimated at around 15 percent in project staff meeting with criminal enforcement officials.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 90

criminal liability was found would require inspection of case documents. The

officials said that while data on both criminal and civil cases are kept, these two

types of enforcement case have unique numbering systems so that connecting

parallel civil and criminal cases against the same defendants is more involved. It

was noted, however, that such parallel cases are infrequent occurrences.

As earlier noted, after several fruitful meetings with EPA officials project

staff efforts to arrange discussions about future EPA data sharing with BJS toward

development of a statistical series on white-collar offenses ran aground when the

officials became unexpectedly unavailable to discuss that prospect. Their

reluctance may be due to perceived budgetary costs of such an arrangement, to a

sense that Agency data are still undergoing major developments including their

integration across all environmental media and therefore they wish to be able to

assure high quality data before sharing them, or to concern for adverse political

ramifications from possible uses of the data. These reasons are speculative as staff

never learned of them or others from EPA staff. However, more recent contact

with EPA in connection with an exploratory study of consistencies and gaps in the

enforcement data kept by the Department of Justice, the Federal Justice Statistics

Program, and the EPA suggests that a promising opening for further work remains

(see below).

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 91

Project staff also met with enforcement and data officials from the

Department of Justice’s Environmental Crimes Section (ECS).

The purpose of

the meeting was to understand the Section’s enforcement data and connections

between them and EPA and Federal Justice Statistics Program data. ECS is staffed

with 37 prosecutors who handle its environmental cases, with U.S. attorneys in the

94 federal districts bringing other environmental crime cases.

ECS staff

estimated that its office handled from 20 to 25 percent of all federal environmental

criminal prosecutions, with the remainder being led by federal prosecutors in the

districts. EPA’s CID refers cases to ECS, but some cases are referred directly to

ECS attorneys by CID enforcement agents with whom the attorneys have

developed relationships over time. ECS staff estimate that the Section opens about

100 cases a year and closes the same number annually (though the opened and

closed cases are not perfectly overlapped).

According to ECS staff, its case management system is outdated but cases

are entered methodically and can be tracked throughout the criminal process. They

also noted that the data permit linking separate cases brought against defendants

for the same violations, whether by sorting on the DOJ number assigned to the

cases or using a “related” tab that links cases. While most ECS cases would

ECS is located within DOJ’s Environment & Natural Resources Division.

State and local governments also prosecute environmental crimes, often in cases in which the offenses violate

both state and federal laws.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 92

appear to constitute white collar offenses, some would not, such as hunting and

other wildlife offenses.

Among the challenges associated with ECS enforcement data is lack of

consistency with other federal environmental criminal data, leading to risks of both

double-counting of cases and missed cases, depending on which data systems one

is using. For example, data kept by the Executive Office for United States

Attorneys does not include all of the ECS enforcement cases. In addition, the ECS

case management system data may not reflect all environmental cases brought by

U.S. attorneys in the districts. But for criminal cases jointly prosecuted by ECS

attorneys and federal prosecutors in the districts, both data systems include them,

so that using both sources without adequate vigilance could lead to double-

counting. Such double-counting may also occur for cases jointly brought by ECS

and other regulatory agencies, such as the Securities and Exchange Commission.

Staff also noted that it is difficult to track repeat offenders in the ECS data system.

ECS staff expressed interest in working with BJS to more precisely locate

data gaps and to develop means of addressing them. See discussion of the

proposed Proof-of-Concept project in the recommendations section below.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 93

Securities and Exchange Commission

The U.S. Securities and Exchange Commission is the principal organization

enforcing the nation’s securities laws. It enforces a substantial number of federal

securities statutes, most prominently including the Securities Act of 1933 (15

U.S.C. § 77a et seq.), the Securities Exchange Act of 1934 (15 U.S.C. § 78a et

seq.), the Investment Company Act (15 U.S.C. § 80a-1 et seq.), the Investment

Advisors Act (15 U.S.C. § 80b-1 et seq.), the Foreign Corrupt Practices Act (15

U.S.C. § 78dd-1 et seq.), the Sarbanes-Oxley Act (PL 107-204, 116 Stat 745,

codified in Sections 11, 15, 18, 28, 29), and the Wall Street Reform and Consumer

Protection Act, also widely known as the Dodd-Frank law (PL 111-203, H.R.

4173). In addition it has statutory authority to oversee the enforcement activities

of a number of self-regulatory organizations (SROs) that police the securities

industry, including the Financial Industry Regulatory Authority (FINRA), the

National Futures Association, the Municipal Securities Rulemaking Board, and 20

registered stock and futures exchanges. The SEC also shares enforcement

authority under some statutory provisions with a number of other federal

organizations, including the Federal Reserve Board, the FBI, the Federal Deposit

Insurance Company, the Office of the Comptroller, the Consumer Financial

Protection Bureau, and the Public Company Accounting Oversight Board-- a

nonprofit corporation established by the Sarbanes-Oxley Act of 2002.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 94

The SEC enforces the federal securities laws against such offenses as insider

trading, misrepresentation of information on securities, manipulating securities’

market prices, and selling unregistered securities. The Commission uses both

administrative and civil law processes. Administrative procedures, which can

involve hearings before the Commission’s administrative law judges, utilize an

array of sanctions, including registration revocation, suspension of stock trading

privileges, and disgorgement of funds. The SEC can also file civil suits in the

federal district courts, seeking such sanctions as injunctions, disgorgement of

funds, and civil monetary penalties. When its investigations reveal evidence of

criminal conduct, the SEC refers the cases to the Department of Justice for

prosecution. However, it refers few such cases, whether considered in absolute

numbers or relative to the securities case referrals of other federal units. For

example, in fiscal year 2010 the SEC referred only 16 suspects for criminal

prosecution for securities fraud, while the FBI referred 435 suspects and the U.S.

Postal Service referred 49.

Most of the SEC’s enforcement is undertaken via

civil and administrative penalty processes.

Project staff conducted a lengthy phone meeting with an official in the

SEC’s Division of Enforcement to discuss the Commission’s enforcement data.

See the project staff agency brief “U.S. Securities and Exchange Commission: Annual Case Statistics, FY 2010,”

in Appendix of this report.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 95

Personnel in the Enforcement Division created its data management system, which

the official described as “robust.” The system tracks data points from the opening

of an inquiry, through its evolution into an SEC investigation, to case conclusions.

The system also included an intake data base which comprises tips, complaints and

referrals (TCR), which are then triaged into the SEC’s case tracking system. In

2010, 15,400 potential cases were entered into the TCR data.

The enforcement data system captures the statutory provisions violated, the

federal districts in which civil court cases are filed, and administrative actions.

The official said that the SEC conducts good collaborative investigations with DOJ

prosecutors, but that the SEC’s enforcement data system does not track parallel

civil and criminal cases. Relatedly, it was noted that if federal prosecutors win

criminal fines in securities cases, the SEC may abandon its pursuit of civil fines in

those cases so as not to over-punish violators.

Finally, the official outlined three model types of Commission cases and

said the quality of the SEC’s electronic data varied by these types. The first type

comprises cases in which there are lengthy investigations (e.g., two years) but for

which it is ultimately determined that no enforcement action should be taken. The

second type involves similarly investigated cases but in which enforcement action

is taken and the case is settled by the SEC and the defendants. In both of these

types of cases the SEC has good electronic data. The third type of case is more

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 96

complex because it involves multiple defendants whose individual allegations are

resolved in different ways. For example, in a particular case one of the defendants

may wish to settle the charges early, while another chooses to litigate the charges

over a period of years. The SEC’s data system contain information of varying

quality in these more complex legal matters.

Unfortunately project staff were unable to pursue questions of data quality

and accessibility with agency officials past this point. Upon request of the official

we sent information on case variables of interest to a data series on white-collar

offenses, and a model version of a memorandum of understanding governing data-

sharing processes and agreements between agencies and BJS. As with the

experience with the EPA, it was at this point that communications ceased when the

SEC declined to respond to further inquiries to set meetings to discuss data

measures and data sharing. Again, we do not know why communications ended,

but in the case of the SEC additional possible factors were the reorganization of the

Division of Enforcement during this period, in particular with the appointment of a

new Director of Enforcement and at a time when the Commission was (1) involved

in a number of major investigations and enforcement actions in connection with

the financial crisis of 2008-9, and (2) under considerable media scrutiny.

However, it is worth noting that the SEC publishes considerable data on its

enforcement cases on its web site, as shown in the project staff report “U.S.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 97

Securities and Exchange Commission: Annual Case Statistics, FY 2010” (see

Appendix). While much of this is in the form of documents rather than in the form

of electronic coded categories of data, the availability of this quantity of data

indicates that the Commission’s enforcement cases are a promising prospect for

inclusion in an eventual data series on white collar offenses.

The Federal Trade Commission

The FTC’s mission includes enforcement of laws that protect market

competition and the interests of consumers. This comprises cases to prevent

mergers that would limit competition in industries and to sanction fraud and

deception in the marketplace. The Commission enforces numerous federal

statutes, principal among them the Federal Trade Commission Act (15 U.S.C. § 41

et seq.). The FTC also has enforcement authority under other federal statutes, such

as the Clayton Act (15 U.S.C. § 12 et seq.), the Truth-in-Lending Act (15 U.S.C. §

1601 et seq.), the Fair Credit Reporting Act (15 USC § 1681 et seq.), and the Do-

Not-Call Implementation Act of 2003 (15 U.S.C. § 6151). It shares enforcement

jurisdiction for the Clayton Act with the Department of Justice, state enforcement

agencies and private parties.

The FTC may sanction offenses by individuals or organizations either

through administrative enforcement or by filing civil cases in the federal courts.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 98

Administrative sanctions include consent orders and monetary penalties; court

sanctions include orders, injunctions and civil penalties. As with all other

regulatory agencies, the FTC refers potential criminal cases to the Department of

Justice.

Project staff met with FTC officials from the Bureau of Competition and the

Bureau of Economics. More information on FTC enforcement data was gleaned in

a follow-up telephone meeting with a member of the Bureau of Economics.

These officials reported that the FTC’s case management system was

antiquated and that case details weren’t routinely logged into it. A five-year-old

electronic data base tracks enforcement of orders but the FTC staff said many cases

had incomplete fields, the data base lacked a rigorous codebook, and there weren’t

good ways to assess data quality. The vast majority of consumer protection cases

are filed as civil matters in the federal courts, and staff members indicated that

PACER case documents are a good source for identifying key case details in

FTC’s court cases.

In sum, FTC case data exist most accurately largely in the Commission’s

legal (text) documents for cases (e.g., complaints, consent orders) rather than in

coded categories, indicating that at present some form of data-scraping would be

required to extract case details with any efficiency. These documents contain

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 99

information on the nature of the violations charged, defendants’ names (allowing

for distinctions to be made between individual and organizational offenders) and

resolutions for each count or offense. There is no data base on case referrals to

DOJ, but the Office of the General Counsel tracks case appeals and their outcomes.

Notably the FTC officials expressed interest in working on a pilot project

with BJS to determine the quality of their enforcement data, as well as interest in

the BJS work with other agencies’ white-collar offense data systems because of the

expected synergies toward improved data collection and management that could be

derived from inter-agency information sharing and collaboration. As earlier noted,

these potential synergies provide one of the brightest prospects for the BJS project

of building a data series on white-collar offenses and the federal enforcement

response.

VI. Federal Data Sources on White-Collar Crime: Strengths and Weaknesses

A. Criminal Data

The Federal Justice Statistics Program (FJSP)

FJSP data are collected by the Bureau of Justice Statistics from five federal

agencies and are maintained by the National Archive of Criminal Justice Data

(NACJD) at the Interuniversity Consortium for Political and Social Research

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 100

(ICPSR). The agencies that contribute to the FJSP are the U.S. Marshals Service,

the Executive Office for United States Attorneys (EOUSA), the Administrative

Office of the United States Courts (AOUSC), the United States Sentencing

Commission (USSC), and the Bureau of Prisons (BOP).

Each agency interacts with offenders at distinct periods in a case’s life cycle,

though multiple agencies may collect data about the same case at the same time.

This presents the opportunity to evaluate the quality of data at the points where

agency data intersect. The FJSP uniquely offers a case identifier that may be linked

across agency files, offering a straightforward method for evaluation of

interagency consistency. However, not all stages of the criminal justice process

contain agency data overlap (see Figure 7, below). This is relevant to a white-

collar crime series in that, unlike street crimes, the early stages of a case are much

less likely to be conducted by the U.S. Marshals. Corporate crime, in particular, is

more likely to involve attorneys than it is law enforcement at these stages. The

Marshals’ data, then, are less useful in a series on white-collar crime that includes

organizations as offenders. However, the data are not completely unserviceable

given that the working definition of white-collar crime for this project includes

individual offenders—and these criminals may well be arrested by the U.S.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 101

Marshals.

Records from the Bureau of Prisons similarly exclude organizational

offenders but, unlike the Marshals Service, the BOP data generally are redundant

with information contained in reports by three other agencies in the FJSP—

AOUSC, EOUSA, and USSC. The only advantage BOP data would bring to the

white-collar crime series is that they contain demographic information about

offenders. However, these benefits apply only to those white collar offenders who

are imprisoned. The lack of demographic data across all sources is one notable

weakness of the FJSP.

Given the role-centered definition of white-collar offenses employed by this

project, the FJSP data also lack variables that would identify the occupational role

of offenders as it relates to their offense. Instead, the best means of isolating white-

USMS

BOP

EOUSA

AOUSC

USSC

Detection and

Investigation

Arrest Case Filing

Case

Resolution

Sentencing

Figure 7: Simple Stages of Criminal Justice

System Response

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 102

collar crime for the series are via offense (categories and statute specific) and

agencies involved. A problem with this approach is that the offense categories

(e.g., environmental or “other regulatory offenses”) are difficult to unpack in order

to isolate white-collar offenders.

Offense category variables are consistent across all datasets in the series,

though the operationalization of the coded values are unavailable—at least to these

researchers. That is, the precise characteristics that result in offenses being coded

in one particular category as opposed to another are not made transparent. Specific

statutes are available for some datasets (for most files in the EOUSA, AOUSC,

USSC, and BOP), though the number of statutes available varies from one to up to

seventeen. While some statutes imply crimes consistent with our definition of

white-collar crime (e.g., violations of securities reporting [15 U.S.C. §§77-78],

failure to certify financial reports [18 U.S.C. §1350]), not all statutes are so easily

cataloged. The archetypical “white-collar” offense, mail fraud, neither requires nor

implies the perpetrator’s role in occupational activities. Thus, offense provides a

means to isolate some but certainly not all white-collar crimes.

Another way in which the FJSP can be used to identify white-collar crime is

to extract cases that are referred by agencies whose enforcement activity typically

corresponds with our definition. Cases referred by the Federal Trade Commission,

Securities and Exchange Commission, Food and Drug Administration, and

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 103

Environmental Protection Agency are almost all consistent with our definition.

Unfortunately, however, not all agencies collect and report this information.

Referring agency is only available from the EOUSA and suffers from (limited)

missing data.

Finally, offenses consistent with our definition may be related to “participant

type,” which indicates whether defendants are organizations or individuals.

Unfortunately, this variable is not a required data element or uniformly available

(only in the four datasets from the EOUSA can participant type be used to

distinguish organizations from individuals), and it does not distinguish between

legitimate and illegitimate organizations. But in combination with other data in

the FJSP (e.g., statute, referring agency) it may increase confidence in the

determination of white-collar offenses as defined in our core definition. Future

work with these data could ascertain the degree of accuracy with which

“participant type” and other variables identify cases of white-collar offenses as

defined.

Executive Office for United States Attorneys (EOUSA)

The EOUSA data are extracted from the agency’s case management system

(Legal Information Office Network System, or LIONS) and assembled into four

criminal data files – matters in, matters concluded, cases filed, and cases

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 104

concluded. The “matters” files contain all alleged violations on which a U.S.

attorney spent one or more hours, while “cases” files contain only those in which a

formal case was filed in court. The matters files provide the only look into the

investigatory process through the FJSP, though it is still limited in scope (as

described below).

The EOUSA are the only data source through which white-collar crime can

be isolated via all three methods discussed above. Only one statute—the lead

statute-- is provided for cases filed and concluded. While data entry personnel are

not required to enter referring agency, the field is well-populated (>90%), and it is

available across all four datasets; the participant type variable is also available

across all EOUSA sources.

Unfortunately, because the EOUSA data are divided into four datasets, it is

not possible to track a case’s life cycle within a single dataset, from referral to

sentencing. However, the series does contain information on case referral, the

decision to file or not to file formal charges, the initial and final charges, the

decision, and any resulting sentences. The BJS’s linking system should permit

tracking between datasets.

One notable concern with the EOUSA data pertains to the “terminated” files

– there is no objective rule for determining the basis of a case’s termination

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 105

because attorneys make the decision on a case-by-case basis. For instance, cases

marked terminated may indicate that an attorney or division is no longer working

on that case, or that it was referred out, or it was terminated for some other reason.

Consequently, the case data may be incomplete (e.g., if a case is “terminated”

before sentencing, disposition and sentencing values will be missing).

Administrative Office of the United States Courts (AOUSC)

The AOUSC data cover cases from the initial court filing through the

sentencing process. Thus, data from this part of the FJSP should include all cases

in the EOUSA “Cases Filed” dataset. However, the linking system is imperfect,

and there is no known formal comparison between the two. The AOUSC’s main

advantage over the EOUSA data is that it contains up to five statutes per case filed,

rather than the one provided by the EOUSA. However, the court data do not

contain agency or participant type. Thus, attempts to identify white-collar crimes

are limited to statute-specific crimes. However, if the linking system to EOUSA is

improved, the AOUSC data may prove useful.

Public Access to Court Electronic Records (PACER)

PACER is a publicly accessible search service that contains all court

documents for federal appellate, district, and bankruptcy courts. The site charges

users a small fee to search records and a separate fee to access documents

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 106

associated with specific cases. The main advantage to PACER is that, with its

publicly accessible court documents, it is by far the richest data source available

for case data (after filing). PACER is searchable by party name (e.g., “Securities

and Exchange Commission”), date of filing and conclusion, and type of case.

However, PACER also has several drawbacks, the primary one being its

cost. Searching is arduous because search results are very sensitive to data entry

variation and the rich text files must be accessed through individual cases.

Documents must be compiled one-by-one. Relatedly, PACER lacks efficient

search parameters that are suitable for our data series. Case type may not be

reliably entered (and is not defined for the user), and party searches (such as by

agency) depend on the accuracy of data entry. Comparisons between PACER and

agency data have revealed that agency data nearly always contain cases that are not

returned by PACER, most often due to a party’s name being misspelled or not

captured at all in the PACER database. PACER’s primary benefit is in its ability to

provide supplemental case data for case lists assembled via other methods and for

cross-validation purposes.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 107

B. Civil Data

The Federal Justice Statistics Program

The FJSP has only recently developed the civil case series to parallel its

criminal data. Rather than being drawn from five agencies, the civil case files are

drawn only from the EOUSA and AOUSC. BJS is still finalizing the civil data, so

our observations about the data and their utility for the white-collar data series

should be understood as preliminary.

EOUSA’s civil data (divided into four datasets, like the criminal data)

contain many of the key variables found in the criminal data. This makes sense

given that the FJSP is an extract from the agency’s case management system,

which contains both criminal and civil cases. This means that, for our purposes, it

is possible to determine whether a defendant is an individual or an organization

(but with the limitations noted for the criminal data). Additionally it is possible to

isolate cases by the agency that referred it, and both statute and offense categories

are provided.

However, the pilot civil data are also missing certain key variables. For

instance, while the data’s unit of analysis is the case-defendant, there is no way to

determine if defendants are related to the same case (unlike the criminal data which

contain the case ID number and the defendant ID number). In addition, the civil

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 108

data must be filtered by the case type – not all civil cases in the system are

affirmative; some are defensive (i.e., a party sues the SEC), and some involve the

enforcement of a previous administrative order. Missing data are problematic, as

are “unknown” values for variables such as “monetary relief”—the key sanction

for civil cases. Less than rigorous data entry all too often compromises data

quality. In addition, like the criminal data, EOUSA data do not contain variables of

interest related to business and individual characteristics. However, unlike the

criminal data, no linking system allows the extraction of these variables from other

agencies’ datasets.

The AOUSC civil data are similarly challenged. And like the criminal

AOUSC data, the civil files lack participant type and agency information. Statute

information is available, but for only a single charge (removing the primary

advantage of the AOUSC data over EOUSA). However, the AOUSC files are more

complete than the EOUSA files for case judgment information – the files contain

the amount awarded for each case in the dataset (unlike the EOUSA files). Thus,

the AOUSC data can be used to assess, verify, and supplement the sanction

information in the EOUSA.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 109

VII. Lessons learned from the Agency Data Briefs

One of the main tasks of this project was to map how regulatory agencies

process offenses, especially identifying similarities and differences in

legal/legislative authority, decision processes, available sanctions, levels of

analysis, and ultimately the kinds of data collected, so that BJS can develop and

implement the means to standardize coding and presentation of data across the

variety of data sources (i.e., create a crosswalk in which criminal, civil, and

regulatory classifications of white-collar offenses are specified). To meet this goal,

the project team prepared briefs on five regulatory agencies: the SEC, EPA, FTC,

FDA, and the CFPB. In each of the briefs (which are attached to this Final Report

in the Appendix), we describe the federal enforcement domain of the agency

(classification), the scope of enforcement activities, and case sources and

processing details. Whenever possible, these data points are compared with

information generated from other sources, such as an extract from LIONS or

PACER. The detailed agency briefs allow us to draw conclusions about the utility

of specific agencies’ publicly available data for the white-collar violations series,

and also to compare defendant and case information with data already held by BJS.

One of the first observations that can be drawn from the agency briefs is that

the scope of enforcement varies considerably by agency. Scope is affected to some

extent by the mission of the agency. The FDA, for instance, is responsible for

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 110

foods, biological products, medical devices, electronic products that give off

radiation, cosmetics, veterinary products, and tobacco products. The Consumer

Financial Protection Bureau emerged out of Dodd-Frank legislation with a more

circumscribed mission--to ensure that consumers are protected from unfair,

deceptive and abusive acts and practices especially in the area of consumer

financial products and services. Scope is also affected, however, by the statutes

that agencies are tasked to enforce and who else operates within the enforcement

domain. While the FTC now oversees the administration and enforcement of more

than 70 laws, in many cases the agency serves in an advisory or reporting capacity,

leaving enforcement to other agencies. Conversely, while the environmental

enforcement domain is vast and populated with many different players, the EPA is

the agency responsible for the highest proportion of investigative and enforcement

activity within the domain.

A review of the agency briefs reveals multiple sources of data across

agencies. Although most agencies report civil judicial and administrative actions,

a few (e.g., EPA) also report referred closed criminal cases. Some agencies are

repositories for consumer complaints (FTC and CFPB), but it is unclear how—if at

all—the complaint data inform enforcement activities. The SEC also has a new

office to manage tips, complaints, and referrals, but there is little information about

this division on the SEC website. It would be helpful to know whether the tips,

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 111

complaints, and referrals result in case generation. Generally, the SEC does not

provide information about case source or origin. However, it does offer a great

deal of detail on case processing, resolution, and outcomes.

The EPA, on the other hand, appears to have the most sophisticated data

systems compared with the other agencies we evaluated. The internal ICIS system

is a multi-media integrated data system that includes information on all stages of

case processing from case inception and handling to resolution. The information in

ICIS is provided at the case level and includes important details that facilitate case

tracking over time. In addition, EPA also collects facility inspection and self-

reported compliance data. The FDA also maintains a database with information

regarding inspections and citations, which is searchable by compliance status of

firm, project area, firm name, location, and inspection date.

With the exception of the CFPB (which only reports administrative cases),

all of the agencies report information about civil judicial cases and some also

report information about criminal cases (either referred to DOJ) or cases in which

the agency worked with the DOJ to assist with case development and prosecution.

(EPA, for instance, keeps these data but not in a publically available database.) The

rich details in the agency data regarding civil enforcement and criminal referrals

can be compared with data in the FJSP (LIONS) and PACER to verify counts and,

with some effort, determine the causes of variation between data sources. Our

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 112

assessment of data quality in some cases finds agency data that are more complete

than counts in LIONS and PACER (see the SEC brief). However, until more

systematic investigation is conducted comparing agency and FJSP counts and case

details, it is premature to draw conclusions about the veracity of any of the data

sources.

Finally, the agency briefs give us a good sense of whether data held by

specific agencies are a good fit with our definition of white-collar offending.

Generally, we found that all agencies hold relevant data for the series, with some

more salient than others. Nearly all of the EPA activity is consistent with the

definition and, in most cases, the SEC, FDA, and FTC data permit the

determination of whether the offenses are committed by legitimate

businesses/professionals in the course of their occupations.

VIII. Conclusions and Recommendations

The development of a data series on federal white-collar offenses by the

Bureau of Justice Statistics would constitute a major advance in the nation’s

catalogue of statistical information on crime and the federal response to it. For

decades researchers have asserted the necessity of such information for enabling a

full understanding of crime and social control in the U.S., and more recently

policy-makers have joined the call to provide such data toward the cause of better

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 113

informed policy responses to such offense patterns as those associated with the

financial crisis of 2008-9. Combined with ever more sophisticated software tools

for compiling and managing data, these important purposes indicate an especially

opportune moment for the development of such a data series.

Such a series would also fill an important gap in BJS’s otherwise wide-

ranging collection of data on the nation’s experience of crime and punishment.

The addition of white-collar offenses would serve well the Bureau’s interest in

creating a taxonomy of crime based on a behavioral definition of offenses that

focuses on the dimensions of actor, intent, mode and seriousness. Federal white-

collar offenses, which comprise violations of federal laws by both individuals and

organizations, commonly locate on unique positions on these dimensions. The

actors are typically offenders who do not commit other kinds of crime; they

“specialize” in taking advantage of their occupational and/or organizational

positions to commit offenses and very often are one-time violators. Intent is tied

to occupationally- and organizationally-situated motives and opportunities that

vary over time and place (e.g., for greater profits, incomes, reduced risks of

detection, etc.). The mode of behavior involves abuses of relations of power,

authority and trust that are inherent in legitimate occupations and organizations.

And white-collar offense are commonly recognized as the most financially serious

violations of law, especially in the aggregate as compared to conventional offenses

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 114

but also in terms of major financial offenses taken singly. It is also the case that

white-collar violations are often costly in other ways. Offenses in the social

regulatory sphere are associated with illnesses, injuries and deaths, particularly in

the case of corporate offenses that put citizens, employees and customers at risk

(e.g., with toxic pollution exposures, dangerous workplaces and products that fail

to meet national safety standards). In this connection it is worth noting that a

number of federal enforcement agencies, such as the EPA, SEC and OSHA, make

formal distinctions as to offense seriousness in their enforcement regimens.

It is also important that such a series be constructed in a way that has

transparent legitimacy in light of the important public purposes it would serve.

This requirement has a number of aspects. First, it must comprise offenses that are

widely recognized as the sort of white-collar violations that observers—

researchers, media and policy-makers—have long noted are both understudied and

less seriously sanctioned relative to conventional crimes. It must, in other words,

minimize inclusion of offenses that are not recognizably white-collar in intent and

mode. The role-centered definition of white-collar offense employed in this

project is designed to ensure this result.

Second, it must include offenses that are sanctioned by federal laws under

any enforcement regime, whether regulatory, civil or administrative processes are

employed. This broad scope of offenses is especially important in connection with

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 115

organizational (e.g., corporate) violations of law. The vast majority of such

violations are sanctioned through either administrative or civil legal processes;

only a very small fraction are criminally prosecuted. Moreover, decisions as to

whether to use criminal versus non-criminal procedures to sanction offenses are

commonly made on criteria other than the seriousness of offense and often even

apart from such criteria as intent to offend.

The administrative or regulatory sanctions data also contribute to an

important analytic and public policy purpose. Combining these data with the civil

and criminal data permits not only analyses of the totality of offenses and the

federal enforcement response to them; it also makes possible analyses of the

“funnel” of cases through which offenses and offenders travel, from weaker to

stronger sanctions. It allows researchers and policy-makers to address questions

such as under what conditions are cases processed in one arena versus another,

what proportions of offenses work their way through to criminal sanctioning, what

decisions and processes drive cases from one to another of the enforcement arenas,

and—given eventual longitudinal data in such a data series—what are the relative

deterrent consequences of the various forms of sanctions.

Third, because the construction of a white-collar offense data series is a

complex undertaking that will require some years of development, it is important

that the early stages of development include samples of offenses that is defensibly

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 116

representative of the universe of such violations. Key dimensions of

representativeness are the nature of offenders—organizations and individuals—and

the nature of offenses. Regarding the latter, we have proposed sampling on a

distinction widely used in the professional literature: that between social and

economic offenses. Our focus on the data of the Environmental Protection Agency

and the Securities and Exchange Commission reflects this important purpose.

As noted above the project reached an impasse with these focal agencies at

the point at which we were seeking to negotiate a memorandum of understanding

for their data sharing with BJS. We believe this barrier is surmountable, and that it

will take convincing agencies that their participation in the project has identifiable

benefits and minimal risks for them. For example, a number of agency officials

expressed interest in the prospect that their participation might inform their

agencies’ own abilities to improve their enforcement data systems and data quality,

given the potential synergies that can arise from discussions and collaborations

with other agencies, including BJS. The effort to build a data series can build in

this interest. For example, working with BJS staff to improve data structures and

management should provide agencies with more reliable and efficient access to

data analyses useful for realizing the agencies’ missions, as would regular

feedback from BJS on the completeness and reliability of their data, and

opportunities to learn from the collective experiences of participating organizations

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 117

about advances in data management practices and state-of-the-art technologies for

data extraction and handling (e.g., software advances for mining quantifiable data

from text documents, or data-scraping techniques).

Overcoming agency resistance to data-sharing for the statistical series on

white-collar offenses will require a “breakthrough” with a major agency to

demonstrate both the feasibility and benefits of participation. We believe that the

“proof-of-concept” project on environmental offenses currently being planned will

provide a solid opportunity for such a breakthrough. The project involves

examining, comparing and cross-validating the case data maintained by the

Department of Justice’s Environmental and Natural Resources Division (ENRD),

the Environmental Protection Agency, and the FJSP civil and criminal case data.

The project would focus on a specific time frame and statute(s) (such as the Clean

Water Act), and would produce evidence of any gaps or variations in cases and

types of information on them, information that can then be used to rationalize the

data and to suggest improvements in data management. Both ENRD and EPA

have agreed to participate in this project, both organizations having an interest in

ensuring the accuracy and reliability of their case data. Importantly, this project

should also help to demonstrate the utility of such collaborative work on

enforcement data and build confidence in the working relationships and data

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 118

sharing mechanisms that participation in the larger project of building a data series

on white-collar offenses requires.

We enthusiastically propose that BJS continue to pursue not only this

demonstration project, but that it use it as a springboard to continue efforts to build

the data series. The success of the demonstration project should provide

legitimacy to future BJS requests for agency involvement, and the reassuring

evidence necessary to securing that involvement.

Toward the future development of the data series, we make several

recommendations:

1. We recommend the formation by BJS of an ongoing working group

that includes representatives of key enforcement units and agencies to discuss and

compare notes on enforcement data management challenges, needs and goals, and

to brainstorm on means of meeting and achieving these. Ideally included in such a

working group would be enforcement units within DOJ and representatives of

major regulatory agencies. Among the latter we recommend minimally the EPA,

SEC, FTC, CFPB and DHHS, which together—based on the findings of this

project—represent both the two broad spheres of offenses (economic and social)

and wide variation in the quality and sophistication of their enforcement data

systems.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 119

Such a working group can achieve a number of important purposes. It can

provide for information sharing between agencies and help to dissolve barriers

between them. It can generate both ideas and increased commitment to improving

agencies’ data systems. And, importantly, with sponsorship by BJS it can clarify

for enforcement units the Bureau’s mission and expertise, make apparent the

mutual benefits of data sharing and the value of the white-collar offenses data

series, and build essential relationships between Bureau and agency personnel. A

successful working group would enhance access to agency data, perhaps initially in

additional demonstration projects, and ultimately provide for regular data sharing

for the series. Despite the access difficulties experienced in this project, we

recommend continued outreach to agency personnel beyond the working group

effort, especially as good relations and ideas are generated.

2. Based on accumulating knowledge regarding the interests and

concerns of agencies, and working closely with the relevant agency personnel, the

development of agency-specific memoranda of understanding that address their

specific interests and concerns.

3. We recommend that the series be built incrementally, over a period of

years, using criminal and civil data held by BJS and regulatory data drawn from

agencies across spheres (economic and social) but within domains. The within

domain comparisons are critical for purposes of data verification and ultimately

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 120

establishing a statistically sound series. Within domain comparisons should be

sensitive to the fact that the data from each source are collected for different

purposes. Key elements to consider include:

i. Standard unit of count (person-case)

ii. Comparable time periods

iii. Tracking and processing of offenders/cases through similar decision points

iv. Classification of disposition/sentence outcomes

4. We encourage research into technological advances in data

management, and in particular advances in software for “reading” text and

extracting key variables for inclusion in the data series. Such advances make

possible the use of agency materials that otherwise would be too labor-intensive to

mine for data, especially on an ongoing basis. Not only would such data increase

the range of enforcement cases included in the series, it would also enable analyses

of data quality across data systems.

It is difficult to overstate the importance of the creation of an ongoing data

series on white-collar offenses. Both research and public policy—closely related

activities, it is worth noting—would benefit greatly from a thoughtfully and

carefully crafted data system. And so would the quality of participating agencies’

enforcement data management systems. Policy in particular is often importantly

driven by what can be seen, and what can be seen most clearly and consistently is

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 121

what can be counted or measured. It seems not too much to say that public safety,

confidence in law, and justice all require the sort of information that a data series

on white-collar offenses would provide.

We envision the data series as a dynamic system—one that will

progressively include additional information about white-collar violations as the

series develops. Over time, working with the different data providers will also

enable upgrades of existing data systems. Pursuing progress in these directions

will present numerous challenges, as illustrated in this Report. But it is difficult to

imagine a more important set of developments in the Nation’s efforts to understand

and address such an important form of offending.

Bibliography

Brown, Darryl K. "Street Crime, Corporate Crime, and the Contingency of

Criminal Liability." University of Pennsylvania Law Review 149 (2001): 1295-

1360.

Bureau of Justice Statistics. "White-Collar Crime: Federal Offenses And

Offenders." Washington, DC, 1987.

Clinard, Marshall B., and Peter C. Yeager. Corporate Crime. New York: The Free

Press, 1980.

—. Corporate Crime. New Brunswick, NJ: Transaction Publishers, 2006.

Clinard, Marshall B., and Richard Quinney. Criminal Behavior Systems: A

Typology. New York: Holt, Rinehart & Winston, 1967.

Clinard, Marshall B., Peter C. Yeager, Jeanne Brissette, and David Petrashek.

Illegal Corporate Behavior. Washington, DC: U.S. Government Printing Office,

1979.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 122

Clinard, Marshall B., Richard Quinney, and John Wildeman. Criminal Behavior

Systems: A typology. New York: Taylor & Francis, 1994.

Comolli, Renzo, and Svetlana Starykh. "Recent Trends in Securities Class Action

Litigation: 2013 Full-Year Review." National Economic Research Association,

New York, 2014.

Dictionary of Criminal Justice Data Terminology. 2nd Edition. Bureau of Justice

Statistics, 1981.

Dunworth, Terence, and Joel Rogers. "Corporations in court: Big business

litigation in U.S. Federal Courts, 1971-1991." Law & Social Inquiry 21, no. 3

(1996): 497-592.

Edelhertz, Herbert. The Nature, Impact, and Prosecution of White-Collar Crime.

Washington, DC: National Institute of Law Enforcement and Criminal Justice,

1970.

Gibbons, Don C. Changing the Lawbreaker: The Treatment of Delinquents and

Criminals. Montclair, NJ: Allenheld, Osmun & Co., 1965.

Gottfredson, Michael, and Travis Hirschi. A General Theory of Crime. Stanford:

Stanford University Press, 1990.

Huff, Rodney, Christian Desilets, and John Kane. The 2010 National Public Survey

on White Collar Crime. Fairmont, WV: National White Collar Crime Center, 2010.

Johnson, David T., and Richard A. Leo. "The Yale White-Collar Crime Project: A

Review and Critique." Law & Social Inquiry 18, no. 1 (1993): 63-99.

Karpoff, Jonathan M, John R. Lott, and Eric Wehrly. "he reputational penalties for

environmental violations: Empirical evidence." Journal of Law and Economics,

2005: 653-675.

Karpoff, Jonathan M., Allison Koester, D. Scott Lee, and Gerald S. Martin.

Database Challenges in Financial Misconduct Research. Social Science Research

Network, 2014.

Karpoff, Jonathan M., D. Scott Lee, and Gerald S. Martin. The economics of

foreign bribery: Evidence from FCPA enforcement sctions. SSRN, 2014.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 123

Karpoff, Jonathan M., Scott Lee, and Gerald S. Martin. "The Cost to Firms of

Cooking the Books." Journal of Financial and Quantitative Analysis 43, no. 3

(2008): 581-612.

Laursen, Christopher. BSA/AML Enforcement Trends and D&O Liability. New

York: National Economic Research Association, 2014.

Mason, Donald A. Tracking Offenders: White-Collar Crime. Washington, DC:

Bureau of Justice Statistics, 1986.

Moffitt, Terrie E. "Adolescence-limited and life-course persistent antisocial

behavior: A developmental taxonomy." Psychological Review 100 (1993): 674-

701.

Reiss, Albert J, and Albert D. Biderman. "Data Sources on White-Collar Law-

Breaking." National Institute of Justice, U.S. Department of Justice, 1980.

Scalia, John. Federal Enforcement Of Environmental Laws, 1997. Washington,

DC: Bureau of Justice Statistics.

Schlegel, Kip, David Eitle, and Gunkel Steven . Securities Lawbreaking: The

Enforcement Response. Final Report, Washington, DC: National Institute of

Justice, 1994.

Simpson, Sally S. "The decomposition of antitrust: Testing a multi-level,

longitudinal model of profit-squeeze." American Sociological Review 51, no. 6

(1986): 859-875.

Simpson, Sally S. "The Economics of Antitrust: Illegal Actions by Survivor Firms

in Basic Manufacturing." PhD Thesis, Department of Sociology, University of

Massachusetts, Amherst, 1985.

Simpson, Sally S., Joel Garner, and Carole Gibbs. Why do corporations obey

environmental law? Assessing punitive and cooperative strategies of corporate

crime control. Washington, DC: National Institute of Justice, 2007.

Sutherland, Edwin H. 1983 [1949]. White Collar Crime: The Uncut Version. New

Haven: Yale University Press.

Weidenbaum, Murray L. The Future of Business Regulation: Private Action and

Public Demand. New York: Amacom, 1979.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 124

Weisburd, David, and Elin Waring. White-Collar Crime and Criminal Careers.

Cambridge: Cambridge University Press, 2001.

Weisburd, David, Stanton Wheeler, Elin Waring, and Nancy Bode. Crimes of the

Middle Classes. New Haven: Yale University Press, 1991.

Wellford, Charles F., and Barton L. Ingraham. Toward a National Uniform White-

Collar Crime Reporting System. Washington, DC: National Institute of Justice,

1990.

Wheeler, Stanton, Austin Sarat, and Kenneth Mann. Sitting in Judgment: The

Sentencing of White-Collar Criminals. New Haven: Yale University Press, 1988.

Yeager, Peter C. "Law, Crime, and Inequality: The Regulatory State." In Crime

and Inequality, edited by John Hagan and Ruth D. Peterson, 247-276. Palo Alto,

CA: Stanford University Press, 1993.

Yeager, Peter C. "Structural bias in regulatory law enforcement: The case of the

U.S. Environmental Protection Agency." Social Problems 34, no. 4 (1987): 330-

344.

—. The Limits of Law: The Public Regulation of Private Pollution. Cambridge:

Cambridge University Press, 1991.

Yeager, Peter Cleary. "The Practical Challenges of Responding to Corporate

Crime." In The Oxford Handbook of White-Collar Crime, edited by Shanna R. Van

Slyke, Michael Benson and Francis Cullen. New York: Oxford University Press,

2015.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 125

Statutes Cited

7 U.S.C. §136

15 U.S.C. § 12

15 U.S.C. § 41

15 U.S.C. § 77

15 U.S.C. § 78

15 U.S.C. § 80a-1

15 U.S.C. §80b-1

15 U.S.C. § 1601

15 USC § 1681

15 U.S.C. § 6151

16 U.S.C. § 1531

18 U.S.C. § 371

18 U.S.C. § 1001

18 U.S.C. § 1341

18 U.S.C. § 1343

18 U.S.C. §1350

18 U.S.C. § 1501

30 U.S.C. § 1201

33 U.S.C. §1251

42 U.S.C. §300f

42 U.S.C. §6901

42 U.S.C. §7401

42 U.S.C. §9601

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 126

Sarbanes-Oxley Act, PL 107-204, 116 Stat 745, codified in Sections 11, 15, 18, 28,

Dodd-Frank Act, PL 111-203, H.R. 4173

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

P a g e | 127

Appendix: Agency Briefs

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

U.S. Securities and Exchange Commission

Annual Case Statistics, FY2010

Miranda A. Galvin, Peter C. Yeager, and Sally S. Simpson

The Securities and Exchange Commission (SEC, or “the Commission”) is tasked with the duty “to protect investors,

maintain

fair, orderly, and efficient markets and facilitate capital information.”

In this regard, the SEC operates within

the domain

of financial regulation. The domain includes numerous other government and independent regulatory

organizations. Other

federal government regulators include the U.S. Commodities Futures Trading Commission (CFTC), the Office of the

Comptroller of the Currency (OCC), and the Federal Reserve Board. Independent regulatory

organizations, known as self-

--regulatory organizations (SROs), include the Financial Industry Regulatory Authority

(FINRA), the National Futures

Association, the Municipal Securities Rulemaking Board, two clearing agencies (e.g. the

National Securities Clearing

Corporation and the Depository Trust Company), and 20 registered stock and futures exchanges.

The SEC is given

statutory authority to oversee these self---regulatory organizations (SROs). The regulation of securities markets may therefore

be conceived of as a nested model of enforcement in which both the SROs and the SEC

are responsible for enforcing federal

securities laws, while the SROs are also responsible for enforcing their own rules.

Federal securities laws encompass a vast number of statutes and even more rules promulgated by either the SEC or the

SROs. However, securities market enforcement may be understood as governed primarily by the Securities Act of 1933,

the Securities Exchange Act of 1934, the Trust Indenture Act of 1939, the Investment Company Act of 1940, the

Investment

Advisers Act of 1940, the Sarbanes---Oxley Act of 2002, the Wall Street Reform and Consumer Protection Act

(Dodd---Frank

Act) of 2010, and the Jumpstart Our Business Startups Act of 2012. According to various statutes, the SEC

shares some

enforcement responsibility with other federal agencies, including the Federal Reserve Board, the Federal

Deposit Insurance

Company, the Office of the Comptroller, and the Consumer Financial Protection Bureau, as well as

with the Public

Company Accounting Oversight Board (PCAOB), a nonprofit corporation established by the Sarbanes---

Oxley Act of 2002.

However, these agencies typically regulate different entities than the Commission; for example,

PCAOB is responsible for

overseeing corporate auditors, while the SEC is responsible for overseeing the public

companies audited. Other than in

criminal cases the SEC has principal jurisdiction in most matters relating to securities

offenses, but shares it with other

agencies for some types of offenses a

nd offenders.

U.S. Securities and Exchange Commission. The Investor’s Advocate: How the SEC Protects Investors, Maintains Market Integrity,

and Facilitates Capital Formation. http://www.sec.gov/about/whatwedo.shtml. Accessed 22 April 2014

Securities offenses may be discovered by any number of agencies that may then refer these offenses for criminal prosecution. In

2010, for example, at least 19 other agency units referred at least one suspect for a securities fraud violation to the United States

Attorney’s Office. The bulk of referrals came from the Federal Bureau of Investigation (435), while 49 suspects were referred by the

United States Postal Service. In comparison, the SEC referred only 16 suspects in 2010, and not all were for a securities offense.

Source: United States Department of Justice. Office of Justice Programs. Bureau of Justice Statistics. Federal Justice Statistics

Program: Suspects in Federal Criminal Matters, 2010

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

The SEC has the power to respond to cases through both administrative and civil means. Administrative cases may or

may not involve a hearing before an administrative law judge (ALJ), and can invoke a wide range of sanctions, including

registration revocation (temporary

permanent), suspension of an

organization’s stock trading

privileges, and disgorgement.

Civil

cases can lead to a variety of

outcomes, including injunctions,

disgorgement, and civil monetary

penalties. The Sarbanes---Oxley Act of

2002 granted the SEC new tools for

enforcement, specifically the

authority to distribute civil monetary

penalties collected to wronged parties (“Fair Funds”), broadened powers to bar, censure, or restrict individuals’

participation in the securities market, and expanded injunctive actions. In the Act’s first year of implementation, the SEC

utilized specific provisions to prevent more than $37 million from being distributed to company officers during the

course

of an investigation and collected more than $750 million for Fair Funds distribution to wronged investors.

The

SEC

cannot bring criminal charges under its own power, but refers cases for criminal prosecution to Department of Justice

prosecutors.

The types of information of interest for a data series on white collar offending include data on case sources, case

processing

and case resolutions or outcomes, as indicated in the graphic. On its website

the SEC provides nearly

exhaustive information

on case processing and resolution, while providing little to no information as to the processes

and sources of case

origination. During this fiscal year the SEC instituted a new office responsible for managing tips,

complaints, and

referrals,

yet little information about this division is available. It is also worth noting that what data are

provided on the

website are embedded within drafted releases and electronic case documents and are not readily available in a

quantitative format.

Case Sources and Processing

As noted in Figure 1, an agency may become aware of a potential violation of law through a variety of means, such as

complaints, referrals, and investigations. Complaints comprise information from consumers or market participants

regarding

suspected illegal activity. In fiscal year 2010 the Office of Market Intelligence became responsible for

receiving and

evaluating this information; previously the SEC had maintained an email tip service.

Referrals refer to

those suspected

violations of securities law or SEC rules detected by an outside government or regulatory agency (e.g.,

the CFTC, SROs, a

state task force)

that subsequently notifies the SEC in order to allow its investigators to more fully

evaluate the behavior

in question. Finally, the SEC may initiate a case based on information generated by internal

U.S. Securities and Exchange Commission. Office of Administrative Law Judges. http://www.sec.gov/alj.shtml. Accessed 22 April

2014

William H. Donaldson, Testimony Concerning Implementation of the Sarbanes---Oxley Act of 2002, before the Senate Committee on

Banking, Housing, and Urban Affairs. Sept 9, 2003.

http://www.sec.gov/news/testimony/090903tswhd.htm. Accessed 22 April 2014.

www.sec.gov.

U.S. Securities and Exchange Commission, FY 2010 Performance and Accountability Report.

http://www.sec.gov/about/secpar/secpar2010.pdf . Accessed 22 April 2014. However, as a result of the Dodd---Frank Act, the SEC

was required to create another tip---related office: the Office of the Whistleblower. It is not clear how this new office and tip line

interact with the responsibilities of the OMI. See Annual Report on Dodd---Franks Whistleblower Program FY 2011,

http://www.sec.gov/about/offices/owb/whistleblower---annual---report---2011.pdf Accessed 22 April 2014.

Infra note 8, page 11.

U.S. Securities and Exchange Commission, “How Investigations Work.”

http://www.sec.gov/News/Article/Detail/Article/1356125787012#.U1cSb01OWUk.

Accessed 22 April 2014.

Case

•

Complaints

•Inves8ga8ons

Case

Processing

•Case Types

•Charges

•Defendants

Case

•Viola8ons

•Penal8es

•Referrals

Fig. 1: Stages of enforcement action and information generated in each stage

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

agency activity, such as might emerge from an ongoing enforcement investigation or during routine compliance

management.

Case origination statistics are not available on the SEC’s website (Table 1),

and the Office of Market Intelligence does not provide statistics on its

activity.

However, some information may be gleaned from annual agency

Performance

and Financial Reports. Specifically, in 2010, 21.9 percent of

investigations were

sparked by internal referrals (interoffice) or prospects

(e.g., generated during

examination processes, data analysis).

Additionally,

303 investigations were initiated as a direct result of complaints. An

unknown number of examinations were also prompted by “cause

Table 1: Case Origination Data

exams.” Cause exams are part of the SEC’s compliance activities, and may or may not result in a formal enforcement

investigation, pending the results of the exam.

The total number of investigations, however, is not provided, nor can it

be assembled from the reported data.

Once a case has been initiated and violations have been substantiated, the SEC may seek either administrative or civil

sanctions against individual and organizational defendants. Most administrative cases are heard before an

Administrative

Law Judge (ALJ). Once formal proceedings are initiated, a hearing is held. After the hearing the ALJ will

issue an order

making findings and, if illegal conduct is found, impose sanctions. In some cases, both the order

instituting proceedings

(OIP) and the order making findings (OMF) are issued on the same day. A myriad of sanctions are

available through an ALJ

hearing, as noted above.

The SEC lists publicly all administrative actions at the points of both case initiation and case conclusion (though when

both an OIP and an OMF are issued on the same day, a case will only be listed once). SEC instituted 425 formal

administrative

proceedings in 2010.

Additionally, the Commission suspended five individuals without formally

instituting proceedings.

In these latter cases, the individuals suspended had been sanctioned by either a criminal court

or a state Bar Association

and thus do not require a formal hearing for sanctions to be imposed, though they arguably

represent new matters. The

SEC also imposes some administrative sanctions without formal proceedings, namely

suspensions of trading privileges for a

company’s stock. In 2010, the SEC issued a total of 51 trading suspensions.

Supra note 8, p.52

Supra note 8, p.48

The number of case---defendants is significantly greater than the number of cases. While case documentation exists for the full

group

of respondents, separating out case defendants for counts of these individuals and organizations would require substantial

coding

efforts. The counts were obtained through use of publicly available case listings. All administrative proceedings (available at

http://www.sec.gov/litigation/admin.shtml) are organized by year and quarter from 1995 to present; counts for FY 2010 include the

fourth quarter of 2009 (Oct. 1 – Dec. 31, 2009) and the first three quarters of 2010 (Jan. 1 – Sept. 30, 2010). In order to avoid double

counting cases in which the initial and final orders were issued separately (and thus the same case appeared in multiple posts), the

case

list was copied from the website into Microsoft Excel. It is important to note that while the SEC maintains a separately

accessible

list of Accounting and Auditing actions, these releases are also posted in the Administrative and Litigation case lists,

reducing the

need to perform separate analysis on this data source. In order to determine which listings were new proceedings, we

coded each

release as containing one or more of the following: an order instituting administrative proceedings (OIP), an order

making findings

or imposing a sanction (OMF), or other (e.g. appointment of a tax administrator, proposal and disbursement of Fair

Funds). Recognizing

that a single case may involve multiple actions (e.g., an OIP followed by up to one OMF per defendant), we

performed both automatic

duplicate field searches and manual comparisons of party name to ensure that case counts of new and

old proceedings did not double-

ount cases with multiple releases. There were no duplicates detected in OIP postings.

Trading suspensions issued by year and quarter can be found on the Commission’s website

(http://www.sec.gov/litigation/suspensions.shtml) and are not included in the listing of administrative procedures, though the two

share a system of release numbers (e.g. release 34---70186 is a trading suspension, but 34---70185 and 34---70187 are both

administrative releases). Trading suspensions are only posted at the start of the suspension and thus an analysis for duplicate

postings was not necessary.

Case

Sources

Data

Availability

Tips

Complaints

Referrals

Investigations

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Suspensions may be reissued if the offensive behavior has not been resolved, though in no case were consecutive

trading

suspensions issued in 2010. Altogether the SEC brought 481 administrative actions during the fiscal year (Table

2).

Civil cases are processed in a similar pattern and are heard in federal district courts. Formal proceedings are initiated

through the filing of a civil complaint; arguments and responses are heard in court, and subsequently findings are made

and sanctions imposed. The SEC instituted 207 civil cases in 196 matters in 2010 (Table 2).

A matter here refers to the

totality of all linked parties against whom action was taken in response to suspected illegal activity, or set of related

illegal activities; thus it is possible (though not common) for the SEC to file multiple civil cases in the same matter.

only 11 instances did the SEC bring separate cases against defendants suspected of involvement in related illegal

activity.

Many civil actions initiated prior to fiscal 2010 were resolved during the year (e.g., by final order or settlement

agreement),

and therefore the total volume of civil cases handled that year is significantly greater than those filed.

Criminal conduct

uncovered during

the course of an

SEC investigation is

subsequently

referred to either

the Department of

Justice or directly

to an Office of the

U.S. Attorney for

prosecution. The

exact processes

involved in this

relationship are not

explicitly clear from

publicly available

data. According to

the SEC’s Annual

Performance

Report, 2010 saw

139 criminal

Table 2: Case Processing Data NA = Not Available NR=Not Relevant

investigations

, though the extent to which these investigations translated into referrals for prosecution is unknown.

The SEC posted 10 litigation releases pertaining to criminal activity in 2010, but only two referenced new filings. The

remaining eight litigation releases pertained to verdicts reached, sentences imposed, and other significant

developments

in cases filed prior to FY2010. The degree of completeness of litigation releases remains unclear, and

they are timed to

later case developments rather than to SEC criminal referrals themselves.

As is the case for administrative proceedings, documents and litigation releases are grouped by case, the number of case---

defendants is significantly greater than the number of complaints filed, and the generation of case---defendant statistics is a labor---

intensive exercise. Litigation releases are available to the general public on the SEC’s website (available at

http://www.sec.gov/litigation/litreleases.shtml). In general, the SEC provides a PDF copy of the civil complaint filed for new cases

(181 matters comprising 192 cases). However, a thorough review of litigation releases found that posted complaints were not

exhaustive of all new action; in 2010, an additional 27 matters referenced a new civil case filed yet did not provide a PDF copy of the

complaint. In addition, in 12 releases a complaint was posted in error (e.g., it was not filed in FY 2010 or pertained to the

appointment of an administrator); none of these erroneous postings contained multiple complaints. Of the correctly posted

complaints, 9 releases included a total of 20 complaints, resulting in the necessity to distinguish between matters and cases.

The SEC’s online data are organized by matter, though can be disaggregated to case---defendant through careful analysis and

coding of provided documentation.

Supra note 8, p. 52

Data

Availability

Counts

Comments

Administrative

Actions

481

Includes formal proceedings instituted,

trading suspensions, and individual

suspensions

Civil Case Filings

207

New cases filed in 2010

Criminal Referrals

Completeness of litigation releases

uncertain; releases timed to post-referral

events

Case Numbers

Admin, Civil

Date Case Initiated

Admin, Civil

Number of

Defendants

Admin, Civil

***

Can be counted in available

administrative, civil and criminal case

documents

Types of Violation

Admin, Civil

Violations are usually provided in both

statute and narrative format.

Dates of Violations

Admin, Civil

Varies in specificity; year readily available

Defendant Chars.

Admin, Civil

Characteristics vary by case but usually

include age, gender, residence, and

employment information for individuals,

or location and business type for

organizations

Case Seriousness

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

It is worth noting that the SEC provides extensive documentation for those determined to query its files. However, the

nature of the data (electronic

Internal and court documents or

summary releases) nevertheless

requires significant time

investments and a systematic

coding system. In Table 2 it is

clear that currently available data

on SEC activity largely provides

most data points of interest.

Case Resolutions

This section presents information

on SEC cases resolved in 2010,

whether by dismissal or with

sanctions that close the matter.

The data available for the case

resolution stage are as rich and

problematic as are the data

to case initiation.

Releases are

structured similarly

and are

archived in the same

space. It is

not possible to

distinguish

between cases filed in

2010 and

cases filed in previous

years that

had significant action

taken in

FY2010 without reading

the

accompanying release or orders.

As discussed above, the SEC has a

myriad of sanctioning outcomes

available to it once a case has

been initiated. In addition, cases

may be suspended or dismissed;

however, these outcomes are

relatively rare and usually are due

to related criminal proceedings. A

case may be appealed to the

Commission, “which performs a

de novo review and can affirm,

reverse, modify, set aside, or

remand for further proceedings.”

direct Commission action is the

source of an appeal, the decision

may be brought to a United

States Co

urt of Appeals.

Not all forms of action have

clearly

differentiated initiation

Table 3: Case Resolution Data

*Available for provided cases

U.S. Securities and Exchange Commission. Office of Administrative Law Judges. http://www.sec.gov/alj#.UyMEhE2YaUk . Accessed

22 April 2014.

FY 2010

Data

Availability

Counts

Comments

Case Numbers

Admin,

Civil

Date(s) of Resolution

Admin,

Civil

Both litigation releases and

administrative files contain dates

filed; if respondents are sanctioned

on different days, the documents

will be filed under the full matter

title but will reference specific

respondents

Administrative Cases

Resolved

528

Includes suspensions, OMFs

issued, and dismissed cases; cases

can be disaggregated into case-

defendant

No. of Admin cases

dismissed

Number of Defendants

***

Number of violations

penalized

Statutes violated usually contained

in the OIP; “counts” of each

violation are not provided

Number of sanctions

imposed

***

Detailed in OMF

Civil Cases Resolved

Number of cases

Number of civil cases

dismissed

Agency suspended charges

Number of Defendants

***

Number of violations

penalized

***

Statutes violated usually contained

in the original complaint if not

release containing final order;

“counts” of each violation are not

provided

Number of sanctions

imposed

***

Releases contain information on

both the number and type of

sanctions imposed for each

defendant, but may not be

exhaustive due to the summary

nature of litigation releases

Detailed information on civil

penalties can be ascertained by

accessing court documents

Criminal Cases Resolved

***

Accuracy of counts uncertain

Cases Declined

Number of Defendants

***

In filed cases, significant actions

Number of violations

penalized

***

Statutes violated usually contained

in the original complaint or court

documents from other sources;

delineated by defendant

Number of sanctions

imposed

***

Appeals

Admin

Cases Only

No civil appeals were found in the

SEC’s litigation releases for

FY2010. The SEC posts appeals of

administrative decisions

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

and conclusion points. For example, trading suspensions represent both the initiation and conclusion of agency action

against a company for reporting violations. In 2010, the SEC concluded

56 matters resulting in suspension.

For administrative actions heard before an ALJ, however, there are separate orders

initiating proceedings and orders making findings, though they may be found in the same document. The SEC concluded

471

such cases in 2010; of these, 258 cases also filed an OIP at the same time, meaning that according to these data, the

case

was formally initiated and concluded on the same day.

In all 528 cases were resolved in FY 2010 (Table 3).

The SEC was also involved in several other administrative matters that continued to be active. Most often, these cases

related to Fair Funds distribution plans. Of the 84 cases that contained neither an OIP nor an OMF, 69 releases contained

information pertaining to 38 ongoing Fair Funds actions. Multiple actions generate releases during the course of Fair

Funds distribution, including the appointment of an administrator, the proposal of a plan, the acceptance of said

distribution

plan, distribution of funds, and the termination of distribution. These are not considered to be new

sanctions; the sanction

of a Fair Funds account is recorded in a case’s OMF. It is also worth noting that SEC was involved

in 13 other administrative

matters that did not pertain to violations of securities laws and are thus not included in the case resolution or processing

counts herein.

Only one appeal and one case of the agency dropping charges were recorded in 2010, suggesting both that this

information

is available publicly and that these actions are rare.

The way in which the SEC makes civil data available limits the amount of case resolution data that may be easily

collected.

Releases contain limited text to describe significant case actions, such as the granting of an injunction, the

reaching of

a verdict, and the reaching of a judgment. In 2010, significant actions were taken in 155 civil cases in which

the SEC was

engaged. These cases were originally filed as early as calendar year 2001. Because sanctions are described

in portable

document files that require individual manual or computerized coding, presently it is difficult to quantify the

types of

sanctions and values of civil monetary awards.

Criminal litigation releases also contain information on significant action in criminal cases related to SEC activity.

Significant

action (e.g. indictment, arrest, verdict) was taken in 10 criminal matters in 2010.

Other Data Sources

The SEC provides relatively complete data on administrative actions, less complete information on civil cases, and the

least complete data on criminal cases. Thus, the final portion of this brief will turn to other available data sources to

supplement and validate the SEC’s data. Civil case files may be supplemented through the federal website Public Access

to Court Electronic Records (PACER). For a nominal fee this site allows users to search for circuit, appeals, bankruptcy,

and civil court files by a number of parameters.

A detailed search initially yielded a significant difference in the number of apparent civil cases filed as reported by

PACER

and SEC records. Both sets of records were cross---checked for completeness. Most complaints posted on the SEC

These 56 matters involve the same 51 companies and 5 individuals that are counted in the case origination statistics.

Most administrative cases resolved in 2010 were initiated in the same fiscal year. In only 37 cases were OMFs (resolutions) reported

for cases that did not also file an OIP in 2010, meaning the case initiated prior to 2010. The majority (n=33) of these cases appeared only

once in the data and contained only an OMF; four cases appeared multiple times in the data but did not provide OIPs at any point.

These cases involved the appointment of tax administrators (6), reinstatement of practice privileges (3), denial or approved motions to

modify or set aside previous OMFs (3), and notice of a formal hearing after a previously issued OIP (1).

In order to best target PACER cases in which the Securities and Exchange Commission brought action against an individual or

entity, the researchers searched for all cases in which “Securities and Exchange Commission” or “SEC” (using exact match) was a

party. The search term “SEC” (not exact match) was not originally used due to the over---inclusion of irrelevant cases. To filter out

cases against the Commission, the case counts only considered those in which the SEC was listed as plaintiff, movant, or petitioner.

This originally returned 533 cases. However, it became apparent that the same matter could appear multiple times within the search

under different case numbers; significant actions (e.g. motions) were recorded under different case numbers despite involving the

same individuals. An effort to identify and remove duplicates was made both automatically (using Microsoft Excel) and manually (by

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

website did appear in PACER filing results, but not all

cases present in the SEC’s litigation releases appeared

the PACER records search using the specified party

search terms “SEC” and “Securities and Exchange

Commission.” The most common reason that a case

did not appear in PACER search results was that the

party name was listed as some variation on the

terms (e.g. “U.S. Securities and Exchange

SEC

PACER

EOUSA

Administrative Cases Initiated

481

Before Data Cleaning

730

Civil Cases Initiated

207

176

Before Data Cleaning

404

533

Civil Cases Referred

Criminal Investigations

139

Criminal Cases Referred

Table 4: Source Comparisons

Commission”, “Securities and Exchange Commission” [sic]). However, in no case was the data absent from the PACER

system in entirety; searching by case number invariably returned case information.

Of the 237 unique cases found in PACER, 27 were absent from the SEC litigation records in 2010. This was primarily due

to actions related to cases initiated in previous years (n=26). In only one case did a new action appear in PACER without

appearing in SEC litigation releases

, suggesting that this is, at worst, a recordkeeping anomaly. PACER appears to over---

count SEC activity. Some search results listing the SEC as plaintiff were in fact actions against the Commission. In

addition,

PACER returns multiple case numbers related to the same civil case;

in two particularly egregious matters,

results returned more than 50 and 70 unique case numbers for each of two cases. Overall, PACER court records

were

found to suffer from data entry errors and inconsistencies that limit their usefulness in determining the volume of

agency

enforcement activity. Duplicate cases (n=296 of 533 search results), secondary enforcement activity [e.g.

motions to enforce

subpoenas] (n=17), and party---role data entry errors (n=18) inflate counts. Taking into account only

new filings of civil

suits PACER returns 176 cases.

PACER also contains criminal case filings. However, a search of criminal court documents revealed no cases in which the

SEC was an interested party in 2010. This should not be viewed as indicative of agency inactivity, but rather as a

limitation in

the data source. Comparing the SEC’s public agency data and PACER’s data, it appears that the SEC’s civil

and criminal

files are more complete than would be generated by broadly searching PACER. The SEC’s litigation releases

contain all of

the matters returned by PACER, in addition to several others that were omitted using a party---based search

procedure.

Unlike PACER, the SEC provides detailed court documents with rich information at no cost. However,

PACER’s services may

be useful for generating supplemental data once a pool of cases has been established using SEC

(or other) records.

Data from the Federal Justice Statistics Program offer new dimensions of SEC involvement in criminal and civil cases.

According to data from the Executive Office of the US Attorney

, 16 suspects in 12 matters were referred by SEC for

prosecution in 2010; this reflects new activity initiated by the SEC (Table 4). In addition, 11 ongoing criminal matters

involving 13 suspects were concluded; the majority (eight) were declined, including both cases with organizational

defendants. Two resulted in guilty verdicts for postal, wire, and radio fraud. In one case the defendant was sentenced to

three years’ probation while in the other the defendant was sentenced to 87 months incarceration. In the third case,

comparing party names). This reduced the cases eligible for count to 244. PACER records were compared with SEC generated data

through a multistage process. First, the authors searched party names in all SEC litigation releases for 2010. If a case could not be

located in the FY2010 data, the authors then searched the SEC website. If this also failed to return SEC generated case information,

the authors reviewed the legal documents in a case to determine the nature of the case, the parties involved, and the dates

initiated. Seven duplicates were detected during the document review phase, leaving a final count of 237 unique matters.

The case did appear in the SEC News (available online), though was not captured by the original data collection.

Case determined by the parties involved in the suit.

The U.S. Attorney’s Office filed 39 civil cases on the SEC’s behalf in FY2010. This number is not meant to be representative of

agency activity in civil court, and thus should not be compared to the number of civil cases filed by the SEC according to agency or

court data.

The FJSP data are available through ICPSR with an approved data protection plan. In order to isolate criminal matters related to

SEC activity, matters were filtered by investigative agency, if one was provided. While this is the best approximation of criminal

matters resulting directly from SEC enforcement activity, the U.S. Attorney’s case management system, LIONS (from which the FJSP

data are drawn), does not strictly require that a value be entered for the agency variable. Thus, a certain amount of missing data can

be expected.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

charges were attached to a different criminal case against the defendant. Further, it is worth noting that fully 50% of

defendants referred by the SEC for prosecution in 2010 faced a lead charge of mail, postal, or wire fraud. Only 12.5% of

the cases specified securities fraud as the lead charge, and 6.25% specified false claims and statements. Given the lack

of context information in these offense data, one cannot determine whether cases involve defendants who are

suspected

of white collar offenses as defined for this project.

The SEC also refers some cases for civil prosecution. According to the Executive Office of the U.S. Attorney, the agency

referred 39 civil matters in 2010. Of these, approximately one---third (14) involved charges against businesses. The U.S.

Attorney’s Office filed 36 civil cases on behalf of the SEC, some of which may have been referred in previous years.

Consistent with the referral patterns, approximately one---third (13) of cases filed were against businesses. The most

common cause for action was violation of securities laws, though other causes included the enforcement of subpoenas

and warrants, torts, Freedom of Information Act requests, and frauds. Of the 32 cases originally referred by the SEC that

closed in FY 2010, one---fourth have an unclear disposition, listed as having “necessary action taken.” In nearly one---third

of cases, the type of relief sought is unknown. However, in every case in which a specified amount of monetary relief

was requested, the amount granted exceeded this value; 71% of cases exceeded an award of $100,000, and 33 exceeded

$1,000,000.

Summary: Strengths and Weaknesses of Publicly Available Data

Publicly available data on securities offenses are plentiful, especially as made available on the SEC web site. They are

strongest on the agency’s own administrative and civil cases. From these data one can determine the numbers of cases,

the

nature of violations, and the numbers and types of defendants, individual or organizational. In most cases the data

permit

the determination of whether the offenses are those of white collar offenders as defined for this project: those

committing

securities violations in the course of legitimate occupations. Since relatively few criminal cases are brought

against white

collar securities offenders, the SEC web site provides a strong substantive source of data on securities

cases of the sort

the series seeks to count.

But these public data also have a number of important limitations for use in a data series on white collar offenses. They

are available only in litigation documents and reports that must be read, not in easily digested and coded tabular

formats.

Hence converting them to statistical data for a series is a labor intensive operation unless and until reliable

software for

such conversions is made available. In addition the large number of such documents posted during a

period (e.g., a fiscal

year) must be closely examined to discern new cases from those filed in earlier periods, and

important developments

from those of less interest. As the SEC is the only source of information on administrative (or

regulatory) cases, they cannot

be validated by other publicly available data sources.

Comparisons of the SEC web site data with PACER data on civil securities cases indicate that the SEC data is as complete

and more comprehensive. All of the 176 civil cases in PACER, for example, were also contained within the SEC records.

The PACER data required substantially more time to clean than SEC data, present similar format limitations, and impose

an additional cost.

The SEC’s public data are less useful with respect to criminal sanctions. The website reports on only some referrals, and

only at later points in the criminal justice process. On the other hand, the nonpublic EOUSA data provide information on

SEC referrals to U.S. attorneys’ offices for cases of suspected criminal conduct. However, it is not clear how complete

these records are in comparison to nonpublic SEC case files.

But the Executive Office of U.S. Attorneys’ data are also limited. Not all fields are required when responsible parties

enter a matter or case into the system, including the field indicating referring party. To the extent this field is not reliably

filled in, there will be an undercount of SEC referrals. Moreover, criminal offense data often do not permit a

determination

as to whether the violations fit the operational definition of white collar offenses.

Future work on the usability and reliability of data sources for measuring white collar securities offenses includes

consideration of nonpublic data from the Administrative Office of the U.S Courts, more closely evaluating

inconsistencies

and gaps in data and in reporting protocols (e.g., via interviews with data managers), and assessing the

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

consistency of case numbering protocols for tracking cases across organizational boundaries (SEC, offices of the U.S.

attorneys,

main Justice, the federal courts) as they work their way through the legal process from complaints and

investigations through

sanctions, dismissals and appeals.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

U.S. Environmental Protection Agency

Annual Case Statistics, FY2010

Megan E. Collins, Sally S. Simpson, and Peter C. Yeager

The following agency brief provides a synopsis of the white collar enforcement activities carried out by the United

States Environmental Protection Agency (EPA) in Fiscal Year 2010. The EPA is a federal regulatory agency tasked with

protecting human health and the environment; as such the EPA is the agency responsible for the highest proportion

of investigative and referral activity within the environmental domain of white collar crimes.

EPA enforcement takes

the form of administrative actions, civil judicial actions, and criminal actions

in addition to cleanup

and federal facility

enforcement. This document will highlight the administrative, civil, and criminal enforcement activities performed by

the EPA within the domain of all federal environmental offenses.

Overview of Federal Environmental Enforcement Domain

The scope of federal environmental enforcement is vast, and—in addition to the EPA—includes many different

agencies within the United States government. The Environmental and Natural Resources Division (ENRD) of the U.S.

Department of Justice, which enforces civil and criminal environmental laws that protect air, land, water, and other

natural resources, reported having eighteen different federal “client agencies” in 2010. ENRD’s federal criminal and

civil enforcement of environmental laws primarily consists of the Clean Air Act (CAA; 42 U.S.C. § 7401 et seq.), Clean

Water Act (CWA; 33 U.S.C. § 1251 et seq.), Resource Conservation and Recovery Act (RCRA; 42 U.S.C. § 6901 et seq.),

the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA or Superfund; 42 U.S.C. § 9601

et seq.), the Safe Drinking Water Act (SDWA; 42 U.S.C. § 300f et seq.), and the Lead Hazard Reduction Act.

Every three years, the EPA sets National Enforcement Initiatives which are reviewed on a yearly basis. For fiscal years

relevant for this brief (2008-2010), initiatives consisted of CWA (stormwater, combined sewer overflow, sanitary

sewer overflow, and concentrated animal feeding operations), CAA (new source review, prevention of significant

deterioration, and air toxics), RCRA (financial assurance and mineral processing), CERCLA (financial assurance), and

Indian country. National initiatives are selected by the EPA’s Office of Enforcement and Compliance Assurance (OECA),

and serve the purpose of setting priorities based on specific environmental problems, risks, or patterns of non-

compliance.

Because initiatives change across time periods (e.g. the fiscal years 2005-2007 enforcement priority was

In this series of agency briefs, the scope of federal white collar offenses is classified into two grand domains, “Social Regulation”, and

“Economic Regulation”. The environmental domain is classified under the “Social Regulation” grand domain.

Administrative actions refer to non-judicial enforcement actions taken by EPA or a state under its own authority to enforce

Federal Statutes, involving a notice of violation or administrative order, Civil judicial actions are formal lawsuits filed in court.

Criminal actions are when EPA or a state enforces a criminal action, which may be followed by a court conviction and penalty

(http://www2.epa.gov/enforcement/enforcement-basic-information

)

Cleanup enforcement involves identifying companies or persons responsible for contamination at a site, and negotiating with them

to perform the cleanup, by ordering them to perform the cleanup, or having them pay for by another party or the Agency to do it.

While the EPA contributes to all three types of enforcement under the environmental domain (criminal, administrative, and civil),

some agencies only contribute to one or two types of enforcement.

ENRD client agencies include USAID, USDA, USFS, DOC, DOD, DOE, EPA, GSA, USGS, DHS, HUD, DOI, NASA, NRC, State, DOT, Treasury,

and VA. (http://www.justice.gov/enrd/ENRD_Assets/ACCOMPLISHMENT_REPORT_2010_FINAL_5_18_11_Internet-508.pdf)

http://www.epa.gov/fedrgstr/EPA-GENERAL/2007/October/Day-12/g20164.pdf

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Petroleum Refining), longitudinal assessments of EPA activities will likely reveal disproportionate enforcement of the

current initiatives. Thus, the data presented in the current report must be interpreted with the 2008-2010 initiatives

in mind.

Within the domain of environmental enforcement, the EPA investigates and refers the highest volume of cases under

relevant acts, compared to other agencies. On the criminal side, for example, in Fiscal Year 2010, 64.6 percent of

environmental matters

that were referred to United States Attorneys were referred by the EPA

(see Table I).

Table I: Criminal Environmental Matters Referred to U.S. Attorneys, FY2010

Department

Referrals

Percent

Department of Agriculture

4.0

Department of Commerce

3.8

Department of Defense

0.3

Department of Education

0.3

Department of Health and Human Services

0.5

Department of Homeland Security

8.3

Department of Interior

9.1

Department of Justice

3.0

Department of Transportation

1.0

Environmental Protection Agency

256

64.6

Independent Agencies

(not including EPA)

2.3

Transferring Districts

0.3

Other

2.5

Total

396

100.0

Data Source: Legal Information Office Network System (LIONS)

The majority of the remaining referrals came from the Departments of Interior, Homeland Security, Agriculture,

Commerce, and Justice. It should be noted that the number of referrals attributed to some federal agencies may be

undercounted due to internal record keeping practices. This can be due to varied regional data management systems

and preferences, or the status of matters referred (i.e. if a matter is not yet a formal case, or if the case is still open, it

is unlikely to be included in publicly available data). Additionally, the EPA was the investigative agency for 81.2

percent of all CWA, SDWA, RCRA, CAA, and Superfund criminal cases terminated by the Executive Office for U.S.

Attorneys (EOUSA)

in fiscal year 2010 (see Table II). Thus it is clear that the majority of federal criminal environmental

enforcement activity stems from the work of the EPA.

“Matters” are records of federal criminal matters that have been received by United States attorneys or filed before United States

magistrates—not all matters are formally charged as cases. Environmental crimes are classified in the LIONS database as: “Criminal

violations of environmental statutes such as Resource Conservation and Recovery Act (RCRA); Clean Air Act (CAA); Clean Water Act

(CWA), as amended by the Oil Pollution Act (OPA); the Act to Prevent Pollution from Ships (APPS); the Rivers and Harbors Act; the

Deepwater Port Act; the Ports and Waterways Safety Act; the Safe drinking Water Act (SDWA); the Marine Resources Research and

Sanctuaries Act (“The Ocean Dumping Act,” Title 33 Chapter 27); the Comprehensive Environmental Response Compensation and

Recovery Act (CERCLA); the Emergency Planning and Community Right to Know Act (EPCRA); the Federal Insecticide, Fungicide, and

Rodenticide Act (FIFRA); the Toxic Substance Control Act (TSCA), the Hazardous Materials Transportation Act (HMTA), Nonindigenous

Aquatic Nuisance Prevention and Control Act; and the Outer Continental Shelf Lands Act, as well as Title 18 offenses arising out of

these violations.”

The unit of analysis is environmental crime suspects in matters referred (suspects can be individuals or organizations).

These data are records of defendants in federal criminal cases terminated; this means that United States attorneys concluded the

cases in United States District Court during fiscal year 2010. Disposition code and reason are provided in the EOUSA’s Federal Justice

Statistics Program Database.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table II: Federal Criminal Cases Terminated by EOUSA, FY2010

CWA

SDWA

RCRA

CAA

Superfund

Percent

Environmental Protection Agency

81.2

All Other Agencies

18.8

Total

100.0

Data Source: FJSP: Defendants in Federal Criminal Cases—Terminated, 2010

Similar to criminal cases, the most common source of environmental domain civil matters referred to the U.S.

Attorney’s Office is the EPA (38.6 percent), followed by the Department of Interior (30.3 percent) (see Table III). The

vast majority of civil cases in the environmental domain (88.5 percent) involved businesses as respondents; only 11.5

percent of cases involved individuals as respondents. Environmental domain cases referred by the EPA

overwhelmingly involved businesses as participants (88.3 percent).

Table III: Civil Environmental Matters Referred to U.S. Attorneys, FY2010

Department

Referrals

Percent

Department of Agriculture

7.3

Department of Commerce

4.8

Department of Defense

3.8

Department of Energy

1.3

Department of Housing and Urban Development

0.3

Department of Homeland Security

1.0

Department of Interior

121

30.3

Department of Justice

2.5

Department of Labor

0.3

Department of State

0.5

Department of Transportation

3.8

Environmental Protection Agency

154

38.6

Independent Agencies (not including EPA)

5.0

Other

0.8

Total

399

100.0

Data Source: Legal Information Office Network System (LIONS)

Case Sources and Processing

Publicly Available EPA Enforcement Data

The EPA utilizes multiple data systems, public and restricted, to quantify and track enforcement activities; a total of

nine national data systems are currently maintained.

A useful database for the purpose of tracking federal judicial

and administrative EPA cases and observing aggregate trends is the Integrated Compliance Information System (ICIS).

Unlike many of the databases produced by the EPA, ICIS is a multimedia data system, and contains information

Executive Office for United States Attorneys (EOUSA) Legal Information Office Network System data.

The data systems include Air, Hazardous Waste, ICIS, Multimedia (ECHO, IDEA, OTIS, Envirofacts), Toxics/Pesticides, and Water. The

largest publicly available database is ECHO (Enforcement and Compliance History Online), which falls within the Multimedia Data

System, and provides inspection information (e.g. permit, inspection, penalty information) on approximately 800,000 EPA regulated

facilities. ECHO provides detailed facility-specific information.

http://www.epa.gov/compliance/data/systems/index.html

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

regarding enforcement cases and outcomes.

It is useful for our purposes because it contains information on federal

administrative and federal judicial cases under nine environmental statutes.

The publicly available ICIS database

is a subset of the larger ICIS database maintained by the EPA. The public version

includes non-sensitive information on Federal enforcement and compliance (FE&C). Data are supplied and updated

by EPA’s Offices of Regional Counsel and Office of Civil Enforcement case attorneys. Information is recorded at the

case level, and includes case details (case number, case name), violation information (law section, violation date),

defendant name(s), dates of milestone events,

and penalty amounts.

Records of EPA’s criminal enforcement also can be found in the Legal Information Office Network System (LIONS)

database, maintained by the case management staff of the Department of Justice’s Executive Office for the U.S.

Attorneys (EOUSA).

However, unlike the ICIS-FE&C data, LIONS contains defendant-level records, including

information about suspects in criminal and civil matters and cases introduced and closed by United States attorneys

or magistrates. These data include cases referred to the United States Attorney’s Office by the EPA, and provide details

regarding lead charge, litigating responsibility, general offense category, and sentence. The EOUSA data do not

capture any of the defendants or cases that are processed or prosecuted outside the U.S. Attorney’s Office, such as

cases prosecuted by individual agencies, or other branches within the Department of Justice working independently

of EOUSA.

Other public sources of data include the Administrative Offices of the United States Courts (AOUSC) and the EPA

Enforcement and Compliance History Online (ECHO), however each of these sources have shortcomings that limit

their utility in building a comprehensive white collar database. For example the AOUSC data, which include civil and

criminal cases processed in federal court, do not identify the referring agency; as such the involvement of the EPA or

other particular agencies cannot be isolated. The ECHO data include comprehensive descriptions and documentations

of inspection records at the facility level. However, at this point in time, it does not appear that these data may be

downloaded from the EPA website; instead each case must be individually queried. This tool appears to be more useful

for researching specific cases than providing useful and systematic details for a comprehensive database.

The EOUSA and AOUSC data (which can be translated to case level data), milestone activities reported by the EPA (in

ICIS-FE&C), and other data available through the EPA website (such as ECHO) each include fields that can be compared

in an effort to verify counts and case reporting (e.g. referrals, cases opened, and cases closed in a particular fiscal

year). The types of information of interest for a data series on white-collar crime include data on case sources, case

processing and case resolution as indicated in Figure 1.

“ICIS-FE&C Data Elements Included in the IDEA Data Download” document, which describes the dataset and its limitations may be

downloaded in pdf format from: http://echo.epa.gov/files/echodownloads/ICIS_FEC_Data_Download.pdf

Federal administrative and federal judicial cases under the following environmental statutes: CAA, CWA, RCRA, EPCRA, TSCA, FIFRA,

CERCLA/Superfund, SDWA, MPRSA.

Publicly available EPA data may be downloaded from: http://echo.epa.gov/data_downloads

Milestones are “a predefined set of tracking events associated with the enforcement action. The individual milestones available

will vary depending on whether the enforcement action is state or federal and whether it is administrative or judicial.”

(http://www.epa.gov/compliance/resources/publications/data/systems/icis/icis-users-guide.pdf)

A non-sensitive subset of the LIONS data can be obtained from the United States Attorneys’ website:

http://www.justice.gov/usao/reading_room/data/CaseStats.htm#data

or via the Federal Justice Statistics Program, maintained by the

Bureau of Justice Statistics: http://www.bjs.gov/index.cfm?ty=dcdetail&iid=262

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Figure I: Stages of Enforcement Action and Information Generated in Each Stage

Case Initiation

Criminal, administrative, and civil EPA cases can originate from a number of sources (see Figure I and Table IV).

Violations or suspected violations can be reported to local bureaus over the phone or online.

For issues beyond the

scope of local enforcement, citizens can contact state and territorial level EPA offices.

Ultimately, tips and complaints

are funneled through the agency, and an EPA employee investigates. Thus, case creation can be initiated at the local,

state, or federal level.

The ICIS-FE&C database capture some of these forms of case as milestone events that took

place in fiscal year 2010. While the language is not always consistent between EPA documents and the data

descriptions provided, Table IV highlights examples of available case initiation data.

Much federal enforcement activity results from voluntary disclosures or inspection. In FY2010, for instance, EPA

reported that 561 companies and 1,218 facilities initiated voluntary disclosures.

The publicly available ICIS-FE&C

data flagged 407 cases as “voluntary self-disclosures” in FY2010.

Incentives such as lower fines or penalties for self-

evaluation and self-disclosure of violations are offered for voluntary disclosures.

In terms of inspections, the agency

conducted 21,000 inspections and evaluations

in fiscal year 2010; 237 inspections were conducted by tribal

inspectors using federal credentials—these were on behalf of the EPA but not considered “EPA Activity.”

These data were calculated using ICIS, and presented in “Fiscal Year 2010 EPA Enforcement & Compliance Annual Results”

http://www.epa.gov/compliance/resources/reports/endofyear/eoy2010/fy2010results.pdf

It is not clear why the number of voluntary disclosures reported by the EPA does not match the number included in the publicly

available data. Potential explanations include the fact that they are measured in different units (i.e. companies and facilities versus

cases); or, possibly because the public data would not include disclosures that were not yet formally charged, or resolved without

formal action being taken.

According to EPA, incentives for self-policing are as follows: “Where violations are found through voluntary environmental audits or

efforts that reflect a regulated entity's due diligence (i.e., systematic efforts to prevent, detect and correct violations, as defined in the

policy), and all of the policy's conditions are met, EPA will not seek gravity-based penalties and will generally not recommend criminal

prosecution against the company if the violation results from the unauthorized criminal conduct of an employee. Where violations are

discovered by means other than environmental audits or due diligence efforts, but are promptly disclosed and expeditiously corrected,

EPA will reduce gravity-based penalties by 75% provided that all of the other conditions of the policy are met.”

http://www.epa.gov/Region2/capp/cip/policy.htm

The EPA conducts inspections for most statutory and regulatory programs. Inspections include visits to a facility or site to gather

information and determine compliance. The Clean Air Act requires evaluations instead of inspections. A full compliance evaluation

“looks for all regulated pollutants at all regulated emission units, and it addresses the compliance status of each unit, as well as the

facility’s continuing ability to maintain compliance at each emission unit.” A partial compliance evaluation is more thorough than a

Case

Source

•Complaints

•Referrals

•Investigations

•Inspections

•Evaluations

•Voluntary

Self-

Disclosures

Case

Processing

•Case Types

•Charges

•Defendants

Case

Resolution

•Violations

•Penalties

•Referrals

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table IV: ICIS-FE&C Case Initiation Data

Case Processing

Close examination of EPA case processing data reveals the complexities and difficulties involved in quantifying the

agency’s annual enforcement actions. Comparing across data sources within the EPA (e.g., the public EPA databases

and reports produced by the EPA), we see different counts of comparable activities for FY 2010. For example, the EPA

Enforcement and Compliance Annual Results reports that 233 civil judicial enforcement cases were referred to the

Department of Justice in fiscal year 2010, in addition to forty-five supplemental referrals. The ICIS-FE&C data indicate

when a case is “Referred to Department of Justice” and “Referred to Headquarters”, and also provides identifiers for

civil judicial cases. However, these public data show only forty-one unique civil judicial cases referred to DOJ in fiscal

year 2010, and one case referred to headquarters. If we expand the parameters to include all types of judicial

referrals, the total cases reported in ICIS-FE&C (74 cases are referred to DOJ and 3 are reported to headquarter) is still

far below that provided by EPA in their Annual Enforcement and Compliance Results. Additionally, EOUSA data reveal

that in fiscal year 2010, 148 respondents were referred to the U.S. attorney by the EPA for civil cases.

It is unclear why the publicly available ICIS and EOUSA data do not align with the data reported in the EPA publication.

One possibility is that some of the referred cases are still open and thus would not be included in a public database

that could potentially expose sensitive information. Yet, this interpretation is speculative. What is clear is that there

is some kind of internal assessment or decision rule at EPA that produces different counts depending on the data

source utilized.

EPA ECHO data, which can be queried on the EPA website, report 241 EPA enforcement cases were filed/issued in

court (judicial enforcement) and 2,970 cases had complaints/proposed orders issued in fiscal year 2010.

Additionally, 2,530 new administrative dockets were filed in fiscal year 2010—some of these dockets were

components of a single case, so the number is likely inflated with regard to the number of administrative cases

“cursory review of individual reports”, and includes “a documented compliance assessment focusing on a subset of regulated

pollutants, regulatory requirements, or emission units at a given facility.”

(http://www.epa.gov/Compliance/monitoring/inspections/index.html#evaluation)

http://www.epa-echo.gov/echo/compliance_report_icis.html

Case Sources

Data Availability

Counts

Cases

Tips

Complaints

Complaint Filed With Court

Complaint/Proposed Order

Amended Complaint

Written Notice of Violation

Notice of Violation Sent

369

1983

220

1589

Referrals

Referred to Department of Justice

Referred to Headquarters

Supplemental Referral

Voluntary Self-

Disclosures

Voluntary Self Disclosure flag

407

Investigations

Data from ICIS-FE&C “Milestones Activities/Dates” table; all events conducted at the state level

have been removed.

”Counts” refer to each time a particular milestone activity occurred in the database in fiscal year

2010

“Cases” refer to the number of unique activity IDs listed for each milestone activity

Not available in a disaggregated format. Number of investigations and evaluations is available in

“Fiscal Year 2010 EPA Enforcement & Compliance Annual Results”.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

initiated.

However, the ICIS-FE&C data contained 11,330 enforcement action milestone events for 4,176 unique

federal civil, judicial, and administrative cases in fiscal year 2010.

Among these milestones were 1,636

complaints/proposed orders, 220 complaints filed with court, twenty-six written notices of violation, three notices of

violation sent, and two pesticide registration notices sent. As with other data presented in this brief, it is apparent

that the reported number of cases or actions taken in a given fiscal year changes depending on the data source

referenced. Again, it may be the result of different terminology, different levels of measurement, and/or different

practices involving the entering/updating of sensitive information.

On the criminal side, EPA’s Enforcement & Compliance Annual Results report indicated that 346 criminal

enforcement cases were opened, and 289 defendants were charged in fiscal year 2010. The data maintained by

EOUSA indicate that 112 criminal cases referred by the EPA were filed in 2010; these cases included 174 defendants,

of which twenty-eight defendants were organizations. Most organizations and individuals were referred for fraud or

transportation offenses.

The EPA also referred a variety of matters including those for violations of immigration,

customs, intimidation, racketeering, auto theft, and embezzlement laws. Of the organizational cases referred, nine

resulted in the filing of new charges, though the disposition of most (seventy-nine percent) is unknown. The

differences in the reported number of criminal cases opened by EPA and EOUSA may reflect the fact that not all

cases opened by the EPA are referred to EOUSA. Additionally, the sanitized data provided by EOUSA are unlikely to

include ongoing or pending cases, whereas the EPA report, which simply included an aggregate total, could report all

cases without risk.

In the Table below (VI), the type of data that are publicly available are compared by source (EPA ICIS FE&C and the

EOUSA). It is clear that data of interest for a white collar crime series are available, including information regarding

civil, criminal, and administrative cases and respective defendants. However, these data are not exhaustive (e.g.

EOUSA data only refer to matters, cases and respondents referred to the U.S. Attorneys for prosecution, and do not

include issues handled internally), are not all maintained in the same database, and are not all measured in the same

unit. Instead the data are in numerous different databases, kept by different agencies, which use different units of

analysis, terminology, and data maintenance procedures. As such, although information is extensive, merging

across sources would not be seamless.

http://yosemite.epa.gov/oa/rhc/epaadmin.nsf/Dockets+by+Year+Filed?OpenView&RestrictToCategory=2010&Start=1601

Unique cases were identified by the “Activity ID” field in the ICIS-FE&C download; this is a unique identifier assigned to individual

cases. Each activity ID number may be included multiple times within the database, as it is referenced with each milestone activity

relevant to the case.

The most commonly cited lead charges for EPA-referred criminal cases involving organizations charged with fraud are: 42§7413

(n=3) and 42§6928 et seq. (n=7); among those charged with transportation offenses, the most common lead charge statute is

33§1319 et seq. (n=4).The most commonly cited lead charges for EPA-referred criminal cases involving individuals charged with

fraud are: 42§7413 (n=36) and 42§6928 et seq. (n=13); among those charged with transportation offenses, the most common lead

charge statute is 33§1319 et seq. (n=37).

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table VI: ICIS-FE&C and EOUSA Data

FY 2010

Data Availability

Counts

Comments

Case Numbers

Incomplete

ICIS-FE&C: “ACTIVID” (activity identifier) is an internal numeric key

identifier used to uniquely identify enforcement case activities.

Date Case Initiated

Yes

Num. of Viols./Def.

Incomplete

Available for some defendants in EOUSA, but has a maximum

number of violations recognized.

Types of Violation

Yes

Both the ICIS-FE&C and the EOUSA criminal datasets include statute

level information; as such, frequencies of charging and/or referring

specific environmental enforcement statutes can be obtained.

Dates of Viols.

Incomplete

Codebook indicates that date of violation should be available; does

not appear to be included in public download

Defendant Chars.

Yes

The ICIS-FE&C data include defendant/respondent name, facility

UIN,” primary name, and location address. The EOUSA civil and

criminal data include a flag that identifies if the

defendant/respondent is an individual or a business

Case Seriousness

Incomplete

EOUSA criminal case data includes a field “priority”, which indicates

if the case is a national and/or district priority

Administrative Cases

Yes

***

ICIS-FE&C includes “referred to DOJ count”—unclear if this is a

complete measure.

Civil Referrals

Yes

148

EOUSA contains civil respondents referred to the U.S. Attorney

Criminal Referrals

Yes

112

EOUSA contains criminal cases referred to the U.S. Attorney

Not available in a disaggregated format. Number of investigations and evaluations is available in “Fiscal Year 2010 EPA Enforcement & Compliance Annual

Results”.

Case Resolutions

Both ICIS-FE&C and EOUSA data provide information about case conclusions, settlements, and sanctions. The ICIS-

FE&C milestone activity variable reports a total of 526 settlements lodged, and 4,005 settlements entered in fiscal

year 2010 (Table VII) across the most common statutes enforced by the EPA. Most settlements were lodged and

entered under the Clean Water Act (95 and 1,015, respectively), Clean Air Act (146 and 701), and CERCLA/Superfund

(202, 441). Additionally, “milestone activities” reported in ICIS-FE&C included Final Order Issued (n=3,556),

Enforcement Action Closed (n=2,397), NPDES Closed (n=162), Pipeline Closed (n=456), Final Order Lodged (n=552),

Final Order Entered (n=529), Expedited Settlement Offered (n=88), Concluded (n=215), Compliance Achieved

(n=112).

Table VII: ICIS-FE&C Settlements by Statute

Settlements

CAA

CERCLA

CWA

EPCRA

FIFRA

MPRSA

RCRA

SDWA

TSCA

Total

Lodged

146

202

526

Entered

701

441

1015

430

239

467

527

184

4005

The publicly available ICIS data also contain information on the penalties incurred in cases settled in fiscal year 2010

(Tables VII).

Federal Penalty Assessed (ENFCFPA) is defined as the “Federal Penalty assessed against the defendant(s)

as specified in the final entered Consent Decree or Court Order” for civil enforcement actions, and “the penalty

assessed in the Consent/Final Order” for administrative actions. Interest payments associated with the fines are not

These data are from the publicly available ICIS database.

http://www.epa-echo.gov/ideadownloads/2012/ICIS_FE&C_summary.pdf

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

included. The Total Compliance Action Amount (ENFCCAA) figure refers to the settlement level value of “injunctive

relief and the physical or nonphysical costs of returning to compliance” (injunctive relief refers to the actions an entity

must undertake to achieve and maintain compliance). Cost Recovery Awarded Amount (ENFCRAA) is “the amount of

cost recovery ordered or agreed to be repaid by the responsible party or parties and due the Superfund in accordance

with either an administrative or judicial settlement”. These data provide additional detail regarding penalties assessed

for the cases captured in the ICIS-FE&C database.

Table VIII: ICIS-FE&C Penalties Assessed

Count Minimum Maximum Mean Total

Federal Penalty Assessed 144 $ 0 $ 15,000,000.00 $ 683,798.52 $ 98,466,989

Total Compliance Action

Amount

159 $ 500.00 $ 3,700,000,000.00 $ 64,126,468.82 $ 10,196,108,542

Cost Recovery Awarded

Amount

99 $ 0 $ 49,549,379.00 $ 2,203,078.24 $ 218,104,746

Criminal outcome data obtained through the EOUSA data are limited. Of eighty-six criminal cases concluded by the

U.S. Attorneys in 2010, fifty-three cases were pled in district court, eleven were pled in magistrate court, three were

disposed by the government from district court, six disposed by the government from District Court, and three by jury

trial verdict in District Court.

The EOUSA also reported concluding 230 matters in 2010 (see description of “matters”

in footnote six); ten of these cases were initially received before 2005, though the majority of cases (seventy-four

percent) were received in 2008 or later. Slightly more than twenty-eight percent of cases concluded in 2010 were

received the same fiscal year. In the Federal Justice Statistics Program’s current form, it is not possible to track cases

from their initial referral to EOUSA (in the EOUSA database) to sentencing (in the U.S. Sentencing Commission

Database), as there are many technical and security obstacles that make linking cases particularly difficult. As such,

we are unable to report on sanctions imposed for fiscal year 2010 criminal cases using the data that are currently

publicly available.

Table IX: ICIS-FE&C and EOUSA Datasets

FY 2010

Data

Availability

Counts

Comments

Case Numbers

Case numbers are not available, however sterile

identifiers for cross-activity comparison are

available in some data:

ICIS-FE&C: “ACTIVID” (activity identifier) is an

internal numeric key identifier used to uniquely

identify enforcement case activities.

EOUSA: “SEQNUM” and “CASE_ID” do not include

actual case numbers, but a within-organization

identifier (cannot be used to link data across

agencies).

Date(s) of Resolution

***

ICIS-FE&C: “ENFCSLD” settlement lodged date; and

“SUBSCAD” sub-activity/milestone date.

Defendant

Characteristics

***

ICIS-FE&C: “DEFFENNM” defendant name;

“FCLTUIN” facility UIN; “FCLTYNM” primary name;

“FCLTYAD” location address.

An additional two cases were declined: one for jurisdictional/venue problems and another cited office policy.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Administrative Cases

Resolved

Case/Defendants

13,524

In 2010 13,791 formal administrative counts were

resolved for 13,524 cases (ICIS-FE&C); defendant

names are also included and can be tabulated

No. of cases dismissed

Inc.

***

ICIS-FE&C: “ENFOOUT” (enforcement outcome)

includes fields such as “DOJ Declined”, “Dismissed

by Tribunal”, and “US Attorney Declined”

Number of violations

penalized

Number of sanctions

imposed

Inc.

ICIS-FE&C: various measures of total penalty

assessed; such as Federal penalty assessed

amount, and total compliance action amount

Civil Cases Resolved

Case/Defendants

219

Civil cases concluded (EOUSA data—unit of analysis

is respondent) ; defendant names are also included

and can be tabulated

No. of cases dismissed

Inc.

Voluntary dismissals (EOUSA data, civil cases

concluded)

Number of violations

penalized

Number of sanctions

imposed

Inc.

ICIS-FE&C: various measures of total penalty

assessed; such as Federal penalty assessed

amount, and total compliance action amount

Criminal Cases Resolved

Case/Defendants

Criminal cases concluded (EOUSA data), 107

criminal defendants

Number of violations

penalized

Cannot identify cases referred by EPA when

analyzing outcome/sentence data

Number of sanctions

imposed

Cannot identify cases referred by EPA when

analyzing outcome/sentence data

Appeals

Inc.

Criminal Appeals (EOUSA Cases Concluded)

Civil Appeals (EOUSA Civil Cases Concluded); the

reliability of this figure is questionable, as zero

appeals were reported for all 75,735 cases in the

database

Table IX identifies the kinds of data publicly available to count and measure EPA-related case resolutions (i.e. ICIS-

FE&C and EOUSA data). Although this table demonstrates that many data fields of interest are publicly available, it

also reveals that all measures are not available from the same data source; and that the fields are not exhaustive or

measured in the same unit.

Supplemental data are available from other sources and comparisons are reported in Table X. For example, data from

the “Fiscal Year 2010 EPA Enforcement & Compliance Annual Results” report summarize enforcement totals using

internal EPA data. The publicly available ICIS data indicate that for administrative formal cases in fiscal year 2010 3,432

final orders were issued, and 2,353 enforcement actions within 1,880 cases were closed.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table X: Supplemental Data Sources

In addition to the supplemental sources highlighted in Table X, resources such as the EPA’s website and its user

interfaces provide supplemental information, often with rich detail. For example, according to an EPA-affiliated

website 43 environmental appeals were filed, and 38 were closed in fiscal year 2010.

Data from the EPA website

indicate that of 31 civil cases concluded in fiscal year 2010 all respondents were organizations. Of the thirty-one

conclusions, twenty-seven resulted in consent decrees, two reached settlement agreements, one received a consent

decree and final order, and one was issued a stop sale, use, or removal order.

Additionally, publicly available ECHO

data report that in fiscal year 2010, 238 cases received judicial settlements, 109 judicial cases were concluded, 3,998

final administrative orders were issued, and 2,055 administrative enforcement actions were closed.

Summary: Strengths and Weaknesses of Available Data

The quantity and availability of data collected and maintained by the EPA is unsurpassed by any other agencies

evaluated in the course of this project. The publicly available ICIS-FE&C data, in addition to internally produced reports,

and interactive interfaces on the agency’s website merely scratch the surface of what is maintained by the EPA. The

data tap into key elements of case processing that are the essential building blocks of any data series, including case

initiation, handling, and resolution. Yet, even with such comprehensive and user friendly interfaces, there are several

challenges to repurposing these data for a white-collar crime statistical series. The publicly downloadable data

provided by the EPA include both civil and administrative cases, but not criminal cases. Criminal data are collected

and held by EPA, but the data are not publicly available in any systematic way. Terminology is often inconsistent with

other data sources, making merging, cross-comparison, and verification efforts difficult. Similarly, the data maintained

and provided by the EOUSA only include civil and criminal matters and cases referred to the Executive Office for U.S.

http://yosemite.epa.gov/oa/EAB_Web_Docket.nsf/Closed+Dockets?OpenView,

http://cfpub.epa.gov/enforcement/cases/

http://www.epa-echo.gov/echo/compliance_report_icis.html

ICIS-FE&C

EPA Report

EOUSA

PACER

Administrative Cases

Initiated

1589 complaints/

proposed orders

Concluded

1880 cases with

enforcement

actions closed

3332 civil

enforcement

cases concluded

(judicial and

administrative)

Civil Cases

Initiated

41 cases referred

172 civil judicial

complains filed

233 civil judicial

referrals

148 respondents

referred

Concluded

205 cases closed

and concluded

200 civil judicial

enforcement

cases

219 respondents

concluded

9 concluded (EPA

was defendant in

seven, plaintiff in

one, movant in

one)

Criminal Cases

Initiated

346 new

environmental

crime cases

opened

112 referrals

Concluded

86 concluded

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Attorneys, and miss anything handled, prosecuted, or dismissed “in-house”, or exclusively by partnering agencies.

Unfortunately, other datasets such as AOUSC or Public Access to Court Electronic Records (PACER) do not indicate

which agency referred the cases being processed; as such they cannot be relied on for information regarding EPA-

initiated or involved enforcement efforts.

Finally, year-to-year trends of EPA enforcement activity should not be interpreted without context. The EPA bases

enforcement priorities on specific environmental problems, risks, or patterns of non-compliance that are salient at

the time. A longitudinal assessment of EPA data might reveal disproportionate enforcement of the current initiatives,

which may not have been a priority area in earlier years. Thus, changes in enforcement activity reflected in the data

may not represent a change in compliance, but instead a change in enforcement focus.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

U.S. Federal Trade Commission

Annual Case Statistics, FY2010

Miranda A. Galvin, Peter C. Yeager, and Sally S. Simpson

The Federal Trade Commission’s (FTC) mission comprises

preventing business practices that undermine competition and

protecting the interests of consumers. In the sphere of economic

regulation, the FTC enforces rules to ensure the competitiveness

of markets and to protect consumers from deception and other

unfair practices in the marketplace. The FTC seeks to accomplish its goals

without “burdening legitimate business activity.”

In practical terms, the FTC is responsible for preventing mergers and other business practices that would unfairly limit

competition (e.g., anticompetitive agreements), and addressing fraud and deception in the marketplace. The consumer

protection mission of the FTC is shared with the Consumer Financial Protection Bureau, created as part of the 2010 Wall

Street Reform and Consumer Protection Act, as well as other agencies such as the Consumer Product Safety Commission.

In addition, the FTC shares its merger review power with the Department of Justice’s Antitrust division and the

Congressional finance committees. The Commission has a working relationship with the Antitrust division that typically

results in the Antitrust division reviewing mergers within regulated industries, while the FTC division oversees all other

mergers.

The FTC was established in 1914 to address anticompetitive mergers and has since grown in importance. The Commission

is now responsible for overseeing the administration and enforcement of more than 70 laws.

Of these, 25 laws relate at

least in part to maintaining competition, including the Clayton Act, the Truth in Lending Act, and the Telemarketing and

Consumer Fraud and Abuse Prevention Act. However, many of these laws require the Commission to act in an advisory

or reporting role, rather than in an enforcement one. Fifty-six laws relate to consumer protection, including the Bankruptcy

Abuse Prevention and Consumer Protection Act of 2005, the Consumer Leasing Act, and the Credit Card Accounting

Responsibility and Disclosure Act of 2003. The FTC shares enforcement authority with the Department of Justice and state-

level and private parties (i.e., injured businesses) under the Clayton and the Sherman Antitrust Acts.

The FTC has the power to respond to cases through both administrative and civil actions, as well as to conduct

investigations and make reports to Congress. Administrative means include both adjudication and rule-making. In the

latter case, the FTC does not respond directly to suspected violations but rather generates a new rule with the intention

of generating reform at the industry-level. Adjudication, however, occurs at the firm level and thus is of interest in the

development of a white-collar crime statistical series. Unless the respondent does not contest the FTC charges and waives

judicial review, a case will be heard before an administrative law judge. Final orders may be appealed all the way to the

U.S. Federal Trade Commission. About the Federal Trade Commission, http://www.ftc.gov/about-ftc. Accessed 15 May 2014.

Staff Meeting at U.S. Federal Trade Commission, 8 May 2013.

U.S. Federal Trade Commission. Statutes Enforced or Administered by the Commission, http://www.ftc.gov/enforcement/statutes.

Accessed 15 May 2014.

See Appendix A. U.S. Federal Trade Commission. Enforcement Authority, http://www.ftc.gov/about-ftc/what-we-do/enforcement-

authority. Accessed 15 May 2014.

Domain

Sphere

Economic

Regulation

Financial

Market

Competition

Consumer

Protection

Fig. 1: The FTC’s Regulatory Domain

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

protection violations are handled through such civil means.

The types of information of interest for a data series on white collar offending include data on case sources, case

processing and case resolutions or outcomes, as indicated in Figure 2. On its website

the FTC provides nearly exhaustive

information on case processing and resolution, while providing little to no information as to the processes and sources of

case origination. It is also worth noting that what data are provided on the website are embedded within drafted releases

and electronic case documents and are not readily available in a quantitative format. All actions within a case are

documented on a continually updated case-timeline webpage.

Case Sources and Processing

An agency may become aware of a

potential violation of law through a

variety of means, such as complaints,

referrals, and investigations (Fig. 2).

Complaints comprise information from

consumers or market participants

regarding suspected illegal activity. The

FTC offers consumers a direct means to

report suspected issues via the

Consumer Sentinel Network. The FTC

releases a calendar year report on the Consumer Sentinel complaints, and reports that more than 1.3 million complaints

were made in calendar year 2010

(Table 1). The bulk of these were for fraud (54%) and identity theft (19%). The total

numbers of investigations, referrals, and tips are not provided, nor can they be assembled from the reported data.

Once a case has been initiated and violations have been substantiated, the

FTC may seek either administrative or civil sanctions against individual and

organizational defendants. Most administrative cases are heard before an

Administrative Law Judge (ALJ), though the FTC also may issue warning

letters.

After the hearing the ALJ will issue an order making findings and,

if illegal conduct is found, impose sanctions. Multiple types of sanctions

are available through an ALJ hearing, including civil monetary penalties

and orders to divest assets.

The Commission uses civil court action to seek injunctions and monetary

equitable relief in antitrust and consumer protection cases. In the latter

type of cases the FTC most often pursues court enforcement rather than

administrative adjudication.

Civil cases are heard in federal district courts. Formal proceedings are initiated through

See Appendix A. U.S. Federal Trade Commission, Enforcement Authority, http://www.ftc.gov/about-ftc/what-we-do/enforcement-

authority. Accessed 15 May 2014.

www.ftc.gov

The FTC recently changed its website. The previous enforcement search tool allowed for the search of cases by mission, opening

date, and closing date, as well distinguished merger from non-merger competition cases.

U.S. Federal Trade Commission. Consumer Sentinel Data Book for January-December 2010,

http://www.ftc.gov/sites/default/files/documents/reports_annual/sentinel-cy-2010/sentinel-cy2010.pdf

. Accessed 27 May 2014.

Calendar-year reports are available on the FTC’s website. In 2009 and 2010, the FTC reports more than 1.3 million annual

complaints. The reports do not allow determination of fiscal year counts.Federal Trade Commission. Consumer Sentinel Network

Reports. http://www.ftc.gov/enforcement/consumer-sentinel-network/reports

Accessed 16 May 2014.

Warning letters were previously available on the Commission’s website but are no longer as of May 2014.

U.S. Federal Trade Commission, A Brief Overview of the Federal Trade Commission’s Investigative and Law Enforcement Authority,

http://www.ftc.gov/about-ftc/what-we-do/enforcement-authority

. Accessed 3 June 2014.

Case

Sources

Data Availability

Tips

Complaints

~1,300,000

Referrals

Investigations

Table 1: Case Origination Data. NA=Not Available.

Case

Source

•Complaints

•Referrals

•Investigations

Case

Processing

•Case Types

•Charges

•Defendants

Case

Resolution

•Violations

•Penalties

•Referrals

Fig. 2: Stages of enforcement action and information generated in each stage

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

the filing of a civil the filing of a civil complaint. Arguments and responses are heard, and subsequently findings are

made and sanctions imposed. In some cases, the FTC case page references a parallel proceeding. To the extent that this

is reliably and consistently done, it contributes to the accuracy of a white-collar crime statistical series.

The FTC initiated 40 administrative cases and 45 civil cases in FY 2010 (Table 2).

Of these 85 cases, 26 related to

competition and 59 to consumer protection. All but one competition case was handled through administrative

procedures, while only 15 consumer protection cases were managed with these procedures.

Suspected criminal conduct that is uncovered during the course of an FTC investigation is subsequently referred to either

the Department of Justice or directly to an Office of the U.S. Attorney for prosecution. The FTC’s Criminal Liaison Unit acts

as a chief source of evidence for federally prosecuted consumer fraud offenses. Very little information is provided about

the unit, and no enforcement reports are available. It is thus not possible to attain information from the FTC’s publicly

available data on criminal cases stemming from the agency’s investigations.

The number of administrative and civil cases opened in FY 2010 was arrived at by searching the FTC’s cases and proceedings

database (http://www.ftc.gov/enforcement/cases-proceedings

). The search function does not operate consistently, though it did

cooperate on 20 May 2014 to allow for a query of cases started 1 October 2009 through 30 September 2010. The list of cases

provided was copied into Excel, and each case was subsequently coded for the type of action (administrative or civil), mission

(competition or consumer protection), and whether the case was opened and/or closed in fiscal year 2010. The number of civil cases

initiated was thus arrived at through the same process. The search returned 95 results, but of these ten were not opened in FY2010,

leaving 85 cases initiated in 2010. It is important to note that the database was also queried on 16 May 2014 and returned only 89

results; it is not immediately clear why some cases were omitted. It is important to note the technological limitations of accessing

public data.

Data

Availability

Counts

Comments

Administrative Actions

In rare instances, a case may not be “tagged” as

administrative or civil; searching on this term alone

would return incomplete counts. Also, it is not

possible to separate types of administrative actions

automatically.

Civil Case Filings

Included in this count are both cases in which the FTC

was plaintiff and cases in which the United States

Attorney filed the complaint on the FTC’s behalf.

Criminal Referrals

Case Numbers

Both administrative docket number and civil case

filing numbers

Date Case Initiated

Number of Defendants

Contained in the complaint document

Types of Violation

Contained in the complaint document; also possible

to search by mission or type of action

Dates of Violations

Defendant Chars.

Corporate characteristics available in complaints

Case Seriousness

Table 2: Cases Initiated Data

NA = Not Available; NR=Not Relevant

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

The FTC files provide extensive documentation for those determined to query them. Cases are organized by matter,

contain detailed timelines of case progression, and contain information on parallel proceedings and appeals. However,

the nature of the data (electronic internal and court documents or summary releases) requires significant time

investments and a systematic coding system. In Table 2 it is clear that currently available data on FTC activity provides

most data points of interest to a data series on white collar offenses.

Case Resolutions

The data available for FTC cases resolved in 2010 are as rich and problematic as are the data related to case initiation.

Case resolution information is available through the same means as case initiation and processing data. Using search

features, it is possible to isolate cases that were “updated” in fiscal year 2010. Cases are updated when they are resolved

with sanctions, and also when earlier and later legal steps are taken in cases. In the latter situation, for example, cases

are updated when consent orders are terminated or funds are disbursed to victims. For FY 2010 79 cases (of 105 updated)

contained case resolutions. Thirty-two

of these cases were administrative in nature, and 47 were the result of civil

procedures. Almost half of cases resolved were filed the same year: 23 administrative and 14 civil cases were both initiated

and resolved in 2010.

Consumer protection made up the bulk of resolutions, with 61 such cases resolved in 2010. The remaining 18 cases

sanctioned violations of the rules of competition. These competition cases were exclusively resolved administratively. In

comparison, only 14 of the consumer protection cases were resolved using administrative action, with the remaining 47

handled in civil court.

It is important to note, however, that using the query tool is assuredly an undercount of resolved cases; cases in which a

sanction was imposed in FY2010, that then were updated in 2012 for an appeal or a consent order was terminated, for

example, will not be located using the query tool’s “updated date” feature. For this reason, Table 3 does not include

definitive case counts at this stage. Of the 105 cases that were updated in fiscal year 2010, 26 were most recently updated

for an event other than case dispositions. Of these, the most common reason for updating a case was for victim

compensation (n=9). Other reasons included the appointment of a monitor or receiver, the filing of contempt charges

after a judgment, modifying or supplementing a previous order, terminating an order, or filing petitions.

This number was determined by coding all cases in the FTC’s cases and proceedings database

(http://www.ftc.gov/enforcement/cases-proceedings

) that were updated between 1 October 2009 and 30 September 2010. The

automatic search feature did not return any cases updated during this time period, however the number could be arrived at

manually by scrolling through case lists (which are listed in time order as updated). The list of cases provided was copied into Excel,

and each case was subsequently coded for the type of action (administrative or civil), mission (competition or consumer protection),

and whether the case was opened and/or closed in fiscal year 2010. The number of civil cases initiated was thus arrived at through

the same process. The search returned 105 results, but of these, 26 were not resolved in FY2010, returning 79 total cases.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

FY 2010

Data Availability

Counts

Comments

Case Numbers

Admin, Civil

In some cases, multiple case numbers will be provided; these are

often consecutive, though some case numbers begin with the letter

X. It is not clear what purpose these multiple case numbers have.

Date(s) of Resolution

Admin, Civil

Both litigation releases and administrative files contain dates

filed; if respondents are sanctioned on different days, the

documents will be filed under the full matter title but will

reference specific respondents

Administrative Cases Resolved

***

This is possible to attain by systematic coding of every case file,

though is not ascertainable through the available query tool.

No. of Admin cases

dismissed

It is not possible to isolate dismissed cases

Number of Defendants

***

Number of violations

penalized

***

Statutes violated usually contained in the Complaint and order;

“counts” of each violation are not provided

Number of sanctions

imposed

***

Detailed in order

Civil Cases Resolved

***

This is possible to attain by systematic coding of every case file,

though is not ascertainable through the available query tool.

Number of civil cases

dismissed

Number of Defendants

***

Electronic files name applicable defendants

Number of violations

penalized

***

Statutes violated usually contained in the original complaint if not

release containing final order; “counts” of each violation are not

provided

Number of sanctions

imposed

***

Releases contain information on both the number and type of

sanctions imposed for each defendant, but may not be exhaustive

due to the summary nature of litigation releases Detailed

information on civil penalties can be ascertained by accessing

court documents

Criminal Cases Resolved

There is no information on criminal cases or investigations

resulting from agency activity

Cases Declined

Number of Defendants

Number of violations

penalized

Number of sanctions

imposed

Appeals

***

Case filings include appeals as they are filed and decided

Table 3: Case Resolution Data

***Available in case timelines but not independently searchable using query tool

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Other Data Sources

The FTC provides relatively complete data on administrative actions, less complete information on civil cases, and the

least complete data on criminal cases. Thus, the final portion of this brief will turn to other available data sources to

supplement and validate the FTC’s data. Civil and criminal case data can be found in both the federal website Public Access

to Court Electronic Records (PACER) and the Federal Justice Statistics Program (see Table 4).

Civil case records form the Commission may be supplemented through PACER. For a nominal fee this site allows users to

search for cases in criminal, appeals, bankruptcy, and civil court files by a number of parameters.

A detailed search initially yielded a significant difference in the number of apparent civil cases filed as reported by PACER

and FTC records. Both sets of records were cross-checked for completeness. Most complaints posted on the FTC website

did appear in PACER filing results, but not all cases present in the FTC’s adjudicative proceedings list appeared in the PACER

records search using the specified party search terms “FTC” and “Federal Trade Commission.” The most common reason

that a case did not appear in PACER search results was that the plaintiff was in fact the Department of Justice. Thus the

FTC appears to track in its publicly available data cases that it had referred to DOJ. One new case filing was listed in PACER

under a different FTC role (i.e. the FTC was listed as a counter-defendant), and thus was excluded from the original counts.

However, in no case was the data absent from the PACER system in entirety; searching by case number invariably returned

case information.

Of the 48 unique cases found in PACER, eight were absent from the FTC adjudication record search results in 2010. This

was primarily due to the FTC case proceedings’ query tool; when records were searched by party name rather than release

date, seven of the eight anomalous records were located. The remaining case was a petition to enforce a civil investigative

demand and was dismissed; thus, no record of the enforcement action exists on the FTC website. PACER appears to over-

count FTC activity; party-based searches return multiple case numbers related to the same civil case.

Overall, PACER

court records were found to suffer from data entry errors

and inconsistencies that limit their usefulness in

determining the volume of agency enforcement activity.

Duplicate cases (n=41 of 98 search results), secondary

filing activity (n=4) and party-role data entry errors (n=8)

inflate counts. Taking into account only new filings of civil

suits PACER returns 45 cases, though this is not a perfect

overlap with the 45 cases found in the FTC data. PACER

returns some cases missed by the FTC’s query tool (due to

listing after the date of filing, e.g.), while not capturing some cases reported by the FTC (e.g. ones brought on behalf of

the FTC by the Department of Justice). Thirty-seven of the 45 cases were present in both data sets.

In order to best target PACER cases in which the Federal Trade Commission brought action against an individual or entity, the

researchers searched for all cases in which “Federal Trade Commission” or “FTC” (using exact match) was a party. (The search term

“FTC” [not exact match] was not used due to the over-inclusion of irrelevant cases.) This procedure returned 98 cases. To filter out

cases against the Commission, the case counts only considered those in which the FTC was listed as plaintiff, movant, or petitioner.

This restricted list contained 89 cases. However, it became apparent that the same matter could appear multiple times within the

search under different case numbers; significant actions (e.g. motions) were recorded under different case numbers despite involving

the same individuals. An effort to identify and remove duplicates was made both automatically (using Microsoft Excel) and manually

(by comparing party names). Forty-one duplicates were detected during the document review phase, leaving a final count of 48 unique

cases. PACER records were compared with FTC generated data through a multistage process. First, the authors searched party names

in all FTC litigation releases for FY 2010. If a PACER case could not be located in the FY 2010 data, the authors then searched the FTC

website using the general search feature. If this also failed to return agency generated case information, the authors reviewed the

legal documents in a case (through PACER) to determine the nature of the case, the parties involved, and the dates initiated.

Case determined by the parties involved in the suit.

FTC

PACER

EOUSA

Administrative Cases Initiated

Civil Cases Initiated

Before Data Cleaning

Civil Cases Referred

Criminal Investigations

Criminal Cases Referred

Table 4: Source Comparisons

NA=Not Available

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Comparing the FTC’s public data and PACER’s data, it appears that the FTC’s civil files are more complete than would be

generated by broadly searching PACER. Unlike PACER, the FTC provides detailed court documents with rich information

at no cost. However, the effort to access the FTC’s own data is hampered by an erratic query tool. Meanwhile, PACER’s

services may be useful for cross-checking the FTC data and generating supplemental data, such as defendant

characteristics.

PACER also contains criminal case filings. While the Department of Justice bears the responsibility for filing criminal cases

investigated by the FTC, the PACER system allows for the listing of referring agencies as “interested parties.” However, a

search of criminal court documents revealed no cases in which the FTC was an interested party in 2010 (or any other year).

Because the EOUSA data also do not report any criminal case filings referred by the FTC in FY 2010, it is not possible to

make a statement about the quality of either data source.

Data from the Federal Justice Statistics Program offer new dimensions of FTC involvement in criminal and civil cases.

According to data from the Executive Office of the US Attorney

, the FTC did not refer any cases for prosecution in 2010,

nor were any existing matters closed.

In addition to filing its own civil cases, the FTC also refers other cases to the Department of Justice for civil prosecution.

According to the Executive Office of the U.S. Attorney, the agency referred 23 civil matters in 2010. Of these,

approximately three-quarters (17) involved charges against businesses; activity against individuals is less common. The

U.S. Attorney’s Office filed 19 civil cases on behalf of the FTC, some of which may have been referred in previous

years.

Approximately three-fourths (14) of cases filed were against businesses. The most common sanctions being

sought were civil penalties or the enforcement of an existing order. Of the 21 cases originally referred by the FTC that

closed in FY 2010, 57 percent were closed without a trial in favor of the U.S., and another 15 percent were settled out of

court. In nearly 60 percent of cases, the type of relief sought is nonmonetary (e.g., injunction, asset freeze).

Summary: Strengths and Weaknesses of Publicly Available Data

Publicly available data on FTC enforcement cases are plentiful, especially as made available on the FTC web site. They are

strongest on the agency’s own administrative and civil cases. From these data one can determine the numbers of cases,

the nature of violations, and the numbers and types of defendants, individual or organizational. In most cases the data

permit the determination of whether the offenses are those of white collar offenders as defined for this project: those

committing securities violations in the course of legitimate occupations. Although relatively few criminal cases are

brought against white collar competition and consumer protection offenders, the FTC web site provides a strong

substantive source of data on cases of the sort the series seeks to count.

But these public data also have a number of important limitations for use in a data series on white collar offenses. They

are available only in litigation documents and reports that must be read, rather than in easily digested and coded tabular

formats. Hence converting them to statistical data for a series is a labor intensive operation unless and until reliable

software for such conversions is made available. As the FTC is the only source of information on administrative (or

regulatory) cases, they cannot be validated by other data sources.

Comparisons of the FTC web site data with PACER data on civil cases indicate that the FTC data is as complete and more

comprehensive. Though the FTC and PACER return the same number of cases filed in FY 2010, PACER’s number is reflective

of the eight cases in which data was available on the FTC site but did not appear in the original search; the counts are thus

not perfectly overlapped; eight cases present each data source were not present in the other. The PACER data require

The FJSP data are available through ICPSR with an approved data protection plan. In order to isolate criminal matters related to

FTC activity, matters were filtered by investigative agency if one was provided. While this is the best approximation of criminal

matters resulting directly from FTC enforcement activity, the U.S. Attorney’s case management system, LIONS (from which the FJSP

data are drawn) does not strictly require that a value be entered for the agency variable. Thus, a certain amount of missing data can

be expected.

The FTC does list at least some referred cases that are subsequently filed by the Department of Justice. However, this is not a

complete tracking of referred cases (i.e. from referral to filing decision).

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

substantial, time-intensive cleaning, present similar format limitations (in that they do not present readily quantified data),

and impose the additional cost of users fees.

The FTC’s public data are of no use in identifying cases in which criminal sanctions are imposed; the criminal enforcement

section of the website does not contain any list of cases. It is not possible to track criminal case activity, for example, at

the prosecutorial decision to file. On the other hand, while the nonpublic EOUSA data should provide information on FTC

referrals to U.S. attorneys’ offices for cases of suspected criminal conduct, no such records were located in the EOUSA

data for fiscal year 2010.

The query tool on the FTC’s website does not always function as designed. In the course of preparing this brief, at multiple

points the tool failed to return results or yielded inconsistent counts with the same search parameters. Systematic

treatment of all cases (i.e., coding and quantifying all agency case data, without imposing date limits using the query tool)

will return more results than use of the query tool to target specific actions or time periods. This is particularly relevant

for documenting case resolutions. The “updated date” returns many cases in which the last action was not a resolution,

and fails to return cases in which a resolution was made but last action was taken at a later date. Any attempt to curtail

the list of cases using date-based queries will thus return inaccurate counts. Initiation dates are also problematic to the

extent that case postings are separated in time from actual case initiation or filing.

The Executive Office of U.S. Attorneys’ data are also limited. Not all fields are required when responsible parties enter a

matter or case into the system, including the field indicating referring party. To the extent this field is not reliably filled in,

there will be an undercount of FTC referrals. Moreover, statute-based data often do not permit a determination as to

whether the violations fit the operational definition of white collar offenses. For example, statutes do not indicate whether

an individual committed a crime in the course of his occupation.

Future work on the usability and reliability of data sources for measuring white collar securities offenses includes

consideration of nonpublic data from the Administrative Office of the U.S Courts, more closely evaluating inconsistencies

and gaps in data and in reporting protocols (e.g., via interviews with data managers), and assessing the consistency of case

numbering protocols for tracking cases across organizational boundaries (FTC, offices of the U.S. attorneys, main Justice,

the federal courts) as they work their way through the legal process from complaints and investigations through sanctions,

dismissals and appeals.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

U.S. Food and Drug Administration

Annual Case Statistics, FY2010

Megan E. Collins, Sally S. Simpson, and Peter C. Yeager

The following brief provides a synopsis of the white collar enforcement activities carried out by the United States Food

and Drug Administration (FDA) in fiscal year 2010. The FDA exists within the U.S. Department of Health and Human

Services with the purpose of overseeing four core areas: medical products and tobacco, foods and veterinary

medicine, global regulatory operations and policy, and operations. The objectives of the FDA are to protect public

health by “assuring the safety, effectiveness, quality, and security of human and veterinary drugs, vaccines and other

biological products, and medical devices. The FDA is also responsible for the safety and security of most of our nation’s

food supply, all cosmetics, dietary supplements and products that give off radiation.”

Overview of Food and Drug Administration Domain

The FDA oversees a wide array of regulatory activity. Specifically, the FDA has authority to undertake administrative

and judicial enforcement activities. The administrative actions taken by the FDA include product recalls, debarment

of individuals and companies convicted of felonies, withdrawals of product approvals, license revocations, and

disqualification of clinical investigators. The judicial actions taken include seizing products in violation of regulations,

injunctions, criminal prosecutions, and certain civil monetary penalties.

The scope of FDA regulation includes: foods, drugs (prescription and non-prescription), biological products (including

vaccines, blood, and tissue), medical devices, electronic products that give off radiation (including microwave ovens,

x-ray equipment, and sunlamps), cosmetics, veterinary products, and tobacco products.

The FDA does not regulate

areas pertaining to advertising (except for prescription drugs, medical devices, and tobacco products), some consumer

products (except those that give off radiation), illegal drugs of abuse, health insurance, meat and poultry (except for

game meats), restaurants and grocery stores, or vaccines for infectious animal diseases.

The agency’s authority and responsibilities are similar to some other federal agencies, which can cause for some

confusion among consumers. For example, the FDA shares regulation responsibility in the area of pesticides with the

U.S. Department of Agriculture (DOA) and the Environmental Protection Agency (EPA). Similarly, regulation of water

is a shared responsibility, as the FDA oversees the labeling and safety of bottled water, while EPA implements

standards for drinking water from municipal supplies.

Other agencies with regulatory authority in areas similar to the

FDA include: the Federal Trade Commission (advertising), Alcohol and Tobacco Tax and Trade Bureau (alcohol),

Consumer Product Safety Commission (CPSC), Drug Enforcement Administration (drugs of abuse), Food Safety and

Inspection Service (meat and poultry), and Animal and Plant Health Inspection Service (vaccines for animal diseases).

http://www.fda.gov/AboutFDA/Transparency/Basics/ucm192695.htm

http://www.fda.gov/AboutFDA/Transparency/TransparencyInitiative/ucm254426.htm

http://www.fda.gov/AboutFDA/Transparency/Basics/ucm194879.htm

http://www.fda.gov/AboutFDA/Transparency/Basics/ucm203499.htm

http://www.fda.gov/AboutFDA/Transparency/Basics/ucm194879.htm

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Table I: Criminal Food and Drug Matters Referred to U.S. Attorneys, FY2010

Department

Referrals

Percent

Food and Drug Administration

57.6

Food and Consumer Service

8.7

U.S. Postal Service

8.1

All other Agriculture

4.7

Federal Bureau of Investigation

4.7

Bureau of Immigration & Customs Enforcement

4.1

Drug Enforcement Administration

2.3

U.S. Marshals Service

2.3

U.S.D.A. Hazardous Materials Account

1.7

DHHS Office of Inspector General

1.2

General Services Administration

1.2

All other Health and Human Services

0.6

Army

0.6

Department of Justice Inspector General

0.6

Other

0.6

Metropolitan Police Department

0.6

National Park Service

0.6

Total

172

100.0

The breadth of agencies involved in the oversight of food and drug regulation is demonstrated in the Department of

Justice’s Executive Office for United States Attorneys (EOUSA) Legal Information Office Network System (LIONS)

database. Table I provides a summary of agencies referring criminal “food and drug” matters to the EOUSA during

fiscal year 2010.

It is clear that while the FDA refer the majority of food and drug matters received by EOUSA, other

agencies such as the Food and Consumer Service, and the U.S. Postal Service, also handle such matters.

Case Sources and Processing

Publicly Available FDA Data

The FDA provides seven data sets as part of the federal Open Government initiative. These datasets include one

database of product recalls initiated by firms (based on company press releases), in addition to six incident-specific

databases.

Additionally the “openFDA” website is set to launch in September 2014. This will allow easy access to

public FDA data, and allow for coding and downloading of data by those in the public and private sectors. These

datasets will include adverse event reporting, recalls, and documentation.

“Matters” are records of federal criminal matters that have been received by United States attorneys or filed before United States

magistrates—not all matters are formally charged as cases. The LIONS codebook does not explicitly define “food and drug” offenses,

beyond that they are considered to be regulatory public order offenses.

http://www.justice.gov/usao/reading_room/data/Info/APPENDIX_A_NOVEMBER_2010.PDF

The incident specific data sets describe the following isolated recalls: Abbott Infant Formula Recall, Shell Egg Recalls, Hydrolyzed

Vegetable Protein Containing Products Recalls, Pistachio Product Recalls, Peanut Product Recalls, Plainview Milk Cooperative

Ingredient Recall, and Fraudulent 2009 H1N1 Influenza Products.

http://www.fda.gov/aboutfda/transparency/opengovernment/default.htm

http://open.fda.gov/about/

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

The FDA maintains three publicly available, downloadable databases with information regarding inspections

conducted; these include Inspection Observations, Citations, and Classifications. The Inspections Observations

database includes descriptions behind any 483 forms (described in detail below), which are completed when a

potential violation may have been observed.

The Inspections Observations data include the center name (e.g.

Biologics, Drugs, Foods, etc.), cite ID, cited statute, short text description, and long text description. The Inspections

Citations data, which report citations issued following an inspection, include the firm name, location, inspection date,

center, cited statute, and long description of the infraction.

The Inspection Classification Database contains

searchable information on the status of firms subject to FDA inspections (although some inspection information is

withheld until enforcement actions are taken). This database includes fields such as classification (compliance status

of a firm), project area (e.g. food safety and applied nutrition, drug evaluation and research, etc.), firm name, location,

and inspection date.

In addition to inspection data, Warning letters, which are typically generated from inspections, may be downloaded

for particular time periods. These publicly available data include information such as the date the warning letter was

issued, the company, issuing office, subject, if a response letter was posted, closeout date, and a link to the letter.

Other enforcement actions are also classified in publicly available data—for example a running list of firms and

individuals debarred by the FDA is maintained on the website.

Clinical Investigator Disqualification Proceedings are

also public and available for download.

These data include the name, center, status, date of status, date the Notices

of Initiation of Disqualification Proceedings and Opportunity to Explain (NIDPOE) was issued, and link to the

NIDPOE letter (to be described in detail below).

Data regarding the FDA’s criminal and civil enforcement activity can be found in the LIONS database, which is

maintained by the case management staff of the EOUSA

. These are defendant-level records, containing information

about suspects in criminal and civil matters and cases referred to, prosecuted, and closed by United States attorneys

or magistrates. These data include cases referred to the United States Attorney’s Office by the FDA, and provide details

regarding lead charge, litigating responsibility, general offense category, and disposition.

Figure I: Case Processing

Case Initiation

http://www.fda.gov/ICECI/Inspections/ucm250720.htm

http://www.fda.gov/ICECI/Inspections/ucm346077.htm

http://www.accessdata.fda.gov/scripts/inspsearch/

http://www.accessdata.fda.gov/scripts/warningletters/wlSearchExcel.cfm

http://www.fda.gov/ICECI/EnforcementActions/FDADebarmentList/default.htm

http://www.accessdata.fda.gov/scripts/sda/sdNavigation.cfm?sd=clinicalinvestigatorsdisqualificationproceedings

A non-sensitive subset of the LIONS data can be obtained from the United States Attorneys’ website:

http://www.justice.gov/usao/reading_room/data/CaseStats.htm#data or via the Federal Justice Statistics Program, maintained by

the Bureau of Justice Statistics: http://www.bjs.gov/index.cfm?ty=dcdetail&iid=262

Case

Source

•Complaints

•Referrals

•Inspections

•Investigations

Case

Processing

•Case Types

•Charges

•Defendants

Case

Resolution

•Violations

•Penalties

•Referrals

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

The FDA Office of Regulatory Affairs (ORA) oversees all field activities conducted by the agency, including inspections

and enforcement. The agency inspects regulated facilities to determine compliance with regulations and the Food,

Drug, and Cosmetic Act (21 U.S.C. Chapter 9).

The FDA conducts numerous types of investigations, such as routine

and for-cause inspections of companies and individuals that manufacture, distribute, or test regulate products, as well

as criminal investigations.

In fiscal year 2010, the FDA conducted a total 21,012 inspections.

Table II: Case Initiation Data

When objectionable conditions are observed by investigators, they are documented on an FDA Form 483. The FDA

provides summaries of regulation areas cited on 483s, by fiscal year. These summaries do not describe all

observations, but demonstrate the number of times an area of regulation was cited as an observation during

inspections. The regulatory areas observed in 483s filed in fiscal 2010 are summarized in Table III (Inspection

Observations Database). What is not clear is if the disproportionate representation of foods in the database is a

function of a higher percent of food centers, more frequent inspections, or disparate enforcement intensity (among

other possible explanations).

Table III: Inspectional Observation Summaries from FDA Form 483, FY2010

Center Name

483s Issued

Foods

2,976

Devices

976

Drugs

746

Incidental Text

362

Bioresearch Monitoring

343

Biologics

275

Veterinary Medicine

275

Parts 1240 and 1250

176

Human Tissue for Transplantation

133

Radiological Health

Special Requirements

Total 483s in System

6,303

This table does not represent the complete set of 483's issued in FY2010;

some were manually prepared and not available in this format. The sum

of 483's for all Product Areas is higher, as a 483 may include citations

related to multiple product areas, and counted more than once, under

each relevant product center. The actual total 483s in the system is 5,710.

http://www.fda.gov/ICECI/Inspections/ucm255532.htm

http://www.fda.gov/ICECI/Inspections/ucm222557.htm

Ibid.

http://www.accessdata.fda.gov/scripts/inspsearch/

Case Sources

Data Availability

Cases

Tips

Complaints

Referrals

Yes (Referrals to EOUSA: LIONS Database)

43 civil

80 criminal

Voluntary Self-

Disclosures

Investigations

Yes (Inspection Classification Database)

21,012

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Case Processing

Before taking formal enforcement actions, the FDA typically notifies companies and individuals of potential

violations.

For example, the FDA may issue Warning Letters, which inform those inspected of violations documented

that could lead to enforcement action if the violation is not corrected. In fiscal year 2010, the FDA issued 661 Warning

Letters. These were most often issued by the FDA’s Center for Devices and Radiological Health (18.0 percent), Center

for Drug Evaluation and Research (13.3 percent), and the Center for Food Safety and Applied Nutrition (8.5 percent).

The Warning Letters were most often due to illegal drug residue (6.2 percent), unapproved new drugs or misbranding

(5.7 percent), medical device reporting regulation or misbranding (4.4 percent), and premarket approved,

misbranded, or adulterated products (4.4 percent).

The FDA may also issue Untitled Letters, which cite violations,

but do not always meet the criteria for a Warning Letter. Of the 21,012 inspections conducted in fiscal year 2010,

7,379 were followed by voluntary actions taken by the firm inspected; 849 firms had official action taken, and 12,784

had no action taken (Inspection Classification Database).

Other types of violation notices include Import Alerts, which include information affecting new imports, and NIDPOEs,

which are sent to clinical investigators, alerting them that the FDA is considering disqualifying them from future

investigations.

In fiscal year 2010, four NIDPOE notices were issued (Disqualification Proceedings Database).

Initial administrative actions are decided on and undertaken by the FDA. These actions can include import refusals;

biologic license revocations or suspensions; orders of retention; recall, destruction and cessation of manufacturing of

human cell, tissue, and similar products; disqualified, restricted, restrictions removed, and assurance lists for clinical

investigators; recalls and market withdrawals; and debarment (these administrative actions can be appealed to

federal courts).

In fiscal year 2010, six inspectors were disqualified and two were restricted following disqualification

proceedings (Disqualification Proceedings Database); additionally nine individuals were permanently debarred, and

one was debarred for five years (FDA Debarment List).

26,27

Unlike administrative actions, judicial actions (such as

criminal convictions) are decided on and undertaken by the federal courts, at the request of the U.S. Department of

Justice (DOJ) and FDA.

In fiscal year 2010, the FDA referred fifty-two civil (forty businesses and twelve individuals) and 272 criminal (fifty-

eight businesses and 214 individuals) matters to the EOUSA (see Table V).

As previously defined, these matters are

not yet formal cases, but issues referred to EOUSA by the FDA. In the same year, EOUSA opened formal cases against

forty-three civil and 119 criminal defendants, based on matters referred by FDA (Table V).

Table V: Civil and Criminal Matters and Cases Referred to EOUSA by FDA, FY2010

http://www.fda.gov/AboutFDA/Transparency/TransparencyInitiative/ucm254426.htm

http://www.accessdata.fda.gov/scripts/warningletters/wlSearchExcel.cfm

http://www.accessdata.fda.gov/scripts/inspsearch/

Ibid.

http://www.accessdata.fda.gov/scripts/sda/sdNavigation.cfm?sd=clinicalinvestigatorsdisqualificationproceedings

http://www.fda.gov/AboutFDA/Transparency/TransparencyInitiative/ucm254426.htm

http://www.accessdata.fda.gov/scripts/sda/sdNavigation.cfm?sd=clinicalinvestigatorsdisqualificationproceedings

http://www.fda.gov/ICECI/EnforcementActions/FDADebarmentList/default.htm

Note that these data are defendant level. Case level data were not available for civil matters or cases, so all are presented at the

defendant level for purposes of consistency and comparison. There were 191 matters (46 business and 145 individual) referred in

FY2010 involving the 272 defendants mentioned in the text.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Civil Criminal

Matters In Cases In Matters In Cases In

Bus.

Indiv.

Bus.

Indiv.

Bus.

Indiv.

Bus.

Indiv.

Asset Forfeiture

Murder

Review of Admin. Action

Embezzlement

Bankruptcy

Fraud

Constitutionality of Statute

Counterfeiting

Commercial Lit./Gen. Claims

Motor Vehicle Theft

Civil Penalty

Drug Trafficking

FOIA

Other Drug Offenses

Fraud

Food and Drug

Health Care Fraud

Custom Laws

Land Condemnation

Other Regulatory Offenses

Post-Convict. Prisoner Lit.

Racketeering and extortion

Subpoena/Summons/Warrant

Perjury, contempt, intimid.

Torts

Nonviolent sex offenses

Conspiracy, aiding, abetting

Unexpected

Total

214

109

These data come from the EOUSA LIONS database for FY 2010

There is clearly a wide array of data available with regard to case processing activities conducted by the FDA. However,

as made clear by Table VI, there still exist gaps in our knowledge. While the FDA may refer some administrative cases

to federal court if compliance is not achieved through other means (e.g. Warning Letters), it is not clear if this figure

is exhaustively captured by the EOUSA LIONS civil database. Additionally, measures of case seriousness do not appear

within the publicly available data—though seriousness may be indicated in the text of some documents such as

Warning Letters or Dismissal Proceedings.

Table VI: Case Processing Data

FY 2010

Data Availability

Counts

Comments

Case Numbers

Date Case Initiated

Yes

Inspection, Warning Letter, Dismissal dates; EOUSA LIONS data

provide partial dates

Num. of Viols./Def.

Incomplete

Not all observations or citations listed from inspections; cap on

how many violations may be listed in EOUSA LIONS data

Types of Violation

Yes

Dates of Viols.

However, dates of inspections are included

Defendant Chars.

Incomplete

Company/Individual name and location included in databases, no

identifiers listed in EOUSA LIONS

Case Seriousness

Administrative Cases

However administrative actions such as warning letters are

available

Civil Referrals

Yes

43 cases

52 matters

EOUSA LIONS (unit of analysis is defendant, not case/count)

Criminal Referrals

Yes

119 cases

272 matters

EOUSA LIONS (unit of analysis is defendant, not case/count)

Case Resolutions

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

As the case processing description indicates, most FDA matters are concluded without the necessity of generating a

formal case, or going to trial. Instead, the high volume of inspections and investigations often result in formal or

informal warning letters, and are followed by voluntary acts of compliance. However, it is also demonstrated in the

LIONS data, that the FDA does refer a number of matters to EOUSA each year, and many are made into formal cases.

In fiscal year 2010, thirty-three civil and 137 criminal defendants’ cases were concluded. These cases were referred to

the EOUSA by FDA, and captured in the LIONS data. Most of the civil respondents’ cases were closed once “necessary

action” was taken (n=13), or after a judgment, order, or decision was reached, without a trial (13 in favor of U.S., 1 in

favor of opposition). The remainder of civil respondents’ cases were concluded by settlement without litigation (n=3),

voluntary dismissal (n=2), and post-disposition action (n=1). Most of the criminal defendants’ cases were concluded

with guilty pleas (n=98), while others were dismissed (n=15), or found guilty by jury trial (n=2). Additionally, cases for

twenty-two criminal defendants were either declined or pending.

Table VI: Case Resolution Data

FY 2010

Data

Availability

Counts

Comments

Case Numbers

Date(s) of Resolution

Yes

Defendant

Characteristics

No identifiers listed in LIONS except whether the

defendant is an individual or business

Administrative Cases

Resolved

Case/Defendants

No. of cases dismissed

Number of violations

penalized

Number of sanctions

imposed

Civil Cases Resolved

Case/Defendants

Yes

EOUSA LIONS (unit of analysis is defendant, not

case/count)

No. of cases dismissed

Yes

Number of violations

penalized

Number of sanctions

imposed

Criminal Cases Resolved

Case/Defendants

137

EOUSA LIONS (unit of analysis is defendant, not

case/count)

Number of violations

penalized

Number of sanctions

imposed

Appeals

Summary: Strengths and Weaknesses of Available Data

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

It appears that the FDA has more publicly available data than some other agencies covered in this series.

Transparency initiatives have resulted in the numerous investigation, warning, and recall databases discussed in this

brief. As a result, thorough information about specific events and companies is made available. However, these data

are not necessarily exhaustive (e.g. the public spreadsheet of post-inspection 483s is missing manually prepared

forms, which are not included in the electronic database). Additionally, the FDA does not provide much data about

incidents that are made into formal cases; the only information available on cases was extracted from LIONS.

Currently, when DOJ files a case for the FDA, the enforcement action may be made public through a press release;

another release may be published following the case’s closure.

While these press releases are available on the

FDA’s website, there is no comprehensive, publicly available list of court actions pursued by the FDA. Legal case

searches using the Public Access to Court Electronic Records (PACER) and Lexis Nexis Academic databases revealed

that the FDA did not bring any cases to trial itself (i.e. independent of DOJ) in fiscal year 2010, but was a defendant

on a number of occasions.

The biggest obstacles that must be overcome for the FDA data to be of use for a white collar statistical series, are

increased access to information regarding matters that are not settled with non-judicial action, such as a warning

letter, and additional opportunities to verify the level of comprehensiveness and accuracy with the data that are

currently available. At the moment, it appears the only validation source is an FDA report published after fiscal year

2010, “FDA Enforcement Statistics Summary, Fiscal Year 2010”.

The numbers included in this document are

consistent, although not identical, to those derived from the publicly available databases on the agency’s website.

For example, the Warning Letter download from the FDA website indicated that there were 661 Warning Letters sent

in fiscal year 2010, whereas the Enforcement Summary stated that 673 were sent; additionally, the online Debarment

List indicated that ten debarments occurred in fiscal year 2010, whereas the Enforcement Summary stated that there

were thirteen. It is possible that these discrepancies can be explained by some entries being manually processed,

and therefore not included in the digital databases, the timing of case filing, or issues involving privacy.

http://www.fda.gov/newsevents/newsroom/pressannouncements/2010/default.htm

http://www.fda.gov/downloads/ICECI/EnforcementActions/UCM247845.pdf

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Consumer Financial Protection Bureau

Annual Case Statistics, FY2013

Megan E. Collins, Miranda A. Galvin, Sally S. Simpson, Peter C. Yeager

The CFPB is an independent federal agency tasked with protecting consumers by supervising the implementation of

federal consumer financial laws. The Bureau was created under Title X of the 2010 Dodd-Frank Wall Street Reform and

Consumer Protection Act (the Dodd-Frank Act). Operations began in July of 2011.

The CFPB’s purpose is “to implement

and, where applicable, enforce Federal consumer financial law consistently for the purpose of ensuring that all consumers

have access to markets for consumer financial products and services and that markets for consumer financial products

and services are fair, transparent, and competitive.”

The agency’s core functions include: writing rules, supervising

companies, and enforcing federal consumer protection laws; restricting unfair, deceptive, or abusive practices; taking

consumer complaints; promoting financial education; researching consumer behavior; monitoring financial markets for

new consumer risks; and enforcing laws that outlaw discrimination and other unfair treatment.

The CFPB’s mission consists solely of consumer financial protection. This

positions its regulatory domain within the broader sphere of economic

regulation (see Figure 1). At its inception the agency assumed the

enforcement responsibilities of over one dozen consumer protection

laws

from seven different primary agencies.

This resulted in a

comprehensive—but not exhaustive

—consolidation of consumer

Note that this brief covers fiscal year 2013, while others in this series report on fiscal year 2010. Because CFPB was not formed

until 2010 and did not begin operating until July 2011, fiscal year 2013 was chosen as more representative of a “typical” year of

activities compared to the earlier start-up years.

The Dodd-Frank Act, Public Law 111-203, Section 1021.

12 U.S. Code § 5511 - Purpose, objectives, and functions.

http://www.consumerfinance.gov/the-bureau/.

The statutes transferred to the CFPB include, but are not limited to: the Alternative Mortgage Transaction Parity Act of 1982 (12

U.S.C. 3801 et seq.); the Equal Credit Opportunity Act (15 U.S.C. 1691 et seq.); portions of the Fair Credit Reporting Act (15 U.S.C. 1681

et seq.); the Home Mortgage Disclosure Act of 1975 (12 U.S.C. 2801 et seq.); the Home Ownership and Equity Protection Act of 1994

(15 U.S.C. 1601); the Real Estate Settlement Procedures Act of 1974 (12 U.S.C. 2601 et seq.); the Secure and Fair Enforcement for

Mortgage Licensing Act (12 U.S.C. 5101 et seq.); and the Truth in Lending Act (15 U.S.C. 1601 et seq.).

http://files.consumerfinance.gov/f/2012/01/Congressional_Report_Jan2012.pdf.

The statutes transferred to the CFPB include, but are not limited to: the Alternative Mortgage Transaction Parity Act of 1982 (12

U.S.C. 3801 et seq.); the Equal Credit Opportunity Act (15 U.S.C. 1691 et seq.); portions of the Fair Credit Reporting Act (15 U.S.C. 1681

et seq.); the Home Mortgage Disclosure Act of 1975 (12 U.S.C. 2801 et seq.); the Home Ownership and Equity Protection Act of 1994

(15 U.S.C. 1601); the Real Estate Settlement Procedures Act of 1974 (12 U.S.C. 2601 et seq.); the Secure and Fair Enforcement for

Mortgage Licensing Act (12 U.S.C. 5101 et seq.); and the Truth in Lending Act (15 U.S.C. 1601 et seq.).

http://files.consumerfinance.gov/f/2012/01/Congressional_Report_Jan2012.pdf.

While the CFPB was designed to serve as a primary agency for consumer protection, the agencies previously tasked with the

enforcement of consumer protection statutes still retain authority over some statutes. For example, the FTC still pursues fraudulent

advertising cases that mislead consumers; the Federal Deposit Insurance Corporation still retains consumer protection

Domain

Sphere

Economic

Regulation

Financial

Market

Competition

Consumer

Protection

Fig. 1: The CFPB’s Regulatory Domain

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

protection authority. The Dodd-Frank Act outlined specific objectives for the CFPB, seeking to ensure that consumers were

appropriately informed and protected from unfair financial practices, and to provide for market transparency and

accountability.

The CFPB focuses on the illegal behavior of companies and individuals within those companies acting for

company gain. While it is possible that individuals acting in organizational roles for personal gain are under the purview

of the CFPB, the laws for which the Bureau has enforcement activity make this type of behavior unlikely to appear in

agency cases. Dodd Frank also directs the agency to conduct education programs, act as a mediator and investigator of

consumer complaints, monitor markets and disseminate relevant information. Finally, the agency is granted rulemaking

authority, akin to that held by the Securities and Exchange Commission or Federal Trade Commission.

While not all of the agency’s directives are relevant to a white collar crime statistical series, several are. The investigation

of consumer complaints and market monitoring, for example, are important sources of discovery of potential violations.

The agency is also tasked with assuring compliance. Accordingly, only some of the data promulgated by the agency are

desirable for inclusion in the series. Consumer complaints, for example, are not necessarily indicative of a violation.

Administrative cases, however, include substantiated violations, and thus are a potentially rich source of data.

The CFPB has, thus far, acted in primarily administrative roles. The Consumer Complaint Database is an administrative

process, and the only other source of agency promulgated data relates to administrative hearings (though civil monetary

penalties can be meted out during these cases). The CFPB only recently referred its first criminal matter to the

Department of Justice

, illuminating the relationship between the two agencies. The CFPB also files civil claims, though

the agency itself releases little to no information about this process.

Publicly Available CFPB Data

The CFPB maintains a number of public and protected databases. Some data can be directly downloaded from the

Bureau’s website, but other datasets are restricted. Access to the restricted data either requires permission from the

Bureau or right of use through a Freedom of Information Act request.

The Consumer Complaint Database is a publicly available compilation of complaints the Bureau has received about

financial products and services.

Conduct alleged in complaints is not verified prior to being included in the database.

However, the Bureau does take some steps to authenticate the complaints.

For example, in order to be included in the

database the complaints cannot be duplicative of anther complaint to the Bureau from the same consumer, cannot be a

whistleblower complaint, must involve a consumer financial product or service within the scope of CFPB jurisdiction, and

must be submitted by a consumer or a representative of a consumer with an authenticated relationship with the

company.

The public database is live and updated each evening with anonymous complaint submission data. Populated

fields include: complaint ID, product, sub-product, issue, sub-issue, state, ZIP code, date received, date sent to company,

company, company response, whether the response was timely, and if the consumer disputed the response.

13,14

Criticisms of the Consumer Complaint Database entries have been raised by trade organizations, who object that it is

inappropriate to publish data that are not generated through randomly selected consumers, leading to potentially

responsibilities. Additionally, consumer protection in the social regulatory sphere (e.g. product and food safety) has not been

affected by the creation of the CFPB.

Public Law 111-203, Section 1021 (b)

Nate Raymond. 2013. “U.S. brings charges in first criminal case for consumer agency.” Reuters.

http://www.reuters.com/article/2013/05/07/us-cfpb-criminal-idUSBRE9460LO20130507

http://www.consumerfinance.gov/complaintdatabase/

Disclosure of Consumer Complaint Data [Docket No. CFPB-2012-0023] http://files.consumerfinance.gov/f/201303_cfpb_Final-

Policy-Statement-Disclosure-of-Consumer-Complaint-Data.pdf

Ibid.

http://files.consumerfinance.gov/f/201403_cfpb_consumer-response-annual-report-complaints.pdf

https://data.consumerfinance.gov/dataset/Consumer-Complaints/x94z-ydhh.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

unrepresentative findings (i.e., selection bias). However, the CFPB has received support from state agencies, asserting the

data do not need to be random to be informative, particularly for the purpose of decreasing risk, increasing customer

service, and identifying best practices.

The data do not distinguish complaints of major and minor significance, or distinguish between those based on confusion

about a requirement from those that are violations. Additionally the database is updated regularly and additional

complaints are added on a rolling basis. Therefore the number of complaints filed in fiscal year 2013 may appear to

increase over time as additional complaints are added to the database.

Another data source, which is available to law enforcement and requires permission for access, is the Repeat Offenders

Against Military (ROAM) Database.

This database is a joint effort with state Attorneys General and the Department of

Defense to track companies and individuals who repeatedly scam military personnel. The database, which was approved

in 2012 and has been developed gradually, synthesizes publicly available data from multiple sources including final

judgments, formal judicial and administrative findings, and other formal allegations, with the intention of supplementing

with private data as the series develops.

While other agencies, such as the Federal Trade Commission, also collect

information on consumer complaints from the military community (e.g., via Military Sentinel), ROAM differs in the kind of

data it collects. Specifically, unlike Military Sentinel ROAM reports on completed enforcement actions.

As part of its yearly reporting activities, the CFPB releases the Loan Application Register data (HMDA). These data,

compelled as part of the Home Mortgage Disclosure Act and collected beginning in 2007, are provided by banks and other

financial institutions. The data comprise approximately fifteen to twenty million mortgage records each year. Records

contain information regarding the property, its location, the loan, the loan applicant, and the lender, and can be sorted

by year or by specific filters (e.g., property type). The CFPB is tasked with conducting assessments to verify the accuracy

of the reported data.

While the HMDA data do not specifically highlight infractions or enforcement activity, they may

be useful for identifying patterns of discriminatory lending (i.e. violations).

In addition to formal databases, CFPB activity can be found on the “Administrative Adjudication” page of the Bureau

website.

This page includes administrative proceeding files (consent orders and stipulations) for CFPB cases. These

documents are provided in portable document format (PDF) for each individual case, and include information such as

business name, violations, fees, and provisions.

These administrative files are the most fruitful publicly available agency

source of Bureau Activity for a white-collar crime series but there are some limitations. For instance, while it appears that

all closed cases are made available through this portal, investigated administrative matters, or ongoing cases, are not

documented publicly. Further, these files do not include civil or criminal actions initiated by the Bureau.

Case Sources and Processing

Disclosure of Consumer Complaint Data [Docket No. CFPB-2012-0023] http://files.consumerfinance.gov/f/201303_cfpb_Final-

Policy-Statement-Disclosure-of-Consumer-Complaint-Data.pdf

For example, in October 2012, credit card complaints dating back to 12/1/11 were added; in March 2013 mortgage complaints

dating back to 12/1/11 and bank account and service complaints, private student loan complaints, and other consumer loan

complaints dating back to 3/1/12 were added; in May 2013 credit reporting complaints dating back to 10/22/12 and money transfer

complaints dating back to 4/4/13 were added; and in November 2013, debt collection complaints dating back to 7/10/13 were

added to the database. http://files.consumerfinance.gov/f/201403_cfpb_consumer-response-annual-report-complaints.pdf.

Data may be accessed by contacting ROAMDatabas[email protected]; http://www.consumerfinance.gov/blog/a-new-tool-for-

protecting-the-military-community/

http://files.consumerfinance.gov/f/2012/01/Project-ROAM-CFPB.pdf

http://files.consumerfinance.gov/f/201310_cfpb_hmda_resubmission-guidelines_fair-lending.pdf.

http://cfpb.github.io/api/hmda/.

http://www.consumerfinance.gov/administrativeadjudication/.

Ibid.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

The CFPB’s many data sources capture multiple stages of the enforcement process (see Figure 2). However, most capture

specific types of cases, rather than specific stages. In addition, each case type may originate from various sources.

However, publicly available data on case sources is quite limited, being largely confined to information on consumer

complaints.

The Bureau’s Consumer Response team receives complaints directly from consumers through its website, toll-free

telephone line, mail, email, facsimile, and referral from other agencies.

The Consumer Response team then screens all

complaints based on numerous criteria (e.g., whether the complaint falls within Bureau’s enforcement authority, is

complete, duplicates prior submissions by the same individual). These cases then appear in the consumer complaint

database, or, if further action is taken, in administrative files. Unfortunately, a case cannot be linked to a specific complaint

unless that complaint is mentioned within the agency drafted administrative consent order.

Because the agency’s enforcement purview is relatively narrow compared with other agencies (e.g., consumer financial

laws and products), the

kinds of complaints it

receives are concentrated

in certain areas. Most

common are complaints

regarding mortgages,

credit cards, bank

accounts or service, and

credit reporting (see Table

1). The types of credit

card complaints that are

commonly submitted by

consumers involve billing

disputes, identity theft,

fraud, embezzlement,

annual percentage or

interest rates, closing or

cancelling of accounts,

credit reporting, credit

card payment, debt

protection, collection

practices, late fees, and

other fees. The most common occasions of mortgage complaints include applying for loans (application, originator,

mortgage broker), receiving a credit offer (credit decision/underwriting), signing the agreement (settlement process and

costs), making payments (loan servicing, payments, escrow accounts), and when unable to pay (loan modification,

collection, foreclosure).

In fiscal year 2013, 99,421 complaints were filed with the CFPB and recorded in the Consumer Complaint Database. The

database also specified sub-products within the eight primary products (i.e., within the primary category of “Mortgage”

the sub product clarifies the specific type of mortgage; see Table 1). There was little substantive overlap in complaints

Consumer Complaint Database. https://data.consumerfinance.gov/dataset/Consumer-Complaints/x94z-ydhh.

http://files.consumerfinance.gov/f/2012/01/Congressional_Report_Jan2012.pdf.

Ibid.

Product

Complaints

Sub-Products

Bank Account or Service

13,121

(13.2%)

Certificate of deposit (673); Cashing a check without an

account (91); Checking account (10,153); Other financial

product/service (1,244); Savings account (960)

Consumer Loan

2,898

(2.9%)

Installment loan (366); Personal line of credit (265);

Vehicle lease (222); Vehicle loan (1,945)

Credit Card

13,590

(13.7%)

Unspecified (13,590)

Credit Reporting

12,384

(12.5%)

Unspecified (12,384)

Debt Collection

4,410

(4.4%)

Unspecified (860); Auto (149); Credit card (1,083);

Federal student loan (113); Medical (405); Mortgage

(222); Non-federal student loan (134); Other (phone,

health club, etc.) (1,444)

Money Transfers

295

(0.3%)

Domestic (US) money transfer (167); International

money transfer (128)

Mortgage

49,890

(50.2%)

Conventional adjustable mortgage (ARM) (5,102);

Conventional fixed mortgage (13,861); FHA mortgage

(4,190); Home equity loan or line of credit (1,859); Other

mortgage (23,528); Reverse mortgage (296); Second

mortgage (271); VA mortgage (783)

Student Loan

2,833

(2.8%)

Federal student loan (36); Non-federal student loan

(2,797)

Total

99,421

(100%)

Table 1: Consumer Complaint Products and Sub-products, FY 2013

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

across products, except for across Debt Collection and Student Loans. However, the CFPB ensures that despite the same

sub-products being referenced, these complaints are unique and not duplicate entries by the same complainant.

Complaints are cataloged only once, though the sub products of federal student loans and non-federal student loans are

specifications of both debt collection and student loan complaints. Sub products should simply be considered as

elaborations of primary product complaints.

The Bureau also has a supervision program that determines if providers are in compliance with federal consumer financial

laws, assesses whether companies appropriately monitor their own compliance, and identifies risks to consumers in

various markets. The Bureau’s activities include a large bank supervision program (focusing on compliance at banks,

thrifts, and credit unions with assets over ten billion dollars, their affiliates, and certain service providers), and supervision

of non-bank providers of consumer financial products and services (e.g. certain financial services companies without a

bank, thrift, or credit union charter, such as mortgage lenders and brokers, credit bureaus, and payday lenders).

On top of its role in mediating complaints,

the CFPB may initiate action against

companies. Investigations can be prompted

by Bureau staff or referred by other

regulators and the United States

Department of Housing and Urban

Development (HUD). Currently, the Bureau

oversees a joint task force on foreclosure

scams with the Special Inspector General

for the Troubled Asset Relief Program

(SIGTARP) and the U.S. Department of the Treasury. The task force was formed to target scams directed at homeowners

applying for Home Affordable Modification Program (HAMP), which is a foreclosure prevention program managed by

Treasury. The whistleblower hotline allows current or former employees, contractors, vendors, and competing companies

to submit information or tips, and provides as much confidentiality as is allowed by the law.

It should be noted that while there is a wide array of data available on the CFPB website, the applicability to constructing

a white collar enforcement database is limited. Table 2 catalogues the limited information on case initiation available on

the Bureau’s website. The Consumer Complaint Database identifies the quantity, nature, and outcome of complaints for

a given year. Unfortunately the same type of comprehensive database is not available for Bureau-led investigations or

referrals from other agencies, nor for tips and self-disclosures.

Case Processing. The Consumer Financial Protection Bureau handles two types of cases – those which are substantiated

violations, and consumer complaints, which may not be formal violations (opposed to undesirable interactions between

consumer and company). Subsequently, the type of case (i.e. the originating source of the case) determines the course of

case processing. When complaints have been entered into the system by consumers, the Office of Consumer Response

reviews complaints to determine if they fall under the agency’s enforcement authority and is authentic. A company then

has 15 calendar days in which to initially respond. Once a company has issued a response, the complainant is allowed to

provide feedback, at which point a complaint is considered “closed”. The Bureau then determines how to proceed by

prioritizing, reviewing and investigating complaints in cases where the company fails to provide a timely response (i.e.,

within fifteen days of the complaint) or if the complainant disagrees with the company’s response, while also considering

http://files.consumerfinance.gov/f/2012/01/Congressional_Report_Jan2012.pdf.

Ibid.

Case Sources

Data Availability

Cases

Tips

Complaints

Consumer Complaint Database

99,421

Referrals

Voluntary Self-

Disclosures

Investigations

Table 2: Case Initiation Data

NA: Not available

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

the nature of the complaint itself.

The Bureau’s Office of Consumer Response works to determine why a company may

have failed to provide a timely response, and/or whether the company’s response warrants additional review in

concordance with consumer financial protection laws. In doing so, Consumer Response may elicit additional information

from consumers and companies, or refer complaints to the Bureau’s Division of Supervision, Enforcement, and Fair

Lending & Equal Opportunity, the unit responsible for enforcement action.

Potential violations may also be found

through the supervision and monitoring activities of the Bureau.

In addition to complaint processing, the Bureau also pursues administrative enforcement actions both in response to and

independently of consumer complaints. The Office of Administrative Adjudication, an independent judicial office within

the Bureau, holds hearings to decide on formal charges and actions brought by the CFPB. Such charges are based on

evidence that federal laws under the purview of the Bureau have been violated.

Administrative cases are heard before

an Administrative Law Judge. Unfortunately, the CFPB only provides documentation on completed cases, precluding a

discussion of administrative matters opened in fiscal year 2013. It is possible to put together case processing data on these

completed cases, though such methods may return only a subset of cases processed. Cases that are terminated before a

consent order, for example, will not be in these files, nor will ongoing cases.

FY 2013

Data

Availability

Counts

Comments

Case Numbers

Complaint identification numbers are available. However, not every

complaint develops into a case or is even representative of a violation.

Docket numbers are available for concluded administrative cases

Date Case Initiated

Incomplete

Month and year of initiation are contained in concluded

administrative cases.

Num. of Violations

***

Statutes violated are listed in concluded administrative case files

Num. of Defendants

Types of Violation

Dates of Viols.

Incomplete

Month and year of violation are contained in administrative case

materials provided.

Defendant Chars.

Incomplete

Defendant name included in Consumer Complaint Database, and in

Administrative Adjudication PDFs

Case Seriousness

Administrative

Cases

Can access only case stipulation or consent orders, and only in PDF

format. It does not appear to be possible to determine how many

cases are initiated per year because the process data are available only

for closed cases.

Civil Referrals

***

Criminal Referrals

***

Table 3: Case Processing Data Available from the CFPB

Case Resolutions

Because companies are given only a 15-day window to respond to a complaint lodged by a consumer, 99.9 percent of

complaints received in fiscal year 2013 have been closed. Most consumer complaints (75.6 percent) were closed with an

Consumer Financial Protection Bureau. Consumer Response: A Snapshot of Complaints Received through

June 30, 2014. http://files.consumerfinance.gov/f/201407_cfpb_report_consumer-complaint-snapshot.pdf

http://files.consumerfinance.gov/f/201403_cfpb_consumer-response-annual-report-complaints.pdf.

Ibid.

http://www.consumerfinance.gov/administrativeadjudication/.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

explanation (see Table 4). Additionally, most consumers (78.3

percent) did not dispute the outcome of their complaint. As

stated earlier, the Bureau’s Office of Consumer Response is

tasked with identifying why in 75 cases companies appear to

have failed to provide a timely response, why consumers may

have disputed the outcome, and what next steps to take. Thus,

while the complaint may be marked closed, it is not necessarily

the conclusion of case activity by the Bureau. Often the closure

of a consumer complaint may initiate an investigation.

Seven administrative cases were adjudicated by the CFPB in fiscal year 2013 (as indicated by the dates the consent order

and stipulation documents were signed)

. The consent orders cover different kinds of conduct which have been resolved

by monetary, compliance, and administrative provisions that were agreed upon by the company and CFPB. Similarly, the

stipulation documents included a consent agreement and waivers, and were signed off by company directors and Cordray.

Based on the content of available PDF consent order and stipulation documents, these seven cases alleged twenty-five

violations by ten companies; multiple companies may be charged in the same administrative case. Violations of statutes

12 USC §§ 5531 and 5536 were listed for each case, which state the Bureau’s

authority in the adjudication (see Table 5).

While these administrative cases provide

richer case details than the consumer

complaint database (which also lacks the

ability to distinguish genuine violations

from unfounded complaints), the

administrative case files still do not

provide all of the information that is

desirable for inclusion in a white-collar

statistical series. As Table 6

demonstrates, defendant characteristics

are not provided within the agency files

(though supplementary sources may be

useful). Civil and criminal case files are

absent entirely from the Commission’s

datafiles; a search of their website

returns not even a discussion of civil or

criminal case processes. This is a

significant disadvantage of using Bureau

data without supplementary sources.

Ibid.

Consumer Financial Protection Bureau. Office of Administrative Adjudication.

http://www.consumerfinance.gov/administrativeadjudication/

Total

Closed with Unknown Response

% down

2,882

2.9

Closed with Explanation

% down

75,115

75.6

Closed with Monetary Relief

% down

9,020

9.1

Closed with Non-Monetary Relief

% down

12,329

12.4

Untimely Response

% down

0.1

Total

% down

99,421

100.0

Table 4: Outcomes of Consumer Complaints

Statute

Description

Count

12 CFR §

1002.6

ECOA Rules concerning evaluation of applications

12 CFR

Part 1026

CFPB Truth in Lending (Regulation Z)

12 USC

§§ 2601

et seq.

Real estate settlement procedures

12 USC §

5531

CFPB authority prohibiting unfair, deceptive, or abusive acts

or practices

12 USC §

5536

CFPB prohibited acts

15 USC §

45(a)(1)

FTC unfair methods of competition unlawful; prevention by

Commission

15 USC

§§ 1601

et seq.

Consumer credit cost disclosure

15 USC

§§ 1681

et seq.

Credit reporting agencies

15 USC

§§ 1691

et seq.

Equal credit opportunity

Table 5: Administrative Adjudication Statutes in Cases Closed in FY2013

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Other Data Sources

In addition, civil and criminal

case data can be found in the

federal website Public Access

to Court Electronic Records

(PACER). For a nominal fee

this site allows users to search

for cases in criminal, appeals,

bankruptcy, and civil court

files by a number of

parameters. A search of

PACER court filings for

“Consumer Financial

Protection Bureau” and

“CFPB” returned eleven civil

cases filed in fiscal year 2013

in which the Bureau is

bringing affirmative action

(i.e., is listed either as a

plaintiff or petitioner).

However, PACER records have

been shown to be less

complete than agency records

in previous work (see data

briefs on SEC and FTC, this

series). Without agency data,

it is not possible to determine

the quality of PACER’s data in

relation to the CFPB.

Data from the Federal Justice

Statistics Program may also

be able to offer new

dimensions of the CFPB’s

involvement in criminal and

civil cases. However, the

series has not yet released FY

2013 data, precluding an

evaluation of the CFPB’s

involvement via this source. In

the future, data collected Executive Office of the US Attorney

may prove useful, as they have with SEC, FTC, FDA and

EPA.

The FJSP data are available through ICPSR with an approved data protection plan. In order to isolate criminal matters related to

agency activity, matters can be filtered by referring agency if that information was provided. The U.S. Attorney’s case management

system, LIONS (from which the FJSP data are drawn) does not strictly require that a value be entered for the referring agency

variable. Thus, a certain amount of missing data can may be expected.

FY 2013

Data

Availability

Counts

Comments

Case Numbers

Inc.

Complaint

identification number

in Consumer

Complaint Database;

file number included

on consent order and

stipulation documents

Date(s) of Resolution

Inc.

Not included in

Consumer Complaint

Database; date consent

order and stipulation

documents signed is

noted

Defendant Characteristics

Inc.

Company name

included in Consumer

Complaint Database

and on administrative

documents

Administrative Cases Resolved

No. of cases dismissed

No. of Defendants

No. of violations penalized

Inc.

***

Docket information

doesn’t clearly link

penalty to specific

violation

Number of sanctions imposed

***

Civil Cases Resolved

No. of cases dismissed

No. of Defendants

***

Number of violations

penalized

Number of sanctions

imposed

Criminal Cases Resolved

Cases Declined

No. of Defendants

No. of violations penalized

No. of sanctions imposed

Appeals

Table 6: Case Resolution Data

NA: Not Available NR: Not Relevant

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.

Summary: Strengths and Weaknesses of Available Data

The CFPB is quite different from other agencies profiled in this series. Because The CFPB primarily serves to oversee

aspects of consumer protection and facilitate exchanges between consumers and companies, the publicly available data

have limited value for a white-collar crime data series. The Consumer Compliance Database contains a high volume of

complaints with details about the complaint and outcome. However, it is unlikely that many of these incidents fall

within the definition of white collar offending utilized in this series, and the details included in the database are

insufficient for identifying relevant cases. Case data may be used by researchers to assess trends in offending, based on

reporting patterns, but such examinations are outside the purview of this series.

The most useful information likely comes from the administrative docket documents, which are currently only available

in PDF format. These are useful in the rich detail that they provide, but limited in their utility. These documents would

likely require more than data scraping software to fully comprehend, given the use of legal and legislative jargon in the

text. Unfortunately, the CFPB does not provide civil or criminal case files in any form, unlike the comprehensive files

provided by other agencies (e.g. SEC, FTC, EPA).

Supplementing the CFPB’s agency managed data may be possible through either PACER or FJSP records. PACER contains

civil cases filed by the Bureau that would otherwise be uncounted in the statistical series of white-collar crime and can

be used both to compile a list of cases as well as mine court documents for defendant and offense characteristics.

Unfortunately, it is not yet possible to assess the quality of these data relevant to the CFPB, as researchers cannot

compare case counts across sources.

This document is a research report submitted to the U.S. Department of Justice. This report has not

been published by the Department. Opinions or points of view expressed are those of the author(s)

and do not necessarily reflect the official position or policies of the U.S. Department of Justice.