the challenge of empirical research on business compliance in regulatory capitalism

ANRV392-LS05-03 ARI 24 September 2009 18:6

The Challenge ofEmpirical Research onBusiness Compliance inRegulatory CapitalismChristine Parker1 and Vibeke Nielsen2

1Melbourne Law School, University of Melbourne, Victoria 3010, Australia;email: [email protected] of Political Science, University of Aarhus, DK-8000 Aarhus, Denmark;email: [email protected]

Annu. Rev. Law Soc. Sci. 2009. 5:45–70

First published online as a Review in Advance onAugust 4, 2009

The Annual Review of Law and Social Science isonline at lawsocsci.annualreviews.org

This article’s doi:10.1146/annurev.lawsocsci.093008.131555

Copyright c© 2009 by Annual Reviews.All rights reserved

1550-3585/09/1201-0045$20.00

Key Words

business regulation, social science methodology

AbstractRegulatory capitalism—a social, political, and economic order charac-terized by a proliferation of both markets and state and nonstate at-tempts to regulate markets and business conduct—creates the opportu-nity for theoretically and politically significant research on compliance.The plural and decentered nature of regulation, and therefore of com-pliance, in regulatory capitalism creates significant complexity and dif-ficulty for social scientists in the conceptual definition and operational-ization of regulatory compliance, however. We survey the different waysin which empirical researchers have studied business compliance withregulation, and their strengths and weaknesses. In doing so, we reviewand interrogate the literature on regulatory compliance to understandwhat it is that researchers study when we study business compliancewith regulation, and what we might have been missing or assuming.

45

Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


INTRODUCTION

Many of our most significant global policy chal-lenges require either radical changes in indus-try behavior or large-scale government action:to arrest global warming (Haines & Reichman2008), to keep people’s livelihoods safe fromfinancial meltdown (Partnoy 2003), and toachieve justice and social inclusion for thoseproducing goods and services in the GlobalSouth (Hutchens 2009). Similar challenges facegovernments and businesses at the local andnational levels, too: environmental degradationand overconsumption, health and safety of workpractices and of food and beverage consump-tion, the security of employment and finan-cial investments, and the capacity to deliver arange of essential services (including power, wa-ter, housing, transport, and communication) ina just and inclusive way.

Yet contemporary Western society trustsneither government nor capital to promote andachieve social goals (Putnam 1993, Fukuyama1995). Instead, Western polities and, increas-ingly, global ones are now built on the hopethat we can govern big business to meet socialgoals as we redesign public governments to fa-cilitate and work with private and civil societygovernance. Ours is an age of regulatory cap-italism. The distribution of power at both na-tional and global levels is shaped by interactionsbetween state-based regulation, business activ-ity, and civil society (Levi-Faur 2008, p. viii; alsoBraithwaite 2008, pp. 1–31; Levi-Faur 2005;Schneiberg & Bartley 2008; Vogel 1996). Theexpanding part of public government is regula-tion, not the direct provision of public and pri-vate services. Capital is increasingly expected tobe responsive to social goals and regulation.

Regulatory capitalism creates a demand forsocial science research on business regulation:to map out the genesis and implementation ofvarious attempts at regulation, uncover whatovert and covert purposes they serve, and inter-pret and explain what intended and unintendedimpacts they have ( Jordana & Levi-Faur 2004a,McCahery et al. 1993). In this review, we areconcerned with the last question—interpreting

and explaining the impacts of regulatory capi-talism on business behavior.

Schneiberg & Bartley (2008, p. 33) havecommented in a previous review that re-searchers “have amassed evidence for therecent proliferation of regulation and newforms at the national and transnational levels,”but researchers “have barely begun to analyzesystematically. . .the extent to which new formsactually reshape markets and organizationalbehavior on the ground.” It is commonsense tosuppose that business regulation “has markedlyimproved the safety of banking, dairy products,electrical systems in housing, pharmaceuticalsand motor vehicles” and that it has drasticallyreduced death rates at work, pollution, and in-equality of employment opportunity, and thereis much evidence that this is so (Coglianese& Kagan 2007, p. xx; see also Braithwaite2008, p. 14). Certainly, any form of businessregulation can be vulnerable to empty ritualism( J. Braithwaite et al. 2007). But many scholarssee in regulatory capitalism an overridingopportunity to harness the power of markets toachieve social goods in efficient, effective, anddemocratically legitimate ways (Braithwaite2008, Gunningham et al. 2003, Sabel 2007).Others, however, are skeptical about theultimate politics and efficacy of regulatorycapitalism: Regulatory activity might create asmokescreen for ever more self-aggrandizingcorporate power. Regulatory capitalism rep-resents the economization of the social—thedomination of economic over social values—rather than the socialization of capital (Shamir2008; see also Garsten & Jacobsson 2007).

Either way, it is clear that empirical researchinto the nature, workings, and impacts of reg-ulatory capitalism is necessary to understandand explain contemporary social, political, andeconomic power relations. Empirically under-standing and explaining the impact of differ-ent regulatory developments in an independentand rigorous way is also an important aspect ofpolicy evaluation and accountability. We cannotevaluate the true “political colors” ( Jordana &Levi-Faur 2004a, pp. 10–11) of regulatory capi-talism without empirical evidence of its impacts.

46 Parker · Nielsen

Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


In the first section of this review, webriefly outline two key features of regulatorycapitalism—the “responsibilization” of busi-ness and the plural, decentered, and networkednature of regulation. We go on to argue thatthese developments create a practical and the-oretical demand for at least two types of socialscience research on regulatory compliance: re-search aimed at understanding how people con-ceptualize and socially construct compliance,and research that seeks to explain the causes andeffects of compliance. In the second and thirdsections, we consider the challenges and op-portunities for each of these types of empiricalcompliance research in regulatory capitalism.

We argue that much empirical research onbusiness compliance has been concerned withthe considerable challenge of interpretive un-derstanding of compliance. Much very fruitfulresearch has focused on conceptualizing com-pliance from the viewpoints of different actorsand the social processes of construction of com-pliance in which they together engage. We goon to argue, however, that the operationaliza-tion of compliance for the purposes of causal ex-planation and prediction is also crucial for theo-retical development of the social sciences as wellas policy accountability. We argue that pluraland decentered understandings of compliancein regulatory capitalism—often pointed to bysocial constructivist research—lead to partic-ular difficulties for researchers doing explana-tory research. These difficulties put a particulardemand on researchers to legitimize and makeexplicit their own operationalization of compli-ance. They also make it even more relevant—but also more difficult—for the researcher todeal with classic methodological problems likebiased data sources and shaky information.

REGULATORY CAPITALISMAND THE NEED TOUNDERSTAND AND EXPLAINREGULATORY COMPLIANCE

Regulatory Capitalism

The regulatory state was invented in the Pro-gressive and New Deal eras in the United States

in response to public social concerns aboutthe rise of big business (Lobel 2004). Late-nineteenth- and early-twentieth-century Euro-pean nations, on the other hand, respondedto the same concerns with nationalization andstate provision of goods and services (includingwelfare, health care, public transport, utilities,industrial infrastructure, steel, coal, banking,airlines, and so on). In the 1970s and 1980s,the European provider state gave way to theregulatory state under the neoliberal agenda ofprivatization, deregulation, and nurturing ofmarkets (see Braithwaite 2008, pp. 4–29;Levi-Faur 2005).

Although neoliberalism preached deregu-lation, in fact it expanded and extended reg-ulation first in Europe and then globally( Jordana & Levi-Faur 2004a, p. 7; Vogel 1996).For example, Levi-Faur (2005) has analyzed thedecision to privatize telecommunications andelectricity and the decision to create regula-tory agencies in 171 countries, showing the “in-timate association” between privatization andthe creation of independent regulatory agenciesand the exponential increase in both since the1980s. Similarly, Jordana & Levi-Faur (2004b;Levi-Faur 2008, p. vii) studied regulation in 16sectors across 49 different countries from the1920s to 2002 and found that the number ofregulatory agencies rose sharply in the 1990s,with an autonomous regulatory agency in about60% of the possible sector niches by the end of2002. The diffusion of regulation is not con-sistent: Gilardi et al. (2006) find that economicregulatory reforms are more widespread thansocial ones and that social regulation is morewidespread in Europe than elsewhere. Yet it isa global phenomenon.

The neoliberal agenda of freer markets and,ultimately, more rules or good governance wasinternationalized and projected globally to Asiaand the developing world via bilateral trade ne-gotiations, the International Monetary Fund,and the United Nations Conference on Tradeand Development in the 1980s and 1990s(Braithwaite & Drahos 2000). It is likely tobe further developed through responses to theglobal financial crisis of 2008. Government and

www.annualreviews.org • Empirical Research on Compliance 47

Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


civil society are responding by increasing theirregulation of financial services with the inten-tion of rehabilitating the capacity of regulatedcorporate capital to provide financial services ina way that adequately protects investors. Evengovernments that are taking an ownership stakein financial institutions are doing so to regulatecorporate capital to function better, not to pro-vide financial services (MacNeil 2008).

The national regulatory state is itself nowgiving way to a situation of global and nationalregulatory governance (Lobel 2004). Nonstate-based regulation, including voluntary indus-try and internal corporate systems of self-regulation and nongovernmental organization(NGO) certification and management systems,is growing at the national and international lev-els (Bernstein & Cashore 2007, Braithwaite &Drahos 2000, King & Lenox 2000, Lehmkuhl2008, Rees 1997, McBarnet et al. 2007, Wolf2008).

Features of Regulatory Capitalism:Responsibilization and NetworkedGovernance

Two distinguishing characteristics of regula-tory capitalism include the responsibilization ofbusiness to self-regulate to achieve policy goalsand values, not just financial profits (Shamir2008, pp. 379–83), and the decentering of reg-ulation away from the state to plural networksof regulation (Black 2001, Parker et al. 2004,Solomon 2008).

These two characteristics are inherent inthe very nature of regulatory capitalism, whichinvolves power sharing between state, business,and civil society. A tendency toward these twocharacteristics has therefore been evident sincethe United States’s invention of the regulatorystate in the late nineteenth century. But neolib-eralism and the challenge of global regulatorygovernance have made responsibilization andnetworked governance more prominent. Sincethe rise of neoliberalism in the 1970s and 1980s,there has also been a much greater demand forstate-based regulation to be less centered onhierarchical rule-based command-and-control

strategies (Lobel 2004). The challenges oftransnational and global governance have madeactivists, academics, and policy makers lookfor less hierarchical governance possibilities atthe international level as well (Braithwaite &Drahos 2000, De Burca & Scott 2006, Teubner1997).

First, regulatory capitalism privileges busi-ness rather than state provision of goods andservices with a concomitant emphasis on busi-ness responsibility for how goods and servicesare provided (Braithwaite 2008). State provi-sion of goods and services is limited, but there isa profusion of regulation that attempts to makebusinesses responsible to state and civil soci-ety regarding the values and goals that shouldbe observed in the provision of goods and ser-vices. For example, although government maynot provide transportation, housing, or con-sumer goods, regulation by government and bysome international and industry organizationsspecifies that provision must meet standards forenvironmental, health and safety, competition,consumer protection, fair trade, and social in-clusion, among other policy goals and values(Shamir 2008).

The fact that businesses are responsibilizedto a large degree by the policy goals and valuesset by regulation means that compliance is anexceedingly broad concept. Compliance is oftennot a unitary response by a regulatee to an of-ficial rule promulgated and enforced by officialagencies. Rather it is the result of a process ofconstruction and negotiation of compliance be-tween official regulator and regulatee (Edelmanet al. 1991, McBarnet 2003, Reichman 1992).Moreover, it is not just official state regulatorswho negotiate compliance with business, but arange of civil society, professional, industry, andNGO regulators. The power of these third par-ties to regulate organizations often depends toa greater or lesser extent on voluntary consentto being regulated (Meidinger 2008).

Second, in regulatory capitalism, regula-tion is plural, decentered, and networked. Stateand nonstate regulators proliferate at both thenational and transnational levels. Empiricalscholarship on regulatory compliance has long


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


pointed out that various stakeholders repre-senting social and economic pressures support,counteract, or compete with pressures to com-ply with state-based regulation (Gunninghamet al. 2003, pp. 35–40; Scott 2001). Increas-ingly, industry, civil society, and NGO associ-ations are explicitly taking on regulatory rolesto require standards of behavior that go beyondcompliance with the law or that fill gaps in le-gal regulation. For example, there are multipleenvironmental certification schemes that busi-nesses can opt in or out of, and state-based reg-ulators recognize and reward or enforce par-ticipation in these schemes to various degrees(Coglianese & Nash 2006, Prakash & Potoski2006, Meidinger 2008). State-based regulatorsoften work purposely with civil society and in-dustry actors to create collaborative regulatoryand self-regulatory schemes. And all these regu-lators compete with and regulate one another atboth the national and transnational levels. Thisresults in plural, networked regulatory regimesin which public and private regulation are in-tertwined (Parker et al. 2004, Schepel 2005,Shearing & Wood 2003). Scholars now oftentalk about regulatory governance rather thanregulation to indicate that hierarchical, one-way regulation of businesses by official stateagencies is no longer typical (De Burca & Scott2006, Solomon 2008).

Consider, for example, responsibilization ofbusiness and decentering of regulation in fi-nancial services (Black 2003, Krawiec 2008).There are global prudential standards (transna-tional regulation) that require state-based regu-lators to give financial service institutions bothassigned responsibilities and degrees of free-dom to define appropriate risk management forthemselves (national regulation that responsibi-lizes business). To a large extent, national reg-ulation and internal corporate self-regulationalso depend on standards of behavior definedby diverse stakeholders and on nonstate-basedregulators (such as securities exchanges, institu-tional investors, and auditors) following publicaccounting standards and quality standards forinternal compliance and risk management sys-tems (decentering of regulation).

The Demand for Social Science toUnderstand and Explain Compliancein Regulatory Capitalism

This proliferation of regulation—in termsof its sources (decentering) and values(responsibilization)—puts demands on socialscience to understand, explain, and predicthow and why those who are the objects ofregulation respond to it and what effects it has.

There is always a policy pull from leg-islators, regulators, businesses, and citizenstoward empirical research that is certain,authoritative, and programmatic (Sarat &Silbey 1988, p. 131) in measuring the impact ofregulation: Does it make any difference to thebehavior of those it seeks to regulate? Whichregulatory enforcement tools, interventions, orstrategies work most effectively and efficiently?In what circumstances and for what reasons arebusinesses most likely to comply with differenttypes of regulation? And, when businessesdo comply, does their compliance behaviorachieve the public policy goals that motivatedthe regulation in the first place?

These traditional policy questions forresearchers have become more complicated inregulatory capitalism. The responsibilizationof business and decentering of regulation meanthere are multiple parties seeking to hold eachother accountable by revealing what each isdoing to comply and whether their compliancebehaviors actually achieve policy goals. To theextent that government regulators use lighter-touch regulation, they depend more than everon external research to find out whether it ishaving any impact on market behavior becausegovernment’s own regulatory enforcementagents are not in the field as much monitoringcompliance. The more optimistic theoriesabout the democratic potential of regulatorycapitalism see this kind of research and eval-uation as fundamental to new forms of policydevelopment and implementation. The “demo-cratic experimentalism” (Dorf & Sabel 1998)and “evidence-based” regulation (Braithwaite2000, Bennear & Coglianese 2005) movementspropose that communities should experiment


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


with regulatory strategies to solve policyproblems, collect evidence to evaluate theirsuccess, and learn from the results for futurepolicy development and implementation.

Understanding and explaining regulatorycompliance is also of critical theoretical signif-icance to the social sciences. The relations be-tween corporate power, state power, and civilsociety are always fundamental issues requir-ing explanation, especially as situations changeand previous explanations may be tested or re-vised. Regulatory capitalism, understood as reg-ulatory governance of business, is a particularinstitutionalization of the relationships amongcorporate power, state power, and civil soci-ety. Research that uncovers whether and howthe regulation of corporate capitalism works, aswell as the power relations, values, and goalsrepresented in the way that compliance is con-structed, should be a core concern of social sci-ence theory building. Research that uncovers,evaluates, explains, and critiques the workingsof regulatory capitalism is therefore importantfor pragmatic, policy-oriented reasons, and alsofor more fundamental theory-building reasons.

Empirical Compliance Research

The goal of the social sciences—and there-fore also of studies of regulatory compliance bybusiness—is for researchers to understand var-ious people’s conceptualizations of these phe-nomena and the social structures and processesby which they arrive at these conceptualiza-tions; that is, it is to produce an accumulat-ing body of valid knowledge that enables usbroadly to understand and explain an empiricalphenomenon that interests us. Looking at com-pliance, this leads to two different—but equallyimportant—research projects.

The first is research that treats the under-standing and conceptualization of complianceas endogenous to the project. The research ob-ject in this kind of study is how compliance isunderstood and conceptualized. Some of thesecompliance-endogenous research projects havean interpretative ambition because their mainfocus is to discover the plural meanings of

compliance that exist among key actors in theregulatory field. Others have a more construc-tivist research ambition, given that their focusis both on uncovering the network of socialconstruction processes that create understand-ings of compliance and on the power relationsbetween the actors involved. These power re-lations might result in one understanding be-ing socially accepted as more legitimate thanothers. The challenge in this kind of researchis to achieve a balanced, not biased and non-normative discovery of understandings and so-cial construction processes.

The second kind of research project treatsthe understanding of compliance as exoge-nous to the research. How compliance isunderstood is not an object for researchin such projects. The researcher predefinescompliance—hopefully in line with generallyaccepted definitions or understandings. There-fore, the understanding of compliance is nottreated as a dynamic and changeable concept.Compliance in this type of research is a fixedvariable. It may be narrowly or broadly defined,but it must be predefined and then used as eithera dependent or independent variable in discov-ering causal relations between compliance andother variables. The ambition in this kind of re-search is to explain what causes compliance (aspredefined), or what effect compliance (as pre-defined) has. The main challenge to this sec-ond type of research is the operationalizationof compliance. How well does the researcheractually measure the definition of complianceshe or he claims to measure?

Even though regulatory compliance may of-ten be constructed in complex social processesthrough networks of regulation and negotia-tion, this does not mean that researchers cannotalso seek to explain and predict compliance be-havior and its effects. It does mean, however,that it is necessary to discuss and argue very ex-plicitly for one’s choice of theoretical and opera-tional definitions of compliance because there isnot one thing called compliance that everybodycan agree on. The policy demand to developquantitative measures of levels of compliancethat can be used to evaluate the effectiveness


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


Research treating understanding

of compliance endogenously

Theories of differences in

understanding of compliance or

about social mechanisms in the

construction process

Data collection and coding of

understandings of and/or social

construction processes of

understandings of compliance

Hypotheses

Test of hypotheses of the

mechanisms of the social

construction process and/or of

the existence of plural

understandings

Research treating understanding

of compliance exogenously

Theories of causation

between compliance

and other variables

Theoretical definition

of compliance:

How the researcher defines the

concept in words

Hypotheses

Operational definition

of compliance:

Classification rules as to what

data the researcher sees as

relevant to the concept’s

definition

Data collection and coding:

Gathering and using data for a

measure following the rules of

the operational definition

Testing hypotheses of causation

Figure 1Two different research projects.

of a particular regulatory intervention and spitout a policy prescription often underestimatesthe debatability of the used measures and for-gets that simple measures are not beautiful ifthey do not reflect legitimate understandingsof compliance.

Examples of the two different kinds ofresearch are illustrated in Figure 1. Bothillustrated research projects are examples ofinductive ways of doing research because thedifference between compliance-endogenous

versus compliance-exogenous research is onlya matter of whether the understanding of com-pliance is the very research object or not.It is not necessarily a question of deduc-tive versus inductive research methods. Like-wise the choice between endogenous and ex-ogenous research projects does not in itselfpredetermine whether one uses quantitativeor qualitative data collection methods, eventhough compliance-endogenous researchers—for good reasons—generally have a preference


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


for qualitative research methods because thesemethods are especially good at collecting infor-mation about meanings and understandings inan open-minded way.

Furthermore, even though the two researchprojects illustrated in the figure are both set outas involving testing of hypotheses, not every en-dogenous or exogenous research project needsto do that. For example, many studies measurethe level of compliance on one or a broad varietyof exogenously defined measures of compliancewithout testing theory-driven hypotheses aboutthe causes or effects of the variation in compli-ance level. Many official reports, statistics, andmapping exercises do this.

Any type of social research also faceschallenges at the data collection stage. Thebiased data source problem—in relation tocompliance—refers to the fact that each datasource is conceptually filtered or biased ac-cording to the understanding of what shouldcount as important for compliance by the in-dividual or agency creating the data. Regula-tory agencies, industry or professional associ-ations, businesses, and other researchers eachmight filter data according to their own con-ceptualizations of what should count as com-pliance. The problem of biased data sources isrelevant to compliance-endogenous research inthat the researcher uses a limited number ofinformation sources, which might result in lim-ited understandings of the meanings of compli-ance and their construction processes. On theother hand, biased data sources are only rele-vant to the compliance-exogenous researcher ifthe bias in the accessible sources does not fit theresearcher’s definition of compliance. For thecompliance-exogenous researcher, biased datasources might mean that he or she ends up mea-suring something other than what he or she pre-defined to be compliance.

The shaky information problem refers tothe fact that there are always communicativeproblems with collecting accurate data fromany source, including that (a) informants donot always understand the question asked andthe information required in the way expectedby the researcher; (b) sometimes informants

lie (whether consciously or subconsciously);and (c) sometimes informants do not knowthe correct answer for practical reasons (evenwhen they understand the question and haveno incentives to lie). Shaky information is agreat challenge in compliance-exogenous re-search projects, as explained further below. Incompliance-endogenous research, the respon-dents’ “understanding of the concept” is the re-search object; hence, the reliability and validityof this research object cannot and shall not beevaluated against an external/predefined defini-tion or standard of compliance. It simply has avalue in itself.

The following two sections respectivelydiscuss the two types of research—discoveringunderstandings and social constructions ofcompliance first, and then research operational-izing and explaining compliance.

DISCOVERINGUNDERSTANDINGS ANDSOCIAL CONSTRUCTIONSOF COMPLIANCE INREGULATORY CAPITALISM

Indeterminate Rules and RegulatoryDiscretion in Regulatory Capitalism

Critical criminology has long recognized thateven apparently clear criminal laws are not nec-essarily clear. Police and other regulators exer-cise a great deal of discretion that influenceswhat gets counted as a crime through their de-cisions about which crimes to record, investi-gate, and prosecute ( Jupp 1989, pp. 90–103;Edwards 2006; Mosher et al. 2002). Regula-tory capitalism amplifies an imprecise approachto sanctioning business misconduct. State leg-islation on corporate crime and misconduct isoften imprecise. Many rules created in regu-latory capitalism are expressed as broad, po-tentially indeterminate principles, rather thanas narrower, more easily determinable prohibi-tions. Many statements of rules explicitly andintentionally leave regulators the tasks of ap-plying and enforcing the rules or at least work-ing out guidance for the regulatees about what


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


is required. Even where this is not intentionaland explicit, the haste with which regulationproliferates makes it more likely that new rulesare poorly drafted. Thus, sorting out what theymean is left de facto to regulators’ discretion.Limited political support for state-based busi-ness regulation means that in many cases reg-ulation may be seen as being promulgated toaddress a problem, but because of fear of busi-ness backlash, prohibitions are not clearly de-fined, authoritatively prohibited, or criminal-ized (Snider 1991). Indeed, much regulation ofcorporate misconduct is done via noncriminalcivil and regulatory offenses or delegated to civilsociety or business and industry themselves.

Regulatory capitalism prefers to be flexibleand vague about what business activity is al-lowed in which circumstances and what busi-ness conduct will be prohibited or limited. Thisis evident at the international level, too, wherethe push for international criminal law andcourts for individuals is not matched by equallycoercive attempts at corporate regulation, butby voluntary, soft regulation by civil society andindustry (McBarnet et al. 2007, Ruggie 2002).The diverse sources of rules (decentered reg-ulation) increase the complication of impre-cise standards, rules, and expectations. Diverseenforcement agents geometrically increase thecomplexity of regulatory capitalism.

Much empirical research is concernedwith understanding what different actors—regulatory agencies (state or nonstate),regulated businesses, and a range of otherstakeholders—mean by compliance and howtheir attitudes and activities converge, diverge,and interact. This research is often particularlyconcerned with what emerges as sociallyaccepted meanings of compliance or what re-mains contested and why (Edelman et al. 1991,pp. 74, 92; Fairman & Yapp 2005; Lange 1999;McCaffrey et al. 2006). Many empirical studiesuse qualitative interviews and participantobservation to investigate how different orga-nizations and individuals within them respondto legal and nonlegal regulation, includingwhat motivations, attitudes, and perceptionsregulatees have toward regulation, regulators,

and compliance; what actions they take inresponse to regulation; and to what extent theytake responsibility for their own regulationand compliance (e.g., Geis 1967; Genn 1993;Haines 1997, 1999; Heimer 1996; Hutter2001; McCaffrey & Hart 1998; Rees 1988,1994; Silbey et al. 2009).

For example, Ericson & Doyle (2006) pointout that rarely do either government or indus-try regulators discover, prosecute, and officiallyrecord criminal fraud by insurance agents. Yettheir own research via interviews and ethnog-raphy with people inside the insurance agencyidentified five sources of moral risk that are “in-stitutionalized in the structure and culture oflife insurance sales” that give rise to “institu-tionally endorsed manipulation, deception andsometimes fraud” that could be prosecuted ascriminal fraud (Ericson & Doyle 2006, p. 994).These activities are euphemistically describedas “misselling” within the life insurance in-dustry itself (Ericson & Doyle 2006, p. 994).This misselling certainly violates industry rules,guidelines, and codes of ethics, yet it rarely evenbecomes an issue whether it constitutes a breachof criminal law.

Such studies also look at regulators’ un-derstandings of, and influence on, meaningsof compliance. Many of the most significantstudies in empirical regulation research havebeen based on surveys, interviews, or partici-pant observation study of regulatory staff andtheir behaviors, what they understand to countas compliance in different circumstances, howthey choose to use their discretion in regu-lation and enforcement, what attitudes theyhold toward different businesses in which cir-cumstances, and how these affect business atti-tudes or behaviors (e.g., de Bruijn et al. 2007;Grabosky & Braithwaite 1986; Hawkins 1989,2002; Hawkins & Hutter 1993; Gilad 2008;May & Wood 2003; May & Winter 2000).

These studies of regulatory enforcementagents’ behavior and enforcement style of-ten reveal hidden biases in official statisticsand statements about compliance. Studies haveshown that official enforcement decisions aboutcompliance or noncompliance are often the


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


result of regulatory agents’ less tangible, sub-jective appraisals of compliance, such as a firm’soverall standard of compliance and its will toimprove. Hawkins’s (2002) study of prosecutiondecision making in the UK Health and SafetyExecutive uncovered the significance of theway individual decision makers framed enforce-ment discretion—especially their evaluations ofmoral (not just legal) guilt, despite the factthat moral judgments are not legally relevantfor prosecution of health and safety offenses inthe UK. As Hawkins had found in an earlierstudy, many regulatory inspectors themselvesjudge compliance quite subjectively: Their as-sessments of compliance are “fluid and abstract,rather than concrete and unproblematic. ‘Atti-tudes’ are judged [by enforcement agents] asmuch as activities” (Hawkins 1989, p. 109).

Similarly, in Nielsen’s (2002) study of fireprecaution, environmental, and health andsafety regulators in Denmark, she asked reg-ulatory inspectors to rank the willingness tocomply of each regulatee and then comparedthe results with their average annual numberand gravity of offenses (as rated by indepen-dent experts on the basis of official recordsof inspections conducted by the same inspec-tors). She found little correlation between thetwo. Inspectors’ rating of willingness to complyseemed to be based on a sense of how thingsare and not on whether various breaches hadoccurred in the past.

Regulators also frequently leave businessesand other organizations alone to construct theirown meaning of compliance. This might oc-cur because the regulator uses enforced self-regulation, designates some businesses as lowrisk, or simply never inspects many businesses.In these circumstances, the regulator never en-forces his or her own interpretation of com-pliance on the regulated business: What theregulatee does to comply with the law is themain influence on what compliance means inpractice. It also influences what regulators andcourts will be able to count as compliance lateron (Reichman 1992; also McBarnet 2003).

It is not just regulators and regulateeswho have a say in negotiating the meaning of

compliance in regulatory capitalism. There area range of other stakeholders affected by busi-ness behavior (employees, customers, share-holders, local communities) with varying in-terests in what regulatory compliance means ineach situation. These stakeholders have variouscapacities to influence both regulators and busi-nesses in their understandings of what it shouldmean to comply with regulation in various sit-uations (Black 2003; Gunningham et al. 2003,pp. 35–40; Hutter & Jones 2007). In regulatorycapitalism, there are especially likely to beplural meanings of compliance where there arevoluntary schemes, networks of regulation, orreliance on third parties to certify or enforcecompliance with legal or voluntary regulation(e.g., Meidinger 2008, Prakash & Potoski2006).

Limits of Compliance-EndogenousResearch

Description and explanation of processes ofsocial construction of compliance should notreplace description and explanation of compli-ance behaviors. Studies of processes of com-pliance construction focus on the definitionsof compliance that emerge from existing so-cial structures. Certain meanings of compliancecan become socially accepted even though theywere self-servingly put forth by interested par-ties, or even though they are quite differentfrom what was originally envisaged by thosewho made the rule. Studies of social processesof construction of compliance can show howexisting power relations influence socially ac-cepted meanings of compliance and thus canaid in critique of current practice. But they donot articulate an alternative standpoint againstwhich actual business behavior and its outcomescan be evaluated and critiqued. The study andcritique of regulatory capitalism needs researchthat evaluates its substantive behavioral andpolicy outcomes, as well as research that exca-vates its social origins.

It would be circular to adopt socially ac-cepted definitions of compliance in researchthat seeks to identify and explain substantive


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


compliance behaviors and their outcomes: Thedefinitions of compliance that emerge fromsocial negotiation have often been put forthspecifically because they represent a standard ofbehavior that can easily be achieved in practice.This standard will likely be lower than or dif-ferent from the standard required by the policygoals that initially motivated the regulation orthat the community or stakeholders might havethought were or should be the intent of the rule.If we adopt the bottom-up perspective of un-derstanding parties’ interpretations of compli-ance in practice, we cannot evaluate and explaincompliance unless we compare it with someoneelse’s. Social science researchers therefore needto be able to conceptualize for ourselves whatcompliance means in each situation we seek toresearch either by developing, justifying, andmaking explicit our own substantive evaluationcriteria or by explicitly acknowledging that weare adopting the evaluation criteria espoused byone or other of the parties with an interest inregulation. In the latter case, we need to be clearabout its biases and limitations.

RESEARCH EXPLAININGTHE CAUSES AND EFFECTSOF COMPLIANCE

Operationalizing Compliance

Research that has a predefined exogenous un-derstanding of compliance begins with relatingabstract concepts to the real world through def-inition, operationalization, and measurement.The validity of any empirical research will al-ways, therefore, be a matter of the degree towhich it captures the concept of interest.

Validly translating between the abstractconcept and the real world is often thought ofas having several stages, as shown in Figure 2.Any social science researcher must deal with anumber of obstacles in moving from an abstractconcept, such as regulatory compliance, tothe point where one has collected valid andreliable data that allow one to draw conclusionsabout one’s research questions, such as levelsof compliance, correlations between different

Abstract concept:

e.g., regulatory compliance

Theoretical definition:

How the researcher defines

the concept in words

Operational definition:

Classification rules as to what

data the researcher sees as

relevant to the concept’s

definition

Data collection and coding:

Gathering and using data for a

measure following the rules of

the operational definition

Figure 2Chain of validity for compliance research measuresthat are exogenously defined (inspired by McDonald2005).

aspects of behavior relevant to compliance,and tests of hypotheses about what explainsvariation in compliance. At the most abstractlevel is the concept, the underlying theoreticalconstruct that is to be studied. Often there willbe a preliminary discovery or exploratory stageof uncovering the concept and the differentmeanings it has for different people or in differ-ent kinds of studies. The theoretical definitiongives the concept meaning through a concretedescription. The operational definition pro-vides rules of classification to distinguish cases.Measures, also called indicators or variables,are generated through the process of scoringor gathering data by following the rules ofclassification.

If the researcher’s hard work at each stageis to be relevant and fruitful for the next stage,the use of the concept at each stage must corre-spond and correlate to its use at the other stages.This does not mean statistical correlation andcorrespondence, although statistical toolsmay be a useful check where the researcher


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


is working with quantitative data. What isimportant to both quantitative and qualitativeempirical research studies is that within eachstudy there is a logical or epistemic correlationthat links the different stages shown in Figure 2together (Bryman 2008, pp. 143–53). If thatdoes not exist, statistical rigor will be restingon a very weak logical foundation that meansthe statistical measures can never be valid.Having established a case for the epistemiccorrespondence between the abstract concept,theoretical understanding, and operational def-inition of regulatory compliance, there are stillpotential obstacles in the sourcing, collection,coding, and analysis of data (the last stage inFigure 2). These can, as mentioned above,create two problems of validity: the biased datasource problem and the shaky informationproblem.

In the subsections below, we first considertwo different ways in which researchers avoidoperationalizing actual compliance by focus-ing instead on business attitudes and motiva-tions or focusing on the policy goals that busi-ness should meet by complying with regulation.We go on to argue that empirical complianceresearchers need also to do projects that op-erationalize compliance by reference to actualcompliance behavior for the purposes of discov-ering causes and effects of compliance. In thefinal subsections, we discuss the challenges ofdata collection that attempts to operationalizeactual compliance behavior.

Operationalizing Complianceby Reference to Attitudesand Motivations

One tendency in the empirical literature onbusiness compliance is to focus on compliancemotivations and attitudes of businesses as thevariable of interest, rather than on actual com-pliance. Studies like this include Valerie Braith-waite and coauthors’ work on motivational pos-tures in which constellations of attitudes andmotivations toward compliance and regulatorsare treated as key dependent variables (e.g.,V. Braithwaite et al. 1994, 2007), and Peter

May and Soeren Winter’s (May 2005, Winter& May 2001) studies of various motivations forcompliance among Danish farmers in rela-tion to environmental regulation, U.S. home-builders in relation to building safety, and U.S.marine facilities in relation to water quality.

A number of studies examine the culture andmanagement systems of businesses using eitherquantitative or qualitative methods or both inorder to understand the compliance motiva-tions and attitudes of the business as a business(e.g., Ashby & Diacon 1996, Florida & Davison2001, Parker & Nielsen 2009, Weaver &Trevino 1999) or its management style, com-mitment, and organizational culture. For exam-ple, Gunningham et al.’s (2003) study of busi-ness compliance with environmental regulationfinds a high level of actual compliance acrossall businesses but distinguishes between reluc-tant compliers, committed compliers, environ-mental strategists, and true believers and drawssome conclusions about what factors influencethese different constellations of compliance at-titudes and motivations.

This sort of research is attractive becausedata can be collected on attitudes and motiva-tions more accurately and reliably than on com-pliant or noncompliant behavior through self-reports (but see discussion of data collection be-low for the limitations of self-report data frombusinesses). Attitudes and motivations are likelyto be very important explanatory variables forcompliance behavior and are therefore worthstudying (Parker & Nielsen 2009). Measuringthe attitudes and motivations of business firmsand individuals about compliance can also bea useful evaluation standard in itself of the suc-cess of regulation in terms of its likely effective-ness and its democratic legitimacy, i.e., the reg-ulatee’s consent to be regulated (V. Braithwaiteet al. 2007). This might be especially useful asregulators and stakeholders may well be morecapable of changing business attitudes and mo-tivations toward compliance than of changingother factors that explain compliance such assize, resources, or management capabilities.

Nevertheless, measuring a firm’s attitudesand motivations about compliance is not the


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


same as measuring compliance behavior, and apositive attitude or motivation may not lead tomatching behavior. They are therefore not aproxy or substitute for explaining compliancebehavior because we cannot be sure that theassumed connection to actual compliancedoes exist. It is therefore useful to distinguishbetween behavioral compliance and noncom-pliance (actual behavior) and psychologicalnoncompliance (attitudes of commitment orresistance). Both are relevant and useful fordifferent research questions (Murphy 2005,p. 570; see also Braithwaite 2003).

Regulatory capitalism’s desire to solve prob-lems of corporate power and irresponsibility byresponsibilization, self-regulation, and businessinvolvement in networked regulation mightmake us too eager to see good intentions as asubstitute for compliance actions. Complianceattitudes and intentions are worth studying intheir own right, but it is important to keep clearthat they may not translate into action becausethey are not made in good faith, they are actedon in a purely symbolic and ineffective way, orthey are stymied by other forces (Krawiec 2003,Parker & Nielsen 2009).

Operationalizing Complianceby Reference to Policy Goals

Another way that researchers have tried to over-come some of the vagaries of defining com-pliance in regulatory capitalism is by focusingon whether businesses meet certain substan-tive goals such as reduced environmental emis-sions (Andrews 2003; Berkhout & Hertin 2001;Johnston 2006; Winter & May 2001, 2002),greater employment of women or racial minori-ties (Braithwaite 1993), fewer worker injuriesand fatalities (Mendeloff & Gray 2005), andso on. The policy goal or value outcome thatresearchers choose to study might come fromthe researchers themselves or from others. Aresearcher might adopt the standpoint of thelocal community or NGO in deciding that heor she wants to evaluate whether certain emis-sions are down or the standpoint of financialservices companies in evaluating whether regu-lation results in a chilling effect on risk-taking

in investment. Or the researcher might seek todetermine what the government’s original pol-icy goal was in adopting a piece of legislationand then measure whether it was achieved. Ora researcher might develop and justify his orher own set of criteria for evaluating the broadoutcomes of regulation.

Such an approach is useful if the aim of theresearch is critique and evaluation of policy im-plementation. It takes a substantive social goaland uses it as the evaluation standard. It avoidsthe problem that rules in regulatory capital-ism might be quite vague about how businessesshould specifically behave and instead evaluateswhether the policy outcome has been achieved.But this approach cannot say whether it is com-pliance with regulatory rules that delivers theoutcome or not. As Tombs (2000) points out,a researcher might use numbers of injuries anddeaths as a measure of compliance in relation tooccupational health and safety regulation, butinjuries and deaths are not always caused by ille-gal conduct. Rather the researcher is conceptu-alizing the thing to be studied not as behavioralcompliance with a rule, but rather compliancewith a substantive objective that the rule is ex-pected to serve, what political scientists wouldcall the overall implementation success of apolicy.

Operationalizing Complianceas Compliance Behavior

Ultimately, compliance researchers wantto evaluate whether people comply withregulation—even very vague, plural, or decen-tered regulation—and whether this leads toachieving substantive goals. Researchers aretherefore still left with the problem of definingand operationalizing evaluation standard(s) forcompliance. Although this task is complex, it islogically important for both policy evaluationand theory testing to be able to explain whichregulatory interventions prompt which behav-ioral responses and whether these behaviorslead to the desired goal.

In doing this, the researcher him-/herselfwill become part of the process of construct-ing the meaning of compliance. The researcher


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


cannot choose an evaluation standard withoutimplicit or explicit use of norms and values.This puts a heavy moral responsibility on theresearcher. He or she is to define the evalu-ation standard for compliance that he or shethinks is reasonable and/or morally defensible.This might mean that the researcher has to de-fine compliance with a rule from scratch, orit might mean owning up to whose norma-tive standpoint the researcher is adopting—aregulator’s, some stakeholder’s, or regulatees’.From a democratic policy-study point of view,the standard should be one that is closely andloyally defined by a reading of the intentionsand interests of the legitimately elected de-cision makers (Mazmanian & Sabatier 1981,1983; Olsen 1978; Winter et al. 1994). But fromanother—more critical—point of view, the re-searchers’ own normative standard or the stan-dard of certain actors in the field (e.g., certaincitizen/interest groups, the regulatory agenciesand inspectors, or the regulatees) is equally—oreven more—legitimate (Guba & Lincoln 1981).There is no standard handbook approach towhat role to play as a researcher. It is a matter ofindividual beliefs and position. But researchers’and evaluators’ ambition should be to defineexplicitly their own positions and measure(s) ofcompliance.

Researchers’ capacities to operationalizecompliance their own way might be con-strained, however, by the available data sources.Thus, we must address biased data sourcesand shaky information and what definitionsof compliance are implicit in the data sourcesavailable for studying compliance in regulatorycapitalism.

Collecting Data by Observationof Regulatory Compliance Behavior

Ideally, once social science researchers have de-fined and operationalized regulatory compli-ance, they will go out and directly observe rel-evant business behavior that they can code ascompliant or noncompliant according to theirdefinitions, as shown in Figure 1 (e.g., Burby& Paterson 1993, Fairman & Yapp 2005). They

might use secret observation so that there is nointerference attributable to the research sub-ject knowing the researcher is watching. Theymight use an experimental design if they aretrying to test the impact of regulation on behav-ior (e.g., McGraw & Scholz 1991). They mightconduct qualitative interviews before observingcompliance behaviors to develop and hone theirdefinitions of the concepts they are interestedin and also their hypotheses or research ques-tions, and then again after observation to helpinterpret the results. But, ideally, researcher-controlled observation of compliance behaviorsshould always form the core of data collectionfor the purpose of causal explanation.

In practice, however, direct observation ofbusiness compliance and noncompliance by re-searchers is generally impractical, and stud-ies that rely on this method of data collectionare extremely rare (but see the studies cited inthe previous paragraph). Time and resourcesfor the researcher and his/her assistants to bepresent for long enough to observe sufficientbehaviors are often a particular problem. Inmany countries, government and university hu-man research ethics committees (institutionalreview boards in the United States) also heav-ily restrict or even prohibit observation of il-legal behavior, especially if such observationis to be done secretly. There are also obviousethical and political problems in experimentingwith regulatory design and interventions: Ran-domly assigning a large number of businesses toa more lenient, cooperative inspection regimewhile a similar number are assigned to a stricter,punitive regime for experimental and statisticalpurposes raises obvious issues. A control groupthat is not regulated at all may be the ideal wayto test the impact of different regulatory en-forcement conditions on compliance behavior,but it is likely to take an unusually persuasiveresearcher and a very farsighted regulator toachieve this sort of experimental design.

Because the researcher is generally inter-ested in studying business organizations—andoften quite powerful ones—he or she can havegreat difficulty obtaining access to observe rel-evant behavior. Moreover, even with adequate


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


access, the researcher will need highly special-ized knowledge and even specialist technologyto accurately detect and identify legally com-pliant and noncompliant behavior. For exam-ple, it is difficult for laypeople (as most reg-ulatory researchers are in the different arenasof regulation) to detect whether or not a com-pany is violating building codes or environmen-tal, health and safety, and fire precaution laws.One solution, if time and resources allow, is forthe researcher to hire an independent rater withspecialized knowledge to judge regulatees’ be-havior (e.g., Braithwaite et al. 1992) or for a re-searcher to train as an inspector (e.g., Fairman& Yapp 2005).

Some compliance researchers overcomethese problems by using quasi-experiments inwhich regulatees are asked to report their mostlikely behavioral responses to various hypothet-ical scenarios in a written questionnaire, inter-view, or role play (e.g., Edwards 1991, Feldman& Lobel 2008, Gunz & Gunz 2002, Simpson2002). Some have even used lab experimentsin the behavioral economics tradition to mea-sure compliance tendencies (Casey & Scholz1991). These give the researcher control overboth the facts and how they define compliance.But the hypothetical or laboratory nature of thisresearch means that at best we find out aboutcompliance tendencies. We cannot predict forsure what the research subjects would do in reallife and to what extent we have elicited an ide-alistic or politically correct answer. That is, theexperiment might result in information that isshaky or biased.

Researcher Dependence onOther Data Sources

Because of the impracticality of observing com-pliance for themselves, researchers often de-pend on other sources for data about actualcompliance behavior: state-based regulatoryagencies, regulated businesses, or various stake-holders and nonstate-based regulators.

Many researchers of regulatory compliancesource their data from state-based regulatoryagencies and the data these bodies collect on

rule violations and enforcement actions: Thisincludes records of the results of compliancemonitoring and inspection visits (e.g., Gray &Scholz 1993; Hill et al. 1992; Mendeloff &Gray 2005; Nielsen 2006; Nyland et al. 2006;Prakash & Potoski 2006, pp. 132, 136; Short &Toffel 2007), publicly available records of en-forcement action (notices, fines, enforcementand discipline, litigation) (e.g., Baucus & Near1991, p. 16; Dalton & Kesner 1988; Harrison1995; McKendall et al. 1999; Simpson 1987),and other records of actions taken to com-ply with regulatory requirements (Braithwaite1993, Johnston 2006, Kleindorfer 2006).

State-based regulatory agencies have tradi-tionally been a rich data source because pub-lic regulatory agencies’ own accountability re-quirements often authorize or mandate them tomake their enforcement and compliance infor-mation available to researchers. But as regula-tory capitalism places more and more respon-sibility on business organizations for their ownregulation, and as other nonstate-based formsof regulation abound, more and more relevantdata may be held by businesses themselves or byindustry or civil society–based regulators (suchas certification agencies), and less and less bypublic regulatory agencies.

A second source of information about reg-ulatory compliance is therefore self-reports byregulated businesses or individual officers andemployees of regulated businesses: Self-reportsmay be collected in surveys, open-ended in-terviews, and, occasionally, internally collecteddata that is passed on to the researcher. Mostresearchers collect their data from individu-als within larger organizations (e.g., Weaver &Trevino 1999) or the principals of small busi-ness organizations (e.g., Kuperan & Sutinen1998; Murphy 2005; Scholz & Lubell 1998;Wenzel 2001; Winter & May 2001, 2002), notthe leaders of larger organizations (e.g., Nielsen& Parker 2008, Parker & Nielsen 2009).

A third source of data is third-partystakeholders: These might include potentialvictims of business noncompliance such as con-sumers or local residents, professional advi-sors and other stakeholders who might have


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


particular knowledge about business compli-ance behaviors (e.g., Beckenstein & Gabel1983), and even nonstate regulators. For exam-ple, employees might be asked how often theyhave experienced workplace harassment or un-reported occupational injuries (Mayhew et al.1997), or consumers might be asked how oftenthey have experienced food poisoning or pur-chased faulty products. But these types of victimsurveys are not capable of measuring victimlesscrimes or misconduct where the harm is diffuseor not able to be detected by victims (e.g., manyinstances of price-fixing). For these reasons, itis generally accepted that victim surveys under-estimate “corporate crime and so-called ‘crimesof the powerful’” ( Jupp 1989, p. 107).

As already mentioned, however, relying onthese various sources of data raises concernsabout the reliability and validity of the data thatcan be obtained in the form of biased sourcesor shaky information. Even though a researcherhas defined and operationalized compliance ina way that makes sense for his or her own re-search project, shaky information and bias inthe data sources might make it challenging orimpossible to collect data that actually addressthe research question. As with the conceptu-alization of compliance, the nature of regula-tory capitalism creates particular challenges ofshaky or biased information. We discuss thechallenges of collecting valid and reliable in-formation from two of the most popular datasources for compliance research—official regu-latory agencies and self-reports from regulatedbusinesses—in turn below.

Data from OfficialRegulatory Agencies

Criminologists have long recognized that of-ficial crime statistics are shaky because theydo not account for the dark figure of crime—crime not reported to police in the first place, ornot recorded by them (Coleman & Moynihan1996; Jupp 1989, p. 87). Compliance datacollected by business regulators are likely tobe even less complete (Shover & Hochstetler

2006, p. 13). Many business violations are neverofficially discovered or recorded. This is espe-cially true of the records of reactive businessregulators (Kagan 1994, p. 387) that only in-vestigate potential breaches when complaintsare made to them, rather than engaging in sys-tematic, proactive monitoring and inspectionto seek out breaches: Victims will often notreport the offense because the harm is diffuseand not seen as worth complaining about (e.g.,many consumers who all lose a small amountof money through purchasing a faulty prod-uct). Organizations might also hide breaches onpurpose (e.g., secret illegal dumping of pollu-tants in a river, cartel conduct) or unintention-ally via complex organizational structures (e.g.,a label that represents that a food product iscertified organic where not all supplies of thatfood are always in fact from organically certifiedproducers). Even proactive regulators that ac-tively monitor compliance through inspectionsor audits miss many violations that come andgo during the intervals between inspections andrarely discover all violations even at the timeof inspection. Of those that are detected, notall are recorded because of inattention, discre-tionary leniency, or even corruption (see Hillet al. 1992; Weil 1996, p. 619).

Regulators’ discretion means that the datathey record about compliance and noncompli-ance are likely to be normatively biased. Wehave already seen that social, political, and eco-nomic factors influence official enforcementdecisions about what counts as compliance ornot. In many instances, regulators’ understand-ing and recording of compliance will be based tosome extent at least on how regulated businesseshave negotiated and constructed the meaning ofcompliance.

Methodologically, shaky information andbiased data sources are two distinct problemsthat do not necessarily correlate: One mightvery well imagine an official database with per-fectly inclusive data established with referenceto a very narrow, biased definition of the breach.But the two problems often feed off each other.Data can be shaky in systematic ways that create


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


institutional biases, or normative biases in theinstitution can create shaky data.

Shaky and biased data from regulators are aproblem not only because they are biased butbecause the bias and its criteria are nontranspar-ent. Often researchers will have access only to aset of official enforcement statistics that are theoutcome of a range of subjective assessments ofbehavior. Moreover, regulatory agency person-nel’s filtering processes and criteria for decid-ing what is illegal are often unconscious, non-explicit, and therefore unknown. Ideally, theresearcher would have access to the full com-plaints or inspection records and uncensoredobservations of individual regulatory staff in or-der to understand what biases and inaccuracieshave affected the creation of official statistics(e.g., Nielsen 2002). This underscores the im-portance of empirical research that uncoversthese biases by examining processes of socialconstruction of compliance and the making ofagency records. In addition to being worthwhileand interesting in itself, this type of researchhelps us interpret and use official statistics ef-fectively, or at least to note their limitations.

In one sense, biased data from regulators arenot a problem if the individual researchers’ con-ceptualization and operationalization of com-pliance fit that which is used in the officialdata. The problem is that in many situations us-ing official data sources will unacceptably limitour understanding of compliance. Existing datasources will tend to be normatively filtered inways that reproduce existing positions of power;it is important for policy and theory that re-searchers also bring different evaluation stan-dards to bear and do not always rely on officialdata sources.

Self-Report Data from Regulatees

Many empirical studies of regulatory compli-ance seek to avoid the limitations—and oftensheer lack of availability—of official statisticsby asking regulatees themselves how much theyhave broken the law or to what extent theycomply with it. In criminology, it is generally

accepted that self-reports are most useful forstudying low-level criminal activity and vic-timless crimes that may not be reported tothe police and have relatively low levels ofsocial stigma attached to them, especiallydelinquency among adolescents (Coleman &Moynihan 1996, p. 55). Self-report data may bequite useful in business regulation complianceresearch for similar reasons. Many regulatoryviolations are not stigmatized so much that wewould never expect people to admit to them in astudy with suitable anonymity and confidential-ity assurances. Moreover, as we have seen, oftenthe victims will not report business offenses toregulators, or even be aware enough of themto be able to report them to researchers whenasked. Self-report data have the advantage thatgenerally the respondents themselves are morelikely to know about their own history of com-pliance with the law than do enforcement offi-cials or anyone else.

There are also, however, a number of well-known shaky information problems with thereliability of self-report data on compliance.Criminological studies have shown that under-or overreporting of crime is common becauseof simple lapses of memory or more complextricks of memory such as telescoping in whichrespondents remember the relevant activity butat the wrong time: Events outside the surveyperiod may be brought forward into it or viceversa (Huizinga & Elliott 1986, pp. 166–67). In-dividuals admitting to fewer criminal acts provemore reliable than do frequent offenders whohave more to remember, report, and potentiallycover up (Huizinga & Elliott 1986). Serious of-fenses are generally recalled with greater accu-racy than trivial ones, when they are recalled(Coleman & Moynihan 1996, p. 66).

Self-reports of illegal activity also, of course,provide biased data. People are generally likely(either deliberately or subconsciously) to in-terpret, remember, and report events in sucha way as to exaggerate their compliance withthe law and underplay or excuse noncom-pliance or to overreport trivial offenses andunderreport serious offenses ( Jupp 1989,


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


p. 102). Researchers generally seek to over-come social desirability bias by guaranteeinganonymity, by framing questions about illegalactivity in neutral and factual ways (e.g., Winter& May 2002, p. 126), and also by giving respon-dents a range of apparently socially acceptableways to confess noncompliance. For example,Scholz & Lubell (1998, pp. 402–3) gave respon-dents the opportunity to report that they defi-nitely did, probably did, probably did not, anddefinitely did not report all income to the taxauthorities—three ways of reporting noncom-pliance.

Where the respondents are business firmsand the topic is regulatory compliance, prob-lems with accurate memory of illegal activityand social desirability bias are more complexthan with individual criminal activity. In anyorganization, but especially in a large organi-zation, the researcher must ask an individualor individuals to fill out the survey or respondto the interview. On the one hand, the indi-vidual respondent may not know enough aboutwhat goes on inside the organization, or per-haps even about the law or technical processes(in complex areas), to report accurately on orga-nizational compliance. Ideally, self-report stud-ies should include representatives of every partof the organization where the relevant illegalactivity might have occurred (e.g., Weaver &Trevino 1999, Key 1999). But this might beimpractical for reasons of access or cost. Onthe other hand, the fact that there are manypeople within the organization who potentiallyknow about illegal activity means that theremay also be a number of access points into theorganization with varying levels of knowledgeand susceptibility to social desirability bias (e.g.,Hutter 2001). There is also the possibility thatthe researcher can gain access to corporaterecords that document compliance contempo-raneously over a long period in a relativelyobjective way, e.g., emissions or customer com-plaints. But any individual or record’s knowl-edge of compliance is institutionally created.Researchers need to understand the inter-nal processes by which any businessperson’s

knowledge of compliance was created in orderto understand the value of that information.

Social desirability is also more complex inregulatory compliance research. It is more so-ciologically and psychologically plausible foroffenders and observers to fail to see corporatenoncompliance as illegal. Indeed, business firmsmay be intentionally organized so that infor-mation about noncompliance is hidden (Tombs2000, p. 75). To the extent that regulatory non-compliance is not seen as morally wrong oris not stigmatized in the same way as streetcrime, respondents may simply be less likely toremember breaking the law because it is notmemorable. Similarly, to the extent that regu-latory compliance and noncompliance are seenas having fuzzy borderlines, then it may be verydifficult to define accurately illegal activity andto get respondents to remember which side ofthe line their own (or their businesses’) activ-ity fell on. However, lack of social stigmati-zation of regulatory noncompliance may alsomake it easier to counteract social desirabilitybias. Questions may be able to be worded rel-atively neutrally about certain activities so asnot to trigger social desirability bias. More-over, because the reality is that many regula-tory breaches are never prosecuted or taken se-riously, respondents should have less fear aboutreporting illegal activity to researchers.

In other areas, researchers generally try tocheck the reliability of self-report data againstother data sources. To the extent that regula-tory capitalism uses more and more decentered,plural, and private regulatory governance, moreof the information relevant for compliance re-search will be in private organizations, whereaccess for researchers is difficult and opportu-nities for reliability checking rare.

Even now, a number of studies that appearto use official statistics from regulators in factuse self-reports because the official statisticsare based on more or less unfiltered self-reportdata on compliance (Bennear & Coglianese2005). For example, because U.S. environmen-tal regulators have required a great deal ofself-reporting of emissions data, a number of


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


studies of compliance in environmental reg-ulation are in fact based on official data thatare themselves based on self-report data fromregulatees. The U.S. Toxic Release Inventoryscheme, for example, is relied upon by manyresearchers, but the legislation “does not set astandard for accuracy and gives wide latitudefor estimates” (de Marchi & Hamilton 2006,pp. 59–60). Research has shown that at leastsome emissions data self-reported by businessesare not reliable and do not match data from ex-ternal monitoring of air pollution (de Marchi& Hamilton 2006), and that firms might alsostrategically manage their operations to fall un-der reporting thresholds leading to inaccuraciesin researchers’ measures of reductions in toxicreleases of up to 40% (Bennear 2008, p. 298).

The same type of difficulties can occur withstudies of occupational health and safety regu-lation compliance, which are based on manda-tory self-reports of injuries and on mandatorydisclosures of incidents of risk (Tombs 2000).Harrison (1995, p. 229) reports that when sheasked the Canadian federal environmental reg-ulator for compliance data she was “met withthe remarkable response that the data could notbe released until it was verified by the regulatedindustry.” This response probably reflects thefact that official compliance data were based atleast partly on business self-reporting. It alsoreflects the power of the industry vis-a-vis theregulator to close off external scrutiny of com-pliance by a researcher.

Regulatory capitalism increasingly func-tions by requiring the regulated to takeresponsibility for regulating themselves toa certain extent by implementing their owninternal compliance management and data col-lection systems and then reporting data abouttheir implementation of management systemsand/or their compliance or other performanceoutcomes back to the regulator and/or tothe public (e.g., environmental emissions,employee days off work, number of women ormembers of racial minorities employed) (Parker2002, pp. 277–83). But the regulator does notnecessarily have any objective data not sourced

from the regulated, and neither therefore doesthe researcher. For example, Winter & May(2001, p. 694) corroborate their self-reportdata against official data. But we have alreadyseen that in many studies (including Winter& May’s) the official data are themselves basedlargely on self-reports. So if they corroborateone another then it may simply mean that thesame biases are replicated in both data sets.

In some cases (e.g., emissions), this will raiseissues only of shaky information because the of-ficial statistics based on self-reported data askthe regulatees to report fairly detailed and ex-plicit information on phenomena that both theregulator and the regulatee have a fairly alikedefinition of. The concerns are that the infor-mation reported is shaky simply because theregulatees lie or do not know the right an-swer. But when regulators ask regulatees toself-report on implementation of, for example,appropriate risk management systems, thenthere is room for bias through different norma-tive judgments about what counts as an appro-priate system. Regulatees may have succeededin influencing the whole way compliance is con-structed by the regulatory enforcement system,and neither self-reports nor official data arelikely to provide a different viewpoint (Black2008, Krawiec 2008).

The various reliability problems with differ-ent kinds of self-report data from regulateesare not necessarily merely technical problemsthat can be controlled or corrected for withthe addition of data, statistical procedures, orappropriate technique in survey questionnairedesign. The danger is that there will develop(or has already developed) a profound ideolog-ical/normative match between regulated busi-nesses and the regulatory enforcement system.As a result, corporate illegality may often re-main invisible not only to regulators and thepublic at large, but also to researchers—andperhaps even the regulatees themselves.

The best solution to this problem is to tryto find information about compliance from dif-ferent sources and stakeholders with differentviewpoints. There are some studies in which


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


researchers have attempted to corroborate theirself-report data against data from a completelyseparate source: For example, in Bussmann &Werle’s (2006) study of corporate fraud insidecompanies in a range of countries, they cor-roborate the overall levels of fraud reported foreach country against each country’s ranking inthe OECD’s corruption perception index. Onemight also test the reliability of self-report orofficial regulatory agency data by asking a rangeof stakeholders surrounding the firm, such asemployees or customers, about the firm’s com-pliance with the law (e.g., Mayhew et al. 1997).

CONCLUSION

In an age of both proliferating regulation andcapital, many of the political, economic, and so-cial research questions that social scientists seeas most significant will raise issues of regula-tory compliance. To answer these questions, weneed to understand what effect regulation hason behavior and policy outcomes. Although thisreview has been concerned with research thataddresses business as the object of regulation,increasingly government bodies, states, and allsorts of agents of civil society and military andquasi-military groups are the subject of regu-lation. Understanding how and why they arelikely to comply is fundamental to understand-ing the prospects for peace and global justicemore broadly (Braithwaite 2002, pp. 169–210;Chayes & Chayes 1998).

The very characteristics of regulatorycapitalism—the responsibilization of businessand the decentering of regulation away fromthe state—however, create difficulties in con-ceptualizing, operationalizing, and measuringcompliance. Different regulators, regulatees,and stakeholders involved in plural regulatoryregimes will each have their own ideas aboutwhat compliance means. To the extent thatthere are any socially accepted meanings ofcompliance, they emerge from a process of in-teraction between these parties. This createsinteresting research opportunities, but it doesnot fit neatly with policy expectations that re-searchers can measure how different attempts at

regulation impact compliance and come up withobjective and certain results. To understandcompliance adequately, researchers must col-lect data on information, processes, attitudes,motivations, and, finally and often most diffi-cult and critical of all, actual behavior and itspolicy outcomes.

The problems this lack of a clear definitionposes for social science researchers are famil-iar. There is no one objective standpoint fromwhich social scientists can study a social phe-nomenon like compliance. The fact that sociallyaccepted understandings of compliance are of-ten negotiated by regulators with regulatedbusinesses and other stakeholders means thatsocial science researchers need to be particu-larly wary of uncritically adopting either officialor business definitions and measures of com-pliance. Compliance researchers have to takeresponsibility to develop, justify, and explicitlyexplain our own definitions of what counts ascompliance or to be clear about the limitationsof the definitions we adopt from parties who areinvolved in the regulation process. Even oncethe researcher has settled on a satisfactory wayof conceptualizing and operationalizing com-pliance for the purposes of his or her own re-search, data sources that fit the researcher’s ap-proach may not be available. A particular issuein regulatory capitalism is that businesses’ po-tentially self-serving views of what should countas compliance for themselves will often have in-sinuated themselves into even apparently objec-tive official data, and much compliance researchwill have to depend on self-report data of onesort or another.

Ultimately, plural conceptualizations ofcompliance and, consequently, the use of mul-tiple methods for measuring compliance aredesirable and necessary. Different researchquestions and research ambitions place differ-ent demands on data—demands that some-times, because of data validity problems, aredifficult to fulfill. Because no one method ofmeasuring compliance is likely to be perfect andcorporate compliance itself is a contested con-cept, multiple methods will build a better socialscience of regulatory capitalism overall.


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


DISCLOSURE STATEMENT

The authors are not aware of any affiliations, memberships, funding, or financial holdings thatmight be perceived as affecting the objectivity of this review.

ACKNOWLEDGMENTS

The authors acknowledge the funding support of the Australian Research Council; helpful dis-cussions and comments from John Braithwaite, Bob Kagan, David Levi-Faur and Susan Silbey;and research assistance from Sharyn Broomhead.

LITERATURE CITED

Andrews R. 2003. Environmental management systems: Do they improve performance? Project Final Rep.,National Database on Environmental Management Systems, Univ. N. C., Chapel Hill. http://ndems.cas.unc.edu

Ashby S, Diacon S. 1996. Motives for occupational risk management in large UK companies. Saf. Sci. 22:229–43

Baucus M, Near J. 1991. Can illegal corporate behaviour be predicted? An event history analysis. Acad. Manag.J. 34:9–36

Beckenstein A, Gabel H. 1983. Antitrust compliance: results of a survey of legal opinion. Antitrust Law J.51:459–516

Bennear L. 2008. What do we really know? The effects of reporting thresholds on inferences using environ-mental right-to-know data. Regul. Gov. 2:293–315

Bennear L, Coglianese C. 2005. Measuring progress: program evaluation of environmental policies. Environ-ment 47(2):22–39

Berkhout F, Hertin J. 2001. Towards environmental performance management. MEPI Study Prepared for DGRes. Eur. Comm., Jan., Sci. Technol. Policy Res., Univ. Sussex, Brighton

Bernstein S, Cashore B. 2007. Can non-state global governance be legitimate? An analytical framework. Regul.Gov. 1:347–71

Black J. 2001. Decentring regulation: understanding the role of regulation and self regulation in a “post-regulatory” world. Curr. Legal Probl. 54:103

Black J. 2003. Enrolling actors in regulatory processes: examples from UK financial services regulation. PublicLaw Spring:62–90

Black J. 2008. Forms and paradoxes of principles-based regulation. Cap. Markets Law J. 3:425–57Braithwaite J. 2000. The new regulatory state and the transformation of criminology. Br. J. Criminol. 40:222–

38Braithwaite J. 2002. Restorative Justice and Responsive Regulation. Oxford: Oxford Univ. PressBraithwaite J. 2008. Regulatory Capitalism. Cheltenham, UK: Edward ElgarBraithwaite J, Drahos P. 2000. Global Business Regulation. Cambridge, UK: Cambridge Univ. PressBraithwaite J, Makkai T, Braithwaite V. 2007. Regulating Aged Care: Ritualism and the New Pyramid.

Cheltenham, UK: Edward ElgarBraithwaite V. 1993. The Australian government’s affirmative action legislation: achieving social change

through human resource management. Law Policy 15:327–54Braithwaite V. 2003. Dancing with tax authorities: motivational postures and non-compliant actions. In Taxing

Democracy: Understanding Tax Avoidance and Evasion, ed. V Braithwaite, pp. 15–39. Aldershot, UK: AshgateBraithwaite V, Braithwaite J, Gibson D, Landau M, Makkai T. 1992. The Reliability and Validity of Nursing

Home Standards. Canberra: Aust. Gov. Public Serv.Braithwaite V, Braithwaite J, Gibson D, Makkai T. 1994. Regulatory styles, motivational postures and nursing

home compliance. Law Policy 16:363–94Braithwaite V, Murphy K, Reinhart M. 2007. Taxation threat, motivational postures and responsive regulation.

Law Policy 29:137–58


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


Bryman A. 2008. Social Research Methods. Oxford: Oxford Univ. PressBurby R, Paterson R. 1993. Improving compliance with state environmental regulations. J. Policy Anal. Manag.

12:753–72Bussmann K, Werle M. 2006. Addressing crime in business: first results from a global survey of economic

crime. Br. J. Criminol. 46:1128–44Casey J, Scholz J. 1991. Beyond deterrence: behavioral decision theory and tax compliance. Law Soc. Rev.

25:821–44Chayes A, Chayes A. 1998. The New Sovereignty: Compliance with International Regulatory Agreements. Cam-

bridge, MA: Harvard Univ. PressCoglianese C, Kagan RA. 2007. Regulation and regulatory processes. In Regulation and Regulatory Processes,

ed. C Coglianese, RA Kagan, pp. xi–xxxvi. Aldershot, UK: AshgateCoglianese C, Nash J, eds. 2006. Leveraging the Private Sector: Management-Based Strategies for Improving

Environmental Performance. Washington, DC: Resour. FutureColeman C, Moynihan J. 1996. Understanding Crime Data: Haunted by the Dark Figure. Buckingham, UK:

Open Univ. PressDalton D, Kesner I. 1988. On the dynamics of corporate size and illegal activity: an empirical assessment.

J. Bus. Ethics 7:861–70De Bruijn H, Heuvelhof E, Koopmans M. 2007. Law Enforcement: The Game Between Inspectors and Inspectees.

Boca Raton, FL: Universal Publ.De Burca G, Scott J. 2006. Law and New Approaches to Governance in the European Union and the United States.

Oxford: HartDe Marchi S, Hamilton J. 2006. Assessing the accuracy of self-reported data: an evaluation of the Toxics

Release Inventory. J. Risk Uncertain. 32:57–76Dorf M, Sabel C. 1998. A constitution of democratic experimentalism. Columbia Law Rev. 98:267–473Edelman L, Petterson S, Chambliss E, Erlanger H. 1991. Legal ambiguity and the politics of compliance:

affirmative action officers’ dilemma. Law Policy 13:73–97Edwards M. 2006. Law and the parameters of acceptable deviance. Crim. Law Criminol. 97:49–100Edwards P. 1991. Choices that increase compliance. Policy Stud. Rev. 10:6–27Ericson R, Doyle A. 2006. The institutionalization of deceptive sales in life insurance: five sources of moral

risk. Br. J. Criminol. 46:993–1010Fairman R, Yapp C. 2005. Enforced self-regulation, prescription, and conceptions of compliance within small

businesses: the impact of enforcement. Law Policy 27:491–519Feldman Y, Lobel O. 2008. Behavioral versus institutional antecedents of decentralized enforcement in orga-

nizations: an experimental approach. Regul. Gov. 2:165–92Florida R, Davison D. 2001. Why do firms adopt advanced environmental practices (and do they make a

difference)? In Regulating from the Inside: Can Environmental Management Systems Achieve Policy Goals?, ed.C Coglianese, J Nash, pp. 82–104. Washington, DC: Resour. Future

Fukuyama F. 1995. Trust: The Social Virtues and the Creation of Prosperity. London: Hamish HamiltonGarsten C, Jacobsson K. 2007. Corporate globalization, civil society and post-political regulation—whither

democracy? Dev. Dialogue 49:143–57Geis G. 1967. White collar crime: the heavy electrical equipment antitrust cases of 1961. In Criminal Be-

haviour Systems: A Typology, ed. M Clinard, R Quinney, pp. 139–51. New York: Holt, Rinehart &Winston

Genn H. 1993. Business responses to the regulation of health and safety in England. Law Policy 15:219–33Gilad S. 2008. Accountability or expectations management? The role of the ombudsman in financial regulation.

Law Policy 30:227–53Gilardi F, Jordana J, Levi-Faur D. 2006. Regulation in the age of globalization: the diffusion of regulatory agencies

across Europe and Latin America. Work. Pap. 1, Inst. Barceloni d’Estudis Int.Grabosky P, Braithwaite J. 1986. Of Manners Gentle: Enforcement Strategies of Australian Business Regulatory

Agencies. Melbourne: Oxford Univ. PressGray W, Scholz J. 1993. Does regulatory enforcement work? A panel analysis of OSHA enforcement. Law

Soc. 97:177–213


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


Guba E, Lincoln Y. 1981. Effective Evaluation: Improving the Usefulness of Evaluation Results through Responsiveand Naturalistic Approaches. San Francisco, CA: Jossey-Bass

Gunningham N, Kagan R, Thornton D. 2003. Shades of Green: Business, Regulation and Environment. Stanford,CA: Stanford Univ. Press

Gunz H, Gunz S. 2002. The lawyer’s response to organizational professional conflict: an empirical study ofthe ethical decision making of in-house counsel. Am. Bus. Law J. 39:241–87

Haines F. 1997. Corporate Regulation: Beyond “Punish or Persuade.” Oxford: ClarendonHaines F. 1999. Innocent deaths and regulatory failure: a case study of change in the absence of punishment.

Int. J. Sociol. Law 27:23–50Haines F, Reichman N. 2008. The problem that is global warming: introduction to special issue. Law Policy

30:385–93Harrison K. 1995. Is cooperation the answer? Canadian environmental enforcement in comparative context.

J. Policy Anal. Manag. 14:221–44Hawkins K. 1989. Environment and Enforcement: Regulation and the Social Definition of Pollution. Oxford:

ClarendonHawkins K. 2002. Law as Last Resort. Oxford: Oxford Univ. PressHawkins K, Hutter B. 1993. The response of business to social regulation in England and Wales: an enforce-

ment perspective. Law Policy 15:199–217Heimer C. 1996. Explaining variation in the impact of law: organizations, institutions and professions. Stud.

Law Polit. Soc. 15:29–59Hill C, Kelley P, Agle B, Hitt M, Hoskisson R. 1992. An empirical examination of the causes of corporate

wrongdoing in the United States. Hum. Relat. 45:1055–77Huizinga D, Elliott DS. 1986. Reassessing the reliability and validity of self-report delinquency measures.

J. Quant. Criminol. 2:293–327Hutchens A. 2009. Changing Big Business: The Globalisation of the Fair Trade Movement. Cheltenham, UK:

Edward ElgarHutter B. 2001. Regulation and Risk: Occupational Health and Safety on the Railways. Oxford: Oxford Univ. PressHutter B, Jones C. 2007. From government to governance: external influences on business risk management.

Regul. Gov. 1:27–45Johnston J. 2006. The promise and limits of voluntary management-based regulatory reform: an analysis of

EPA’s strategic goals program. See Coglianese & Nash 2006, pp. 167–200Jordana J, Levi-Faur D. 2004a. The politics of regulation in the age of governance. See Jordana & Levi-Faur

2004b, pp. 1–28Jordana J, Levi-Faur D, eds. 2004b. The Politics of Regulation: Institutions and Regulatory Reforms for the Age of

Governance. Cheltenham, UK: Edward ElgarJupp V. 1989. Methods of Criminological Research. London: Unwin HymanKagan R. 1994. Regulatory enforcement. In Handbook of Regulation and Administrative Law, ed. D Rosenbloom,

R Schwartz, pp. 383–422. New York: Marcel DekkerKey S. 1999. Organizational ethical culture: real or imagined? J. Bus. Ethics 20:217–25King A, Lenox M. 2000. Industry self-regulation without sanctions: the chemical industry’s responsible care

program. Acad. Manag. J. 43:698–716Kleindorfer P. 2006. The risk management program rule and management-based regulation. See Coglianese

& Nash 2006, pp. 87–110Krawiec K. 2003. Cosmetic compliance and the failure of negotiated governance. Wash. Univ. Law Q. 81:487–

544Krawiec K. 2008. The return of the rogue. Ariz. Law Rev. 50:878–918Kuperan K, Sutinen J. 1998. Blue water crime: deterrence, legitimacy, and compliance in fisheries. Law Soc.

Rev. 32:309–37Lange B. 1999. Compliance construction in the context of environmental regulation. Soc. Legal Stud. 8:549–67Lehmkuhl D. 2008. Control modes in the age of transnational governance. Law Policy 30:336–63Levi-Faur D. 2005. The global diffusion of regulatory capitalism. Ann. Am. Acad. Polit. Soc. Sci. 598:12–32Levi-Faur D. 2008. Foreword. See Braithwaite 2008, pp. vii–x


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


Lobel O. 2004. The renew deal: the fall of regulation and the rise of governance in contemporary legal thought.Minn. Law Rev. 89:342

MacNeil I. 2008. Risk-based regulation, the credit crisis and the compliance conundrum. Presented at MelbourneLaw Sch., Dec. 9

May P. 2005. Compliance motivations: perspectives of farmers, homebuilders, and marine facilities. Law Policy27:317–47

May P, Winter S. 2000. Reconsidering styles of regulatory enforcement: patterns in Danish agro-environmental inspection. Law Policy 22:143–73

May P, Wood R. 2003. At the regulatory front lines: inspectors’ enforcement styles and regulatory compliance.J. Public Adm. Res. Theory 13:117–39

Mayhew C, Quinlan M, Ferris R. 1997. The effects of subcontracting/outsourcing on occupational health andsafety: survey evidence from four Australian industries. Saf. Sci. 25:163–78

Mazmanian D, Sabatier P. 1981. Effective Policy Implementation. Lexington, MA: Lexington BooksMazmanian D, Sabatier P. 1983. Implementation and Public Policy. Glenview, IL/London: Scott, ForesmanMcBarnet D. 2003. When compliance is not the solution but the problem: from changes in law to changes in

attitude. See Braithwaite 2003, pp. 229–43McBarnet D, Voiculescu A, Campbell T, eds. 2007. The New Corporate Accountability: Corporate Social Respon-

sibility and the Law. Cambridge, UK: Cambridge Univ. PressMcCaffrey D, Hart D. 1998. Wall Street Polices Itself: How Securities Firms Manage the Legal Hazards of Competitive

Pressures. New York: Oxford Univ. PressMcCaffrey D, Smith A, Martinez-Moyano I. 2006. “Then let’s have a dialogue”: interdependence and nego-

tiation in a cohesive regulatory system. J. Public Adm. Res. Theory 17:307–34McCahery J, Picciotto S, Scott C, eds. 1993. Corporate Control and Accountability: Changing Structures and the

Dynamics of Regulation. Oxford: ClarendonMcDonald M. 2005. Validity, data sources. In Encyclopedia of Social Measurement, ed. K Kempf-Leonard, 3:939–

48. Amsterdam: ElsevierMcGraw K, Scholz J. 1991. Appeals to civic virtue versus attention to self-interest: effects on tax compliance.

Law Soc. Rev. 25:471–99McKendall M, Sanchez C, Sicilian P. 1999. Corporate governance and corporate illegality: the effects of board

structure on environmental sanctions. Int. J. Organ. Anal. 7:201–23Meidinger E. 2008. Competitive supragovernmental regulation: How could it be democratic? Chicago J. Int.

Law 8:513–34Mendeloff J, Gray W. 2005. Inside the black box: How do OSHA inspections lead to reductions in workplace

industries? Law Policy 27:219–37Mosher C, Miethe T, Phillips D. 2002. The Mismeasure of Crime. Thousand Oaks, CA: SageMurphy K. 2005. Regulating more effectively: the relationship between procedural justice, legitimacy, and tax

non-compliance. J. Law Soc. 32:562–89Nielsen VL. 2002. Dialogens Pris. Uformelle spilleregler, ressourceasymmetri og forskelsbehandling i offentligt tilsyn.

PhD thesis. Forlaget Politica, Univ. ArhusNielsen VL. 2006. Are regulators responsive? Law Policy 28:395–416Nielsen VL, Parker C. 2008. To what extent do third parties influence business compliance? J. Law Soc.

35:309–40Nyland C, Smyth R, Zhu C. 2006. What determines the extent to which employers will comply with their

social security obligations? Evidence from Chinese firm-level data. Soc. Policy Adm. 40:196–214Olsen J. 1978. Folkestyre, burakrati og korporativisme. In Politisk Organisering, ed. J Olsen, pp. 13–104. Oslo:

UniversitetsforlagetParker C. 2002. The Open Corporation: Effective Self-Regulation and Democracy. Cambridge, UK: Cambridge

Univ. PressParker C, Nielsen VL. 2009. Do compliance programmes influence compliance? Adm. Soc. 41:3–37Parker C, Scott C, Lacey N, Braithwaite J. 2004. Introduction. In Regulating Law, ed. C Parker, C Scott, N

Lacey, J Braithwaite, pp. 1–12. Oxford: Oxford Univ. PressPartnoy F. 2003. Infectious Greed: How Deceit and Risk Corrupted the Financial Markets. London: Profile Books


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


Prakash A, Potoski M. 2006. The Voluntary Environmentalists: Green Clubs, ISO14001, and Voluntary Regulations.Cambridge, UK: Cambridge Univ. Press

Putnam R. 1993. Making Democracy Work: Civic Traditions in Modern Italy. Princeton, NJ: Princeton Univ.Press

Rees J. 1988. Reforming the Workplace: A Study of Self-Regulation in Occupational Safety. Philadelphia: Univ. Pa.Press

Rees J. 1994. Hostages of Each Other: The Transformation of Nuclear Safety Since Three Mile Island. Chicago: Univ.Chicago Press

Rees J. 1997. Development of communitarian regulation in the chemical industry. Law Policy 19:477–528Reichman N. 1992. Moving backstage: uncovering the role of compliance practices in shaping regulatory pol-

icy. In White Collar Crime Reconsidered, ed. K Schlegel, C Weisburd, pp. 244–68. Boston, MA: Northeast.Univ. Press

Ruggie J. 2002. The theory and practice of learning networks: corporate social responsibility and the GlobalCompact. J. Corp. Citizensh. 5:27–36

Sabel CF. 2007. Rolling rule labor standards: why their time has come, and why we should be glad of it.In Protecting Labour Rights as Human Rights: Present and Future of International Supervision, pp. 257–77.Geneva: Int. Labour Organ.

Sarat A, Silbey S. 1988. The pull of the policy audience. Law Policy 10:97–166Schepel H. 2005. The Constitution of Private Governance: Product Standards in the Regulation of Integrating Markets.

Oxford: HartSchneiberg M, Bartley T. 2008. Organizations, regulation, and economic behavior: regulatory dynamics.

Annu. Rev. Law Soc. Sci. 4:31–61Scholz J, Lubell M. 1998. Trust and taxpaying: testing the heuristic approach to collective action. Am. J. Polit.

Sci. 42:398–417Scott C. 2001. Analysing regulatory space: fragmented resources and institutional design. Public Law

Summer:329–53Shamir R. 2008. Corporate social responsibility: towards a new market-embedded morality? Theor. Inq. Law

9:371–94Shearing C, Wood J. 2003. Nodal governance, democracy, and the new “denizens.” J. Law Soc. 30:400–19Short J, Toffel M. 2007. Coerced confessions: self-policing in the shadow of the regulator. J. Law Econ. Organ.

24:45–71Shover N, Hochstetler A. 2006. Choosing White-Collar Crime. Cambridge, UK: Cambridge Univ. PressSilbey S, Huising R, Coslovsky SV. 2009. The sociological citizen: recognizing relational interdependence in

law and organizations. Annee Sociol. 59:201–29Simpson S. 1987. Cycles of illegality: antitrust violations in corporate America. Soc. Forces 65:943–63Simpson S. 2002. Corporate Crime, Law and Social Control. Cambridge, UK: Cambridge Univ. PressSnider L. 1991. The regulatory dance: understanding reform processes in corporate crime. Int. J. Sociol. Law

19:209–36Solomon J. 2008. Law and governance in the twenty-first century regulatory state. Tex. Law Rev. 86:819–56Teubner G, ed. 1997. Global Law Without a State. Aldershot, UK: DartmouthTombs S. 2000. Official statistics and hidden crime: researching safety crimes. In Doing Criminological Research,

ed. V Jupp, P Davies, P Francis, pp. 64–81. London: SageVogel SK. 1996. Freer Markets, More Rules: Regulatory Reform in Advanced Industrial Countries. Ithaca, NY:

Cornell Univ. PressWeaver G, Trevino LK. 1999. Compliance and values oriented ethics programs: influences on employees’

attitudes and behavior. Bus. Ethics Q. 9:315–35Weil D. 1996. If OSHA is so bad, why is compliance so good? RAND J. Econ. 27:618–40Wenzel M. 2001. The impact of outcome orientation and justice concerns on tax compliance: the role of taxpayer’s

identity. Work. Pap., Cent. Tax Syst. Integr., Aust. Natl. Univ., CanberraWinter S, Elmore RF, with Abelmann C. 1994. Youth employment policy in the USA and Denmark: political culture,

policies, and implementation. Presented at Annu. Res. Conf. Assoc. Public Policy Manag., Chicago, Oct.Winter S, May PJ. 2001. Motivation for compliance with environmental regulations. J. Policy Anal. Manag.

20:675–98


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.


Winter S, May PJ. 2002. Information, interests, and environmental regulation. J. Comp. Policy Anal.: Res. Pract.4:115–42

Wolf K. 2008. Emerging patterns of global governance: the new interplay between the state, business andcivil society. In Handbook of Research on Global Corporate Citizenship, ed. A Scherer, G Palazzo, pp. 225–48.Cheltenham, UK: Edward Elgar


Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.

AR392-FM ARI 30 September 2009 19:8

Annual Review ofLaw and SocialScience

Volume 5, 2009Contents

Morality in the Law: The Psychological Foundations of Citizens’Desires to Punish TransgressionsJohn M. Darley � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 1

Experimental Law and EconomicsRachel Croson � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �25

The Challenge of Empirical Research on Business Compliancein Regulatory CapitalismChristine Parker and Vibeke Nielsen � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �45

Welfare, Workfare, and Citizenship in the Developed WorldJoel F. Handler � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �71

Willpower and Legal PolicyLee Anne Fennell � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �91

More Religion, Less Crime? Science, Felonies, and theThree Faith FactorsJohn J. DiIulio, Jr. � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 115

The Political Economy of ProsecutionSanford C. Gordon and Gregory A. Huber � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 135

Lineups and Eyewitness IdentificationAmy-May Leach, Brian L. Cutler, and Lori Van Wallendael � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 157

Punitive DamagesNeil Vidmar and Matthew W. Wolfe � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 179

Does the Process of Constitution-Making Matter?Tom Ginsburg, Zachary Elkins, and Justin Blount � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 201

The New Legal PluralismPaul Schiff Berman � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 225

Global Legal PluralismRalf Michaels � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 243

Recursivity of Global Normmaking: A Sociolegal AgendaTerence C. Halliday � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 263

v

Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.

AR392-FM ARI 30 September 2009 19:8

Rethinking Sovereignty in International LawAntony Anghie � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 291

Does Torture Work? A Sociolegal Assessment of the Practicein Historical and Global PerspectiveLisa Hajjar � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 311

The Empirical Study of Terrorism: Social and Legal ResearchGary LaFree and Gary Ackerman � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 347

Public Support for Civil Liberties Pre- and Post-9/11John L. Sullivan and Henriet Hendriks � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 375

The Expanding Purview of Cultural Properties and Their PoliticsRosemary J. Coombe � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 393

Indexes

Cumulative Index of Contributing Authors, Volumes 1–5 � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 413

Cumulative Index of Chapter Titles, Volumes 1–5 � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 415

Errata

An online log of corrections to Annual Review of Law and Social Science articles may befound at http://lawsocsci.annualreviews.org

vi Contents

Ann

u. R

ev. L

aw. S

oc. S

ci. 2

009.

5:45

-70.

Dow

nloa

ded

from

ww

w.a

nnua

lrev

iew

s.or

gby

Uni

vers

ity o

f Pe

nnsy

lvan

ia o

n 05

/04/

13. F

or p

erso

nal u

se o

nly.

the challenge of empirical research on business compliance in regulatory capitalism

Documents